Seismic Hazard Analysis for the UK: Sensitivity to Spatial Seismicity Modelling and Ground Motion Prediction Equations

The uniformity and consistency of probabilistic seismic hazard assessments for the U.K. is investigated by exploring the sensitivity of analysis results to spatial seismicity modelling and ground motion modelling. To evaluate the effects of different seismicity smoothing approaches, conventional source-zone-based methods as well as alternative seismicity smoothing methods, Kernel Smoothing (KS), and geometrical Epicentral Cell (EC) methods, are employed. Moreover, ground motion models that are applicable to lower magnitude ranges and those based on stochastic point-source simulations are considered for the assessment. The analysis results indicate that the choice of seismicity smoothing methods has significant influence on seismic hazard estimates; for the U.K., on average, the source-zone method based on the British Geological Survey model (Musson and Sargeant, 2007) leads to higher seismic hazard values than the KS and EC methods. The effects of using ground motion models that are applicable to smaller events can have moderate impact on seismic hazard estimates (up to about 20% reduction). The controlling factors of such effects are the characteristics of dominant earthquakes. Furthermore, dominant scenario events identified based on different smoothing approaches vary significantly, which may have important implications for advanced earthquake engineering applications. (e.g., response spectral. shape and record selection for nonlinear dynamic analysis).