IRUS Total

Consistent Incorporation of topography effects into ground motion prediction equations

File Description SizeFormat 
Skiada-E-2019-PhD-ThesisThesis76.29 MBAdobe PDFView/Open
Title: Consistent Incorporation of topography effects into ground motion prediction equations
Authors: Skiada, Evangelia
Item Type: Thesis or dissertation
Abstract: The substantial effect of irregular surface features, broadly known as topographic irregularities, on earthquake ground-motion has been identified in a number of case histories, involving both empirical observations from past events and numerical studies. The ground response during earthquakes at areas close to convex topographic features such as ridges and slope crests is particularly altered in terms of amplitude, duration and frequency content. Several parametric studies have been performed on the subject, illustrating its importance and identifying the main parameters that control the ground surface response. These parameters are shown to be related to the different geometric configurations, ground stratigraphies and characteristics of the input motion. However, topographic amplification is not accounted for in a consistent manner in simplified design according to codes. For this reason, this thesis aims to develop a rigorous methodology to account for topographic effects within design codes or probabilistic seismic hazard analysis. A methodology to estimate topographic amplification factors for the horizontal component of ordinates of spectral acceleration for canyon topographies lying on a soil layer over rigid bedrock is developed with the aid of numerical modelling. It should be noted that although this study focuses on canyon geometries, the results for wide canyons are also applicable to single slope geometries. With the aid of finite element analysis an extensive dataset is firstly populated by computing the spectral accelerations along the canyon surface under a broad range of parametric scenarios. Topographic amplification factors are defined as the ratios of these surface motions with respect to the free-field response without topographic effects. This set of numerical topographic amplification factors are then regarded as an 'empirical dataset' from which design equations for the definition of topographic factors are derived. Topographic amplification is seen to be affected by the distance of the site of interest from the crest or the toe of the canyon, the slope angle, the slope height, the canyon width and the depth to rigid bedrock. The effect of different types of input motion on the resulting ground-surface response is also examined in this study. Two different types of wavelet motions, as well as 30 earthquake recordings, are used as input excitations in both the horizontal and the vertical direction. The use of wavelet input motions results in comparable topographic amplification variation across the canyon surface to that caused by the earthquake records. The effect of parameters such as non-uniform soil stiffness with depth and different values of the water compressibility on topographic amplification is also investigated for a single canyon geometry. This later set of analyses highlights that topographic amplification is affected by the presence of multiple soil layering because the wave scattering is affected both by the irregular topography and the impedance contrast between the successive soil layers. These analyses are performed to gain intuition about the robustness of the design equations, however, these analyses have not considered a sufficient range of realistic stratigraphies to incorporate their effect within the design equations at this point. The necessity for performing further numerical simulations to include the variation of these parameters in the design methodology is then highlighted. An approach for considering the horizontal component of the topographic amplification factors within the design at locations close to canyon geometries is finally proposed. Equations for the estimation of the topographic factors are derived for the crest and the toe areas of the canyon based on the dataset corresponding to the case of a homogeneous soil layer over rigid bedrock. Horizontal topographic amplification varies with the distance of the site of interest from the crest or the toe of the canyon slope and the spectral period of interest, unlike the constant value of topographic amplification recommended in Eurocode 8. Moreover, an estimation of the expected topographic amplification at the toe area of a topographic irregularity is provided by the proposed design equations, contrary to other literature design proposals which focus on the expected ground amplification at the crest area of a single slope. The amplitude of the topographic factors in this study is higher than the topographic amplification values reported in the literature. This is due to the rigid bedrock presence in the numerical simulations which effects the amplitudes of the proposed horizontal topographic factors. A framework to use these equations within the design code recommendations or when performing PSHA is also proposed. This framework serves as a first approximation of the expected horizontal topographic amplification at the canyon. Several other parameters such as variations in the canyon geometry, subsurface soil characteristics and soil nonlinearity should be also considered to further refine the proposed equations as mentioned above.
Content Version: Open Access
Issue Date: Oct-2018
Date Awarded: Feb-2019
URI: http://hdl.handle.net/10044/1/86492
DOI: https://doi.org/10.25560/86492
Copyright Statement: Creative Commons Attribution NonCommercial Licence
Supervisor: Kontoe, Stavroula
Stafford, Peter
Sponsor/Funder: Imperial College London
Department: Civil and Environmental Engineering
Publisher: Imperial College London
Qualification Level: Doctoral
Qualification Name: Doctor of Philosophy (PhD)
Appears in Collections:Civil and Environmental Engineering PhD theses

This item is licensed under a Creative Commons License Creative Commons