International Efforts in Lifeline Earthquake Engineering

Some recent research developments of reliability-based design of lifeline systems are summarized in this paper, including testing of buried pipeline under non-uniform excitation, seismic reliability analysis of pipeline system and reliability-based optimization design of lifeline systems. The applications of reliability-based design on the water distribution networks are introduced as well. Finally, the possible developing tendency in the future is presented.

An assessment model for post-earthquake serviceability of three utility lifeline systems was applied to seven recent earthquakes in Japan. First, using the "Quick estimation system for earthquake maps triggered by observation records (QuiQuake)," population exposures to shaking intensities were obtained for each earthquake. The assessment model was then applied, and preliminary results for the time histories of the number of people without lifeline service were obtained.

Water system services may be characterized through five categories: water delivery, quality, quantity, fire protection, and functionality. The first four categories relate the provision of water to customers with the systems' basic intended purpose. Functionality describes a water system's overall operational ability to perform in a damaged state. This study describes how to quantify the service categories in a post-earthquake environment with a focus on incorporating systemic capabilities into functionality. The model quantifies water system functionality by directly accounting for customer service restorations resulting from damage repairs and incorporating inter-system dependencies. An example water system is used to explain the model and show its applicability from earthquake initiation through complete restoration. The example shows how a system can provide water soon after an earthquake in a manner customers are used to receiving, by use of redundancy and resources, while the system itself may remain in a highly damaged and vulnerable state. This attribute that water systems may possess has never before been modeled. The proposed model intends to improve the overall understanding of water system functionality and its relation to other services. The information is applicable to other lifeline systems.

Understanding and improving the resilience of infrastructure is critical to the broader goal of community disaster resilience - minimization of loss and facilitation of recovery. Empirical research shows that infrastructure damage or disruption from hazard events and ensuing reduced service levels has a strong, negative impact on business and household performance. Conceptual and computer models relating the socio-technical facets of infrastructure to community disaster resilience are underdeveloped. Community disaster resilience does not make theoretical or practical sense without viewing infrastructure as the combination of capitals and services. Specifically, infrastructure is characterized by its relationship to other infrastructure, agents, possible disruptions, possible interventions, jurisdictions and markets. This paper presents a subset of evolving conceptual, algorithmic and geo-visual representations of infrastructure-dependent community disaster resilience. This builds upon existing work on ResilUS - a prototype simulation model of community disaster resilience - and other contributions from the literature. The ultimate goal of this research is to provide the means for making lifeline infrastructure mitigation planning and restoration decisions that account for socio-technical and recovery-based goals.

The infrastructures are interconnected and interdependent on multiple levels. The failure of one infrastructure can result in the disruption of other infrastructures, which can cause severe economic disruption and loss of life or failure of services. A methodological approach to analyze vulnerability of interdependent infrastructures has been introduced and two types of vulnerability are studied: structural vulnerability and functional vulnerability. Infrastructure topologies are only used for analysis on structural vulnerability, while operational roles of different infrastructures are further considered to analyze functional vulnerability. For these two types of vulnerability, interdependent effects are mainly studied and the effects of interdependence strength between infrastructures have also been analyzed. The analysis on structural vulnerability will be helpful to design or improve the infrastructures in the long run, while the discussion on functional vulnerability will be useful to protect them in the short term.

A 1999 study published by the American Society of Civil Engineers Technical Council on Lifeline Earthquake Engineering, entitled "Overcoming Barriers: Lifeline Seismic Improvement Programs", developed and analyzed six utility service providers and one major freeway management agency to understand their approach to seismic improvement. The study described seismic improvement policies and techniques, as well as the legal, historical, and organizational context that either facilitated or hindered seismic improvement. Since the original study, there have been significant changes in the scale and nature of these risk reduction efforts. This paper describes preliminary results of a follow-up study using the same data collection strategy to explore longitudinal trends of lifeline seismic improvement. The goal of the current study is to get an idea of the relative success of previously identified seismic improvement programs and study recommendations. More importantly, this study seeks to gain insight regarding whether these service providers have increased or decreased their seismic improvement efforts, how this has occurred and why.

On October 23^rd and November 9^th, 2011, the province of Van in Eastern Turkey was hit by two earthquakes (Mw=7.2 and M=5.6) causing damage in buildings, industrial structures and utilities damages in Van and Ercis, with nearly 700 fatalities. This paper provides a brief overview of these damages through field observations. Observed pipe breaks due to permanent ground deformations (PGD) and wave propagation (WP) effects are quantified. Comparisons are given for the estimated vs. observed repair rates for segmented and continuous pipes as per relevant codes. Results reveal that the code-based repair rate formulation for buried segmented and continuous pipes is compatible with the observations.

Post-earthquake fire refers to the fire caused by an earthquake, directly or indirectly, which is an interlocking disaster with a complex process, involving a large number of systems engineering and uncertain factors. Post-earthquake fire has caused enormous loss of life and property in many cities. Thus, it's necessary and eager to conduct research about post-earthquake fire. In this paper, the causes, characteristics, preventive and fire-fighting strategies of post-earthquake fire are analyzed and discussed. At first, it introduces the causes and features of post-earthquake fire. Then, it shows the sources of ignition in the city. On the above basis, the preventive and fire-fighting strategies of post-earthquake fire are put forward finally.

Based on the site investigation, the disaster chain and cascading effect are discussed for Xinyuan-Hejing Ms 6.6 earthquake and Wenchuan Ms 8.0 earthquake. Regarding an active seismic region, a medium earthquake (5.0≤Ms<7.0) could also lead to serious building damage and enormous economic loss due to multi-earthquake disaster cascading effect, although the earthquake magnitude is not so large. On the other hand, a strong earthquake (Ms≥7.0 ) usually triggers a cascade of secondary disasters. The disaster risk assessment (DRA) for a single incident may underestimate the severity of earthquake disaster. The damage case of a Beichuan County town in the 2008 Wenchuan earthquake shows that the disaster chain and cascading effect should be also considered.

From a system perspective, telecommunication has been performing very well in a number of major post-disaster environments. The above-normal demand for circuits right after a medium-to-large earthquake invariably created a perception of system failure. The fact is that the telephone system is not built to accommodate everybody making a call at the same time. Even if there are no physical damages to the telecommunication system, the symptom of telephone not working will exist. This paper will examine the current level of telecommunication system resilience based on past earthquake performance investigations. Reviews of preparedness and network upgrades before and after earthquakes will hopefully provide us with a path to develop a more resilient system. This includes the Internet network system, which is a key part of today's business environment. Performance of major elements of the telecommunication system will be presented. Post-disaster system emergency policy and recovery planning will be suggested for discussion purposes. The expectation is to generate interest in telecommunication service providers to develop standards or practices that will lead us to a more resilient telecommunication system. Lifeline interdependence will also be discussed. All lifelines have to co-exist as a unit in order to provide the community with security and business continuity.

When a large-scale earthquake occurs, since a huge number of electric power distribution equipment has been installed in many electric power supply areas, many electric power distribution equipment, such as poles, wires, transformers and service lines, are simultaneously damaged. During the emergency restoration work after an earthquake occurrence, it is usually difficult to quickly collect the damage information. Therefore, my research team developed a seismic damage estimation system, called RAMPEr, with an actual electric power company. RAMPEr was applied to the Miyagi area damaged by the 2011 earthquake off the Pacific coast of Tohoku (the 311 earthquake) and supported the actual emergency restoration work. This paper presents the damage and emergency restoration process for the damaged electric power distribution system due to the 311 earthquake and illustrates the operation process of RAMPEr to clarify the effectiveness of the damage estimation technologies for the emergency restoration work.

Buried high voltage power cables failed in many locations in the February 22, 2011, M 6.3 Christchurch earthquake, due to liquefaction. The empirical evidence of these failures is presented. The original design of the cables is considered, and what made the cables so vulnerable. The failure mechanism is described. Recommendations are made for the design of new buried cables in liquefaction zones.

Ensuring the safety of structure is especially important because of the complex environment in service and direction to large-scale. Two types of environment are taken into account, mountain and offshore, and height of structures is respectively 50.0 m and 88.5 m. Bearing behaviors of wind turbine towers in different kinds of load combination are discussed according to Japanese standard in 2010. Then, influence of height, quality and stiffness on displacement and stress are analyzed. Also, the dynamic responses of structures in different types and levels of earthquake are studied. It shows that the revised edition of Japanese standard in 2010 is needed to assess the bearing capacity comparing with edition in 2007. The dynamic response of structure is affected by height. When the height increases 177%, the maximum displacement in the top of the tower would increase 231% in 8-degree rare earthquake. Other factors are quality and stiffness and so on. Far-field seismic waves have the largest impact on structures which are high-rise and great flexibility.

With the rapid development of the fire power plant, an extra-large air-cooling tower (1000MW grade) will be designed and constructed in the high seismic risk areas of China such as mid-north and west-north regions. The dimension of the huge tower structure can reach up to 220 meters high and 188 meters in diameter. It's constructed with hyperboloid shell supported by X type R/C columns and the X column's length-width ratio can reach up to 1:40. It's really a challenge, but very necessary and urgent to know the seismic behavior and design weak points of the huge tower under strong earthquake attacks. This paper provides procedures and results about the modal analyses and nonlinear dynamic finite element analyses for a 220-m-high super cooling tower and its small scale (1:30) model for shaking table test under design earthquakes. The vibration modal and the seismic performance of the cooling tower, as well as the seismic damage pattern of its critical element of the support leg columns (X-pillar), were studied preliminarily. In the three-dimensional, nonlinear, dynamic time history analyses, a kind of failure criterion is adopted to simulate the possible collapse by using the LS-DYNA program. This paper provides useful reference to the seismic design practice for the extra-large tower structures.

Hydraulic simulation of earthquake-damaged water distribution system (WDS) is essential to evaluate seismic service capacity of WDS. This paper first deduces outflow models of the damages (leaks and breaks) induced by earthquake with the help of flow pattern analysis, then presents comparisons on outflow models of pipeline damages and topology models of damaged-pipeline networks. The transformations between leakage outflow and breakage outflow is obtained by conversions of minor head loss at the leak and the frictional head loss along the disconnected pipelines of the break. Model comparisons and application results shown that the outflow of the pipe leakage and breakage can be deduced from the minor head loss at the leak and the frictional head loss along the disconnected pipe, respectively.

The damage of the flexible and the rigid joints is analyzed. Based on the analysis, a new expression for leak area which owns clear physical background is given to update the old leakage model. Meanwhile, an important parameter of the new model, leak coefficient, is derived from a leakage experiment of pipeline. The seismic functional reliability of a small water distribution network is analyzed using the old and the new model, respectively. The result shows that the old model overestimates the performance of the water distribution network subjected to earthquake, while the new model agrees well with engineering practice.

As the Seismic Design Code (SDC) has revised its requirement following major earthquakes in the past decades, old water infrastructure in Japan must be seismically upgraded to keep up with the latest SDC. As a first step, such upgrade usually requires an assessment whether a specific civil structure is seismically resistant. The Detailed Evaluation Method (DEM) is well-known as a full evaluation method; but, due to its complexity and required high technical knowledge, the Simplified Evaluation Method (SEM) is normally applied to prioritize the structures for DEM testing. This simple assessment method, however, falls behind the current civil engineering technology in some respects, being thirty years old. Therefore, we propose a new SEM with a revised procedure which adopts a three-step approach where the target structure is evaluated according to "threat of liquefaction," "year of construction" and then "structural resistibility."

The proposed 110-million-gallon reinforced concrete Headworks Reservoir structures are planned by the Los Angeles Department of Water and Power (LADWP) to be part of the water supply system of the City of Los Angeles. Resilience of the water supply system is crucial for continued water supply in the event of a disaster, such as an earthquake. The seismic resilience of the reservoir structure is a function of the cracking and associated leakage that would be expected due to static loading and during the design level earthquake. Evaluation of the seismic deformation of the structure was accomplished utilizing a soil-structure interaction (SSI) model to evaluate performance of the reservoir structure in an earthquake, after the initial design based on standard code-based design procedures. SSI was used to provide information on the structural behavior of the reservoir and to understand relative movement of inlet and outlet pipelines. In addition, a leak detection system was incorporated into the design.

Table 1 shows the damage rate for each type of pipe used in the sewers in Sendai City. The damage rate of the Fiberglass Reinforced Plastic Mortar (FRPM) pipes, which is the subject of this investigation, was the lowest at 0.21% in the actual disaster assessment by the local government. Therefore, it is evident that FRPM is a pipe material that has the best earthquake resistance. We conducted a follow-up investigation of Fiberglass Reinforced Plastic Mortar (FRPM) pipes constructed by using Shield Secondary Lining Construction Method in Ishinomaki City where the Great East Japan Earthquake attacked. The result of the investigation shows that the pipes were not affected by the earthquake. Subject of the investigation was a section of Ishinomaki trunk sewer, which was extended by 1362 m (completed inside diameter: 1200; soil covering: 9.4 m) in 1995 in the lower reaches of Kitakami River in Miyagi Prefecture. In the investigation, conditions of sewers (deflection, gaps of joints, and damages to pipes) were investigated by entering the sewer pipes. At the same time, damages to above-ground parts of the sewers and manholes were investigated. The trunk sewer is constructed in the area near the offshore of Miyagi Prefecture, where earthquakes occurred frequently. Thus, follow-up investigations of the trunk sewer were conducted after Minamisanriku Earthquake (M7.0) and Northern Miyagi Earthquakes (M6.2) in 2003, and the Great East Japan Earthquake in 2011. In this report, an analysis of quantitative measurements of the trunk sewers through the investigations is discussed.

During the 2011 Great East Japan Earthquake in Japan, extreme liquefaction caused extensive damage to structures in the Kanto Plain region with the magnitudes of the settlements and tilts larger than that was observed during past earthquakes. This paper deals with an application of ground reinforcement technique by installing logs into loose sand layer as a countermeasure against soil liquefaction for water purification and a sewage treatment plant. Field surveys were conducted after the earthquake to investigate structures supported by wood piles in the water purification and sewage treatment plant. Then small scale shaking table tests were carried out to clarify the effectiveness of installing logs into liquefiable sand layer against settlement of structures.

The Canterbury earthquake sequence, 2010-2011, seriously impaired the wastewater network across much of the Canterbury region, New Zealand, by both seismic shock and liquefaction impact. This paper provides a preliminary overview of how infrastructure resilience is highlighted and built into rehabilitation practices and decision-making processes adopted by asset owners to reinstate the wastewater system in Christchurch. The physical damage suffered by the Christchurch wastewater system is briefly presented. Secondly, the decision-making processes for reconstructing the affected sewer system conducted by rebuild delivery agency - Stronger Christchurch Infrastructure Rebuild Team (SCIRT) as agreed by its clients: Christchurch City Council (CCC), Canterbury Earthquake Recovery Authority (CERA) and New Zealand Transport Agency (NZTA) are discussed with particular attention to the aims of achieving and enhancing infrastructure resilience. Furthermore, the paper describes how local and international experts have engaged with SCIRT, CCC and CERA to provide know-how in support of development of new Infrastructure Technical Standards and Guidelines that will underpin new design and construction of wastewater systems, describing the decision-making process that has resulted in the resilient rehabilitation of the Christchurch wastewater network.

In major cities, buried pipelines are more and more important in our life. Its types and layouts are various and its surroundings are very complicated. Now, it is a research hot topic in the field of world earthquake engineering and how to ensure its safety when extreme earthquakes occur in the future. So, in this paper, based on the ABAQUS software, seismic responses of the soil-pipeline systems under different cases are studied. Then, the effects of some factors on seismic response of buried pipeline are analyzed, including the vibration direction of the earthquake, embedded depth of pipe, pipe diameter, wall thickness of pipe, and tube spacing. Finally, some effect laws are proposed.

A simple model is used to study the response of deeply buried pipeline laid through a fault to random ground motion, in which the fault is simplified as a two-media site and seismic ground motions are considered as stationary stochastic processes. The cross-PSD function is calculated by wave propagation theory and random vibration theory. Then, the cross-PSD function, regarded as seismic input, is used to calculate the seismic response of deeply buried pipeline. The results show that: (1) the peak normal stress is always larger than the peak shear stress; (2) the peak normal stress for incident P-wave or SV-wave, which appears in the pipeline in the soft media, is larger than the one for incident SH-wave, which appears in the pipeline in the firm media; and (3) the peak shear stress for incident SH-wave is larger than the one for incident P-wave or SV-wave, and the peak shear stress always appears in the pipeline in the firm media.

A numerical model of buried pipeline-soil system subject to reverse-fault is developed, which is used in discontinuous deformation analysis (DDA) coupled with the finite element method (FEM). The pipe is regarded as plane structure and modeled by beam elements, while the soil is discredited into two-dimensional plane strain elements according to FEM. The contact status of discontinuous media is described by the contact analysis of DDA. The initial stress state of the pipe-soil system is also estimated for response analysis. The mechanical responses of buried pipeline and the discontinuous contact analysis between buried pipeline and surrounding soil are analyzed using pseudo-static analysis method without considering soil fracture. Some regular conclusions are drawn. The results verify the feasibility and effectiveness of using DDA coupled with FEM to study buried pipeline. The study provides a new approach and technique for studying pipe-soil discontinuous medium interaction.

Under the action of earthquake, the buried pipelines in nuclear power plant damage may become a disaster. Research for earthquake buried pipeline damage of all sorts of reasons is very important. The different subsidence is one of the important factors which makes the pipeline damaged. In the paper, the pipe is discretized into 4 nodes thin shell elements and 8 nodes solid elements in the soil entity. The nonlinear contact element is created between the pipeline and soil, and the subsidence of soil achieved by loading is assumed as the linear displacements. Some conclusions of buried pipelines subjected to subsidence are obtained according to the comparative result of some factors. Finally, an example is demonstrated for buried pipelines subjected to the subsidence and the proposed method above will be base research on response analysis for buried pipelines subjected to the subsidence.

Shaking table tests of the underground cement pipe with connection joints under non-uniform excitation are investigated in this paper. In order to simulate the non-uniform earthquake excitation, different seismic waves with spatial correlation are applied to two tables along the axis of the pipe. The experimental results of the pipe joints displacement are presented in the paper. The experimental study could provide useful data for further pipe-soil interaction mechanism research and numerical analysis.

This paper focuses on damage to water supply pipelines during the 2011 Great East Japan Earthquake. Severe damage occurred not only to houses and buildings, but also lifeline facilities (including water supply facilities) by this earthquake and tsunami. Water supply pipelines were damaged by strong ground motion and/or soil liquefaction. Outline of the damage is given first. Then the damage analysis is done by using GIS database of water supply pipelines and its damage. Lessons learned regarding the earthquake performance of water supply pipeline are summarized.

A number of water pipelines were damaged in the 2011 Great East Japan Earthquake. After the earthquake, we measured behaviors of the earthquake-resistant ductile iron pipeline (hereafter, ERDIP) located in a collapsed road embankment by removing soils around the pipes, and investigated their correlation with road subsidence and lateral spread of the embankment. This paper describes the results of this on-site investigation and analysis.

This study is focusing on behavior of ductile iron pipelines with earthquake-resistant joints buried across a fault used widely for water pipelines in Japan. It is necessary to design a pipeline carefully in case of crossing a fault, because the partially large displacement occurs on the pipelines when a fault moves by an earthquake. Although there is some research on behavior of steel pipelines which cross a fault, there are few studies of ductile iron pipelines. In this study, we investigated the behavior of pipeline buried across a fault by simulation analysis and verification experiment. As a result, we clarified that the earthquake resistant joint nearest to a fault began to move when a fault began to move, and then the joints in line began to move when the expansion, contraction and deflection angle in the joint nearest to a fault reached its capacity. In addition, we confirmed that we could evaluate the safety of pipeline against the fault movement by deflection angle in each joint. As a result, using colors with large bending performance rather than straight pipe sockets of shortening the pipe length was effective to increase the safety of pipeline near a fault.

For the seismic design of an oil and gas pipeline crossing fault, the fault displacement is usually estimated with the empirical relationship between Magnitude and Fault displacement. It should be noted that this fault displacement may under-estimate the fault displacement imposed on the buried pipeline. Based on those strong ground motion records near the causative faults, the maximum displacement of fault movement (MFD) is larger than permanent fault displacement (PFD). With the baseline correction, fault displacement time histories were obtained from those recordings near fault during the Chi-Chi earthquake and the Wenchuan earthquake, and the relationship between MFD and PFD is developed. As an example, the response of an oil pipeline crossing the Xiaojiang fault was analyzed. The axial strain of static analysis under PFD is less than the maximum strain, but a little larger than the last residual strain in the pipeline of dynamic analysis considering the process of fault movement. It should be noticed that the recordings near fault are not enough to obtain a good model for the process of fault movement. At present, the static analysis can satisfy the need of seismic design for a pipeline crossing the fault. To consider the uncertainty, the design-basis fault displacements corresponding to the different important pipelines are suggested in this article. A new empirical relationship between the maximum fault displacement and the surface-wave magnitude is obtained with the supplemented earthquake data in East Asia.

In this research, behavior of The Greater Tehran gas distribution pipelines is evaluated in case of seismic wave propagation. By synchronizing and adjustment of distribution maps of the pipelines, ground conditions and seismic inputs, various different cases for evaluation are selected and then analyzed by the finite elements method. In the numerical evaluations, different pipe specifications (including material, size, end conditions and bends), soil conditions along with seismic wave attributes (including peak ground acceleration, time history and directivity) are considered. After conducting numerical results, the basic criteria for damage assessment of the pipelines are created, which are used for network analysis. The results showed a relatively satisfying overall performance for the network; however, in several cases, some countermeasures for rehabilitation and improvement of the system reliability and safety are suggested.

The Ruby Pipeline is a 1,067-mm (NPS 42) diameter natural gas transmission pipeline extending 1,086 km from southwest Wyoming to southern Oregon in the northwestern United States. The alignment crosses or projects to within 0.1 km of 31 traces of active or potentially active normal faults. It crosses the Wasatch fault, the largest active fault in Utah, at a location approximately 3 km north of Brigham City and about 150 m east of a state highway. A maximum vertical displacement of 2.1 to 2.4 m is expected from a moment magnitude (M_w) 7.1 earthquake rupturing the 37-km-long fault (Brigham City) segment of the Wasatch fault. The resilience of the pipeline in response to fault displacement was investigated using a three-dimensional, finite-element stress analysis following recommendations in seismic guidelines developed for the Pipeline Research Council International, Inc. The pipeline was modeled over a length of 300 m, which included the state highway crossing and two three-dimensional bends. The finite-element analysis was performed by applying 4 m of vertical displacement oriented at a 60° dip angle in 120 steps with computed longitudinal strains extracted at eight equally-spaced circumferential positions for each model element at every step. The displacement capacity was defined by a tension strain of 4.0% or compression strain of 3.6%. For the 22.0-mm pipe wall thickness at the fault crossing, the finite-element analysis showed that the fault displacement capacity was greater than 4 m, which corresponds to more than 1.7 times the expected 2.4-m maximum fault displacement in a single M_w 7.1 earthquake event.

Transportation systems in the eastern part of Japan were suspended after the 2011 Tohoku earthquake. The earthquake, which occurred at 14:46 JST, resulted in the interruption of almost all railway services in Tokyo until the next day. There was a significant increase in road traffic the day after the earthquake and congestion was observed all over the city. The present study on the causes of the temporary traffic congestion was based on the analysis of probe-car data collected from approximately 2600 taxis. We also analyzed the traffic demands of Tokyo and compared the results with the probe-car data to simulate the road traffic volume in the city after the earthquake.

In order to reasonably obtain the earthquake alarm threshold value of high-speed train, the vehicle-track coupling model is established based on calculation method of wheel-rail contact relation, the track irregularities are simulated with trigonometric series method and several typical seismic records are selected which are used as the external excitations of the train. A calculation program is developed to solve the vehicle-track coupling vibration equations, the earthquake dynamic responses of vehicle-track coupling system under the excitations of track irregularities and typical seismic records are analyzed. The main indexes which are used to judge whether the running train is safe in the correlative specifications are extracted and compared with calculated results, the corresponding relation between vehicle speed, ground motion intensity and train operational safety is obtained. An analysis method for threshold value of seismic urgent handling system of high-speed railway train is explored and a preliminary suggestion of the earthquake alarm threshold values under different running speeds are presented.

A healthy and sustainable urban society depends heavily on the proper functioning of a lifeline system and its resilience in case of a disaster. While engineers seek to reduce failure probability, risk managers focus on reducing negative consequence and improving recovery time when failure occurs. Researchers at AIR Worldwide have developed an earthquake model to assess railway damage due to shaking, liquefaction and tsunami for the insurance industry. This paper describes the model and presents its application to the 2011 Tohoku earthquake. The entire Japan railway system is discretized at every 1km and the model is utilized to estimate the severity and geographic distribution of damage to the aggregate of railways' components. The model is validated by comparing the results from simulations with damage observations in the affected areas. Typical results from the model can help risk managers allocate critical resources in disaster management, repair and recovery processes.

To clarify correlation between seismic energy and damage indices of mountain tunnels, comprehensive inelastic seismic analyses are carried out on four types of mountain tunnels (i.e., tunnel with and without inverted arch, tunnel with vertical side walls, and multiple-arch tunnel). To trace the plastic behavior of concrete lining, a viscoplastic-damage model was adopted as the concrete constitutive model. Then, local and overall lining damage indices, as well as the seismic energy inputted and dissipated by tunnels, can be defined and calculated from numerical results. It is found that input seismic energy can provide a good estimation of the overall lining damage evolution time history. In addition, local damage indices are directly proportional to dissipated energy.

In order to achieve understanding of the distribution of internal force of the subway shield tunnel at the earthquake, as well as understanding the influence of the earthquake to the tunnel so as to ensure the safety of the tunnel structure, one tunnel project is studied and an appropriate numeric modeling by quasi-static method is conducted for shield tunnel in soft ground. The research shows that seismic design method is considered more appropriate for subway shield tunnels in soft ground.

The earthquake vulnerability model is a mean that used to estimate the loss of earthquake rapidly. For developing a fast seismic loss estimation method, the tunnel seismic damage was divided into minor, moderate, major and collapse according to the extent of seismic damage. In developing the tunnel vulnerability model, the peak ground acceleration (PGA) was selected to be ground motion parameter and calculated according to the modified John Zhao et al. (2006) attenuation model. The tunnel vulnerability models were established. Through the vulnerability model, it was found that tunnels have a good seismic performance in resisting the earthquake damages.

Entering the 21st century, earthquake activity is frequent world-wide, among which the number of strong earthquakes is increasing, as well. When the earthquake occurs in the mountain area, highways become the lifeline project for anti-seismic and earthquake relief works, which presents absolute importance for rehabilitation and reconstruction after earthquake. How to provide the timely rescue and how to ensure the lifeline project unlocked turn out to be vital and urgent issues. Based on the surveys of the 5.12 Wenchuan earthquake, the damage to slopes and tunnels is analyzed and characterized into different types in this paper. The author concludes that more tunnels, rather than slopes, should be selected for the high seismic region in the future. Some thinking for the design and research of tunnels is proposed, as well.

This paper presents a review of the shaking table test of mountain tunnels from five aspects, as follows: shaking table, simulation material, similarity ratio of the material, testing box, and data acquisition method. The traditional testing methods are summarized, especially their disadvantages, literally. Finally, assumptions for solving such problems are described.

Tunnel portal section is one of the most easily damaged parts in a mountain tunnel during a strong earthquake. Tunnel opening is the only part exposed outside the surrounding rock. Failures of this part may block roads, which greatly influences the transportation of the local area. Examples studied in this paper focus on failures on tunnel portals. Damage types are categorized. According to these damage types, influencing factors and mechanism of these damages are analyzed including thickness of cover layers, interfaces and discontinuities, and construction methods. Suggestions of seismic resistance and reduction methods are proposed regarding these influencing factors.

Based on the background of a complex tunnel constructed in Tibet, a large-scale shaking table test is carried out to study the seismic responses of the tunnel through fault. In this paper, decision of model similarity ratio, model box and the treatment of boundary design, the manufacture process, layout of monitored points and input scheme of seismic wave of the test are introduced in detail. From the test, it concludes that there are some similarities in seismic response between the tunnel through fault and the tunnel through homogeneous rock. With seismic wave upward propagating, seismic response of rock and soil is increasing and tunnel lining in fault was damaged severely. There is a certain influence on seismic wave propagation because of fault.

In recent years, with the increasing size of shield tunnels, the influence of foundation's shear deformation on transverse direction can no longer be neglected. Many earthquake-resistant codes in Japan have been revised by adopting two design levels representing low-to-moderate and strong earthquakes, especially after the Hyogo-ken Nanbu Earthquake. So far, the earthquake-resistant calculation to the transverse direction of shield tunnel is still employing elastic analysis that is obviously inappropriate to the high-level (strong) earthquakes. It is necessary to take the discontinuity of the shield tunnel and nonlinear characteristic of joints and segments into account in terms of high-level (strong) earthquake. Due to the above-mentioned reasons, this paper analyzes the following items on the shield tunnel applying the seismic deformation method based on finite element method, and rational earthquake resistant calculation method is put forward as well: (1) the influence of nonlinear characteristic of joints and segments; (2) the influence of longitudinal discontinuity; and (3) the application scope of the simplified method based on an elastic ground beam theory.

Since the existing model boxes cannot meet requirements in shaking table model tests of complex underground structures' seismic response under two-dimensional horizontal earthquake shaking, a suspended laminar multidirectional shear model box has been developed. Shaking table tests were carried out to verify the performance of the box, and the test result showed that shear deformation without torsion occurred in the sand specimen under the shear wave action induced by the shaking table. The suspended laminar multidirectional shear box is very satisfactory in simulating lateral boundary conditions of the horizontal prototype soil layers and can be used in the shaking table model tests to be developed.

Many subway lines in China are located in earthquake zones, and the seismic resistance problem is particularly prominent. In seismic design, subway stations are usually three-dimensional, nonlinear, large systems. In order to understand their behavior, an actual double subway station was chosen as the object of this study. A comprehensive, three-dimensional, numerical model - including subway station structure with surrounding soil - was established and nonlinear seismic response analysis was performed. The dynamic soil-structure interaction considering interface, material nonlinearity and large deformation was included in the analysis. The following results were shown, such as the deformation and stress of the surrounding soil, the deformation and internal forces of the station structure, and the velocities and accelerations at several important locations. The possible failure locations in the subway station and their mechanisms were analyzed and discussed. Several suggestions were provided as measures of earthquake-resistant construction.

According to the cross-structure between subway station and tunnel in formation condition of silt sand in Beijing, a series of shaking table tests on the dynamic interaction between soil and structure are carried out. In the case of inputting waves, monitor the acceleration, displacement of soil, the acceleration, strain of structure and dynamic earth pressure between soil and structure. Research on the seismic force and deformation characteristics of station, and analyzed the interaction law of seismic response of cross-structure of subway station and tunnel. The result shows that the top displacement of model box is larger than the bottom, and the distribution law of the acceleration is contrary. The acceleration response of shallow soil is weakened with the structure buried. The interaction force between soil and structure is comparatively large. The strains of columns and walls are larger than which of beams and plates. Stress concentration is generated on the joints. The bottom tunnel can absorb the earthquake waves coming from shaking table test, and reduce the seismic response of the upper station structure. The week effect reduces with soil thickness increasing.

On the basis of Biot dynamic theory for saturated, porous, elastic media, analytical solutions are presented for two-dimensional scattering and diffraction of transient plane SH waves by deep-buried, cylindrical, lined cavity in saturated porous elastic soil employing the Laplace transform and wave function expansion method according to the continuous conditions on saturated soil and lining structure interface and the stress-free condition on the surface of lining structure. The solutions can be degenerated under the condition of deep-buried, cylindrical cavity or rigid inclusion in saturated, porous, elastic soil or in the elastic medium. The solutions also are proofed by being compared with the existing related solutions. Numerical results are given, the effect of the lining thickness and stiffness on the stress concentration factor of the outside boundary of lining is shown.

Centrifuge model tests on rectangular structures buried in unsaturated soils are being conducted to simulate an underground restrained wall subject to earthquake shaking. In order to compare the centrifuge testing results with soil-structure-interaction (SSI) results from a numerical modeling approach that is becoming increasingly common for practical design, a pre-centrifuge testing Class A prediction has been performed. This prediction will enable: (1) comparison of the "blind" numerical modeling with the centrifuge testing results; (2) engineers to identify any potential analysis issues with emerging numerical modeling approaches; and (3) developing design improvements for underground restrained walls. Class A prediction results are presented for a buried rectangular structure with zero and 5 feet of soil above the roof, respectively. The centrifuge results, additional analytical studies, and recommendations for seismic analysis and design improvements will be published elsewhere after completion of the centrifuge testing program.

In this paper, the 3D nonlinear finite element model of a high-pier bridge is built based on OpenSees. It then analyzes the reaction rule under different ground motions and focuses on the differences of the effects of seismic pounding between uniform excitation and multi-excitation considering the effect of a traveling wave. The seismic responses of the high-pier bridge are closely related with its natural dynamic characteristics and the spectral characteristics of ground motion. The traveling wave effect amplifies the seismic pounding response, while the ends of piers are adversely affected, exacerbating the damage, and which has obvious influences to all bearings. Therefore, to consider traveling wave effect is necessary in the analysis of bridge seismic collision response.

Concrete arch bridge with steel webs is a modification of the conventional reinforced concrete arch bridge by replacing the concrete webs with the corrugated or plain steel webs. In this study, a finite element model of the reinforced concrete arch bridge and the concrete arch bridge with corrugated or plain steel webs were developed using a nonlinear finite element program, "NL_Beam3D". Their nonlinear seismic responses were studied and compared. The results demonstrate that axial force and out-of-plane bending moment are predominant under out-of-plane seismic excitations, and the materials at the arch springing have reached the plastic stage. On the other hand, axial force and in-plane bending moment are predominant under longitudinal seismic excitations, and the materials still remain in the elastic state. It is also found that seismic responses under out-of-plane excitations are greater than that under longitudinal excitations. For concrete arch bridges with steel webs, it is unnecessary to consider the effects of synchronous out-of-plane and longitudinal seismic excitations. RC-PSW arch bridge has better seismic performance than RC-PSW arch bridge.

The current probability-based AASHTO LRFD Bridge Design Specifications are fully calibrated against gravity dead load and truck load. However, it does not cover extreme load effects in a consistent fashion. In this paper, a methodology is described to combine extreme and non-extreme loads. To illustrate this method, common truck load and earthquake load combinations are used. This illustration quantitatively shows that the current design practice is non-conservative if only considering peak truck load or peak earthquake load alone; and it is too conservative if the design is based on the combined peak truck and earthquake.

Based on the ideas of passive energy dissipation, multi-failure and damage reduction, a multi-level control mode of seismic unseating failure for concrete girder bridge is presented. The control mode, respectively, provides: (1) displacement restriction by energy dissipation-based pier-girder connection mode; and (2)unseating prevention by girder-girder connection mode. The two-level control modes can be automatically transformed according to the preset threshold value of "structural fuse". Firstly, the analysis model of the multi-level control mode is established and the control mechanism is investigated. Furthermore, the nonlinear seismic responses of a concrete girder bridge equipped with multi-level control mode are analyzed based on a simplified mechanical model to examine the control effectiveness. Finally, parametric influence laws of unseating failure control effectiveness are researched. It is found that the multi-level control mode can achieve unseating failure prevention and structure safety protection effectively for girder bridges during earthquakes.

Fluid viscous dampers (FVDs) are well suited to vibration control of highway bridges subjected to seismic excitations. However, the parametric selection of dampers for a real project is lacking a simple and quick estimation method. The research of seismic behavior of a 3-pylon cable-stayed bridge with fluid viscous dampers has been conducted to investigate two parameters (damping coefficient C and damping exponent α) of the FVDs through nonlinear, dynamic, time-history analysis. The parameter sensitivity study indicates that setting dampers in the bridge longitudinal direction can reduce the response of the bridge. Design proposals are suggested following the results of the parametric study, as well.

Complex flexural, shear and torsion failure of reinforced concrete (RC) bridge columns for skewed and curved bridges will be taken place during seismic excitation. An experimental investigation on the seismic performance of RC bridge columns subjected to combined cyclic bending and torsional loading was carried out. Five test models of circular RC bridge columns were tested under various loading conditions: cyclic uniaxial bending, cyclic torsion, and combined cyclic torsion and cyclic uniaxial bending. Several combinations of cyclic bending and cyclic torsion were applied to evaluate the interaction between torsion and flexural capacity of columns. The results reveal that: (1) the flexural capacity in flexure decreases when the torsion increases; (2) the torsional capacity in torsion decreases when the bending moment increases under combined bending and torsion; and (3) the damage tends to occur upward outside the flexural plastic hinge region when the level of applied cyclic torsion increases.

Reinforced concrete (RC) curved bridges suffered severe damage, or even collapsed, in the 2008 Wenchuan Earthquake. A field survey of the seismic performance of RC curved bridges was examined. Probable causes of the damage patterns to RC curved bridges - including Baihua Bridge and the ramp bridge of the Huilan interchange - are presented and analyzed in this paper. The main failure was pure shear failure or shear-flexural failure of the pier columns. Damage to bearings and expansion joints was another common failure pattern. Lessons learned from the damage to RC curved bridges in this earthquake and the recommendations on the seismic design of RC curved bridges are presented involving: ductility of bridge columns; design of curved bridges; and, design of bearings and devices preventing girders from falling down. Suggestions for the future seismic design and retrofitting of RC curved bridges in moderate-to-severe earthquake areas are also presented.

Based on the structural dynamics theory, the fluid-solid coupling dynamic equation of ice-water-bridge pier is established. Taking a bridge pier surrounded by ice as the research object, the dynamic response of the pier in different types of earthquake was studied. The results show that, in different earthquake conditions, the pier top displacement, acceleration and hydrodynamic pressure without ice are smaller than that with ice, and increases with the increasing of ice thickness. Bending moment and shear of pier bottom are also increased with the increasing thickness of the ice. They are both biggest under the action of T2-II-1 seismic wave, but the bending moment and shear rate of change are biggest under the action of El-Centro seismic wave, respectively 494.51% and 374.76%.

A novel three-dimensional (3D) isolation bearing for bridge structures was proposed. The new isolation device is composed of several devices, mainly including: combined disk spring, rhombic steel plate damper, and conventional horizontal isolation device (e.g., lead rubber bearing). The configuration component and mechanical design method are introduced. Experimental testing of every main part of the novel bearing was performed to verify its mechanical performance. Horizontal and vertical pseudo-static mechanical behavior of the proposed 3D isolation bearing (3DIB) were tested on a full-scale 3DIB, respectively. Experimental testing indicated that the proposed 3DIB has reasonable configuration and an explicit mechanical transmission mechanism. The employed component in this paper can satisfy the need for 3D isolation technology from the viewpoint of configuration and mechanical aspect. Pseudo-static tests and case study revealed that appropriate energy dissipation performance can be achieved by reasonable design, keeping favorable stability at plentiful displacement.

This paper presents a long-term condition assessment approach of cables under in-service loads collected by SHM system. For the cables of the bridge, the stochastic axial force response can be collected by the SHM system and described by a filtered Poisson process, through which the maximum value distribution of axial forces in its design reference period can be derived using Poisson Process theory. The long-term deterioration process of steel wires in the cables considers simultaneously the uniform and pitting corrosion due to environmental attack and the fatigue propagation induced by cyclic stress. By employing first order reliability method, the reliability of the cables under the monitored responses is further estimated in terms of the safety under the extreme traffic load distribution in the design reference period and the serviceability specified in the design specification. The discussions of the life-cycle condition assessment of the cable-stayed bridge provide guidance to the future decision-making related to maintenance and replacement.

To investigate the seismic performance of fiber-reinforced polymer (FRP) jacketing constrained reinforcement concrete (RC) hollow section bridge piers under multi-dimensional earthquake action, cyclic tests of five RC rectangular hollow section benchmark and carbon fiber reinforced polymer (CFRP) jacketing confined piers specimens were carried out under unidirectional and bidirectional lateral horizontal repeated loading. The failure characteristics, lateral force-deformation curves, effective loading stiffness, ductility and energy dissipation of specimens were analyzed. The experimental and analytical results show that: the damage is more severe; the failure process is accelerated; and bearing capacity, ductility, rigidity and energy dissipation are decreased under bidirectional earthquake action. The CFRP constraints may: change the failure location and mode of bridge piers; improve the deformation ductility, plastic deformation and energy dissipation, but have less contribution to horizontal force.

Bridge system is an important part of the lifeline systems. In order to study the effects of fluid-structure interaction on the dynamic response behavior of deep water bridge structure, an investigation on hydrodynamic pressure is developed in two ways, including theoretical analysis and experimental study. For the purpose of establishing the seismic response-based criterion under earthquake of submerged bridge structure, a set of equations are derived by utilizing Morison theory. Then, a model test of submerged bridge pier system, including pile groups-cap-pier and inertia mass, has been performed to investigate the effect of hydrodynamic pressure on seismic response. Based on the results, the effect of hydrodynamic pressure was analyzed experimentally theoretically. Study results indicate that the surrounding water can change the characteristics of structure, so the effect of hydrodynamic pressure should be paid much attention.

In order to provide a tool for making retrofit decisions on damaged bridges, a multi-criteria, fuzzy, comprehensive evaluation system is established considering economic-social facts and the engineering circumstances. Within this system, overall, a bridge's damage scale is divided into five damage states, which is assembled with three damage levels based on fuzzy logic. By means of observation, comparison and summary of post-earthquake damage assessment and retrofitting database made up of damage data from 469 concrete bridges due to Wenchuan earthquake, a novel relationship can be derived between damage states and seismic loss parameters such as the feasibility, safety of retrofitting plan and retrofit fee. Furthermore, this paper introduces a procedure to Bayesian update scheme for fragility curve utilizing bridge empirical damage data as the prior distribution obtained from the Wenchuan earthquake. By composition of the seismic hazard and updated fragility curve of bridge before and after retrofit, the risk of seismic damage can be derived and the retrofitting decision can be made. Moreover, the method is applied to making decision of Huilan bridge retrofitting engineering and it indicates that, regarding seismic damage risk as standards, the reasonable retrofitting decision for damaged bridges can be made effectively and accurately with multi-criteria fuzzy comprehensive evaluation system.

Seismic collisions can result in damage or ruin of the adjacent buildings or bridges based on many investigations of earthquake damage. In this paper, various contact collision models and methods employed by researchers at home and abroad were introduced, especially on the three-dimension (3D) contact-friction model of pounding considering tangential friction proposed by Zhu (2001), etc.; moreover, not only to realize this numerical model based on OpenSees which is open source software, but also to testify this model by numerical simulation.

This paper presents an experimental study of the seismic behavior of reinforced concrete beam to concrete-filled steel tube column connections with ring-beam. Three specimens are tested under low cyclic loading conditions. The displacements, stresses, failure patterns, and energy dissipation capacities of the tested specimens are studied. In addition, the influence on the sectional dimension of the ring beam is investigated. To validate the experimental results, numerical simulation is also conducted to compare the stress distributions of the joints with and without the ring beam. It is shown that damages of the joints with the ring beam develop as the longitudinal bar inside the reinforced concrete frame beam begins to yield. On the other hand, the concrete-filled steel tube column is found to behave essentially in the linear range and no significant damage is observed in the reinforced concrete ring beam. It is also noticed that plastic hinges typically nucleate in the frame beam and the hysteretic curves generally take on a spindle-shaped morphology.

Experiments of axial compression were conducted using four full-scale reinforced concrete columns. One group of three columns used high-strength steel with the yield strength of 1000 MPa, and the second group used the ordinary strength steel with yield strength of 400 MPa for transverse steels. The performances between these two groups, specifically in bearing capacity and deformation ability, are discussed. Compared to the specimen with ordinary-strength steels, the bearing capacity of specimens with high-strength transverse steels increases 29% at most. The deformation ability increases about 2 times when the bearing capacity declines 50%. At peak confined concrete stress, the effectiveness of high-strength transverse steel depends on configuration and volumetric ratio of transverse steel. The better the concrete is confined, the higher the effectiveness of transverse steel. Five stress-strain models of confined concrete are used to calculate the bearing capacity of the specimens. Compared with experimental results, the calculation results show that the results calculated by the Razvi and Saatcioglu (1999) model agree with the experimental results well, as well as the stress in transverse steel at peak concrete stress.

Experimental studies on four shear wall specimens under low-cyclic reversed loading were carried out, in which high-strength rebar with the yield strength of 1000 MPa was used in longitudinal and transverse reinforcement in boundary embedded columns of reinforced concrete shear walls, and tested variables involved yield strength of longitudinal rebar, yield strength of hoops and volumetric ratio of hoops in embedded columns. The behavior mechanism of high-strength rebar in boundary embedded column of shear wall is studied. As well, the bending bearing capacity, ductility, energy dissipation capacity and the failure characteristics of high-rise shear wall with high-strength rebar were researched. The experimental results showed that, compared to RC shear wall with normal-strength steel, the strength of specimens with high-strength steel used as longitudinal reinforcement in boundary concealed columns was increased, and the yield and ultimate displacement were improved greatly.

High-damping rubber-bearing isolator (HDR) has become as an effective device to enhance the seismic resistance for ridge, the lead rubber isolators (LRB) as well. However, the dynamic characteristics of the HDR isolators have not always been clarified so far because of their scatter nature of mechanical properties and the lack of reliable test data. In this study, a new type of HDR isolator is tested dynamically under different conditions. Using the test results, the relationships between the equivalent stiffness, damping ratio and number of cycles, as well as the magnitude of shear strain, were examined.

The super-large diameter steel pipe piles have been used more and more in China in recent years. The driving possibility of pile has already become a major engineering problem for the designer and contractor. Due to the lack of realistic analysis method and construction experience, some piling engineering accidents often appeared, such as hammer refusal, pile damage, etc.. A case study of wave equation drivability analysis for the super-large diameter steel pipe pile in a large-scale wind power plant project was performed in this paper. The performance of pile-soil-hammer system and the stress condition of pile movement during driving were described. The driving possibility, a suitable hammer and a rational installation criterion were obtained. The wave equation drivability analysis is to utilize the stress wave and its dynamic effect in the pile shaft generated by hammer impact vibration. This successful application has shown that the wave equation drivability analysis is a very practical tool for the super-large diameter steel pipe pile.

Zigong Seismograph Array (ZGSA), located in Sichuan province of China, was installed in 2007 and included 8 irregularly-spaced seismograph stations. The array successfully recorded strong motions during the 2008 Wenchuan earthquake (Ms 8.0). Because the separation distances between every two stations are less than 400 meters, these data offer some interesting insights into spatial variations of seismic ground motions that suits for engineering scale. In this study, authors firstly calculated the coherency functions of accelerations at different directions recorded by ZGSA through application of digital signal processing technology. Then, the variation mechanisms of the coherency functions at various directions, along with the separation distance and frequency changes, were analyzed. The results showed that when the circular frequency was less than 3Π, the coherency function was more than 0.8. However, the degressive trend of the coherency function became significant with the separation distance elongation when frequency gradually increased. Consequently, based on the recordings, a piecewise fitting model was put forward to simulate the spatial variation of seismic ground motion under the conditions of large-earthquake and far-field. The spatial coherency model proposed in this paper has applicable value in simulation study of ground motion field.

Potential sources are simplified as point sources or linear sources in current probabilistic seismic hazard analysis (PSHA). Focus size of large earthquakes is considerable, and fault rupture attitudes may have great influence upon the seismic hazard of a site which is near the source. Using the potential rupture surface model, we analyze the seismic hazard near the Longmenshan central fault that generated the Wenchuan earthquake with magnitude 8.0 in 2008. The following conclusions are reached: (1) The seismic zoning results show hanging-wall effect, although it is not very clear because of the high dip of the causal fault, in agreement with the isoseismal lines of intensity IX or higher; (2) As far as the seismic hazard analysis result is concerned, it is more reasonable to use the potential rupture surface model in the PSHA than to use the current potential source model because it can avoid hazard dilution near the causal faults.

Based on the explicit FE method and 32 CPU parallel computing cluster platform of ABAQUS, and considering the inhomogeneity of ground soils and concave-convex fluctuation of topography and bedrock surface, large-scale refined two-dimensional (2D) FE nonlinear analytical models for Fuzhou Basin were established in the time domain. Seismic responses of the basin site considering topography effect and nonlinear characteristic of soil are studied. Compared with one-dimensional (1D) analysis models by equivalent linearization wave propagation method that takes no account of the topography effect, the 2D models indicated that there is a focusing effect caused by the fluctuant bedrock surface and topography. The acceleration response spectra present a double- or multi-peak phenomena. The response amplitudes and predominant period are relatively larger than the 1D model; besides, the PGA displays a non-monotonic decreasing characteristic with depth and a greater focusing effect of some particular soil. The 2D models have a better ability to simulate the magnification effect of low frequency components of earthquake wave.

The frequency content of earthquake ground motion is one of the important factors which affect the dynamic response of structural systems. Because of the uncertainty of characteristic period T_g and amplification coefficient β, the results of seismic response under the ground motions will be affected by its uncertainty. So, the research on this topic becomes so necessary. Based on the ground motions during the 2008 Wenchuan earthquake, the uncertainty of T_g and β is analyzed and the maximum value, the minimum value, the mean value, the standard deviation and the coefficient variation are presented finally.

Studies have shown that long-period ground motion can be amplified due to basin structure, the amplification effects generate such as the amplitude increased, and the duration lengthened. The basin effects of ground motion are extremely detrimental to engineering structures of long natural vibration period, such as high-rise buildings and bridges. The paper carries out the numerical simulation of long-period ground motion on basin effects with far-field earthquake input, which uses previous developed by the applicant structure-preserving algorithm, simplistic discrete singular convolution differentiator (SDSCD). The preliminary test shows that the algorithm is feasible to simulate long-period ground motion on basin effects. It will lay the foundation on proposing evaluation method to meet the needs of engineering for long-period ground motion on basin effects, to provide more reliable scientific basis for the seismic risk management of major projects.

In this paper, three methods are presented for computing response spectra for torsional ground motions. It is also shown that the response spectra obtained from synthetic torsional earthquake wave conform better to the shape characteristics than results of the other two methods. The response spectra suggested by Wang and Jiang (1993) takes the dispersion effect of the apparent wave velocity into account which makes it more precise than the method proposed by Rutenburg et al. (1985). To be conservative, the envelope torsional response spectrum and artificial torsional earthquake wave are proposed to apply in structure analysis. However, there are still some differences among results of the three methods which need further discussion.

The circle PGA attenuation relationship model is very popular in China because it is useful and convenient in seismic risk analysis. It is obvious, though, that the model is not suitable for large earthquakes, especially some with a long fault. We propose one circle model in which the "epicentral distance" is a mapping of the epicentral distance in the usual circle model and is named as "mapping circle" model. This model is used to analyze the PGA attenuation relationship in Wenchuan Earthquake (Ms 8.0). The results show that the model could be used to simulate the ground motions distribution along the Wenchuan earthquake fault and it is better than the circle attenuation model and the fault-rupture attenuation model.

Currently, there is not yet a good solution to rational and effective seismic response spectrum and ground motion of engineering site with inclined bedrock in engineering practice. Using numerical simulation method of decoupled near-field wave propagation, seismic response of engineering site with inclined bedrock, when δ pulse wave and artificial ground motion are inputted into this two-dimensional model, are analyzed and discussed. The results show that: (1) By the weighted average method of strong motion response spectrum of the left and the right in one-dimensional model, seismic response of engineering site with inclined bedrock can be approximately obtained; but, the coefficient factor α is gotten from "trial and error" and the adjustment for many times through the results of the two-dimensional model; (2) For SV wave seismic action, the seismic response results of the two-dimensional model have significant and obvious wave-type conversion phenomenon which cannot be solved in a one-dimensional model with SH wave as input; and (3) Soft soil characteristics of two-dimensional model sometimes reflect isolation effect, weakening effect and soft soil damping energy. In short, to solve this issue, the two-dimensional model (including tilt and inclined bedrock medium), rather than using the left and the right one-dimensional model, are recommended and suggested.

The SYNACC method for synthesizing earthquake ground motion time histories is reviewed, with emphasis on the recent extension to generating motions on an array of points in space. This method is a combination of empirical and physical model-based approach, and involves unfolding in time a site-specific Fourier amplitude spectrum of ground acceleration, obtained for a scenario earthquake by an empirical scaling model, or for an ensemble of earthquakes by probabilistic seismic hazard analysis. The unfolding consists of representing the ground motion as a superposition of traveling surface Love and Rayleigh waves and of body P and S waves, which propagate with phase and group velocities consistent with the dispersion characteristic of the site geology, approximated by parallel layers. The synthesized motions are site-specific and consistent statistically with observations within the recording range of typical accelerographs (0.02-25 Hz). The output consists of synthetic accelerations, velocities and displacements, and also of point strains, rotations and curvatures, at a point or at an array of points. Such time histories are useful for the design of spatially extended structures, like pipelines, bridges and tunnels.

Liquefied and post-liquefied sand can be treated as a kind of fluid. Previous research has suggested that liquefied or post-liquefied sand is a shear-thinning, non-Newtonian fluid. The flow characteristics of pre-liquefied sand under high pore pressure ratio are worth discussing. According to the theory of dropping ball viscosimeter, the test apparatus of flow characteristics of pre-liquefied sand is developed. The saturated sand was put in a model box. A water tank was linked with the bottom of the sand box. The height of the water tank can be lifted to provide the excess pore pressure required in the sand soil. A steel sphere, embedded in the sand box, can move in the horizontal direction under a dragging force. Resistance and ball velocity were measured during dragging and, meanwhile, apparent viscosity is evaluated. The velocity of the sphere and excess pore pressure ratio were the main focuses in the tests. The results suggest that the apparent viscosity of the pre-liquefied sand decreases with the growth of the strain rate. The pre-liquefied sand is shear thinning, non-Newtonian fluid which is similar with liquefied and post-liquefied sand. The apparent viscosity of the pre-liquefied sand also decreases with the growth of the excess pore pressure ratio.

Evaluation of liquefaction potential of soil is important in geotechnical earthquake engineering. Significant phenomena of gravelly soil liquefaction in the 2008 Wenchuan earthquake were reported. Thus, further studies on the liquefaction potential of gravelly soil are needed. This paper investigates the potential of an artificial neural networks-based approach to assess the liquefaction potential of gravelly soils form actual shear wave velocity field data. The success rates for occurrence and non-occurrence of liquefaction cases in Wenchuan earthquake could be up to 100% and 94%, respectively. The study suggests that neural networks can successfully model the complex relationship between seismic parameters, soil parameters, and the liquefaction potential of gravelly soils.

Shaking table tests were performed to evaluate the dynamic interactive behavior of soil and pile group foundations in a liquefiable ground. The test results give detailed information on the pile acceleration time histories, distributions of bending strain within the pile group, as well as the ground response and pore water pressure generation in the soil around the piles. Slight liquefaction occurs under moderate shaking, while complete liquefaction occurs under strong shaking. The soil accelerations at the ground surface are both amplified at moderate and strong shaking. Accelerations of the pile near tip and superstructure are slightly amplified under both moderate and strong shaking, with larger amplification factors occurring in the latter case. The accelerations of pile cap are de-amplified under moderate shaking and significantly amplified under strong shaking. For the pile, the maximum bending moment occurs at the connections between pile and cap. Regarding the pier bottom, the maximum bending moments are almost equal to those of the pile-to-cap connections. The above results may contribute to the deep understanding of soil-pile-structure interaction in liquefiable ground.

The pile bending moments are particularly important for engineers to design pile foundations in liquefiable soils. In order to investigate the effect of width-to-thickness ratio of pile cap, pile diameter, initial stiffness ratio of soil and pile and mass of superstructure on the pile peak bending moments subjected to dynamic loading, the three-dimensional (3D) finite difference method considering the coupling of soil-fluid is employed to simulate shaking table experiment for soil-pile group interaction in liquefiable soils under sinusoidal wave. For comparison, a representative single-pile reference simulation is also studied. In the numerical modeling, a modified form of the Mohr-Coulomb plasticity model called as Finn Model (Itasca, 2002) that can perform coupled dynamic-groundwater flow calculations is used to simulate the effects of soil liquefaction. The numerical modeling is well verified through the shaking table experiment. A number of results and discussions are presented for describing the simulated behavior of pile group in liquefiable soils.

Regarded as fluid, the flow characteristics of the liquefied saturated sand is similar to a kind of shear thinning non-Newtonian fluid. According to the theoretical analysis of liquefaction of saturated sand based on non-Newtonian fluid mechanics, the relationship between shear strain rate and dynamic apparent viscosity, in zero effective stress state, can be properly described by Gross mode. Through fitting the flow curve of the dynamic torsional shear test results, the flow constitutive model of post-liquefied sand is established. Considering the influences of relative density, confining stress, stress path and history and the direction of loading cycles on the model parameter selection, the functional relationships of the model parameters and their physical significances are illustrated, which establishes theoretical foundation for further research on large deformation of liquefied flows based on fluid mechanics theory.

The soil is usually under complex initial stress before being subjected to earthquake and wave loading. The conventional triaxial test cannot simulate the dynamic behavior of soils under complex consolidation stress states. The true cyclic triaxial tests of coarse materials are studied by using partial flow code under three-directional anisotropic consolidation. The effects of initial intermediate principal stress coefficient on dynamic deformation characteristics of coarse materials are investigated systematically based on PFC3D. The results show the initial intermediate principal stress coefficient has little influence on shear modulus and damping ratio. The normalized shear modulus and damping ratio vs. normalized shear strain are located in a narrow space, respectively. The maximum absolute axial deformations are not always in the loading direction, which is determined by the initial stress state.

Small strain shear modulus G_max is a key parameter in soil dynamics. Although the cyclic loading on saturated soils induces a decrease of effective stress and a rearrangement of soil skeleton which may both lead to a degradation in undrained stiffness, only the contribution of effective stress reduction to the G_max degradation is considered in Hardin-Richart equation, and that of soil fabric change is neglected. In this paper, untrained cyclic trysail tests in combination with bender element tests were conducted on saturated clays. The influences of cyclic loading history on the small strain shear modulus were studied. The test results show that, in the beginning of tests, G_max of saturated clays under cyclic loading history is moderately lower than the corresponding value under non-cyclic loading history at the same effective stress. However, when the effective stress reduces to some extent, G_max decreases suddenly and reflects the failure of soil skeleton.

The stability of the slope was significantly reduced based on seismic action, which will lead to slope failure. According to relevant design specifications to determine the seismic load ,then use Strength reduction method respectively solve the slope stability of natural conditions and earthquake condition in which the horizontal earthquake inertia force and vertical earthquake inertial force have different combinations. Through the analysis of the results found when the direction of horizontal earthquake inertia force outside toward slope, no matter whether the vertical earthquake inertial force, the stability of the slope is lower than the stability of nature condition obviously. When the vertical earthquake inertia force is downward, the slope is the most unstable. Accordingly, we suggest this kind of condition to evaluate the stability of the slope and start to design the slope of seismic resistance, and verified that seismic design of engineering examples is reasonable.

The instability of slopes triggered by earthquake has received much attention recently. In engineering practice, a slope consists of unsaturated and saturated soils. The seismic analysis should consider the dynamic behavior not only of saturated soils, but also of unsaturated soils. This paper extended the use of the strength reduction FEM to the seismic stability analysis of unsaturated soil slopes. A new procedure based on static and dynamic finite for analyzing unsaturated soil slope seismic stability without considering the change of matric suction under short-term seismic loading was proposed. The seismic stability analysis was validated through numerical simulations selecting the generalized cohesion of unsaturated soil and the internal friction angle of saturated soils as strength parameters. Results show that the convergence of time-history curves of relative displacement on key points and plastic zone distribution should be considered simultaneously as the dynamic instability criterion of the slope.

Tsunami scour has received increasing attention over the past decade, given its role as a critical failure mode for building damage and, in particular, the wide scale damage to infrastructure resulting from the 2004 Sumatra Andaman Tsunami and the 2011 Tohoku Tsunami. This paper uses observed scour from these events to distinguish four primary scour mechanisms: local scour (often associated with pore pressure softening), overtopping scour, channelized scour (in which flow is concentrated in a channel), and general scour or wide area erosion. The first two of these are commonly observed to undercut buildings and other structures. Simple models are presented for these two scour mechanisms correlated to field measurements and observations, providing upper limits on the observed scour depth values. Correlation of scour behavior by soil type, common in theoretical models, is not supported by the present data set, so a general condition is recommended for current practice.