Sensitivity analysis in a radiological impact assessment of a nuclear power plant discharge. A comparison of the Morris, Spearman and Sobol' approaches.

This paper compares the Morris, Spearman and Sobol' methods of sensitivity analysis in radiological risk assessment. The determination of the most influential parameters on model with regards to the propagation of their uncertainties to output variables, is of greatest interest. This study aims to determine the relative importance of parameters uncertainties on the dose calculation uncertainty in the framework of a scenario of routine discharges discussed in the context of an IAEA working group. The scenario considers atmospheric and liquid discharges of three different types of radionuclides (14C, tritium as HTO and 110mAg) from a nuclear power plant located by the side of a river. It is concluded that the most reliable and practical method according to the ability of ranking influential parameters and the easiness of its application is the Spearman method. As key result, the three first influential variables for annual total dose for all pathways and all radionuclides were the water dissolved inorganic carbon concentration, the volatilisation rate constant and the soil layer solid liquid distribution in 14C.

Factors controlling the variability of 137Cs concentrations in 5 coastal rivers around Fukushima Dai-ichi power plant.

The Fukushima Dai-ichi Nuclear Power Plant (FDNPP) accident led to the contamination by radiocesium (137Cs) of large drained areas. Cesium-137 concentrations in rivers result from complex transfer processes, depending on multiple forcings. Better knowledge of the factors controlling these concentrations is therefore a prerequisite to improve predictions of 137Cs transfers within river catchments. This study aimed at analyzing the spatial and temporal variability of 137Cs concentrations in rivers and identifying the key factors controlling their variability. Published values of 137Cs concentrations in rivers in the north of FDNPP were collected, characterizing 122 sampling sites from May 2011 to October 2014. It resulted in three datasets: dissolved concentrations CW (Bq/L), concentrations in suspended sediment CSS (Bq/kg) and total concentrations CT (Bq/L). The resulting database reflected a large variety of catchments and hydrological conditions. Observed 137Cs concentrations varied by 2-4 orders of magnitude and were poorly explained (R2 = 0.13-0.38) by the average contamination density. Indices summarizing the complex spatial and temporal properties of the catchments were proposed as candidate explanatory variables of concentrations in rivers. They were selected by stepwise regression for each dataset (CW, CSS, CT). For the three datasets, the selection and combination of 5-10 indices significantly better explained this variability (R2 = 0.69-0.83). Deposit indices were identified as first drivers of concentrations in rivers. A deposit index was selected for each dataset, indicating no effect of the contamination distribution for CW, whereas CT and CSS required considering the distribution of contamination and connectivity, as well as the presence of dams for CSS. The others selected variables significantly contributed to explain the concentration variability. This meta-analysis emphasizes the importance of structural (e.g. slope, land-cover) and functional (e.g. delay, season, rainfall) properties in the dissimilarities of catchments responses, stressing that assessments could be improved by including more these properties in models.

Improving transfer functions to describe radiocesium wash-off fluxes for the Niida River by a Bayesian approach.

This paper proposed methodological refinements of the generic transfer function approach to reconstruct radiocesium wash-off fluxes from contaminated catchments, by the integration of hydrological descriptors (passed volume of water, flow rate fluctuations and antecedent flow conditions). The approach was applied to the Niida River (Fukushima prefecture, Japan) for the period 03/2011-03/2015, for which daily flow rate (m3/s) and infrequent total radiocesium concentration (Bq/L) values were available from literature. Three models were defined, generic TF (Φ0), flow-corrected time variant (Φ1) and antecedent-flow corrected variant (Φ2). Calibration of these models' parameters was performed with a Bayesian approach because it is particularly adapted to limited datasets and censored information, and it provides parameters distributions. The model selection showed strong evidence of model Φ2 (indicated by marginal likelihood), which integrates current and recent hydrology in its formulation, and lower prediction errors (indicated by RMSE and ME). Models Φ1 and Φ2 better described wash-off dynamics compared to model Φ0, due to the inclusion of one or several hydrological descriptors. From March 2011 to March 2015, model Φ2 estimated 137Cs export from Niida catchment between 0.32 and 0.67 TBq, with a median value of 0.49 TBq, which represents around 0.27% of the initial fallout and could represent a significant source-term to the Ocean compared to the direct release from Fukushima Dai-ichi Nuclear Power Plant (FDNPP). Moreover the remaining 99% of the initial radiocesium fallout within the catchment may constitute a persistent contamination source for wash-off. Although the proposed methodology brought improvements in the assessment of wash-off fluxes, it remains an empirical interpolation method with a limited predictive power, particularly for recent low activities. To improve predictions, modelling approaches require more observed data (particularly more activity values corresponding to more hydrological conditions), and the inclusion of more hydrological descriptors.

Identifiability of sorption parameters in stirred flow-through reactor experiments and their identification with a Bayesian approach.

This paper addresses the methodological conditions -particularly experimental design and statistical inference- ensuring the identifiability of sorption parameters from breakthrough curves measured during stirred flow-through reactor experiments also known as continuous flow stirred-tank reactor (CSTR) experiments. The equilibrium-kinetic (EK) sorption model was selected as nonequilibrium parameterization embedding the Kd approach. Parameter identifiability was studied formally on the equations governing outlet concentrations. It was also studied numerically on 6 simulated CSTR experiments on a soil with known equilibrium-kinetic sorption parameters. EK sorption parameters can not be identified from a single breakthrough curve of a CSTR experiment, because Kd,1 and k- were diagnosed collinear. For pairs of CSTR experiments, Bayesian inference allowed to select the correct models of sorption and error among sorption alternatives. Bayesian inference was conducted with SAMCAT software (Sensitivity Analysis and Markov Chain simulations Applied to Transfer models) which launched the simulations through the embedded simulation engine GNU-MCSim, and automated their configuration and post-processing. Experimental designs consisting in varying flow rates between experiments reaching equilibrium at contamination stage were found optimal, because they simultaneously gave accurate sorption parameters and predictions. Bayesian results were comparable to maximum likehood method but they avoided convergence problems, the marginal likelihood allowed to compare all models, and credible interval gave directly the uncertainty of sorption parameters θ. Although these findings are limited to the specific conditions studied here, in particular the considered sorption model, the chosen parameter values and error structure, they help in the conception and analysis of future CSTR experiments with radionuclides whose kinetic behaviour is suspected.

Experimental quantification of radiocesium recycling in a coniferous tree after aerial contamination: Field loss dynamics, translocation and final partitioning.

After foliar interception of radioactive atmospheric fallout by forest trees, the short-term recycling dynamics of radiocesium from the tree to the soil as well as within the tree is a primary area of uncertainty in the modeling of the overall cycle. The partitioning of radiocesium transfers in a spruce tree exposed to aerial deposits was investigated during one growth season to reveal the dynamics and significance of underlying processes. The rate of radiocesium loss resulting from foliage leaching (wash-off) was shown to have a functional dependence on the frequency of rainy episodes in a first early stage (weathering 60% of initial contamination during 70 days) and on the amount of precipitation in a second stage (weathering 10% of initial deposits during the following 80 days). A classical single exponential decay model with offset and continuous time as predictor lead to a removal half-life t1/2 of intercepted radiocesium of 25 days. During the growth season, the similar pattern of the internal (134)Cs content in new shoots and initially contaminated foliage confirmed that radiocesium was readily absorbed from needle surfaces and efficiently translocated to growing organs. In the crown, a pool of non-leachable (134)Cs (15-30%) was associated with the abiotic layer covering the twigs and needle surfaces. At the end of the growth season, 30% of the initial deposits were relocated to different tree parts, including organs like stemwood (5%) and roots (6%) not directly exposed to deposition. At the scale of the tree, 84% of the residual activity was assimilated by living tissues which corresponds to a foliar absorption rate coefficient of 0.25 year(-1) for modeling purposes. According to the significant amount of radiocesium which can be incorporated in tree through foliar uptake, our results support the hypothesis that further internal transfers could supply the tree internal cycle of radiocesium extensively, and possibly mask the contribution of root uptake for a long time.

Evaluating variability and uncertainty in radiological impact assessment using SYMBIOSE.

SYMBIOSE is a modelling platform that accounts for variability and uncertainty in radiological impact assessments, when simulating the environmental fate of radionuclides and assessing doses to human populations. The default database of SYMBIOSE is partly based on parameter values that are summarized within International Atomic Energy Agency (IAEA) documents. To characterize uncertainty on the transfer parameters, 331 Probability Distribution Functions (PDFs) were defined from the summary statistics provided within the IAEA documents (i.e. sample size, minimal and maximum values, arithmetic and geometric means, standard and geometric standard deviations) and are made available as spreadsheet files. The methods used to derive the PDFs without complete data sets, but merely the summary statistics, are presented. Then, a simple case-study illustrates the use of the database in a second-order Monte Carlo calculation, separating parametric uncertainty and inter-individual variability.

Kinetics of selenate sorption in soil as influenced by biotic and abiotic conditions: a stirred flow-through reactor study.

This study (i) quantified the kinetics of selenate sorption and (ii) measured the influence of biotic processes in soil selenate stabilisation. Stirred flow-through reactor experiments were conducted on samples of a silty clay soil (pH = 8, Eh = 240-300 mV) from Bure (France) in both non-sterile and sterile conditions. Parameters of the proposed two-site sorption model (EK), adapted from van Genuchten and Wagenet (1989), were estimated by nonlinear regression. Fast selenate sorption on type-1 sites was moderate, with an equilibrium constant of 25.5 and 39.1 L/kg for non-sterile and sterile conditions. Rate-limited sorption on type-2 sites increased with time, and was predominant for longer periods of time in non-sterile conditions. At equilibrium, it would represent over 96% of the sorbed inventory, with mean sorption times of 17 h and 191 h for non-sterile and sterile conditions. Our results showed for Bure soil that (i) selenate sorption in flowing and mildly-oxidising conditions was strongly kinetically controlled, especially in non-sterile conditions, (ii) selenate desorption was much slower than sorption, which suggests its pseudo-irreversible stabilisation, and (iii) microbial activity increased the contribution of rate-limited sorption on type-2 sites, for which it increased sorption rate by a factor 7 but also facilitated its reversibility. This work stresses the limits of the Kd approach to represent selenate sorption in flowing conditions and supports an alternative formulation like the EK model, but also points out that biotic conditions are significant sources of variability for sorption parameters.

Effects of radionuclide and rainfall characteristics on field loss parameters of grass.

The decrease of foliar activity in vegetation after its initial contamination by foliar deposition is termed "field loss" (Chamberlain, 1970). This work investigated further laboratory data concerning field loss of (134)Cs, (137)Cs, (85)Sr, (133)Ba and (123m)Te deposited on grassland (Madoz-Escande et al., 2005). Treatments consisted in rainfall scenarios cumulating 14 mm per week, combining two levels of intensity (8 or 30 mm/h) and two levels of frequency/precocity (late once or early twice-a-week). The time course of field loss was monitored in the edible tissues which were sampled by mowing between the rainfalls. Data were analyzed with an offset exponential loss model which is applicable to chronic contamination and is consistent with approaches adopted in radiological assessment models. Its parameters were estimated by the maximum-likelihood method, and their accuracy was determined by nonparametric bootstrap. Radionuclide and rainfall conditions significantly affected the estimated rate (lambda(1)) and extent (A(1)) of field loss. Field loss rate (lambda(1)) and nonentrainable fraction (1-A(1)) varied by a factor 1.5-3. Cesium was very mobile but persistent. On the contrary Tellerium was found less labile, but eventually was almost completely eliminated. Strontium and Barium had intermediate behaviors. Field loss was more efficient for moderate late once-a-week rainfalls (8mm/h). Higher rainfall intensity reduced more the radionuclides losses than higher rainfall frequency/precocity. This paper reports statistically relevant effects that should be considered for more realistic assessments.

Watershed wash-off of atmospherically deposited radionuclides: a review of normalized entrainment coefficients.

Radionuclide wash-off is the transport of activity by flowing water over the soil surface (runoff). To complete existing reviews on long-term removal rates, this paper focuses on short-term wash-off fluxes, quantified in the literature by soil-runoff transfer factors called normalized liquid and solid entrainment coefficients (noted K(l)(*), K(s)(*)). Compiled data concerned essentially (137)Cs and (90)Sr wash-off measured under simulated rainfalls on small experimental plots after Chernobyl fallout in the exclusion zone. K(l)(*) and K(s)(*) values span approximately one order of magnitude. Their validity is limited to a season, and their representativeness is limited by restricted studied situations, notably dominant unsoluble forms in fallout, light soils and intense rainfalls. Formulas based on a simplified representation of the soil-runoff system were proposed to generalize the existing values for other conditions. However, their implementation requires a more systematic compilation of the available information, including decisive influence factors such as the fraction of exchangeable form, distribution coefficient, suspended matter enrichment ratio. Entrainment coefficients K(l)(*) and K(s)(*) were mathematically related to the transfer function approach. The proposed relationships proved their complementarity in terms of time support and captured fluctuations. Both approaches should be used in assessments to estimate average fluxes and their variability.

Watershed wash-off of atmospherically deposited radionuclides: review of the fluxes and their evolution with time.

Radionuclide transport with water from contaminated watersheds is an intermittent flux--also called watershed wash-off--which causes radionuclide redistribution in terrestrial ecosystems, and more critically radionuclide delivery to downstream waterbodies. Mean wash-off fluxes and their evolution with time have been generally quantified with transfer functions, which can be seen as transfer factors varying with time. But unfortunately the published quantifications rely on many analytical formulations for the transfer function. This paper aims at unifying the heterogeneous information concerning radionuclide wash-off. A generic transfer function model was proposed. Published wash-off quantifications were gathered in a database and converted within this new framework. This extensive review covered various radionuclides, source terms, processes, time and space scales reported in the literature since 1960. It demonstrated the feasibility of a generic wash-off model. Intervals of variations for the introduced coefficients were summarised and the influence of wash-off conditions were discussed. In some cases, the influence of wash-off conditions should be clarified to propose narrower parameter intervals for practical assessments.

Influence of rainfall characteristics on elimination of aerosols of cesium, strontium, barium and tellurium deposited on grassland.

This work is aimed at quantifying foliar transfer of cesium, strontium, barium and tellurium under the influence of rainfall characteristics (intensity, frequency and time elapsed between contamination and first rainfall). Grassland boxes were contaminated by dry deposition of multi-element aerosols of (137)Cs, (85)Sr, (133)Ba and (123m)Te. They were grown in a greenhouse under controlled conditions. The treatments consisted of mowing and applying rainfalls (8 and 30 mmh(-1)) at different times after the contamination. At a leaf area index of 5.9+/-1.9, interception of the aerosols was similar for the 4 radionuclides (83.8+/-5.9%). Dew produced significant radionuclide accumulation in the base of the vegetation and transfer to the soil. For moderate intensity, an early (2 days after contamination) first rainfall was as efficient, in terms of leaf wash-off, as a longer rainfall occurring later (6 days after contamination). For early rainfalls, eliminated activities were comparable because the influence of rain intensity was compensated by rain duration. However, for late rainfalls, wash-off efficiency increased with rainfall intensity. Total transfer factors (TTF) were determined on whole grass immediately after 4 rainfalls and at harvest. After 4 medium intensity rainfalls, rain frequency did not influence total transfer factors (TTF) of strontium, barium and tellurium (about 0.2, 0.3 and 0.35 Bq kg(fresh weight)(-1) by Bq m(-2), respectively). Cesium TTF value was lower in the case of a weekly rain (0.1 against 0.2 Bq kg(fresh weight)(-1) by Bqm(-2)). TTF values were similar for twice-a-week rainfalls, whatever their intensity. They were higher for weekly rains of high intensity (between 0.3 and 0.75 Bq kg(fresh weight)(-1) by Bqm(-2) against 0.1-0.35 Bq kg(fresh weight)(-1) by Bq m(-2), depending on the radionuclides). TTF values attested that wash-off was more efficient when rainfalls lasted longer. Field loss on the top of the leaves was well described by an offset exponential model. The half-lives varied with rainfall characteristics from 4 days for cesium, strontium and barium to 20 days for tellurium. The offset value varied between 0% for tellurium (high intensity rainfalls) and 14% for cesium (medium intensity rainfalls).

Erosion of atmospherically deposited radionuclides as affected by soil disaggregation mechanisms.

The interactions of soil disaggregation with radionuclide erosion were studied under controlled conditions in the laboratory on samples from a loamy silty-sandy soil. The fate of 134Cs and 85Sr was monitored on soil aggregates and on small plots, with time resolution ranging from minutes to hours after contamination. Analytical experiments reproducing disaggregation mechanisms on aggregates showed that disaggregation controls both erosion and sorption. Compared to differential swelling, air explosion mobilized the most by producing finer particles and increasing five-fold sorption. For all the mechanisms studied, a significant part of the contamination was still unsorbed on the aggregates after an hour. Global experiments on contaminated sloping plots submitted to artificial rainfalls showed radionuclide erosion fluctuations and their origin. Wet radionuclide deposition increased short-term erosion by 50% compared to dry deposition. A developed soil crust when contaminated decreased radionuclide erosion by a factor 2 compared to other initial soil states. These erosion fluctuations were more significant for 134Cs than 85Sr, known to have better affinity to soil matrix. These findings confirm the role of disaggregation on radionuclide erosion. Our data support a conceptual model of radionuclide erosion at the small plot scale in two steps: (1) radionuclide non-equilibrium sorption on mobile particles, resulting from simultaneous sorption and disaggregation during wet deposition and (2) later radionuclide transport by runoff with suspended matter.

Characterization of an rRNA operon (rrnB) of Mycobacterium fortuitum and other mycobacterial species: implications for the classification of mycobacteria.

Mycobacteria are thought to have either one or two rRNA operons per genome. All mycobacteria investigated to date have an operon, designated rrnA, located downstream from the murA gene. We report that Mycobacteriun fortuitum has a second rrn operon, designated rrnB, which is located downstream from the tyrS gene; tyrS is very close to the 3' end of a gene (3-mag) coding for 3-methylpurine-DNA-glycosylase. The second rrn operon of Mycobacterium smegmatis was shown to have a similar organization, namely, 5' 3-mag-tyrS-rrnB 3'. The rrnB operon of M. fortuitum was found to have a single dedicated promoter. During exponential growth in a rich medium, the rrnB and rrnA operons were the major and minor contributors, respectively, to pre-rRNA synthesis. Genomic DNA was isolated from eight other fast-growing mycobacterial species. Samples were investigated by Southern blot analysis using probes for murA, tyrS, and 16S rRNA sequences. The results revealed that both rrnA and rrnB operons were present in each species. The results form the basis for a proposed new scheme for the classification of mycobacteria. The approach, which is phylogenetic in concept, is based on particular properties of the rrn operons of a cell, namely, the number per genome and a feature of 16S rRNA gene sequences.