Skip to main content

2022 | Buch

Proceedings of the 4th International Conference on Performance Based Design in Earthquake Geotechnical Engineering (Beijing 2022)


Über dieses Buch

The 4th International Conference on Performance-based Design in Earthquake Geotechnical Engineering (PBD-IV) is held in Beijing, China. The PBD-IV Conference is organized under the auspices of the International Society of Soil Mechanics and Geotechnical Engineering - Technical Committee TC203 on Earthquake Geotechnical Engineering and Associated Problems (ISSMGE-TC203). The PBD-I, PBD-II, and PBD-III events in Japan (2009), Italy (2012), and Canada (2017) respectively, were highly successful events for the international earthquake geotechnical engineering community. The PBD events have been excellent companions to the International Conference on Earthquake Geotechnical Engineering (ICEGE) series that TC203 has held in Japan (1995), Portugal (1999), USA (2004), Greece (2007), Chile (2011), New Zealand (2015), and Italy (2019). The goal of PBD-IV is to provide an open forum for delegates to interact with their international colleagues and advance performance-based design research and practices for earthquake geotechnical engineering.



Keynote Lectures

Performance-Based Seismic Assessment of Slope Systems

Seismic slope displacement procedures are useful in the evaluation of earth embankments and natural slopes. The calculated seismic slope displacement provides an index of performance. Newmark-based sliding block models are typically employed. The manner in which the key components of the analysis are addressed largely determines the reliability of a particular procedure. The primary source of uncertainty in assessing the seismic performance of an earth slope is the input ground motion. Hence, sliding block procedures have advanced over the last two decades through the use of larger sets of ground motion records. Recent updates of the procedures developed by the authors are highlighted. The nonlinear fully coupled stick-slip sliding block model calculates reasonable seismic slope displacements. Displacements depend primarily on the earth structure’s yield coefficient and the earthquake ground motion’s spectral acceleration at the effective fundamental period of the sliding mass. Through their use, the sensitivity of the seismic slope displacements and their uncertainty to key input parameters can be investigated. These procedures can be implemented within a performance-based design framework to estimate the seismic slope displacement hazard, which is a more rational approach.

Jonathan D. Bray, Jorge Macedo
Recent Advances in Helical Piles for Dynamic and Seismic Applications

Helical piles have become popular foundation option owing to their many advantages related to ease of installation and large load carrying capacity. They are typically manufactured of straight steel shafts fitted with one or more helices and are installed using mechanical torque. They can sustain static and dynamic loading and are increasingly used in applications that induce complex loading conditions on them. The behavior and design of single vertical helical piles subjected to static loading is well investigated. However, a few studies investigated the dynamic or seismic behavior of single helical piles and their group behavior. This paper presents recent advances in evaluating the axial and lateral capacity and performance of helical piles and their response to dynamic and seismic loads.

M. Hesham El Naggar
Seismic Response of Offshore Wind Turbine Supported by Monopile and Caisson Foundations in Undrained Clay

The seismic performance of offshore wind turbine (OWT) becomes increasingly critical as more wind farms are built/planned in sites subjected to active seismic events. Time-domain, continuum-based methods can provide comprehensive analyses of OWT during seismic shaking yet its application in engineering practice can be restricted by the complexity of soil constitutive models. This work presents a constitutive model for undrained clay that is simple yet replicates the essential soil behavior, including mechanical anisotropy, cyclic degradation of stiffness and strength, and path-dependent stiffness nonlinearity at small strains. The capacity of the soil model is assessed at various levels, by simulating soil element tests, the response of cyclically-loaded pile in centrifuge test, and the response of caisson-supported OWT in seismic centrifuge test. These assessments show that the seismic response of OWT can be reasonably represented by the proposed method. Based on the soil model, dynamic finite element analyses are performed to explore the seismic response of OWT supported by different foundation types: monopile and caisson. This investigation highlights: (1) the dynamic response of OWT foundation is governed by the kinematic-inertia interactions between soil, foundation, and superstructures; (2) the resonance of OWT structure system (including higher modes) can be a critical mechanism behind the developments of excessive foundation movements.

Maosong Huang, He Cui, Zhenhao Shi, Lei Liu
Development of the Earthquake Geotechnical Engineering (EGE) in ISSMGE

After the original studies at the University of California led by H.B. Seed and I.M. Idriss, the study of EGE was introduced to ISSMFE in 1985 as one of the activities of a technical committee (TC) of ISSMFE. Since then, it has grown to an area of importance through the activities of TC-4 (1985 – 2009) and TC-203 (2009-present).

Kenji Ishihara
Transient Loading Effects on Pore Pressure Generation and the Response of Liquefiable Soils

Like ground motions, the loading applied to potentially liquefiable soils in earthquakes is complex and unique. Conventional procedures for characterizing liquefaction hazards represent seismic loading in simplified manners that do not account for characteristics that can influence pore pressure generation and potential ground deformation. This paper presents the results of cyclic direct simple shear tests performed with transient, irregular loading derived from recorded earthquake ground motions. Data from these tests show the influence of the order in which individual pulses of shear stress occur on generated pore pressure and provide insight into limitations of evolutionary intensity measures based on time integration to provide efficient predictions of liquefaction. The form of a new intensity measure for triggering of liquefaction is proposed and calibrated on the basis of available data. The response of cyclic simple shear specimens subjected to transient loading superimposed upon static shear stresses is also illustrated and comments on existing procedures for estimation of the adjustment factor, $$K_{\upalpha }$$ K α , are provided.

Steven L. Kramer, Samuel S. Sideras
Performance-Based Design for Earthquake-Induced Liquefaction: Application to Offshore Energy Structures

Liquefaction has been a major challenge to design of structures founded on loose silt and sand in moderate and specially highly seismic regions. While assessment of liquefaction susceptibility and potential have been largely based on empirical methods, the design of structures on liquefiable soil requires reliable numerical tools and clear performance criteria. In this paper, solutions are provided based on the well-established SANISAND model and its more recent extension, SANISAND-MSu, implemented in the open-source finite element platform OpenSEEs. Applications are presented for structures commonly encountered in offshore energy sector such as conventional subsea facilities on mudmats and offshore wind turbines founded on large-diameter monopiles. The impact of pore-water pressure, and ultimately liquefaction, on the offshore structures is assessed by performing both quasi-static cyclic loading and earthquake shaking. The general behavior of these offshore structures during liquefaction are presented from a numerical modelling perspective. The simulation results indicate that the response of these structures is considerably affected by structural features and environmental loading conditions. The results presented in this work motivates the use of SANISAND-MSu model in enhanced 3D finite element modelling in offshore structural dynamic analyses.

Haoyuan Liu, Amir M. Kaynia
Factors Affecting Liquefaction Resistance and Assessment by Pore Pressure Model

Assessment of the liquefaction resistance of clean sand still involves considerable uncertainties, which are a current research topic in the field of soil liquefaction. Factors to be considered include shaking history, overconsolidation, degree of saturation and partial drainage. The effects of these factors on liquefaction resistance have been studied in the laboratory and empirical relationships are derived. This paper describes the development of pore pressure generation model similar to that of Martin et al. [18] but based on stress-controlled triaxial tests. The effects of various factors on the pore pressure generation and liquefaction resistance of clean sand are explained using the unique index of volumetric strain. The model is verified through comparisons with the results of laboratory tests. It is confirmed that the plastic volumetric strain accumulated in sand either by drained or undrained loading dominates the increase in liquefaction resistance of pre-sheared, overconsolidated and unsaturated sand. The model provides a better understanding of the physical processes leading to liquefaction of saturated and unsaturated sand with and without stress histories.

Mitsu Okamura
How Important is Site Conditions Detailing and Vulnerability Modeling in Seismic Hazard and Risk Assessment at Urban Scale?

This work investigates the effect of the level of detailing of site conditions as well as the selection of the fragility and vulnerability models on large scale seismic risk assessment. For this we consider the application of selected components of the recent European Seismic Hazard (ESHM20) and Risk (ESRM20) Models and we focus on Thessaloniki, Greece, a city which is very well documented in terms of local site conditions and exposure. Seismic risk results are compared in terms of expected damages and economic losses for a seismic hazard with a 475-year return period. The results indicate that the level of site conditions modelling does not significantly affect the estimated aggregate damages and economic losses at city scale, however significant discrepancies may occur at local scale. On the other hand, the selection of the vulnerability model for the building stock may considerably affect the intensity and the spatial distribution of damages, resulting in a considerable differentiation in the economic losses estimate.

Kyriazis Pitilakis, Evi Riga, Stefania Apostolaki
The Influence of Soil-Foundation-Structure Interaction on the Seismic Performance of Masonry Buildings

Most of the damage and of the casualties induced even by the most recent strong-motion earthquakes which stroke Central and Southern Italy can be attributed to the extreme seismic vulnerability of the ordinary residential buildings. Both in small villages and in mid-size towns, these latter are mainly constituted by two- to four-story masonry structures built without anti-seismic criteria, with direct foundations corresponding to an in-depth extension of the loadbearing walls or to an underground level. For such structures, especially when founded on soft soils, soil-foundation-structure interaction can significantly affect the seismic performance; on the other hand, its influence must be handled with methods which should be as simple and straightforward as possible, in order to be cost-effective and accessible by practitioners. The contribution wishes to summarize the studies carried out in the last years at University of Napoli Federico II, based on parametric numerical analyses on complete soil-foundation-structure models reproducing the most recurrent building configurations combined with different subsoil conditions. The analyses provided calibration criteria for: i) predicting the elongation of the fundamental period of the structure, ii) defining and optimizing fragility functions for different damage mechanisms accounting for soil-foundation-structure interaction. The effectiveness of these simplified tools was validated against well-documented case studies at the scale of single instrumented buildings or of extended areas, with building properties and subsoil conditions comparable to those adopted in the parametric analyses.

Francesco Silvestri, Filomena de Silva, Fulvio Parisi, Annachiara Piro
Coseismic and Post-seismic Slope Instability Along Existing Faults

The world in the 21st Century faces increased hazards of gigantic earthquakes and heavy rains as compared with the previous century. Under this circumstance, coseismic landslide is one of the important topics in geotechnical earthquake engineering and is promoted by the increased precipitation both before and after the gigantic earthquakes. This paper sheds light on this topic, paying attention not only to the effects of strong shaking but also to the important role played by faults, which are either active or inactive. The fault action consists of the rock fissures along the fault and possible water ejection from the fault plane. The rock fissures made by the fault dislocation cause the long-term instability of mountains slopes for years or for centuries after strong shaking.

Ikuo Towhata
Static Liquefaction in the Context of Steady State/Critical State and Its Application in the Stability of Tailings Dams

Mining operations produce large quantities of tailings that must be conducted and stored safely in a technical framework that reconciles economic restrictions and environmental sustainability. Unfortunately, the history of tailings deposits is marked by episodes of catastrophic failures that have caused many victims and have cost enormous material losses. This is confirmed by the recent catastrophic collapses of tailings dams that occurred in Canada (Mount Polley), Australia (Cadia) and Brazil (Samarco and Brumadinio), where the tailings that flowed downstream severely affected the environment and, in the Brazilian case, caused many casualties. Due to this alarming empirical evidence left by the mining industry worldwide, there is international concern about the stability of tailings dams and a demand to build these deposits safely from every point of view. In this scenario, the correct evaluation of the actual shear strength of the deposited tailings is crucial. The critical state soil mechanic (CSSM) and/or the steady state of deformation provide the conceptual framework for evaluating the ultimate strength mobilized by particulate materials. In the context of a performance-based design, these states are discussed, and based on them, procedures and a flow index, If, are presented to analyze the physical stability of tailings.

Ramon Verdugo
Seismic Behaviour of Retaining Structures: From Fundamentals to Performance-Based Design

This lecture summarises research carried out by the Authors on the seismic behaviour of displacing or yielding retaining structures, i.e., structures that can undergo permanent displacements during strong earthquakes without failing. For these systems, energy dissipation on shaking, leading to reduced inertia forces, can be achieved by allowing the activation of ductile plastic mechanisms. These must be correctly identified to guarantee the desired strength hierarchy, and depend on the specific retaining structure under examination. It will be shown that the critical acceleration, or the smallest value of acceleration corresponding to the activation of the critical plastic mechanism, is a key ingredient for performance based design of yielding retaining structures. In fact, the critical acceleration controls both the maximum internal forces on the structural elements and the magnitude and trend of post-seismic permanent settlements and rotations, required for quantitative serviceability and post-earthquake operability assessment of infrastructure. Based on a clear understanding of the physical mechanisms governing the dynamic behaviour of these systems, pseudostatic limit equilibrium solutions and simplified dynamic methods can be developed for their seismic design. Theoretical results are validated against data from reduced scale centrifuge models and the results of pseudo-static and fully dynamic numerical analyses. Finally, all the results presented in the paper, including experimental, numerical and theoretical findings, are used to provide suggestions for the performance-based design of retaining structures.

Giulia M. B. Viggiani, Riccardo Conti
Study and Practice on Performance-Based Seismic Design of Loess Engineering in China

Loess is a kind of special soil with porous structure and weak cohesion, which widely deposits in China with an area of 640,000 km2. Especially, it is continuously distributed in the Loess Plateau of China with an area of 440,000 km2 and a thickness ranging from tens meters to more than 500 m, where is a region with the biggest thickness and the most complicated topography of loess deposit in the globe. On the other hand, the Loess Plateau is a strong earthquake-prone region, where 120 earthquakes with Ms ≥ 6.0 and 7 events of Ms ≥ 8.0 occurred in history. These earthquakes killed more than 1.4 million people in the region. The investigation shown that so large casualties should be attributed to a large scale of landslides, liquefaction, subsidence and amplification of ground motion. In this paper, the characteristics of engineering geology and seismic hazards with different probabilities of exceedance in the Loess Plateau of China is introduced. Based on the field investigation and exploration, observation data, in-situ tests, laboratory tests, and numerical analysis, the method of evaluating amplification of ground motion on topography and deposit of loess sites is proposed. The method of risk assessments of seismic landslides is developed. Seismic design for engineering loess slopes with single-step and multi-steps is provided. The methods of evaluating liquefaction and seismic subsidence of loess ground and its treatment measurements are respectively presented. Moreover, the seismic design methods of pile foundation in loess ground are proposed for considering the negative friction on piles due to seismic subsidence and horizontal pushing force on piles under liquefaction. The above-mentioned methods and techniques have been adopted by a national code and a provincial code, which have been proved to be practical, efficient and rational through a large number of major engineering projects and buildings’ construction.

Lanmin Wang, Jinchang Chen, Ping Wang, Zhijian Wu, Ailan Che, Kun Xia
Large-Scale Seismic Seafloor Stability Evaluation in the South China Sea Incorporating Soil Degradation Effects

Submarine landslides are severe threats to the safety of offshore facilities. Earthquake is considered as a main trigger of submarine landslides. This study characterizes the large-scale seismic seafloor stability in the South China Sea with consideration for the earthquake induced degradation of soil. The digital elevation models, Peak Ground Acceleration (PGA) maps, and 3-D continuous soil models are used for the large-scale analysis. A seismic soil degradation model is developed to simulate the degradation of soil subjected to earthquake loading. The infinite slope model with bidirectional seismic load is adopted to evaluate seafloor stability. The results indicate that the degradation effects of soil can significantly reduce seafloor stability, especially in areas of high PGA. The findings emphasize the importance of soil degradation characteristics when performing large-scale seismic seafloor stability evaluation.

Yuxi Wang, Rui Wang, Jian-Min Zhang
Seismic Inspection of Existing Structures Based on the Amount of Their Deformation Due to Liquefaction

If structures such as houses, or structures with very long lengths, such as river dikes, railway embankments, and sewers, have not been constructed with liquefaction in mind, it is necessary to extract repair their parts that could be easily damaged in the event of a future earthquake. In inspecting these structures, it is necessary to quantitatively estimate the points where countermeasures are required based on the amount of settlement or the amount of uplift of the structures. For that purpose, it is necessary to properly consider the process of settlement or uplift of the structures due to liquefaction based on past cases of damage and the results of model tests, and develop a method to easily estimate the amount of settlement or uplift. In Japan, there are many cases of structural damage due to liquefaction, and many model tests using shaking tables have been conducted. Simple estimation methods have been developed based on these and are being used for inspections. Furthermore, measures are being taken based on the inspection results.

Susumu Yasuda

Invited Theme Lectures

Site Characterization for Site Response Analysis in Performance Based Approach

The local seismic hazard analysis would yield probabilistic uniform hazard acceleration response spectrum on the engineering bedrock outcrop. Thus, site-specific response analyses need to produce a probabilistic uniform hazard acceleration response spectrum on the ground surface. A possible performance based approach for this purpose requires a probabilistic estimation of soil stratification and engineering properties of encountered soil layers in the soil profile. The major uncertainties in site-specific response analysis arise from the variabilities of (a) local seismic hazard assessment, (b) selection and scaling of the hazard compatible input earthquake time histories, (c) soil stratification and engineering properties of encountered soil and rock layers, and (d) method of site response analysis. Even though the uncertainties related to first two items have primary importance on the outcome of the site-specific response analyses, the discussion in this article focuses on the observed variability and level of uncertainty in site conditions, related to soil stratification, thickness and type of encountered soil layers and their engineering properties, depth of ground water table and bedrock and properties of the engineering bedrock. Thus, one option may be conducting site response analyses for large number of soil profiles produced by Monte Carlo simulations for the investigated site to assess probabilistic performance based design acceleration spectra and acceleration time histories calculated on the ground surface based on 1D, 2D, or 3D site response analysis with respect to different performance levels.

Atilla Ansal, Gökçe Tönük
Seismic Landslide Susceptibility Assessment Based on Seismic Ground Motion and Earthquake Disaster Analysis

Seismic motion is one of the significant factors triggering slope instability. Landslides induced by intense earthquakes pose a great threat on the public security and traffic safety. About 720 landslides were caused by the 2014 Ms6.5 Ludian earthquake. The seismic ground motion records of Ludian earthquake and the influencing factors of seismic landslides are analyzed, and the susceptibility of landslides was evaluated in combination with machine learning models (BP neural network and SVM). The characteristics of Ludian earthquake motion records in time domain, frequency domain and time-frequency domain reveal that, the maximum value of PGA reached 949.2 cm/s2 (EW), and the duration of seismic wave within the nearest station to epicenter was 5.0 s (EW), 4.9 s (NS) and 4.3 s (UD), respectively. The result of Hilbert Huang Transform suggests that large instantaneous and cumulative energy concentrated in low frequency region (0–5 Hz), which can be considered to be related to large area landslides. Twelve factors are selected as the influencing factors of seismic landslides under Ludian earthquake, and the spatial correlation analysis indicates that the relation between factors and seismic landslide is characterized by strong nonlinear properties, of which ground motion parameters show a strong positive correlation with landslide distribution. The receiver operating characteristic (ROC) curve is adopted to compare the performance of two models. The results show that the AUC values of two curves are 90.1% and 89.5%, respectively, showing that BP neural network has higher accuracy of seismic landslide susceptibility results, and illustrating dependency on spatial distribution of seismic ground motion parameters.

Ailan Che, Hanxu Zhou, Jinchang Chen, Yuchen Wu, Ziyao Xu
SEM-Newmark Sliding Mass Analysis for Regional Coseismic Landslide Hazard Evaluation: A Case Study of the 2016 Kumamoto Earthquake

Coseismic landslides have been observed to cause severe damage during many historic earthquakes. Numerical simulation of fault rupture process, wave propagation and triggering of landslides considering realistic topography and geological conditions helps identify the key factors in the landslide triggering and evaluate the potential slope instability in seismically active areas.In this study, a physics-based regional coseismic landslide evaluation framework is constructed by integrating the flexible sliding analysis into spectral element model (SEM). The framework combines advantages of SEM model for its capability of simulating complex 3D large-scale wave propagation and a flexible mass sliding analysis for capturing local soil response in coseismic slope stability evaluation. Besides, equivalent linear method is implemented to incorporate the effect of soil nonlinearity.The developed model is adopted to simulate the landslides in the Aso Volcano area triggered by the 2016 Kumamoto earthquake. Realistic digital elevation model, site conditions and fault rupture model of simulated region (51 km × 43 km) are used in the simulation. The resulting sliding response is compared with the inventory of the triggered landslides to validate the proposed model.

Zhengwei Chen, Gang Wang
Understanding Excess Pore Water Dissipation in Soil Liquefaction Mitigation

Soil liquefaction has conventionally been studied in cyclic laboratory tests as a pure undrained condition, assuming that excess pore water pressure is unable to dissipate during rapid shearing. As such, dissipation of excess pore water pressure generated during shearing is often not modelled in classic cyclic simple shear and triaxial tests. In contrast, dissipation of excess pore pressure does occur in the field following major earthquakes in the form of severe ground settlement and sand boils where fine granular soils were found. In this theme lecture, two aspects of soil liquefaction are discussed. First, a study on the interaction of excess pore water pressure generation and dissipation on liquefiable clean sand in the cyclic triaxial test setup is conducted. Results show that pore water dissipation at merely a fraction of the permeability of the soil can have a significant effect to the soil’s susceptibility to liquefaction. This calls for future cyclic laboratory tests to adopt similar near-perfect undrained condition to better reflect a more realistic representation of soil liquefaction in the field. Second, a critical cyclic shear stress ratio is obtained, where excess pore pressure dissipation dominates excess pore pressure generation, beyond which the soil is no longer susceptible to liquefaction. Complementing this ratio is the identification of a critical void ratio in which the excess pore pressures do not build up despite considerable shearing amplitude, implying the soil is no longer susceptible to liquefaction at that degree of densification. Such information would serve as a guidance to mitigate soil liquefaction in soil densification operations.

Siau Chen Chian, Saizhao Du
Buried Pipeline Subjected to Ground Deformation and Seismic Landslide: A State-of-the-Art Review

Pipelines are commonly used for transporting different materials namely water, gas, sewage, and oil from one place to another. The different past earthquakes (1923 Kanto earthquake, 1971 San Fernando earthquake, 1994 Northridge earthquake, 2010 Chile earthquake) induced hazards such as landslide, fault movement, liquefaction, etc., resulting in the damage of buried pipelines. These hazards induced ground deformations are known as permanent ground deformation (PGD), and the deformation resulting from wave propagation is called transient ground deformation (TGD). Further, soil can move along or normal to the pipe axis, and accordingly, it can be further categorized as axial and transverse ground deformation respectively. Apart from seismic excitation, ground deformation and vibration can also be generated from other sources like pipe bursting, underground explosion etc. Failure of pipelines due to ground deformation can cause the source of firing, contamination to the environment, explosion, economic loss etc. Therefore, it is vital to design the buried pipeline incorporating the effect of possible ground movement on buried pipelines. Thus, the focus of the present review study is to understand the various possible patterns of ground deformation, estimation of additional forces on pipeline due to ground deformation, and their influence on the response of buried pipeline, which can be implemented in practice to carry out performance-based design of buried pipelines subjected to earthquake loadings.

Deepankar Choudhury, Chaidul H. Chaudhuri
Performance-Based Assessment and Design of Structures on Liquefiable Soils: From Triggering to Consequence and Mitigation

Effective liquefaction mitigation requires an improved fundamental understanding of triggering in terms of excess pore pressures in realistically stratified deposits that experience cross-layer interactions as well as performance-based consequence procedures that account for 3D soil-structure interaction (SSI), all mechanisms of deformation, total uncertainty, and the impact of mitigation. In this paper, we first present a series of centrifuge experiments to evaluate site response and pore pressure generation in layered liquefiable deposits, soil-structure interaction (SSI), and the impact of ground densification as a mitigation strategy on SSI and building performance. Second, experimental results are used to validate 1D and 3D, fully-coupled, nonlinear, dynamic finite element analyses of layered sites and soil-foundation-structure systems with and without mitigation. Third, numerical parametric studies (exceeding 167,000 1D and 63,000 3D simulations) are used to identify the functional forms for predicting liquefaction triggering in the free-field based on the capacity cumulative absolute velocity ( $$CA{V}_{c}$$ C A V c ) required to achieve a threshold excess pore pressure ratio (ru), settlement of unmitigated structures, and the relative impact of ground densification on foundation’s permanent settlement. And finally, a limited case history database is used validate the triggering and consequence models, accounting for field complexities not captured numerically or experimentally. This integrative approach yields a set of procedures that are the first to consider variations in soil layering and geometry, layer-to-layer cross interactions, foundation and structure properties (in 3D), contribution of all mechanisms of deformation below unmitigated structures, geometry and properties of densification, ground motion’s cumulative characteristics, total inherent model uncertainties, and the explicit conditionality of structural settlement on free-field triggering—which are necessary to realize the benefits of performance-based engineering in liquefaction assessment.

Shideh Dashti, Zachary Bullock, Yu-Wei Hwang
Calibration and Prediction of Seismic Behaviour of CFRD Dam for Performance Based Design

Dams have an important role in ensuring climate change resilience with limited available resources while balancing the needs for water and energy security for future. In this paper the behaviour of a tall CFRD dam located in a highly seismic area will be discussed. During the first impoundment, this concrete-faced rockfill dam (more than 150 m high) showed cracking of the parapet wall and damage to the joints between the parapet wall and the upstream face slabs. The broad objectives of this paper are to evaluate the available monitoring data and create a three-dimensional Finite Element (FE) model for dynamic behaviour. An extensive review of the deformation material parameters combined with the monitoring data during construction and impoundment was used to calibrate a stress-deformation 3D FE model. Material properties from both geophysical and geotechnical site investigations were used to develop the inputs for the non-linear time history analyses to estimate the permanent earthquake-induced deformations and stability within the dam body. Performance based criteria were used to confirm that the seismic displacements (horizontal and vertical) within the body of the dam are acceptable.

Barnali Ghosh, Vipul Kumar, Sergio Solera
Uncertainties in Performance Based Design Methodologies for Seismic Microzonation of Ground Motion and Site Effects: State of Development and Applications for Italy

Performance-Based Design (PBD) is a more rational and general design approach, particularly for seismic regions, in which the design criteria are expressed in terms of achieving stated performance objectives. In this approach it is relevant the performance required to structures and to geotechnical works subjected to stated levels of seismic hazard, as well as the geotechnical constitutive model used to predict the performance. The parameters of the constitutive models are related in turn to soil properties. Earthquake hazard zonation in urban areas is the first and most important step towards a seismic risk analysis in densely populated Regions. The Seismic Microzonation is nowadays a world-wide accepted tool for the mitigation of seismic risk. It is a complex process involving different disciplines ranging from Geology and Applied Seismology to Geotechnical and Structural Engineering. The aim achieved in seismic hazard microzonation studies throughout the last 20 years performed at the presented typical case histories in Italy was to quantify the spatial variability of the site response on some typical historical scenario earthquakes that would be expected in the area. In order to quantify the expected ground motion, the manner in which the seismic signal is propagating through the subsurface was defined. Propagation was particularly affected by the local geology and by the geotechnical dynamic ground conditions. Large amplification of the seismic signals generally occurs in areas where layers of low seismic shear wave velocity overlie material with high seismic wave velocity, i.e. where soft sediments cover bedrock or more stiff soils. Therefore, essential key issue here is to obtain a good understanding of the local subsurface conditions. The study builds on the recent experience of seismic microzonation studies in Sicily (Italy), after the effects of the 2018 seismic sequence. Examples of ground response analysis are presented by using some 1-D and 2-D codes, including methodologies taking into account soil uncertainties for site characterisation.

Salvatore Grasso, Maria Stella Vanessa Sammito
Mechanisms of Earthquake-Induced Landslides: Insights in Field and Laboratory Investigations

In this lecture, the author will share his experience in investigating several earthquake-induced landslides that occurred in Japan, Indonesia, and China, and discuss the mechanisms of slope failures and contributing factors. This work will present data from reconnaissance field surveys conducted at landslide sites and analysis of geological conditions related to slope instability. It will also discuss the results from a series of laboratory tests performed on soil specimens subjected to various loading modes and issues that may occur during investigations. This lecture will be of interest to researchers in the field of soil dynamics as well as engineers and decision-makers who are interested in the causes and mechanisms of earthquake-triggered landslides.

Ivan Gratchev
Regionalization of Liquefaction Triggering Models

The stress-based simplified liquefaction triggering procedure is the most widely used approach to assess liquefaction potential worldwide. However, empirical aspects of the procedure were primarily developed for tectonic earthquakes in active shallow-crustal tectonic regimes. Accordingly, the suitability of the simplified procedure for evaluating liquefaction triggering in other tectonic regimes and for induced earthquakes is questionable. Specifically, the suitability of the depth-stress reduction factor (rd) and magnitude scaling factor (MSF) relationships inherent to existing simplified models is uncertain for use in evaluating liquefaction triggering in stable continental regimes, subduction zone regimes, or for liquefaction triggering due to induced seismicity. This is because both rd, which accounts for the non-rigid soil profile response, and MSF, which accounts for shaking duration, are affected by the characteristics of the ground motions, which can differ among tectonic regimes, and soil profiles, which can vary regionally. Presented in this paper is a summary of ongoing efforts to regionalize liquefaction triggering models for evaluating liquefaction hazard. Central to this regionalization is the consistent development of tectonic-regime-specific rd and MSF relationships. The consistency in the approaches used to develop these relationships allows them to be interchanged within the same overall liquefaction triggering evaluation framework.

Russell A. Green
Different Aspects of the Effects of Liquefaction-Induced Lateral Spreading on Piles, Physical Modelling

Devastating pile failures due to liquefaction induced lateral spreading during or after major earthquakes in gently sloping ground or level grounds with free end, especially the pile foundations located under structures in or near ports and harbors, have been observed and studied for decades. Many physical and numerical modellings have been implemented or developed to study and understand the insight into different aspects of this phenomenon. In this regard, 1-g shake table and N-g dynamic centrifuge tests using both rigid box and laminar shear box have been implemented to physically model the problem and measure the parameters that may affect the impact of lateral spreading on deep foundations. A number of countermeasures have also been examined for tackling this problem. In this paper and theme lecture, the author tries to describe shortly the physical modelling researches and studies that have been conducted by him and his coworkers on this subject in more than a decade, and discuss the various parameters that are involved in physical modelling for studying the behavior of pile foundations subjected to liquefaction induced lateral spreading. A number of limitations involved in such physical modellings are also mentioned and some solutions to the involved challenges are discussed as well.

S. Mohsen Haeri
Hybrid Type Reinforcement of Highway Embankment Against Earthquake Induced Damage

In this research, a hybrid type pile system is proposed as a countermeasure of embankment failures during earthquakes. In the proposed technique, inclined piles are added in addition to the rows of vertical steel pipe piles. Inclined piles are expected to reduce ground deformation and settlement due to earthquakes, especially when the embankment widths are large. The focus of the research is to evaluate the stability of embankments by the external force of the earthquake and to elucidate the mechanism of the reinforcement effect by the vertical steel pipe piles and inclined piles. Model tests using shaking table and their numerical simulations were performed to evaluate the performance of the hybrid type reinforcement measures. Model tests were conducted by taking into the consideration the pile rigidity and embedment depth. The numerical simulations were also performed in which the effect of the inclination of the inclined piles was also taken into the consideration. Both the test results and numerical simulations have confirmed that in embankment with hybrid reinforcement measures, the increase in excess pore water pressure during an earthquake could be suppressed, and the settlement of the embankment could be significantly reduced.

Hemanta Hazarika, Chengjiong Qin, Yoshifumi Kochi, Hideo Furuichi, Nanase Ogawa, Masanori Murai
Multiple Liquefaction of Granular Soils: A New Stacked Ring Torsional Shear Apparatus and Discrete Element Modeling

In this study, we develop an innovative stacked ring torsional shear apparatus for multiple liquefaction tests of fully saturated sands. Stacked rings are built upon an existing hollow cylinder torsional shear apparatus to impose lateral constraint to the soil sample, such that multiple liquefaction can be tested in a fully saturated state. In this project, we used pressure compensation technique to reduce vertical friction between the sample and the ring direct contact force between the membrane and the rings such that the vertical stress will be much more uniformly distributed along the sample height. The new stacked ring device facilitates investigation of fabric evolution and its effect on reliquefaction resistance of sands.To further study the fundamental mechanism of the multiple liquefaction phenomenon, 3D clumped discrete model is used to construct realistic particle packings, and simulate fabric evolution during liquefaction-reconsolidation-reliquefaction process. The study reveals that strain history significantly influences the number of inter-particle contact and fabric anisotropy after reconsolidation, hence the subsequent liquefaction resistance greatly varies. The increase in relative density due to reconsolidation has only secondary effects. The state-of-the-art DEM simulation provides micromechanical insights into the fundamental mechanism of the multiple liquefaction phenomenon.

Duruo Huang, Zhengxin Yuan, Siyuan Yang, Pedram Fardad Amini, Gang Wang, Feng Jin
Liquefaction-Induced Underground Flow Failures in Gently-Inclined Fills Looser Than Critical

In two similar and unprecedented case histories during recent earthquakes in Hokkaido Japan, liquefied sand strangely flowed underground in gentle man-made fill slopes of a few percent gradient, leaving large surface depression behind. In both of them, a large amount of non-plastic fines was involved in loose fine fill sands. That particular sand with fines content Fc≈35% tested in undrained triaxial tests was found far more contractive with strain-softening and easier to flow than that of the same density deprived of fines. This strongly suggests that high fines content was the major cause of the strange flow failures because it destined the sand flowable on the contractive side of Steady State Line under sustained shear stress. Another series of cyclic simple shear tests on contractive sands with non-plastic fines under initial shear stress indicated that flow failure tends to occur in gentler slopes when the effective stress path comes across a yield line uniquely drawn from the origin on τ ~ σc’ diagram irrespective of stress paths. Thus, a scenario to realize the unprecedented flow failures has been clarified based on the field observations and test results.

Takaji Kokusho, Hazarika Hemanta, Tomohiro Ishizawa, Shin-ichiro Ishibashi
Prediction of Site Amplification of Shallow Bedrock Sites Using Deep Neural Network Model

Site amplification models are widely used with ground prediction equations to estimate ground motion intensity measures. The time-averaged shear wave velocity of top 30 m (VS30) is the primary site proxy in site amplification models. A large number of models have been developed for a range of site conditions. However, the simplified nature of all models produce large residuals compared with the computed responses. The prediction accuracy of the models can be greatly enhanced through use of machine learning technique. In this study, the outputs of nonlinear one-dimensional site response analyses are used to train the deep neural network (DNN) model. The linear and nonlinear components are separately trained. The comparisons highlight that the DNN model successfully captures the amplification characteristics of the shallow bedrock sites and produces significantly lower residual compared with the available simulation based model.

Duhee Park, Yonggook Lee, Hyundong Roh, Jieun Kang
Response of Pumice-Rich Soils to Cyclic Loading

Soils containing pumice are frequently encountered on engineering projects in the North Island of New Zealand. The presence of pumice is known to result in different material behaviours, including the resistance to cyclic loading. In this paper, results of triaxial testing on undisturbed specimens of dense, pumice-rich soils are presented, and examined to identify the apparent effects that differing pumice content has on the observed behaviours. It is shown that significant reductions in the cyclic resistance were observed in these soils compared with expectations for hard-grained materials, but that this effect appears to be fully developed with limited amounts of pumice in the soil. It is further shown that the undrained strength is significantly reduced by increasing amounts of pumice and that typical predictions of post-cyclic reconsolidation strains are unconservative in pumice bearing materials.

Mark Stringer
In-Situ Liquefaction Testing of a Medium Dense Sand Deposit and Comparison to Case History- and Laboratory-Based Cyclic Stress and Strain Evaluations

Observations of the dynamic loading and liquefaction response of a deep medium dense sand deposit to controlled blasting have allowed quantification of its large-volume dynamic behavior from the linear-elastic to nonlinear-inelastic regimes under in-situ conditions unaffected by the influence of sample disturbance or imposed laboratory boundary conditions. The dynamic response of the sand was shown to be governed by the S-waves resulting from blast-induced ground motions, the frequencies of which lie within the range of earthquake ground motions. The experimentally derived dataset allowed ready interpretation of the in-situ γ-ue responses under the cyclic strain approach. However, practitioners have more commonly interpreted cyclic behavior using the cyclic stress-based approach; thus this paper also presents the methodology implemented to interpret the equivalent number of stress cycles, Neq, and deduce the cyclic stress ratios, CSRs, generated during blast-induced shearing to provide a comprehensive comparison of the cyclic resistance of the in-situ and constant-volume, stress- and strain-controlled cyclic direct simple shear (DSS) behavior of reconstituted sand specimens consolidated to the in-situ vertical effective stress, relative density, and Vs. The multi-directional cyclic resistance of the in-situ deposit was observed to be larger than that derived from the results of the cyclic strain and stress interpretations of the uniaxial DSS test data, indicating the substantial contributions of natural soil fabric and partial drainage to liquefaction resistance during shaking. The cyclic resistance ratios, CRRs, computed using case history-based liquefaction triggering procedures based on the SPT, CPT, and Vs are compared to that determined from in-situ CRR-Neq relationships considering justified, assumed slopes of the CRR-N curve, indicating variable degrees of accuracy relative to the in-situ CRR, all of which were smaller than that associated with the in-situ cyclic resistance.

Armin W. Stuedlein, Amalesh Jana
Flowslide Due to Liquefaction in Petobo During the 2018 Palu Earthquake

The 2018 Palu Earthquake in Indonesia triggered liquefaction that was followed by many incidents, including in massive flowsides in Petobo. In this location, flowsides with deformations reaching 1 km occurred on gentle slopes. The liquefaction mechanism causing this enormous lateral deformation is not fully understood yet. In fact, a comprehensive understanding of this phenomenon mechanism can be used to mitigate similar disasters in the future. This study describes the flow slide observations in Petobo based on geotechnical investigations, including trench, geophysical survey, CPT, boreholes, and laboratory testing. The investigation showed that the flow slide location consisted of alternating layers with high and low permeabilities. The high permeability difference during pore water pressure dissipation in the liquefied layer resulted in void redistribution in layers overlain by low permeability layer. This void ratio increase triggered a significant shear strength loss in the corresponding layer. This study also shows that differences in the soil condition and the water table depth play important roles whether a location experiences a flow slide or not. The results of this study are expected to enrich our knowledge related to liquefaction induced flow slide and can be used as a basis to evaluate flow slide potential in other locations.

Mahdi Ibrahim Tanjung, Masyhur Irsyam, Andhika Sahadewa, Susumu Iai, Tetsuo Tobita, Hasbullah Nawir
Challenge to Geotechnical Earthquake Engineering Frontier: Consideration of Buildings Overturned by the 2011 Tohoku Earthquake Tsunami

A simplified pseudo-static analysis is proposed to estimate the safety factors with time against instability of a building subjected to tsunami loads, based on the results of a 2D shallow water equation. The proposed analysis is used to examine the key factors affecting the performance of buildings. A 3D analysis is also conducted for one building, the performance of which has not been explained by the simplified analysis. It is shown that: (1) if the tsunami hydrodynamic and buoyant forces are the major driving forces, the proposed analysis is capable of predicting the difference in observed building performance; (2) the safety factors against overturning and sliding become minima near when the peak landward and seaward flow velocities occur, while one against uplift around when the peak inundation depth occurs; (3) the instability of building tends to occur when any of the safety factors first becomes less than one, i.e., mostly at the peak landward flow velocity; (4) the hill immediately backward of a building might have induced a lower peak landward flow velocity at a lower inundation depth and a peak seaward flow velocity at a deeper inundation depth than any other location, having affected the seaward overturning of the building; (5) the major cause of the overturning in orthogonal to the tsunami propagation direction of one building is likely due to the collision of the drifting section of other building originally located on the seaside; and (6) the interdisciplinary collaboration was enormously useful to the progress of this study.

Kohji Tokimatsu, Michitaka Ishida, Shusaku Inoue
Multipoint Measurement of Microtremor and Seismic Motion of Slope Using Small Accelerometers

The authors have developed a real-time monitoring system. The monitoring sensors are in small size, power-saving, low-cost. The sensor unit is embedded with a 1D micro seism accelerometer and 3D MEMS (Micro Electro Mechanical Systems) accelerometer (seismic motion/vibration) and has been verifying its field performance since 2019. In the microtremor measurement, compared the result of acceleration Fourier amplitude spectrum with speed Fourier amplitude spectrum of conventional microtremor speed sensor at the same site, almost the same results of the predominant period were obtained. The measured results of the Mj 6.9 earthquake in Miyagi prefecture (March 20, 2021) will also be introduced as a case study. The developed microseism sensor units were applied to the field test of slope over two years, the result is that it can withstand long-term measurement with an accuracy comparable to a conventional seismometer.

Lin Wang, Takemine Yamada, Kentaro Kasamatsu, Kazuyoshi Hashimoto, Shangning Tao
Cyclic Failure Characteristics of Silty Sands with the Presence of Initial Shear Stress

The liquefaction of silty sands remains an outstanding issue since it triggers catastrophic hazards in recent earthquake events. The cyclic failure pattern is one of the fundamental aspects of liquefaction analysis. However, due to various influencing factors, such as packing density, confining pressure, initial shear stress, cyclic loading amplitude, etc., the failure patterns are less well understood and the underlying mechanism remains unclear. This study presents a series of laboratory testing results to identify the cyclic failure patterns of silty sands considering different soil states, fines contents, initial static shear stress, etc. The key finding is that the failure patterns are related to the states of soils and the cyclic loading characteristics, i.e., the combination of initial shear stress, density, confining stress and cyclic loading amplitude. Limitations of the existing prediction methods are discussed, and future work is also suggested.

Xiao Wei, Zhongxuan Yang, Jun Yang
Particle Fabric Imaging for Understanding the Monotonic and Cyclic Shear Response of Silts

The current knowledge from experimental research has shown the significant effect of particle structure (fabric) on the monotonic and cyclic shear behavior of silts, in addition to the well understood influence of void ratio and effective confining stress. Advancing the knowledge on this matter requires systematic quantification of particle fabric in a given silt matrix in terms of individual particle parameters (e.g., dimensional sizes, volumes, shapes, orientations) as well as inter-particle contact arrangements. In the research work undertaken with this background, new methodologies were developed for X-ray micro-computed tomography (μ-CT) imaging of silts with specific attention paid to: sampling and preparing of silt specimens, scanning parameters to obtain the needed image resolutions, and digital processing of images to capture individual particle data. It is shown that μ-CT imaging is able to effectively image and capture 3-dimensional fabric of silt. The particle fabric in silt specimens reconstituted via gravity deposition is illustrated using the μ-CT images produced using standard-size silica particles. The fabric(s) developed from the imaging of natural low plastic silt is also presented, and the findings are shown to be well in accord with those inferred from the mechanical laboratory element testing of the same silt. The work contributes to the accounting for fabric in understanding the macroscopic shear behavior of natural silts.

Dharma Wijewickreme, Ana Maria Valverde Sancho
Technical Framework of Performance-Based Design of Liquefaction Resistance

Seismic liquefaction hazard is always a very challenging problem in earthquake geotechnical engineering and it is necessary to develop the performance-based technology of liquefaction hazard governance. In the paper, by investigating the characteristics of seismic liquefaction hazard and combined with the existing performance-based earthquake engineering (PBEE) thought, the connotation and key points of the performance-based liquefaction hazard governance are proposed, and a technical framework of performance-based design of liquefaction resistance (PBDLR) is constructed according to the technical system construction rules. The results indicate that the liquefaction hazard has its own characteristics in the target, scale, mechanism, mode, chain effect, treatment technology and uncertainty of the influencing factors, and its technical system of hazard governance needs special consideration. The presented PBDLR, which integrates advantages of both the advanced risk management theory and the existing performance-based earthquake engineering technology system, and the elements and structure consider the sociality of the target, the integrity of the function, the hierarchy of the composition, the relevance of elements and the completeness of technology, and may provide guidance and reference for the technology development of liquefaction hazard governance in earthquake geotechnical engineering.

Jinyuan Yuan, Lanmin Wang, Xiaoming Yuan
Deformation Mechanisms of Stone Column-Improved Liquefiable Sloping Ground Under Earthquake Loadings

Liquefaction-induced deformation in sloping ground caused heavy damage to lifelines and overlaid structures in the past earthquake events. However, there is still a great need for estimation of large deformation of sloping ground during earthquakes and the associated deformation mitigation method. In this paper, a series of element tests were conducted to investigate residual volumetric strain and shear strain of soil element with initial shear stress, corresponding to infinite sloping ground condition. The element test results indicate that residual volumetric strain estimation method proposed by Shamoto and Zhang for level ground could be able to estimate that in sloping ground with initial static stress condition, and the residual shear strain is well correlated with maximum shear strain. In addition, a series of centrifuge model tests with respect to stone column improved sloping ground were designed and conducted to explore the effect of densification and drainage of stone column on deformation in sloping ground. The mechanisms of settlement and lateral spreading mitigation by densification and drainage effect of stone columns in sloping ground were analyzed and discussed in combination with centrifuge model tests and numerical simulation results. The present study provides an effective method for evaluating post-liquefaction settlement and deep insights of deformation mitigation of stone column in sloping ground, which is of great help for developing the performance-based mitigation method for sloping ground in the future.

Yan-Guo Zhou, Kai Liu
Liquefaction-Induced Downdrag on Piles: Insights from a Centrifuge and Numerical Modeling Program

Earthquake-induced soil liquefaction can cause settlement around piles, which can translate to negative skin friction, resulting in the development of drag load and settlement of the piles. Despite significant research progress on the effects of liquefaction on structures and the seismic response of piles, there is still a knowledge gap in the assessment of liquefaction-induced downdrag loads on piles. The interrelationships between mechanisms affecting negative skin friction are not accounted for in current practice, typically leading to over-conservative or unsafely designed piles. A series of centrifuge model tests were performed to assess liquefaction-induced downdrag and understand the interplay of pile embedment, pile-head load, excess pore pressure generation, dissipation, reconsolidation, and ground settlement on the axial load profile during and post shaking. The two tests featured instrumented piles, loaded with a range of axial loads and embedded at a range of depths into the bearing layer, in a uniform and a layered liquefiable profile respectively. The tests showed that most of the settlement of the piles occurred during shaking. The mechanisms behind the development of liquefaction-induced drag load on piles and settlements will be described and ramifications concerning the design of piles in liquefiable soils will be discussed.

Katerina Ziotopoulou, Sumeet K. Sinha, Bruce L. Kutter

Performance Design and Seismic Hazard Assessment

A Simplified Method to Evaluate the 2D Amplification of the Seismic Motion in Presence of Trapezoidal Alluvial Valley

The seismic response of an alluvial valley is ruled by a complex combination of geometric and stratigraphic factors, which makes it hard to define a simplified amplification factor, to be adopted by the codes as it usually happens in the case of topographic and 1D stratigraphic amplification factors. This paper reports the results of an extensive parametric study aimed at analysing the influence of geometrical (e.g. shape ratio and inclination of the edges) and stratigraphic factors (e.g. impedance ratio) on the ground motion at surface of a trapezoidal valley. The soil behaviour has been assumed as linear visco-elastic. The results of the numerical analyses have been synthesised in terms of Valley Amplification Factor (VAF), defined as the ratio between the spectral ordinates obtained from 2D and 1D analysis, the latter carried out along a soil column corresponding to the centre of the valley. Based on the results of the parametric analysis, a simplified method was finally proposed, to calculate straightforward the VAF distribution along the valley.

Giorgio Andrea Alleanza, Anna d’Onofrio, Francesco Silvestri
Critical Acceleration of Shallow Strip Foundations

This paper focuses on the assessment of the critical acceleration of soil-foundation systems and examines the influences of various relevant parameters. Using a formula recently proposed for the evaluation of the bearing capacity of shallow strip foundations in seismic conditions, the critical acceleration of the soil-foundation system has been evaluated imposing the equilibrium condition between the limit load of the soil-foundation system and load transmitted from the foundation onto the ground surface. The proposed solution led to an iterative equation that permits the numerical evaluation of the critical acceleration.An extensive parametric analysis allowed to examine the effects of many relevant parameters, such as the angle of shear strength of the foundation soil, the value of the surcharge acting aside the footing, the static vertical load transmitted by the foundation, the direction of the resultant of the inertial forces acting in the soil and in the superstructure.The results obtained are provided in design charts for given sets of the relevant parameters and allow a quick evaluation of the critical acceleration of the soil-foundation system.

Orazio Casablanca, Giovanni Biondi, Ernesto Cascone
Seismic Vulnerability Analysis of UHV Flat Wave Reactor Based on Probabilistic Seismic Demand Model

UHV Flat Wave Reactor is one of the important equipment of UHV converter station. Once damaged in an earthquake, it may cause the breakdown of the power system, resulting in serious economic losses and casualties. Therefore, analyzing its seismic vulnerability is significantly important. Firstly, failure modes, key vulnerable parts (weak parts), and seismic damage limit states of the structure under seismic action were analyzed by seismic damage investigation and finite element simulation. Secondly, based on the incremental dynamic analysis (IDA), the peak ground acceleration (PGA) was used as the input parameter to obtain the seismic response of key vulnerable parts. Thirdly, the seismic vulnerability curves of each key vulnerable part under different seismic damage states (complete, extensive, moderate, and slight) were generated by probabilistic seismic demand model (PSDM). Finally, through the reliability theory, the seismic vulnerability curves of UHV Flat Wave Reactor can be obtained. The results shown that this seismic vulnerability curve can give the probabilities of four seismic damage states for UHV Flat Wave Reactor under different PGA and reflects the seismic performance level of the structure, which provide reference for performance design and earthquake relief work.

Jiawei Cui, Ailan Che
Seismic Performance of the “Mediterranean Motorway” Piers Founded on Soft Soil

The seismic performance of bridge piers is significantly influenced by stiffness and strength of the foundation soil. In this study, the effects of soil-structure interaction are investigated with reference to the piers of the viaducts of the Mediterranean Motorway, a 442 km long road connecting the southernmost part of Italy. Non-linear dynamic analyses were carried out on numerical models including soil, deep foundation and pier, under unscaled records of real earthquakes. The characteristics of the analyzed prototypes were inferred from the most recurrent features of the viaducts. The results of the analyses show that soil-structure interaction reduces the bending drift, due to significant peak and permanent foundation rotations. Hence, the overall effect of soil-structure interaction is a reduction of structural damage while, on the other hand, pier instability and tilting are enhanced. The results of the analyses were finally exploited to compute the demand hazard curve of the pier prototype, expressed in terms of mean annual rate of exceedance of the foundation rotation.

Filomena de Silva, Michele Boccardi, Anna d’Onofrio, Valeria Licata, Enrico Mittiga, Francesco Silvestri
Performance-Based Estimation of Lateral Spread Displacement in the State of California: A Case Study for the Implementation of Performance-Based Design in Geotechnical Practice

Within the last decade, many researchers have demonstrated that performance-based methods for estimating the hazard from liquefaction triggering and its effects can effectively be approximated using simplified, map-based methods. However, the development of the reference parameter maps that are necessary for the implementation of these simplified performance-based methods is major endeavor and has proven to be a significant impediment for the implementation of these methods in engineering practice. This study presents a case history of how the simplified performance-based reference parameter maps for a liquefaction-related hazard (lateral spread displacement) were recently developed for a single state known for its high seismicity (California) in the United States. Through a mentored research experience involving several undergraduate and graduate students, the development of the lateral spread reference parameter maps corresponding to three commonly used return periods (475 years, 1033 years, and 2475 years) for the State of California is summarized and presented. When combined with an existing simplified performance-based lateral spread method, the reference parameter maps described in this paper become a powerful design resource for engineers in California. Example lateral spread displacement values are computed for various parts of the state to validate the maps/approach and to demonstrate the performance-based methodology and potential uses/benefits.

Kevin W. Franke, Clay Fullmer, Delila Lasson, Dallin Smith, Sarah McClellan, Ivy Stout, Riley Hales
On a Novel Seismic Design Approach for Integral Abutment Bridges Based on Nonlinear Static Analysis

This paper focuses on the seismic performance and design of a single-span integral abutment bridge (IAB), as a structural system characterised by a monolithic connection between deck and abutments. Although this is becoming a popular design solution due to its low maintenance requirements, there is still the need of developing robust design criteria for such structures under seismic conditions, mainly because of the complex soil-abutment-deck interaction. This study proposes an application of a novel design method for IABs to a reference case study inspired by a real integral bridge recently built in Italy. In the proposed method, the seismic capacity of the bridge is obtained through a nonlinear static analysis of the entire soil-structure system, in which the soil domain is perturbed by a distribution of equivalent forces aimed at reproducing the effects associated with the significant modes of the bridge. This approach is validated against the results of several dynamic analyses carried out on an advanced, full soil-structure model of the reference bridge implemented in OpenSees. Several seismic scenarios are taken into account, as well as the possibility to use an average response spectrum prescribed by technical provisions. This study demonstrates that the proposed design approach is able to reproduce quite satisfactorily the performance of the structure, in terms of maximum internal forces and displacements, with a very low computational demand.

Domenico Gallese, Davide Noè Gorini, Luigi Callisto
Influence of Seismic Displacement Models on Landslide Prediction at the Egkremnoi Coastline During the 2015 Lefkada Earthquake

Strong earthquakes can trigger thousands of landslides in mountainous areas, and accurate prediction of landslide occurrence can reduce the risk of infrastructure and communities to earthquake-induced regional landslides. However, current methods for regional landslide prediction are still in their infancy and prediction of landslide occurrence remains a challenge. In this study, a pseudo-three-dimensional (pseudo-3D) procedure was implemented to a 1.35 km2 area along the Egkremnoi coastline of the Lefkada island in Greece to assess the triggering of landslides and their geometry. To better understand the influence of permanent seismic displacement models on landslide prediction, seven models were employed in the prediction procedure. The adopted displacement models include four Newmark-type rigid block models and three flexible models. The results show that in this specific case and for the input parameters used in the analyses, the Bray and Macedo model achieves the best performance in landslide prediction in terms of the percentage of correctly predicted landslides and the centroid distance between mapped and predicted landslides, which are used to quantify the location accuracy of predicted landslides. The Rathje and Antonakos model predicts more landslides that overlap with mapped landslides, but also overpredicts more overall.

Weibing Gong, Dimitrios Zekkos, Marin Clark
Assessment of the Seismic Performance of Large Mass Ratio Tuned Mass Dampers in a Soil-Structure System

The use of Tuned Mass Dampers with a large mass ratio (LM-TMD) can represent an efficient means for seismic retrofitting of existing structures, intended as a superelevation of the pre-existing layout having a mass up to 20–30% of the structural mass. The mechanical properties of the LM-TMD should be designed to reproduce a proper dynamic coupling with the remaining structural system but there is still a limited knowledge about its effectiveness. Hence, this study presents a coupled numerical modelling of a soil-structure system equipped with a LM-TMD, implemented in OpenSees, as a benchmark to test the effectiveness of the non-conventional device. The full model simulates the nonlinear response of the soil-structure-TMD system and the multi-directionality of the seismic motion. The soil-structure interaction effects are quantified by comparing the results of the full model with those obtained in the case of structure with fixed-base. The results highlight the significant improvement of the structural performance when the LM-TMD is used, and the influence of soil-structure interaction on its effectiveness, as an additional source of energy dissipation not affected by the presence of the anti-seismic device.

Davide Noè Gorini, Guglielmo Clarizia, Elide Nastri, Pasquale Roberto Marrazzo, Rosario Montuori
Required Strength of Geosynthetics in Seismic Reinforced Soil Retaining Wall in Multi-tiered Configuration

Based on the limit equilibrium method, seismic stability analyses of the reinforced multi-tiered wall are conducted to calculate the required strength of geosynthetics. The available reinforcement tensile capacity against the front and rear-end pullout is used to determine the distribution of the required strength and the strength of connection to the facing for each layer. The calculated solutions are compared with the results obtained from the experiments and numerical simulations for the verification of the presented method. A series of parametric study is carried out to investigate seismic effects on the stability of reinforced multi-tiered wall and its failure mechanism. The positions of maximum value of the required tensile force are closer to the end rear of the reinforcements resulting in the pullout failures at the reinforcements of the uppermost wall. Increasing offset distance and tier number can yield less reinforcement strength required for seismic stability. The height of the uppermost wall can be selected as 0.6–0.7 times of the total height for a two-tiered wall seismic design.

Shilin Jia, Fei Zhang, Xiaoyi Lu, Yuming Zhu
Study on Design Method of Vertical Bearing Capacity of Rock-Socketed Piles Based on Reliability

The traditional vertical bearing capacity design method of rock-socketed piles uses the constant safety factor K to assess the reliability of pile foundations, and this method leads to the high cost or the insufficient safety reserve of some pile foundations. The Bayesian optimization estimation was used to process the complete data of 144 rock-socketed piles, 65 sets of skin resistance data and 31 sets of end resistance data in rock-socketed segment. The analysis of the probability limit state design method was carried out for the five partial resistance coefficients: the whole pile, the skin resistance of soil section, the resistance of rock section, the skin resistance of the rock section and the end resistance of the rock section. The relationship between the partial coefficients, embedment ratio and uniaxial saturation compressive strength of rock was established, and the suggested partial coefficients was proposed. Finally, the partial coefficients were introduced into the calculation formula of the vertical bearing capacity of rock-socketed piles in the Code for Design of Ground and Foundation of Highway Bridge and Culvert. This formula can provide reference for further revision and improvement of existing normative practices.

Zhongwei Li, Hanxuan Wang, Guoliang Dai, Fan Huang
Performance-Based Probabilistic Assessment of Liquefaction Induced Building Settlement in Earthquake Engineering

The Current engineering procedures for estimating liquefaction-induced building settlements (LIBS) utilize deterministic or pseudoprobabilistic approaches that separate the estimation of ground motion intensity measures (IMs) hazard from the estimation of LIBS hazard. In contrast, in a performance-based probabilistic approach, the estimation of the IM hazard is coupled with the estimation of the LIBS hazard. As a result, engineers can directly obtain LIBS estimates corresponding to a selected design hazard level (or return period), which is more consistent with performance-based engineering design. In this study, we make new developments for the performance-based probabilistic assessment of LIBS hazard, including 1) the performance-based assessment of LIBS considering the hazard from a single IM in terms of scalar probabilistic seismic hazard assessment (PSHA), 2) the performance-based assessment of LIBS considering the hazard from multiple IMs in terms of vector PSHA, 3) deaggregation of earthquake scenarios from LIBS hazard curves, and 4) treatment of aleatory variability and epistemic uncertainties. We implement the developments in a computational platform named “LIBS” to facilitate their use in engineering practice. Finally, we share the insights from the comparison between the performance-based and pseudoprobabilistic-based estimates of LIBS hazard.

Chenying Liu, Jorge Macedo, Gabriel Candia
Seismic Hazard Analysis, Geotechnical, and Structural Evaluations for a Research Reactor Building in the Philippines

The Philippine Research Reactor-1 (PRR-1) Building, which is operated by Philippine Nuclear Research Institute (PNRI) but has been shut down since 1988, is planned to be used as housing for a subcritical assembly. Considering that the building is old and not in use, there is a need to assess and evaluate its overall integrity.This paper presents the seismic hazard analysis, geotechnical, and structural evaluations done for the Reactor-1 Building. The geotechnical evaluation includes the subsurface characterization and derivation of shear strength parameters from field and laboratory testing. Probabilistic Seismic Hazard Analysis (PSHA) was performed to quantify the seismic hazard on site for different hazard levels based on recurrence interval. Upon performing PSHA, site specific response spectra were developed for different return periods and damping ratios.Structural evaluation, which includes structural idealization, analytical procedures, and calculations, was also conducted to determine the building integrity. Retrofitting is recommended for structural members shown to be inadequate to achieve the performance objective of life safety subjected to a 475-year return period earthquake. Under classification of Hazard Category 4 according to IAEA guidelines, PRR-1 facilities may proceed with operation of a subcritical assembly below 0.1 MW provided retrofit measures are undertaken to meet the objective.

Roy Anthony C. Luna, Patrick Adrian Y. Selda, Rodgie Ello B. Cabungcal, Luis Ariel B. Morillo, Stanley Brian R. Sayson, Alvie J. Asuncion-Astronomo
Integrating Local Site Response Evaluations in Seismic Hazard Assessments

In recent years, site-specific seismic hazard studies have steadily been incorporated in large infrastructure projects as performance-based earthquake assessment is continuously being assimilated into engineering practice. The first component of this framework is the characterization of (engineering) bedrock or “seismic base” accelerations, often using probabilistic seismic hazard analysis (PSHA) with ergodic (global) ground motion models. A major part of the analysis effort is the accounting of local site effects, which have been known to alter surface ground motions significantly. Moreover, for sites with soft/loose deposits, the ground shaking intensity is expected to be high, and thus, the onset of liquefaction may also be a concern. Liquefaction effects are often evaluated using simplified methods via triggering curves, but there is an increasing trend towards using nonlinear effective stress analysis to better simulate surface ground motions and obtain a more appropriate structural response.In this paper, a case study involving a site situated on soft sedimentary deposits is discussed to show the integration of seismic hazard and site response analyses. This study highlights the importance of advancing the state-of-practice to integrate local site response evaluations in seismic hazard assessment.

Roy Anthony C. Luna, Ramon D. Quebral, Patrick Adrian Y. Selda, Francis Jenner T. Bernales, Stanley Brian R. Sayson
Performance-Based Design Review of a Reinforced Earth Retaining Wall for a Road Embankment Project in the Philippines

Several infrastructure projects such as light rail transit system, national roads, expressways, and underground rapid transit lines are currently being developed in the Philippines to address the huge gap in transportation infrastructure. One of the major infrastructure projects is an elevated toll expressway that connects several business districts in the southern areas of Metro Manila. Located near the coastal area of Manila Bay, the expressway alignment is underlain by soft clay and loose sand layers where long-term settlements and liquefaction are expected. In this context, a need to transition from conventional design methods to Performance-Based Design is essential in providing a cost-effective solution.This paper presents the methodology utilized in evaluating stability and deformation of road embankment protection designs involving Mechanically Stabilized Earth walls. The assessment of the proposed ground improvement by Soil-Cement Columns to minimize settlements and to mitigate the onset of geohazards such as liquefaction are discussed. This paper also shows the instabilities observed by performing Slope Stability Analysis, and their corresponding deformations evaluated by Finite Element Analysis. Recommendations for optimization and design improvements based on Client’s risk tolerance, and further studies to be undertaken are identified.

Roy Anthony C. Luna, Jenna Carmela C. Pallarca, Patrick Adian Y. Selda, Rodgie Ello B. Cabungcal, Marvin Renzo B. Malonzo, Helli-Mar T. Trilles
Geotechnical and Seismic Design Considerations for Coastal Protection and Retaining Structures in Reclaimed Lands in Manila Bay

In recent years, more and more reclamation projects along the coastlines of the Philippines’ capital, Manila, are being envisioned and realized. With the commonly known poor ground conditions in coastal areas, characterized by soft and loose sedimentary soils, coupled with the combination of hazards that endanger the country, both seismic and climate related, the need for performance based, advanced engineering analyses has never been more important in the local design scene. This paper presents the geotechnical and seismic considerations in the long-term design of coastal protection and retaining structures of land reclamation projects in Manila Bay, aimed at addressing the multiple hazards inherent on the geographical location of the country. State-of-the-art technologies and approaches in geotechnical characterization and seismic analysis are discussed, including the seismic tests using seismic velocity logging (SVL) to obtain relevant dynamic properties of the soil, probabilistic seismic hazard analysis (PSHA) to quantify the overall seismic hazard of the area and obtain appropriate ground motions, and nonlinear time-history analysis (NLTHA) by finite element method (FEM) to simulate the response of the land mass and structures during earthquake events.A case study on a massive land reclamation development along the shorelines of Manila Bay is presented.

Gian Paulo D. Reyes, Roy Anthony C. Luna, John Michael I. Tanap, Marvin Renzo B. Malonzo, Helli-mar T. Trilles
A Framework for Real-Time Seismic Performance Assessment of Pile-Supported Wharf Structures Incorporating a Long Short-Term Memory Neural Network

Past earthquake disasters have shown that pile-supported wharf structures are susceptible to severe damage during earthquakes, and thus it is important to assess the seismic performance of pile-supported wharf structures. Effective post-earthquake performance assessment requires a prompt and accurate assessment. However, existing performance assessment methods cannot simultaneously meet these requirements. In recent years, with the progress of software and hardware technology, machine learning has surpassed the traditional methods in many fields such as computer vision and natural language processing. The time history of ground motion acceleration is a typical time series data, and most of the traditional machine learning methods are not good at dealing with such high dimensional time series data. Recurrent Neural Network (RNN), as a technology suitable for extracting and learning features of high-dimensional time-series data, has attracted more and more attention. A graphical user interface is developed to combine nonlinear dynamic time history analysis of pile-supported wharf structures with the implementation of Recurrent Neural Network (RNN). A framework for real-time seismic performance assessment of pile-supported wharf structures incorporating a Long Short-Term Memory (LSTM) neural network is proposed. The framework is built around a workflow that establishes mapping rules between ground motions and structural performance via pile-supported wharf structures models. In this paper, the analysis framework and the main components of the graphical user interface are presented.

Liang Tang, Yi Zhang, Zheng Zhang, Wanting Zhang, Xianzhang Ling
Evaluation of the Liquefaction Hazard for Sites and Embankments Improved with Dense Granular Columns

Dense granular columns (DGC) have become a common soil improvement strategy for critical embankment structures founded on potentially liquefiable deposits. The state-of-practice for the design of DGCs is limited to simplified methods that consider, separately, the three-primary liquefaction-mitigation mechanisms provided by these columns: (i) installation-induced densification; (ii) enhanced drainage; and (iii) shear reinforcement. Critical aspects, such as the effects of soil-column-embankment interaction, site characteristics and layer-to-layer interaction, ground motion characteristics beyond the peak ground acceleration, and the total uncertainty, are not included in current engineering design procedures. In this work, we present results from a numerical parametric study, previously validated with dynamic centrifuge test results, to evaluate the liquefaction hazard in layered profiles improved with DGCs. The criteria for various degrees of liquefaction are based on peak excess pore pressure ratios and shear strains observed within each layer. Our study includes different properties and geometries for both soil and DGCs, various confining pressures induced by an overlying embankment, as well as a large collection of ground motions from shallow crustal and subduction earthquakes. We performed a total of 30,000 3D, fully coupled, nonlinear, dynamic finite-element (FE) simulations in OpenSees using a state-of-the-art soil constitutive model (PDMY02), whose properties were calibrated based on both element level laboratory tests and a free-field boundary-value problem modeled in the centrifuge. The results from this parametric study are used to develop a probabilistic predictive model for the triggering of liquefaction in embankment sites treated with DGCs.

Juan Carlos Tiznado, Shideh Dashti, Christian Ledezma
History of Liquefaction Hazard Map Development and a New Method for Creating Hazard Maps for Low-Rise Houses

This paper introduces the development history, current situation and problems concerning hazard maps for soil liquefaction. A zonation manual produced by the TC 4 of ISSMFE in 1993, is introduced first. Then, the development of hazard maps for liquefaction in Japan is reviewed. A hazard map for liquefaction was first created by Ishihara and Ogawa in 1978. Currently, hazard maps exist for all administrative divisions of 47 administrative zones. However, they are not being fully used. In order to be fully used, hazard maps must be upgraded, mainly improving the following three points: i) accuracy and reliability, ii) maps specifically for low-rise housing, and iii) maps showing timelines of emergency risks immediately after an earthquake. In addition, there is a lack of risk communication between the government, home builders and residents. Therefore, in order to solve this problem, Japan’s Ministry of Land, Infrastructure, Transport and Tourism organized a technical committee to propose a method for creating liquefaction hazard maps, and published the “Liquefaction Hazard Map Guide for Risk Communication” in February 2021. It is expected that such risk communication will promote measures against liquefaction.

Susumu Yasuda
Dynamic Response Analysis of Slope Based on 3D Mesh Model Reconstruction and Electrical Resistivity Topography

Seismic motion is one of the important factors triggering slope instability and failure. In areas with frequent earthquakes, such as the southwestern region of China, the motion of earthquakes with small magnitude could induce the reactivation of slope, resulting in destruction to the surrounding public transportation infrastructure. Along the Dayong expressway in Yunnan province, China, a slope is situated in Damieju village, Chenghai town, Yongsheng county, having the trend developing into geological disaster. In order to investigate the potential risk of Damieju slope under earthquakes, electrical resistivity measurement containing 3 measuring lines were conducted in the field. Based on the 3D mesh model reconstruction method, the numerical model of Damieju slope is built using resistivity data, considering the spatial distribution characteristic of silt clay and mudstone. And the dynamic response of Damieju slope model is simulated using the commercial software Abaqus. The seismic wave record under the 2019 Yongsheng 4.9 earthquake is adopted as the input wave. The amplification coefficient of PGA reaches the maximum value of 4.18. The maximum value of displacement appears at the area with approximate height of 60 m, where could be the most unstable area of Damieju slope under the influence of seismic motion.

Hanxu Zhou, Ailan Che

Ground Motions and Site Effects

In-Situ Characterization of the Near-Surface Small Strain Damping Ratio at the Garner Valley Downhole Array Through Surface Waves Analysis

The quantification of attenuation properties of soils has great relevance in geotechnical earthquake engineering. The small strain damping ratio is generally obtained from either direct laboratory tests on small samples or generic empirical relationships. Alternatively, some promising techniques for extracting in-situ small strain damping ratio rely on the analysis of surface wave data. This paper presents a subset of results from a massive dynamic site characterization study at the Garner Valley Downhole Array, wherein in-situ damping ratio profiles have been extracted from several multichannel analysis of surface waves (MASW) datasets. Waveforms generated from both a sledgehammer and a dynamic shaker were recorded, allowing for comparisons between the damping estimates obtained from both types of sources. Dispersion data and attenuation curves were derived from the waveforms using several approaches presented in the literature, and one new approach developed by the authors. This paper documents the inter-method differences and similarities across approaches in terms of uncertainties in the wavefield attenuation, together with the impacts on the amplification of the soil deposit. This study contributes towards better, in-situ characterization of the attenuation properties of soil deposits, enhancing the accuracy of ground models used in dynamic analyses. It is expected that progress in this area will lead to greater reliability of predicted ground motion amplification.

Mauro Aimar, Mauro Francavilla, Brady R. Cox, Sebastiano Foti
A Seismic Microzonation Study for Some Areas Around the Mt. Etna Volcano on the East Coast of Sicily, Italy

On the night of December 26th, 2018, a strong earthquake, with a magnitude of 4.9 on the Richter scale, hit the southeastern side of the Mt. Etna Volcano (Sicily), with the epicenter between the Municipalities of Viagrande and Trecastagni. The hypocenter of the strong earthquake was located just 1 km deep and, for this reason, the effects of the shock were greatly amplified on the ground surface. They have been counted not only material damages to churches and buildings, but also 10 injured around the epicenter area. Therefore, an investigation campaign was carried out with the aim of planning the reconstruction of the damaged areas. In situ soil investigations were carried out in order to determine the soil profile and the geotechnical parameters for the area under consideration. Among in situ tests, borings, Down Hole Tests (D-H), Multichannel Analysis Surface Wave Tests (MASW), Seismic Refraction Tomography and Horizontal to Vertical Spectral Ratio Tests (HVSR) were carried out, with the aim to evaluate the soil profile of shear wave velocity (Vs). Moreover, laboratory tests were carried out on undisturbed samples in the static field: Shear Tests (ST). The results have been grouped and allowed the characterisation of the following soil categories: clay, sandy clay, silica sand, volcanic sand, volcanic rock. The seismic behaviour of these soil categories has been used for the site response analysis and for the seismic microzonation of the studied area.

Antonio Cavallaro, Salvatore Grasso, Maria Stella Vanessa Sammito
Influence of Local Soil Conditions on the Damage Distribution in Izmir Bay During the October 30, 2020, Samos Earthquake

On October 30, 2020, a damaging earthquake of moment magnitude 6.6 struck about 14 km northeast of the island of Samos, Greece, and about 70 km from the center of the city of Izmir in Turkey. Even though the epicenter was relatively far away, the effects of the seismic event in the highly populated city center of Izmir were destructive causing over 100 fatalities and significant structural damage. Multiple failures of high buildings constituted the major source of the fatalities.This paper aims to understand the link between the localized damage distribution and the nature of amplification effects that have been observed in Izmir Bay, starting from collection and data analysis interpretation of seismic records and targeted damage assessment of the built environment, as well as geological and morphological characteristics of the area and the geotechnical properties of soils. Critical analysis of the numerous recorded signals shows the key role of the young alluvium and shallow marine deposits of the basin on which Izmir Bay was growing. The coupling mechanism between the frequency content of the shaking and the fundamental frequencies of the damaged buildings contributed to exacerbating the inertial forces acting on the collapsed buildings.

Anna Chiaradonna, Eyyub Karakan, Giuseppe Lanzo, Paola Monaco, Alper Sezer, Mourad Karray
Topographic Amplification of Seismic Ground Motion in Hills

Effect of slopes on topographic amplification has been rarely studied. This is an important issue for performance based design of foundations in hilly areas. The presented numerical study examines the seismic response of slope topography for different site conditions under vertically propagating SV waves. A single face slope, of constant base width, is considered. The influence of slope angle (β) and frequency of excitation on amplification of seismic ground motion is investigated. The two-dimensional Finite Element Analysis (2D-FEA) is adopted for the present seismic analysis of slope-topography. The behavior of soil is assumed as linear elastic. The side boundary is considered as a free field boundary to avoid reflection of ground motion. Complaint base condition is assigned at the bottom of the model. The incident ground motion is applied at the base of the FEA model and a far-field point from toe and crest is defined. The Seismic-Slope Topographic Amplification Factor (S-STAF) is expressed as a ratio of seismic response at the near field (along and surrounding the slope) and response at free field condition (horizontal ground surface). It was observed that the amplification increases as the slope angle and frequency of excitation increases. Further, the distance of point of the maximum peak ground acceleration from the crest increases as the slope angle increases.

Sukanta Das, B. K. Maheshwari
Input Ground Motion Selection for Site Response Analysis at the Port of Wellington (New Zealand)

Semi-empirical liquefaction evaluation procedures have several limitations and challenges when applied to nonstandard soils such as those found in reclaimed land. Hence, there is a growing need and benefit of additionally performing advanced numerical analyses. Such techniques can provide unique insights on the dynamic response of soils which are beyond the scope of simplified procedures. A key issue in applying dynamic analyses for back-analysis of case history records is the determination of appropriate input earthquake motions. In this study, recorded ground motions of three earthquakes at sites around Wellington city are deconvolved and used to derive input motions for 1D site response analysis of the reclaimed land at the port of Wellington, New Zealand (CentrePort). Three of the deconvolved sites sit on native soil deposits 200–600 m away from the port, and one nominal rock site is approximately 1 km away. The deconvolved motions are then used for forward analysis at neighboring sites and rigorously compared to recorded motions to scrutinize the appropriateness of the input motions. Surface ground motions of the three sites within the basin incorporate the effects of the complex basin geometry (i.e., basin-edge effects) which are not present at the nominal rock site. Since 1D site response analysis does not simulate these features, the study finds that the three basin sites provide reasonable estimates of the recorded surface motions for 1D analysis compared to the nominal rock site. The recommendations in this study will inform the definition of input motions for future analyses of case-history sites in CentrePort.

Riwaj Dhakal, Misko Cubrinovski, Jonathan Bray
Responses of a Cantilever Retaining Wall Subjected to Asymmetric Near-Fault Ground Motions

Fault rupture produces pulse-like ground motions in the near-fault regions, which are significantly different from ordinary ground motions in the far-field regions with respect to both intensity and frequency contents. The pulse-like ground motions are characterized as high amplitude and long period pulses that could cause severe damage to structures. In this study, we investigated how the pulse-like ground motions have effects on the responses of a cantilever retaining wall by performing a series of numerical simulations. The seismic behaviors of the retaining wall were numerically modeled by adopting a finite difference scheme. High amplitude ground motions scaled to a fixed PGA were first collected as an input dataset and classified into pulse-like, non-pulse-like, and ambiguous motions by a pulse indicator. Then, differences in the development of displacements at the wall were quantitatively compared between the types of the motions. Additional simulations were carried out with the original and inverted input ground motions to investigate the effect of asymmetry of ground motion on the wall responses. It turned out that the asymmetrical ground motions with larger velocity amplitudes in the direction coincide with the relative wall movement could generate significant wall displacements.

Seong Jun Ha, Hwanwoo Seo, Hyungseob Kim, Byungmin Kim
Modeling Two-Dimensional Site Effects at the Treasure Island Downhole Array

Numerous studies have challenged the wide applicability of one-dimensional (1D) ground response analyses (GRAs), seemingly due to the variable subsurface geologic conditions present at most sites. However, while two- and three-dimensional (2D and 3D, respectively) GRAs could be used to overcome these limitations, the lack of site-specific 3D subsurface models and the daunting computational costs of current software are impediments to improved modeling of site effects in engineering practice. In this paper, we aim to address both of these challenges by utilizing: (1) a practical framework called the ‘H/V geostatistical approach’ to develop large-scale, site-specific, 3D shear wave velocity models, and (2) an optimized open-source, finite element software called ‘Seismo-VLAB’ to perform large-scale 2D/3D finite element analyses. The investigations are performed at the Treasure Island Downhole Array (TIDA). By incorporating cross-sections across different lateral extents and azimuths and validating the site response predictions relative to recorded earthquake motions in the downhole array, we show that the site-specific 3D Vs model from the H/V geostatistical approach is capable of replicating wave scattering and more complex wave propagation phenomena observed in the field.

Mohamad M. Hallal, Peyman Ayoubi, Domniki Asimaki, Brady R. Cox
Variability in Kappa () Estimated with Coda Waves for California

Characterizing and quantifying the effects of local soil conditions are essential for site-specific seismic hazard assessment and site response analysis. The high-frequency spectral decay parameter $${\kappa }_{r}$$ κ r and its site-specific component, $${\kappa }_{0}$$ κ 0 , have gained popularity due to their abilities to characterize near-surface attenuation in situ. Values of $${\kappa }_{0}$$ κ 0 for rock conditions are of particular interest for site-specific seismic hazard analysis for critical facilities. However, ground motions (GM) recorded at sites underlain by stiff soils or rocks are scarce, which limits the computation of $${\kappa }_{0}$$ κ 0 values via the classic acceleration spectrum method. Recent research has found that $$\kappa $$ κ values computed using the coda wave of a GM (i.e., the multiple-scattered wave that is less sensitive to the earthquake source and local site effects) can capture regional variations of the attenuation of hard rock materials regardless of the subsurface conditions near the surface. However, there are still large uncertainties in $$\kappa $$ κ estimates based on the coda wave per GM, $${\kappa }_{r\_coda}$$ κ r _ c o d a , associated with the absence of consistent guidelines for the computation procedure and a user-orientated GM processing protocol. This work uses California GMs to examine the variability associated with the computation process of $${\kappa }_{r\_coda}$$ κ r _ c o d a , including the choice of onset of the coda wave and its duration. The objective of this paper is to understand and quantify the variabilities in $$\kappa $$ κ values based on coda waves, which has potentially large implications in its applicability in future ground motion models.

Chunyang Ji, Ashly Cabas, Marco Pilz, Albert Kottke
Structural Response to High Frequency Motion Released in Nonlinear Soil

High frequency motion is often registered in experimental measurements in laminar boxes placed on shaking tables in centrifuge or 1-g laboratory tests when investigating soil free field response under sinusoidal input motions. The source of the high frequency motion is often attributed to inaccuracies of experimental setups. On the other hand, some numerical studies suggested physical explanations, due to soil complex mechanical behavior, such as general soil nonlinearity or dilation. The most recent numerical studies suggest initially another potential source of the high frequency motion in tested soil specimens, i.e. the hypothetical release of unloading elastic waves in the steady state response. Moreover, these studies show that soil-released high frequency motion can potentially impact structural response.This paper presents a brief example of a numerical study on dynamic soil-structure interaction analyzed under harmonic input motion of a single sinusoidal driving frequency. The soil is modelled using an advanced soil constitutive model to accurately represent soil cyclic behavior. The analyzed structure is shown to experience the motion of the driving frequency and the motion of high frequency in the steady state response. Importantly, the computed high frequency motion is not present at the specimen base and is apparently introduced into the dynamic system by soil nonlinearity.

Piotr Kowalczyk
Linear Combination of Ground Motion Models with Optimized Weights Using Quadratic Programming

The epistemic uncertainty of a prediction model can be reduced by linearly combining independent models with appropriate weights. Linear combination indicates that weights are assigned at each model, where their sum is unity, and weighted models are summed for the final combined model. This paper suggests a framework selecting optimized weights minimizing standard deviation of combined models using the quadratic programming technique. Estimating optimized weights for the combination of two models is straightforward and a mathematical equation can be easily derived. However, finding optimized weights for multiple models is not trivial and a numerical approach such as a grid search technique was often used. The quadratic programming, optimizing a quadratic objective function, can be used to find the optimized weights for multiple models effectively. We applied the quadratic programming to the combination of ground motion models to evaluation the effectiveness of the proposed method. The spectral accelerations from seismic records observed at the downhole sensors in South Korea were used as the dataset. We found that the quadratic programming successfully suggested optimized weights minimizing the uncertainty of the combined model.

Dongyoup Kwak, Dongil Jang, Jae-Kwang Ahn
An Evaluation of Estimates from the P-wave Seismogram Method for Sites in California

The P-wave seismogram method estimates the average shear wave velocity over a representative depth ( $$V_{SZ}$$ V SZ ) from earthquake recordings at a site. The $$V_{SZ}$$ V SZ is computed from the amplitudes of the radial and vertical P-wave arrivals on the earthquake recordings and an estimate of the seismological ray parameter ( $$p$$ p ). The ray parameter is estimated from the depth of the event, the epicentral distance, and the regional crustal velocity model. We evaluated the P-wave seismogram approach to estimating $$V_{SZ}$$ V SZ at 153 seismic recording stations in California for which shear wave velocity profiles are available and tested the effect of different crustal models on the estimated ray parameter and the resulting $$V_{SZ}$$ V SZ . Across all the sites, the estimated $$V_{SZ}$$ V SZ values were, on average, about 24% larger than the measured $$V_{SZ}$$ V SZ , although the difference was negligible for softer sites and as large as 45% at stiffer sites (>1000 m/s). Two crustal velocity models for California were considered: a simplified four-layer crustal model for the entire state and a set of more detailed crustal models used for different parts of the state. The effect of the assumed crustal velocity profile was not significant for earthquake events with focal depths greater than about 3–5 km, but for shallower events the detailed crustal velocity model produced $$V_{SZ}$$ V SZ values significantly smaller than the $$V_{SZ}$$ V SZ from the simplified crustal velocity model due to the effect of the shallow low velocity layers and large gradient on the ray parameter.

Meibai Li, Ellen M. Rathje
The Uncertainty of In-situ S and P Wave Velocity Test at Xichang Experimental Field of CSES

In this study, a blind in-situ wave-velocity experiment was carried out in the Xichang Experimental Field of the CSES (China Earthquake Science Experiment Site). Relying on 286 times of boreholes velocity results, a statistics model was established to describe the test uncertainty of Vs and Vp. Based on the blind test, the uncertainty of the Vs and Vp test does not relay on the objective factors of the field. The relative deviation from the mean of the velocity can be described by Standard Normal Distribution N(0,0.082) at Xichang. Furthermore, when consider the impact of the uncertainty on the site rigidity judgement, it is only need to consider the situation when the overburden depth is less than 10 m and the site is stiff. In conclusion, all these points can be used for the construction of CSES.

Yongbo Liu, Zhuoshi Chen, Xiaoming Yuan, Longwei Chen
Assessment of Spatial Variability of Site Response in Japan

Local soil conditions influence earthquake-induced ground shaking and deformation; a phenomenon known as site effects. The latter correlate with the concentration of damages to the built environment in areas prone to ground motion (GM) amplification. However, the characterization of geotechnical parameters affecting spatially varying earthquake GMs is often oversimplified. Spatial variability of GMs stems from source heterogeneities, varying ray propagation paths, varying local soil conditions, wave scattering, and directivity effects. Regionalization of site effects may also be necessary when specific geologic structures found within a given region control systematic amplifications in the area (e.g., deep sedimentary columns in the US Atlantic and Gulf coastal plains). Evaluating spatially variable GMs is also essential when investigating large, distributed infrastructure, such as water distribution systems. Because soil properties can be spatially correlated at nearby locations, the expected site response will also be spatially correlated. This study focuses on quantifying spatial correlations in site parameters from the Japanese databases, Kyoshin Network (K-Net) and Kiban-Kyoshin Network (KiK-net). Current spatial correlation models for intensity measures are based on correlation length, which neglects for instance, the effects of depositional processes in the spatial distribution of local soil conditions. In this work, we use Kriging to evaluate the significance of the spatial correlation for different site parameters and evaluate potential causes for the observed differences.

Cristina Lorenzo-Velazquez, Ashly Cabas
Ground Deformation Evaluation Using Numerical Analyses and Physics-Based Mega-thrust Subduction Zone Motions

A large earthquake is anticipated for the Pacific Northwest of the United States and western Canada from the Cascadia Subduction Zone (CSZ). The earthquake is expected to cause significant damage to infrastructure, including at Oregon’s Critical Energy Infrastructure (CEI) hub which extends for over 9 km and handles 90% of Oregon’s liquid fuel. There is particular concern about damage to fuel storage tanks from earthquake-induced ground deformations. CSZ earthquake events are infrequent and there do not exist any recorded ground motions. Recently, a suite of broadband synthetic physics-based broadband ground motions was developed by modeling full rupture of the CSZ as part of The M9 Project. The present study estimates likely lateral displacements in the CEI hub using numerical simulations with the M9 ground motions. Two alluvial soil types are modeled: coarse-grained soils with the PM4Sand model and fine-grained soils with the PM4Silt model. Lateral displacements are estimated with one-dimensional (1D) site response analysis combined with strain-potential and empirical procedures, and two-dimensional (2D) nonlinear dynamic simulations. The sensitivity of estimated lateral displacements to ground motion characteristics are presented to evaluate the use of M9 motions for geotechnical seismic studies, and to evaluate the use of 1D and 2D methods with fine-grained soils.

Diane Moug, Arash Khosravifar, Peter Dusicka
VS30 Correlations from Shear Wave Velocity Profiles in the UAE

The time averaged shear wave velocity to 30-m depth (VS30) is used for seismic hazard assessment for National Earthquake Hazard Reduction Program (NEHRP) site classification. There are several models to predict VS30 from average shear wave velocity to a depth (z) less than 30 m (VSz). This study evaluates the prediction capability of the existing models in the United Arab Emirates (UAE). The analyses show that there is a significant prediction biases in the existing models developed in different regions. By reviewing borehole data from the selected sites, it reveals that these biases are mainly because of the characteristic geology of the UAE where weak bedrock is encountered at shallow depths. This bedrock mostly shows a considerable variation in both clay content and the porosity which also have a direct effect on the prediction uncertainty.

Deepa Kunhiraman Nambiar, Tadahiro Kishida, Tareq Fahmy Abdallatif, Mohammad H. AlHamaydeh
Insight into the Relationship Between Dynamic Shear Strain and Vibration Velocity of Horizontally Layered Sites

Equivalent linear method in frequency domain, represented by SHAKE2000, is the mainstream approach for seismic response analysis of soil layers. Due to the seriously unreasonable results in soft soil sites, its improvement becomes a research hotspot, which mainly adopts frequency-dependent method (FDM), but there has been no substantial effect. In this paper, the assumption of dynamic shear strain and vibration velocity behave a constant proportional relationship in FDM was studied. Based on wave motion equation, exact solutions for the relationship of the two variables were derived on one-dimensional horizontally layered sites. The reasonability and deviation degree of the assumption were studied through numerical experiments. Results show that the constant proportional hypothesis only holds in the case of unbounded homogeneous medium with one-way traveling wave. For actual layered soil sites, the relationship between dynamic shear strain and vibration velocity strongly depends on the frequency of the wave and the location of the observation point. If we ignore the reflection wave in ground seismic response analysis, the use of constant proportional assumption under one-way traveling wave will make significant deviation to the results. The deviation caused by the constant proportional relationship assumption may span four order ranges. Even for a single uniform deposit, the deviation is also very significant. For practical ground motion calculation, this assumption clearly exists qualitative error in theory, and the quantitative deviation can not be acceptable.

Qi Xia, Rui-shan Li, Xiao-ming Yuan
Investigating the In-Situ Properties of Poisson’s Ratio Based on KiK-Net Data

Poisson’s ratio is an important parameter in the engineering field. However, most related studies have been performed using laboratory test or in-situ test, the studies based on seismic observations are limited. In this study, the methods for estimating the site Poisson’s ratio and its variation with shear strain based on HVSR are proposed combined with the theories of wave propagation and one-dimensional site response analysis. Then, the in-situ properties of Poisson’s ratio are investigated by applying the proposed method to KiK-net seismic database. It is observed that the shear strain change threshold of the site Poisson’s ratio is around 10–5, and the maximum Poisson’s ratio is smaller than 0.5, without the effects of groundwater. The results of this study are verified by the comparison with the laboratory test results, and they can be utilized as the reference to the engineering practice and related studies.

Yang Shi, Hao Zhang, Yu Miao
Effectiveness of Distributed Acoustic Sensing for Acquiring Surface Wave Dispersion Data Using Multichannel Analysis of Surface Waves

Distributed acoustic sensing (DAS) is a rapidly expanding tool to sense vibrations and system deformations in many engineering applications. In terms of site characterization, DAS presents the ability to make static and dynamic strain measurements on a scale (e.g., kilometers), density (e.g., meter-scale), and fidelity (e.g., microstrain) that was previously unattainable with traditional measurement technologies. In this study, we assess the effectiveness of using DAS to extract surface wave dispersion data using the multichannel analysis of surface waves (MASW) technique. We utilized both highly-controlled, broadband vibroseis shaker truck and more-variable, narrow-band sledgehammer sources to excite the near surface and compared the DAS-derived dispersion data directly with concurrently acquired traditional geophone-derived dispersion data. We report that the differences between the two sensing approaches are minimal and well within the uncertainty bounds associated with each individual measurement for the following DAS testing conditions: (a) a tight-buffered or strain-sensing fiber optic cable is used, (b) the cable is buried in a shallow trench to enhance coupling, and (c) short gauge lengths and small channel separations are used. Our deployed conditions are more promising than previous attempts documented in the literature, thereby demonstrating that DAS can provide accurate measurements of surface wave dispersion data of the same quality as geophones. We show that frequency-dependent normalization of the dispersion image removes the effects of scaling, integration, and differentiation of the measured data, thereby removing the need to post-process the geophone-derived and DAS-derived waveforms into equivalent units before performing dispersion processing. Finally, we summarize the important effect of gauge length on the dispersion data for future reference. This study demonstrates that DAS, when appropriate considerations are made, can be used in-lieu of traditional sensors (i.e., geophones) for making high-quality measurements of surface wave dispersion data using the MASW technique.

Joseph P. Vantassel, Brady R. Cox, Peter G. Hubbard, Michael Yust, Farnyuh Menq, Kyle Spikes, Dante Fratta
Relationships Between Ground-Motion Intensity Measures and Earthquake-Induced Permanent Slope Displacement Based on Numerical Analysis

In engineering applications, the permanent displacement (D) commonly serves as a useful indicator of the seismic performance of slopes. When developing empirical displacement models as a function of ground-motion intensity measures (IMs), the IMs that are best correlated to D are preferred. On the other hand, the predictability of IMs, in terms of the standard deviations using ground motion models, is also of concern in developing D models. This study aims to: (1) investigate the efficiency of IMs in developing D models for a cohesive-frictional slope based on numerical analysis; and (2) compare the means and standard deviations of randomized D by considering uncertainties in predicting both the IMs and D via Monte Carlo simulation (MCS). A total of 10 scalar IMs and 38 vector-IMs, are employed to develop D models. The results indicate that the spectral acceleration at a degraded period of the soil layer (SA(1.5Ts,layer)) and Arias intensity (IA) are the two most efficient scalar IMs. Additionally, the vector-IMs consisting of [IA, spectrum intensity] and [IA, mean period] are the two most efficient vectors. The MCS results illustrate that the rankings for standard deviations of D models and total standard deviations (i.e., including ground motion variability) may be considerably different. The results are also found to be dependent on earthquake magnitudes and site conditions. This study could provide guidance on the development of numerical-based D models especially within a probabilistic seismic slope displacement analysis framework.

Mao-Xin Wang, Dian-Qing Li, Wenqi Du
Accuracy of Complex Moduli in Seismic Response Analysis of Ground

The complex moduli used for the seismic response analysis of ground, the Sorokin model used in original SHAKE, the Lysmer model proposed to improve SHAKE, and the YAS model proposed by the authors are discussed. The Sorokin model overestimates maximum stress. The Lysmer model shows maximum stress the same as the test but underestimates the damping ratio. The YAS model gives the maximum stress and the damping ratio the same as the test result. The case study using 269 ground and 10 observed strong earthquake motions in Japan shows that the Sorokin model overestimates PGA about 15% at maximum compared with the YAS model. Overestimation of PGA by the Lysmer model is as small as 3% or less, although it is theoretically incorrect.

Nozomu Yoshida, Kenji Adachi
Scattering of Incident Plane Waves by Underground Cylindrical Cavity in Unsaturated Poroelastic Medium

At present, the study of the dynamic characteristics of unsaturated porous media is still in its infancy. Therefore, it is particularly important to study the dynamic characteristics of cavity in unsaturated media, which is similar to the actual situation of cylindrical underground structures such as tunnels are encountered with seismic wave. Based on the dynamic theory of unsaturated porous media considering the mixture theory, analytical solutions are presented for two-dimensional of scattering of plane P-wave and SV-wave by cylindrical cavity by using wave function expansion method. The validity of the solutions is confirmed by comparing with the one in which the infinite medium is assumed as a pure elastic medium. A detailed parametric study is presented to illustrate the effect of saturation, Poisson’s ration, and porosity of the unsaturated poroelastic medium on the dynamic stress concentration factor around the cavity is analyzed.

Aichen Zhang, Weihua Li, Fengcui Feng
Dynamic Soil-Structure Interaction for a SDOF with Rigid Foundation Embedded in a Radially Inhomogeneous Bearing Stratum Under SH Waves

Dynamic soil-structure interaction will occur when the seismic wave passes through a soil-structure system. It involves the scattering of incident wave by the foundation of superstructure, the transfer of incident wave energy to the superstructure and the radiation of structural vibration energy back into the soil. In this kind of wave process, the local soil stratum condition of the bearing stratum has an important role, especially for those engineering sites with either sedimentation- or load-induced inhomogeneous shear modulus profile. In this study, an analytical model of dynamic interaction between inhomogeneous bearing stratum and superstructure in bedrock half space under plane SH wave is established. The shear modulus of the inhomogeneous bearing stratum is assumed to vary with radius in a power-law manner, and the superstructure adopts the classical single degree of freedom (SDOF) oscillator model. By using a wave function expansion method, an analytical series solution for the foundation displacement and relative displacement of superstructure is derived. The influence of the inhomogeneous bearing stratum on the system response is studied. When the inhomogeneous bearing stratum is more flexible than the bedrock half space, the peak value of foundation displacement is obviously larger than that without considering the inhomogeneous bearing stratum. With the decrease of the stiffness of the superstructure, the peak frequency of the system response shifts to the low frequency.

Ning Zhang, Xinyu Sun, Haijun Lu, Denghui Dai
Ground Motion Amplification by a Rectangular Tunnel in a Saturated Poroelastic Half-Space

The ground motion around underground tunnels can vary significantly from the free-field site response due to wave scattering, thus influencing the seismic demand for the design of aboveground structures in the vicinity. This paper numerically investigates the effects of a rectangular tunnel on the ground motion of a saturated poroelastic half-space for obliquely incident seismic waves. The results show that the ground surface accelerations can be significantly modified by the rectangular tunnel, while the modification is a function of the soil property. It is shown that the differences in the ground motion amplification between the saturate and dry poroelastic half-spaces are noticeable, both in terms of the magnitude and period contents of the ground surface accelerations, and moreover, the saturation degree is a critical factor that can influence the ground motion amplification to a great degree.

Jun Zhu, Xiaojun Li, Jianwen Liang, Mianshui Rong

Foundations and Soil-Structure Interaction

The Effect of Soil Damping on the Soil-Pile-Structure Interaction Analyses in Liquefiable and Non-liquefiable Soils

Preliminary design of piles is performed by considering the static loads, but the final design must include the dynamic loads, especially in earthquake-prone regions. The soil nonlinearity under seismic loading is evaluated using the modulus degradation curves in the total-stress approach. In this study, two different centrifuge tests were simulated in FLAC3D. The nonlinear elastic method (hyperbolic model) and the elastoplastic Mohr-Coulomb (MC) model were employed in the study. The soil-single pile-structure systems were analyzed under the specific earthquake events, and the soil-pile-structure response was compared. The analyses with the low-intensity input motions show that the superstructure accelerations and the bending moments in the single pile are estimated with reasonable accuracy. However, the superstructure accelerations might be underestimated, especially in the MC model, compared to the centrifuge test results due to the increase in the amplitude of the input motion. The low accelerations can be attributed to the high damping ratios in the perfectly plastic constitutive model. Although the nonlinear elastic model is less complex, closer results might be obtained since the more realistic damping ratios are implemented. The results show that even the less elaborate models, such as the hyperbolic model with an indefinite failure criterion, might give reasonably accurate results in the total-stress approach, thanks to the limited damping ratios. As a result, the responses of the superstructure and the pile in soil-pile-structure interaction problems are highly dependent on the soil damping; in turn, the due account must be given to the selection of the constitutive model.

Ozan Alver, E. Ece Eseller-Bayat
Bearing Capacity of Shallow Strip Foundations Adjacent to Slopes

In this paper the static and seismic bearing capacity factors for shallow strip foundations adjacent to slopes have been evaluated using the method of characteristics, extended to the seismic case by means of the pseudo-static approach. Bearing capacity factors have been evaluated for different values of the slope angle and, under seismic conditions, accounting for the effect of inertia forces arising in the foundation soil and transmitted by the superstructure. These effects are dealt with independently and it was also demonstrated that they can be superimposed without significant error.The proposed solutions, obtained assuming Hill’s and Prandtl’s failure mechanisms, have been checked against those obtained through finite element analyses and compared with results already available in the literature.Closed-form solutions and an empirical formula have been provided to evaluate corrective coefficients accounting for the effect of the sloping ground on bearing capacity that, under static conditions, allow a straightforward application of the bearing capacity trinomial formula.

Orazio Casablanca, Giovanni Biondi, Giuseppe Di Filippo, Ernesto Cascone
Seismic Response of Anchored Steel Sheet Pile Walls in Dry and Saturated Sand

Anchored Steel Sheet Pile (ASSP) walls are complex systems whose behaviour during an earthquake depends on the interaction between the soil, the anchor and the wall. The design of ASSP retaining walls is frequently carried out using pseudo-static methods, where the earthquake-induced inertial forces are represented using an equivalent pseudo-static coefficient that is constant in time and space, usually estimated neglecting any effects of soil-structure interaction, whereas their permanent displacements are often estimated using pseudo-dynamic approaches based on Newmark’s sliding block method. However, ASSP walls are complex systems and understanding their seismic response is necessary to determine the extent to which the design assumptions of pseudo-static methods are acceptable and the extension of Newmark’s method to the prediction of their seismic-induced permanent displacements is sound. This paper presents the results of two dynamic centrifuge tests on reduced scale models of ASSP walls in dry and saturated uniform medium dense sand performed on the Turner Beam Centrifuge at the Schofield Centre, University of Cambridge. Both trains of sinusoidal waves and realistic earthquake motions were applied to the base of the models. Digital image correlation was used to measure the displacements of a cross-section of the models during each applied earthquake. In dry sand, the largest amplification of the horizontal accelerations was observed near the retaining wall, and the main wall accumulated significant outward rotation with limited horizontal displacement of the toe. In saturated sand, the generation of excess pore pressure led to significant de-amplifications and phase lag in the accelerations, and caused larger permanent displacements of the structure.

Alessandro Fusco, Giulia M. B. Viggiani, Gopal S. P. Madabhushi, Riccardo Conti, Cécile Prüm
Response of Suction Bucket Foundation Subjected to Wind and Seismic Loading During Soil Liquefaction

Offshore wind turbines (OWTs) are now being constructed in seismic areas as part of a plan to achieve net-zero carbon dioxide emissions. Liquefaction caused by earthquakes poses a serious threat to the operation of OWTs. The dynamic response of the OWT on the suction bucket foundation in liquefiable sand under earthquakes combined wind loads was investigated in the present study. The FE-FD software named DBLEAVES-X with the Cyclic Mobility constitutive model was used in the research. The nonlinear dynamic analyses were carried out using a three-dimensional model of the OWT. The effects of earthquake, the wind load, sand density, and aspect ratio of the suction bucket on the behavior of the OWT were studied. The findings show that OWTs will be permanently tilted and displaced caused by wind, earthquake, and liquefaction. In seismic areas, this numerical model could be used in a comparative sense for the design of the OWT.

Bin Gao, Guanlin Ye, Qi Zhang, Wenxuan Zhu
A Class of Thermodynamic Inertial Macroelements for Soil-Structure Interaction

The seismic performance of structures can be significantly influenced by the interaction with the foundation soils, with effects that depend on the frequency content and the amplitudes of the ground motion. A computationally efficient method to include these effects in the structural analysis is represented by the macroelement approach, in which a geotechnical system is modelled with a single macroelement that describes the generalized force-displacement relationship of the system. While this method has been mainly developed for shallow foundations, the present study proposes a class of macroelements representing the macroscopic response of different foundation types, including abutments, piled and caisson foundations. The generalized force-displacement relationships for these models are elastic-plastic and are derived using a rigorous thermodynamic approach. The plastic responses of the macroelements are bounded by the ultimate capacities of the geotechnical systems, while the inertial effects associated with the soil mass involved in the dynamic response of the structure are simulated by introducing appropriate participating masses. The macroelements are implemented in OpenSees; in this paper they are applied to assess the seismic performance of a tall viaduct showing highly nonlinear features.

Davide Noè Gorini, Luigi Callisto
Simulation Analyses of Centrifuge Model Tests on Piled Raft Foundation with Deep Mixing Walls

Three-dimensional finite element analyses were conducted to simulate the dynamic centrifuge model tests on piled raft foundation with deep mixing walls (DMWs). The DMWs is one of the most efficient liquefaction countermeasure methods. After the shaking model tests, the DMWs were observed that several cracks were locally induced in the wall. However the DMWs were quite effective at reducing the sectional force of the piles to an acceptable level under the large earthquake load. The stresses in the DMWs, which was difficult to measure in the tests, were investigated by the numerical analyses. The shear stresses on the DMWs in the analyses were almost consistent with the inertia force of the structure similar to the experimental results. In addition, tensile stresses of about 1.5 MPa were generated, which were more than 0.2 times higher than an unconfined compressive strength of 6.33 MPa, confirming the experimental results of cracking in the DMWs. The model test results were reproduced by the numerical analyses considering the separation from the ground around the pile and at the base of the foundation and the sliding.

Junji Hamada, Takehiro Okumura, Yoshimasa Shigeno, Yoshihiro Fujiwara
Seismic Interactions Among Multiple Structures on Liquefiable Soils Improved with Ground Densification

Current guidelines for evaluating the performance of ground densification as a liquefaction countermeasure near buildings are based on free-field conditions. However, particularly in urban areas, where structures are constructed in the vicinity of each other, structure-soil-structure interaction in liquefiable deposits near two adjacent buildings (SSSI2) and multiple buildings (≥3) in a building cluster (SSSI3+) have been shown to be consequential on key building engineering demand parameters (EDPs). Yet, their potential tradeoffs associated with ground densification are currently not well understood. In this paper, three-dimensional, fully-coupled, nonlinear, dynamic finite element analyses, validated with centrifuge models of SSSI2, were used to explore the influence of ground densification and the building spacing to width ratio (S/W) on key EDPs of buildings experiencing SSSI2 or SSSI3+ (particularly for four structures in a square-shape configuration). For the conditions considered, ground densification reduced the permanent settlement of an isolated structure by up to 58% compared to a similar unmitigated structure. Both SSSI2 and SSSI3+ had a minor impact on the mitigated structures’ average settlement compared to SSI. In contrast, SSSI2 and SSSI3+ strongly amplified the permanent tilt of the mitigated structure by up to 6.5 times compared to SSI at S/W < 0.5, due to the enhanced degree of asymmetry in the properties of and demand on the underlying soil. Increasing spacing reduced the permanent tilt of structures under SSSI2 or SSSI3+. Overall, the results suggest that SSSI2 and SSSI3+ can adversely affect the foundation performance and hence, damage potential of the mitigated soil-foundation-structure system compared to SSI, particularly at S/W < 0.5. Such complexities need to be considered in mitigation planning and design of urban structures.

Yu-Wei Hwang, Shideh Dashti, Juan Carlos Tiznado
Numerical Simulation of Real-Scale Vibration Experiments of a Steel Frame Structure on a Shallow Foundation

Dynamic Soil-Structure-Interaction (DSSI) phenomena can considerably affect the structural response under dynamic loading. Time domain finite element analysis allows to study these phenomena in depth, but several computational challenges need to be addressed first to achieve rigorous modelling of both the structure and soil domain. Within this context, this study presents three-dimensional FE analyses of real-scale forced vibration tests of a steel frame structure founded on a shallow foundation. The field tests were carried out with the real-scale prototype structure of EUROPROTEAS at the Euroseistest experimental facility located in the Mygdonian Valley in Northern Greece. The structure has outer dimensions of 3 × 3 × 5 m and it is placed in an area whose geotechnical properties have been well documented by previous studies. The study focuses on the modelling of the soil-foundation interface, which is one of the major challenges of such complex SSI simulations. A novel approach to model potential contact imperfections at the interface is proposed, showing very good agreement with the field data and hence improving the reliability of numerical predictions. The analyses show that contact imperfections affect considerably the predicted motion in both the structure and the soil.

Marios Koronides, Stavroula Kontoe, Lidija Zdravković, Athanasios Vratsikidis, Dimitris Pitilakis, Anastasios Anastasiadis, David M. Potts
Experimental Behavior of Single Pile with Large Mass in Dry Medium Sand Under Centrifuge Shaking Table Test

In this study, a single pile centrifuge shaking table model tests was conducted. The foundation soil was dry medium sand with 50% relative density which was made of pouring method. In prototype, the pile diameter was 1.0 m, and a mass of 312.5 ton was fixed on pile head which was 5 m above the ground surface. The strain gauges were uniformly instrumented along the pile to measure the bending moment response of the pile, the accelerometers were instrumented to obtain the acceleration response of soil and pile. The input seismic excitation was the sinusoidal wave of 0.3 g. The results show that: the obvious acceleration amplification effect was observed in the soil above 15 m; the maximum bending moment of pile occurred in 1.65 m depth, and the residual bending moment in the lower part of the pile was observed.

Longyu Lu, Chunhui Liu, Mengzhi Zhang, Tiqiang Wang
Numerical Investigation on Dynamic Response of Liquefiable Soils Around Permeable Pile Under Seismic Loading

Permeable piles have been widely adopted as a new anti-liquefaction treatment measure in the liquefiable site. The dynamic numerical analysis to study the anti-liquefaction effects of liquefiable soils around the permeable pile under the seismic wave was carried out to estimate the excess pore water pressure ratio (EPWPR), pile settlement, and the soil settlement. Numerical results show that the EPWPR has exceeded the critical level of liquefaction at 2 m far from the pile at the top surface of the liquefiable soil layer. The drainage effect of the soil around the pile weakens as the depth of the soil layer increases, that is, the shear stress and strain are being strong rapidly. Additionally, the effect of pile permeability on dynamic responses of liquefiable soil indicates that the increase of pile permeability will considerably weaken dynamic responses. Furthermore, the failure of liquefaction never takes place in the site if the permeability ratio of the pile to that of soil exceeds the critical value of 10.

Chi Ma, Guo-Xiong Mei, Jian-Gu Qian
Rotation of a Cantilevered Sheet-Pile Wall with Different Embedment Ratios and Retaining a Liquefiable Backfill of Various Relative Densities

Damage to retaining structures at the waterfront during major earthquakes can be attributed to the development of positive excess pore water pressure in the retained or foundation soils, leading to liquefaction. As part of Liquefaction Experiments and Analysis Project (LEAP), dynamic centrifuge tests of a cantilevered sheet-pile wall model floating in dense sand and retaining a liquefiable backfill were carried out. The embedment ratio (embedment depth / excavation height) and the relative density (Dr) of the backfill were varied to observe the rotation of the sheet-pile during a tapered 1 Hz sinusoidal input base motion. The Dr of the backfill affected the wall rotation during gravity loading and the initial static shear stress in the backfill; smaller Dr backfill led to a larger wall rotation and initial static shear. During seismic loading, the model with a higher initial static shear had a dilative pore pressure response which increased the wall stability. On the other hand, the model with a smaller initial static shear showed a contractive response leading to initial liquefaction in the active failure wedge, which resulted in a large increase of wall rotation. Moreover, the model with a higher embedment ratio had a smaller wall rotation during initial loading cycles, but the wall rotation increased rapidly after the soil in front of wall softened due to positive excess pore water pressure buildup. These results showed the importance of considering the initial static shear stress prior to earthquake loading and the vulnerability of wall stability due to excess pore pressure increase in front of the wall.

Satish Manandhar, Seung-Rae Lee, Gye-Chun Cho
Foundation Alternatives for Bridges in Liquefiable Soils

Ground failure during major seismic events associated with soil liquefaction can lead to major structural damage in both the columns and bridge upper deck, due to large seismic-induced displacements in the support foundation. Liquefaction driven ground motion incoherence during the dynamic event, and permanent soil deformations are key variables in the observed damage. This paper summarizes a numerical study of an alternative bridge foundation design proposed to reduce support displacements during and after an earthquake, as well as relative settlement associated with partial loss of bearing capacity when the bridge column is founded on a potential liquefiable layer. Three dimensional numerical models were developed with the program FLAC3D. The seismic environment was characterized by a uniform hazard spectrum, UHS, developed for a nearby rock outcrop, considering a return period of 1000 years. Initially, a one-dimensional analysis was performed with the software SHAKE, to evaluate the liquefaction susceptibility. Later, to consider changes in both topography and ground subsurface layering of the site, a two-dimensional model was developed with the program QUAD4M. Based on the results gathered in here, it was concluded that structured cell foundation is a sound alternative to improve the seismic performance of bridges located in liquefiable soils and allows reducing detrimental effects associated with liquefaction-induced ground deformations.

Juan Manuel Mayoral, Daniel De La Rosa, Mauricio Alcaraz, Nohemi Olivera, Mauricio Anaya
A Case Study of Seismic Design of Pile Foundation Subject to Liquefaction, Cyclic Softening, and Lateral Spreading

Pile foundations are frequently subjected to the lateral load imposed by soil movement during and after an earthquake event. This lateral load transfer from the soil to the pile is a complex soil-structure interaction problem that involves an iterative process and requires effective communication between structural and geotechnical engineers. Several methodologies for the design of laterally loaded pile foundations have been proposed by researchers. However, limited practical research papers are available which explain the design procedure of piles subject to lateral ground movement due to liquefaction and cyclic softening. This paper presents the details of the methodology that was implemented for a structure located on an area subject to the risk of kinematic ground movement due to cyclic softening and liquefaction during an earthquake event and lateral spreading after the earthquake shaking. The foundation comprises 82 piles in very soft cohesive soil. Separate design scenarios representing the ground and structure behavior at different stages of earthquake shaking have been considered. Structural software ETABS is used to check the buckling load capacity of the pile-supported laterally by soil springs.

Yasin Mirjafari, Malcolm Stapleton
Dynamic Centrifuge Model Tests on Plate-Shaped Building Supported by Pile Foundation on Thin Load-Bearing Stratum Overlying Soft Clay Layer

In Japan, many plate-shaped buildings are built as residences in coastal areas. As a result, some of them are sustained by the pile foundation embedded on the thin load-bearing stratum overlying the soft clay layer because a continuous load-bearing stratum is too deep to construct economically. During a large earthquake, plate-shaped buildings cause large lateral force and varying axial force in their pile foundations and make their seismic behavior more complicated because they each have a high aspect ratio and single-span composed of two columns and several beams. In addition, the thin load-bearing stratum overlying soft clay layer makes the seismic behavior of a plate-shaped building more complex because of the less stiffness of soft clay. In this study, two dynamic centrifuge model tests in a 50 g field were carried out to compare the seismic behavior of the plate-shaped building and pile foundation on the thin load-bearing stratum overlying soft clay layer with that on the continuous load-bearing stratum. From the several shaking tests, it was observed that the large instantaneous acceleration of the superstructure occurred when the gap between the pile tip and the load-bearing stratum was closed in the model on the continuous load-bearing stratum. On the other hand, the bending moment at the pile head in the thin load-bearing stratum was almost equal even though the inertial force differed. However, the pile foundation on the thin load-bearing stratum did not settle critically, although the peak axial force was almost double the preload.

Takehiro Okumura, Junji Hamada
The Behaviour of Low Confinement Spun Pile to Pile Cap Connection

An experimental and numerical study of the low confinement spun pile to pile cap connection typically used in Indonesia was performed. A cyclic loading test on two spun pile connections was performed. One was an empty spun pile (SPPC01) and another was filled with reinforced concrete (SPPC02). No shear failure was observed during the test. Filling the reinforced concrete into the spun pile increase the capacity by 54% and also improve the ductility from 4.26 to 4.70. A numerical study was performed to simulate the effect of the soil. A similar spun pile with a fixed connection to the pile cap was embedded in clay. Soil affects the strength and ductility of the connection, but it is not significant. Stiffer soil results in higher bending capacity. The capacity could be approximated based on the P-M interaction. To conclude, the connection of the spun pile to pile cap perform well though the spun pile has insufficient confinement.

Mulia Orientilize, Widjojo Adi Prakoso, Yuskar Lase, Carolina Kalmei Nando
Rocking Pilegroups Under Seismic Loading: Exploring a Simplified Method

Recent research on the rocking response of bridges founded on pilegroups has demonstrated the potential benefits of allowing full mobilization of pile bearing capacity during seismic loading. Especially for the retrofit of existing bridges, allowing strongly nonlinear foundation response may allow avoiding foundation retrofit, which can be a challenging and costly operation. However, this calls for 3D numerical analysis of the bridge–foundation system with adequately sophisticated constitutive models, which can be time-consuming. To promote practical application of such performance-based design philosophy, this study explores the efficiency of a simplified analysis method, where the pilegroup is replaced by an assembly of nonlinear (rocking) and linear (horizontal and vertical) springs and dashpots. Such a simplified approach has been shown to offer reasonable predictions for rocking shallow foundations, and it is therefore of interest to explore its applicability to pilegroups. A bridge pier founded on a 2 × 1 pilegroup in homogeneous clay is chosen as an illustrative example. The bridge-foundation system is initially modelled with 3D finite element (FE) analysis, accounting for soil and structural nonlinearity of the reinforced concrete (RC) piles. The system is subjected to monotonic and cyclic pushover loading, based on which the nonlinear rotational spring and the corresponding dashpot are calibrated. The simplified model is subjected to dynamic time history analyses and compared to the full 3D model. The comparison yields promising preliminary results, but also offers insights on the reasons of the observed discrepancies and the unavoidable limitations of such simplified technique.

Antonia Psychari, Saskia Hausherr, Ioannis Anastasopoulos
Remediation of Structure-Soil-Structure Interaction on Liquefiable Soil Using Densification

Earthquake induced soil liquefaction has drawn significant attention in urban areas where buildings are closely spaced due to the extensive damage it has caused on buildings in recent earthquakes. This study investigated the effect of targeted soil densification as a remediation method on one of a pair of adjacent buildings with different storey levels on liquefiable soil using dynamic centrifuge modelling. The effects of remediation on structural and foundation performance of both treated and untreated buildings are presented. It was demonstrated that localised soil densification can be effective and beneficial in treating detrimental foundation behaviours of the treated building on the liquefiable soil and an untreated adjacent building can also be beneficially affected. From a structural perspective, soil densification can reduce the structural demand on the treated building with negligible detrimental effect or even some beneficial effect on the untreated building on the liquefiable soil. In the case studied, localised soil densification when building new structures adjacent to existing ones can be an effective remediation method in improving seismic performance on liquefied soil. This study also suggests that the decision of remediating any building in an urban setting should be made after considering effects of SSSI.

Shengwenjun Qi, Jonathan Adam Knappett
Seismic Responses Analysis on Basements of High-Rise Buildings Considering Dynamic Soil-Structure Interaction

Soil-structure interaction is a research hotspot in the geotechnical and civil engineering, especially with the further accelerated development of urban high-rise buildings. Unlike previous studies focusing on the superstructures, this paper presents a numerical study on the seismic response characteristics of underground structures, i.e., the basements of high-rise buildings. To this end, a comprehensive model considering soil-basement-superstructure system was analyzed firstly, and the dynamic response of basement was divided into the kinematic interaction (KI) effect and inertial interaction (II) effect based on a theoretical framework. Then, a typical 25-story symmetric wall-frame structure with a 3-story basement was selected as the case to verify the founds obtained in theoretical analysis. The results demonstrate that: 1) Unlike the single underground structure, the dynamic response of basement would be significant amplified by the inertial interaction effect of superstructure (averagely enlarger by 1.524 times); and 2) There is a larger KI effect on basement whetever the superstructure exists or not. Therefore, it is necessary for seismic design of building basements to comprehensively consider the KI and II effects. In addition, this paper investigates the possibility on acquiring the KI and II effects with simplified methods according to the characteristics of two effects.

Yan-Jia Qiu, Hong-Ru Zhang, Zhong-Yang Yu
Centrifuge and Numerical Simulation of Offshore Wind Turbine Suction Bucket Foundation Seismic Response in Inclined Liquefiable Ground

With the rapid development of offshore wind farms, suction bucket foundation is gaining attention due to installation convenience, cost efficiency, and reusability. Currently, increasing number of offshore wind turbines are built in seismically active areas, the seismic stability of bucket foundation is becoming a major concern, especially in liquefiable soil. By comparing the results of centrifuge shaking table test with solid-fluid coupled dynamic finite difference analysis, the seismic response of suction bucket foundations in liquefiable ground is analyzed. A unified plasticity model for the large post-liquefaction shear deformation of sand is used to provide high fidelity representation of the behavior of saturated sand under seismic loading. The soil-bucket-superstructure dynamic interaction mechanism is investigated, via analysis of the rotation and translation of the bucket foundation and the acceleration and displacement of the superstructure. The results show that excess inclination angle of the bucket foundation and amplified turbine response during the earthquake in the inclined liquefiable foundation should be considered in design.

Xue-Qian Qu, Rui Wang, Jian-Min Zhang, Ben He
Unconventional Retrofit Design of Bridge Pile Groups: Benefits and Limitations

This paper investigates an example problem of bridge retrofit due to widening, focusing on the foundation. Conventional foundation retrofit is compared to an unconventional “do-nothing” approach, where the existing pile group is maintained, allowed to fully develop its moment capacity. The two systems are comparatively assessed employing the finite element (FE) method, accounting for all sources of material and geometric nonlinearities. The soil is modelled with the hypoplastic model for sand, while the Concrete Damaged Plasticity (CDP) model is used for all reinforced concrete (RC) members (piles and bridge pier). The soil-pile interface is modelled with frictional interface elements. Both foundation systems (retrofitted and existing) are initially subjected to pushover loading to comparatively assess their moment capacities. Subsequently, the widened bridge with and without foundation retrofit is subjected to dynamic time history analysis using excitations representative of Swiss seismicity. The performance of the two approaches is comparatively assessed, considering structural and geotechnical performance criteria. The paper provides useful insights regarding the benefits and limitations of unconventional retrofit design of piled foundations.

L. Sakellariadis, S. Alber, I. Anastasopoulos
Shaking Table Tests on Level Ground Model Simulating Construction of Sand Compaction Piles

The sand compaction pile (SCP) method is a ground improvement method that increases the density of the ground by enlarging the diameter of the sand pile through penetration and repeated withdrawal/re-driving of the casing pipe. To confirm the improvement effect of the SCP method by model tests, it is desirable to reproduce the actual construction as accurately as possible, taking into account the effect of stress history during construction. The authors compacted horizontal model ground using a newly-developed sand pile driving apparatus that simulates the static SCP construction, and conducted shaking table tests on the compacted ground. In these tests, the effect of casing rotation for the ground behavior during sand pile driving was investigated, and the behavior of the ground models was measured by the changes in the driving force, the earth and water pressure gauges installed into the ground models when performing ground improvement with the SCP driving system. The relationship between excess pore water pressure and acceleration during shaking was investigated after improvement. From these measurements, it was confirmed that the driving force and earth pressure were smaller and the ground was denser in the model compacted with rotation of casing than in the model without rotation. The behavior of the improved model ground, which was made to simulate the real construction process, was studied based on the measured records of acceleration, pore water pressure, and settlement. The results showed that the pore water pressure and the amount of settlement were larger in the order of the unimproved ground, the improved ground without rotation of casing, and with rotation. In addition, a simple dynamic cone penetration test was conducted as a sounding test before and after shaking table tests, and from the results, the ground conditions before and after shaking and the improvement effect were confirmed.

Hiroshi Yabe, Junichi Koseki, Kenji Harada, Keiichi Tanaka
3D Numerical Lateral Pushover Analysis of Multiple Pile Group Systems

In the development of a performance-based design approach for pile groups, it is important to identify aspects that may have significant effects. To achieve that objective, a series of 3D numerical lateral pushover analyses of pile groups is performed, and the results are subsequently evaluated and synthesized. The effects of pile structural capacity, soil conditions, and foundations systems (single pile groups and multiple pile group systems) are examined explicitly. The highlighted observations include: 1) the structural performance criteria of piles as structural elements alone may not be directly applicable for pile group systems, 2) the performance criteria for the pile group systems would need to consider explicitly the soil conditions, 3) the behavior of pile groups in softer soils appears to be displacement-controlled, while that in stiffer soils appears to be force-controlled, and 4) the progression of plastic hinge development would not most likely lead to a sudden decrease in the pile group system stiffness.

Amelia Yuwono, Widjojo A. Prakoso, Yuskar Lase
Responses of Adjacent Building Pile to Foundation Pit Dewatering

In water-rich area, foundation pit pumping has adverse effects on adjacent buildings. In this study, a series of numerical simulations are carried out, based on a practical dewatering test in an excavation for metro station in Tianjin, to investigate the characteristics of near building pile deformations induced by dewatering. The effect of different pumping depth (Hd) on the pile axial force and lateral friction are revealed. The results indicate that the axial force of the pile foundation experiences a substantial increment; accordingly, the pile shaft resistance becomes negative in the same depth range, and the neutral point appears at the position of 1/2 times the pile length from the pile crown. The research results can provide guidance to the safety assessment of the existing building pile during pumping of adjacent foundation pit.

Chao-Feng Zeng, Hai-Yu Sun, Hong-Bo Chen, Xiu-Li Xue, Yun-Si Liu, Wei-Wei Song
Dynamic Interaction Between Adjacent Shallow Footings in Homogeneous or Layered Soils

Seismic design of buildings is generally carried out without accounting for the interaction between adjacent structures through the underlying soil. The increasingly growth of urbanization in modern metropolitan areas or the urgent need for seismic requalification of historical centres with very closely-spaced structures require the Structure-Soil-Structure Interaction problem to be properly investigated and quantified. The paper tries to shed lights on cross interaction phenomena arising between two identical rigid shallow foundations excited by harmonic loads. Through a 3D continuum approach, solved by the finite difference method, the stiffness matrix of a footing in presence of a neighbouring one was obtained. Different values of foundation-foundation spacing and subsoil type (halfspace or layered soil deposit) were considered.

Enza Zeolla, Filomena de Silva, Stefania Sica
Effects of Nonliquefiable Crust on the Seismic Behavior of Pile Foundations in Liquefiable Soils

Pile foundations located in liquefiable soil deposits are vulnerable to liquefaction-induced damage during earthquakes. The constraint conditions of both the pile head and pile tip are vital for the dynamic response of a pile. In this paper, a validated three-dimensional finite element model, where a PressureDependMultiYield constitute model based on multisurface plasticity theory is utilized for soil, is established to study the seismic behavior of a single pile foundation in liquefiable ground with variable nonliquefiable crust thicknesses. The effects of the properties of the nonliquefiable crust and pile head fixity on the dynamic behavior are investigated. Analysis shows that the existence of a nonliquefiable crust significantly affects the behavior of end-bearing piles and floating piles. The effect of a pile head constraint varies with the properties of the nonliquefiable crust.

Gang Zheng, Wenbin Zhang, Haizuo Zhou
Numerical Implementation of Ground Behaviors Beneath Super-Tall Building Foundations During Construction

Recently, in Japan, the number of super-tall buildings has been increasing, and weight per area for some has exceeded 1000 kN/m2. The behavior of soil in Japan under such high pressure has not been fully studied, and it is not clear whether such behavior can be simulated using conventional soil constitutive models. As a first step, a series of K0 triaxial tests using Toyoura sand was conducted to study the behavior of the soil beneath the super-tall building foundation during excavation and construction. In this study, a new constitutive model for geomaterials is proposed, where both the confining pressure dependency and strain dependency are considered. Finally, a series of numerical simulations of the CD triaxial test were carried out using the proposed model, along with conventional models for comparison. The proposed model best describes the stress–strain relation obtained from the test because both the confining pressure dependency and strain dependency are properly modelled.

Youhao Zhou, Takatoshi Kiriyama

Slope Stability and Reinforcement

Failure Mechanism Analysis of Loess Slope Under the Coupling Effect of Rainfall and Earthquake Using Shaking Table Test

Rainfall and earthquake are two main factors that trigger landslides. In practice, landslides would occur more easily under the coupling effect of them and present long sliding distance and large sliding scale characteristics, which brought huge losses and casualties to people. Shaking table test of a model loess slope was performed in the study. The wave of Minxian station in 2013 Minxian-Zhangxian Ms6.6 earthquake with intensities of 50 gal, and 100 to 800 gal (interval 100gal) were exerted on the slope successively. Then, the rainfall was applied after 800 gal by using artificial rainfall device and 1300 gal is applied at last. Based on Hilbert-Huang transform (HHT), we investigated the acceleration response of the slope in time-frequency-energy space. The change characteristics of Hilbert energy and predominate frequency under the single effect of earthquake and the coupling effect of rainfall and earthquake were analyzed based on marginal spectrum. According to the dynamic response results of the slope, failure mechanism and damage evolution law of the landslide under the coupling effect of rainfall and earthquake were clarified. The study shows that the coupling effect greatly decreased the stability of the slope and the final damage presents soil flow. The instantaneous energy and cumulative energy have a great dissipation under the coupling effect, which provide the kinetic energy to the soil flow.

Jinchang Chen, Lanmin Wang, Ailan Che
Seismic Stability Analysis of Earth Slopes Using Graphical Chart Solution

The stability charts for static and pseudo-static (PS) stability analysis of homogenous slopes have been widely used for preliminary design work. Based on the concept of a graphical stability chart, this study presents a series of improved stability charts solution based on PS analysis of a wide range of homogeneous slope models with inclination angles varying from 10° to 60°, and considering both horizontal and vertical seismic loads. The concept of limit state line (LSL) is presented, and their mathematical expressions are obtained from extensive regression analyses. The LSL-based graphical stability charts’ solutions are proposed for PS stability analysis of homogeneous slopes with any given combination of physical parameters. The proposed charts show very good performances for obtaining the PS FoSs accurately and rapidly; the calculation errors are generally within 10%. It is anticipated that the developed charts, which consider the relatively realistic conditions of seismic loads, may serve as a promising tool for practicing engineers to estimate the FoS of homogenous slopes equivalent to PS stability analysis in less time.

Hong-zhi Cui, Jian Ji
Influence of Cyclic Undrained Shear Strength Degradation on the Seismic Performance of Natural Slopes

The consequences of several large earthquakes occurred worldwide, reveal that the seismic stability conditions and the post-seismic serviceability of natural slopes are largely affected by the cyclic behaviour of soils. In fact, failures of natural slopes and large permanent displacements may be triggered by the weakening effects due to cyclic shear strength degradation, which may arise simultaneously with inertial effects. In this paper, effect of the cyclic behaviour of soils on the slope seismic response is accounted for in a displacement-based analysis using a simplified degradation model for the soil undrained shear strength. To assess the seismic performance of the slope, a modified Newmark-type analysis is proposed in which a time-dependent value of the slope yield acceleration coefficient kh,c is assumed. kh,c is expressed as a function of the number of equivalent loading cycles of the input motion, a degradation parameter T and a degradation ratio R, the latter quantifying the reduction of kh,c with respect to its initial value. The results of a parametric analysis were used to derive empirical predictive models for a safe estimate of the influence of cyclic undrained shear strength degradation on the seismic-induced permanent displacements.

Giuseppe Di Filippo, Orazio Casablanca, Giovanni Biondi, Ernesto Cascone
Dynamic Analysis of Geosynthetic-Reinforced Soil (GRS) Slope Under Bidirectional Earthquake Loading

For the past few decades, geosynthetic-reinforced soil slopes (GRS slopes) have been increasingly used in geotechnical, hydraulic and geoenvironmental engineering applications, due to their great earthquake resistance. Near-field strong ground motion usually involves a vertical component, which is very large in some cases. However, existing design guidelines do not provide a clear approach of earthquake resistant design for GRS slopes subjected to combined horizontal and vertical accelerations. In this study, a nonlinear Finite Element procedure was further validated by a centrifuge shaking table test, and then employed to investigate the seismic responses of a GRS slope model considering a large range of bidirectional earthquake loadings, based on a highway project located in Xinjiang. The results showed that the vertical acceleration had a great effect on permanent displacement, if the corresponding horizontal acceleration was large, and it also played an important role on the stiffness and natural resonant frequency of the soil due to soil compaction. There existed good correlations between the earthquake intensity parameter ars at the center of gravity of active wedge and seismic responses of the GRS slope after horizontal earthquake loading.

Cheng Fan, Kui Cai, Huabei Liu
Seismic Performance of Slopes at Territorial Scale: The Case of Ischia Island

Ischia is an active volcanic island in the gulf of Naples (Italy), historically hit by several earthquakes which caused extensive structural damage as well as landslides, especially localized in the north-western area of the island. The aim of this study is to assess the seismic performance of slopes at territorial scale in the three municipalities which were mostly damaged by a recent earthquake occurred in 2017. The main results are presented through different maps individuating the areas potentially unstable in terms of earthquake-induced slope displacements. Validation of the results was carried out comparing such areas with those historically affected by earthquake-induced landslides, as reported by the chronicles and the literature. The comparison confirmed the effectiveness of the proposed approach, while the resulting maps may provide a significant planning tool for managing the emergency after a strong-motion event and defining a priority scale of interventions to mitigate the instability risk. Finally, they permitted to identify the most critical areas where further site-specific investigations as well as simplified to advanced numerical analyses will be carried out.

Francesco Gargiulo, Giovanni Forte, Anna d’Onofrio, Antonio Santo, Francesco Silvestri
Numerical Simulation of Seismic Performance of Road Embankment Improved with Hybrid Type Steel Pile Reinforcement

Highway embankments are part of strategic infrastructures that play a critical role in connecting and transporting critical rescue components and disaster relief materials during natural disasters such as earthquakes, tsunamis, typhoons, and rainstorms. The failure of such structures during disasters can exacerbate the extent of damage to human lives due to delays in transporting rescue services. The 2017 NEXCO report points to a recent example of embankment damage in the town of Mashiki in the Kumamoto region as a result of the 2016 Kumamoto earthquake. Therefore, it is important to develop a sustainable and economical geotechnology applicable to existing and new highway embankments. This research proposes a new seismic mitigation technique that uses hybrid reinforcement to protect highway embankments and mitigate the associated ground subsidence. The technique aims to reduce lateral and vertical deformation of the embankment and the buildup of excess pore water pressure by reinforcing the underlying liquefiable soil with two types of steel piles deployed around and below the embankment foundation. The performance of the proposed technique is evaluated through a dynamic effective stress analysis using the LIQCA FEM program.

Chengjiong Qin, Hemanta Hazarika, Divyesh Rohit, Nanase Ogawa, Yoshifumi Kochi, Guojun Liu
Distribution of Deformations and Strains Within a Slope Supported on a Liquefiable Stratum

Ground slopes with an underlying liquefiable layer are particularly prone to failure due to seismic excitation. For an embedded foundation, the extent of ground deformation at its specific location within the slope, will dictate the level of detrimental consequences. As such, knowledge about the spatial configuration of expected ground deformation along the slope’s length and height will provide insights towards assessment and mitigation of the consequences. For that purpose, calibrated Finite element (FE) simulations of ground slopes are conducted, and derived insights from the seismic response analyses are gleaned, where properties of upper crust layer and thickness of the liquefiable layer are varied. Generally, lower levels of deformation are to be expected with distance away from the crest and toe of the sloping zone, and the study aims to quantify this effect. In addition, it is shown that properties of the upper crust may have a significant influence on the pattern and level of accumulated downslope permanent deformation.

Zhijian Qiu, Ahmed Elgamal
Probabilistic Seismic Hazard Curves and Maps for Italian Slopes

The seismic performance of an earth slope is commonly evaluated through the permanent displacements developed at the end of an earthquake. In this paper a probabilistic approach is adopted to assess the displacement of the slope for a given hazard level using an updated database of ground motions recorded during the earthquakes occurred in Italy. The results are presented in terms of hazard curves, showing the annual rate of exceedance of permanent slope displacement evaluated using ground motion data provided by a standard probabilistic hazard analysis and a series of semi-empirical relationships linking the permanent displacements of slopes to one or more ground motion parameters. The probabilistic approach permits to take into account synthetically the characteristics of the slope through the yield seismic coefficient, the aleatory variability of the ground motions and the different subsoil classes of the recording stations. Finally, the procedure has been extended on a regional scale to produce seismic landslide hazard maps for Irpinia, one of the most seismically active regions in Italy. Seismic landslide hazard maps are very attractive for practitioners and government agencies for a screening level analysis to identify, monitor and minimise damages in zones that are potentially susceptible to earthquake-induced slope instability.

Fabio Rollo, Sebastiano Rampello
New Soil-Pile Spring Accounting for a Tree-Root System in the Evaluation of Seismic Slope Stability

Tree-root system plays an important role in the stabilization of soil slopes. Many failures of soil slopes covered with trees were observed during the Hokkaido Eastern Iburi earthquake in 2018. Tree vibration may have induced the slope failure. The response of trees on the slope is not well understood under seismic conditions. The influence of a tree-root system on the slope failures may not be ignored under the seismic event. Therefore, it is important to evaluate the seismic behavior of a tree-root system including surrounding soil. The 2D FEM analysis for the slope stability was performed by adding the macroscopic properties such as resistance strength (i.e., cohesion) and stiffness directly to the soil, which simplifies the modeling. The cohesion predicting is influenced by the definition of the rooted zone since the shape of the root differs depending on the tree species. This study aims to investigate the soil-pile spring accounting for tree-root systems using the 2D FEM analysis. The tree-root system is simply modeled as a connection spring between the soil and the pile. The root-soil spring is validated with the tree pull-down tests. The simulation revealed that the root-soil spring differs depending on the tree species. The soil-pile spring could be applied for evaluating the tree-root system. The concept of the soil-pile spring could be applied for evaluating the tree-root system including the effect of the horizontal root growth.

Yoshikazu Tanaka, Kyohei Ueda, Ryosuke Uzuoka
Numerical Study on Delayed Failure of Gentle Sloping Ground

The 2018 Sulawesi, Indonesia, earthquake (Mw7.5) triggered massive flow slide on very gentle slopes of 1% to 5%. Thousands of casualties and missing persons were reported due to such unprecedented disaster. Although detailed mechanism of the flow has yet been speculative, liquefaction is identified as a possible suspect. In the present study, cause of such a flow slide is numerically investigated using a 2D finite element method by simulating the dynamic behavior of the 30 m thick and 200 m long gentle sloping ground (slope angle 2°) of alternating silt and sand layers with the recorded acceleration as an input motion. Flow slide was simulated by varying the permeability of silt (5 m thick) located on top of the saturated loose sand (5 m thick). When the permeability of silt (ksilt = 1 × 10–6 m/s) is set lower than that of the saturated loose sand (ksand = 1 × 10–4 m/s), the surface layer above the boundary between the silt and sand starts to flow long after the ground shaking ended. In total 1,000 s of simulation, the ground displacement of about 40 m was obtained. Close observation at the boundary elements revealed that the excess porewater pressure is gradually increasing in the silt layer while that in the liquefied layer dissipates. Mechanism of this type of failure has been studied as the void redistribution mechanism during liquefaction or the formation of water film, which may cause delayed failure of the gentle sloping ground. Although this study successfully simulates the delayed failure in 2D model, there is room for further research into seeking causes of the flow slides occurred in specific areas in Palu, Sulawesi, Indonesia.

Tetsuo Tobita, Hitomi Onishi, Susumu Iai, Masyhur Irsyam
Seismic Fragility Assessment for Cohesionless Earth Slopes in South Korea

Seismic performances of earth slopes have been primarily assessed from the predictive models of a horizontal slope displacement as a function of characteristics of slope and ground shaking, based on Newmark or Newmark-type sliding block analyses. A seismic fragility curve of earth slopes has been barely investigated, but it could be useful for resilient analysis (e.g., repair cost, recovery time, and loss of life and property). This study aims to develop the seismic fragility curves of earth slopes in South Korea using finite element (PLAXIS 2D) simulations. We selected a single height of slope (10 m) and three slope angles of 20°, 30°, and 40° defining slope geometries that are representative of slope conditions in South Korea with cohesionless soil types. These conditions resulted in a total of 6 slope models. We considered 280 ground motions calculated by one-dimensional site response analyses for various site conditions. The fragility curves were constructed for three damage states (i.e., minor, moderate, and major) using the computed displacements from the simulations. It turned out that the fragility curves were highly dependent on slope angle and ground motion characteristics.

Dung Thi Phuong Tran, Hwanwoo Seo, Youngkyu Cho, Byungmin Kim
Liquefaction-Induced Lateral Displacement Analysis for Sloping Grounds Using Long-Duration Ground Motions

Earthquake-induced liquefaction occurring in saturated granular deposits would generally yield permanent deformation of sloping grounds. Many previous studies have highlighted the important role of ground motion duration in triggering liquefaction, yet its effect on liquefaction-induced lateral displacement (LD) of sloping grounds remains ambiguous. In this study, a one-dimensional gently sloping ground model containing a loose-sand liquefiable layer is implemented in OpenSees. A total of 126 pairs of long-duration and spectrally equivalent short-duration ground motions are selected from recently occurred giant earthquakes and the NGA-West2 database, respectively. Numerical analyses are then conducted for the sloping ground using the selected ground motions as input. The results indicate that the long-duration ground motion has a higher potential to cause larger LD than the spectrally equivalent short-duration one. The median of the LDs for the long-duration suite is about 10 times larger than that of the short-duration one, attributing to the much more energy contained for the long-duration ground motions. Moreover, it is indicated that cumulative absolute velocity (CAV) is highly correlated with LD for the long-duration ground motions. Consequently, a simple CAV-based empirical model is proposed to predict LD caused by long-duration ground-motion records.

Qiang Wu, Dian-Qing Li, Wenqi Du
Performance of Slopes During Earthquake and the Following Rainfall

Earthquake is a major cause of landslides in a lot of mountainous areas throughout the world. However, when struck by earthquake, some slopes with certain topographies or geological conditions only experienced a certain loss of stability, resulting in tensile cracks on the surfaces rather than a complete failure. These slopes with shaking-induced cracks become vulnerable to rainfall due to the reduced soil strength and water infiltration. Previous studies also suggested that, rainfall that followed an earthquake could cause a significant failure of the slope with cracks. Therefore, in areas where earthquake and rainfall are common, the prediction and mitigation of slope failure caused by earthquake and rainfall are of great importance, which calls for a thorough investigation into the performance of slopes to earthquake and rainfall. This study aimed to examine the performance of slopes to post-earthquake rainfall through the finite element method. First, the slope deformation caused by earthquake was simulated. Although the generation of cracks were not incorporated, the main features of slope deformation such as the location of slip surface and the deformation scale were reproduced. After that, the effects of permeability change and damage due to shaking were included in the analysis of slope deformation during rainfall. Simulation results showed that, the proposed method performed well to reproduce the slope deformation caused by post-earthquake rainfall.

Jiawei Xu, Kyohei Ueda, Ryosuke Uzuoka
Seismic Stability Analysis of Anti-dip Bedding Rock Slope Based on Tensile Strength Cut-Off

For the anti-dip bedding rock slopes (ABRSs), tensile stresses are likely to be generated near the top of the slope, and the results obtained by using the classical Mohr-Coulomb yield criterion are often on the safe side, because the criterion over-estimates the tensile strength of the rock containing joints. In this paper, based on the basic principle of limit analysis upper limit method, a computational failure mode for stability analysis of ABRSs considering tensile strength cut-off is proposed. From the viewpoint of power, the stability coefficient expressions of ABRSs under seismic effects are derived by combining the limit analysis upper limit method and tensile strength cut-off theory. Analysis of the results shows that parameters such as slope angle, horizontal seismic force coefficient, inclination of structural plane and spacing of structural plane have important effects on slope stability coefficient and failure modes.

Qiangshan Yu, Dejian Li, Yingbin Zhang
Seismic Stability Analysis of High Steep Slopes Considering Spatial Variability of Geo-Materials Based on Pseudo Static Method

In recent years, with development of underground space and further emphasis on the ecological environment protection, some engineering projects aimed at the ecological restoration of underground abandoned mines have emerged. The Changsha Xiangjiang Happy City project is based on an abandoned mine pit, which was excavated due to quarrying, and the loadings from the super-structure acts directly on the rock mass of the pit wall. Therefore, the stability of the high steep slope of the mine is of great importance for safety of the construction projects. Considering the potential earthquake action and the inherent spatial variability of physical and mechanical parameters of rock and soil, the stability analysis of the high steep slope of the mine under self-weight from the super-structuresis carried out. The pseudo-static method is used to simulate the earthquake action, and the random field method is adopted to characterize the spatial variability of the cohesion $$c$$ c and the internal friction angle $$\varphi$$ φ . The results show that the seismic action significantly affects the stability of the slope, and the failure probability of the slope gradually increases with increase of seismic coefficient. The influence of spatial variability on slope failure probability cannot be ignored, and neglecting spatial variability may lead to overestimation of failure probability.

Wengang Zhang, Fansheng Meng, Jianxin Li, Changjie He

Liquefaction and Testing

Urban Scale Fragility Assessment of Structures Considering Soil-Structure-Interaction

Fragility curves for structures are typically calculated considering fixed-base structures, i.e., neglecting the interaction between soil, foundation, and structure (SFSI). The state-of-the-art literature proves that considering foundation flexibility, especially for structures resting on soft soil, may lead to different fragility or loss estimates than the fixed-base-on-rock assumption. Including these effects on the city-scale vulnerability analysis is considered a challenging task due to the high exposure concentration and complexity of all the interacting urban systems. For this reason, large-scale analyses are commonly carried out applying existing fragility curves which may have been assessed not correctly accounting for the variation in frequency and amplitude contents imposed by each site’s local geotechnical and topographic conditions. To this aim, a new simplified methodology is proposed in this study to perform an urban-scale vulnerability assessment of structures considering the influence of SFSI and local site-effects. The applicability of the proposed approach is based on globally available data regarding the soil parameters, the foundation, and the building taxonomy. The main findings demonstrate that the conventional way of calculating fragility curves may lead to an incorrect evaluation of the seismic risk, especially in soft soil formations.

C. Amendola, D. Pitilakis
Seismic Performance Assessment of Port Reclaimed Land Incorporating Liquefaction and Cyclic Softening

The seismic performance of a reclaimed land part of a New Zealand Port facility was examined through Nonlinear Dynamic Analyses (NDA), under the Operating (OLE) and the Contingency Level (CLE) seismic performance levels described in ASCE/COPRI 61-14 (2014), using advanced tools to investigate the resilience of the asset under the various performance levels and evaluate the seismic hazard, investigate the failure mechanisms leading to ground deformations, and identify key factors contributing to earthquake-induced ground deformations. The geotechnical models were implemented in the finite difference software FLAC and the behaviour of the sand-like and clay-like soils was simulated using the PM4Sand and PM4Silt constitutive models, respectively. Results from the NDA suggest that a compounded effect of both liquefaction of sand-like soils and cyclic softening of clay-like soils may lead to excessive ground deformations that can have detrimental effects on this asset. Liquefaction starts to become a dominant phenomenon under the CLE NDA. Shear failures associated with ratcheting of the strains being developed within the adjacent soils are developed immediately under the revetments. This phenomenon is initiated under the OLE and becomes also dominant under the CLE performance level simulations. Although liquefaction presents a significant challenge, the seismic response of the clay-like soils is indicated to pose a much more significant problem across the site.

Ioannis Antonopoulos, Alex Park, Grant Maxwell
An Attempt to Evaluate In Situ Dynamic Soil Property by Cyclic Loading Pressuremeter Test

Laboratory experiments, such as cyclic triaxial tests, are generally conducted to evaluate the dynamic behavior of soil for earthquake-proof design. The disturbance and stress release of soil specimens caused by sampling considerably affect the testing results. To avoid sampling disturbance effect, the authors attempted to measure the dynamic soil properties under cyclic loading by directly implementing an in situ pressuremeter test in the borehole. In the feasibility test on loose sand, it was confirmed that cyclic loading was achieved in situ successfully. Besides, in the tests on medium dense sand, hardening behaviors of surrounding soil were observed.

Keigo Azuno, Tatsumi Ishii, Youngcheul Kwon, Akiyoshi Kamura, Motoki Kazama
Effect of Refinements to CPT-Based Liquefaction Triggering Analysis on Liquefaction Severity Indices at the Avondale Playground Site, Christchurch, NZ

The 2010–11 Canterbury Earthquake Sequence (CES) resulted in significant and widespread liquefaction of the soils underlying Christchurch, New Zealand, offering an opportunity to evaluate factors contributing to seismic ground failure. This study presents the results of a field campaign undertaken to examine the subsurface stratigraphy at a selected liquefaction case history site in Christchurch with cone penetration tests (CPTs) at an unparalleled horizontal resolution. The field data is used to evaluate the role of selected refinements to liquefaction triggering procedures on commonly-used liquefaction severity indices for comparison to observed liquefaction severity. Subsurface cross-sections were constructed at the alluvial Avondale Playground site using 18 closely-spaced CPTs. Refinements to the selected CPT-based liquefaction triggering procedure included the use of a site-specific fines content correlation, corrections for the increased cyclic resistance of partially-saturated soils, and rigorous correction for thin layer and layer transition effects on CPT data. Five liquefaction severity indices (LSIs) were evaluated and compared against observed liquefaction severity. In general, application of the refinements to the liquefaction triggering analyses resulted in improved correspondence between the calculated and observed LSIs by reducing over-prediction of severity when evaluated on a site-wide basis. However, when considering individual soundings, the presence of local and deeper fine-grained lenses appear to explain the lack of observed surficial expression of liquefaction which could not have been anticipated from the magnitude of computed LSIs alone.

John R. Cary, Armin W. Stuedlein, Christopher R. McGann, Brendon A. Bradley, Brett W. Maurer
Effect of Membrane Penetration on the Undrained Cyclic Behavior of Gravelly Sands in Torsional Shear Tests

While there is extensive research conducted on liquefaction of sandy soils, only a limited number of studies have been done to understand the liquefaction behavior of gravelly soils. Even with the few case histories that have been documented, several disasters have highlighted the need for further research. To investigate the liquefaction characteristics of gravelly soils in laboratory tests, due consideration must be made on the effect of membrane penetration (MP). MP may be defined as the intrusion/extrusion of membrane into the specimen voids as confining pressure is increased/decreased, respectively. In addition, the effect of membrane force on the measured shear stress must be considered for large strains. This study presents a method for minimizing the effect of MP in undrained torsional shear tests. In this method, a thin layer of sand is used as a MP minimization layer on both the inner and outer surfaces of the specimen. Undrained cyclic torsional shear tests while keeping the specimen height constant were performed on saturated gravelly sand. Based on the results, the undrained cyclic behavior of gravelly soils was investigated.

Matthew Gapuz Chua, Takashi Kiyota, Masataka Shiga, Muhammad Umar, Toshihiko Katagiri
Implementation and Verification of an Advanced Bounding Surface Constitutive Model

The description of the mechanical behavior of loose saturated sandy soils under large cyclic loading represents a complicated task in earthquake geotechnical engineering because of the possible occurrence of liquefaction. The physical and numerical reproduction of this phenomenon is of great support for the protection of the built environment and the assessment of new constructions. To this end, a large number of constitutive models were developed over the years and integrated into computational platforms to describe the soil response to seismic actions. The present paper deals with the implementation and testing of the constitutive model developed by Papadimitriou and Bouckovalas (2002) into the OpenSees environment with the aim of investigating its performance at the element level. The model provides some enhancements with reference to the previous common elastoplastic models, such as a non-linear elastic hysteretic Ramberg-Osgood formulation, and an empirical index that accounts for the fabric evolution. The reliability of the implementation is analyzed through the simulation of drained and undrained, cyclic and monotonic laboratory tests from literature, regarded as a benchmark for the verification and validation of the constitutive model. Overall, the implementation procedure produced satisfactory results and the OpenSees potential in hosting user-defined codes is demonstrated.

Tony Fierro, Stefano Ercolessi, Massimina Castiglia, Filippo Santucci de Magistris, Giovanni Fabbrocino
Performance of Advanced Constitutive Models in Site Response Analyses of Liquefiable Soils

The prediction of liquefiable soils response to earthquake excitations represents one of the most challenging achievements in geotechnical earthquake engineering. For this reason, many constitutive models were formulated to capture the main features of such a complex phenomenon, in the different forms of flow liquefaction and cyclic mobility. These models can catch the soil response at the element level, and nowadays their reliability is increasing with reference to boundary value problems too. To this aim, this paper presents the results of mono-dimensional site response analyses by using an existing advanced bounding surface constitutive model that we implemented in the OpenSees framework. The results are compared to those obtained by adopting other constitutive models already included in the platform. The comparison was possible thanks to the availability of the calibration parameters for the Nevada Sand. The outcomes highlight that this newly implemented constitutive model can be used to address the response of a liquefiable soil profile. However, the discrepancies in the response should be further investigated.

Tony Fierro, Massimina Castiglia, Filippo Santucci de Magistris
A Study on Liquefaction Characteristics of Sandy Soil in Large Strain Levels to Improve the Accuracy of Large Deformation Analysis

An analytical method to accurately predict the large deformation due to liquefaction is required, especially for the design considering a large earthquake. Therefore, as a basic study of large strain liquefaction behavior for numerical analysis, elemental simulations were carried out based on the results of large strain liquefaction tests for Toyoura sand up to γsa = 60%. The results showed that the normal ϕf could not reproduce the test results. However, by using ϕf after reduction by cyclic shear stress loading, the results of large strain liquefaction tests up to γsa = 60% could be generally reproduced. In order to further verify the applicability of using the reduced ϕf, a two-dimensional large-deformation analysis was conducted to reproduce the centrifuge model test that obtained large deformation. As a result, compared with the results using the normal ϕf, the results using the reduced ϕf were closer to the experimental results. Future work includes setting up a method to evaluate the continuous reduction of ϕf due to cyclic shear stress, including after reaching the limiting shear strain γL, confirming the rate of reduction of ϕf in materials other than Toyoura sand, and applying the method to two-dimensional analysis.

Noriyuki Fujii, Takashi Kiyota, Muhammad Umar, Kyohei Ueda
A New Biaxial Laminar Shear Box for 1g Shaking Table Tests on Liquefiable Soils

In this paper, the design of a new laminar shear box at the Laboratory of Earthquake Engineering and Dynamic Analysis (L.E.D.A.) of the University of Enna “Kore” (Sicily, Italy), has been presented. The laminar box has been developed to investigate the liquefaction phenomenon and to validate advanced numerical models and/or the numerical approaches assessed to simulate and prevent related effects.The paper describes in detail the types of soil container that have been used in the last three decades. Particular attention has been paid to the laminar shear box and liquefaction studies. Moreover, the most important factors that affect the performance of a laminar shear box are reported. The last part of the paper describes components, properties and design advantages of the new laminar shear box for 1g shaking table tests at L.E.D.A.The new laminar box is rectangular in cross section and consists of 16 layers. Each layer is composed of two frames: an inner frame and an outer frame. The inner frame has an internal dimension of 2570 mm by 2310 mm, while the outer frame has an internal dimension of 2700 mm by 2770 mm. Between the layers, there is a 20 mm gap making the total height of 1.6 m. Aluminum is chosen in order to reduce the inertial effect of the frame on the soil during shaking. Each internal frame is supported independently on a series of linear bearings and rods connected to the external frame, while each external frame is supported independently on a series of linear bearings and rods connected to the surrounding rigid steel walls.Shaking table tests will be carried out on saturated sandy soils using the 1g 6-DOF 4.0 m × 4.0 m shaking tables at L.E.D.A.

Salvatore Grasso, Valentina Lentini, Maria Stella Vanessa Sammito
Assessing the Limitations of Liquefaction Manifestation Severity Index Prediction Models

The severity of surface manifestation of liquefaction is commonly used as a proxy for liquefaction damage potential. As a result, manifestation severity index (MSI) models are more commonly being used in conjunction with simplified stress-based triggering models to predict liquefaction damage potential. This paper assesses the limitations of four MSI models. The different models have differing attributes that account for factors influencing the severity of surficial liquefaction manifestations, with the newest of the proposed models accounting more factors than the others. The efficacies of these MSI models are evaluated using well-documented liquefaction case histories from Canterbury, New Zealand, with the deposits primarily comprising clean to non-plastic silty sands. It is found that the MSI models that explicitly account for the contractive/dilative tendencies of soil did not perform as well as the models that do not account for this tendency, opposite of what would be expected based on the mechanics of liquefaction manifestation. The likely reason for this is the double-counting of the dilative tendencies of medium-dense to dense soils by these MSI models, since the liquefaction triggering model, to some extent, inherently accounts for such effects. This implies that development of mechanistically more rigorous MSI models that are used in conjunction with simplified triggering models will not necessarily result in improved liquefaction damage potential predictions and may result in less accurate predictions.

Russell A. Green, Sneha Upadhyaya, Brett W. Maurer, Adrian Rodriguez-Marek
Assessment of Stone Column Technique as a Mitigation Method Against Liquefaction-Induced Lateral Spreading Effects on 2 × 2 Pile Groups

Two shake table experiments were conducted on two 2 × 2 pile groups to investigate the efficacy of stone column technique as a mitigation method against liquefaction-induced lateral spreading. The experiments were performed employing a designed and manufactured laminar shear box by Sharif University of Technology. Two separate 2 × 2 pile groups (with and without superstructure weight) were installed in the models to investigate the effects of superstructure weight on the behavior of pile groups in treated and untreated liquefiable layers as well. The models were shaken with a sinusoidal base acceleration having a frequency of 3 Hz and amplitude of 0.3 g. The piles and the soil at far field were instrumented to measure various parameters during and after shaking. The results including acceleration, pore water pressure, displacement at free field and bending moment of the piles are briefly presented and discussed in this paper. The results illustrate that stone column technique significantly decreased lateral displacemnts in the free field and bending moments in the piles, while increased the accelearions in the pile caps and rate of dissipation of excess pore water pressure in the liquefiable layer.

S. Mohsen Haeri, Morteza Rajabigol, Milad Zangeneh, Mohammad Moradi
Influence of Lateral Stress Ratio on N-value and Cyclic Strength of Sands Containing Fines

The compaction method, such as the sand compaction pile (SCP) method, is a method to increase the strength of the ground by installing the material into the ground to be improved. However, in the standard penetration test (SPT) and cone penetration test (CPT) of the in-situ sounding test to confirm the improvement effect, the effect of density increase has been sufficiently evaluated, but the effect of lateral stress has not been evaluated. The authors investigated the effect of the lateral stress ratio, Kc-value and presented a relationship between the normalized N-value, N1 of the standard penetration test and cyclic strength based on the results of soil chamber tests and hollow torsional shear tests under anisotropic condition of clean sand. This relationship was analyzed using the same method that led to this relationship against sands containing fines. As a result, a design chart of the ground (recommended chart), which is the relationship between the two considering the fine content, was presented.

Kenji Harada, Kenji Ishihara, Hiroshi Yabe
Effect of Water Flow Rate and Insertion Velocity on Soil Disturbance Due to Insertion of Small-Scale Self-boring Tube

The freezing sampling is considered as a method for high quality undisturbed soil sample especially for evaluating the liquefaction strength on cohesionless and saturated sandy soils accurately. However, since the conventional freezing sampling method takes much coolant consumption and time, a newly “small-scale” freezing sampling method was developed recently. Due to the procedure of inserting the freezing tube into the ground, some disturbances around the freezing tube might be occurred before the soil has been frozen. In this study, in order to obtain undisturbed soil samples for the laboratory test, attempts were made to evaluate the area of disturbance in the collected sample by the “small-scale” freezing sampling. A 750 mm-height, 186 mm-diamater sand box with an insert machine is adopted in this research to observe the disturbance after insert the freezing tube. From the changes in density, soil hardness and other factors after the test, it was found that flow rate of water and the insert velocity of freezing tube affected the area of sample disturbance around the freezing tube.

Pei-Chen Hsieh, Takashi Kiyota, Toshihiko Katagiri, Masataka Shiga, Manabu Takemasa
Liquefaction Countermeasure for Existing Structures Using Sustainable Materials

Many liquefaction countermeasures have been developed by researchers all over the world. The drainage approach is considered to be one of the most successful among those. However, most research focused on utilizing this method for newly constructed structures. In this research, a sustainable and low-cost technique for existing infrastructures is proposed. Gravel-tire chips mixture (GTCM) as an alternative drainage enhancing geomaterial has been adopted here. A series of model tests were conducted using shaking table. The results indicated that excess pore water pressure beneath the building quickly dissipated through GTCM drains during the shaking. The drainage system successfully prevented the liquefaction leading to the significant reduction of the settlement of the structure.

Yutao Hu, Hemanta Hazarika, Gopal Santana Phani Madabhushi, Stuart Kenneth Haigh
Undrained Monotonic Compression, Cyclic Triaxial and Cyclic Simple Shear Response of Natural Soils: Strength and Excess Pore Water Pressure Response

The mechanical response of soils is governed by several factors including loading and boundary conditions. Under undrained boundary conditions, the nature and magnitude of excess pore water pressure (PWP) control the evolution of effective confining pressure (p′) which in turn controls the evolution of shear stress. In this study, we investigate the shear strength and excess PWP response of natural soils under monotonic triaxial compression (TX), cyclic triaxial (CTX) and cyclic simple shear (CSS) testing conditions. The experimental study consisted of evaluating the undrained response of 31 natural soils collected from 10 locations (including 5 dams) in the Kutch region of India. The significance of the investigation lies in the fact that the region is seismically active with a proven history of devastating earthquakes. The most recent earthquake, the 2001 Bhuj earthquake, created large scale destruction with incidences of widespread earthquake liquefaction. The experimental investigation revealed that the undrained response of the soils at the in-situ density is controlled by both the fines content (FC) and plasticity index (PI). For cohesionless soils, FC governed the soil behaviour whereas for cohesive soils PI dominated the soil behaviour. Cohesionless soils exhibited intense strain softening (SS) under monotonic triaxial compression whereas cohesive soils displayed limited strain softening (LSS). Under CTX and CSS testing conditions, cohesionls soils exhibited very low liquefaction resistance (less than 10 cycles) whereas cohesive soils did not liquefy in 50 cycles. However, cohesive soils did exhibit significant degradation in cyclic strength, which was controlled by PI. The excess PWP was found to be contractive for all three conditions. For cyclic loading, PWP was found to be 30% higher for CSS conditions compared to the CTX conditions. Cyclic simple shear simulates the earthquake conditions better and should be considered for seismic and liquefaction analysis.

Majid Hussain, Ajanta Sachan
Simple Countermeasure Method to Mitigate the Settlement and Tilting of Existing Detached Houses Owing to Liquefaction

In this study, we propose a new liquefaction countermeasure to suppress the settlement and tilting of detached houses that can satisfy the three conditions of “narrow construction space,” “cost,” and “impact on neighboring houses and life.” The concept of this liquefaction countermeasure is to compensate for the reduced bearing capacity of a spread foundation when the bearing layer liquefies using the axial force of the pile and tension of the wire. In this paper, the mechanism of the countermeasure effect of this method is discussed using 1/25 scale 1G shaking-table tests under the experimental conditions of pile arrangement and wire arrangement between piles. From a series of shaking table experiments, it was observed that the piles were placed in pairs near the corners of the building model, and the horizontal displacement of the pile heads and pile tips was constrained, which significantly reduced the amount of settlement. However, the deformation did not cause the pile to yield.

Keisuke Ishikawa, Susumu Yasuda, Motomu Matsuhashi, Toshifumi Fukaya
Effect of Artesian Pressure on Liquefaction-Induced Flow-Slide: A Case Study of the 2018 Sulawesi Earthquake, Indonesia

This paper discusses the trigger of the long-distance flow-slide that occurred in the 2018 Sulawesi earthquake, Indonesia. As a result of post-earthquake field surveys and interviews with local residents, it was confirmed that artesian groundwater existed before the earthquake in the area where the long-distance flow-slide occurred. This fact may be due to the liquefaction of the deep ground by the earthquake and the inflow of a large amount of groundwater from the aquifer below the liquefaction to the surface layer, which may have caused the long-distance flow-slide. In this study, the possibility of the presence of artesian pressure was examined by preparing a soil cross section of the affected areas estimated from the results of borehole surveys conducted after the earthquake and grain size analysis. In addition, the cyclic resistance ratio of the supposed liquefied soil was obtained by undrained triaxial cyclic loading test using in-situ undisturbed samples, and the effect of artesian pressure on the occurrence of liquefaction was examined by simplified liquefaction analysis.

Takashi Kiyota, Masataka Shiga, Toshihiko Katagiri, Hisashi Furuichi, Hasbullah Nawir
Physical Modeling and Reliability Assessment of Effectiveness of Granular Columns in the Nonuniform Liquefiable Ground to Mitigate the Liquefaction-Induced Ground Deformation

Granular columns have been widely used to mitigate the liquefaction-induced effects on the built environment. Previous studies based on physical and numerical modeling and post-earthquake site investigations consolidate the efficacy of granular columns to mitigate the liquefaction-induced effects under small earthquakes. The increment in lateral stress due to densification, shear reinforcement, and drainage capacity of granular columns are believed to increase the liquefaction resistance of the ground. However, several case histories and recent research development exhibited the limitations of the effectiveness of granular columns under strong earthquakes. Therefore, a series of dynamic centrifuge experiments are carried out to investigate the effectiveness of granular columns in the liquefiable ground under strong ground motion recorded at Hachinohe Port during the 1968 Tokachi-Oki Earthquake. The performance of granular columns is evaluated by examining the evolution of excess pore water pressure, evolution of co-shaking and post-shaking settlement of foundation-structure system. A series of three-dimensional nonlinear stochastic analyses are also carried out using the OpenSees framework with PDMY02 elasto-plastic soil constitutive model to map the reliability of the performance of equally-spaced granular columns. The spatial nonuniformity of the ground should be considered for a reliable engineering assessment of the performance of granular columns which is implemented with stochastic realizations of overburden and energy-corrected, equivalent clean sand, (N1)60cs values using spatially correlated Gaussian random field. The reliability of the performance of the granular column is assessed based on the stochastic distributions of average surface settlement and horizontal ground displacement associated with the degree of confidence.

Ritesh Kumar, Akihiro Takahashi
Experimental Study on the Effect of Coexistence of Clay and Silt on the Dynamic Liquefaction of Sand

In order to study the effect of coexistence of clay and silt on the dynamic liquefaction of sand under different void ratios, the characteristics of sand with two different fines contents (FC = 5% and 10%), three different clay silt ratios (CS = 0.25, 1 and 4) under two different void ratios were investigated by the cyclic undrained triaxial test. The test results show that dynamic test results of specimens with different FC, CS and void ratios are different. Liquefaction happens for all specimens and there are two liquefaction types: flow liquefaction (brittle failure) and cyclic mobility (ductility failure). With the increase of FC or CS, two liquefaction types change mutually. When FC is same, the liquefaction resistance of specimens (with different CS) is different. When FC is different, the liquefaction resistance is different with the monotonous change of clay or silt content in fines. When the void ratio is different, the dynamic strength of specimens is different with same FC and CS. The clay and silt play roles of filling, lubrication, bonding and skeleton effect on the sand particles. With the change of clay and silt content in fines, the proportion of four roles is different under different fines content.

Tao Li, Xiao-Wei Tang
The Prediction of Pore Pressure Build-Up by an Energy-Based Model Calibrated from the Results of In-Situ Tests

The excess pore water pressure generation induced by rapid forms of loading can lead to a significant reduction in soil stiffness and strength until reaching liquefaction. The estimate of earthquake-induced pore pressure is important to predict accurately the response of soil deposits and consequently earthquake effects on built environment. Traditionally, the excess pore water pressure generation is linked to stress and strain, however, recently, more innovative and promising energy-based models are developing. Although they allow avoiding the conversion of irregular earthquake load in an equivalent number of cycles – necessary for traditional models – their calibration is often complex, limiting considerably their application. In this paper, the calibration procedure proposed by Mele et al. [10] for the model of Berrill and Davis [5] is used. It consists of linking the two parameters of the model to the well-known equivalent cone tip resistance (qc1Ncs) or the corrected SPT blowcount ((N1)60cs). This procedure has been used to predict the excess pore water pressure of the case history of Scortichino (Italy), affected by liquefaction phenomena during the 2012 Italian earthquake. The results of 1D site response analysis of Scortichino dyke, performed with a non-linear code, in which the energy-based pore pressure generation model is implemented, show that the results are in good agreement with those deriving from susceptibility analysis and with the experimental evidence, confirming the effectiveness of the calibration procedure and the reliability of the energy-based models in the prediction of pore pressure build-up.

Lucia Mele, Stefania Lirer, Alessandro Flora, Alfredo Ponzo, Antonio Cammarota
CDSS Tests for Evaluation of Vibration Frequency in Liquefaction Resistance of Silica Sand

The frequency of ground motions during earthquakes is typically on the order of a few hertz. In the evaluation of the liquefaction resistance of soil in laboratory tests, it is necessary to consider various vibration frequencies generated by real earthquakes. The effect of vibration frequency has been studied by cyclic triaxial tests; however, it has rarely been investigated by cyclic direct simple shear (CDSS) tests, which are more similar to the cyclic loading conditions associated with earthquakes. In this study, a series of CDSS tests was performed on relative density of 40% of sand obtained from Nakdong River. Two different initial vertical effective stresses (σ′v0, 100 and 200 kPa) and four different frequencies (f, 0.05, 0.1, 0.5, and 1 Hz) were applied to evaluate the effect of the vibration frequency on the liquefaction resistance of clean sand for both the undrained and drained conditions. For the undrained CDSS tests, the liquefaction resistance of the sand was observed to increase with f, regardless of σ′v0. The maximum increase in the cyclic resistance was 15% when f was increased from 0.1 to 1 Hz. For the drained CDSS tests, with an increase in f, the rate of volumetric strain accumulation decreased and the shear modulus ratio increased.

Zhen-Zhen Nong, Sung-Sik Park, Peng-Ming Jiang
Some Important Limitations of Simplified Liquefaction Assessment Procedures

A subset of a recently compiled database of CPT-based liquefaction case histories from Canterbury, New Zealand, is used to scrutinize the performance of simplified liquefaction assessment procedures, for different soil types and ground conditions. In general terms, simplified procedures are shown to perform well (i.e. correctly predict severe manifestation of liquefaction at the ground surface) for deposits that have a critical zone of vertically continuous low-resistance liquefiable soils at shallow depth (true-positive sites), and also (correctly predict the absence of liquefaction manifestation) for relatively uniform high-resistance clean sand or fine sand deposits (true-negative sites). In contrast, the severity of liquefaction manifestation in intermediate-resistance clean sand to silty sand deposits is generally underpredicted by the simplified procedures (false-negative sites). Lastly, systematic overprediction of liquefaction manifestation is observed in deposits with interbedded non-liquefiable soils and liquefiable soils of low-resistance (false-positive sites). The poor performance of the simplified procedures for the false-negative and the false-positive sites can be attributed to the neglect in the evaluation of important system response effects which, on the one hand, intensify the severity of liquefaction manifestation for the false-negative sites and, on the other hand, mitigate liquefaction manifestation for the false-positive sites.

Nikolaos Ntritsos, Misko Cubrinovski
Dynamic Behavior of Pipe Bend Subjected to Thrust Force Buried in Liquefiable Sand

At pressure pipe bends, thrust forces cause the lateral displacement of pipes and subsequent detachment of joints. The current design requires the calculation of the safety factor for pipe sliding based on the passive earth pressure behind the pipe bends; however, the resistance force of the ground and the dynamic behavior of the pipe during liquefaction are not elucidated. Therefore, efficient countermeasures and their design have not yet been established. The objective of this study is to understand the dynamic behavior of buried pipes subjected to thrust forces during liquefaction through a series of dynamic centrifuge tests. One-dimensional horizontal shaking is applied to the model at 40g whereas the lateral load, which simulates the thrust force, is applied to the model pipe. The lateral movement mechanism of the pipe is clarified by measuring the excess pore water pressure, acceleration response, and horizontal displacement of the pipe. Based on the test results, a countermeasure using a gravel layer is devised and its effectiveness is examined.

Kohei Ono, Mitsu Okamura
Liquefaction Resistance of Solani Sand Under Normal and Sequential Shaking Events

Earthquake damage caused by liquefaction is intense and innumerable. This led to the research worldwide on understanding the mechanism and designing suitable ground improvement systems. In the present study, 1g shake table experiments were conducted to investigate the liquefaction resistance and the intrinsic mechanism associated with the sequential shaking events compared to normal shaking events. A tank of dimension 1 × 0.6 × 0.6 m was used for preparing the saturated sand-bed of 25% relative density. Poorly graded liquefiable Solani sand collected from the bed of Solani river near Roorkee, India was used for testing. A total of 8 shaking events were conducted with varying acceleration amplitudes from 0.1g to 0.4g under constant frequency 2 Hz and 1-min shaking duration. In sequential shaking events, the prepared sand-bed was shaken four times in succession with increment in acceleration after the dissipation of pore water pressure. The variation in the generated excess pore-water pressure was monitored continuously using pore-pressure transducers placed at different depths. Sand-bed subjected to sequential events were found susceptible to reliquefaction despite improved sand density under incremental acceleration loading. The experimental data shows that liquefaction resistance increased under the sequential shaking compared to normal events even at higher acceleration amplitude.

Gowtham Padmanabhan, B. K. Maheshwari
Numerical Simulation of Caisson Supported Offshore Wind Turbines Involving Uniform Liquefiable Sand Layer

In this research, a series of the non-linear finite element (FE) analyses was conducted to analyze the influence of the contact pressure caused by the offshore wind turbine and motion characteristics on the settlement pattern and seismic demand of the structure. The procedure was validated against a database of well-documented centrifuge test. The FE results suggested that motion, and offshore wind turbine (OWT) characteristics greatly control liquefaction-induced OWT settlement.

Alfonso Estepa Palacios, Manh Duy Nguyen, Vladimir Markovic, Sina Farahani, Amin Barari, Lars Bo Ibsen
Pore-Pressure Generation of Sands Subjected to Cyclic Simple Shear Loading: An Energy Approach

The energy-based approach to assess the residual pore water pressure build-up and the cyclic liquefaction resistance of sandy soils is receiving increasing attention in the last years since it has been found to better capture the effect of irregular loading cycles compared to the conventional equivalent stress-based methods. This paper reports the results of an experimental study carried out on two sands (Ticino and Emilia) under undrained cyclic simple shear loading with a focus on the development of specific correlations between residual pore water pressure (PWP) build-up and dissipated energy (WS) up to the onset of liquefaction. Tests were carried out on specimens reconstituted at various relative densities (DR) and subjected to various cyclic stress ratios (CSR). The results obtained were compared with two energy-based PWP models provided by other researchers leading to positive conclusions as for the general trend of WS with the main influencing factors. A new simple energy-based PWP model was also proposed for sands and the good performance of the model for cyclic simple shear experiments was shown.

Daniela Dominica Porcino, Giuseppe Tomasello, Roohollah Farzalizadeh
Constitutive Modeling of Undrained Cyclic Shearing of Sands Under Non-zero Mean Shear Stress

SANISAND-MSf is one the latest members of the SANISAND family of models within a critical state compatible bounding surface plasticity framework with kinematic hardening of the yield surface. In pursuance of enhancing the undrained cyclic response, the model incorporates a memory surface for controlling the stiffness affecting the deviatoric and volumetric plastic strains in the pre-liquefaction stage, and the concept of a semifluidized state for controlling stiffness and dilatancy in the post-liquefaction stage. The new model can precisely simulate the cyclic liquefaction of sands in undrained cyclic shearing in the absence of a mean shear stress under isotropic initial stress conditions. In this study we assess the performance of SANISAND-MSf in modeling the undrained cyclic shearing response of sands in the presence of non-zero mean shear stresses. Simulations have been compared with relevant experimental data, along with an assessment of the components of the model responsible for simulating such loading conditions.

Andrés Reyes, Mahdi Taiebat, Yannis F. Dafalias
Investigation of Lateral Displacement Mechanism in Layered and Uniform Soil Models Subjected to Liquefaction-Induced Lateral Spreading

The phenomenon of liquefaction-induced lateral spreading has been responsible for causing catastrophic damage during the past earthquakes. LEAP project was formed with an objective of developing a large centrifuge database using different centrifuge facilities in an effort to characterize the median response of a sloping ground during lateral spreading. Further these databases provide a unique opportunity to verify and validate numerical modeling and contribute to the further enhancement of constitutive modeling involving soil liquefaction. However, such centrifuge and numerical analysis did not consider the presence of a layered sloping ground, which is most likely to be encountered in an ideal scenario. At the same time, it is also important to assess the lateral displacement mechanism within different models having larger soil variability. This shortcoming hinders the development of a performance-based design involving liquefaction-induced lateral spreading considering larger variability in soil conditions. To overcome this, we developed a series of centrifuge experiments. In the centrifuge plan, tests were developed for a uniform loose sand model, a uniform dense sand model and a multi-layer soil model. The result portrays the significantly different deformation mechanism for the uniform denser soil model as compared to a multi-layer soil model and a uniform loose sand model. The lateral soil displacements were found to be largest near the center array for all the models. Due to the occurrence of liquefaction-induced lateral spreading within the loose sand layer, the above denser soil layer could be dragged alongside, resulting in significant mobilization of shear strains towards the ground surface.

Anurag Sahare, Kyohei Ueda, Ryosuke Uzuoka
Probabilistic Calibration and Prediction of Seismic Soil Liquefaction Using quoFEM

Liquefaction under cyclic loads can be predicted through advanced (liquefaction-capable) material constitutive models. However, such constitutive models have several input parameters whose values are often unknown or imprecisely known, requiring calibration via lab/in-situ test data. This study proposes a Bayesian updating framework that integrates probabilistic calibration of the soil model and probabilistic prediction of lateral spreading due to seismic liquefaction. In particular, the framework consists of three main parts: (1) Parametric study based on global sensitivity analysis, (2) Bayesian calibration of the primary input parameters of the constitutive model, and (3) Forward uncertainty propagation through a computational model simulating the response of a soil column under earthquake loading. For demonstration, the PM4Sand model is adopted, and cyclic strength data of Ottawa F-65 sand from cyclic direct simple shear tests are utilized to calibrate the model. The three main uncertainty analyses are performed using quoFEM, a SimCenter open-source software application for uncertainty quantification and optimization in the field of natural hazard engineering. The results demonstrate the potential of the framework linked with quoFEM to perform calibration and uncertainty propagation using sophisticated simulation models that can be part of a performance-based design workflow.

Aakash Bangalore Satish, Sang-ri Yi, Adithya Salil Nair, Pedro Arduino
Fluid-Solid Fully Coupled Seismic Response Analysis of Layered Liquefiable Site with Consideration of Soil Dynamic Nonlinearity

The seismic response study of a layered liquefiable site is crucial in the seismic design of both aboveground and underground structures. This study introduces one-dimensional dynamic site response processes with advanced nonlinear soil constitutive models for non-liquefiable and liquefiable soils in the OpenSees computational platform. The solid-fluid fully coupled plane-strain u-p elements are used to simulate the soil elements. This study investigates the seismic response of a layered liquefiable site with specific focus on the development of excess pore water pressure, acceleration and post-earthquake ground surface settlement under two typical earthquake excitations. The numerical results show that the ground motion characteristics as well as the site profile have significant effects on the dynamic response of the layered liquefiable site. The loose sand layer with 35% relative density is more prone to liquefaction and contractive deformation under the same intensity of ground motion, resulting in irreversible residual deformation and vertical settlement. The saturated soil layer may efficiently filter the high-frequency components of ground motions while amplifying the low-frequency components. Meanwhile, during the post-earthquake excess pore pressure dissipation, the soil produce a large consolidation settlement.

Yiyao Shen, Zilan Zhong, Liyun Li, Xiuli Du
Variation in Hydraulic Conductivity with Increase in Excess Pore Water Pressure Due to Undrained Cyclic Shear Focusing on Relative Density

This paper presents the variation in the hydraulic conductivity of a sandy soil with the increase in the excess pore water pressure by conducting falling head permeability tests under undrained cyclic shear while focusing on the relative density of the soil. A pressurized drainage tank with a double-pipe burette for water injection and a standpipe for drainage were newly installed in a hollow torsional shear apparatus. The testing results showed that the hydraulic conductivity remained practically unchanged after consolidation until the excess pore water pressure ratio increased to approximately 0.9 under the cyclic loading. However, when the excess pore water pressure ratio increased above 0.9 owing to the cyclic loading, the hydraulic conductivity rapidly increased regardless of the relative density.

Toshiyasu Unno, Akiyoshi Kamura, Yui Watanabe
Framework and Demonstration of Constitutive Model Calibration for Liquefaction Simulation of Densified Sand

Two distinct mechanisms contribute to the increase in liquefaction resistance of densified granular soils: (1) the increased relative density, Dr, and the corresponding peak friction angle, $${\phi }_{p}^{{\prime}}$$ ϕ p ′ , and (2) the increased mean effective stress, $${p}^{{\prime}}$$ p ′ , resulting from vibro-compaction-induced lateral stresses. This paper presents a framework for simulating liquefaction phenomena in densified soil that accounts for the two mechanisms. The proposed framework is implemented in the PDMY03 constitutive model calibrated to both laboratory tests and earthquake motions. The direct simple shear tests provide reference CRR-N curves at selected Dr of reconstituted, normally-consoldiated specimens, whereas prior experimental data is used to capture the effects of increased lateral stresses. The selected constitutive model is then calibrated to provide the correct cyclic resistance for the number of equivalent cycles inherent within selected ground motions, Dr, and coefficient of earth pressure at rest, K0. The calibrated PDMY03 constitutive model is then used within free-field nonlinear dynamic analyses to investigate the effect of densification on the free-field response of simple liquefiable soil profiles.

Hao Wang, Armin W. Stuedlein, Arijit Sinha
Three-Dimensional Numerical Simulations of Granular Column Improved Layered Liquefiable Soil Deposit

This paper presents numerical analysis on the mitigation effect of granular columns for layered liquefiable soil deposit. 3D simulations of the centrifuge shaking table tests conducted at the University of Colorado Boulder’s 400 g-ton centrifuge facility are using a plasticity constitutive model for large post-liquefaction deformation of sand (CycLiq), via the FLAC3D finite difference code. The simulations investigate the response of a layered liquefiable soil deposit improved by granular columns, under horizontal Kobe earthquake motion input. The results suggest that the constitutive model and FLAC3D simulation method can be used to successfully capture the seismic response of granular columns improved liquefiable ground. Granular columns can significantly enhance drainage, which suppresses the build-up of excess pore pressure and increases the rate of excess pore pressure dissipation. Meanwhile, it also reinforces the ground by reducing the horizontal and vertical displacements.

Zhao Wang, Rui Wang, Jian-Min Zhang
Fundamental Study on Laboratory Test Method for Setting Parameters of Effective Stress Analysis

The authors of this paper discussed a practical method of setting parameters for an effective stress analysis using results of a laboratory test. As an example of effective stress models, we used GHE-Bowl model, which was composed of GHE model for a skeleton curve and Bowl model for a dilatancy model. The first step of the method is that the parameters of Bowl model are set so that the turning points of the dilatancy curve simulated by Bowl model coincide to those of the test results. The second step is that the parameters of GHE model are determine by fitting deformation properties of the model with those of the test results. In addition, an optimization method was applied for searching parameters, in which the error is minimized using the Shor’s algorithm. Furthermore, to verify the validity of the method, we compared the ground response obtained from the effective stress analysis modeled by GHE-Bowl model with that of hybrid ground response analysis. As the results, we demonstrated that the effective stress analysis using the proposed method for setting parameters can give almost appropriate results.

Masanori Yamamoto, Ryuichi Ibuki, Yasutomo Yamauchi, Taku Kanzawa, Jun Izawa
Experimental and Numerical Study of Rate Effect in Cone Penetration Tests

This paper investigates the rate effect of cone penetration tip resistance. The focus is on the relationship between the penetration resistance and the penetration rate under partially drained conditions. When the penetration rate is increased from a relatively low rate, soil resistance decreases. Beyond a certain value, any further increase in penetration rate causes an increase in the resistance. Numerical simulation of constant rate and variable rate cone penetration tests were carried out using the two-phase material point method (MPM). The penetration rate at which the calculated resistance started to increase was used to estimate the consolidation coefficient. Centrifuge modelling of variable rate cone penetration tests explained the rate effect in terms of consolidation and viscosity.

Jian-Hong Zhang, Hao Wang

S1: Special Session on Liquefaction Experiment and Analysis Projects (LEAP)

Lessons Learned from LEAP-RPI-2020 Simulation Practice

In accordance with the Liquefaction Experiment and Analysis Projects (LEAP)-RPI-2020 guidelines, a model was built using finite element tool OpenSees for blind prediction of centrifuge experiments to investigate the seismic response of a sheet pile retaining structure supporting liquefiable soils. This paper discusses different aspects of a representative numerical model for the LEAP experiment, implications of the decisions made in the process, and presents the lessons learned from this effort. These aspects include modeling of the soil-fluid coupling, soil behavior, structure element, soil-structure interface, boundary conditions, and initial static conditions.

Long Chen, Alborz Ghofrani, Pedro Arduino
Numerical Simulations of the LEAP 2020 Centrifuge Experiments Using PM4Sand

Soil liquefaction is a complex phenomenon that has been studied for several decades, and there is still much to learn about it and its consequences. In this context, as part of Liquefaction Experiments and Analysis Projects (LEAP), in 2020, eleven universities around the globe performed centrifuge tests of a model representing a sheet-pile retaining system supporting liquefiable soils. In this study, we document the numerical simulation performed to predict the 2020 LEAP centrifuge tests response of the Rensselaer Polytechnic Institute and the University of California Davis tests. The numerical simulations are performed using the PM4Sand model developed by Boulanger and Ziotopoulou (2017). Towards this end, we inspected the available laboratory information for the Ottawa sand, and, interestingly, we find discrepancies in the critical state line (CSL) definition, a key input for critical state-based numerical models. Independent of this finding, we selected a CSL to calibrate the PM4Sand model further using the information from cyclic tests. Once calibrated, the PM4Sand model is used in the context of a boundary value problem in finite-difference calculations using the software FLAC. The results show that displacements are well predicted, excess pore pressures are reasonably well predicted, and high-frequency dilation peaks cannot be predicted, which deserves further examination.

Renzo Cornejo, Jorge Macedo
LEAP-2021 Cambridge Experiments on Cantilever Retaining Walls in Saturated Soils

Retaining walls are important geotechnical structures that are used worldwide. The lateral movement of retaining walls can cause damage to structures behind them or their collapse can endanger the infrastructure on the excavated side. This is particularly an issue when retaining structures are located in saturated soils in seismic areas. To address the need to understand the behaviour of retaining walls in saturated soils, an international collaborative project called LEAP (Liquefaction Experiments and Analysis Projects) was conducted. This paper describes the dynamic centrifuge tests on a retaining wall with two different embedment ratios conducted for LEAP-2020 and LEAP-2021. The behaviour of the retaining wall is studied using Particle Image Velocimetry (PIV). Additionally, the acceleration and the excess pore water pressure in the saturated Ottawa sand are presented. It is found that more dilation occurred in the test with the lower embedment ratio. However, the embedment ratio was still crucial in terms of the stability of the retaining wall.

Xiaoyu Guan, Alessandro Fusco, Stuart Kenneth Haigh, Gopal Santana Phani Madabhushi
Repeatability Potential and Challenges in Centrifuge Physical Modeling in the Presence of Soil-Structure Interaction for LEAP-2020

The LEAP (Liquefaction Experiment and Analysis Project) is a continuing international collaboration to create a reliable databank of high-quality experimental results for the validation of numerical tools. This paper investigates the response of a floating rigid sheet-pile quay wall under conditions of seismically induced liquefaction, embedded in dense sand and supporting a saturated liquefiable soil deposit. The experimental challenges related to repeatability in physical modeling in such a soil-structure-interaction regime are also discussed. To this end, three experiments performed at Rensselaer Polytechnic Institute (RPI) as part of the experimental campaign for the LEAP-2020 are discussed herein. Models RPI_REP-2020 and RPI10-2020 investigate the repeatability potential in centrifuge modeling in the presence of soil-structure-interaction. Model RPI_P-2020 is the pilot test of the LEAP-2020 experimental campaign at RPI and investigates the effect of the wall’s initial orientation on the system’s dynamic response and soil liquefaction, as a possible “defect” in the model construction procedure. The three models were built in a consistent way, employed comparable instrumentation layout while simulating the same prototype and comparable soil conditions. The three models were subjected to the same acceleration target input motion, which was repeated across all three models with high consistency.

Evangelia Korre, Tarek Abdoun, Mourad Zeghal
Numerical Modeling of the LEAP-RPI-2020 Centrifuge Tests Using the SANISAND-MSf Model in FLAC3D

The SANISAND-MSf model is an extension of a reference and established critical state bounding surface plasticity model for sands with two new constitutive ingredients: a memory surface and a semifluidized state. Each ingredient is designed to significantly enhance the reference model in capturing the progressive reduction of mean effective stress during undrained cyclic shearing for various cyclic stress ratios in the pre-liquefaction, and the subsequent development of large shear strains in the post-liquefaction, respectively. This paper presents the validation of this model using the results of ten centrifuge tests from LEAP-RPI-2020. These experiments consisted of submerged liquefiable soil deposits, supported by a 4.5 m sheet-pile wall, and subjected to ramped sine wave motions. Fully coupled dynamic analyses were conducted in the finite difference computational platform FLAC3D. Comparisons between the numerical simulations and experimental results reveal that the employed modeling approach is effective in reproducing several essential aspects of the system response.

Keith Perez, Andrés Reyes, Mahdi Taiebat
Numerical Simulations of LEAP Centrifuge Experiments Using a Multi-surface Cyclic Plasticity Sand Model

Numerical simulations of LEAP-UCD-2017 centrifuge tests are conducted using a multi-surface cyclic plasticity sand model implemented with the characteristics of dilatancy, cyclic mobility and associated shear deformation. This model extends the OpenSees PDMY03 material to include the Lade-Duncan failure criterion as the yield function, thus allowing for considerable accuracy in three-dimensional shear response conditions. For this study, the model parameters are calibrated based on a series of available cyclic stress-controlled triaxial tests of Ottawa F-65 sand with relative density Dr. = 65%. Using the calibrated model parameters, Finite Element (FE) simulations are performed for dynamic centrifuge tests of a liquefiable sloping ground. The computed results are systematically presented and directly compared to the centrifuge test data. An overall good match between the simulations and measurements demonstrated that the multi-surface cyclic plasticity model has capabilities for simulating the stress path of cyclic stress-controlled triaxial tests as well as the response of the liquefiable sloping ground subjected to seismically-induced liquefaction.

Zhijian Qiu, Ahmed Elgamal
Validation of Numerical Predictions of Lateral Spreading Based on Hollow-Cylinder Torsional Shear Tests and a Large Centrifuge-Models Database

This paper aims to present a complete validation exercise that explores the capabilities of numerical predictions to simulate the lateral spreading phenomena in clean sands under a diverse range of densities and input motions. The validation exercise used the “Strain Space Multiple Mechanism Model” to simulate the lateral spreading phenomena (although the methodology presented might be used for the validation of other numerical tools as well), and was based on multiple, cross-checked, and high-quality physical models (Centrifuge Models) and element tests (Hollow Cylinder Cyclic Shear Tests), developed for LEAP Project.The validation exercise showed that the numerical model is able to predict the displacements for the median trend and the 95% probability confidence bounds for PGA < 0.25 g.

R. Vargas, Z. Tang, K. Ueda, R. Uzuoka
Centrifuge Modeling on the Behavior of Sheet Pile Wall Subjected Different Frequency Content Shaking

Sheet pile wall is often used as a retaining system at the riverbank owing to its economy, convenience, and constructability. The soil deposit nearby river is composed of alluvium soil with a high ground water level. Therefore, the soil deposit usually has high potential of liquefaction. The shaking may induce soil liquefaction when the earthquake occurs, causing the sheet pile wall damage or failure. Each earthquake has different frequency content and acceleration amplitude in real conditions. Thus, it would lead to different behavior of the wall-soil system. In this study, three dynamic centrifuge tests were carried out by NCU geotechnical centrifuge and shaking table under 24 g centrifugal acceleration field. Ottawa sand was used to prepare the liquefiable ground with a prototypical excavation depth of 3 m. The models were subjected to the input motions with different frequency content of 1 Hz and 3 Hz. The horizontal displacement of the sheet pile wall and ground surface were measured by the linear variable differential transformers and surface markers. Test results indicate that the model subjected to input motion with higher 3 Hz content and higher peak base acceleration has higher excess pore water pressure excitation and excitation rate. The shallow layer soil in the backfilled area of it achieved initial liquefaction. Moreover, it also has larger lateral displacement of sheet pile wall and ground surface as compared to the others.

Yi-Hsiu Wang, Jun-Xue Huang, Yen-Hung Lin, Wen-Yi Hung

S2: Special Session on Liquefaction Database

Liquefaction Cases and SPT-Based Liquefaction Triggering Assessment in China

In the 60s to 70s of last century, major damaging earthquakes hit China and trigged tremendous liquefaction. The post-earthquake in-situ investigation and site tests on liquefied sites and comparative non-liquefied sites were conducted that relevant data have been collected and studied. Using the data, liquefaction evaluation methods have been formulized and accepted by Chinese codes, in which the standard penetration test (SPT) based formula is widely used in engineering practice. This paper introduces liquefaction data collected in China mainland and the revision process of SPT-based liquefaction evaluation formulation. Adopting the established procedures, the SPT blows were corrected on trial by influence factors. Comparison with liquefaction evaluation methods, which base on cyclic-stress-ratio (CSR) framework, indicates the CSR-based liquefaction formulae basically overestimated the liquefaction data in low CSR range less than 0.1. Further investigation has to be performed on the data analysis to study the consistency of data collected from China with those from worldwide.

Longwei Chen, Gan Liu, Weiming Wang, Xiaoming Yuan, Jinyuan Yuan, Zhaoyan Li, Zhenzhong Cao
Hammer Energy Measurement of Standard Penetration Test in China

Hammer efficiency, which is defined as the energy transfer ratio, is an important index in standard penetration test which is widely used for measurement of soil penetration resistance and subsequent correlation with soil properties. In the routine procedure, the energy measurement should be made to correct penetration blow counts. However, the energy correction for standard penetration test are not considered in Chinese codes. To circumvent the shortcoming, in-situ tests at three selected sites in Xichang city region are performed to measure the energy transmitted into rod during hammer fall in the current commonly used standard penetration test setups. The energy transfer ratio data recorded are well consistent, that most energy transfer ratio values range in 60% to 90% with mean ratios exceeding 70%. The dependence of energy transfer ratio on testing depth is weak. The energy transfer ratio increases by about 10% when the penetration depth increases down to 20 m beneath surface. The analysis results can prove the reliability of current standard penetration test setups and provide a good reference for energy correction of blow counts.

Longwei Chen, Tingting Guo, Tong Chen, Gan Liu, Yunlong Wang
Empirical Magnitude-Upper Bound Distance Curves of Earthquake Triggered Liquefaction Occurrence in Europe

This article presents empirical magnitude – maximum distance threshold curves developed starting from the European interactive Catalogue of earthquake-induced soil Liquefaction phenomena (ECLiq). Based on the latter, European regressions were computed to predict the maximum distance of liquefaction occurrence starting from the main seismological information of an earthquake. The interactive catalogue ECLiq is a unique digital archive, which includes documented historical information regarding earthquake-induced manifestations of soil liquefaction occurred in Europe in the latest 1000 years or so. ECLiq is publicly available as web-based Geographical Information System (GIS) platform at the link . The catalogue includes both the main seismological data of the earthquakes that triggered liquefaction in Europe and the features of liquefaction manifestations. Based on ECLiq, new empirical European relationships between earthquake magnitude and maximum distance for liquefaction were developed and presented hereinafter. ECLiq was used to identify magnitude – maximum threshold distance pairs above which liquefaction is unlikely to occur. It is important to emphasize that since these models were developed based on historical data of liquefaction occurrence, their use to predict the occurrence/non occurrence of liquefaction at a given site and for a given magnitude and location of an earthquake in Europe, must be linked to the actual ground conditions of the site of interest and in particular of its susceptibility/non susceptibility to liquefaction.Estimating the location where soil liquefaction can possibly occur in the immediate aftermath of a strong earthquake is valuable for rapid loss estimation, post reconnaissance surveys and site investigations. The proposed equations can also be adopted to estimate the magnitude of an earthquake in paleoseismic studies.

Mauro De Marco, Francesca Bozzoni, Carlo G. Lai
Laboratory Component of Next-Generation Liquefaction Project Database

Soil liquefaction and resulting ground failure due to earthquakes presents a significant hazard to distributed infrastructure systems and structures around the world. Currently there is no consensus in liquefaction susceptibility or triggering models. The disagreements between models is a result of incomplete datasets and parameter spaces for model development. The Next Generation Liquefaction (NGL) Project was created to provide a database for advancing liquefaction research and to develop models for the prediction of liquefaction and its effects, derived in part from that database in a transparent and peer-reviewed manner, that provide end users with a consensus approach to assess liquefaction potential within a probabilistic framework. An online relational database was created for organizing and storing case histories which is available at ( , [1]). The NGL field case history database was recently expanded to include the results of laboratory testing programs because such results can inform aspects of liquefaction models that are poorly constrained by case histories alone. Data are organized by a schema describing tables, fields, and relationships among the tables. The types of information available in the database are test-specific and include processed-data quantities such as stress and strain rather than raw data such as load and displacement. The database is replicated in DesignSafe-CI [2] where users can write queries in Python scripts within Jupyter notebooks to interact with the data.

Kenneth S. Hudson, Paolo Zimmaro, Kristin Ulmer, Brian Carlton, Armin Stuedlein, Amalesh Jana, Ali Dadashiserej, Scott J. Brandenberg, John Stamatakos, Steven L. Kramer, Jonathan P. Stewart
Liquefaction Characteristics of 2011 New Zealand Earthquake by Cone Penetration Test

The liquefaction-induced damage is one of the most serious damage in the earthquake. At present, the discrimination of sand liquefaction is one of the most effective and direct methods to prevent liquefaction-induced damage. Investigation of liquefied sites and data collection are the basis of the formation of the liquefaction discrimination method. The 2011 Christchurch Mw 6.3 earthquake caused large-scale liquefaction. The earthquake is the first earthquake in which sand liquefaction is the main cause of seismic damage and provides a mass of liquefaction data. This paper collects liquefaction data of 132 survey points after the earthquake. Through the analysis of the collected static cone penetration test data, groundwater level, peak acceleration and other data, the liquefaction characteristics of the earthquake are obtained. At the same time, it was discovered that multiple deep liquefaction phenomena occurred in the earthquake. By testing the existing CPT liquefaction evaluation method in Chinese code using the collected data, it is found that the method has obvious errors in the evaluation of deep sand liquefaction.

Zhao-yan Li, Si-yu Zhang, Xiao-Ming Yuan
Insights from Liquefaction Ejecta Case Histories from Christchurch, New Zealand

Liquefaction ejecta were a primary contributor to the damage of land and light-weight structures in Christchurch from the 2010–2011 Canterbury earthquakes. It occurred predominantly in east Christchurch, which is typically characterized by thick, clean sand deposits. By contrast, ejecta tended to be absent in the stratified silty soil swamp deposits of west Christchurch. To advance understanding of ejecta production, 235 well-documented case histories of ejecta occurrence, its quantity, and its effects on infrastructure were developed. The ejecta database includes 61 sites that underwent four major earthquakes that produced no-to-extreme ejecta. The case histories take advantage of numerous CPTs, boreholes, pre- and post-earthquake airborne LiDAR surveys, aerial photographs, liquefaction land damage documentation from insurance claims, and earthquake-specific PGAs and groundwater estimates. Ejecta coverage and amounts for each of the four main Canterbury earthquakes were extracted using photographic- and LiDAR-based approaches because direct measurements of ejecta quantities were not made. The database provides the basis for the development of procedures to evaluate the occurrence and quantity of ejecta.

Z. Mijic, J. D. Bray, S. van Ballegooy
Chilean Liquefaction Surface Manifestation and Site Characterization Database

Seismic sources differ significantly from great megathrust to shallow crustal earthquakes in several aspects, yet we analyze them in many engineering applications as if they were similar. This work is framed within the study of the liquefaction behaviour of sites subjected to subduction events. To pursue the analysis of subduction triggered liquefaction, we must first have a database with observations of surface manifestation of liquefaction on such events, seismic demand, and site characterization. We present a database focused on the last three large events that affected Chile since 2010, but include some sites from other subduction events as well. This database has more than 200 sites, including surface manifestation of liquefaction; geotechnical data such as CPT and SPT logs, grain size distribution, and Atterberg limits; geophysical data including shear-wave velocity profiles, dispersion curves, and horizontal to vertical spectral ratios; surface intensity estimations at each site including PGA, PGV, and spectral ordinates.

Gonzalo A. Montalva, Francisco Ruz, Daniella Escribano, Felipe Paredes, Nicolás Bastías, Daniela Espinoza

S3: Special Session on Embankment Dams

Deterministic Seismic Hazard Analysis of Grand Ethiopia Renaissance Dam (GERD)

This paper presents the Deterministic Seismic Hazard Analysis for Grand Ethiopia Renaissance Dam (GERD) (11.2183ºN, 35.0941ºE). This area is of special importance because GERD is the largest hydroelectric power plant in Africa and the seventh-largest in the world. The Hazard analysis was done using linear sources present near the dam site (Inferred faults and Indicated faults); 713 faults were used for the analysis. Earthquake data was collected for the dam region, and homogenization of data was done to moment magnitude followed by declustering foreshock and aftershock events. 1375 earthquakes of moment magnitude 4–7 were considered and represented on a seismotectonic map. Ground Motion Prediction Equations (GMPEs) were identified for the dam site, and suitability was checked by comparing the Data Support Index (DSI) value of various prediction equations. Considering the selected GMPEs and input data, DSHA was performed by using a MATLAB program developed for this purpose. The dam site was further divided into a grid cell of 100 m × 100 m, and hazard parameters were obtained at the midpoint of the grid cells, bearing in mind all seismic sources within a 500 km radius. Obtained PGA values were compared with USGS developed Instrumental Intensity scale. The hazard map showing the spatial variation of the risk associated with the dam site is presented.

Mohammed Al-Ajamee, Abhishek Baboo, Sreevalsa Kolathayar
Parametric Study of Seismic Slope Stability of Tailings Dam

Tailings are the byproduct of the mining industry. The minerals mined by these mining industries are only about 3 to 5% pure, so the rest, 97 to 95%, become the tailings. It is being said that the increasing population has increased the demand for minerals for various uses. This demand produces a massive volume of tailings that, when disposed of inadequately, causes several failures and has cost lives in some cases. Therefore, a proper study of the design and safety of these structures is needed. This study aims to understand the stability variation of different tailings construction methods with different seismic loads. In this paper, the seismic slope stability of tailings dam is done by varying the slope angle of each dyke and material properties for three methods of construction – upstream, centerline, and downstream method of construction under different seismic loading. The numerical study is done in GeoStudio software.

T. S. Aswathi, Ravi S. Jakka
Seismic Performance of a Zoned Earth Dam

2D dynamic numerical analyses of a zoned earth dam located in a high seismic hazard area of Southern Italy have been carried out. In the analyses soil behaviour has been described through an isotropic hardening elasto-plastic hysteretic model. A set of 47 real accelerograms recorded on rock outcrop has been used as input motion. The activation of plastic mechanisms induced by the seismic loading and the amplification of the horizontal accelerations in the dam are investigated. Also, assuming that the vertical settlement wc of the crest of the dam can be regarded as an index of the seismic performance of the dam, the effects of various parameters of the input motion on wc have been examined.

Orazio Casablanca, Andrea Nardo, Giovanni Biondi, Giuseppe Di Filippo, Ernesto Cascone
Evaluation of the Seismic Performance of Small Earth Dams

The evaluation of the seismic response of earth dams is a nontrivial task, which generally requires the use of advanced soil models able to accurately reproduce the material behaviour under dynamic loadings. The reliability of the numerical simulations is however constrained by the level of knowledge of the geotechnical model of the dams. The issue is particularly relevant when small earth dams, characterized by a reduced height and a limited reservoir volume, are considered. Such structures indeed frequently lack proper characterization of the materials constituting the dam body and its foundation. The implementation of dynamic numerical analyses is therefore limited in the common practice and the seismic performance of the dams are frequently assessed through simplified empirical methods. This study investigates the seismic behaviour of two small earth dams for which a reliable geotechnical model, based on both laboratory and in situ tests, is available. The seismic responses of the dams are analyzed through fully coupled effective stress dynamic analyses. The analyses are developed within the context of the ReSba European project for the French-Italian Alps area. The results have allowed comparing the seismic performance of the structure as predicted by simplified and advanced approaches in terms of stability conditions and seismic-induced settlement of the crest.

Andrea Ciancimino, Renato Maria Cosentini, Francesco Figura, Sebastiano Foti
Liquefaction Potential for In-Situ Material Dams Subjected to Strong Earthquakes

Dam safety assessment is a crucial task to ensure the safety of the existing or the new dam which can be subjected to high risks such as heavy floods and/or strong earthquakes. In some old existing dams in Vietnam, the filling and foundation materials were consisted of sand which could possibly trigger a liquefaction when subjected to strong earthquakes. Thus, liquefaction potentials of some in-situ material large dams in the west-north of Vietnam were analyzed based on the seismic data, dynamic soil properties and finite element modeling with various scenarios regarding earthquake return periods and reservoir water levels. The simulation results suggested that the sand zones in the upstream dam body foundation and under the downstream drainage rockfill could be liquefied when subjected strong earthquakes. The extent of liquefied zone increased with the increase of the peak ground acceleration and reservoir water levels.

Hong Nam Nguyen
Seismic Assessment of a Dam on a Clayey Foundation

The seismic performance of a rockfill dam founded on a 20 m clayey foundation layer is presented. This stratum is the foundation of a 45 m high rockfill dam. A specific site seismic hazard study indicates that a Mw 8.0 event with peak ground acceleration of about 0.45 g is appropriate for design. The final aim of this ongoing study is to study the fragility of the dam-foundation system. The geotechnical characterization of the foundation involved the in-situ measurement of index properties (e.g., water content and PI) and the retrieval of undisturbed samples. Next, a battery of compressibility and strength testes were performed. The inferred compressibility characteristics of the clay layer and the measured monotonic stress-strain curves allowed the calibration of the modified Cam-Clay model. For the dynamic evaluation, a series of monotonic and cyclic direct simple shear tests results representative of this clay layer were collected to calibrate the PM4Silt model. Shear wave velocity profiles allowed the dynamic characterization of the core material and rockfill shells for which the UBCHyst model was employed. The static and dynamic analysis results are presented in terms of permanent deformations for design level motions.

Franklin R. Olaya, Luis M. Cañabi
Seismic Fragility Analysis of Two Earth Dams in Southern Italy Using Simplified and Advanced Constitutive Models

The seismic performance of earth dams can be assessed using various procedures spanning from simple deterministic methods to fully-probabilistic approaches, such as fragility analyses. In this study, analytical seismic fragility functions (i.e., based on results of numerical analyses) for two earth dams (the Farneto del Principe and the Angitola dams) in the Calabria region (Southern Italy) are developed. The Farneto del Principe dam does not have liquefaction-related issues, while the Angitola dam is founded on soils susceptible to liquefaction. The analyses are performed using numerical simulations based on the so-called multiple-stripe method. This framework takes inputs from site-specific probabilistic seismic hazard analysis results and it is used here to develop fragility functions for intensity measure (IM) values at the same return period of the seismic action. Separate calculations are presented, using the relatively simple hysteretic soil model implemented in the 2D finite difference method software FLAC and two advanced constitutive models (PM4Sand and PM4Silt) specifically developed to simulate the stress-strain response of sands and silts in geotechnical earthquake engineering applications. Fragility functions for several damage mechanisms and IMs are obtained. The analyses show that the IMs with the highest predictive power (i.e., those with the lowest fragility function standard deviations) are the peak ground acceleration and velocity (PGA and PGV) for the Farneto del Principe dam and the cumulative absolute velocity (CAV), the cumulative absolute velocity after application of a 5 cm/s2 threshold acceleration (CAV5), and PGV for the Angitola dam.

Gianluca Regina, Paolo Zimmaro, Katerina Ziotopoulou, Roberto Cairo
Comparison of the Monotonic and Cyclic Response of Tailings Sands with a Reference Natural Sand

This article considers the monotonic and cyclic behaviour of a tailings sand material, tested over a wide range of confining pressures and relative densities. The collated experimental data are used to interpret the behaviour of this material in the framework of Critical State Soil Mechanics (CSSM) and to calibrate an advanced bounding surface plasticity model. The response of this man-made material is then contrasted, within the same framework, to that of Toyoura sand, which is a well-characterised reference natural sand.This process aims to identify the principal behavioural differences between the two types of geo-materials and the implications that these may have on the calibration process of an advanced constitutive model. Ultimately, the process also supports the numerical modelling of Tailings Storage Facilities under both monotonic and cyclic/earthquake loading.

David Solans, Stavroula Kontoe, Lidija Zdravković
Seismic Performance of a Bituminous-Faced Rockfill Dam

Assessing the seismic safety of an existing earth dam is much more cumbersome than the design of a new structure since several actions and internal processes might have occurred over time. A detailed knowledge of both dam and foundation soils together with a proper characterization of the embankment response under the actual past static loading history, is a prerequisite before handling the dam seismic assessment through a performance-based approach. The paper describes the analysis procedures developed for analyzing the static and seismic response of a bituminous upstream-faced rockfill dam placed in Sicily (IT). The outcomes of the numerical study, performed through a complete dynamic formulation, are presented and discussed focusing on the embankment settlements at the end of the construction stage and on the seismic performance of the dam under the Collapse Limit State scenario. The paper tries to shed light on the role of a strong heterogeneity (limestone block) detected within the foundation soil to ascertain if its presence, which was completely disregarded during the original design of the dam, is beneficial or detrimental for the dam overall response.

M. Tretola, E. Zeolla, S. Sica
Liquefaction-Induced River Levee Failure at Miwa of Inba Along Tone River During 2011 Tohoku Earthquake

A numerous number of river levee failures were observed in Kanto region far from the epicentre during 2011 Tohoku earthquake. In this study, the earthquake reconnaissance investigations were carried out on the liquefaction-induced sliding failure of the river levee located at Miwa of Inba along Tone river. A series of weight sounding tests were conducted at a dozen of locations near the failed river levee. Soil sampling was also carried out at each location of weight sounding, and laboratory sieve and sedimentation analyses were performed to determine the grain size distributions of the soil samples. At this particular site, the outer face of the river levee slid laterally, and the evidence of sand boils was clearly found at the foot of the river levee. The liquefaction resistance of soils was estimated based on the results of weight sounding and soil sampling, by using the empirical formula developed by the authors, which explicitly incorporates the effect of fines. The values of liquefaction resistance of foundation soils at which the river levee failed and did not fail are estimated, and the cause of the sliding failure is discussed in detail.

Yoshimichi Tsukamoto, Naoki Kurosaka, Shohei Noda, Hiroaki Katayama

S4: Special Session on Earthquake Disaster Risk of Special Soil Sites and Engineering Seismic Design

Effect of Rice Husk Ash and Sisal Fibre on Strength Behaviour of Soil

Due to an increase in population, sustainable development, climate change, construction, waste management, and production of agriculture being the major challenges to be solved by researchers and scientists since the depletion of non-renewable resources has rapidly also increased. In this study the use of rice husk ash and treated sisal fibre on the strength behaviour of soil will help in the reduction of effects caused by rice husk, also increasing space for agricultural production. Rice husk ash (RHA) and single length sisal fibre (SF) (10–15 mm) at 2.5, 5.0, and 7.5% of RHA and 0.3, 0.6, and 0.9% of SF were used for this research investigation, and the optimum value of 7.5% RHA and 0.3% SF by weight of soil sample resulting to cohesion value which was over fifty (50) times the value natural soil (from 0.13 kg/cm2 to 6.71 kg/cm2) and also the angle of internal friction increased three times the natural soil value (from 26° to 72°).This research also helps to obtain SDG (1, 2, 6, 9, 12, 13, 14, and 15).

Benard Obbo, Gideon Okurut, Pragnesh J. Patel, Nisha. P. Soni, Darshil V. Shah
Cyclic Behaviour of Brumadinho-Like Reconstituted Samples of Iron Ore Tailings

These days, there is an urgent call for a transition to a new low-carbon energy paradigm enabling sustainable development. This is expected to increase the demand for minerals and metals in the next decades and intensify the already massive and problematic problem of mining waste, which represents a huge proportion of the excavated material and is often accumulated in large tailings dams that have caused serious disasters around the world. To promote future sustainable mining and ensure the safety of tailings deposits, namely in seismically active regions, performance-based design needs to be applied in practice based on suitable geotechnical-based characterization of mining waste materials. This paper focus on the assessment of the undrained behaviour of iron ore tailings, which has been observed as a major factor in historic disasters in the recent past, including Brumadinho’s failure in Brazil. Samples reconstituted through a slurry-based method and tested in undrained triaxial monotonic and cyclic conditions are used to investigate the response of Brumadinho-like reconstituted iron ore tailings. The results show that the samples have physical and identification properties similar to those in situ, the mechanical behaviour obeying to the principles observed in granular soils tested in similar conditions. In fact, phase transformation and critical state lines are suitably defined in undrained monotonic and cyclic loading, particular features being justified by the high angularity of crushed particles.

Paulo A. L. F. Coelho, David D. Camacho, Felipe Gobbi, Luís M. Araújo Santos
Dynamic Properties of Organic Soils

A database of laboratory stiffness degradation data of 30 samples, cering a wide range of soil types and percentage of organic content, was assembled to allow a thorough investigation of the dynamic properties of organic soils. The collated datasets are first compared with existing, commonly used empirical, hyperbolic stiffness degradation and damping curves, which are expressed mainly as a function of plasticity index and effective confining pressure. The identified discrepancies between the collated data and the empirical curves of the literature were ascribed to the presence of organic matter, leading to the introduction of a new shape variable through nonlinear regression. Based on statistical analysis of the collated information a new empirical model for organic soils is proposed, describing the reduction of normalised shear modulus and variation of damping ratio with cyclic shear strain as function of organic content in addition to other key parameters (mean effective confining pressure, plasticity index, number of cycles and loading frequency). The proposed empirical model can be used in engineering practice to infer the dynamic properties of organic soils in the absence of site-specific data.

Vincenzo d’Oriano, Stavroula Kontoe
Experimental Study on the Correlation Between Dynamic Properties and Microstructure of Lacustrine Soft Clay

Due to the influential mesostructure of lacustrine soft clay on the dynamic characteristic parameters, the dynamic shear modulus and damping ratio of the lacustrine soft clay are obtained by selecting different consolidation confining pressures for indoor resonance column test and scanning electron microscope (SEM) test, and the microstructure parameters of the lacustrine soft clay are acquired by analyzing the SEM image through the IPP (Image-Pro Plus) image processing software as well. By analyzing the relationship between some dynamic characteristic parameters and microstructure parameters is analyzed to explain the dynamic shear modulus and damping ratio from the perspective of soil microstructure characteristics. It is shown that the mean maximum dynamic shear modulus has a strong correlation with the maximum radius of the particles and the shape of the soil pores (abundance, circularity, fractal dimension), indicating that the size of the soil particles and the shape of the pores have a strong influence on dynamic shear modulus of soil; the correlation between the maximum damping ratio and the microstructure characteristics is weak, and the influence law is more complicated.

Yurun Li, Zhongchen Yang, Jingjuan Zhang, Yingtao Zhao, Chuang Du
Small-Strain Shear Modulus of Coral Sand with Various Particle Size Distribution Curves

Particle size distribution plays an important role in the small-strain shear modulus G0 of granular soils. Numerous expressions were proposed for predicting the G0 of siliceous sands in the literature. There are significant differences between the mechanic characteristics of coral sand and siliceous sand. In this paper, a series of resonant column tests were conducted on the coral sand with various particle size distribution curves. The influences of uniformity coefficient Cu, mean-grain size d50, and fines content Fc on G0 of coral sand are investigated. For the given void ratio and confining stress, the G0 of coral sand is smaller than that of siliceous sand. The G0 of coral sand decreases with the increase of Cu, increases with the increase of d50, and first decreases then increases with the increase of Fc. The minimum void ratio emin, instead of the Cu, d50, and Fc, is adopted as a unified index for capturing the characteristics of the particle size distribution of granular soils. The relationships between the parameters in Hardin’s equation and emin are established and a minimum void ratio-based G0 expression for coral sand is proposed. The validity of the new expression for the other types of granular soils is confirmed using the G0 data in the literature.

Ke Liang, Guoxing Chen, Qing Dong
Flowability of Saturated Calcareous Sand

The potential for damaging flow liquefaction failure of calcareous sands subject to seismic, wave, or other dynamic loading is of increasing importance owing to its predominance along coastal areas and increased use as a fill material. Undrained cyclic triaxial tests were performed on saturated calcareous sand specimens prepared with different relative densities and subjected to various effective confining pressures to study the flowability of viscous, liquefied calcareous sand. The cyclic shear stress-strain rate relationship for the saturated calcareous sand specimens transitioned from an elliptical shape to a dumbbell shape as excess pore pressures accumulated under cyclic loading. The dumbbell-shaped relationship demonstrates that the saturated calcareous sand exhibited low shearing resistance and high fluidity under elevated excess pore pressures for certain conditions evaluated. The average flow coefficient, defined as the maximum shear strain rate triggered by the unit average cyclic shear stress, and the flow curve describing the flowability, of the saturated calcareous sand are used to quantify the cyclic failure potential of the calcareous sand. The relative density and cyclic stress ratio has a significant influence on the average flow coefficient: the smaller the relative density and larger the cyclic stress ratio, the smaller the number of cycles to failure. In contrast, the effective confining pressure has little effect on the magnitude of flow potential or number of cycles to triggering flow-like behavior.

Lu Liu, Jun Guo, Armin W. Stuedlein, Xin-Lei Zhang, Hong-Mei Gao, Zhi-Hua Wang, Zhi-Fu Shen
Liquefaction Characteristics of Saturated Coral Sand Under Anisotropic Consolidations

A series of undrained cyclic triaxial tests were conducted on saturated coral sand by using GDS dynamic traixial apparatus. This research mianly investigates the characteristics of excess pore water pressure (ue), axial strain (εa), stress-strain relationship and liquefaction resistance of coral sand under different consolidation ratios (R) and cyclic stress ratios (CSR). The test results indicate that for the specimens under anisotropic consolidations, the development of pore water pressure always show the trend from fast to slow, and eventually tends to be stable. Two development modes of axial strain can be found in saturated coral sand: cyclic mobility and plastic cumulative deformation. The stress-strain relationships of saturated coral sand under isotropic and anisotropic consolidations are quite different, and the difference could be associated with stress reversal. The relation of static deviatoric stress (qs) and cyclic deviatoric stress (qd) plays an important role in liquefaction characteristics. Compared with the Fujian sand, the saturated coral sand has higher liquefaction resistance due to the special physical properties. The liquefaction resistance of saturated coral sand increased with the increasing of consolidation ratio.

Weijia Ma, Guoxing Chen
Microscopic Test Analysis of Liquefaction Flow of Saturated Loess

On the basis of existing research, triaxial shear tests were carried out on the remolded samples after liquefaction. The research shows that the apparent viscosity of the remolded sample has been increasing, indicating that the flow characteristics have been “shear thickening”. The samples in different test stages were tested by scanning electron microscope, and the microstructure parameters of the two samples in different test stages were obtained. The analysis shows that the difference of particle and pore size between undisturbed loess and remolded loess leads to the rising mode of pore water pressure in the process of liquefaction test, and then affects the change of apparent viscosity in the process of shear.

Xingyu Ma, Lanmin Wang, Qian Wang, Ping Wang, Xiumei Zhong, Xiaowu Pu, Fuqiang Liu
Geotechnical Behavior of the Valley Bottom Plain with Highly Organic Soil During an Earthquake

The valley bottom plains on the plateau around the center of Tokyo form a dendritic pattern. During the Great Kanto Earthquake of 1923, destruction of houses and damage to buried pipes were concentrated along the Kanda River and the valley bottom plain of Tameike. This damage might have been caused by the amplification of seismic motions due to the presence of soil with abundant organic matter and soft clay near the ground surface in the valley bottom plain and the unconformity of the valley bottom. However, the geological structure of the valley (including its width and depth) varies greatly depending on the formation process and location, and it is unknown how these factors affect the seismic response. The purpose of this study was to investigate the effect of the ground structure on the seismic response in a valley bottom plain using the two-dimensional finite element method for seismic response analysis (FLUSH). The study area consisted of three cross-sections with different valley widths, specifically, upstream, midstream, and downstream sections in the valley bottom plain of Magome, Ota-ku, where the thickest organic-rich soil layer (9 m) in Tokyo was deposited. The results showed that the different valley widths affected the position and magnitude of the maximum velocity response at the valley floor and the magnitude of the horizontal ground strain. In the midstream section with medium valley width, the maximum amplitude of the ground strain within the ground was approximately 0.2–0.47%, which could cause damage to buried pipes.

Motomu Matsuhashi, Keisuke Ishikawa, Susumu Yasuda
Development of Constitutive Model Describing Unsaturated Liquefaction Characteristics of Volcanic Ash Soil

In this paper, the numerical simulations were conducted for reproducing the tendency obtained in the unsaturated liquefaction tests (cyclic tri-axial tests) using volcanic soils damaged in the 2018 Hokkaido Iburi eastern earthquake. In the numerical simulations, the three-phase porous media theory was applied. However, the simulated results underestimated the unsaturated liquefaction resistance in unsaturated conditions because the constitutive models used in the simulations were developed for describing the tendency of liquefaction characteristics for saturated soils. Therefore, the evolution rules focused on the change of void ratio were proposed for modifying the dilatancy characteristic of unsaturated volcanic ash soils. The simulated results using proposed method could successfully reproduce the tendency obtained in the tests.

Takaki Matsumaru, Toshiyasu Unno
Field Observations and Direct Shear Tests on the Volcanic Soils Responsible for Shallow Landslides During 2018 Hokkaido Eastern Iburi Earthquake

A numerous number of landslides were observed in the local town of Atsuma during 2018 Hokkaido Eastern Iburi Earthquake. The area of landslides was found the largest ever recorded in Japan during the last 150 years. It was noteworthy that there were many landslides observed on relatively gentle slopes of less than 20°, and most of the landslides were shallow-seated. This area is known to be covered by several layers of volcanic pyroclastic fall deposits, intervened with their weathered clayey and loam soils. The effects of rainfall recorded one day before the earthquake may not be negligible for triggering of so many shallow slips of the local volcanic soil deposits. In this study, field investigations were conducted to scrutiny the layers of soil deposits responsible for shallow landslides, and two different soil samples of Shikotsu and Tarumae-d pumice fall deposits were retrieved. Multiple series of volume-constant direct shear tests were then conducted on these two volcanic soils, and the effects of degree of saturation were especially examined. In order to mechanically characterise the crushability of particles of these pumice fall deposits, a series of particle crushing tests were also conducted.

Shohei Noda, Hiroya Tanaka, Yoshimichi Tsukamoto
Investigation on Saturation State of Loess Using P-wave Velocity

Pore pressure parameter, also known as B-value, is used to characterize the saturation state of soils. Recent advances have established a relationship between B-value and P-wave velocity based on the poroelasticity theory, hence it sheds a light to evaluate the degree of saturation of soils using P-wave velocity. Previous studies in this regard often focus on granular materials, the feasibility of using the relationship for loess was less extensively studied. This paper presents experimental results on the P-wave velocity of loess using a pair of piezoceramic extender elements at a sequence of saturation states. It was found that, the P-wave velocity do not change with excitation frequency. At an excitation frequency of 5 kHz, the wave signal is most clear. Besides, the effects of initial water content and sand content on P-wave velocity were both examined. Though the P-wave velocity increases with the B-value, that is approximately in line with theoretical predictions, departures from predictions were observable at higher B-values. The outcome of this study suggests a potentially good use of P-wave velocity as an indicator of saturation in loess.

Zehua Qin, Yuchuan Wang, Xin Liu
Relationship Between Shear Wave Velocity and Liquefaction Resistance in Silty Sand and Volcanic Sand

Quantification of influencing factors to liquefaction is essential for accurate prediction of in-situ liquefaction resistance. Previous studies have shown that there is a significant relationship between shear wave velocity and liquefaction resistance for a limited number of samples under the same density as in-situ with different soil fabric. However, samples such as silty sand or volcanic sand have not been investigated. In this study, reconstituted specimens were prepared using in-situ samples collected from two locations, and shear wave velocity measurements and undrained cyclic loading tests were conducted. The results show that shear wave velocity and liquefaction resistance rise with increasing over-consolidation ratio, and when they are normalized to specimens without over-consolidation history, the results are in good agreement with previous trends.

Masataka Shiga, Takashi Kiyota
Small to Medium Strain Dynamic Properties of Lanzhou Loess

Loess is a kind of world-recognized problematic soil and, as a consequence, it is a major hazard in geotechnical engineering. This study examines the small to medium strain dynamic properties of Lanzhou intact and recompacted loess specimens from different burial depths through a set of resonant column (RC) tests. The influence of soil structure on the initial shear modulus G0, the variations of G/G0 and damping ratio DT (%) with shear strain γ were analyzed and discussed, including also comparisons between undisturbed and recompacted loess. The test results show that the structure effect has an important influence on the dynamic properties of dry loess at small and medium strain, whereas, with the increase of moisture, the structure effect weakens, causing the dynamic properties (i.e., G0, G/G0-γ and DT (%)-γ) of undisturbed loess are comparable to that of recompacted ones. Finally, it is proved that a normalized correlation between G/G0 and γ/γ0.7, where γ0.7 is a reference soil strain is practically identical for undisturbed and recompacted loess in Lanzhou, irrespective of soil structure, which may be useful in practical applications.

Binghui Song, Angelos Tsinaris, Anastasios Anastasiadis, Kyriazis Pitilakis, Wenwu Chen
Cyclic Simple Shear Tests of Calcareous Sand

A series of simple shear tests under different test conditions were carried out on calcareous sand from the South China Sea. Notably, constant stress and volume tests under monotonic and cyclic testing conditions were conducted. This study was focused on the simple shear behavior of calcareous sand and the relationship between the relative breakage Br and the input energy E. The test results showed that the calcareous sand has obvious particle breakage during the simple shear test procedure, and there is a unique relationship between the relative breakage Br and the input energy E at the end of the tests, regardless of the testing conditions. In addition, the study also found that a unique critical state line (CSL) exists for both constant stress and volume monotonic and cyclic tests, and there is a unique cyclic phase transformation line (CPTL) for constant volume cyclic tests. The volumetric strain accumulation increases with cycle number, vertical stress, and cyclic shear strain during constant stress cyclic tests.

Kai-Feng Zeng, Hua-Bei Liu

S5: Special Session on Soil Dynamic Properties at Micro-scale: From Small Strain Wave Propagation to Large Strain Liquefaction

Effect of Fabric Anisotropy on Reliquefaction Resistance of Toyoura Sand: An Experimental Study

Recent earthquakes in New Zealand and Japan demonstrated that preshaking histories can significantly influence the liquefaction resistance of sandy soils, which has not been fully understood yet. It has also been reported that soils prepared with different depositional methods referred to as initial or inherent anisotropy show different liquefaction behavior. Therefore, a comprehensive experimental study was carried out for the first time to explore the combined effects of inherent anisotropy and induced anisotropy due to cyclic shearing on the reliquefaction resistance of loose (Dr = 45%) and dense (Dr = 70%) Toyoura sands subjected to various cyclic stress ratios (CSR). A hollow cylinder torsional shear apparatus (HCTSA) was used in this study since it can mimic the field stress conditions more realistically during earthquake events. In order to investigate the influence of inherent anisotropy on liquefaction and reliquefaction resistance of Toyoura sand, the hollow cylindrical specimens were reconstituted with different methods of dry deposition (DD) and moist tamping (MT), representing naturally and artificially deposited soils, respectively. Furthermore, sand specimens were cyclicly sheared up to medium and large shear strain levels, then reconsolidated at different states to evaluate the effect of induced anisotropy with different shear histories on reliquefaction resistance in the sand. A large preshearing reduces the reliquefaction resistance, while the reliquefaction resistance increases when the sandy soil is moderately presheared, irrespective of relative density and reconsolidation state. It was also found that the effect of reconsolidation state is more significant in the dense sand than the loose sand. Another key finding of this study is that the effects of initial fabric acquired by different reconstitution methods and reconsolidation states are lost in the reliquefaction tests once the loose Toyoura sands experienced large preshaking.

Pedram Fardad Amini, Gang Wang
Variation of Elastic Stiffness of Saturated Sand Under Cyclic Torsional Shear

For soil elements surrounding the foundations of offshore wind turbines or oil platforms, very many loading cycles of varying amplitudes and frequencies are quite common. The influence of such cyclic loadings on the soil stiffness is a critical concern in the evaluation of the long-term performance of the foundation-structure system during its service life. Current understanding of this influence is however inadequate due to the scarcity of experimental data. This study investigates the variation of small-strain shear modulus (G0) of saturated Toyoura sand subjected to cyclic torsional shear stress cycles with a small amplitude by using a RC/TS apparatus. Sand specimens of different densities have been tested. It is found that the G0 value tends to decrease with loading cycles and the decrease is mainly associated with the initial loading cycles. When the number of cycles is beyond around 3500, the G0 value tends to become stable. The degree of reduction of G0 is not sensitive to the density of sand specimen and is generally within 10% for the range of densities investigated.

Yutang Chen, Jun Yang
Liquefaction Resistance and Small Strain Shear Modulus of Saturated Silty Sand with Low Plastic Fines

The liquefaction resistance and the small strain shear modulus G0 of saturated Fujian sand with low plastic fines were evaluated with cyclic triaxial and bender element tests. The testing program encompasses a wide range of initial void ratios, confining stresses, and fines content from 0% to 30%. The results show that the influence of fines content on the liquefaction resistance and G0 can be characterized predominantly by the equivalent granular void ratio (e*). The liquefaction resistance of Fujian sand with Shanghai silt decreases as e* increases, which resembles the results of sands containing non-plastic fines. Furthermore, a notable relationship between the liquefaction resistance and the G0 was presented, which can potentially be used to evaluate the liquefaction resistance of silty sand using G0.

Xiaoqiang Gu, Kangle Zuo, Chao Hu, Jing Hu
Volumetric Strains After Undrained Cyclic Shear Governed by Residual Mean Effective Stress: Numerical Studies Based on 3D DEM

A series of numerical tests is performed using the three-dimensional discrete element method to study the volume contraction characteristics of granular material during the reconsolidation process following undrained cyclic shear. Results show that post-liquefaction reconsolidation can be categorized into a liquefied and a solidified portion. The decrease in the void ratio in the liquefied portion is not accompanied by an increase in the mean effective stress, while the opposite is true for the solidified part. The residual mean effective stress affects the volumetric strain significantly during reconsolidation. The decrease in void ratio during the reconsolidation beginning from effective stress reduction ratios of 0.1, 0.5, and 0.9 is 86%, 37%, and 7% of that beginning from the initial liquefaction state, respectively. In addition, the volumetric strain during reconsolidation is associated with the change in pore uniformity.

Mingjin Jiang, Akiyoshi Kamura, Motoki Kazama
One-Dimensional Wave Propagation and Liquefaction in a Soil Column with a Multi-scale Finite-Difference/DEM Method

The paper describes a multi-phase, multi-scale rational method for modeling and predicting the free-field wave propagation and liquefaction of soils. The one-dimensional time-domain model of a soil column uses the discrete element method (DEM) to track stress and strain within a series of representative volume elements (RVEs), driven by seismic rock displacements at the column base. The RVE interactions are unified with a time-stepping finite-difference algorithm. The Darcy's principle is applied to resolve the momentum transfer between a soil's solid matrix and its interstitial pore fluid. The method can analyze numerous conditions and phenomena, including site-specific amplification, down-slope movement of sloping ground, dissolution or cavitation of air in the pore fluid, and drainage that is concurrent with shaking. Several refinements of the DEM are necessary for realistically simulating soil behavior and for solving a range of propagation and liquefaction factors: most importantly, the poromechanic stiffness of the pore fluid and the pressure-dependent stiffness of the grain matrix. The model is verified with successful modeling-of-models simulations of several well-document centrifuge tests. The open-source DEMPLA code is available on the GitHub repository.

Matthew R. Kuhn
Microscopic Insight into the Soil Fabric During Load-Unload Correlated with Stress Waves

Soils have a complex contact network (i.e. soil fabric), which affects the mechanical characteristics (e.g. stiffness anisotropy, shear strength) and further determines the performance in geotechnical constructions. Conventional laboratory tests on soils can give macro-scale mechanical responses as a physical relevance, while the discrete element method (DEM) has been used as a numerical tool to gain insights into micro-scale mechanical characteristics such as soil fabric. In this study, laboratory element tests are conducted using spherical glass beads subjected to several cycles of load-unload reversals at selected pre-peak strain ranges, where stress waves are continuously measured during the loading process. DEM is adopted using spherical particles to simulate the equivalent load-unload reversals and wave propagations. Based on the experimental and DEM results, the compression velocity (Vp) and shear wave velocity (Vs) are found to be approximately reversible when the load-unload cycles are at low stress levels, while a significant reduction in both velocities can be observed at higher stress levels. The variation of soil fabric anisotropy during load-unload reversals is explored using DEM; the fabric anisotropy increases markedly beyond a stress ratio threshold of 1.8 and cannot be fully recovered during the subsequent unloading process. Fabric anisotropy is found to be linearly correlated with wave velocity ratio (Vp/Vs) at the pre-peak stage, although both quantities are respectively affected by the given stress states.

Yang Li, Masahide Otsubo, Reiko Kuwano
Shear Work and Liquefaction Resistance of Crushable Pumice Sand

Crushable volcanic soils, such as pumice sands, are often encountered in engineering projects in the North Island of New Zealand. Due to the highly crushable nature of the pumice sand components, current empirical correlations, derived primarily from hard-grained sands, are not applicable when evaluating the liquefaction potential of pumice-rich soils. To better understand their liquefaction characteristics, cyclic undrained triaxial tests were performed on high-quality undisturbed soil samples sourced from various pumice-rich sites in the North Island. The undrained response, expressed in terms of the development of excess pore water pressure and axial strain with the number of cycles, and the shear work (or cumulative dissipated energy), defined as the energy consumed by the soil during plastic deformation until liquefaction, is examined vis-à-vis the pumice contents of the specimens. When compared to published trends for normal sands available in the literature, the results indicate that the shear work for pumice-rich sands sand is larger in specimens with higher pumice contents because some energy is spent as the particles undergo crushing. As a result, the liquefaction resistance of crushable pumice sand is higher than that of natural sand for the same level of loading applied.

Rolando P. Orense, Jenny Ha, Arushi Shetty, Baqer Asadi
Digital Particle Size Distribution for Fabric Quantification Using X-ray μ-CT Imaging

Past research works on monotonic and cyclic shear behavior of natural silts at the University of British Columbia (UBC), Canada, has shown that the soil fabric and microstructure has a significant influence on the mechanical response of natural silts. With this background, a research program has been undertaken at UBC to capture non-destructive 3D images of Fraser River low-plastic silt using X-ray micro-computed tomography (X-ray µ-CT) for assessing particle fabric. In this regard, identifying individual grains to analyze the particle size distribution (PSD) of a given soil specimen forms a vital step in studying the fabric of silts.Advancements in computing power have allowed for 3D X-ray imaging of fine-grained soil (silt) specimens. Through computer processing of these 3D images, it is possible to obtain the main dimensional and directional parameters to represent individual particles in digital form; in turn, some of this information can be used to obtain a digital PSD of a given silt matrix. A good way to assess the outcomes of the image processing approach is by comparing the PSDs from the image-based analysis to those from mechanical methods and laser technologies.In the present work, specimens from a natural silt are analyzed using laboratory methods (i.e., mechanical sieve along with hydrometer analysis) and laser diffraction techniques to establish the benchmark PSDs, and the results are compared with those digitally derived from X-ray µ-CT imaging. In addition, the digital PSDs obtained by imaging of specimens made of pre-calibrated standard-size silica-based beads were also compared with their counterpart physical PSDs. The observed good agreement between the results obtained from physical and digital techniques support the suitability of using the X-ray µ-CT to understand the particulate fabric of silty soils.

Ana Maria Valverde Sancho, Dharma Wijewickreme
The Relationship Between Particle-Void Fabric and Pre-liquefaction Behaviors of Granular Soils

When driven by undrained cyclic shearing, saturated granular soils will experience the increase of excess pore water pressure and the decrease of effective stress. This phenomenon is termed as “cyclic liquefaction” or “cyclic softening”. Revealing the evolution of fabric in accompany with effective stress reduction provides a significant insight into the fundamental mechanism of cyclic liquefaction. In this study, numerical tests were conducted in DEM simulations to explore the cyclic liquefaction of granular packings with different void ratios. With the decrease of mean effective stress p′ during cyclic liquefaction process, the decrease of particle-void descriptor Ed can be observed for all packings. From the micromechanical perspective, large size voids are redistributed and the local void distribution around particle becomes relative uniform. The change of particle-void fabric from consolidated state to initial liquefaction state is irrelevant to density of the packing. A power function is further adopted to describe the negative correlation between Ed and p′.

Jiangtao Wei, Minxuan Jiang, Yingbin Zhang
Changes in Sand Mesostructure During Sand Reliquefaction Using Centrifuge Tests

Previous studies demonstrated that the sand reliquefaction resistance may decrease though the sand deposits were densified after reconsolidation, and attributed this counterintuitive phenomenon to the changes in sand mesostructure after liquefaction. However, no previous studies visualized how sand mesostructure evolved during the whole process of a liquefaction event. In this study, a centrifuge model of the saturated sand deposit liquefied successively under the same input seismic motion. Excess pore pressure at the upper and lower deposit were measured for assessing the sand liquefaction resistance. Besides, sand mesoscopic images during the whole process of each liquefaction event were recorded for analyzing the sand particle behaviors under multiple liquefaction events. The experimental results show that large voids formed under the upward seepage effects during sand reconsolidation, and the long-axes of the sand particles neighboring the large voids rotated vertically. The sand mesostructure after reconsolidation caused the quick contraction of the sand deposit under the subsequent shaking event.

Xiaoli Xie, Bin Ye
Examining the Seismic Behavior of Rock-Fill Dams Using DEM Simulations

High rock-fill dams are planned or under construction in high seismic intensity area, challenging the seismic design of those dams. This study aims to examine the seismic behavior of rock-fill dams using DEM simulations. The focus is put on the permanent deviatoric strain field and the shear stress-strain loops in rock-fills which can hardly be measured in the experiments. The acceleration response and crest settlement observed in the simulations are consistent with those monitored in centrifuge shaking table tests. The key feature of permanent deformation is surface sliding, and the size of the associated shear zone is influenced by the slope angle. Moreover, there also exist deeper shear zones with a relatively low level of shear strain compared with the surface sliding zone. The rock-fills near slope surface may experience a sharp increase in shear strain after several loading cycles, and the associated strain accounts for a large amount of the permanent shear strain. The variation rate of the shear strain can be decreased, and then the shear strain can hardly develop although the input motion is still strong, reflecting the effect of pre-shaking on the seismic behavior of dams. Moreover, the timing of sharp increase of shear strain is different for rock-fills at various locations, suggesting the complexity of the seismic behavior of rock-fill dams.

Zitao Zhang, Rui Wang, Jing Hu, Xuedong Zhang, Jianzheng Song
DEM Simulation of Undrained Cyclic Behavior of Saturated Dense Sand Without Stress Reversals

Two-dimensional discrete element method (DEM) is used to study the undrained behavior of dense granular materials under cyclic loading without stress reversals, and to clarify the effect of initial static shear on cyclic resistance and the underlying mechanism. A series of undrained stress-controlled cyclic triaxial tests were simulated with varying values of cyclic stress ratio (CSR) and initial static shear stress ratio (α), and the type of “residual deformation accumulation” cyclic response was identified. The evolution of internal microstructure of the granular materials was quantified using a contact-normal-based fabric tensor and the coordination number. The higher α (i.e., smaller consolidation stress ratios in tests) leads to higher stress-induced initial fabric anisotropy. The cyclic resistance of dense granular materials increases with initial fabric anisotropy. During the loading process, the dense granular materials with higher initial fabric anisotropy exhibited slower reduction in coordination number. The study shed lights on the underlying mechanism that why the presence of initial static shear is beneficial to the cyclic resistance for dense sand.

Xin-Hui Zhou, Yan-Guo Zhou

S6: Special Session on Underground Structures

Seismic Behaviour of Urban Underground Structures in Liquefiable Soil

Soil liquefaction has been often one of the most significant causes of damage to aboveground structures in urban areas during recent earthquakes, e.g. 2012 Emilia (northern Italy), 2011 Tohoku Oki (Japan) and particularly 2011 Canterbury- Christchurch (New Zealand), where about half of the €25 billion loss was directly caused by such a phenomenon.In some cases, sewer pipes or open-cut tunnels in liquefied deposits have been affected by floatation and large uplift. The current and future construction of relatively shallow and light underground structures in seismic regions may involve areas that are exposed to the risk of liquefaction thus increasing possible associated damages.This paper investigates the behaviour of a tunnel during soil liquefaction, from an experimental and numerical point of view, focusing on the combined effects of soil liquefaction in urban areas, where underground structures are likely to interfere with buildings. Such an aspect is rather unexplored, and the research in this field may contribute to the performance-based design of urban underground facilities.

Emilio Bilotta
Parametric Analyses of Urban Metro Tunnels Subject to Bedrock Dislocation of a Strike-Slip Fault

As an essential part of lifeline engineering, the subway tunnel is the hub of traffic and transportation. With the development of the economy, the social demand for it is increasing. Although faults should be avoided as much as possible in the process of route selection for tunnels, the tunnel, as a long linear structure, in some cities will be inevitably crossed by faults in the construction. In this paper, the seismic response of a shallow buried subway tunnel crossing an active strike-slip fault is analyzed subject to bedrock fault dislocations. Based on the 3D modeling of the tunnel using the finite element software ABAQUS, the displacement and stress of the tunnel lining structure are calculated with the change of five factors, including fault-tunnel crossing angle, bedrock overburden thickness, tunnel buried depth, lining thickness, and soil properties. The influence of the various factors on the structural seismic response of the tunnel under the bedrock dislocations is discussed. The analysis results can be used as a reference for performance-based design for tunnels crossing active faults.

Zhanpeng Gan, Junbo Xia, Jun Du, Yin Cheng
Resilience Assessment Framework for Tunnels Exposed to Earthquake Loading

The present paper proposes an integrated framework for the seismic resilience assessment of tunnels by using the appropriate fragility, restoration and functionality models, which consider both geotechnical and structural effects. Typical circular tunnel in Shanghai city of China is examined in this study, and the corresponding numerical model is built in ABAQUS. A couple of earthquakes are chosen to conduct non-linear dynamic analysis, so as to get the tunnel responses under increasing levels of ground shaking intensity. Fragility curves are constructed accounting for the main sources of uncertainties. Moreover, based on the proposed fragility curves and the existing empirical tunnel restoration functions, the development of resilience index (Re) with the peak ground acceleration (PGA) at the free-field ground surface for circular tunnel is evaluated and quantified. The results indicate that this type of tunnel is good of seismic resilience. The proposed framework is expected to facilitate city managers to support adaptation with preventive or retrofitting measures against seismic hazards, toward more resilient metro systems.

Z. K. Huang, D. M. Zhang, Y. T. Zhou
Stability Analysis of Tunnel in Yangtze Estuary Under Dynamic Load of High-Speed Railway

The section from Shanghai to Hefei along the Yangtze River is located in the territory of Shanghai, Jiangsu and Anhui provinces in East China. There is one tunnel in the whole line, which is the Yangtze River tunnel along the Yangtze River. The total length of the tunnel is 14,150 m, in which the total length of the section under the Yangtze River is 10,360 m, with the maximum soil covering of about 39 m and the maximum underwater burial depth of about 67 m. The project is located at the mouth of the Yangtze River, and there are some unfavorable geology, such as saturated soft soil (silty) and fine sand formation. From the perspective of train-track-structure-stratum coupling effects, the interaction mechanism between tunnel and stratum under long-term cyclic load is explored. Combined with the actual geological parameters and engineering parameters, the dynamic response of the train structure under cyclic load is numerically simulated. Through the results of numerical simulation, the tunnel life under the train load is evaluated from the perspective of cyclic load fatigue failure, and the necessity of the tunnel lining structure is analyzed from the perspective of structural dynamic response.

Liqun Li, Qingyu Meng, Leming Wang, Zhiyi Chen
Seismic Performance of an Integrated Underground-Aboveground Structure System

Underground structures such as subway stations, which are operated at a high capacity and frequency, are prone to damage by large earthquakes. The seismic behavior of underground-ground structure system, which refers to underground structures that directly connect to above ground buildings, is a complex issue involving the dynamic interaction between ground building and underground structures, as well as the kinematic interaction between underground structures and surrounding soil. To study the seismic performance of integrated underground-aboveground structure systems, this paper presents a three-dimensional finite element simulation analysis, addressing the influence of aboveground high-rise buildings on the seismic response of underground structures. The results show that above ground buildings significantly increase the underground structure’s racking responses and dynamic forces during earthquakes.

Wen-Ting Li, Rui Wang, Jian-Min Zhang
A Simplified Seismic Analysis Method for Underground Structures Considering the Effect of Adjacent Aboveground Structures

Adjacent aboveground structures can significantly affect the seismic response of underground structures, and the structure soil-structure dynamic interaction should be considered in the seismic design of underground structures. Three-dimensional numerical models of underground-aboveground structure systems in saturated soil are established and fluid-solid coupling elastoplastic dynamic analysis is conducted. Results show that the aboveground structure can amplify the seismic internal force and deformation response of the underground structure due to the change of the horizontal stress distribution of the near-field soil, which is an adverse seismic effect. A simplified seismic analysis method is proposed based on this mechanism. The effect of adjacent aboveground structure can be approximately considered by applying the additional loading generated by the overground structure to both sides of the underground structure in the form of pseudo-static force. The effectiveness of this simplified method is preliminarily verified by dynamic analysis.

Jianqiang Liu, Tong Zhu, Rui Wang, Jian-Min Zhang
Seismic Response Analysis on the Tunnel with Different Second-Lining Construction Time

For large deformation of soft rock tunnel, the construction time of secondary line plays a key role in the research of controlling the large deformation. In this paper, taking the Qamdo Tunnel of Sichuan-Tibet Railway, which is the greatest engineering project in construction in the 21th century, the stress states of secondary lining with different installation times of secondary lining are analyzed in the statics and dynamics. The results indicated that dynamic stress has a more remarkable decreasing tendency than static stress as the construction time for secondary lining; tunnel structure will have a lower sensitivity to dynamic excitation when the displacement ratio is up to 98% to construct secondary lining. If the secondary lining has to be constructed with a displacement ratio less than 98%, the concrete strength must be strengthened based on the real construction time of secondary lining, so that it may safely bear more partial load for controlling large deformation of the tunnel.

Weigong Ma, Lanmin Wang, Yuhua Jiang
Coupled Seismic Performance of Underground and On-Ground Structures

Seismic performance evaluation of on-ground structures can be significantly affected by their interaction with underground structures. This effect becomes more relevant in densely populated areas where transportation systems such tunnel metro lines are located nearby urban overpasses, and buildings. This paper presents the findings of a numerical study carried out to conduct the seismic evaluation of a typical bridge-building-tunnel system, located in the high plasticity Mexico City clay. Through a series of three-dimensional finite difference numerical models developed with the software FLAC3D, both detrimental and beneficial interactions effects among these structures are analyzed, establishing zones around each structure in which this interaction leads to ground motion variability, as well as the impact on both the building and bridge seismic performance. The parametric study was carried out considering both normal and subduction events for a return period of 250 years. Distances among each structure was varied to cover a wide range of scenarios. From the results gathered in here, the effect of this interaction on the modification of surface accelerations, seismic demand in on-ground structures, as well as deformations and structural stresses in the tunnel lining was established.

Juan Manuel Mayoral, Daniel De La Rosa, Mauricio Alcaraz, Enrique Barragan
Simplified Numerical Simulation of Large Tunnel Systems Under Seismic Loading, CERN Infrastructures as a Case Study

Seismic analysis of large tunnel systems using the continuum (Finite Element; FE) approach can be complex and computationally expensive. The inefficiency stems from the extended length of tunnels over very long distances, compared to the tunnel diameter, different terrain and lithological profiles, complex fixity conditions provided by the intermediate station boxes, and ground motion asynchronicity. This paper proposes an uncoupled numerical methodology to model and analyse the seismic response of large tunnel systems that is able to consider various tunnel alignments. The method is capable of simplifying the computationally intensive FE models into a lower-order, practically affordable numerical solution while still accounting for the aforementioned key features. This was achieved using a Beam-on-Non-linear Winkler Foundation (BNWF) model. The soil-structure interaction was considered using non-linear springs and frequency dependent dashpots. The springs were subjected to a free-field displacement time history obtained from 1-D wave propagation analysis. The proposed method is implemented for the case study of the circular Large Electron-Positron Collider (LEP) tunnel network at CERN in Geneva, Switzerland, the forerunner of the Large Hadron Collider (LHC). The tunnel system is 100m below the ground surface and completely embedded within a competent layered rock. The pre-LHC-upgraded tunnel complex contains four large underground cavern structures housing the particle detectors (‘station-boxes’) along its alignment. The study investigates forces developed along the circular tunnel alignment assuming a synchronous ground motion.

A. Mubarak, J. A. Knappett
Shaking Table Test of the Seismic Performance of Prefabricated Subway Station Structure

In order to study the seismic performance of prefabricated subway station structure (PSSS), the shaking table test was carried out to analyze the seismic response of PSSS. According to the shaking table test results, a three-dimensional finite element model considering the interaction of PSSS, enclosure structure and soil was established. Numerical simulations of the seismic response of PSSS under different test conditions were implemented. Through the comparative analysis of numerical results and shaking table test results, the seismic response characteristics of PSSS were revealed, and the seismic damage mechanism of PSSS was summarized. The results showed that the numerical results and shaking table test results reflected similar regularities, indicating that the established finite element model and analysis method were reliable and effective. PSSS had good performance in earthquake resistance, prefabricated joints had outstanding performance in deformation resistance, which enabled the prefabricated components to work together. Under extreme earthquakes, the vault, the upper and lower ends of the side walls, the internal non-load-bearing structure, and the envelope structure of PSSS were the most severely damaged areas. It was reasonably predicted that the seismic damage mechanism of PSSS is divided into three stages: firstly, the enclosure structure suffered seismic damage; secondly, the non-load-bearing structure inside the structure loses stability; finally, the top arch structure degenerates into a three-hinged arch structure.

Lianjin Tao, Cheng Shi, Peng Ding, Linkun Huang, Qiankun Cao
Numerical Study on Seismic Behavior of Shield Tunnel Crossing Saturated Sand Strata with Different Densities

This study investigated the seismic response of a shield tunnel crossing two saturated sand stratums with different relative densities using a finite difference method. A bounding surface plasticity model was employed to model the liquefaction of saturated sand. A deformable force-displacement link model was proposed to simulate the interaction between circumferential joints, in which the face contact between adjacent lining rings was discretized into point contact and characterized by a series of multi-degree-of-freedom springs together with the bolted connections. A three-dimensional numerical model was then established to investigate the acceleration, pore pressure, and deformation of the tunnel–soil systems. The numerical results indicate that the seismic responses of tunnel structures are more significant near the soil interface. The liquefaction-induced horizontal and vertical displacements of the tunnel linings are coupled with each other, which results in the most unfavorable position of the tunnel section. In addition, different degrees of liquefaction between the two sand stratums lead to a large dislocation of circumferential joints and then contribute to a stress concentration on the segment joints. Therefore, the circumferential joints near the soil interface should be paid more attention to in the seismic design of tunnel structures.

Hong-Wu, Zhi-Ye, Hua-Bei Liu, Yu-Ting Zhang
Seismic Response of a Tunnel-Embedded Saturated Sand Ground Subject to Stepwise Increasing PGA

In this study, the interaction between a model tunnel and the saturated sand ground was investigated based on a series of shaking table tests. The model tunnel was made of stainless-steel tube and the model ground was composed of China ISO standard sand through water sedimentation technique. In order to re-duce the boundary effect, polystyrene foam boards were used as a buffer layer to absorb vibration energy due to the reflection from the boundaries. A record from the 1995 Kobe earthquake in Japan was used as the input motion, of which the PGA (peak ground acceleration) was scaled from 0.1g to 0.6g. The input motion was applied in the direction perpendicular to the tunnel axis. The earth pressure, the pore water pressure, the accelerations of both the ground and tunnel, and the ground settlement were recorded during the tests. The model was settled between different testing groups until the recorded pore water pressure was stabilized. A new definition of pore pressure ratio was proposed which can be used to deter-mine the degree of liquefaction at different PGAs. The liquefaction phenomenon was observed during the tests when the PGA reached 0.6g. It was also showed that the ground near the tunnel is more prone to liquefaction than in other regions.

Mingze Xu, Zixin Zhang, Xin Huang
Groundwater Response to Pumping Considering Barrier Effect of Existing Underground Structure

Existing underground structure may block the water flow or extend the seepage path, which should affect the groundwater response induced by pumping. In this study, finite element models are established by ABAQUS, based on a practical pre-excavation pumping test in Tianjin, to investigate the influence of existing underground structure on seepage flow under pumping. It is found that the existing underground structure has barrier effect in pumping. In addition, the effect of the spacing between the existing structure and the excavation on the groundwater response was revealed. This study can provide guidance for the design of groundwater recharge.

Xiu-Li Xue, Long Zhu, Shuo Wang, Hong-Bo Chen, Chao-Feng Zeng
Ranking Method for Strong Ground Motion Based on Dynamic Response of Underground Structures

It is well known that different input ground motions result to significantly different underground structural response. Hence, rational selection of input Intensity Measures (IMs) is the basis for accurate seismic design of underground structures. In order to find out the most destructive ground motion for seismic design of underground structures, a ranking method for strong ground motions was presented to get their failure potential sequence which may result to maximum structural response of underground structures against ground motion excitations. In this paper, dynamic time-history analyses under a ground motion cluster were carried out on a typical subway station structure in Class IV site. In the analysis, the general finite element code ABAQUS was used to set up a refined numerical modeling, 64 real ground motion records were selected as seismic wave input, and 12 IMs of 3 groups were selected as original IMs. The moment, shear, axial force of central columns and relative displacement of central column of a subway station structure was used as the Structural Response Parameters (SRPs). Based on the Partial Least Squares (PLS) regression method, a series of composite IMs were developed and investigated. The results show that the composite IMs constructed by PLS regression method have better efficiency. This method can be used to rank the design input ground motion according to the degree of SRPs, which represents the failure potential of structure, and the results can provide a reference for the selection of input ground motions for seismic design of underground structures.

Wei Yu, Zhiyi Chen
Proceedings of the 4th International Conference on Performance Based Design in Earthquake Geotechnical Engineering (Beijing 2022)
herausgegeben von
Lanmin Wang
Jian-Min Zhang
Rui Wang
Electronic ISBN
Print ISBN