The role of physical processes controlling the behaviour of radionuclide contaminants in the aquatic environment: a review of state-of-the-art modelling approaches

This paper is aimed at presenting and discussing the methodologies implemented in state-of-the-art models for predicting the physical processes of radionuclide migration through the aquatic environment, including transport due to water currents, diffusion, settling and re-suspension. Models are briefly described, model parameter values reviewed and values recommended. The different modelling approaches are briefly classified and the advantages and disadvantages of the various model approaches and methodologies are assessed.     

Short and long term dispersion patterns of radionuclides in the atmosphere around the Fukushima Nuclear Power Plant

The Chernobyl accident and unfortunately the recent accident at the Fukushima 1 Nuclear Power Plant are the most serious accidents in the history of the nuclear technology and industry. Both of them have a huge and prolonged impact on environment as well as human health. Therefore, any technological developments and strategies that could diminish the consequences of such unfortunate events are undisputedly the most important issues of research. Numerical simulations of dispersion of radionuclides in the atmosphere after an accidental release can provide with a reliable prediction of the path of the plume. In this study we present a short (one month) and a long (11 years) term statistical study for the Fukushima 1 Nuclear Power Plant to estimate the most probable dispersion directions and plume structures of radionuclides on local scale using a Gaussian dispersion model. We analyzed the differences in plume directions and structures in case of typical weather/circulation pattern and provided a statistical-climatological method for a “first-guess” approximation of the dispersion of toxic substances. The results and the described method can support and used by decision makers in such important cases like the Fukushima accident.     

Sources of anthropogenic radionuclides in the environment: a review

Studies of radionuclides in the environment have entered a new era with the renaissance of nuclear energy and associated fuel reprocessing, geological disposal of high-level nuclear wastes, and concerns about national security with respect to nuclear non-proliferation. This work presents an overview on sources of anthropogenic radionuclides in the environment, as well as a brief discussion of salient geochemical behavior of important radionuclides. We first discuss the following major anthropogenic sources and current developments that have lead, or could potentially contribute, to the radionuclide contamination of the environment: (1) nuclear weapons program; (2) nuclear weapons testing; (3) nuclear power plants; (4) uranium mining and milling; (5) commercial fuel reprocessing; (6) geological repository of high-level nuclear wastes that include radionuclides might be released in the future, and (7) nuclear accidents. Then, we briefly summarize the inventory of radionuclides ⁹⁹Tc and ¹²⁹I, as well as geochemical behavior for radionuclides ⁹⁹Tc, ¹²⁹I, and ²³⁷Np, because of their complex geochemical behavior, long half-lives, and presumably high mobility in the environment; biogeochemical cycling and environment risk assessment must take into account speciation of these redox-sensitive radionuclides.     

Vertical migration of radionuclides in undisturbed grassland soils

Literature data on numerical values obtained for the parameters of the two most popular models for simulating the migration of radionuclides in undisturbed soils have been compiled and evaluated statistically. Due to restrictions on the applicability of compartmental models, the convection–dispersion equation and its parameter values should be preferred. For radiocaesium, recommended values are derived for its effective convection velocity and dispersion coefficient. Data deficiencies still exist for radionuclides other than caesium and for soils of non-temperate environments.     

Modelling of long-term behaviour of caesium and strontium radionuclides in the Arctic environment and human exposure

In this paper a compartment model of the highly vulnerable Arctic terrestrial food chain "lichen-reindeer-man" is outlined. Based upon an analysis of measured (137)Cs and (90)Sr contents in lichen and reindeer meat from 1961 up to 2001, site specific model parameters for two regions in north-western Arctic Russia and for Kautokeino municipality in Arctic Norway have been determined. The dynamics of radionuclide activity concentrations in the "lichen-reindeer-man" food chain for all areas was satisfactorily described by a double exponential function with short-term and long-term effective ecological half-lives between 1-2 and 10-12 years, respectively, for both (137)Cs and (90)Sr. Using parameter values derived from the model, life-time internal effective doses due to consumption of reindeer meat by reindeer-breeders after an assumed single pulse deposit of 1 kBq m(-2) of (137)Cs were estimated to be 11.4 mSv (Kola Peninsula), 5 mSv (Nenets Autonomous Area), and 2 mSv (Kautokeino, Norway). Differences in vulnerability to radiocaesium deposition were due to differences in transfer between lichen and reindeer and in diet between the three regions.     

Fractionation of radionuclide species in the environment

Naturally occurring and artificially produced radionuclides in the environment may be present in different physico-chemical forms (i.e., radionuclide species) varying in size (nominal molecular mass), charge properties and valence, oxidation state, structure and morphology, density, degree of complexation, etc. Low molecular mass (LMM) species are believed to be mobile and potentially bioavailable, while high molecular mass (HMM) species such as colloids, polymers, pseudocolloids and particles are considered inert. Due to time-dependent transformation processes such as mobilisation of radionuclide species from solid phases or interactions of mobile and reactive radionuclide species with components in soils and sediments, the original distribution of radionuclides deposited in ecosystems will change over time. To assess the environmental impact from radionuclide contamination, information on radionuclide species deposited, interactions within affected ecosystems and the time-dependent distribution of radionuclide species influencing mobility and biological uptake is essential. The development of speciation techniques to characterize radionuclide species in waters, soils and sediments should therefore be essential for improving the prediction power of impact and risk assessment models. The present paper reviews available fractionation techniques which can be utilised for radionuclide speciation purposes.     

Modelling the reworking effects of bioturbation on the incorporation of radionuclides into the sediment column: implications for the fate of particle-reactive radionuclides in Irish Sea sediments

A microcosm laboratory experiment was conducted to determine the impact of biological reworking by the ragworm Nereis diversicolor on the redistribution of particle-bound radionuclides deposited at the sediment-water interface. Over the course of the 40-day experiment, as much as 35% of a ¹³⁷Cs-labelled particulate tracer deposited on the sediment surface was redistributed to depths of up to 11 cm by the polychaete. Three different reworking models were employed to model the profiles and quantify the biodiffusion and biotransport coefficients: a gallery-diffuser model, a continuous sub-surface egestion model and a biodiffusion model. Although the biodiffusion coefficients obtained for each model were quite similar, the continuous sub-surface egestion model provided the best fit to the data. The average biodiffusion coefficient, at 1.8 ± 0.9 cm² y⁻¹, is in good agreement with the values quoted by other workers on the bioturbation effects of this polychaete species. The corresponding value for the biotransport coefficient was found to be 0.9 ± 0.4 cm y⁻¹. The effects of non-local mixing were incorporated in a model to describe the temporal evolution of measured ⁹⁹Tc and ⁶⁰Co radionuclide sediment profiles in the eastern Irish Sea, influenced by radioactive waste discharged from the Sellafield reprocessing plant. Reworking conditions in the sediment column were simulated by considering an upper mixed layer, an exponentially decreasing diffusion coefficient, and appropriate biotransport coefficients to account for non-local mixing. The diffusion coefficients calculated from the ⁹⁹Tc and ⁶⁰Co cores were in the range 2-14 cm² y⁻¹, which are consistent with the values found by other workers in the same marine area, while the biotransport coefficients were similar to those obtained for a variety of macrobenthic organisms in controlled laboratories and field studies.     

Assessing field-scale migration of radionuclides at the Nevada Test Site: “mobile” species

Many long-lived radionuclides are present in groundwater at the Nevada Test Site (NTS) as a result of 828 underground nuclear weapons tests conducted between 1951 and 1992. In conjunction with a comprehensive geochemical review of radionuclides (³H, ¹⁴C, ³⁶Cl, ⁹⁹Tc and ¹²⁹I) that are presumably mobile in the subsurface, we synthesized a body of radionuclide activity data measured from groundwater samples collected at 18 monitoring wells, to qualitatively assess their migration at the NTS over distances of hundreds of meters and over timescales of decades. Tritium and ³⁶Cl showed little evidence of retardation, while the transport of ¹⁴C may have been retarded by its isotopic exchange with carbonate minerals in the aquifer. Observed local reducing conditions (either natural or test-induced) will impact the mobility of certain redox-sensitive radionuclides (especially ⁹⁹Tc) that were otherwise soluble and readily transported under oxidizing conditions. Conversely, strongly oxidizing conditions may impact the mobility of ¹²⁹I which is mobile under reducing conditions. The effect of iodine speciation on its transport deserves further attention. Indication of delayed transport of some “mobile” radionuclides (especially ⁹⁹Tc) in the groundwater at the NTS suggested the importance of redox conditions of the natural system in controlling the fate and transport of radionuclides, which has implications in the enhanced performance of the potential Yucca Mountain repository, located adjacent to the NTS, to store high-level nuclear wastes as well as management of radionuclide contamination in legacy nuclear operations facilities.     

Distribution of radionuclides among the main components of Lake Chervyanoe in the Eastern Ural Radioactive Trace

Radioecological conditions have been estimated in Lake Chervyanoe located on the central axis of the Eastern Ural Radioactive Trace (EURT) about 100–110 km northeast of the Mayak Production Association. Data on the ⁹⁰Sr and ¹³⁷Cs contents in the main components of the lake, the distribution of these radionuclides over the profile of bottom sediments, and their amounts in the lake are presented. The experimental data have been used to construct mathematical models for estimating changes in the radionuclide concentrations and amounts in the water and bottom sediments of the lake over a long period after the nuclear accident, as well as the tritium content of the lake water in different periods of time.     

Regional-scale application of the decision support system MOIRA-PLUS: an example of assessment of the radiological impact of the Chernobyl accident on the fresh water ecosystem in Italy

The present work illustrates the customisation and application of the decision support system MOIRA-PLUS (a MOdel-based computerised system for management support to Identify optimal remedial strategies for Restoring radionuclide contaminated Aquatic ecosystems and drainage areas) to the fresh water environment in Italy. MOIRA-PLUS is aimed at evaluating the behaviour of radiocaesium and radiostrontium in fresh water ecosystems and at assessing the appropriateness of suitable strategies for the management of contaminated water bodies by the application of multi-attribute analysis techniques. MOIRA-PLUS can be applied to complex networks of lakes, rivers and tributaries and can be straightforwardly customised utilising data and information from readily accessible sources such as official websites provided by scientific or government organisations. The present work shows an application of the decision system to 10 lakes and 18 rivers in Italy contaminated with ¹³⁷Cs of Chernobyl origin. Site-specific values of some aggregated transfer parameters were estimated for the most important Italian lakes. Although high values of fish and water consumptions were hypothesised, very low doses to public from the fresh water pathway following the accident were calculated.     

Modelling the long-term consequences of a hypothetical dispersal of radioactivity in an urban area including remediation alternatives

The Urban Remediation Working Group of the International Atomic Energy Agency's EMRAS (Environmental Modelling for Radiation Safety) program was organized to address issues of remediation assessment modelling for urban areas contaminated with dispersed radionuclides. The present paper describes the second of two modelling exercises. This exercise was based on a hypothetical dispersal of radioactivity in an urban area from a radiological dispersal device, with reference surface contamination at selected sites used as the primary input information. Modelling endpoints for the exercise included radionuclide concentrations and external dose rates at specified locations, contributions to the dose rates from individual surfaces, and annual and cumulative external doses to specified reference individuals. Model predictions were performed for a “no action” situation (with no remedial measures) and for selected countermeasures. The exercise provided an opportunity for comparison of three modelling approaches, as well as a comparison of the predicted effectiveness of various countermeasures in terms of their short-term and long-term effects on predicted doses to humans.     

Simulating mesoscale transport and diffusion of radioactive noble gases using the Lagrangian particle dispersion model

In order to simulate the impact of mesoscale wind fields and to assess potential capability of atmospheric Lagrangian particle dispersion model (LPDM) as an emergency response model for the decision supports, two different simulations of LPDM with the mesoscale prognostic model MM5 (Mesoscale Model ver. 5) were driven. The first simulation of radioactive noble gas ((85)Kr exponent) emitted during JCO accident occurred from 30 September to 3 October 1999 at Tokai, Japan showed that the first arriving short pulse was found in Tsukuba located at 60km away from the accidental area. However, the released radioactive noble gas was transported back to the origin site about 2 days later due to the mesoscale meteorological wind circulation, enhancing the levels of (85)Kr with the secondary peak in Tsukuba. The second simulation of atmospheric dilution factors (the ratio of concentration to the emission rate, chi/Q), during the underground nuclear test (UNT) performed by North Korea showed that high chi/Q moved to the eastward and extended toward southward in accordance with the mesoscale atmospheric circulations generated by mesoscale prognostic model MM5. In comparison with the measurements, the simulated horizontal distribution patterns of (85)Kr during the JCO are well accord with that of observation in Tsukuba such as the existence of secondary peak which is associated with the mesoscale circulations. However, the simulated level of (85)Kr anomaly was found to be significantly lower than the observations, and some interpretations on these discrepancies were described. Applications of LPDM to two mesoscale emergency response dispersion cases suggest the potential capability of LPDM to be used as a decision support model provided accurate emission rate of accident in case of a large accident.     

A study of the atmospheric dispersion of a high release of krypton-85 above a complex coastal terrain, comparison with the predictions of Gaussian models (Briggs, Doury, ADMS4)

Atmospheric releases of krypton-85, from the nuclear fuel reprocessing plant at the AREVA NC facility at La Hague (France), were used to test Gaussian models of dispersion. In 2001-2002, the French Institute for Radiological Protection and Nuclear Safety (IRSN) studied the atmospheric dispersion of 15 releases, using krypton-85 as a tracer for plumes emitted from two 100-m-high stacks. Krypton-85 is a chemically inert radionuclide. Krypton-85 air concentration measurements were performed on the ground in the downwind direction, at distances between 0.36 and 3.3 km from the release, by neutral or slightly unstable atmospheric conditions. The standard deviation for the horizontal dispersion of the plume and the Atmospheric Transfer Coefficient (ATC) were determined from these measurements. The experimental results were compared with calculations using first generation (Doury, Briggs) and second generation (ADMS 4.0) Gaussian models. The ADMS 4.0 model was used in two configurations; one takes account of the effect of the built-up area, and the other the effect of the roughness of the surface on the plume dispersion. Only the Briggs model correctly reproduced the measured values for the width of the plume, whereas the ADMS 4.0 model overestimated it and the Doury model underestimated it. The agreement of the models with measured values of the ATC varied according to distance from the release point. For distances less than 2 km from the release point, the ADMS 4.0 model achieved the best agreement between model and measurement; beyond this distance, the best agreement was achieved by the Briggs and Doury models.