Estimation of strong ground motion: Aleatory variability and epistemic uncertainties

• Main Author: Douglas, John
• Other Authors: Bureau de Recherches Géologiques et Minières (BRGM), Université Joseph-Fourier - Grenoble I, M. Pierre-Yves Bard
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: HAL CCSD 2010
• Subjects: incertitudes epistemiques; mouvement fort du sol; aléa sismique; équations de prédiction de mouvements du sol; variabilité aléatoire; [SDU.STU.GP]Sciences of the Universe [physics]/Earth Sciences/Geophysics [physics.geo-ph]; Iceland
• Online Access: https://theses.hal.science/tel-00545546; https://theses.hal.science/tel-00545546/document; https://theses.hal.science/tel-00545546/file/douglas2010hdr.pdf

This report summarises the research I have undertaken since finishing my Ph.D. thesis in autumn 2001. The work reported was undertaken as part of various projects and in collaboration with many researchers at: Imperial College London, UK (2001–2004); Bureau de Recherches G´eologiques et Mini`eres (BRGM), France (2004–now); and the Earthquake Engineering Research Centre, University of Iceland (2009–2010). In addition, this report lists the teaching, supervision and consultancy work that I have been involved with since 2001. My research has mainly been on the prediction of earthquake ground motions for engineering purposes, e.g. design and retrofit projects and seismic hazard and risk assessments. Most of my studies relate to empirical shaking estimation through ground-motion prediction equations (GMPEs, also called ground-motion models and attenuation relations). It has focussed on: improving estimates of the median ground motion and associated variability; quantifying, understanding and potentially reducing variability; assessing and modelling the regional dependence of strong ground motions; and combining simulations and empirical estimates. My research shows that, although significant progress has been made in the past couple of decades in improving the accuracy of estimates of the median ground motion for a given scenario, epistemic uncertainty remains high and this must be accounted for in all seismic hazard and risk evaluations. In addition, all methods of reducing the standard deviations of GMPEs proposed so far, although they looked promising, have proved to be largely ineffective. My studies have demonstrated yet again that earthquake ground motions can vary significantly between earthquakes and sites and this also must be taken into account when conducting seismic hazard and risk assessments. Any attempt to reduce epistemic uncertainties further and derive GMPEs with lower standard deviations are reliant on increasing the density of strong-motion networks but, perhaps more importantly, improving the ...