Sensitivity of surface hazard to different factors and site response analysis approaches: a case study for a soft rock site

Near-surface effects for a soft rock site (average shear-wave velocity of the top 30 m, Vs 30 ≈ 800 m/s) for a proposed nuclear power station in the UK are integrated into the “bedrock” results of a probabilistic seismic hazard analysis (PSHA) by application of US Nuclear Regulatory Commission (USNRC) Approach 3 and employing a partially non-ergodic PSHA. The sensitivity of the surface hazard to the site response analysis method is assessed, employing both random vibration theory (RVT) and time series (TS) approaches. The effects of different assumptions relating to strong-motion duration, selection of target frequency in the surface uniform hazard spectrum (UHS) and the incorporation of the variability of site properties through Monte Carlo simulations are also quantified. The results show that for the examined stiff site, with response concentrated at high frequencies, the use of RVT site response analysis does not introduce a systematic bias in the low frequency ground motion predictions and the duration used in the definition of the input ground motions is demonstrated to have a secondary effect on the site response. The incorporation of the variability of site properties and the selection of the target frequencies in the convolution are shown to be important in the derivation of the uniform hazard spectrum.