Proceedings of the Specialist Meeting on the Seismic Probabilistic Safety Assessment of Nuclear Facilities

The main objectives of the Meeting were to review recent advances in the methodology of Seismic Probabilistic Safety Assessment (SPSA), to discuss practical applications, to review the current state of the art, and to identify methodological issues where further research would be beneficial in enhancing the usefulness of the methodology. Applications of the Seismic Margin Assessment methodology (SMA), a methodology related to SPSA, were also discussed. One specific objective was to compare the situation today with the situation at the time of the 1999 Tokyo workshop, and to develop a set of findings and recommendations that would update those from that earlier workshop. There was a consensus at the Specialists Meeting that SPSA is now in widespread use throughout the nuclear-power industry worldwide, by the operating nuclear power plants (NPPs) themselves, by the various national regulatory agencies, and by the designers of new NPPs. It was also widely agreed that it can systematically accomplish several very important objectives; specifically, it can contribute: - To understanding the seismic risk arising from NPPs. - To understanding the safety significance of seismic design shortfalls. - To prioritizing seismic safety improvements. - To evaluating and improving seismic regulations. - To modifying the seismic regulatory/licensing basis of an individual NPP. Compared to the situation in 1999, when the first Workshop was held in Tokyo, there have been significant expansions in the use of SPSA in many different areas. Some countries provided detailed information on their regulatory framework for using seismic PSA. Many other countries also provided some information in their papers as background for conducting SPSA. During the Meeting, a small number of important weaknesses in SPSA methodology were identified. None of these are new, all having been widely recognized for many years. However, for some of the weaknesses, extensive discussions during the Meeting provided insights into how the community of SPSA practitioners could improve matters. Although a PSHA report from Japan was a very useful contribution as an example from a high seismicity area, the main discussions at the Meeting concentrated on PSHAs conducted in areas with low-to-moderate seismicity. Problems in PSHA studies for regions with low to moderate seismicity may arise from the fact that, due to the small number of strong-motion earthquakes in such regions, attenuation relationships must start with those taken from other regions with available strong motions. This could lead to inconsistencies or to large uncertainties, depending on experts' choices. Consequently, the PSHA hazard distributions in such low-to-moderate seismicity areas may have this large uncertainty, so that the mean values become comparable in size to PSHA mean hazard results obtained from areas with relatively high seismicity. Much discussion occurred on this topic. The meeting observed that one vast area of continuing uncertainty is in quantifying the response of the operating crew and emergency organizations after earthquakes. Partly, the problem is generic with all human reliability analysis: uncertainties and the lack of data. However, there are specific characteristics of earthquakes that make post-earthquake actions more difficult to analyze and quantify. Among these characteristics are physical and mental consequences of a seismic shock. Starting with the very first SPSAs in the early 1980's, analysts have struggled with the problem of how to quantify the correlations in the failures of similar equipment or similar structures due to the earthquake. Discussions during the Meeting did not provide any novel insights about how to improve the treatment of correlations. This area therefore remains one where uncertainties in the numerical bottom-line results of SPSAs will remain large. The Meeting concluded that there is an urgent need for a comparison of PSHA results from countries with high, medium and low seismicity because of the differences that have been identified. The problem i s especially important for countries with low and medium seismicity, because little actual earthquake data exist for them. The Meeting also gave a clear recommendation that because of the rapid progress in using seismic PSA, the state-of-the-art report issued in 1997 by the NEA CSNI Working Group Risk should be updated. Also, developing a short document about the latest situation for generalists, for example a technical opinion paper, should be considered