Faire avancer la sûreté nucléaire

La Recherchev2


Advantages and limitations of dose assessments to the members of the public based on environmental monitoring



Email :

Mot de passe :

Despres, A.; Mansoux, H.
IRPA-10: 10. international congress of the International Radiation Protection Association Hiroshima (Japan) 14-19 May 2000, T-19(2)-3, P-4b-235;4 refs., 2 tabs.


The assessment of doses received by members of the public living in the vicinity of a nuclear facility is usually based on the modelization of the transfer of radionuclides discharged by the plant. Due to the uncertainties in the transfer parameters and to the approximations inherent to the modelization, this evaluation was usually applied to the assessment of an upper bound rather than to the estimation of a realistic figure. International recommendations now request for realistic dose assessments. Related improvements call for further developments as regards on modeling, and knowledge of transfer parameters. They also lead to investigate other methods. In this purpose, an assessment of doses received by the critical group living in the vicinity of the Marcoule reprocessing plant has been performed, using the results of environmental monitoring. This paper describes the methodological options of this assessment and point out the difficulties encountered. Finally, doses estimated according to this approach are compared with those obtained by modeling. Dose predictions derived from actual releases were in the range of 5-20 muSv.a-1, while those derived from measurements were on the order of 50 muSv.a-1. More important, relevant radionuclides and pathways were not always the same. This comparison shows that only a fraction of the radiological impact of the plant operation can be deduced from environmental monitoring, the main reasons being: -the difficulty to know the background level (natural, other sources) for non specific radionuclides concentrations in the environment and to define the origin of measured concentrations. -the lack of data: all the radionuclides are not measured in all the environmental compartments of the environment (air, soils, water, agricultural productions,...). -the monitoring design: routine measurements are devoted to the detection of abnormalities and to the follow up of trends rather than to dosimetric evaluations. As a consequence, results are often expressed in a form that does not allow an easy dosimetric exploitation (e.g. global activities,...). -the detection level: radionuclide concentrations are often below the threshold. This paper concludes that these two approaches (dose calculations based on radionuclides transfer modeling and on environmental survey) are complementary when performing more realistic dose assessments. lt suggests further developments, both on the methodology and in the field of metrology.