Approaches to modelling radioactive contaminations in forests - Overview and guidance

Effect of soil organic matter on the fate of 137Cs vertical distribution in forest soils

Understanding the fate of the vertical distribution of radiocesium (¹³⁷Cs) in Japanese forest soils is key to assessing the radioecological consequences of the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident. The ¹³⁷Cs behavior in mineral soil is known to be primarily governed by interaction with clay minerals; however, some observations suggest the role of soil organic matter (SOM) in enhancing the mobility of ¹³⁷Cs. Here we hypothesized that soil organic carbon (SOC) concentration profile determines the ultimate vertical pattern of ¹³⁷Cs distribution in Japanese forest soils. In testing this hypothesis, we obtained soil samples that were collected before the FDNPP accident at four forest sites with varying SOC concentration profiles and quantified the detailed vertical profile of ¹³⁷Cs inventory in the soils roughly half a century after global fallout in the early 1960 s. Results showed that the higher the SOC concentration in the soil profile, the deeper the ¹³⁷Cs downward penetration. On the basis of the data for surface soils (0-10 cm), the ¹³⁷Cs retention ratio for each of the 2-cm thick layers was evaluated as the ratio of ¹³⁷Cs inventory in the target soil layer to the total ¹³⁷Cs inventory in and below the soil layer. A negative correlation was found between the ratio and SOC concentration of the layer across all soils and depths. This indicates that the ultimate fate of ¹³⁷Cs vertical distribution can be predicted as a function of SOC concentration for Japanese forest soils, and provides further evidence for SOM effects on the mobility and bioavailability of ¹³⁷Cs in soils.

Contamination processes of tree components in Japanese forest ecosystems affected by the Fukushima Daiichi Nuclear Power Plant accident 137Cs fallout

In forests affected by the Fukushima Daiichi Nuclear Power Plant accident, trees became contaminated with ¹³⁷Cs. However, ¹³⁷Cs transfer processes determining tree contamination (particularly for stem wood, a prominent commercial resource) remain insufficiently understood. We propose a model for simulating dynamic behavior of ¹³⁷Cs in a forest tree-litter-soil system and applied it to contaminated forests of cedar plantation and natural oak stand in Fukushima to elucidate relative impact of distinct ¹³⁷Cs transfer processes determining the tree contamination. The transfer of ¹³⁷Cs to the trees occurred mostly (>99%) through surface uptake of ¹³⁷Cs trapped by needles and bark during the fallout. Root uptake of soil ¹³⁷Cs was several orders of magnitude lower than the surface uptake over a 50-year period following the accident. As a result, internal contamination of the trees proceeded through an enduring recycling (translocation) of ¹³⁷Cs absorbed on the tree surface. A significant surface uptake of ¹³⁷Cs through bark was suggested, contributing to 100% (leafless oak tree) and 30% (foliated cedar tree; the remaining uptake occurred at needles) of the total uptake by the trees, although that pathway still needs to be evaluated by experimental evidence. It was suggested that the activity concentration of ¹³⁷Cs in stem wood of the trees at these sites are currently (as of 2021) decreasing by ~3% per year, mainly through radioactive decay of ¹³⁷Cs and partly through dilution effect from tree growth. Although further refinement of the model is recommended, for example by including tree species specific ¹³⁷Cs transportation in stem, these findings provide vital information for planning of forestry reactivation in Fukushima; e.g., removal of forest floor organic layer will not reduce the tree contamination for a long term because of the ¹³⁷Cs absorption via the tree surface substantially greater than root uptake of ¹³⁷Cs deposited to the floor.

Dynamics of radiocaesium within forests in Fukushima-results and analysis of a model inter-comparison

Forests cover approximately 70% of the area contaminated by the Fukushima Daiichi Nuclear Power Plant accident in 2011. Following this severe contamination event, radiocaesium (¹³⁷Cs) is anticipated to circulate within these forest ecosystems for several decades. Since the accident, a number of models have been constructed to evaluate the past and future dynamics of ¹³⁷Cs in these forests. To explore the performance and uncertainties of these models we conducted a model inter-comparison exercise using Fukushima data. The main scenario addressed an evergreen needleleaf forest (cedar/cypress), which is the most common and commercially important forest type in Japan. We also tested the models with two forest management scenarios (decontamination by removal of soil surface litter and forest regeneration) and, furthermore, a deciduous broadleaf forest (konara oak) scenario as a preliminary modelling study of this type of forest. After appropriate calibration, the models reproduced the observed data reliably and the ranges of calculated trajectories were narrow in the early phase after the fallout. Successful model performances in the early phase were probably attributable to the availability of comprehensive data characterizing radiocaesium partitioning in the early phase. However, the envelope of the calculated model end points enlarged in long-term simulations over 50 years after the fallout. It is essential to continue repetitive verification/validation processes using decadal data for various forest types to improve the models and to update the forecasting capacity of the models.

Quantitative modeling of radioactive cesium concentrations in large omnivorous mammals after the Fukushima nuclear power plant accident

Large quantities of radionuclides released by the Fukushima nuclear power plant accident entered terrestrial and marine ecosystems. The resulting radioactive contamination of large omnivorous wild mammals such as wild boar (Sus scrofa) and Asian black bear (Ursus thibetanus) varied greatly depending on location, season, and time after the accident. Quantitative modeling of how such factors influence radionuclide burdens in these species is important for enhancing current knowledge of chronic radionuclide exposure consequences in mammalian populations, and for assessing potential human risks from consumption of contaminated animal meat. Here we modeled the time course of radioactive cesium (¹³⁴Cs + ¹³⁷Cs) concentrations in boar and black bears from Fukushima Prefecture over ~ 7 years after the accident, using nonlinear robust and quantile regressions and mixed-effects modeling. To estimate predictive performance, models fitted to the full data set were compared with those fitted only to the first 3.5 years of data, and tested on the last 3.5 years of data. Ecological half-lives for radioactive cesium, and magnitudes and phase shifts for sinusoidal seasonal oscillations in cesium burdens, were estimated by each analysis method for each species. These results can improve the understanding and prediction of radionuclide concentrations in large mammals that inhabit radioactively contaminated areas.

Meta-analysis of radiocesium contamination data in Japanese cedar and cypress forests over the period 2011-2017

Since Fukushima accident, dozens of field studies have been conducted in order to quantify and understand the behaviour of atmospheric radiocesium (¹³⁷Cs) fallouts in contaminated forests of Fukushima and neighbouring prefectures. In this paper, we carry out a detailed review of data acquired over 2011-2017 in Japanese cedar and cypress plantations, focusing on aerial tree organs, soil layers and tree-to-soil depuration fluxes. To enable comparison and reinforce the consistency between sites, radiological measurements were normalized by the deposit and interpolated onto the same spatio-temporal frame. Despite some (poorly explained) residual variability, we derived a "mean" pattern by log-averaging data among sites. These "mean" results were analysed with the help of a simple mass-balance approach and discussed in the light of post-Fukushima literature. We demonstrated that the activity levels and dynamics in all compartments were consistent and generally well reproduced by the mass balance approach, for values of the interception fraction between 0.7 and 0.85. The analysis indicated that about 5% of the initial deposit remained in the aerial vegetation after 6 years, more than two thirds of intercepted ¹³⁷Cs being transferred to the soil due to throughfall. The simulations indicated that foliar uptake might have contributed between 40% and 100% to the activity transferred to stem wood. The activity concentration in canopy organs rapidly decreased in the first few months then more slowly, according to an effective half-life of about 1.6 years. The activity level in the organic layer peaked in summer 2011 then decreased according to an effective half-life of 2.2 years. After a rapid increase in 2011, the contamination of mineral horizons continued to increase more slowly, 85% of ¹³⁷Cs incoming through the organic layer being retained in the 0-5 cm layer according to a mean residence time longer than in the upper layer (7 against 1.5 years).

Uncertainties in the use of concentration ratios for modelling NORM waste sites

The activity concentrations of ²³⁸U, ²²⁶Ra and ²¹⁰Pb were modelled in Pinus sylvestris (Scots pine trees) on a uniform CaF₂ sludge heap in Belgium. The aim of this work is to enhance the knowledge of how transfer factors behave in NORM landfills. The simplest possible model in radioecology is used, which is based on Concentration Ratios (CR-s) measured in equilibrium and activity concentrations of the above-mentioned radionuclides measured in the substrate where pine trees grow. Two alternative CR-s were used: (1) international CR compilations by the IAEA (2014) and (2) CR-s specifically determined for pine trees studied in British Columbia (Mahon and Mathews, 1983). Both CR-s were applied assuming lognormal distributions fitted from data reported in the literature. The results were compared with activity concentrations measured in trees sampled on-site. Modelled concentrations match the measured ones best in the case of ²³⁸U. For the studied NORM waste site, the approach using generic IAEA concentration ratios does not fulfill the conservatism requirement in the cases of ²³⁸U and ²²⁶Ra, as the concentration of radionuclides in trees is underestimated. On the other hand, the ratios from Mahon and Mathews, (1983) produce wide distributions, ensuring conservatism due to larger CR-s. The measured concentrations are narrowly distributed in general, which can be expected on a small sampling site on a uniform substrate. The generic approach outlined here is practical but, as a result of the uniqueness of the site considered, should be applied cautiously in other NORM situations.

TRIPS 2.0: Toward more comprehensive modeling of radiocaesium cycling in forest

Because internal transfers can play a key role in radiocaesium persistence in trees, a reliable representation of radiocaesium recycling between tree organs in forest models is important for long-term simulations after radioactive fallout in Chernobyl and Fukushima. We developed an upgraded 2.0 version of the initial TRIPS ("Transfer of Radionuclides In Perennial vegetation System") model involving explicit differentiation between tree organs (i.e., foliage, branches, stemwood and bark). The quality of TRIPS 2.0 was evaluated by testing model outputs against independent datasets for pine stands in Belarus and Ukraine. Scenarios involving "hot particle" deposits in forest remained challenging, but in all other scenarios generally positive verification results for soil and tree compartments indicated that the TRIPS 2.0 model adequately combines the major relevant processes. Interestingly, the response of stemwood contamination to changes in radiocaesium availability in soil, as determined by soil conditions, was shown to be more sensitive than for other tree compartments. We recommend the conceptual tree discretization of TRIPS 2.0 for generic forest modeling for two reasons: 1) regardless of different soil conditions, there was concurrent good agreement between simulations and data for individual tree compartments (foliage, branches, stemwood and bark), and 2) the measurements necessary to estimate internal tree transfers are easily accessible to usual field monitoring in forest biogeochemistry (for details, see Goor, F. & Thiry, Y., 2004. Science of the total environment, 325(1-3), 163-180).

A long-term study of radionuclide concentrations in mushrooms in the 30-km zone around the Mining-and-Chemical Combine (Russia)

¹³⁷Cs concentrations were measured in mushrooms in an area affected by radioactive discharges of the Mining-and-Chemical Combine (MCC) (Rosatom, Russia) in 2002-2017. The sources of radionuclides in the study sites were global fallouts and waterborne and airborne radioactive discharges of the MCC. The mushroom species Suillus granulatus and S. luteus showed the highest ¹³⁷Cs concentrations (140-7100 Bq kg^-1) for this area. Over the entire monitoring period, no significant change in ¹³⁷Cs concentration was observed in the Suillus spp. samples collected from the sites with the aerial deposition of radionuclides. In the floodplain site with the radionuclide deposition from water and air, a significant decrease in the average ¹³⁷Cs concentration was observed in the period between 2004 and 2017: a three-fold decrease in Suillus spp. and a nine-fold decrease in Lactarius deliciosus. The effective half-lives of ¹³⁷Cs in fruiting bodies of the mushrooms L. deliciosus and Suillus spp. in this site were 3.6 ± 0.6 and 9.2 ± 2.7 years, respectively.

Fit-for-purpose modelling of radiocaesium soil-to-plant transfer for nuclear emergencies: a review

Numerous radioecological models have been developed to predict radionuclides transfer from contaminated soils to the food chain, which is an essential step in preparing and responding to nuclear emergencies. However, the lessons learned from applying these models to predict radiocaesium (RCs) soil-to-plant transfer following the Fukushima accident in 2011 renewed interest in RCs transfer modelling. To help guide and prioritise further research in relation to modelling RCs transfer in terrestrial environments, we reviewed existing models focussing on transfer to food crops and animal fodders. To facilitate the review process, we categorised existing RCs soil-to-plant transfer models into empirical, semi-mechanistic and mechanistic, though several models cross the boundaries between these categories. The empirical approach predicts RCs transfer to plants based on total RCs concentration in soil and an empirical transfer factor. The semi-mechanistic approach takes into account the influence of soil characteristics such as clay and exchangeable potassium content on RCs transfer. It also uses 'bioavailable' rather than total RCs in soil. The mechanistic approach considers the physical and chemical processes that control RCs distribution and uptake in soil-plant systems including transport in the root zone and root absorption kinetics. Each of these modelling approaches has its advantages and disadvantages. The empirical approach is simple and requires two inputs, but it is often associated with considerably uncertainty due to the large variability in the transfer factor. The semi-mechanistic approach factorises more soil and plant parameters than the empirical approach; therefore, it is applicable to a wider range of environmental conditions. The mechanistic approach is instrumental in understanding RCs mobility and transfer in soil-plant systems; it also helps to identify influential soil and plant parameters. However, the comlexity and the large amount of specific parameters make this approach impractical for nuclear emergency preparedness and response purposes. We propose that the semi-mechanistic approach is sufficiently robust and practical, hence more fit for the purpose of planning and responding to nuclear emergencies compared with the empirical and mechanistic approaches. We recommend further work to extend the applicability of the semi-mechanistic approach to a wide range of plants and soils.

Development and assessment of a simple ecological model (TRIPS) for forests contaminated by radiocesium fallout

The management of vast forested zones contaminated by radiocesium (rCs) following the Chernobyl and Fukushima fallout is of great social and economic concern in affected areas and requires appropriate dynamic models as predictive or questioning tools. Generally, the existing radio-ecological models need less fragmented data and more ecological realism in their quantitative description of the rCs cycling processes. The model TRIPS ("Transfer of Radionuclide In Perennial vegetation Systems") developed in this study privileged an integrated approach which makes the best use of mass balance studies and available explicit experimental data for Scots pine stands. A main challenge was the differentiation and calibration of foliar absorption as well as root uptake in order to well represent the rCs biocycling. The general dynamics of rCs partitioning was simulated with a relatively good precision against an independent series of observed values. In our scenario the rCs biological cycling enters a steady-state about 15 years after the atmospheric deposits. At that time, the simulations showed an equivalent contribution of foliage and root uptake to the tree contamination. But the root uptake seems not sufficient to compensate the activity decline in the tree. The initial foliar uptake and subsequent internal transfers were confirmed to have a great possible impact on the phasing of tree contamination. An extra finding concerns the roots system acting as a buffer in the early period. The TRIPS model is particularly useful in cases where site-specific integrated datasets are available, but it could also be used with adequate caution to generic sites. This development paves the way for simplification or integration of new modules, as well as for a larger number of other applications for the Chernobyl or Fukushima forests once the appropriate data become available. According to the sensitivity analysis that involves in particular reliable estimates of net foliar uptake as well as root uptake not disconnected from rCs exchange reactions in soil.

Calibration of forest 137Cs cycling model "FoRothCs" via approximate Bayesian computation based on 6-year observations from plantation forests in Fukushima

Predicting the environmental fate of ¹³⁷Cs in forest ecosystems along with the concentrations of ¹³⁷Cs in tree parts are important for the managements of radioactively contaminated forests. In this study, we calibrate the Forest RothC and Cs model (FoRothCs), a forest ecosystem ¹³⁷Cs dynamics model, using observational data obtained over six years from four forest sites with different levels of ¹³⁷Cs contamination from Fukushima Prefecture. To this end, we applied an approximate Bayesian computation (ABC) technique based on the observed ¹³⁷Cs concentrations (Bq kg^-1) of five compartments (leaf, branch, stem, litter, and soil) in a Japanese cedar plantation. The environmental decay (increment) constants of the five compartments were used as the summary statistics (i.e., the metric for model performance) to infer the five parameters related to ¹³⁷Cs transfer processes in FoRothCs. The ABC technique successfully reconciled the model outputs with the observed trends in ¹³⁷Cs concentrations at all sites during the study period. Furthermore, the estimated parameters are in agreement with the literature values (e.g., the root uptake rates of ¹³⁷Cs). Our study demonstrates that model calibration with ABC based on the trends in ¹³⁷Cs concentrations of multi compartments is useful for reducing the prediction uncertainty of ¹³⁷Cs dynamics in forest ecosystems.