A critical reappraisal of some problems in engineering seismology

Research output: ThesisDoctoral Thesis

140 Downloads (Pure)


The estimation of strong-motion characteristics is important for engineering design. Such an estimation, often in terms of peak ground acceleration and spectral ordinates, is usually based on the combination of physical models that describe the process with observed ground motions recorded during earthquakes.

A multitude of results have been derived over the past thirty years, based on different models and different quantities and qualities of input data. However, there is still little consensus on their validity and on the associated uncertainties which are important for the estimation of expected ground motions in design.

This thesis describes investigations of whether best use is being made of the strong-motion observations now becoming available, given the assumptions underlying the relationships to estimate ground motions, in the hope that this estimation can be improved. Potential sources of scatter, from each stage of the derivation of attenuation relations are highlighted, and many of these are critically
examined to assess their importance. This is achieved by: assessing the inherent uncertainty of the input strong-motion data including that arising from accelerogram processing, examining the importance of independent parameters and the effect of uncertainties and errors in these variables and by investigating the effect of the data distribution with respect to the independent variables.

This thesis presents updated relations for horizontal and vertical near-field strong-motion characteristics including peak ground acceleration and spectral acceleration, examines the assessment of permanent ground displacements in the near field due to faulting and estimates the effect of vertical ground motion on horizontal response. It concludes that any further improvement of the scaling of ground motions with seismological parameters and local site conditions depends primarily on the acquisition of more high-quality observational data.
Original languageEnglish
Awarding Institution
  • Imperial College of Science, Technology and Medicine
  • Ambraseys, N. N., Supervisor, External person
Award date30 Nov 2001
Place of PublicationLondon
Publication statusPublished - 30 Nov 2001


  • engineering seismology
  • seismic hazard
  • earthquake engineering
  • ground-motion prediction equations
  • single degree of freedom
  • permanent displacements
  • strong-motion data


Dive into the research topics of 'A critical reappraisal of some problems in engineering seismology'. Together they form a unique fingerprint.

Cite this