Over the last decade, a number of approaches have been developed to determine the exposure of, and in some instances risk to, non-human biota (i.e. wildlifeFootnote 1) from ionizing radiation (USDOE 2002; Copplestone et al. 2001; Brown et al. 2008). These have been developed in response to national legislation and changes in international recommendations with regard to the need to demonstrate that the environment is protected from radioactive releases (e.g. IAEA 2006; ICRP 2007) rather than relying on previous, anthropocentric approaches.

Early in the process of developing these approaches, it was suggested that the parameterization of the variables that describe the transfer of radioactivity between ecosystem components would contribute significantly to the uncertainty in the estimated dose rate (Avila et al. 2004). Recent international evaluations of the available models have shown this to be the case (Vives i Batlle et al. 2007, 2010; Beresford et al. 2008a, 2010a, b; Yankovich et al. 2010a). There are many reasons for this including lack of empirical data for many radionuclide–organism combinations and the theoretical approaches used to derive transfer values when no data exist; use of national data sets; and the conservative nature of some of the parameter values selected to meet the purpose of specific tools (Beresford et al. 2008a, 2010a; Yankovich et al. 2010a). Recognizing the need to improve the model parameterization, the International Atomic Energy Agency (IAEA) is in the process of producing a handbook of radionuclide transfer parameters for application in environmental assessments similar to that available for human food chain modelling (IAEA 2010). To facilitate this, an online database has been established to collate and summarize data (http://wiki.ceh.ac.uk/x/-QHbBg); the database will also be used by the International Commission on Radiological Protection (ICRP) to help further develop their framework for environmental protection (ICRP 2003, 2008). In 2009, the IAEA supported three workshops to bring together scientists who may be able to contribute to the handbook, with one aim being to stimulate the provision of previously unpublished data. The papers in this issue of Radiation Environmental Biophysics result from presentations made at these workshops.

The majority of models developed to assess the exposure of wildlife to ionizing radiation use concentration ratios (CRs) to estimate whole-body activity concentrations and, consequently, absorbed dose rates for internal exposure, where the concentration ratio is defined as:

$$ {\text{CR = }}\left( {{\frac{{{\text{Radionuclide}}\,{\text{whole}{\text{-}}{body}}\,{\text{activity}}\,{\text{concentration}}\,\left( {{\text{Bq}} . {\text{kg}}^{ - 1} \,{\text{fresh}}\,{\text{mass}}} \right)}}{{{\text{Radionuclide}}\,{\text{activity}}\,{\text{concentration}}\,{\text{in}}\,{\text{media}}\,\left( {{\text{Bq}}\,{\text{per}}\,{\text{unit}}\,{\text{media}}\,{\text{mass}}\,{\text{or}}\,{\text{volume}}} \right)}}}} \right) $$

where the media for terrestrial ecosystems is usually soil (Bq kg−1 dry mass) and for aquatic ecosystems water (Bq l−1), although for some radionuclides (e.g. 3H, 14C, 35S), whole-body activity concentrations are related to activity concentrations in air (Bq m−3).

This is obviously a simplistic approach that amalgamates many biogeochemical processes. However, to enable consideration of the wide range of radionuclides and wildlife species that may need to be considered in assessments, it currently represents a pragmatic approach that meets the needs of ‘prospective’ and ‘current’ exposure (ICRP 2007) assessments. The estimation of whole-body activity concentrations and, consequently, whole-body dose rates (rather than consideration of target tissues) also enables comparison with the available effects data, which originate from studies of external gamma exposure (Garnier-Laplace et al. 2010).

A number of the papers in the current issue of Radiation and Environmental Biophysics present data for poorly studied organism types: reptiles (Wood et al. 2010), bats (Gashchak et al. 2010), marsupials (Johansen and Twining 2010) and invertebrates (Dragović and Mandić 2010; Johansen and Twining 2010). In some regulatory assessments, such organisms are often the object of protection (e.g. Copplestone et al. 2003). Other papers present compilations of data previously unavailable in the open literature: Fesenko et al. (2010) provide a review of data for the transfer of radionuclides to marine organisms from the Russian language literature, and Johansen and Twining (2010) summarize data from the grey literature reporting studies conducted in Australia. Takata et al. (2010) demonstrate (for estuarine species) the value of techniques such as inductively coupled plasma (ICP) mass spectrometry and ICP optical emission spectrometry. These techniques can, from a given sample, provide data for a wide range of elements of relevance to radiological assessments many of which are currently poorly understood. Such methodologies are a cost- and resource-effective tool and are also used in the works described by Higley (2010) and Tagami and Uchida (2010).

Whilst these papers, and other recent works, make valuable additions to the data underlying wildlife assessment models, we have to acknowledge that, realistically, we will never have parameters for all of the potential radionuclide–species combinations that may be required in assessments. To overcome this lack of data, a number of approaches have been derived and used in the existing models (Copplestone et al. 2003; Beresford et al. 2008b; USDOE (United States Department of Energy) 2002). In some cases, the resultant parameters used in the models are only intended to be applicable in initial screening level assessments, which aim to be conservative (Copplestone et al. 2003; Brown et al. 2008). Recent intercomparison exercises have shown that the applications of such parameters would generally result in conservative (i.e. high) estimates of internal dose rate as intended, although in some instances, they would underestimate (Beresford et al. 2008a, 2010a, b; Yankovich et al. 2010a). There is a need for scientifically robust approaches to provide parameters for radionuclide–species combinations for which there are no data. The requirements for such parameters will vary and include the need to be inclusive in initial screening level assessment models and the need to consider protected species (for which it may not be possible to obtain data) in more refined assessments.

Three of the papers in this issue of Radiation Environmental Biophysics evaluate approaches to estimate radionuclide transfer to wildlife when appropriate data are not available (Higley 2010; Tagami and Uchida 2010; Willey 2010). Biological scaling, or allometry, is the consideration of the effect of size on biological variables. Many different variables have been shown to be dependent on mass (e.g. Peters 1983). The dependence of a biological variable Y on a body mass M is typically characterized by allometric equations of the form Y = aM b. Radioecological transfer parameters for terrestrial and aquatic animals for, to date, a limited number of radionuclides have been shown to fit such allometric relationships (e.g. Higley et al. 2003; Vives i Batlle et al. 2009). However, Higley (2010) reports that evidence to support the concept of using allometric scaling functions to estimate radionuclide activity concentrations in plants was inconclusive. This is perhaps not surprising as Tagami and Uchida (2010) present an analysis of tree and crop leaf data that suggest that CR values derived for crops [which are relatively numerous (IAEA 2010)] could be applied to predict radionuclide transfer to trees if data for trees were lacking. The data presented by Higley (2010) suggest that the ionic potential of an element could be used to predict its CR for plants. Willey (2010) demonstrates that radionuclide transfer to plants can be related to plant evolutionary history (or phylogeny), with, for instance, monocots being shown to have lower CR values for the elements considered (Cs, Sr, Co, Cl and Ru) than eudicots. The potential phylogenetic influence of radionuclide uptake has also recently been demonstrated for marine fish by Jeffree et al. (2010).

Much of the available radioecological data for wild species originate from measurements taken for human food chain assessments. These data are, therefore, for tissues consumed by humans (typically for animal muscle). To utilize these data for the purposes of environmental assessment, we need to be able to convert tissue-specific data to whole-body activity concentrations. To enable best use of tissue-specific data to be made, Yankovich et al. (2010b) provide a series of look-up tables for a wide range of elements with whole-body/tissue-specific concentration ratios for marine, freshwater and terrestrial organisms.

The forthcoming IAEA handbook of transfer parameters for wildlife will make a significant contribution to the ongoing development of models to enable assessments of the potential impact of ionizing radiation on wildlife. However, it will be far from complete and is likely to present collated CR values at broad ecosystem and organism levels. There will be an ongoing need to better represent the variability observed in wildlife radionuclide CR values and further develop scientifically robust approaches, such as phylogenetic and allometric relationships, to provide transfer parameter values when data are lacking. The level of ‘acceptable’ uncertainty in transfer parameters varies with assessment level and, potentially, radionuclide (as some radionuclides contribute relatively little to overall releases). For more refined assessments, there may be the need to develop more mechanistically based models. As far as possible, such developments should not be conducted in isolation from the development of models to assess humans. Many of the factors influencing transfer are predominantly the same for humans (via foodstuffs) and wildlife; some existing human food chain models, especially those for aquatic systems, should be readily adaptable for wildlife assessments. Similarly, there has been much international effort to establish databases of parameters describing the transfer of radionuclides to human foodstuffs (IAEA 2010) and wildlife (http://wiki.ceh.ac.uk/x/-QHbBg). However, as noted by Copplestone et al. (2010), these activities are being conducted separately and in future should be combined to better use resources and ensure consistency between the data underpinning human and wildlife assessments.