Download PDF Summer Bridge Issue on Engineering for Disaster Resilience July 1, 2019 Volume 49 Issue 2 The articles in this issue present examples of engineering innovation to develop resilient infrastructure. Increasing Community Resilience through Improved Lifeline Infrastructure Performance Wednesday, July 3, 2019 Author: Christopher Rojahn, Laurie Johnson, Thomas D. O’Rourke, Veronica Cedillos, Therese P. McAllister, and Steven L. McCabe The concept of community resilience is complex and multidimensional, relying on engineering and other disciplines to help communities break the cycle of destruction and recovery and reduce the impacts of earthquakes and other hazards. This article presents proposed prioritized actions to improve lifeline infrastructure resilience based on an assessment of lifeline infrastructure performance commissioned and funded by the National Institute of Standards and Technology (NIST). Introduction Resilience involves the ability of people and communities to adapt to changing conditions and to withstand and rapidly recover from disruptions (White House 2011). At the community level, this concept is complex and multidimensional, relying on contributions from the social sciences, engineering, earth sciences, economics, and other disciplines to improve the ways communities prepare for, resist, respond to, and recover from disruptions due to either natural hazards or man-made causes. Resilience is intended to reduce both the impact of -hazards by restoring community functions within a specified timeframe and the duration and cost of -recovery. This requires planning for recovery and restoration prior to hazard events. Disasters interfere with electric power, natural gas and liquid fuel, telecommunications, transportation, and water and wastewater infrastructure systems. Such systems are commonly referred to as “lifelines” because they are vital for the economic well-being, security, and social fabric of the people they serve (NIST 2014). Lifeline systems are often distributed over large geographic regions and have numerous interdependent -levels of operation, making them especially exposed to the impacts of earthquakes, hurricanes, and other hazards that affect broad areas.They are therefore vulnerable to distress and malfunction at many locations, which, in turn, impedes response and recovery. Lifeline infrastructure system failures and disruptions displace people, obstruct social and economic institutions, and in the worst cases lead to death and long-term negative societal consequences. The NIST study overviewed in this article (a) assessed societal expectations of acceptable lifeline infrastructure system performance levels and (b) proposed actions pertaining to policy, modeling, systems operations, and research needs that will facilitate improved lifeline infrastructure performance during disasters. The NIST assessment was carried out to inform users of the Community Resilience Planning Guide for Buildings and Infrastructure Systems (NIST 2016a), which provides a practical and flexible approach to help communities improve their resilience by setting priorities and allocating resources to manage risks for their prevailing hazards. The assessment is described in detail in the NIST (2016b) report, Critical Assessment of Lifeline System Performance: Understanding Societal Needs in Disaster Recovery. Background We begin with an overview of (a) the current state of lifeline infrastructure design and construction codes, standards, guidelines, and performance requirements; and (b) impacts of disasters on lifeline infrastructure systems. Codes and Standards of Practice Codes, standards, guidelines, and manuals that govern the design, construction, and performance of lifeline infrastructure systems and components vary considerably from system to system and represent various levels of consensus, typically among operators, regulators, and engineering experts. Variability in Requirements and Enforcement Performance requirements in codes and guidelines focus mainly on engineering for system design, construction, and operation and tend to emphasize minimum levels of safety or the performance (e.g., for extreme loading conditions) of components as opposed to system response and levels of service. Most address day-to-day operations and do not cover the full range of hazard types that affect infrastructure systems, particularly low--probability, high-consequence events. For electric power, natural gas, and liquid fuel systems, certain measures of system performance are routinely assessed and even required to be reported to regulators. But these measures generally address outages during normal operations and often exclude disruption caused by hazards. The three systems have regulatory data requirements for safety and reliability, and for gas and liquid fuel pipelines there is a national regulatory framework defined by federal legislation. Federal regulations also guide telecommunication systems. While most US jurisdictions adopt the latest codes and standards and some add more stringent requirements to them, others may not adopt them in their entirety or may even reduce some requirements. Even if codes and standards are adopted, their effectiveness may be compromised by poor enforcement during the planning, design, and construction of infrastructure components. This disparity in code adoption and enforcement can significantly degrade community resilience through regional dependencies and cascading consequences. Societal Considerations Codes and standards give the greatest emphasis to failures in particular infrastructure systems (e.g., natural gas, liquid fuel, transportation) and to infrastructure service outages (e.g., power, telecommunications , natural gas, water, wastewater) that can contribute to mortality and morbidity. They also take into account different but more limited societal considerations such as life safety, public health, emergency response, critical service provision, property and monetary loss prevention, and environmental protection. As discussed below, most system performance measures are not informed directly by or linked to societal expectations and needs. The direct and indirect costs to customers will likely vary with the duration of system disruptions (i.e., mere inconvenience in the first few hours to severe hardship after weeks of lost service), but system performance measures rarely consider the differential hardships imposed on society from varying durations of outage. Disaster Impacts on Specific Infrastructure Systems There is a substantial body of literature on the social and economic impacts of infrastructure service disruptions, spanning studies on actual events—both hazard and nonhazard related—and scenario-based and probabilistic loss projections. Most studies of disaster-related impacts on infrastructure systems are event specific; systematic, multievent studies are generally rare. Also, in general there is more information and a better understanding of the societal impacts and restoration patterns of short-term rather than longer-term disruptions. We briefly describe disaster-related impacts on several types of lifeline infrastructure. Electric Power Most major and widespread electric power outages are due to storms or other weather events, and there are more data on such events than other hazards, particularly the technical aspects of component and system failure and restoration. Common information for electric power performance in hazard events includes standard industry measures, peak number of customers without service, and time to restore service to all (or nearly all) customers. Gas and Liquid Fuel Gas and liquid fuel production, transmission, and distribution systems are susceptible to damage in most hazard events. Loss of power to oil refineries and pipeline pump stations during high wind and coastal inundation events causes loss of production and transmission, interrupting fuel supply for businesses and consumers. After Hurri-cane Sandy, for example, disruptions at nearly every level of the fuel supply chain reduced all fuel flow into and within the New York City metropolitan area. Ground faulting and liquefaction caused by earthquakes can lead to the rupture of gas and fuel pipelines, but appropriate design measures can mitigate these effects, as demonstrated by the good performance of the Trans-Alaska Pipeline System during the 2002 Denali Fault earthquake (Hall et al. 2003). But the fact that some lifeline infrastructure systems are confined within states and others distributed nationally leads to differential impacts. The spatial distribution of the nation’s natural gas delivery system, for example, is nationwide (figure 1). Pipelines originating in -Louisiana and crossing the Mississippi Valley convey natural gas for heating and cooling in heavily populated areas of the Northeast and Midwest. They are subject to disruption by hurricanes, river flooding, and earthquakes that originate in the New Madrid Seismic Zone, which crosses Kentucky, Missouri, Tennessee, and Arkansas. Figure 1 Moreover, some locations are serviced by a single pipeline. For example, more than 90 percent of refined petroleum products bound for Portland, Oregon, follow a single route that is at risk from liquefaction-induced ground failure during an earthquake as well as from hurricanes and floods (NIST 2016a). Transportation Transportation systems are susceptible to damage or disruption due to a variety of natural events. Common earthquake damage includes bridge failures (figure 2) and landslides that can hamper emergency response, particularly to remote communities. Scour induced by inundation or flooding can also result in bridge, rail, and roadway failures. Figure 2 Weather events involving wind, snow, and ice can cause disruptions but physical damage tends to be more limited, so data on transportation performance in earthquakes, -tsunamis, coastal inundation, and riverine flooding are more readily available. Water and Wastewater Systems Water and wastewater systems are susceptible to damage during earthquakes, tsunamis, and -other forms of inundation. Disruptions to water supplies can have serious impacts on fire-fighting capacity and sanitation with adverse public health, safety, and economic consequences. Figure 3 Water service restoration times for Los Angeles after the 1994 Northridge earthquake, for example, are shown in figure 3. They varied significantly for different services (Davis 2014; Davis et al. 2012). Restoration of potable water to households and the fire department, for instance, lagged that of nonpotable water deliveries and normal quantity. And a return to preevent system functionality, as well as achievement of improved reliability, was a long-term process that lagged the restoration of other services and actually required many years to achieve. Overarching Societal Considerations Most infrastructure system outages last from hours to weeks (short- to intermediate-term recovery). In severe cases, outages can last for months or even years. Long-term outages are associated with the most destructive events, when critical, large, and/or multiple components of lifeline infrastructure systems that are time consuming to replace—such as bridges, piping, and essential equipment—must be reconstructed or replaced to restore system operability. Societal Expectations and Tolerance Empirical data on public expectations of acceptable infrastructure performance after disruptions are sparse, and gauging societal expectations is challenging because US society is highly diverse. Expectations vary among individuals, households, and businesses as a function of a number of factors, such as vulnerability and resilience characteristics; geographic location; hazard characteristics (e.g., severity, probability, and duration); infrastructure type; available information on the impacts of disruption; prior experience and knowledge about service disruptions; levels of resulting losses; public perceptions of the trustworthiness and competence of service providers; and availability of substitutes and contingencies that can compensate for system outages. Additionally, there is evidence to suggest that societal expectations and tolerances may be changing as social and economic activity becomes more dependent on highly reliable service provision, particularly electric power and telecommunication systems. One approach to assessing likely societal expectations is to look more closely at how infrastructure performance and disruption can have deleterious effects on what society members value most. This approach is consistent with NIST (2016a), which uses Maslow’s “hierarchy of needs” framework to prioritize different building and infrastructure systems in communities. For this reason, human health and safety, the functionality of healthcare systems, and economic well-being are deemed priority areas when exploring societal considerations regarding performance. Risk Perception and Communication Public expectations and tolerances for infrastructure service disruptions are dynamic and likely to be shaped by both risk perception and risk communication. Factors that affect risk perception include prior experience with hazards and outages, substitutability and dependency on lost services, and available information about the impacts of disruption. Other things being equal, disruptions may be tolerated for longer periods in severe and catastrophic events than in less serious ones, because the public will be more willing to accept the difficulties that extreme hazard events pose for service providers to anticipate and mitigate. Public confidence and the past performance of infrastructure service providers (during both routine operations and hazard events) can also influence expectations and tolerances. Public perceptions and expectations are also shaped by communications about the risks associated with past disasters and outages. Infrastructure system service providers can work with emergency management, public safety, and other governmental agencies to ensure that risk communication messages reach and are understood by the affected public. Differential Vulnerability Various segments of the population and sectors of the economy are differentially exposed, sensitive, and adaptable to infrastructure service disruptions. Risks associated with infrastructure service disruptions are not borne equally by all members of society but are imposed disproportionately on already vulnerable social and economic groups. Infrastructure Interdependency Considerations Dependent and interdependent relationships among infrastructure systems have evolved over time, with various systems and technology advances expanding and linking systems together. The NIST (2016b) assessment considered interactions among different infrastructure systems during normal operations and restoration after hazard-related events. Interdependency mechanisms broadly classified as physical, geographic, cyber, and logical are not necessarily mutually exclusive (Rinaldi et al. 2001): Physical interdependence: the performance of one network depends on the outputs of others. Geographic interdependence: systems are colocated or in close proximity. Cyber interdependence: the interdependence between two networks is based on shared information (e.g., the “smart grid,” which relies on telemetry and situational awareness data). Logical interdependence: systems are interconnected through channels different from the preceding three, for example based on human decisions related to restoration prioritization among systems. The NIST assessment revealed critical dependencies and interdependencies across infrastructure systems. Virtually every infrastructure system today depends on electric power and telecommunications for control and monitoring. All infrastructure systems also depend on fuel and transportation, particularly for service restoration and system repairs. Fuel is a critical contingency for power when outages occur. Water is critical for cooling in the generation processes for electric power. Water also helps with pollution control and supports other infrastructure, such as natural gas and liquid fuel systems. Some interdependencies are increasing, as evidenced by the expanding role of telecommunications and electric power in monitoring and remote control of infrastructure systems as well as household management, personal choices for renewable power, and community planning for decentralized energy. A simplified set of dependencies for communication systems is shown in figure 4. Figure 4 The physical proximity and colocation of multiple infrastructure systems can enhance efficiencies but also increase risks of cascading failures and complex -interactions in restoration, as well as risks posed by multi-hazard effects. Interdependent systems may also require a cross-system, cross-organizational, and integrated approach to planning that is difficult to implement. And choke points may amplify interdependencies within systems as well as between infrastructure systems and community-level processes. Proposed Actions for Improving Lifeline Infrastructure Performance Infrastructure resilience is clearly essential to the country’s economic and social well-being. And according to the National Institute of Building Sciences, the nation saves $4 in future disaster costs for every $1 invested in utility and infrastructure mitigation activities (MMC 2018). The federal Disaster Recovery Reform Act, signed into law in October 2018, established a National Public Infrastructure Pre-Disaster Hazard Mitigation Grant Program for public infrastructure projects to increase community resilience before a disaster occurs. In this context, the NIST (2016b) assessment identifies and prioritizes needed improvements to codes, standards, and guidelines; modeling; system operations; and proposed research topics. Codes, Standards, and Guidelines Ten actions, in ranked order, are proposed to address needs related to codes, standards, and guidelines that govern the design, construction, and performance of various lifeline infrastructure systems (“lifelines”) and system components. The priority rankings reflect organizational and framework needs, available information, new knowledge needs, guideline and standards development needs, and scoping breadth. Proposed actions pertaining to broad issues and improved community resilience have higher priority than proposed actions for specific lifelines. Identify or establish an organization and process for advocating, harmonizing, and unifying consensus procedures for lifeline guidelines and standards development. Develop more consistent terminology for lifeline standards. Develop an up-to-date and complete suite of codes, standards, and guidelines for all lifeline systems to reflect the current state of practice, knowledge, and performance requirements. Develop a methodology to combine component-based design criteria with system-level performance targets. Develop lifeline system performance requirements that relate to community resilience and better reflect societal considerations. Develop consensus-based guidelines and standards for the design of new lifelines and the retrofit of existing lifelines to reflect community resilience performance requirements and societal -considerations. Develop guidelines to inform the design, interoperability, and upkeep of lifeline system dependencies. Reduce inconsistencies in codes and standards for the design, construction, and resilience of the built environment (e.g., fire codes, building codes, and codes, standards, and guidelines for lifeline systems). Develop consistent policy and standards on accessing information and databases about critical infrastructure systems that are coordinated with activities of the Department of Homeland Security. Provide updated guidance for evaluating gas and liquid fuel pipeline and facility response to seismic hazards, floods, coastal storms, and tsunami-related inundation. Modeling System modeling for lifeline systems and their interdependencies can be leveraged to improve resilience across such systems. The following proposed improvements to address limitations in scope, outputs, integration, and validation are considered high priority. Aggregate existing infrastructure modeling tools and create a user-friendly interface so communities can properly assess their lifeline-related system performance and restoration risks, including -uncertainty. Develop first-generation models and practical tools for community resilience analysis that account for lifeline system dependencies and -interdependencies. Improve numerical modeling of water and waste-water systems, with emphasis on validation of -models, development of the most effective simulation procedures, and applications in real systems. Infrastructure System Operations The following proposed high-priority actions address needs related to lifeline system operations and operational design. These needs must be addressed to improve community resilience and bridge the gap between the postevent capabilities of lifeline systems and societal expectations of their performance and restoration. Develop a process for major utilities to conduct self-assessments of their preparedness for various natural hazard events, as a basis for prioritizing improvements to system robustness and postevent response. Develop guidance for lifeline service providers on how to engage and collaborate with communities, including emergency management agencies and other key community institutions, in developing resilience strategies and preparing system restoration and contingency plans. Develop guidance for local planning (e.g., for fuel delivery to emergency responders and critical -infrastructure). Develop guidance for lifeline service providers to evaluate the effects of system component failures, both in isolation and in combination, and considering upstream and downstream dependencies. Design protocols for lifeline service providers, working with emergency management and other community institutions, to communicate to the public the likely impacts of different hazard events on service provision and disruption. Research Fifteen research topics are proposed to address gaps in data and knowledge needed to improve understanding of acceptable infrastructure performance. All the topics are considered high priority. Gather information on and systematically study the relationships between service disruptions and societal impacts and expectations to better understand lifeline system performance. Assess societal expectations associated with lifeline system performance. Systematically study and compare design approaches and methods for addressing societally based performance requirements as set forth in current codes, standards, and guidelines for lifeline systems. Investigate the differential vulnerability among social groups to lifeline system outages. Systematically collect and review various “proxies” and secondary evidence for societal expectations of lifeline performance and restoration timeframes. Assess lifeline performance programs and practices for public safety and develop guidance on their application to other critical lifelines, including multiple, interdependent systems and colocated facilities. Conduct research on needed service restoration times, including the role of system operability, as a performance metric, in supporting community -resilience. Study lifeline system operator organizational issues and how they affect community-scale lifeline performance and resilience planning. Enhance understanding of infrastructure-related failures and cascading effects from low-probability/high-consequence events. Develop postevent data collection protocols to assess lifeline system recovery and restoration timeframes and improve understanding of restoration processes across individual and interdependent lifeline systems. Develop tools to identify (a) interdependent infrastructure systems and services and (b) their restoration criteria. Establish procedures to quantify hazards for spatially distributed systems. Enhance understanding of lifeline system supply sources and endpoint facilities and their role in system performance, restoration, and community and regional recovery with the goal of improving databases and modeling of such sources and facilities. Study changes in water demand considering an array of hazards as well as seasonal and longer-term climate variability (e.g., drought). Improve knowledge, databases, and modeling for impacts of widespread flooding and storm damage on regional fuel supplies. Conclusion The NIST assessment of lifeline system performance identifies a number of significant deficiencies in the current state of lifeline infrastructure design and construction codes, standards, guidelines, and performance requirements. It proposes prioritized actions to advance policy, modeling, systems operations, and research. In these ways it provides a roadmap to ensure that future investments in infrastructure resilience not only improve lifeline infrastructure performance during disasters but also better match societal expectations of acceptable system performance. Acknowledgments The authors gratefully acknowledge Stephanie Chang, Craig A. Davis, Leonardo Dueñas-Osorio, Ian -Robertson, Henning Schulzrinne, and Kathleen -Tierney for their contributions in developing the NIST (2016b) assessment and the resulting recommendations. We similarly appreciate and recognize the attentive review of the NIST (2016b) document by Bruce -Ellingwood, Timothy J. Lomax, Douglas J. Nyman, Dennis Ostrom, Jon Peha, and Kent Yu. -Finally, we sincerely appreciate Cameron Fletcher’s insightful edits to this article. References Davis CA. 2014. Water service categories, post-earthquake interaction, and restoration strategies. Earthquake Spectra 30(4):1487–1509. Davis CA, O’Rourke TD, Adams ML, Rho MA. 2012. Case study: Los Angeles water services restoration following the 1994 Northridge earthquake. 15th World Conference on Earthquake Engineering, Paper No. 364, International Association of Earthquake Engineering, Tokyo. Hall WJ, Nyman DJ, Johnson ER, Norton, JD. 2003. Performance of the Trans-Alaska Pipeline in the November 3, 2002 Denali Fault Earthquake. TCLEE Monograph No. 25. Reston VA: American Society of Civil Engineers. MMC [Multihazard Mitigation Council]. 2018. Natural Hazard Mitigation Saves: 2018 Interim Report. Washington: National Institute of Building Sciences. NIST [National Institute of Standards and Technology]. 2014. Earthquake-Resilient Lifelines: NEHRP Research, Development and Implementation Roadmap. Report GCR 14-917-33, prepared by the NEHRP Consultants Joint Venture, a partnership of the Applied Technology Council and Consortium of Universities for Research in Earthquake Engineering. Gaithersburg MD. NIST. 2016a. Community Resilience Planning Guide for Buildings and Infrastructure Systems, Vols I and II. NIST Special Publication 1190. Gaithersburg MD. NIST. 2016b. Critical Assessment of Lifeline System Performance: Understanding Societal Needs in Disaster -Recovery. Report GCR 16-917-39, prepared by the Applied -Technology Council. Gaithersburg MD. Rinaldi SM, Peerenboom JP, Kelly TK. 2001. Identifying, understanding, and analyzing critical infrastructure interdependencies. IEEE Control Systems 21(6):11–25. White House. 2011. Presidential Policy Directive/PPD-8: National Preparedness. Washington. About the Author:Christopher Rojahn is director emeritus of the Applied Technology Council (ATC). Laurie Johnson is principal, Laurie Johnson Consulting | Research. Thomas O’Rourke (NAE) is the Thomas R. Briggs Professor of Civil and Environmental Engineering in the College of Engineering at Cornell University. Veronica Cedillos is president and CEO of GeoHazards International (GHI). Therese McAllister is leader of the Community Resilience Group and Steven McCabe is leader of the Earthquake Engineering Group, both at the National Institute of Standards and Technology (NIST).