Use of machine learning and deep learning to predict particulate 137Cs concentrations in a nuclearized river.

Cesium-137, discharged by nuclear installations under normal operations and deposited in watersheds following atmospheric testing and accidents (i.e. Chernobyl, Fukushima …), has been studied for decades. Thus, modelling of 137Cs concentration in rivers have been developed based on geochemical approaches and equilibrium assumptions (solid/liquid ratio) as this radionuclide has moved into rivers and oceans due to soil erosion. Recently a new approach is possible to model these concentrations with the popularization of data-driven models based on data acquired in the environment by monitoring networks. In this study, the concentrations of particulate cesium-137 measured near the mouth of the Rhône River (France), a highly nuclearized river, are simulated using two data-driven models, a Hierarchical Attention-Based Recurrent Highway Networks (HRHN) and a Random Forest Regressor (RF). The data-driven predictions were done using only hydrological data (water discharge and suspended solid fluxes) and industrial input of 137Cs. Although the data-driven models provided a better prediction than a recent empirical model, the best prediction (R2 = 0.71) was obtained with HRHN, a model that considers the temporal aspect of the monitoring data. The most important predictors were the hydrological data at the monitoring station and of the tributary that generate the most sediment flux (Durance River). In fact, the concentration of 137Cs in the perimeter of this study was more related to hydrology than to nuclear release, as there were few events with high 137Cs concentrations (concomitant nuclear release and low water discharge). However, the HRHN approach, which is more complex to implement than RF, can predict the concentrations of such events correctly despite their low representation of these events. The results of this study demonstrate the usefulness of data-driven models to assist monitoring programs by filling in gaps or helping to understand observed concentrations.

Systematic influences of gamma-ray spectrometry data near the decision threshold for radioactivity measurements in the environment.

Several methods for reporting outcomes of gamma-ray spectrometric measurements of environmental samples for dose calculations are presented and discussed. The measurement outcomes can be reported as primary measurement results, primary measurement results modified according to the quantification limit, best estimates obtained by the Bayesian posterior (ISO 11929), best estimates obtained by the probability density distribution resembling shifting, and the procedure recommended by the European Commission (EC). The annual dose is calculated from the arithmetic average using any of these five procedures. It was shown that the primary measurement results modified according to the quantification limit could lead to an underestimation of the annual dose. On the other hand the best estimates lead to an overestimation of the annual dose. The annual doses calculated from the measurement outcomes obtained according to the EC's recommended procedure, which does not cope with the uncertainties, fluctuate between an under- and overestimation, depending on the frequency of the measurement results that are larger than the limit of detection. In the extreme case, when no measurement results above the detection limit occur, the average over primary measurement results modified according to the quantification limit underestimates the average over primary measurement results for about 80%. The average over best estimates calculated according the procedure resembling shifting overestimates the average over primary measurement results for 35%, the average obtained by the Bayesian posterior for 85% and the treatment according to the EC recommendation for 89%.