Seismic Hazard Assessment for Nuclear Plants: Current Practices in France, Uncertainties and Key Scientific Issues

The objectives of this contribution are (i) to provide a description of the current practices in France to assess the seismic hazard for nuclear safety and (ii) to highlight scientific key questions and issues. Historically, the French nuclear regulation is anchored on a deterministic approach. Accounting for seismic risk is currently guided in France by the Fundamental Safety Rule (RFS2001-01) for assessing the seismic hazard, and by the ASN/2/01 Guide providing design rules of nuclear civil engineering structures. These references respectively updated by the Nuclear Safety Authority in 2001 and 2006 are used to design new plants but also for re-assessing the seismic loadings of existing plants. Since then, the 2011 Tohoku earthquake that triggered a huge tsunami caused the severe accident at the Fukushima Daïchi nuclear plant. The analysis of the observations demonstrated that the seismic and tsunami hazards were underestimated for this region. Consequently all around the world nuclear operators were asked by their authority to perform "stress tests" to estimate their plant capacity sustaining extreme seismic loadings. French operators then defined a ''Hard Core'' of materials and organizational arrangements to ensure the control of crucial safety functions in extreme situations. This hard core is associated to a specific seismic level, obviously higher than the one considered for designing or re-assessing the plant safety: originally proposed in the context of the French deterministic regulatory context, these levels have been finally justified conducting state of art probabilistic seismic hazard assessments, as recommended by a European peer review. Using French representative cases we emphasize some questions or issues more specific of low to moderate seismic areas (e.g. data and meta-data highly incomplete and uncertain, lack of site specific strong motions, use of oversimplified assumptions or methods, treatment of uncertainties …): regarding the topic of the workshop, the way in which the seismic source is described and the strong motion assessed will be particularly discussed. The impact of using either deterministic or probabilistic approach will be also illustrated. (author)