Lifelines 2022: 1971 San Fernando Earthquake and Lifeline Infrastructure

San Fernando Earthquake in 1971 was one of the significant devastating hazards despite a moderate magnitude. Therefore, a substantial number of studies were performed from different perspectives to understand the impact of the earthquake and the behavior of the structures that had structural and non-structural damages. In this study, 383 research documents from 749 authors were downloaded from the web of science and analyzed between 1971 and 2019. This way, the variety of the contents of this research, linkage between them, the contribution of different countries and institutions were established. This study conducts a bibliometric analysis and a review of existing literature on the San Fernando Earthquake, focusing on the last five decades to achieve a better understanding of the research work on this subject. Bibliometric analysis was chosen as a methodology, which enables understanding of the relationship between subsequent studies that were performed within the last half-century. It presents the annual contribution, author and country production and collaboration, citation details, keyword, and content-based analysis in both digital and visual format. It is believed that the study will provide useful information to the researchers, and connect the individual studies to others, and identify the gaps. This study has shown that a disastrous event like 1971 San Fernando Earthquake still leads scientists of today to learn from the mistakes and shed light on the future after 50 years.

As a result of the M_w6.6 1971 San Fernando earthquake the Lower San Fernando Dam (LSFD) suffered significant damage that could have resulted in a breach and uncontrolled release of the reservoir (URR). The LSFD embankment was damaged as a result of liquefaction that led to a massive slide of the upstream slope into the reservoir. The slide lowered the crest of the dam well below the reservoir level. However, a section of the downstream slope of the embankment remained, with about 5 ft of freeboard. Fortunately, the reservoir was lowered soon after the earthquake, avoiding URR into the heavily populated San Fernando Valley below. Experience suggests that system failures are a result of a combination of events or circumstances (a sequence of events) that collectively conspire to produce the outcome that was observed. This paper describes a probabilistic model of the events at the LSFD as a result of the 1971 earthquake to assess just how rare (in terms of its annual frequency) of occurrence it was. This retrospective evaluation of the events associated with the LSFD as a result of the 1971 earthquake provides insights and lessons for conducting prospective seismic risk studies in the future.

The 1971 San Fernando and 1994 Northridge earthquakes significantly shook the Upper San Fernando Dam (USFD) and substantially deformed its hydraulic fill embankment. Extensively analyzed, this dam has become a landmark case study for validating numerical models, despite its poorly documented construction methods. Following the 1971 earthquake, most analysts assumed a semi-hydraulic fill method of construction as well as a firm foundation. They also thought that the outlet line, damaged in 1971, was within the hydraulic fill materials and deformed with them. These assumptions have influenced the interpretation of the USFD performance and every numerical simulation of its seismic behavior. Based on a thorough review of archives, we confirm that the dam was constructed using the full hydraulic fill method and that the outlet line was within alluvium soils and suffered damage in 1971 when those deformed. We propose cross-sections to refine numerical models and describe what aspects are needed to validate the dam’s seismic performance accurately.

The Joseph Jensen Filtration Plant of the Metropolitan Water District is located in the area of strongest shaking generated by the 1971 San Francisco Earthquake. One or more concrete basins which had been precisely leveled settled by amounts of up to 6 in., of which about 4 in. were estimated to be due to compaction, rendering the plant inoperative. The primary lesson that was learned is that the amount of settlement that was observed could not have been predicted based on the calculated history of cyclic shear strains using a single component of the earthquake motion. However, it was found that the settlements caused by each component acting separately could be added to approximate the settlements caused by multi-directional loading. It is now practical to conduct bi-directional nonlinear effective stress site response analyses running the two horizontal components simultaneously and adding the settlements and excess pore pressures that are calculated for each component at each time step.

Prior to the February 9, 1971, California’s Department of Transportation (Caltrans) bridges were designed with very limited seismic design criteria. Consequently, several bridges and interchanges were severely damaged during the San Fernando earthquake. Within a few days after the earthquake, Caltrans managers had written change orders that all bridges under construction were required to have much more transverse reinforcement in the columns, end the practice of using undeveloped starter bars between foundations and columns, and increase seat lengths to at least 18 in.; all to prevent the types of damage that was observed after the earthquake. However, at that time the interstate highway system was almost completely built. Consequently, most of California’s bridge inventory had been designed with little resistance to large earthquakes. Caltrans has been playing catch-up ever since, retrofitting and re-retrofitting vulnerable bridges as new lessons were learned after every earthquake.

During past earthquakes in the world, industrial parks in waterfront areas have been repeatedly damaged. The main causes of the damages to the industrial facilities were the soil liquefaction of man-made islands reclaimed from waters, its induced large ground displacements, and the earthquake ground motions amplified by soft reclaimed soil, and tsunamis. Heavy damages to energy industries by future earthquakes and tsunamis will seriously affect living and safety of the people, as well as the world-wide economic activities. The enhancement of earthquake resistance of energy industries is one of the most urgent subjects around highly urbanized areas such as the Tokyo and Osaka Bays in Japan. A national project has been carried out by the Japanese government for the reinforcement of the energy industries all around Japan since 2014.

In recent years, malfunctions of submarine telecommunication cables caused by abrupt seabed movements have been observed worldwide. However, the mechanism of cable failure under geohazards is not completely clear, and methods for evaluating cable safety are still limited. Physical model tests and finite element analyses were conducted in this study to investigate the change in cable axial strain under abrupt lateral seabed movements. The sensitivities of soil properties, offset characteristics, and cable anchored lengths were investigated. The results show that under the same offset distance, the increase in the crossing angle and cable anchored length can lead to a decrease in the axial cable strain. A regression relationship between the maximum cable strain and the related parameters was obtained based on the numerical results, which can clarify the cable-seabed interaction and provide a basis for the safety evaluation of submarine telecommunication cables subjected to abrupt seabed movements. Support is gratefully acknowledged.

Seismic analysis performed in support of design of the offshore segment of a high voltage (230 kV) transmission line that connects two substations in San Francisco, California, is the focus of this study. The offshore portion of the transmission line is located approximately 2,000 feet from the western shoreline of the San Francisco Bay, has a total length of approximately 2.5 miles, and consists of three parallel double armored cross-linked polyethylene (XLPE) cables. The transitions from the offshore portion to onshore portions occur in high density polyethylene (HDPE) conduits, approximately 1,500 ft each, installed from the land using horizontal directional drilling (HDD) and are included in the seismic analysis of the offshore segment. Seismic design criteria for the line are based on 84th percentile ground motions from an Mw7.8 earthquake on the San Andreas Fault, which is located approximately 9 miles to the west. Geologic conditions along the offshore portion of the line were evaluated by a review of bathymetric and geotechnical data and consist of very soft to soft fine-grained sediment of Young Bay Mud underlain by older alluvium deposits over bedrock. Nonlinear geotechnical analysis was performed to calculate the expected permanent ground deformations along the cable alignment. Seismic assessment of the offshore marine cable was performed using non-linear soil–structure interaction analysis to compute maximum axial force under the earthquake-induced permanent ground deformations. The analyses included consideration of material and geometric nonlinearity of the cable, nonlinear response of the surrounding soil, as well as the effects of the operating cable temperature and the friction between the cable and the conduit within the HDPE pipes. Cable properties were obtained from mechanical and electrical cable tests on a prototype cable performed by others. Several parameter sensitivity studies on cable–soil interaction, compression stiffness, and anchorage capacities were performed. Multiple alignments with S-shape slack were evaluated for optimal combined seismic and thermal loading. Overall seismic reliability of the cable was evaluated through analysis that considered uncertainty in earthquake ground motions, geotechnical response, and the corresponding response of the cable. Monte Carlo simulations were performed to estimate the scenario and annual probabilities of failure. The target audience for this study includes engineers as well as public utility company stakeholders, who can benefit from the methods and lessons learned as presented here.

The disruption of power systems on community resilience is particularly important because many other lifelines and community functions depend on commercial power. Power systems have been design to be fault tolerant by incorporating a high degree of redundancy in substations and the network. Unlike most power system elements, a damaged transformer cannot be bypassed to restore service and the restoration of transformer bushings, their most vulnerable component, require lengthy restoration times. Limited shake-table test data from a transformer tank indicates that bushings are sensitive to resonance on resonance damage, which the current standard does not assess. This paper describes the need for modifying the standard to address this deficiency. There is a need to have the standard require a dynamic analysis of the transformer-bushing system to assess this vulnerability and assure that mitigation measures are implemented. Testing of fully dressed transformers is needed to support these changes.

Although the gas pipeline infrastructure in the United States is vulnerable to the seismic hazards of (i) strong ground shaking, and (ii) ground failures induced by surface faulting, liquefaction, or landslides, limited national guidance exists for operators to consistently evaluate the earthquake response of their pipelines. To provide additional information for stakeholders and establish more consistency at a national scale, we attempt to quantify seismic risk for gas transmission pipelines in the conterminous United States using a metric such as average annual loss, which helps readily distinguish geographic areas of high and low relative risk. Specifically, we integrate the 2018 National Pipeline Mapping System, the 2018 National Seismic Hazard Model, and several candidate models from the literature for estimating pipeline damage. Through this effort, we highlight major research needs for ultimately reducing the many uncertainties associated with a comprehensive seismic risk assessment of gas pipelines.

There is a lack of study in the literature for flood evaluation of gas pipelines. This paper shows the evaluation of flooding potential on the high-pressure gas pipelines located within flood zone per FEMA maps. The gas pipelines passing the creek are divided into two categories: the elevated pipelines and the buried pipelines. The proper vertical distance between the thalweg and the elevated pipeline is determined per the flood depth. The optimized cover of buried pipeline also is determined per the scour depth of flood. To calculate the scour depth, the equations in the literature are used in this study. The results of the study show the inconsistency of these equations. It is suggested that more data needs to be collected to fit a better equation for calculating the scour depth. The graphs are also provided to determine the minimum cover for the buried gas pipelines for the various sample median of soil and flood flow along with the maximum flood forces for the elevated gas pipelines. These graphs can be utilized to analyze the impact of the flood on gas pipelines.

This paper proposes a design method to calculate the inelastic strain distribution along unpressurized or pressurized pipelines induced by temporary ground deformations. The inelastic strain distribution is predicted by using the elastic solution of unpressurized pipelines and a strain conversion method. This paper also proposes a calculation method to predict the critical compressive strain (CCS) in compression of pipes with and without pressure by using an analytical solution. The proposed methods are applicable to pipelines and pipes with a roundhouse type or yield plateau type stress-strain curve.

This paper presents an integrative analysis framework combining natural hazards with network mobility to provide insights on disaster preparedness and relief. In particular, this framework characterizes the impact of seismically induced landslides on network mobility to reveal the mobility changes immediately after the events and throughout the course of restoration and recovery efforts. Landslides not only undermine the structural integrity of roadways, but also deposit a significant amount of material on the road surface, usually resulting in partial or complete road closure to traffic. The highly populated Portland, Oregon, Metro is selected as a case study to demonstrate this framework given that the Pacific Northwest is highly prone to large earthquakes as part of the Cascadia Subduction Zone as well as highly susceptible to landslides given its high topographic relief and wet climate. In this case study, travel time to the west and east sides of Willamette River, which divides the Portland Metro area, shows an abrupt change in mobility. In particular, the Portland Hills region with its steep topography is identified as the most vulnerable region. Based on a temporal analysis of recovery, the majority of the network mobility is expected to be restored after 30 days. The results of this study serve as a preliminary assessment of the impact of landslides on network mobility and can facilitate decision making in emergency planning.

A static push-over analysis method was employed to assess nonlinear behavior of embankments. Then, based on the results of the analysis, a simple method for carrying out dynamic nonlinear seismic response analysis of an embankment using an equivalent-single-degree-of-freedom (SDOF) model was proposed. It was confirmed that the sliding displacement of embankment calculated using the proposed method showed good agreement with that obtained using the finite element method. It consequently followed that the proposed method made it possible to carry out the dynamic responses of embankments in an economical manner compared to the conventional two-dimensional finite element method.

We evaluated the post-earthquake functional damage and recovery of highway networks in Japan using dataset obtained from major earthquake disasters during the period of 1995–2018. Firstly, the spatial extents of road closure were compared for four earthquakes in conjunction with corresponding seismic intensity distributions. On this basis, serviceability fragility functions, describing the relationship between shaking intensity and the probability of occurrence of road closure, were developed for each earthquake. It was found that the functions reflect traffic control criteria and related rules in highway companies. Secondly, spatiotemporal recovery processes of daily traffic volumes were compared for three earthquake disasters. Highway network performance was then compared in terms of three kinds of measures: (1) total number of sections open to traffic, (2) total distances open to traffic, and (3) total traffic volumes weighted by travelled distance. The third measure tends to be lower than the other two. Serviceability measure reflecting traffic volumes should be used in evaluating damage and restoration of highway networks.

Road networks aim to provide mobility and accessibility to population, goods, and services. To promote and ensure this major goal road networks must cover most diverse geographies and territories being frequently affected by natural hazards. The effects of natural events directly affect the role of providing access in remote areas as well as ensuring an efficient mobility in major highways. Available risk management systems generally study the physical effects of natural hazards over road infrastructure centering efforts in the vulnerability of network components and the consequent economic losses due to certain levels of damage. Limited literature is available regarding the operational consequences over network mobility and accessibility and the socio-economic consequences to users and non-users. This article proposes an integrated management system that accounts for network risk in terms of travel time delays, the optimization of mitigation strategies to reduce risk, and the effects of considering other dimensions in decision-making such as social vulnerability and accessibility to critical infrastructure. The system considers the effects and consequences of seismic, flooding, and volcanic hazards. The models accounted by the management system consider hazard modeling and characterization, fragility of road infrastructure (pavement structures, tunnels, bridges, and cut slopes), social vulnerability of exposed population, network risk estimated in terms of travel time delays, and the optimization of mitigation strategies. The optimization process in addition considers a greedy randomized adaptive search procedure algorithm, funding allocation with budgetary restrictions, and network topological analysis. The proposed management system is applied to a road network exposed to seismic hazard in Central Chile. The system is intended to assist road authorities in the assessment of network risk and the development of effective mitigation strategies to increase road network resilience to natural hazards.

The negative stiffness control is one of the structural control methods where the control force is negatively proportional to deformation. It has been clarified by past works that negative stiffness control makes it possible to decrease the existing stiffness of the structure to be installed, thereby suppressing the absolute acceleration and absolute displacement responses that are related to the damage of structure. When it comes to build an actual damper device, however, it can only be achieved by using sophisticated but complicated active controlled actuators or semi-active devices. This paper proposed a new vibration control device realizing negative stiffness in a passive manner was proposed. The prototype of the proposed device was assembled, and its performance was investigated through sinusoidal loading tests. It was found from the periodical loading tests that the proposed device generated the designated negative stiffness as well as friction energy dissipations in a good accuracy. It consequently follows that the proposed mechanism could successfully realize a negative stiffness control in a passive manner.

The state of bridge infrastructure in the United States has received a mediocre grade which indicates the vulnerable condition of these lifelines in response to extreme events as well as their poor serviceability during normal conditions. This assessment manifests challenges in the areas of condition, capacity, funding, and future needs, as well as innovation. Aging and structurally deficient bridges require prompt attention to enhance their performance and extend their service life. Presented case studies discuss opportunities to achieve these objectives for existing and new bridges through the application of sustainable practices for both superstructure and substructure components. The key focus in these opportunities is the application of rotary-kiln manufactured lightweight aggregates. These applications cover a wide range of techniques to (1) enhance the service life of bride decks, (2) improve the capacity of the superstructure, and (3) reduce demands on the abutment walls, all contributing to the resilience of bridge infrastructure and their performance during extreme events. Further discussions include ratings and assessment of various practices and their interdependencies with the infrastructure resilience measures, as well as environmental and social resilience, and sustainability objectives.

Seismic isolation has been extensively used around the world for seismic protection of structures. Several high seismic regions around the world experience seasonal freezing that can drastically alter isolation bearing properties and structural response under earthquake loads. Under subfreezing temperature, mechanical properties of bearings are affected as well as the properties of concrete and steel reinforcement. The objective of this study is to evaluate the seismic performance of a base isolated bridge equipped with Friction Pendulum System and Lead Rubber Bearing located in extreme cold climate. Considering the change in mechanical properties of isolation bearings such as, shear stiffness, yield strength, and friction coefficient, due to temperature change, the seismic performance of isolated bridges is evaluated. Detailed nonlinear three-dimensional finite element models of the isolated bridges considering the material properties at cold temperature are developed. Finally, fragility curves are developed for seismic vulnerability assessment of isolated bridges at subfreezing temperature.

In general, only one fixed pier is set in a continuous bridge. It is difficult to satisfy seismic demand of the fixed pier in a highly seismic area. To solve this problem, a pier-bottom seismic isolation structure with self-centering (PSIS-S) was placed at the bottom of the fixed pier, whose contact surface was elliptical. Under a strong earthquake, there was a relative swing between the bottom of the fixed pier and bearing platform, preventing the seismic energy from transferring to the fixed pier and superstructure. After the peak value of the ground motion, the continuous bridge can basically return to the initial equilibrium position under the action of the self-weight and seismic residual energy. A shake table test and simulations of a three-span continuous bridge were conducted. Then, taking a five-span continuous bridge as an example, the influence of main parameters of the PSIS-S structure on its seismic isolation effect, was studied. The results showed that the PSIS-S structure can effectively reduce the seismic response of the fixed pier, and that the residual deformation of the bridge was small, meaning that the PSIS-S structure had a self-centering capability. The ratio of the long radius to the short radius of the ellipse had a great influence on the seismic isolation effect of the PSIS-S structure, but the effect of friction coefficient could be ignored.

Using the 2018 National Seismic Hazard Model and the 2018 National Bridge Inventory, an annualized earthquake loss (AEL) study was conducted for approximately 610,000 bridges in the conterminous United States, quantifying both direct and indirect economic losses. The typical AEL framework has been augmented with new replacement unit cost data and bridge-specific parameters for modifying default fragility curves. Earthquake hazard is defined using spectral acceleration hazard curves that account for location-specific soil conditions. Hazard is integrated with bridge-specific fragility curves to compute annual probabilities of exceeding various damage states. Further, economic loss for each bridge was estimated using the repair costs associated with specific damage states and indirect costs incurred from downtimes. Quantitative assessments of seismic risk, especially those that account for downtime-related impacts, enable us to illustrate the distribution of risk with respect to geographic region, era of construction, or type of bridge.

In the last 40 years, earthquakes in California have caused freeway bridge collapses in urban areas resulting in considerable economic disruption and fatalities. Following the 1989 Loma Prieta earthquake, where 43 people died as the result of a single bridge failure, the State of California embarked on a significant bridge retrofitting program to help curb future impacts. Under the Multihazard Mitigation Council (MMC) project, the cost efficacy of these efforts was assessed using REDARS (Risk from Earthquake Damage to Road Systems), a software tool that estimates damage to bridges and analyzes disruption experienced road networks after an earthquake event. The analysis also considered secondary economic impacts, which in many cases are far greater than the direct impacts, resulting in a benefit cost ratio of 3 to 1. This paper presents the results of this analysis, and suggests methods of preserving this early 2000s era computational framework using cloud-based technologies.

Adequate seismic performance of marine container terminals is critical to supply chain resilience. Seismic response of cranes and wharves and their performance are key factors in the seismic performance of a container terminal. This paper will provide a basic understanding of the following: seismic performance of container cranes and contributing factors; concepts for designing new cranes and retrofitting existing cranes to improve seismic performance; seismic performance of wharves; how container cranes and wharves interact and respond in earthquakes and factors.

During the Cephalonia, Greece, 2014 earthquake sequence (M_w=6.1 and M_w=6.0), the quay walls in the port of Lixouri, Cephalonia, displaced laterally up to 1.5 m, while liquefaction of the gravelly port earthfills resulted in the manifestation of ground cracking and coarse-grained ejecta. To evaluate the seismic performance of the Lixouri port quay walls, numerical analyses using the finite difference method were performed at the location where the largest horizontal displacements were observed and the results are compared to the observed response. Three commonly used constitutive models (PM4SAND, UBCSAND, and URS/ROTH), informed by data from site investigation efforts (Dynamic Penetration Test, DPT, and Multichannel Analysis of Surface Waves, MASW), were considered for the simulation of the behavior of the liquefiable earthfills. The results of the numerical analyses indicate that liquefaction of the earthfills significantly contributed to the total lateral deformations of the quay walls; without liquefaction, the computed horizontal displacements are only approximately 30% of the observed lateral spreading when liquefaction occurs. Overall, the numerical analyses using best-estimate input parameters for the PM4SAND and UBCSAND models significantly under-predict the observed response, while the computed horizontal displacements are approximately 25% smaller than observed when the URS/ROTH model is used. The difference between model predictions can be partially attributed to the incorporation of residual shear strength once liquefaction is triggered when using the URS/ROTH model. Finally, the results of the numerical analyses show the strong influence of the pulse-like characteristics as well as the polarization of the input motion, indicating that forward directivity also significantly contributed to the observed response.

In order to study the dynamic characteristics of the prefabricated subway station structure (PSSS), a finite element model was developed to simulate interaction among the soil, the envelope structure, and the underground PSSS during earthquakes. Numerical simulation and analysis of the nonlinear seismic responses of the PSSS were implemented. The dynamic relative horizontal deformation of structure and the stress responses on the components were analyzed in detail. The seismic damage characteristics of the PSSS were discovered preliminarily. The results indicated that within the seismic fortification levels, the PSSS had excellent seismic performance and met the seismic performance requirements of the current standard. The additional influence of vertical ground motion on the seismic response of PSSS could not be ignored. Due to the significant increase in damage under the giant earthquakes, the side walls, the crown arch, and the envelope structure were the weak positions for earthquake resistance, which should be given sufficient attention. The research conclusions could provide reasonable reference and guidance for the seismic design of the PSSS.

The Los Angeles Metro transit system has been undergoing a rapid expansion since the 1980s. The expansion includes complex structures comprising elevated structures with major bridges, circular tunnels, oval caverns, rectangular tunnels, and underground stations. In addition, the Metro transit system overlaps with the lifelines of other jurisdictions, including the California Department of Transportation, Los Angeles Bureau of Engineering, and many others. The Los Angeles area is also subject to a significant seismic hazard. The seismic hazard creates risks of damage due to fault rupture, ground shaking, liquefaction, landslides, and seismic settlement. Accordingly, Metro provides the Supplemental Seismic Design Criteria as part of its overall Design Criteria to provide a comprehensive basis for design of new structures for seismic hazards. Metro’s design criteria have progressed over time based on capabilities in the engineering community, new research, and changes in parallel standards such as Caltrans standards and ASCE 7. Metro’s criteria have served as models for other lifeline system design criteria, and are anticipated to continue to change over time based on additional capabilities and knowledge in the engineering community. This work reviews the history of changes of the Metro Supplemental Seismic Design Criteria. Metro’s criteria anticipate changes to computational methodologies used for design of Metro structures, including three-dimensional seismic response analysis and structure-soil-underground structure interaction with Metro facilities. The Metro criteria could be applicable to other codes and standards of practice utilized by the profession for transportation and other lifelines structures in seismic regions.

The Los Angeles Metro transit system has been undergoing a rapid expansion since the 1980s. The expansion includes extensive segments with tunnels and underground stations. The Los Angeles area is subject to a significant seismic hazard, the greatest impact being ground shaking, which affects a large area compared to more localized earthquake effects such as fault rupture. Therefore, all underground stations in the Metro system are designed according to high standards of analysis and design to accommodate ground shaking. With the restraint provided by the ground, underground subway stations are inherently more resilient when subject to ground shaking than above-ground building structures. Nevertheless, Metro provides Supplemental Seismic Design Criteria as part of its Design Criteria to ensure reliable and robust design of new underground structures for resilience to ground shaking. The Supplemental Criteria requires designing for the fundamental racking (side sway) impact of ground shaking due to vertically propagating shear waves. However, some subway stations are adjacent to large building structures with deep basements that modify the racking deformations that otherwise would occur in the free field without major adjacent structures. The change in racking deformations results from inertial effects of the adjacent structure mass during shaking. A further complication is the reflection of shaking energy between the two structures (station and adjacent structure), with an associated period of vibration altered by the relative stiffness and mass of the adjacent affected structure. Therefore, computational seismic analyses of subway station response must include the impacts of nearby structures where such structures are present in the overall structural model as simplified single- or multi-degree-of-freedom (SDOF and MDOF, respectively) systems.

The Regional Connector Transit Corridor (RCTC) Project is a $1.75B design-build project currently being constructed by the Los Angeles County Metropolitan Transportation Authority (Metro). The 3.1 km (1.9 mile) long light rail transit project includes major underground elements: twin bore tunnels, cut-and-cover box tunnels, three cut-and-cover underground stations, and a rail crossover cavern. Final design of the underground structures has been completed and construction has substantially advanced with revenue service planned for late 2022. Located in downtown Los Angeles, California, one of the most active seismic regions, the project underwent comprehensive seismic design for its underground structures in compliance with the Metro Rail Design Criteria. This paper overviews Metro’s seismic design criteria and presents the project-specific design approach that has been taken for the major underground elements of this project.

This study makes a report on functional damage of water supply systems after the 2018 Hokkaido Eastern Iburi earthquake and that after Typhoon Faxai, the 15th typhoon in 2019. The functional damage of water supply systems during these two events were mainly due to the power outages. In Japan, it is anticipated that extensive power outages may occur after the Tokyo Inland and the Nankai Trough earthquakes in the near future. This study will be helpful to establish a resilient water supply system against future earthquakes.

This paper presents a deterministic optimization model that can be utilized for budget allocation to implement retrofitting measures for steel and concrete water pipelines. This model can be used by decision-makers to develop budget allocation strategies and to determine which strengthening and retrofit methods are the most practical and economical to be used considering a predefined set of limitations and criteria. The deterministic model determines the optimal retrofit measures based on minimizing the pipe retrofit costs, post-earthquake replacement costs, and especially large seismic losses. The main constraints considered for this model are mitigation budget constraint, mitigation strategy constraint, and risk of especially large loss constraint. A case study is performed for a pilot area per the return of certain earthquake to demonstrate the application and practical aspects of the model. Pipeline exposure database, built environment occupancy type, pipeline vulnerability functions, and regional seismic hazard characteristics are used to calculate a probabilistic seismic risk for the pilot area. The results provide the variation of retrofit costs for pipelines in the pilot area based on their geographical location and the selected strengthening measures.

The designer of the Los Angeles Aqueduct was aware of the location and importance of the San Andreas fault, and that it was crossed in the Elizabeth Tunnel. Fairmont Reservoir, which provided storage at the upstream end of the tunnel, was taken out of service after the 1971 San Fernando earthquake because its design was the same as a dam at the southern terminus of the aqueduct that failed. A new dam was constructed to provide head with piping that included a bulkhead in the tunnel to prevent backflow into the reservoir. The 2008 Great Southern California ShakeOut scenario revealed that all three aqueducts bringing water to Southern California could be damaged in a single earthquake. The Los Angeles Department of Water and Power considered constructible “enhancements” that might allow water to flow through the tunnel after fault displacement. A short section of high density polyethylene (HDPE) pipe placed across the fault zone inside the tunnel was considered because of its excellent performance in the 2010–2011 Christchurch earthquakes. Some unstable locations were indicated in 1907–1911 as-built tunnel construction geology notes, but most of the concrete-lined tunnel had no notes and it has not been tested by nearby earthquakes. Considerations began in 2012 for anchoring an HDPE pipe in the semicircle invert. The existing bulkhead was to be replaced with a removable-door bulkhead to allow construction access from the out-of-service reservoir. Design alternatives were studied; steel ribs were added for structural support across the 146-m-wide zone of most recent faulting; shotcrete overlays were added in very poor rock quality zones identified with a seismic refraction survey inside the tunnel. HDPE pipe was eliminated from the design concept during final design, which was completed in 2020 by Los Angeles Department of Water and Power engineers. Construction was planned to begin in late 2021 and be completed by 2026.

The Metropolitan Water District of Southern California (Metropolitan) owns and operates an extensive conveyance, storage, treatment, and distribution system, which provides water to approximately 19 million people over a 5,200 square mile service area. Metropolitan’s Colorado River Aqueduct (CRA) conveys on average approximately 25% of its service area annual demand. As part of its seismic resilience strategy to evaluate the overall seismic risk of its water delivery system, Metropolitan has conducted studies to assess the seismic vulnerability from a Moment Magnitude (Mw) 7.8 earthquake on the Southern San Andreas Fault, estimate the potential damage to the CRA, and assess the potential impacts to water delivery. This paper describes the methodology utilized in the risk assessment, development of performance criteria, structural evaluation, pre-earthquake mitigation, and post-earthquake restoration strategies for the existing Whitewater Tunnel No. 2 based on anticipated damage from a Mw7.8 SSAF earthquake that ruptures through the tunnel.

The Casa Loma Siphon Barrel No. 1, as part of the Metropolitan Water District of Southern California’s (Metropolitan’s) Colorado River Aqueduct (CRA), is a 148-in. (3,759-mm) diameter concrete pipe that extends 5 miles and crosses the Casa Loma Fault Segment of the San Jacinto Fault Zone. This paper discusses the design, construction, repair, and seismic resilient design of this siphon over a time span of almost a century. Metropolitan’s strategy to improve the seismic resilience includes: (1) geologic and geotechnical investigations for the siphon at the crossing of the Casa Loma Fault; (2) development of seismic design criteria that meet Metropolitan’s performance objectives; (3) mitigation strategy and application of earthquake-resistant pipe joints; (4) modeling of pipeline and fault interaction; (5) development of detailed design; and (6) construction contract delivery and sequencing considerations. The retrofitted pipeline is anticipated to remain functional and operational during and following a design seismic event.

The Metropolitan Water District of Southern California (Metropolitan) initiated the Replacement of Casa Loma Siphon Barrel No. 1 (Project) to improve the seismic resilience of the Colorado River Aqueduct (CRA) facilities. The Casa Loma Siphon Barrel No. 1 is part of the CRA that conveys water from the Colorado River to Metropolitan’s water supply system and is a critical lifeline facility. The existing siphon crosses the Casa Loma Fault near San Jacinto, California (within the San Jacinto fault zone) and is susceptible to damage due to fault rupture and ground subsidence. The siphon was originally constructed in 1935 and is a 148-in. diameter concrete pipe. The most critical objectives of the Project are to retrofit the existing siphon for the anticipated fault displacement and long-term settlement due to groundwater withdrawl and to allow Metropolitan to quickly return the pipeline to service should damage occur during an earthquake. The Project will replace the existing siphon with dual 104-in. diameter, earthquake-resistant ductile iron pipelines (ERDIP). The project team evaluated several alternatives during design including different pipeline alignments, alternative backfill materials, and various pipeline joint layouts and configurations. Because of the large anticipated fault deformation, the preliminary analyses led to a design that incorporates ERDIP with Expanded Polystyrene (EPS) Geofoam backfill material to distribute the fault displacement across many pipeline joints. The Project team developed a novel approach to modeling the ERDIP joints by utilizing the finite element analysis program ABAQUS to simulate the pipeline joint deformations along axial and rotational degrees of freedom combined with traditional soil-pipe interaction modeling to examine the pipeline performance. This project has several unique challenges, including: (1) design for large fault displacements estimated at 12.8 ft horizontal and 2.6 ft vertical (coseismic surface displacement) and 3.3 ft of settlement over a period of 50 years; (2) novel application of EPS Geofoam as pipe backfill at the fault crossing to distribute displacement and reduce stress in the pipeline joints; (3) sophisticated finite element analysis using a three-dimensional model of the fault and pipeline interactions; and (4) full-scale testing of the 104-in. pipe joint performance to verify the maximum rotation and moment capacities.

To prepare for future seismic events, effective resilience frameworks informed by lessons from past disasters will help to address this complex problem. This work discusses a seismic resilience design approach for a new water infrastructure system supported by an effective framework and lessons learned from the 2011 Tohoku earthquake and tsunami crisis. This work discusses four main topic areas including describing the $1.3 billion Willamette Water Supply System (WWSS) and the seismic hazard it faces. It discusses key insight drawn from the 2011 Tohoku earthquake and tsunami crisis in Japan. From this an overall resilience framework that addresses the complexities and interconnected elements associated with natural disasters is proposed. Finally, the resilience framework, together with the lessons learned from the Tohoku earthquake and tsunamic crisis and other related thinking are applied to the WWSS to illustrate how the framework can be practically and effectively implemented to improve the resilience of water infrastructure systems. This work provides a case study of how these measures can be implemented in the development of a new seismically resilient water infrastructure system, as well as highlights fundamental policy decisions related to establishing level of service priorities.

Seismic fragility models predict the probability of different damage states for a given earthquake intensity measure. This paper outlines a seismic fragility framework for earth dams, which uses the relative crest settlement as the engineering demand parameter and involves two statistical models: (1) an engineering demand model that predicts the relative settlement as a function of ground motion intensity, and (2) a seismic capacity model that predicts the probability of a damage state given the relative settlement. These two models are used together within a Monte Carlo simulation to compute the resulting fragility model. This paper focuses on developing seismic capacity relationships that are based on the original damage description for each dam without consideration of the observed relative settlement. The development of a seismic fragility model is demonstrated through an example that uses the proposed seismic capacity relationships and a simplified engineering demand model.

Seismic performance analysis of water distribution system (WDS) is an essential step in the seismic design and resilience evaluation of WDS. Network connectivity reliability analysis and the flow-based hydraulic simulation methods are widely used in the seismic performance analysis of WDS. Many researchers choose one of these methods for seismic performance evaluation of WDS, but the comparison between the application results of these two methods is insufficient. This work compares the analysis results of the two methods through the application of benchmark WDS cases in which the connectivity reliability and the water demand performance of user nodes are used as quantitative measures; the two methods were conducted based on the Monte Carlo Simulation to consider the uncertainty of seismic damages of pipelines. Three widely-used benchmark WDSs (Modena, C-Town, Exnet) with different characteristics were used for case study. Through the application results of the benchmark WDSs under different seismic damage scenarios, the correlation, consistency, and statistical characteristics between the results of the two methods were compared. The results show that the correlation and consistency between the connectivity reliability and the water demand performance measure is low; the seismic connectivity reliability usually has a larger value than that of the water demand performance; user nodes with lower connectivity reliability may not necessarily have lower levels of water demand performance. The hydraulic simulation method is considered to be a more reliable analysis method because it includes both the geometry topology of pipelines and the hydraulic characteristics of WDS.

In order to efficiently operate and maintain seismically resilient water lifeline systems, it is crucial to characterize the damage potential of the pipeline system in future earthquakes. Mitigation strategies to address known vulnerabilities can then be identified and prioritized in risk reduction programs to meet system performance criteria. Assessing water pipeline damage potential involves considering both the probabilistic geographic distribution of earthquake-induced shaking and ground deformations, and the locations of the pipe network. This study of hazards on a regional scale, rather than a site-specific scale, is a necessity of lifeline engineering. A recent major study was undertaken to evaluate system-level consequences for the entire City of Los Angeles water pipeline network, measured by the estimated total number of pipeline repairs and subsequent repair costs and times due to earthquakes. In order to compute such consequences from earthquakes, new methods were required for quantifying ground shaking, surface fault rupture, liquefaction, and landslide hazards at a regional scale. This study involved mapping zones of surface fault rupture hazard and assigning probability of fault rupture displacement, mapping zones of liquefaction hazard and assigning probability of liquefaction and median estimated liquefaction-induced settlement and lateral spread displacement as a function of peak ground acceleration (PGA) and moment magnitude (M), and mapping zones of landslide hazard with associated probability of earthquake-induced landslide and median estimated landslide displacement as a function of PGA and M. The methods and results of this study enable a detailed analysis of the City of Los Angeles water pipeline network’s resilience to seismic hazard and deaggregation of the risk by hazard type (ground shaking, surface fault rupture, liquefaction, and earthquake-induced landslide), by source faults, and by pipeline service zones.

Seismic vulnerability studies are essential for resiliency planning and prioritization of seismic retrofit projects for spatially distributed critical infrastructure. However, detailed seismic evaluations of medium-sized water and/or wastewater systems are not always economically viable. We performed regional-scale vulnerability studies for two medium-sized water and wastewater systems in areas of moderate and high seismic hazard. The seismic hazards considered in these studies included strong ground shaking, fault rupture, liquefaction, and seismically induced landslides. The studies built upon publicly available information to assess seismic-related hazards and identify the most vulnerable areas with consideration for the full system. Developing a backbone system for each study allowed for prioritization of recommended mitigation projects. Areas of uncertainty and high vulnerability were identified to inform more detailed and targeted studies. The approach used in this study can be applied by engineers and owners to vulnerability studies for other distributed systems to inform resiliency planning and prioritization of mitigation projects.

The City of Portland’s water system is the largest in the state of Oregon covering an area of approximately 225 square miles and serving almost one-quarter of the population of the state. The water system services 165 pressure zones and has over 2,000 miles of pipelines, two major dams, 38 pump stations, 59 distribution system tanks, and 10 terminal storage reservoirs. Some of the oldest components of the system are over 100 years old. In 2009 dollars, the replacement value of the system was estimated to be $6.7 billion. A comprehensive seismic study of the Portland’s water system was completed to assess its performance in an M9 earthquake on the Cascadia Subduction Zone. A long-term system improvement plan was developed to meet the stated recovery goals in the Oregon Resilience Plan, a plan developed under the direction of Oregon House of Representatives to protect lives and maintain economic activity following an M9 earthquake.

Seismic resiliency is a major concern for the stewards of our wastewater infrastructure throughout California. This includes the Orange County Sanitation District (OC San), who provides wastewater collection, treatment, and disposal services for approximately 2.6 million people in Southern California. Because OC San’s facilities are exposed to local seismic hazards, including potential for surface fault rupture, liquefaction, and liquefaction-induced differential settlement and lateral spread, the risk of earthquake damage and associated service disruptions is a significant threat to their $11 billion infrastructure. With this realization, OC San has taken a proactive approach and conducted a study to evaluate structures at their two treatment plants for potential seismic hazards, including identification of mitigation strategies to improve operational resiliency following a major earthquake. Based on the findings from this study, a plan was developed to identify and prioritize mitigation projects using a risk-based methodology that serves as the foundation for integrating seismic resilience into OC San’s overall capital improvements program.

This paper addresses the seismic vulnerability, risk, and resilience of the water systems of four municipalities of Mexico City affected by the September 19, 2017, earthquake and their coincidences with what happened in 1985. Most relevant aspects and results regarding these events are analyzed and discussed. Historic information of the seismic damage produced to the water systems in Mexico is presented. The physical and operational characteristics of the damaged systems and the damage produced by this event and its relation with local soil conditions and earthquake characteristics is presented and discussed. Details of the damage to the investigated aqueducts and networks are described. Based on this information, the seismic vulnerability, risk, resilience and loss, and recovery of functionality of these systems after this earthquake are evaluated suggesting for future events the simulation of damage scenarios. Finally, general conclusions and recommendations for the strengthening of these systems aimed to reduce their seismic vulnerability and to improve their resilience and the functionality recovery time.

The Hetch-Hetchy Reservoir located 160 miles from the San Francisco Bay Area is a major source of water to 2.7 million people. Just east of the greater Bay Area, Hetch-Hetchy pipelines, bringing close to 300 million gallons per day of water, cross the narrow Sunol Valley in an east-west direction. A north-south trending seismically active Northern Calaveras Fault located along the eastern margins of the Sunol Valley cuts across the running pipelines. Considered one of the most active faults in the Bay Area, with the expected rupture displacement of more than four feet lateral and one foot vertical, the fault poses a major threat to reliable transmission of water to the Bay Area. To improve water transmission reliability across the Sunol Valley, a new pipeline and related infrastructure was designed to withstand the expected fault rupture displacement and provide a continuous supply of 120 million gallons per day.

Pipeline damage correlations are used for risk studies of water distribution systems. One of the parameters correlated with pipeline damage in permanent ground deformation zones is horizontal (lateral) ground strain. Some damage correlations use the maximum absolute value of the two principal strains around the pipelines as the lateral ground strain parameter. However, strains caused by lateral ground movements can be resolved into components such as axial strains in the direction of the pipelines. Performing such a process for a large urban water distribution system is a relatively complicated task and development of damage relationships requires comprehensive ground deformation and pipeline data. The unprecedented database following the Canterbury Earthquake Sequence in New Zealand (NZ) provides an opportunity to explore the effects of ground strain and pipeline orientation on pipeline damage during earthquakes. This study presents the GIS methodology for using pipeline orientation to evaluate pipeline network damage caused by ground deformation during earthquakes. The geospatial analyses results indicate differences in the responses of asbestos cement (AC) and cast iron (CI) segmental pipelines under compressive and tensile axial strains.

This study focuses on the development of a pipeline performance classification for Axial Connection Force Capacity (CFC) to support the ongoing efforts of developing seismic design guidelines for water and wastewater systems. The only existing seismic design standard that includes a performance classification system for pipelines is ISO 16134. While this standard recommends performance, levels based on axial CFC (referred to as joint “slip-out resistance” under axial tension loading), it only considers ductile iron (DI) pipe and does not apply to other common pipe materials, connection types, or system components. Previous studies have defined the axial CFC of segmented PVCO (molecularly oriented polyvinyl chloride) pipes relative to Earthquake-Resistant DI Pipe (ERDIP) systems that have performed well during past seismic events and associated earthquake-induced permanent ground deformations. This study expands on previous work by adapting the ISO 16134 classification system for quantifying the CFC to additional pipeline materials and connections common to water distribution and wastewater collection systems. These pipeline materials and their system characteristics, such as joint/connection geometry, burial depth, and backfill soil conditions, affect seismic demands and deformations by influencing the frictional resistance along the pipe length and joint/connection as it undergoes axial deformations. This study quantifies a ratio between the CFC of a particular system of interest and the ISO ERDIP CFC, defined as K₁. The study demonstrates how this conversion factor, calculated based on analytical and experimental results, can be used to classify a pipeline systems’ CFC performance class. The enclosed evaluation of the frictional forces generated under significant ground movement (seismic demand), and the CFC required for a particular system to accommodate these various levels of potential demand, is intended to support industry guidelines for the seismic design of buried water and wastewater pipeline systems.

Metro Vancouver plans to install twin 914-mm internal diameter sewer forcemains under the Fraser River. Liquefaction-induced permanent ground displacement presents a major seismic hazard to the pipeline. A pipe-soil interaction finite element model was built in Abaqus to verify the pipeline design. The pipeline was modeled using a beam-type element that can respond to bending loads with ovalization of the pipe cross section. Ground displacements were applied in multiple modeling steps to capture the anticipated loading sequence and permit the alteration of the soil restraint. Excess pore-water pressure generated during shaking was used to reduce the soil restraint provided by liquefied/partially liquefied soil along the alignment. Steel and HDPE pipeline materials were analyzed in the model. This paper presents the analysis methodology used to justify a pipe wall thickness increase from 6.4 to 9.5 mm in a portion of the alignment and the placement of a critical pipeline joint.

This paper reports on a half-year study to assess the seismic risk of Sacramento Regional County Sanitation District (SRCSD) and Sacramento Area Sewer District (SASD) pipelines and critical facilities. SRCSD mainly collects and conveys bulk sewage while SASD is a local sewage collection agency. While the Sacramento area is not subject to the same level of seismic hazard as nearby San Francisco, it has experienced the effects of earthquakes in the past (e.g., 1989 Loma Prieta earthquake). Since 1950, roughly two dozen seismic events have occurred within 100 km of Sacramento. To understand the risk that these systems currently face, a high-level desktop study that incorporated our best understanding of seismic hazards in the region and how pipelines will respond to various levels of earthquake effects, including strong ground shaking and liquefaction-induced ground failure, was performed. The study quantified current risks and then evaluated whether cost-effective solutions existed to mitigate these risks. Four different products were delivered where each product was useful in evaluating different mitigation strategies to reduce future seismic risks. In some cases, the information was important in determining whether earthquake insurance was a viable means of risk transfer. Additionally, we designed planning strategies to reduce overall impacts by relying on proven effective response and recovery planning activities.

This paper presents the results of a seismic resiliency evaluation program of water mains lined a structural Cured-in-Place Pipe (CIPP) technology, which was used for deteriorated legacy water main renewal and strengthening against earthquake and natural hazard effects. This CIPP product is a composite lining consisting of an inner and outer layer or jacket, produced by circular weaving of continuous textile yarns, impregnated with an epoxy. This product is used to renew water mains by insertion at access pits, without the need for complete excavation of the pipe to be replaced, and was evaluated through a series of large-scale tests at Cornell University’s Large-Scale Lifelines Laboratory, ultimately involving large-scale fault rupture. During the fault rupture test, the CIPP-lined water main was progressively subjected to very large tensile and bending loads, resulting in the water main failure due to joint opening, until debonding and full detachment of the CIPP from the water main, leading to extensive extension and bending of the CIPP. Debonding loads of at least 120 kN were evaluated by means of direct tension testing on CIPP-lined water mains, with lining extension at a water main joint of at least 60 mm prior to debonding. Despite such large deformations the product remained structurally sound and retained water pressure as no water leakage was observed. Post-fault-rupture testing performed on liner samples taken from the highest deformation joint area indicated that residual mechanical properties were similar to pristine liner properties.

The results of three-dimensional numerical modeling are compared with the results of full-scale tests on a pipeline reinforced with cured-in-place pipe (CIPP) performed at the Cornell Large-Scale Lifelines Facility. The results pertain to axial elongation to accommodate earthquake-induced ground deformation of a 6-in. (152-mm)-diameter ductile iron (DI) pipeline lined with ALTRA Proven Water Solutions, formerly known as Aqua-Pipe, Generation 1. Full-scale lab tests were designed to evaluate the axial resistance to relative movement between the host pipe and CIPP. A 24-in. (610-mm)-long section of DI pipe was pulled 36 in. (914 mm) across the CIPP under different rates of axial movement and internal pressure. Both the DI pipe and CIPP were modeled as elastic isotropic materials with Young’s modulus and Poisson’s ratio from tensile coupon test results. Three-dimensional, 8-node, linear interpolation, and full integration elements were used for the pipe and the liner in ABAQUS simulations. The number of elements per pipe and liner circumference was 32. Full-scale tests for internal pressures of 30, 45, and 75 psi (0.21, 0.31, and 0.52 MPa) indicated that the CIPP can sustain higher axial loads as internal pressure increases. The experimental results were shown to compare favorably with the numerical results. The numerical models are used to investigate the effects of initial CIPP diameter, changes in diameter with length, and diameter changes as a function of internal pressure. Rate of loading is shown experimentally to have negligible effects on axial load as a function of axial movement.

An approximately 36-ft (11-m)-long, five-piece section of a ductile iron (DI) pipeline was lined with ALTRA Proven Water Solutions, formerly known as Aqua-Pipe Generation 2 and tested in fault rupture at the Cornell Lifelines Large-Scale Testing Facility. The pipeline had a total of four joints located 5 ft and 15 ft (1.5 m and 3.6 m) north and south of the fault. The test basin was displaced until the full 4-ft (1.2-m) of actuator travel. The lined pipeline maintained its integrity and water pressure of 81.3 psi (561 kPa) throughout the test. Pipeline extension was accommodated by longitudinal movement at the joints, with maximum movement measured at 14.0 in. (356 mm) at the joint closet to the southern side of the fault. Maximum north joint rotation closest to the fault was about 14.0°. The measured and theoretical rotations are in good agreement. The pipeline was able to accommodate a maximum average 30.85 in. (784 mm) of axial extension, corresponding to average tensile strain of 6.15% along the pipeline. No loss of water pressure or structural failure of the lined pipeline occurred. Axial extension during the test is large enough to accommodate the great majority (over 99%) of liquefaction-induced tensile ground strains measured by high resolution LiDAR after each of four major earthquakes during the recent Canterbury Earthquake Sequence (CES) in Christchurch, NZ.

Seismic retrofit of critical underground water pipelines, serving as a key component of lifeline networks, has become an urgent need in engineering practice to improve the performance and serviceability of these critical lifelines. The corrosion protection liner (CPL) technology consists of installing flexible plastic liners with V-shaped anchoring studs inside existing pipelines and subsequently filling cement mortar to the gap between the waterproof liner and the pipeline. With the excellent chemical resistance, impermeability, and fast construction cycles, CPL provides an economic and environmentally friendly alternative for pipeline rehabilitation without large-scale excavation. However, the lack of verification and quantification of the seismic performance of pipelines rehabilitated with CPL remains a critical deficiency for its application as an effective seismic retrofit method for underground pipelines. This study focuses on the response of CPL-strengthened ductile iron (DI) push-on joint by evaluating the liner debonding, relative joint openings produced by axial loadings simulating the seismic loading experienced by the pipeline during an earthquake event. Full-scale quasi-static tests were performed on two water-pressurized DI pipelines with push-on joints before and after reinforcement with the CPL. Moreover, based on the joint response under quasi-static loading, numerical models were developed for capturing the axial behaviors of CPL-reinforced joints that considers strength degradation and energy dissipation. Both the test results and numerical analyses indicate that CPL provides substantial axial strength to the joints of DI pipelines and improve their seismic behavior under high intensity ground motions. But the quality of the heat melting welding of CPL remains a critical uncertain factor, and poor-quality control may be adverse to the seismic performance of the retrofitted pipeline.

Carbon Fiber Reinforced Polymer (CFRP) was designed per AWWA C305-18 Standard to renew sections of 90, 70, and 52-in. diameter riveted steel penstock pipes at a power plant in a seismically active area of California. The primary purpose of the CFRP lining was in 2018 not only to increase the structural capacity of the riveted steel pipe but also increased durability, strength, and corrosion resistance for the pipe. For the design of CFRP renewal, AWWA C305-18 Standard focuses on primary structural loads applied on the pipe but does not consider seismic resiliency In the present study, the seismic resiliency of CFRP liner renewed 90-in. riveted steel pipes was modeled using finite element modeling (FEM). Seismic performance of pipe samples was simulated using permanent ground deformation (PGD) obtained using American Lifeline Alliance 2005 (ALA) guidelines. The seismic resilience of the pipeline depends on the loading direction. In this study, the ground deformation (movement) was considered normal to the pipe axis. The three modes of failure of the renewed pipe for the load resistance factor design (LRFD): (1) tension rupture, (2) local buckling due to compression, and (3) general buckling were investigated and are discussed. This work shows that CFRP lining designed per AWWA C305-18 Standard increased the seismic resilience of a damaged or even an undamaged steel pipe, in addition to the other benefits identified above.

During the last four years, an extensive experimental project has been launched to determine the structural performance of lap welded joints under ground-induced deformations, mainly due to severe seismic events. The research consists of full-scale physical experiments, supported and validated by numerical finite element simulations. Previous experimental results indicated a remarkable strength and deformation capacity of the standard lap welded joints without loss of water containment. The latest phase of the research focuses on the behavior, analysis, and design of a new seismic resistant lap welded joint. Results of a series of additional full-scale experiments on the structural performance of the new lap welded joint under strong axial and bending loading conditions are presented, supported by finite element simulations. The new lap welded joint comprises the standard lap weld configuration with an additional small geometric projection introduced at a specific location near the field applied fillet weld, resulting in consistent buckling of the steel pipe cylinder and not the lap weld joint, during extreme loading. The proposed joint, referred to as “Seismic Resilient joint” or “SR-joint” (patent pending), offers an efficient, reliable, yet economical solution for welded joints in steel water pipelines constructed in seismic areas.

Numerous pipeline-fault crossing studies related to design methods on fault crossings such as fault displacement prediction, impact range setting, and development of countermeasure construction methods were reported recently. It is easy to take proactive countermeasures because the fault map, assumed fault displacement, and tolerance angle between the fault and the pipeline are known. However, the “Seismic Design Guidelines for Buried Pipes” in Japan mentioned the need of countermeasures for fault-crossing buried pipes, but specific design methods and countermeasures were not yet defined. The Japan Society of Civil Engineers Subcommittee has proposed a design formula for calculating the strain generated in buried pipelines (integrated pipelines) subject to fault displacement, and pipeline integrity can be easily judged. However, in this formula, the pipeline is modeled as a beam element, and there is concern about the effect of deformation in the cross-section of large diameter, thin walled pipelines such as water transport pipes. Herein, we report on the application of the design formula to a large diameter pipes which were verified using FEM analysis.

Ever since 1906 San Francisco Earthquake, earthquake damage to buried pipelines due to fault movement, liquefaction, and ground settlement has been seen in large-scale earthquakes. Steel Pipe for Fault (SPF) settlement type was developed to absorb ground displacement and prevent water leakages at connections in ground structures and buried pipelines in order to maintain water flow without cracks, even if ground settlement occurs. The development of this SPF is based on the technology of the SPF original type, which is a solution for pipelines in fault zones. Since the pipe thickness of SPF settlement-type is the same as that of the main pipe, the reaction force is larger when compared to that of general flexible pipes. In order to reduce the reaction force, numerical analysis and modeling were carried out to improve SPF shape. This paper shows several case studies of SPF settlement type in Japan.

Large-scale ground motions have often caused embankment failures. If water pipelines were buried in the embankment, the pipelines have a risk of breaking during embankment failure. Therefore, we need to clarify pipeline behavior in embankment failure and take safety measures to maintain the water flow function of the pipeline. When we analyze pipeline behavior in embankment failure, we must consider the decrease of ground stiffness influenced by large ground deformations. However, a specific method to consider has not yet been established. This study proposes an analysis method considering the decrease of ground stiffness to analyze behavior of pipelines connected with joints in embankment failure. In addition, results of the analysis were compared with measured values of pipeline behavior in actual embankment failure. As a result, the validity of the analysis method was confirmed because the results of analysis could coincide with the measured values well.

In 2018, a moderate earthquake occurred in northern Osaka, Japan, that damaged large-diameter ductile iron pipes used to supply water. The pipe body had large cracks, and the water supply to residents in the surrounding area was stopped. This failure mode is unexpected based on the ductile material properties of the pipes. In this study, tensile and bending strength tests were conducted on damaged and new pipes, and microscopy was used to observe the tensile fracture surfaces to clarify the material properties. The results showed that the excavated pipes did not meet the material strength standard of current ductile iron pipes and exhibited brittle properties. The finite element method was used to simulate the seismic response of a large-diameter pipe along its cross-section. The generated section forces of pipes with different wall thicknesses were compared according to the material strengths obtained from the above tests and current standards to clarify the damage mechanism.

Natural disasters such as floods and slope failures due to heavy rainfalls and strong typhoons have frequently occurred in recent years in Japan. Many water supply pipelines were damaged by deluges and debris flows. No damage, however, was observed for water supply pipelines using ductile iron pipes and seismic resistant joints, which are referred as Earthquake Resistant Ductile Iron Pipes: ERDIPs, originally developed for earthquake countermeasures. The purpose of this paper is to clarify why ERDIPs were effective against deluges and debris flows. To this purpose, the behavior of ERDIPs installed in the liquefied ground at the site of the 2011 Great East Japan Earthquake was studied. Then, the behavior of ERDIPs that were actually undamaged by floods and landslides was clarified in the same way. Finally, the use of hazard maps to create a water supply pipeline that is resistant not only to earthquakes, but also to storms and floods is proposed.

This paper focuses on a complete water supply, transmission, and distribution system for pipe, fittings, valves, and fire hydrants which employ joint systems that deflect, expand, contract, and provide end-wise thrust capacity. Such a comprehensive system from a single manufacturer is of value and interest for all utility system owners and providers, especially those in areas of seismic and storm risk. The system has been tested by Cornell University’s Geotechnical Large Scale Lifelines Testing Facility and certified in compliance with ISO 16134. The audience will learn of recent product developments demonstrating performance through 30-in. diameter and including fittings, valves, and fire hydrants. This investigation will be informative for municipal water system owners and operators and their design consultants for the purpose of writing performance-based material specifications which strengthen water system resiliency. Utilizing pipe products available to everyone plus modern and innovative joint technology, this is a robust and user-friendly system from an installation, operation, and maintenance standpoint. Specifically, the system utilizes AWWA Standards, joints, and construction practices familiar with utilities and installing contractors and meets or exceeds ISO 16134 seismic requirements. A step-wise approach to system-wide design and construction will be presented which addresses areas of highest risk and moves outward to areas of lower risk.

The intensity, severity, and frequency of wildfires throughout the world have significantly increased in recent years, causing damage to structures and lifeline infrastructure in wildland urban interface communities. The 2018 Camp Fire in California damaged nearly 19,000 structures in Butte County and caused widespread contamination of the water within the water distribution system (WDS) in Paradise. The water contamination led to a town-wide water advisory shutoff, impeding recovery of the town as schools, hospitals, and commercial businesses were unable to reopen immediately following the fire. The objective of this research is to explore the relationship between pipe material, pipe temperature, heating duration, and chemical contamination of water within buried water distribution networks, specifically during wildfire heating of service lateral water pipes. This research aims to investigate lifeline resilience through understanding the impact of wildfire on different service pipeline materials and the effect that pipe material may have in the recovery of WDSs after a wildfire. To better understand the impact of temperature and heating duration on pipe materials, the authors perform a series of experiments to determine the critical temperature and duration of heating at which contaminants migrate from pipes into contained water. This paper will help inform utilities of the potential water contamination risk posed by wildfire hazards.

The East Bay Municipal Utility District (District) is a publicly owned utility that supplies water to 1.4 million customers in the San Francisco Bay Area. To address replacement of the District’s 4,200 miles of aging pipelines, the organization embarked on a major initiative to increase the replacement rate of the distribution system from 10 to upwards of 40 miles per year. To accomplish this, a research team at the District was formed to identify efficiencies, develop metrics, and evaluate innovative methods, technologies, and materials used in pipeline replacement projects. Essential to this sustainable pipeline replacement program is considering various AWWA certified pipe materials, products, and fittings for pilot installations to evaluate where cost-effective efficiencies can be gained during design and construction as well as mitigate risk to hazards posed by the District’s proximity to active seismic features, including the Hayward fault which intersects the entire service area. Moreover, to mitigate and reduce the risk of distribution pipeline breaks during a seismic event, the District recently installed AWWA C900 iPVC pipe with RCT Flex-Tite self-restraining fittings at various locations. In parallel, the Center for Infrastructure, Energy, and Space Testing (CIEST) at CU Boulder and the District initiated a collaborative experimental study of this system to evaluate seismic response under axial tension, axial compression, axial cyclic, and four-point bending loading. The study includes full-scale tests on 6-in. diameter DR-14 iPVC pipe specimens and the results indicate excellent performance of the pipeline system. Thus, results of this utility-academic collaboration between the District and CU Boulder highlight experimental testing results, an overview of project-specific geohazards, and lessons learned during construction and installation of this innovative seismic resilient pipe system. Critical research findings established during this collaborative partnership ensure sustained system reliability, sound infrastructure investment strategies, and long-term customer satisfaction.

In this paper, we report a calculation method and calculation example of a fault countermeasure pipeline in a shield tunnel built with earthquake-resistant ductile iron pipes (ERDIPs). When a pipeline is installed in a shield tunnel, it is necessary to consider the interaction between the ground and shield tunnel, the shield tunnel and tunnel backfilling filler, and the filler and pipeline. Therefore, the pipeline performance against fault displacement was evaluated using a composite pipeline model to simulate the pipeline, shield tunnel, ground, and tunnel backfilling filler. We propose a method for designing a pipeline subjected to fault displacement based on the analysis results of the composite pipeline model. The use of a collar or long collar effectively improved the performance of pipelines against fault displacement.

The existing water supply to Wellington (New Zealand’s Capital City) relies on water transmission pipelines that cross the active Wellington Faultline in several locations. These pipelines are expected to break or sustain significant damage in a major Wellington Fault rupture event resulting in extensive parts of Wellington being left without water for up to 100 days whilst repairs are completed. The construction of a Cross-Harbour Pipeline (CHP) (approximately 800 mm diameter and 20 km in length) will significantly improve the disaster resilience and the operational resilience of Wellington’s water supply by providing an alternative pipeline route that does not cross the Wellington Fault. An innovative design approach was required that involved significant geotechnical work at an early stage in the project life cycle to inform the selection of a resilient pipe route and pipe design. This paper will discuss how an approach to a lifelines pipe route selection process was developed (to assess over 700 routes); that was informed by considerable geotechnical assessment and targeted site investigations at an early stage in the project, such that pipe routes being considered were the most resilient available (a primary project objective).

The existing water supply to Wellington (New Zealand’s Capital City) relies on water transmission pipelines that cross the active Wellington Faultline in several locations. These pipelines are expected to break or sustain significant damage in a major Wellington Fault rupture event. The construction of a Cross-Harbour Pipeline (CHP) will significantly improve both the disaster and operational resilience of Wellington’s water supply by providing an alternative pipeline route that does not cross the Wellington Fault. The design approach and route selection process is discussed in the companion paper “The Design Approach to the Cross-Harbour Pipeline, Wellington, New Zealand.” Several geo-challenges may affect the construction and performance of the over 12 km long marine segment of the new pipeline. This, and its alternatives, straddle over soft marine fine-grained sediments, liquefiable coarse-grained soils, the shallow aquiclude layer that caps the aquifer that is the main source of water for Wellington, and some active faults. These challenges and their combinations under various loading conditions are taken into consideration and assessment to inform the selection of a resilient route and the design of it.

Pipelines in regions with moderate and high seismic hazard may be at risk of structural damage and service disruption due to permanent ground deformations (PGD), including earthquake-triggered landslides and liquefaction-induced lateral spread and settlement. Efficient methods of reliability assessment may be required when finite element soil-structure interaction and detailed fragility-based analyses are not possible. This study presents such methods based on a seismic reliability assessment study of large diameter buried pipelines as part of a comprehensive water system seismic study for the Portland Water Bureau (Portland, Oregon). The target audience for this study is engineers as well as water system owners and operators. A case study is presented of three approximately 20 mile long steel pipelines ranging in size from 44 to 66 in. that transport water to East Portland from the Bull Run watershed, and five Willamette River crossings that transport water from East to West Portland. The river crossings include 30 to 60-in. diameter buried pipelines of material types ranging from cast iron and ductile iron to welded steel and bar-wrapped concrete cylinder pipes with welded joints. Segments of these pipelines are subject to as much as 10 ft of combined landslide and lateral spread permanent ground deformations for an M9 scenario earthquake on the Cascadia Subduction Zone. The locations, widths, and magnitudes of the PGD demands on the pipelines were calculated in a GIS framework at the intersection of the pipelines with locations susceptible to landslides and liquefaction. Using available information on the pipeline size, internal pressure, and nominal wall thickness, a simplified analytical analysis was performed to compute maximum strains in the pipelines resulting from transverse PGD. Analytical analysis results were confirmed using nonlinear finite element analyses for representative pipeline segments. Uncertainties in the PGD estimates, width of PGD zones, analytical procedures, and peak ground acceleration were used to perform Monte Carlo simulations to compute individual and joint failure probabilities for the pipelines. Reasonable parameter distributions and values for the analytical pipeline strain and Monte Carlo simulation analyses are provided. The methodology presented in this study uses analytical procedures to assess pipeline vulnerability and provides likely locations of damage compared to more general evaluations typically performed using empirical pipeline fragilities. This approach can be used by engineers performing similar studies to better quantify pipeline vulnerability and provide a more realistic assessment to the owners.

Buried pipelines are an essential underground infrastructure of modern societies, which is susceptible to damage during earthquake events. The characteristics of seismic excitation that impact pipelines are affected by the spatial distribution of seismic waves and the large extent of long pipelines. A shaking table testing program was conducted on scaled buried pipeline model to investigate its seismic response under non-uniform ground motion. The pipelines-sand model was enclosed in a suspension continuum soil box excited using three shaking tables that can induce uniform and non-uniform seismic excitations. The soil bed was uniform dry sand and the model pipeline was 6.0 m in length and 150 mm in diameter. The soil was instrumented with accelerometers along the soil profile, while the pipeline was instrumented with strain gauges and accelerometers around the pipeline. The soil-pipeline model was subjected to twenty-four different ground motions. The recorded data from all instrumentation were analyzed to evaluate the influence of non-uniform seismic excitation on the pipeline response. The results demonstrated that the pipeline response to longitudinal acceleration was larger than the response of surrounding soil under non-uniform excitation. In addition, the peak pipeline tensile and compressive strains to non-uniform ground motion was about twice that under uniform ground motion.

The Metropolitan Water District (MWD) owns and operates the Robert B. Diemer Water Treatment Plant in Yorba Linda, California. The plant, which serves approximately 4 million people in Los Angeles and Orange Counties, was constructed in 1963 in accordance with standards in place at the time. The plant is in a seismically active region and is approximately 0.5 mile from the Whittier Fault, capable of generating Mw6.8 earthquake. The Diemer plant was constructed by cutting off the top of a ridge and placing the excavated materials in the adjacent ravines to produce a level pad. The plant’s 121-in. diameter filter outlet conduit (FOC) is buried about 7 ft while crossing a 400-foot length of the northeast fill slope (NFS). The fill materials were placed without removing colluvium of very low shear strength, and were compacted to the practices of the early 1960s. As such, it was not benched into a competent sedimentary rock as modern practice would dictate. In addition, the fill slope has an inclination of 1.5H:1V, which is steeper than the current practice of 2H:1V. Previous studies and field explorations have indicated that the fill slopes placed in ravines around the plant site including the NFS are only marginally stable under static loading conditions and do not have adequate stability under the MCE event. This led to concerns about seismic slope instability which was confirmed by various analyses. Seismic performance of the FOC was evaluated by performing dynamic soil-structure-interaction analyses using the FLAC 3D computer program. Based on nonlinear seismic deformation analyses, constructability, and consideration of minimum interference with plant operation, a caisson wall (CW) with a length of about 400 ft was selected to laterally support the fill slope housing the FOC. Analysis results indicated that 6-foot diameter caissons spaced at 8-foot on center would seismically stabilize the upper portion of the NFS supporting the conduit. This paper presents the analysis results in terms of lateral displacement, moment, and shear distributions for the CW, the maximum shaking-induced displacements of the FOC and the top of caissons. The effectiveness of the CW in terms of ensuring the structural integrity of the FOC during earthquake shaking is also provided. The results indicated that shaking-induced axial stresses along the FOC are well below the best-estimate buckling stress and the envelope of maximum principal wall stresses of the FOC remain in the elastic range.

Recent extreme coastal flooding caused by hurricanes [e.g., Hurricane Ike (2008) and Sandy (2012)] led to extreme devastation and damage to buildings and critical infrastructure. Typical construction practices for coastal residences in the Gulf of Mexico and on the East Coast involve the use of piles or columns in order to elevate the main living area above the expected inundation level. Such elevated structures can be very effective for extreme events below the design level, since they can be designed to avoid the impact with the waves; however, it is unknown what the structural performance will be during beyond-the-design events. Given the important role of essential facilities, such as healthcare facilities and emergency operation centers, especially during and after the occurrence of natural disasters, this paper investigates the effectiveness of elevating the bottom floor of a simplified structure in reducing the hurricane-induced loads. An extensive numerical investigation was conducted utilizing a FEM-based computational fluid dynamics (CFD) solver and this manuscript will present preliminary results that give an insight into the expected loads. A typical three-story building was simulated in two configurations: (a) supported directly on the soil, and (b) elevated on three-meter tall columns. Following the validation of the numerical method with experimental data and the determination of the optimal numerical settings, a parametric study was conducted for several storm surge levels and solitary wave heights. The study included cases where either the storm surge or the waves reached the bottom slab of the elevated building, in order to account for future sea level rise and intensified hurricane events. The CFD analyses revealed that elevating the structure reduces the horizontal hydrodynamic forces as expected. However, the uplift forces are large especially for the scenarios where it is assumed that part of the 1st story is already inundated by the storm surge and the waves are slamming on the bottom slab of the elevated structure. Moreover, significant overturning moment (OTM) is also generated, which together with the simultaneously large hydrodynamic uplift could introduce significant demand in individual structural components, such as columns, foundations, and piles, and lead to global failure of the building. Given the complexity of the hydrodynamic loading and the observed structural vulnerability in recent coastal events, the results of this study are expected to be of interest to the research community, as well as structural engineers, building owners or disaster response planners that wish to improve the resilience of critical essential facilities on the coast.

This study describes a seismic probabilistic risk assessment (SPRA) of a supplemental cooling water delivery system at a nuclear power plant subjected to multiple earthquake-induced hazards. The system consists of an offsite groundwater well used concurrently for municipal water supply, submersible pumps, an adjacent pumphouse, a pipeline that transports water to a storage reservoir on a hilltop at the plant site, the hilltop reservoir structure and equipment, and buried piping that runs downhill to the plant. During an earthquake, all the system components are subjected to ground shaking hazard. Several of the system components are also subjected to the following seismic-induced ground failure hazards during or immediately following an earthquake: liquefaction, lateral spreading, and slope failure. The system relies on both long-existing infrastructure and recently added components. As a result, in addition to the varying susceptibility of the system components to different earthquake-induced hazards, these components embody a spectrum of construction vintages and prevailing seismic design criteria. This study summarizes the seismic fragility evaluations performed to assess the system component contributions to the overall system failure probability at increasing levels of seismic hazard. Seismic fragility functions were used to characterize the conditional probability of component failure as a function of seismic input. Component failure criteria were explicitly identified or screened out for potential seismic-induced failure modes considering that success of the system function require the ability to deliver cooling water from start to end. Risk insights from the SPRA identified the relative importance of individual components to the overall system risk. These insights can be used to support risk-informed decision-making.

This paper documents outcomes of San Francisco’s efforts to assess post-earthquake interdependencies and restoration performance of lifeline systems serving the city. Two studies, completed in 2014 and 2020, used expert interviews with lifeline providers and cross-sector workshops to understand lifelines impacts from a major earthquake, restoration timelines and processes for each system, and lifelines interdependencies. New issues and changes in restoration performance between the first and second study were assessed. The studies resulted in current and target restoration timelines for all San Francisco lifelines using a common metric of level of service disruption and a clear understanding of interdependencies between lifelines. They also helped identify geographic areas with lifeline system concentrations or other vulnerabilities, key actions needed to achieve target restoration timelines, and high priority mitigation opportunities benefitting multiple systems. These efforts demonstrate the value of qualitative, collaborative approaches to assessing lifeline interdependencies, and restoration performance at a community-scale.

We propose a global sensitivity-analysis-based method for prioritising bridge seismic retrofits in a regional road network with respect to the pth percentile of the road network performance cost (P_p) over a set of disruptive earthquake scenarios. This modifies a previous method that considered the expected cost of the road network performance. The pth percentile of the road network performance cost captures the costly consequences of low-probability earthquakes. We estimate bridges’ total-order sensitivity (Sobol’) indices with respect to P_p using the hybrid-point Monte Carlo approximation method. We show that retrofitting bridges in decreasing order of their total-order Sobol’ indices reduces P_99.9 more efficiently than other heuristic strategies on a network of B = 71 unique bridges in San Francisco. This method leverages existing risk assessment tools and remains computationally tractable while accounting for the probabilistic nature of the seismic hazard, the uniqueness of individual bridges, and network effects.

The Metropolitan Water District of Southern California (Metropolitan) provides approximately 1.4 billion gallons of water every day to 19 million people over a 5,200 square miles service area within southern California. Metropolitan’s system includes the 242-mile Colorado River Aqueduct, 830 miles of conveyance and distribution pipelines and tunnels, 20 dam and reservoir facilities, and 5 water treatment plants. The system traverses an area with numerous faults capable of generating large earthquakes that have the potential to impact water deliveries. Moreover, as an owner and operator of a critical lifeline system, Metropolitan has long recognized the importance of preparing for earthquakes and the associated seismic hazards. Accordingly, the organization has taken steps to improve the seismic resilience of the system to limit the impacts on water deliveries to our customers. Metropolitan employs a defense-in-depth approach for its seismic resilience strategy with the goal of preparing for and effectively responding to seismic events. The major components of this strategy include maintaining adequate emergency water supplies, hardening facilities to limit damage from seismic hazards, and maintaining a robust emergency response program. This paper will describe the implementation of Metropolitan’s seismic resilience strategy, focusing on how the activities support the organizations goals for seismic resilience.

To develop seismically resilient water systems, it is essential that the damage potential of a pipeline system be quantified for future earthquakes. In this study, a large stochastic catalog of earthquake simulations that adapt the Uniform California Earthquake Rupture Forecast Version 3 (UCERF3) model was used to represent the regional seismicity of the Los Angeles Basin. Random event footprints for earthquake simulations were constructed by utilizing the NGA West 2 ground motion models. This set of earthquake simulations was utilized to evaluate system-level consequences for the City of Los Angeles water pipeline network, measured by the total number of pipeline repairs and subsequent repair costs and times due to strong ground shaking and ground deformations. These estimates of damage and impact were based on empirical pipeline fragility models and restoration data from two past events that affected the water system in the past (1971 San Fernando and 1994 Northridge earthquakes). System-level performance was then evaluated at various targeted probability levels and influential seismic sources were identified. This study was performed as part of a long-term program administered by the City of Los Angeles Department of Water and Power to quantify and ultimately enhance the seismic resilience of all city trunklines and distribution pipelines.

Water pipe networks are one of the lifeline networks of modern cities. Despite the importance of these networks, empirical evidence shows a substantial chance of these networks being damaged by seismic events causing significant disruption to their serviceability. To avoid such disruptions, utility managers often need to identify effective maintenance schemes while minimizing use of a limited rehabilitation budget. This identification is often made challenging by the inherent uncertainties associated with seismic events, uncertainty associated with seismic fragility estimation of water pipes, exponentially increasing decision space, and limited maintenance budget available to the utilities. Hence, there is a critical need for an effective approach to identify optimal seismic maintenance plans for minimizing post-earthquake serviceability loss. In this study, a stochastic combinatorial optimization formulation was used to model the problem. Monte Carlo simulation integrated with quasi-pressure-driven hydraulic analysis was used to evaluate the serviceability of the network for each maintenance scheme. Two commonly used metaheuristics, namely genetic algorithm and simulated annealing, were used to identify the optimal maintenance scheme that minimizes the post-earthquake serviceability loss. The effectiveness of the methodology was demonstrated by using two city-scale water pipe networks as the testbeds. The performance of these metaheuristics was then compared in terms of their computational runtime and solution quality. The comparison showed that both genetic algorithm and simulated annealing based optimization can effectively identify economical seismic rehabilitation plans for city-scale water pipe networks. The findings of this study are expected to be highly valuable to utility managers managing water pipe networks exposed to significant seismic risks.

The effects of hurricanes with respect to infrastructure resilience are reviewed with reference to Hurricanes Katrina and Sandy. The effects of Hurricane Sandy on New York City and subsequent programs to improve the City’s infrastructure are described. Special attention is focused on the restoration of the L Line Tunnel, which was flooded by Hurricane Sandy. The paper describes how a team from Cornell and Columbia Universities was assembled at the request of Governor Andrew Cuomo to help re-engineer a $0.5 billion project to rehabilitate the tunnel, and still keep the subway in service. The new approach integrates several advanced technologies, including distributed fiber optics and light detection and ranging (LiDAR). The agents of change that lead to improved policies and approaches are explored, including the technical, institutional, and social challenges of introducing new technologies and engaging community support.

The resilience of underground infrastructure to large ground deformation depends on the ability of pipelines, cables, and conduits to accommodate the geometric nonlinearities in soil by changing shape through axial elongation/compression, flexure, and rotation at joints. This paper focuses on the development of the next generation hazard resilient infrastructure through large-scale testing and numerical modeling. With the assistance of the Cornell Lifelines Large-Scale Testing Facility, ten new pipeline and conduit systems have been developed and commercialized using a protocol of large-scale tests and fault rupture experiments that define and confirm performance under extreme conditions of ground deformation. Resilience involves the capacity of the pipelines to accommodate large ground deformation from earthquake-related movements associated with fault rupture, liquefaction, and landslides. It also involves the accommodation of ground movement caused by hurricanes, floods, tunneling, excavations, and subsidence related to mining and dewatering. The development and validation of analytical and numerical models for soil-structure interaction are described. The performance of the new systems is discussed. Examples of ductile iron, polyvinyl chloride, and steel pipelines, as well as those reinforced with cured-in-place pipe and pipe linings, are used to illustrate the performance of next generation hazard resilient infrastructure. Next steps in the development of hazard resilient infrastructure are discussed, which include the incorporation of smart sensor technologies.

The Southern California Earthquake Center CyberShake platform was designed as the first probabilistic seismic hazard analysis (PSHA) model using physics-based earthquake simulations. Provided that the simulations have been properly validated, they should, in theory, capture the source, path, and site effects of a specific region. For this work, we build on a recent comprehensive probabilistic seismic risk analysis (PSRA) study of the underground water pipeline network for the City of Los Angeles, where the system-level performance (measured by the expected number of pipeline repairs, repair cost, and repair time) was established as a function of exceedance probability based on a large set of earthquake simulations using empirical ground motion models (GMMs). By repeating this study using the events and ground motions from the CyberShake simulations, we are exploring the impact of this region-specific PSHA models on seismic risk assessments of distributed infrastructure. Our results show that using the same set of earthquake ruptures and pipe fragility models, CyberShake simulations result in significantly lower estimates of system-level risks than those computed from empirical GMMs at relatively long return periods. More specifically, at a 500-year return period, the expected number of pipeline repairs computed from CyberShake is about 26% lower than that from empirical GMMs. At a 2,500-year return period, the reduction reaches 41%. While a reduction was expected, more work is needed to fully understand the sources of these difference.

Water distribution networks (WDNs) are critical infrastructure that provides essential support for the community life. Its service can be interrupted by natural hazards such as earthquakes. The resilience of a WDN is related to its capability to resistant damages as well as its speed to recover the service function. This paper introduces the development of a machine learning (ML) model to ensure fast restoration of WDN subjected to earthquake damages, and therefore improves the system resilience. The reinforcement learning (RL) model is trained to optimize WDN restoration sequence considering consumer needs and evolving water demands. The results show that compared with the repair sequence by the commonly used method of static importance ranking, repair sequence by RL model substantially improved the recovery speed of WDN. Besides, under different damage scenarios, the repair sequence by the RL agent consistently achieved efficient recovery (and therefore enhanced resilience), demonstrating the robustness of the RL model. This study demonstrates the potentials of artificial intelligence (AI) techniques to support optimal decision in managing the WDN under emergency conditions.

Rural power distribution systems are evidentially vulnerable to extreme hazards, particularly hurricanes in coastal regions due to low or zero redundancy, low resourcefulness, and aged infrastructure. Some resilience measures exist for quantifying the resilience of infrastructure systems, which can be applied to these rural distribution systems. However, objective decisions based on such measurements are not ready to make due to the lack of framework to assess the change in resilience measure effectively while altering the information of the input parameters. Hence the questions, such as if the infrastructure system is more resilient when subject to some changed input parameters and, if obtained resilience measure is the optimal value for the given infrastructure system considering the available constraints, are not readily answered in terms of existing measures. Thus, this study explores several statistical and information-theoretic distance measures to characterize how a system evolves as a function of system variables from the realm of materials, hazards, or socioeconomic resourcefulness. Numerical evaluation is conducted with stochastically modeled rural distribution systems, which confirms the effectiveness of the proposed framework. Two resilience distance measures, Cramer-Von Mises Distance (CVMD) and Earth Mover’s Distance (EMD), are selected and recommended.

Infrastructure system resilience prior to or following disruptions due to natural or technological hazards is intimately linked with and supports community resilience. This paper presents a framework, consisting of eight key elements, connecting processes and tools for assessment, management and governance related decisions, and the community outcomes. It recognizes infrastructure as interdependent socio-technical systems capable of achieving resilience through optimized flow and provision of services to users that satisfy community-level objectives by reducing social and economic losses while enhancing community wellbeing. In this paper, an overview of the framework is provided.

The seismic vulnerability and risk assessment of infrastructure and utility systems are essential to prevent or mitigate sufficiently the negative consequences, implement resilience management strategies, and recover efficiently after a major earthquake. In a complex urban environment, having multiple interacting and interdependent infrastructures becomes even more important. Earthquake hazards not only affect a single asset, but also their impact is much greater because of the inter- and intra-dependences among various infrastructure, utility systems, and lifelines. Therefore, we urgently need efficient tools to quantify and assess the systemic vulnerability and risk of urban infrastructure and utility systems. This is a challenging topic that is nowadays receiving more attention from the research community, the industry domain, and the policymakers. This paper aims to review the available modelling approaches and tools for the seismic risk analysis of interconnected systems, including advantages and limitations. It focuses in particular on the European funded SYNER-G project that encompasses interdependencies, delivers a holistic methodology, and implements a comprehensive framework based on the Object-Oriented Modelling paradigm. The capacities of the SYNER-G framework are illustrated through a selected application regarding the seismic risk analysis of interconnected infrastructure and utility systems in the city of Thessaloniki, Greece. Among other aspects, the paper discusses hazard modelling issues of the two common approaches, the probabilistic and the scenario-based procedure and illustrates in a specific example the impact of mitigation strategies, based on their effect on the performance of the interconnected systems and the overall loss reduction. The integration of interdependencies into the risk analysis and resilience strategies facilitates a better understanding of critical infrastructure operation and enables well-informed proactive and reactive decision-making and efficient disaster risk management, by infrastructure owners and operators, insurance companies, consulting agencies, and local authorities.

Housing disaster resilience is crucial for community disaster resilience as it provides two essential functions: shelter from the elements and functional housing. We present a methodology to quantify community housing disaster resilience considering the unmet post-disaster community demand for both shelter and functional housing. Total Sobol’ indices are used to evaluate the importance of the vulnerability and recoverability of individual components of community infrastructure systems and building stock units for community housing disaster resilience, while requiring no prior knowledge of these component properties. We illustrate the application of the proposed importance analysis on a virtual community considering the functionality of the electric power supply, water supply and communication infrastructure systems, and a number of building stock units.

Debris removal is a critical activity in the aftermath of natural disasters such as earthquakes and tsunamis to enable community and lifeline network recovery. This activity is hampered by logistical bottlenecks including the non-availability of equipment and inadequate capacity of temporary debris management sites (TDMS). This paper enables analysis of debris removal and lifeline repair operations quantifying recovery times for informed decision-making about equipment allocation and TDMS selection before the disaster. The developed framework was applied to the case study of a Cascadia Subduction Zone event for the coastal town of Astoria in Oregon. The proposed framework enables decision-makers with an objective means of evaluating decision alternatives both before and after disasters to analyze and improve their community’s capability of handling disaster debris. Furthermore, this framework will serve as a platform upon which interdependencies between transportation network and debris removal operations will be analyzed in the future.

This research proposes a framework that allows to define a debris removal strategy from a road transportation network after a seismic event. The case study is a hypothetical large-scale city consisting of many interdependent infrastructure. Once the debris generated by the collapse of buildings have been estimated, blocked roads are identified. Cleanup operations are then prioritized based on road importance and travel time. The goal is to first verify that evacuation routes and important paths connecting strategic facilities such as hospitals, shelters, fire stations, etc., are available. In case some roads within these paths are blocked, alternative routes are considered. If the pre-event travel time does not significantly increase, clearing equipment and resources could be managed accordingly and directed towards other areas. The objective of this work is to help emergency managers to successfully improve disaster response avoiding delays during rescue and recovery operations.