Uncertainties of the transfer of radionuclides in terrestrial ecosystems

The Nord-Cotentin radioecological study was conducted in different stages. First, an assessment tool was developed and implemented to estimate the number of leukaemia radio-induced by the operation of nuclear facilities of the La Hague site (Nord-Cotentin region, France). Then, an uncertainty analysis was initiated to verify if the confidence intervals around the best estimate previously calculated allows to confirm the low impact of the nuclear facilities. The modelling of the terrestrial ecosystems was especially studied because of the paucity of results of measurements. That is also why the methodology built to reconstruct the statistical laws for each parameter of the model mixes the bibliographic knowledge and exploitation of a national database of environmental measurements. The application of this methodology for up to forty parameters shows that the possibility of fitting an accurate distribution such as a normal or a lognormal law is strongly linked with the availability and the number of measured data.