Directional dependence of site effects observed near a basin edge at Aegion,Greece

We study site effects using 520 weak motion earthquake records from a vertical array in Aegion, Greece. The array is inside a basin, has four stations in soil, and one in bedrock (178 m depth). The site is marked by high seismicity and complex surface geology. We first use the records to establish the downhole accelerometer orientations and their evolution with time. Then we estimate site effects using empirical spectral ratios with and without a reference site (standard and horizontal-to-vertical spectral ratio). We find significant site amplification which cannot be accounted for by 1D model predictions, along with a significant difference in the amplification level between the two horizontal components. These are indications of 2D effects, namely surface waves generated at the basin edge. The difference in amplification between the horizontal components is maximised when these are rotated with respect to the orientation of the basin edge. The strongest amplification takes place in the direction parallel to the basin edge (SH, or out-of-plane motion), and is up to 2 times higher than in the perpendicular direction (SV, or in-plane motion). This directional effect on the amplification is corroborated by numerical 2D modelling using incident SH and SV waves, with the former possibly generating strong Love waves. In the records, the directionality is clear for windows containing the largest amplitudes of the records (S waves and strong surface waves), while it tends to vanish for coda-wave windows. This directionality is also observed when using response spectral ratios rather than Fourier ratios. We compute soil-to-rock amplification factors for peak ground acceleration (PGA) and find it is significantly higher than what is predicted by current design codes. We attribute this difference to the basin edge amplification, linear soil behaviour, and to the inability of simple scalar values like PGA to describe complex amplification effects. Finally, we analyse the earthquake records at a surface station near the slope crest and do not observe significant topographic amplification.     

New code site classification, amplification factors and normalized response spectra based on a worldwide ground-motion database

A global dataset of more than 3,000 ground motion records from 536 sites from Greece, Italy, Turkey, USA and Japan is compiled and used to propose code-oriented elastic acceleration response spectra and soil amplification factors for a new site classification system, which, besides the classical geotechnical parameters $N_{SPT}, S_{u}$ and PI, uses also the fundamental period of the site, the thickness of soil deposits and the average shear wave velocity to the seismic bedrock, instead of $V_{s,30}$ . We propose a new classification system with the associated amplification factors and normalized response spectra for two seismicity levels, i.e. $M_{s}\le 5.5$ and $M_{s}>5.5$ . Uncertainties in the estimation of soil amplification factors are captured using a logic-tree approach, which allows the efficient use of alternative models and methods. The aim of this work is to improve the present EC8 soil classification. The effectiveness of the proposed classification system is compared to that of EC8 classification system using an error term, which represents the average dispersion of data within all categories of a given classification scheme. Error terms for the new classification system are lower than the error terms for EC8 classification system at all periods.     

Site Effects and Design Provisions: The Case of Euroseistest

—?Modern seismic codes usually include provisions for site effects by considering different coefficients chosen on the basis of soil properties at the surface and an estimate of the depth of bedrock. However, complex local geology may generate site amplification on soft soils significantly larger than what would be expected if we assume that the subsoil consists of plane soil layers overlaying a homogeneous half-space. This paper takes advantage of the large number of previous studies of site effects done at Euroseistest (northern Greece). Those studies have supplied a very detailed knowledge of the geometry and properties of the materials filling this shallow valley. In this paper we discuss the differences between site effects evaluated at the surface using simple 1-D computations and those evaluated using a very detailed 2-D model of the subsoil structure. The 2-D model produces an additional amplification in response spectra that cannot be accounted for without reference to the lateral heterogeneity of the valley structure. Our numerical results are extensively compared with observations, which show that the additional amplification computed from the 2-D model is real and affects by a significant factor response spectra, and thus suggests that some kind of aggravation factor due to the complexity of local geology is worthy of consideration in microzonation studies and seismic codes.     

Earthquake risk assessment of lifelines

This paper presents the RISK-UE methodology for the seismic risk assessment of utility systems (potable water, waste-water, gas system, telecommunication, electric power) and transportation infrastructures (port, airport, road and railway system). The proposed methodology provides a uniform basis for the reduction of the consequences of lifeline damages in urban areas and an efficient mitigation strategy and prioritization policies for pre-earthquake and post earthquake actions. A detailed inventory for every element at risk together with a reliable seismic hazard assessment, appropriate selection of fragility models, estimation of the “global value” and economical impact of lifeline damages and losses are the main steps of the proposed methodology. The consideration of European distinctive features of lifelines and utility systems in the construction of seismic scenarios and the proposition of fragility curves are among the basic aims of the proposed methodology. Different modules of the methodology were applied in seven pilot cities (Thessaloniki, Catania, Nice, Bucharest, Sofia, Barcelona, Bitola). We present herein few representative examples (case studies) in order to illustrate the methodology and to prove its efficiency.     

Seismic analysis of coupled soil-pile-structure systems leading to the definition of a pseudo-natural SSI frequency

The elastodynamic response of coupled soil-pile-structure systems to seismic loading is studied using rigorous three-dimentional (3D) finite element models. The system under investigation comprises of a single pile supporting a single degree of freedom (SDOF) structure founded on a homogeneous viscoelastic soil layer over rigid rock. Parametric analyses are carried out in the frequency domain, focusing on the dynamic characteristics of the structure, as affected by typical foundation properties such as pile slenderness and soil-pile relative stiffness. Numerical results demonstrate the strong influence on effective natural SSI period of the foundation properties and the crucial importance of cross swaying-rocking stiffness of the pile. Furthermore, the notion of a pseudo-natural SSI frequency is introduced, as the frequency where pile-head motion is minimized with respect to free field surface motion. Dynamic pile bending is examined and the relative contributions of kinematic and inertial interaction, as affected by the frequency content of input motion, are elucidated.     

Numerical simulation of dynamic soil–structure interaction in shaking table testing

This paper provides an insight into the numerical simulation of soil–structure interaction (SSI) phenomena studied in a shaking table facility. The shaking table test is purposely designed to confirm the ability of the numerical substructure technique to simulate the SSI phenomenon. A model foundation–structure system with strong SSI potential is embedded in a dry bed of sand deposited within a purpose designed shaking-table soil container. The experimental system is subjected to a strong ground motion. The numerical simulation of the complete soil–foundation–structure system is conducted in the linear viscoelastic domain using the substructure approach. The matching of the experimental and numerical responses in both frequency and in time domain is satisfying. Many important aspects of SSI that are apparent in the experiment are captured by the numerical simulation. Furthermore, the numerical modelling is shown to be adequate for practical engineering design purposes.     

Understanding single-station ground motion variability and uncertainty (sigma): lessons learnt from EUROSEISTEST

Accelerometric data from the well-studied valley EUROSEISTEST are used to investigate ground motion uncertainty and variability. We define a simple local ground motion prediction equation (GMPE) and investigate changes in standard deviation (σ) and its components, the between-event variability (τ) and within-event variability (φ). Improving seismological metadata significantly reduces τ (30–50%), which in turn reduces the total σ. Improving site information reduces the systematic site-to-site variability, φ _S2S (20–30%), in turn reducing φ, and ultimately, σ. Our values of standard deviations are lower than global values from literature, and closer to path-specific than site-specific values. However, our data have insufficient azimuthal coverage for single-path analysis. Certain stations have higher ground-motion variability, possibly due to topography, basin edge or downgoing wave effects. Sensitivity checks show that 3 recordings per event is a sufficient data selection criterion, however, one of the dataset’s advantages is the large number of recordings per station (9–90) that yields good site term estimates. We examine uncertainty components binning our data with magnitude from 0.01 to 2 s; at smaller magnitudes, τ decreases and φ _SS increases, possibly due to κ and source-site trade-offs Finally, we investigate the alternative approach of computing φ _SS using existing GMPEs instead of creating an ad hoc local GMPE. This is important where data are insufficient to create one, or when site-specific PSHA is performed. We show that global GMPEs may still capture φ _SS , provided that: (1) the magnitude scaling errors are accommodated by the event terms; (2) there are no distance scaling errors (use of a regionally applicable model). Site terms (φ _S2S ) computed by different global GMPEs (using different site-proxies) vary significantly, especially for hard-rock sites. This indicates that GMPEs may be poorly constrained where they are sometimes most needed, i.e., for hard rock.     

Dynamic stiffness and kinematic response of single piles in inhomogeneous soil

The effect of soil inhomogeneity on dynamic stiffness and kinematic response of single flexural elastic piles to vertically-propagating seismic SH waves is explored. A generalized parabolic function is employed to describe the variable shear wave propagation velocity in the inhomogeneous stratum. A layered soil with piece-wise homogeneous properties is introduced to approximate the continuous inhomogeneity in the realm of a Beam-on-Dynamic-Winkler-Foundation model. The problem is treated numerically by means of a layer transfer-matrix (Haskell–Thompson) formulation, and validated using available theoretical solutions and finite-element analyses. The role of salient model parameters such as pile-head fixity conditions, pile-to-soil stiffness ratio, surface-to-base shear wave velocity ratio and rate of inhomogeneity is elucidated. A new normalization scheme for inertial and kinematic response of such systems is presented based on an average Winkler wavenumber. With reference to long piles in moderately inhomogeneous soils, results indicate that: (a) kinematic pile response is essentially governed by a single dimensionless frequency parameter accounting for pile-to-soil stiffness ratio, pile slenderness and soil inhomogeneity and (b) definition of a characteristic pile wavelength allows an approximate estimation of pile elastodynamic response for preliminary design or analysis. Issues related to domain discretization and Winkler moduli are discussed.