The concept of “resilience” is fundamental to a roadmap for realizing the major technical goals of the 2008 NEHRP Strategic Plan within 20 years. The Strategic Plan articulates a vision, mission, and goals that aim to “improve the nation’s earthquake-resilience in public safety, economic strength, and national security” (NIST, 2008; p. iii). The meaning of “resilience,” however, is far from clear. Numerous definitions of “resilience” exist, and the term is often used loosely and inconsistently. To provide a context and vision for the roadmap, this chapter sets out a working definition of “national earthquake resilience” that includes a brief discussion of conceptual and measurement issues. The discussion draws on committee discussions, the rapidly expanding literature on resilience, and input from more than 50 leading earthquake professionals at an August 2009 workshop sponsored by the committee. Two examples are then provided—Evansville, Indiana, and San Francisco, California—to illustrate how a community might work toward a vision of resilience.
Dozens of definitions of “resilience” can now be found in the literature, reflecting a range of perspectives and a lack of consensus on the meaning of the term. In the context of hazards and disasters, three definitions of resilience that are often cited are:
The capability of an asset, system, or network to maintain its function or recover from a terrorist attack or any other incident (DHS, 2006).
The capacity of a system, community or society potentially exposed to hazards to adapt, by resisting or changing in order to reach and maintain an acceptable level of functioning and structure. This is determined by the degree to which the social system is capable of organizing itself to increase this capacity for learning from past disasters for better future protection and to improve risk reduction measures (UN ISDR, 2006; also SDR, 2005).
The ability of social units (e.g., organizations, communities) to mitigate risk and contain the effects of disasters, and carry out recovery activities in ways that minimize social disruption while also minimizing the effects of future disasters. Disaster Resilience may be characterized by reduced likelihood of damage to and failure of critical infrastructure, systems, and components; reduced injuries, lives lost, damage, and negative economic and social impacts; and reduced time required to restore a specific system or set of systems to normal or pre-disaster levels of functionality (MCEER, 2008).
Of these, the Department of Homeland Security’s National Infrastructure Protection Program (NIPP) definition is narrower in scope than the MCEER (Multidisciplinary Center for Earthquake Engineering Research) definition, and the concept of maintaining function is somewhat vague in the former. It could include maintaining as high a function as possible at the moment the disaster strikes. Alternatively, resilience might refer only to maintaining function through activities undertaken after the event, and hence would not necessarily include pre-event mitigation. This focus on post-shock activities (both inherent and adaptive) and the emphasis on recovery as both goal and process are more consistent with the origins of the term resilience. The United Nations International Strategy for Disaster Reduction (ISDR) definition, in contrast, departs further from the origins of the term and appears to emphasize pre-disaster mitigation and preparedness, with the only allusion to the idea of rebounding from a disaster relating to the speed of recovery. It does, however, emphasize that resilience is a process. This definition is also used in the National Science and Technology Council’s Grand Challenges for Disaster Reduction.
Although the 2008 NEHRP Strategic Plan (NIST, 2008; p.47) adopts this latter definition, for purposes of the roadmap, it is important to consider several issues:
• “National earthquake resilience” should primarily involve building resilience at the level of communities. It is also important, however, to prepare for the rare instances where earthquake disasters could extend beyond localities and have national-level consequences (see Box 2.1).
• In order for communities to be more resilient, support from both state and federal levels is required.
• Building national earthquake resilience should foster synergies between resilience to earthquakes and to other hazards.
• Communities should consider developing multi-tier resilience goals and strategies, i.e., different performance expectations for different scale events. In some cases, it may be effective to focus actions on containing the effects of “expected” events, rather than very rare, “extreme” events.
• Resilience involves both pre-disaster mitigation (activities to reduce the amount of loss in an event) and the ability to mute post-event losses and rapidly recover from an event.
• Resilience should allow for systemic change, especially in low-probability, high-consequence events. Resilience does not necessarily entail a return to “normal” or “pre-disaster” conditions. Reducing future risk should also be a goal of recovery activities.
With these considerations in mind, the committee recommends that NEHRP adopt the following working definition for “national earthquake resilience” (applicable more generally to all-hazards resilience):
A disaster-resilient nation is one in which its communities, through mitigation and pre-disaster preparation, develop the adaptive capacity to maintain important community functions and recover quickly when major disasters occur.
Reflecting the lack of a consensus definition, no standard metric exists for measuring disaster resilience. Indeed, one of the priorities in the National Science and Technology Council’s (NSTC’s) Grand Challenges for Disaster Reduction is to “assess disaster resilience using standard methods” (SDR, 2005; p. 2). As this report noted, such metrics are needed for several reasons: “With consistent factors and regularly updated metrics, communities will be able to maintain report cards that accurately assess the community’s level of disaster resilience. This, in turn, will support comparability among communities and provide a context for action to further reduce vulnerability. Validated models, standards, and metrics are needed for estimating cumulative losses, projecting the impact of changes in technology and policies, and monitoring the overall estimated economic loss avoidance of planned actions” (SDR, 2005; p. 2). Perhaps most importantly, standardized methods are needed to gauge improvements in resilience as a result of disaster risk reduction planning and mitigation.
Metrics of disaster resilience differ from the familiar metrics of disaster risk in several ways. Standard risk measures include expected casualties, property damage, and business interruption loss—that is, estimates of
Widespread Consequences of a Central U.S. Earthquake
An analysis of the impacts of a magnitude-7.7 earthquake on all three New Madrid faults was performed by the Mid-America Earthquake Center under the FEMA New Madrid Catastrophic Planning Initiative (Elnashai et al., 2009). Results indicated that this event would have widespread, catastrophic consequences (Figure 2.1), including:
• Nearly 715,000 buildings damaged in eight states.
• Substantial damage to critical infrastructure (essential facilities, transportation, and utility lifelines) in 140 counties: 2.6 million households without electric power; 425,000 breaks and leaks to both local and interstate pipelines; and 3,500 damaged bridges, with 15 major bridges unusable.
• 86,000 casualties for a 2:00 am scenario, with 3,500 fatalities.
• 7.2 million people displaced, with 2 million seeking temporary shelter.
• 130 hospitals damaged.
• $300 billion in direct economic losses, including buildings, transportation, and utility lifelines, but excluding business interruption costs.
Moreover, infrastructure damage would have a major impact on interstate transport crossing the Central United States.
FIGURE 2.1 Distribution of top) the nearly 86,000 total casualties, including 3,500 fatalities, and bottom) the more than 713,000 buildings damaged, in the eight-state study region from a magnitude-7.7 scenario earthquake at 2:00 am on the New Madrid faults. SOURCE: Elnashai et al. (2009); Courtesy of the Mid-America Earthquake Center, University of Illinois.
these losses in potential earthquakes weighted by the probability of such events occurring. Resilience differs from risk in three important ways. First, resilience includes performance in the post-disaster (response and recovery) timeframes, including aspects such as business interruption and the time required to recover, while risk typically focuses on immediate property damage. Second, resilience embodies some sense of goals and considerations of what risk is acceptable. Third, it also encompasses ideas of capacity-building and process, rather than being limited in scope to goals and outcomes.
Because the concept of resilience is specific to the context of the specific community and its goals, it can be expected that no single measure will be able to capture it sufficiently. Moreover, different measures will be needed for different purposes. Thus for federal agencies, a national-scale overview may be useful; a simple measure might be the percentage of these states with active seismic safety programs. For a state government, a useful marker may be the percentage of communities that are actively engaging in seismic risk reduction. For a city, however, more specific measures would be needed. An overall metric of the time required to recover “community wellness” (e.g., an aggregation of casualties, property, and economic losses) in the event of an “expected” earthquake may be one possibility. Annualized expected earthquake losses in that community may provide another alternative. Within a community, organizations such as local fire departments may have yet more specific measures in relation to seismic performance goals. Thus multi-level assessments are needed, rather than searching for a “one size fits all” metric.
Researchers and practitioners have proposed a number of approaches for measuring disaster resilience at the community level. These approaches can be broadly categorized into two types—those emphasizing resilience as a goal, and those emphasizing it as a process. A few examples are briefly reviewed here.
Bruneau et al. (2003), on which the NEHRP definition of resilience is based, treats resilience in terms of performance outcomes or goals. They propose as a measure of resilience the functional or performance loss of a system (such as a city) evaluated over the timeframe for recovery. This is illustrated schematically in Figure 2.2. The smaller the initial drop in a disaster, and the more rapid the recovery, the smaller the aggregate loss (“loss triangle”) and the higher the assessed resilience.
Within this framework, recovery is assumed to entail a return to normal (without-disaster) conditions. Thus, it is difficult to address some of the aspects of resilience discussed above, such as allowing for system change and rebuilding in ways that reduce future risk. However, the framework can be generalized to accommodate these considerations. A summary of recent progress includes:
FIGURE 2.2 Measuring resilience using the “Loss Triangle” concept. Note that the degree of “robustness” depends upon both the system’s inherent resilience and the additional effect of any pre-disaster mitigation actions. SOURCE: Modified from Bruneau et al. (2003) and McDaniels et al. (2008). Reprinted from McDaniels et al. (2008) with permission from Elsevier.
• Several researchers have proposed operational metrics (e.g., Chang and Shinozuka, 2004; Rose, 2004, 2007). The most basic of these provides a starting point for measurement as the avoided losses due to resilience actions divided by the maximum potential losses for a given event.
• An important distinction has been made between system resilience and broader concepts such as economic resilience. The latter is more encompassing because it focuses on the contribution these services make to the economy, including not just the supply but also demand (not just to the first line of customers but also to successive ones down the customer chain, e.g., Cox et al. (2011).
• Recent programs have embraced the resilience concept. The SPUR (San Francisco Planning and Urban Research Association–Resilient City Initiative) approach (SPUR, 2009) also focuses on outcomes (see Figure 2.3 and related discussion). Data for these outcomes are derived, however, from expert judgments, rather than either community consultation or a computer model.
• Broader measures of resilience emphasize the capacity, or process, dimensions of resilience. These typically characterize resilience through
describing features of more disaster-resilient communities or identifying specific actions, adaptations, or tactics both pre- and post-disaster (e.g., Tobin, 1999; Godschalk, 2003; Berke and Campanella, 2006; Cutter et al., 2008a, 2008b; Norris et al., 2008). More recently, progress has been made on developing indices of community resilience (e.g., Emmer, 2008; Cutter et al., 2010; CARRI, 2011). These prospective measures of resilience are facilitated by the use of census or other generally available data and self assessments.
These examples illustrate the range of approaches that have been applied to assess the disaster resilience of communities. As noted earlier, no one resilience indicator can suit all purposes, and different measurement approaches may be appropriate in different contexts for assessing current levels of disaster resilience and incremental progress in developing resilience.
The NSTC’s Grand Challenges for Disaster Reduction identified four key characteristics of disaster-resilient communities (SDR, 2005; p. 1):1
• Relevant hazards are recognized and understood.
• Communities at risk know when a hazard event is imminent.
• Individuals at risk are safe from hazards in their homes and places of work.
• Disaster-resilient communities experience minimum disruption to life and economy after a hazard event has passed.
Within the context of this broad vision, more specific, tangible characterizations of a more earthquake-resilient community are proposed here in order to guide prioritization of efforts. In a major disaster:
• No systematic concentration of casualties. Important or high-occupancy structures (e.g., schools, hospitals, and other major institutional buildings; high-rise commercial and residential buildings) do not collapse, and significant numbers of specific building types (e.g., hazardous unreinforced masonry structures) do not collapse. There are no major hazardous materials releases that would cause mass casualties.
1 A number of other similar characterizations have also been proposed (e.g., Godschalk, 2003; Foster, 2007). Tierney (workshop presentation) notes that resilience has multiple aims—reduced loss of life and economic impact; equity and fairness (addressing disparities in vulnerability); and sustainability (laws, processes, etc. are robust over time and support social values of quality of life, environmental quality, community safety, and livability).
• Financial loss and societal consequences are manageable, not catastrophic. Damage to the built environment is reduced to avoid catastrophic financial and societal losses due to overwhelming cost of repair, casualties, displaced populations, government interruption, loss of housing, or loss of jobs. Community character and cultural values are maintained following disasters; there is not wholesale loss of iconic buildings (including those designated as historic), groups of buildings, and neighborhoods of architectural, historic, ethnic, or other significance.
• Emergency responders are able to respond and improvise. Roads are passable, fire suppression systems are functional, hospitals and other critical facilities are functional. It is noteworthy that during the 9/11 attacks, New York City’s response was hampered by the need to set up a new Emergency Operations Center, the existing one having been located in the World Trade Center.
• Critical infrastructure services continue to be provided in the aftermath of a disaster. Energy, water, and transportation are especially critical elements. Telecommunications are also very important. Continued service is needed for critical facilities such as hospitals to function, as well as for households to remain sheltered in their homes.
• Disasters do not escalate into catastrophes. Infrastructure interdependencies have been anticipated and mitigated, so that disruptions to one critical infrastructure do not cause cascading failures in other infrastructures (e.g., levee failures in New Orleans escalated the disaster into a catastrophe). Fires are quickly contained and do not develop into major urban conflagrations that cause mass casualties and large-scale neighborhood destruction.
• Resources for recovery meet the needs of all affected community members. Resources for recovery are available in an adequate, timely, and equitable manner. To a large extent, local governments, nonprofit organizations, businesses, and residents would have already materially and financially prepared for a major disaster (e.g., are adequately insured; have undertaken resilience activities on their own and in cooperation with others). Safety nets are in place for the most vulnerable members of society.
• Communities are restored in a manner that makes them more resilient to the next event. Experience is translated into improved design, preparedness, and overall resilience. High-hazard areas are rebuilt in ways that reduce, rather than recreate, conditions of disaster vulnerability.
Each community will face unique gaps and challenges in meeting these resilience goals. The priorities and mix of strategies and actions will differ from one community to the next. Each community could translate these general goals into specific, transparent performance goals appropriate for the locality and scaled for different size disasters. These perfor-
mance goals can then provide a basis for developing consistent design standards and retrofit guidelines.
Two examples are provided below to illustrate different approaches that proactive communities have undertaken to enhance their disaster resilience. The Evansville, Indiana, example is noteworthy for the long-term, cumulative efforts of multiple stakeholder groups. Evansville focused largely on traditional pre-disaster mitigation and planning actions—that is, enhancing “robustness” as noted in Figure 2.2. In contrast, the San Francisco example is noteworthy for pioneering community discussions and prioritizing activities that focus explicitly on the “rapidity” dimension of resilience in the aftermath of an earthquake.
Example 1: The Process of Developing Resilience in Evansville
This example outlines the history of Evansville, Indiana’s Disaster Resistant Community (DRC) efforts as an example of one community’s long-term, multi-faceted approach to developing disaster resilience. After a 1987 central U.S. earthquake and the 1989 Loma Prieta earthquake, geologists and emergency response planners recognized that Evansville, Indiana, was at greater risk from earthquakes than most Indiana cities because parts of the city are built upon thick soft soils. In 1990, long before the national programs to improve resiliency of communities, Evansville started its own effort with support by the Indiana Department of Fire and Building Services (IDFBS) and the City of Evansville. Initial activities involved gathering subsurface soil property information by the Indiana Geological Survey and Ball State University. The geologic, geotechnical, and shear wave velocity data provided the basis for risk analysis for the IDFBS and Vanderburgh County Building Commission and emergency management response planning.
In 1997, the Central U.S. Earthquake Consortium (CUSEC) embarked on a pilot disaster-resistant community project involving two communities—Evansville, Indiana, and Henderson, Kentucky. To launch the pilot project, a workshop was held to bring together a multi-disciplinary group of hazards specialists, emergency managers, and community leaders to develop a model disaster-resistant community program. This workshop was cosponsored by Federal Emergency Management Administation (FEMA), Insurance Institute for Property Loss Reduction (IIPLR), and the Disaster Recovery Business Alliance along with the cooperating organizations of the American Red Cross, Risk Management Solutions, Inc., International City and County Management Association, and Evansville community leaders. Working groups developed a mitigation strategy and implementation plan that addressed the key elements of a DRC program: Education and Public Outreach, Existing Development, New Development, Com-
munity Land Use, and Business Vulnerability Reduction. A steering committee identified three key components of an Evansville Model Disaster Resistant Community Program: (1) use of the HAZUS loss estimation software as a central feature of the community’s hazard and risk assessment; (2) application to become a “Showcase Community” in a national program administered by the IIPLR; and (3) formation of an Evansville Business Alliance. The committee outlined objectives and sample activities with the recognition that becoming more disaster resistant would require a long-term, phased approach under the guidance of a partnership of local and national interests.
In applying for the Showcase Community program, the committee agreed to meet 14 criteria:
1. Adopt the latest model building code without modifications.
2. Receive the Building Code Effectiveness Grading Schedule grade and develop an improvement strategy.
3. Participate in the National Flood Insurance Program, and receive a Community Rating Service grade and develop an improvement strategy.
4. Have a minimum of 8 on the fire suppression rating system.
5. Undergo a community risk assessment conducted by the IIPLR and the partnership of local and national interests.
6. Develop and offer mitigation training to professionals (e.g., engineers, architects, building officials, contractors).
7. Conduct nonstructural retrofit assessment of all nonprofit child care centers so that the partnership can retrofit them.
8. Provide public education of natural hazards and mitigation techniques to certify homeowners to qualify them for incentives.
9. Develop K-12 school curriculum teaching about natural hazard risks and mitigation.
10. Ensure that the community has a land-use plan and a planner, and makes zoning decisions in compliance with its land-use plan.
11. Develop an emergency recovery plan and post-disaster recovery plan.
12. Develop a Disaster Recovery Business Alliance to formulate and implement a business mitigation strategy.
13. Develop public- and private-sector incentives.
14. Participate in the Partnership Seal of Approval inspection and certification.
The steering committee worked with the Institute for Business and Home Safety (IBHS) to complete the list of projects. Upon completion in 1997, Evansville was named the nation’s first Showcase Community.
In 1998, Evansville applied to FEMA to be part of Project Impact and
was chosen in the second round. When the grant from FEMA was received, a decision was made to incorporate into the Southwest Indiana Disaster Resistant Community Corporation (DRC). The nonprofit corporation has representation from five counties in southwestern Indiana. In the fall of 1997, a movement to develop an alliance of area businesses was begun by the executive director of the Metropolitan Evansville Chamber of Commerce and other regional business executives. The Southwest Indiana Disaster Recovery Business Alliance (DRBA) was to develop disaster recovery initiatives. This effort was a good fit with the DRC, and a combined office was established with a full-time director in 1999.2
The DRC efforts resulted in numerous accomplishments, a few examples of which are highlighted below to illustrate the range of partnerships involved, the types of activities undertaken, and the spillover benefits of earthquake risk reduction to multi-hazard resilience activities:
• Seismic retrofits were completed in critical and other facilities. Several fire stations were structurally and nonstructurally retrofitted. Nonstructural retrofits were completed at 36 nonprofit daycare centers using materials donated by area businesses and labor provided by volunteers from the local building commission, a youth group, insurance agencies, and the DRC. The school corporation adopted several mitigation policies and was involved in building the ECO House, the first house to be certified “disaster resistant” by the Institute for Business and Home Safety. The DRC coordinated volunteers in nonstructural mitigation of dozens of Habitat for Humanity homes. The City of Evansville’s Housing Rehab Services agreed to strap down all water heaters as part of its housing rehabilitation program provided to low-moderate income households.
• Other accomplishments served to incorporate hazard and risk considerations into urban development. The Area Plan Commission considered hazard and loss estimation information in updating the comprehensive plan. Evansville-Vanderburgh County committed to taking natural hazards into account in all its land-use decisions. A new building code amendment required new buildings to be constructed to withstand 110 mph winds. Vanderburgh County and Evansville, Indiana, received National Flood Insurance Program community ratings that provided 10%
2 The uniqueness of the Evansville DRC gained other national recognition. For example, Sandia National Laboratories partnered with the DRC in 1999 to develop a disaster management system proposal to identify vulnerability issues related to critical infrastructures. Also, the U.S. Geological Survey picked Evansville in 2003 for one of its Urban Hazard Mapping Programs. The project’s goal is to provide state-of-the-art urban seismic hazard maps reflecting the variations in materials and thicknesses that govern the amount of amplification by the soils and locations of liquefaction. The scenario earthquakes, representing reasonable maximum magnitude earthquakes for these areas, are being used to produce ground motion and liquefaction potential maps.
and 5% reductions, respectively, in National Flood Insurance premiums for local residents.
• Training sessions were conducted for professionals. These included city-county building officials, architects and engineers, and fire department personnel. The HAZUS initiative involved numerous participants. Data development involved University of Evansville students, the Indiana Geological Survey, and the Disaster Recovery Business Alliance, among others. Training workshops and a HAZUS Technical Subcommittee were formed to develop and maintain the capacity to use HAZUS for hazard and risk assessment.
• The DRC and its partners developed and disseminated disaster preparedness and mitigation information to educate the general public. Print materials included a disaster preparedness calendar and mitigation tip sheets by the Southern Indiana Gas & Electric Company and the Red Cross. Fox 7 produced a documentary of the Project Impact initiative. The DRC worked with local schools to incorporate K-12 educational programs on disaster preparedness, response, and mitigation.
• Members of the DRC organized a number of community events—including Earthquake Preparedness Week, Fire Prevention Week, Severe Weather Week, and Building Safety Week—and participated in others, such as CPR/Family Safety Day and a local hospital’s safety fair. These events provided opportunities to educate local residents on preparedness and mitigation.
Even at the end of 2009, with no funding, DRC participants continued to perform walk-through inspections in schools and businesses for preparedness, as well as make presentations to various groups and have a presence at area fairs.
The Evansville plan is admirable for its attention to major concerns of reducing the losses from earthquakes and for moving toward an all-hazards approach. However, it focuses almost entirely on pre-event mitigation, and only three of its major tenets refer to post-disaster recovery and reconstruction. The emphasis in theory and practice since the time of the development of the Evansville plan has been much more focused on post-disaster resilience as defined in this report—an emphasis on maintaining function of the economy and broader society, as well as hastening recovery. The San Francisco example described below is more in accord with the concepts of resilience described in this report, with its design of pre-disaster mitigation activities—utilizing a broad definition of “performance”—which emphasizes not just a reduction in building damage but also an emphasis on maintaining and restoring the services that buildings provide.
Example 2: Defining Resilience Goals and Measures in San Francisco
In 2006, as part of the activities surrounding the 100-year anniversary of the 1906 Great San Francisco Earthquake, the Earthquake Engineering Research Institute (EERI), Seismological Society of America (SSA), California Emergency Management Agency (CalEMA), and U.S. Geological Survey (USGS) commissioned the development of a comprehensive simulation and analysis of potential losses if a repeat of the 1906 earthquake were to happen now. The report, When the Big One Strikes Again (Kircher et al., 2006), estimated that many of Northern California’s nearly 10 million residents would be affected. It would cost $90-$120 billion to repair or replace the more than 90,000 damaged buildings and their contents, and as many as 10,000 commercial buildings would sustain major structural damage. Between 160,000 and 250,000 households would be displaced from damaged residences. Depending upon whether the earthquake occurs during the day or night, building collapses would cause 800 to 3,400 deaths, and a conflagration similar in scale to the 1906 fire is possible and could cause an immense loss. Damage to utilities and transportation systems would increase losses by an additional 5% to 15%, and economic disruption from prolonged lifeline outages and loss of functional workspace would cost several times this amount. Considering all loss components, the total price tag for a repeat of the 1906 earthquake is likely to exceed $150 billion. In such a scenario, the city of San Francisco might not be able to recover from the cascading consequences and might lose its central place in the region.
Motivated to reverse this prognosis, earthquake professionals and policy-makers in San Francisco joined forces soon after the conference and began a two-year effort to prioritize policies and actions to help ensure that San Francisco could rebound quickly from a major event. Their efforts resulted in four major policy papers, summarized in “The Resilient City,” a policy paper adopted by the Board of the San Francisco Planning and Urban Research Association in 2008 (SPUR, 2009). The panel of experts took a community-wide perspective, describing their vision of resilience as:
Resilient communities have an ability to govern after a disaster strikes. These communities adhere to building standards that allow the power, water and communications networks to begin operating again shortly after a disaster and that allow people to stay in their homes, travel to where they need to be, and resume a fairly normal living routine within weeks. They are able to return to a “new” normal within a few years … (and the disaster) does not become a catastrophe that defies recovery (SPUR, 2009; p. 1).
Key elements of this vision include:
• Establishment of performance objectives for buildings and lifeline infrastructure systems, including power, gas, water, communications, and transportation.
• Seismic retrofit of a sufficiently large number of homes so that the vast majority of city residents are able to shelter in place (i.e., remain at home) following an earthquake.
• Establishment of a Lifelines Council with influence over the preparation of critical services. This council would ensure that the utility services are restored within days of the earthquake.
• Establishment of a new voluntary rating system, designating Seismic Silver and Seismic Gold buildings, which performs so well that these standards quickly becomes a model for all new housing in the region.
• Ability of the entire city to get back on its feet in four months.
To achieve this vision, the panel established performance targets for new and existing buildings and lifelines, at different phases in the recovery process, for an “expected” earthquake (ATC, 2010). The panel chose to analyze an “expected” earthquake, rather than an “extreme” event, in order to focus on a large event that can reasonably be expected to occur during the useful life of a structure or lifeline system. It chose a scenario earthquake that was also being used by another seismic study under way in the city, with the expected earthquake being a magnitude-7.2 earthquake on the Peninsula segment of the San Andreas Fault. It also established a series of transparent performance measures, based upon usability, for both buildings and infrastructure after the expected event. For buildings, there are three categories: safe and operational, safe and usable during repairs, and safe and usable after moderate repairs. Relying on expert input, the panel assessed the current status of expected performance of buildings and infrastructure. It then set performance targets for four post-earthquake time periods—immediately, 1 to 7 days, 7 days to 2 months, and 2 to 36 months.
SPUR developed a series of near- and long-term recommendations for existing and new buildings as well as infrastructure by considering: (1) the goals for seismic resilience for each component of the city; (2) the gap between current seismic performance and the goal; and (3) the general level of cost to make the necessary improvements or retrofits. In all cases, SPUR’s performance targets require a substantial improvement in seismic performance compared to the current situation. However, SPUR did not recommend that all buildings and infrastructure be upgraded to a level that would make them “damage-proof,” as this was assessed to be cost-prohibitive. Instead, by defining an acceptable level of damage for the
expected earthquake, it focused its recommendations on those improvements considered most likely to yield a quick recovery or level of resilience desired for each phase of recovery. Recommendations were guided by the recognition that two “missing pieces” needed to be addressed in dealing with the earthquake problem—lifelines (critical infrastructure) and the workforce.
The panel emphasized pre-disaster mitigation actions in its recommendations, but some post-disaster actions would also be required to achieve these performance targets. For example, ensuring that “95% of all residences are deemed to be safe for occupancy within 36 hours after the expected earthquake” would require that enough existing structures be seismically retrofitted so that the vast majority of San Francisco residents would be able to shelter in place. It also required substantial changes to inspection procedures and post-earthquake occupancy standards, because residents would need to be allowed to remain in superficially damaged buildings even if utility services are not functioning.
Earthquakes other than the “expected” one are possible, of course, but, in smaller earthquakes, better performance is expected. In larger, more extreme events, lesser performance will have to be tolerated.
Figure 2.3 provides an example of specific resilience goals recommended by SPUR in San Francisco. The figure indicates the expected performance of buildings and infrastructure if the earthquake were to occur today (marked as X’s), the post-earthquake performance targets for each category (shaded boxes), and the gap between them. For example, critical response facilities, such as hospitals, police and fire stations, and emergency operations centers, are categorized as buildings that must be “safe and operational” immediately after the expected earthquake. Currently, these buildings are more likely to be “safe and operational” within 24 hours or, as long as 36 months, after an expected earthquake. For residential housing, buildings must be “safe and usable during repairs” and there is a target to have 95% of residents able to shelter-in-place within 24 hours after an expected earthquake. Currently, it is more likely to take up to 36 months before 95% of San Francisco’s residents would be able to re-inhabit their homes after an expected earthquake.
Other Examples of Resilience
The Evansville and San Francisco examples described above both represent concerted public programs to improve earthquake resilience. Such programs are needed because there is a lack of information and awareness of the earthquake threat, and a lack of adequate incentives to address it, when the rewards for the entity undertaking the investment in resilience involve spillover effects to other segments of society. In the
latter case of a “public good,” the entity making the expenditure cannot capture all of the broader gains, and hence an under-investment occurs from the standpoint of society. Otherwise, in a predominantly market economy like that of the United States, many individual decision-makers and public institutions do make appropriate decisions regarding resilience in response to market signals—the marketplace is an important resource for developing resilience.
Prices reflect the value of economic resources, and price increases following a disaster are often characterized as gouging. Nevertheless, some price increases are warranted and serve as indicators of the increased scarcity of specific goods and services. When markets are working effectively, these price signals need to be considered in making decisions regarding the allocation of resources. When markets are not working effectively, as when market institutions are destroyed or prone to various types of market failure (including price gouging due to asymmetric information or market power), it may be necessary for authorities to override market signals and make decisions with other approaches, such as rationing. This may be the case especially where equity, or fairness, is concerned. Free markets are known to lead to the efficient allocation of resources, but are effectively blind to equity concerns.
Individual decision-makers also capitalize on many types of resilience embodied in the economic system, referred to as “inherent” sources of resilience, including the marketplace itself (NRC, 2007; Rose, 2009). These conditions include inventories of critical materials, the ability to substitute other inputs for those in short supply (e.g., use of bottled or trucked water for piped water serves), and excess productive capacity to be accessed when facilities in use are damaged (e.g., relocating to empty office space or factories). Although many of these types of resilience are taken for granted because they are in place during the normal course of doing business, there is still potential for enhancing them. They also have an advantage in reducing losses over mitigation because they can be accessed at little or no extra cost.
Another category of resilience refers to the ingenuity, or “adaptive ability,” that often is inspired by necessity after an earthquake to keep households, business, and government organizations going (e.g., Comfort, 1999; Mileti, 1999). Examples include making organizations more efficient, finding new substitutes for critical materials, and establishing new social networks. They are also part of the nation’s resilience capability. They may not require large-scale programs as in the previous case study examples, but they do merit attention and further nurturing. Not all decision-makers are aware of these opportunities, and more generic programs, rather than region-specific ones may be the preferable vehicle. For both inherent and adaptive resilience, the dissemination of information on best practice
methods has the potential to be a valuable national project to promote resilience.
A new “business continuity industry” has arisen over the past 10 years, consisting of private-sector professionals that help businesses prepare, clean up, and recover from disasters (the majority of examples relate to information technology backup and business relocation). Such services are especially important to small business, which cannot take advantage of economies of scale or otherwise afford their own in-house hazard professionals.
Another reason for focusing on the role of the individual business or household is the importance that self-reliance can provide. It helps reduce dependence on government bailouts. Flynn (2008) has taken a profound view of this by focusing on how resilience can be “empowering” to the general citizenry.
This discussion of economic considerations and reliance practices is related to the object of resilience—what types of losses are we really trying to reduce. The focus of much of this report is on property damage. However, property damage from earthquakes and most other natural disasters takes place at a given point or short period in time. It is, rather, the flow of goods and services from the property (capital assets) that sustain people’s lives. This reduction in the flow of goods and services (often referred to as “business interruption,” or BI) starts at the point of the earthquake but continues until recovery is complete. Resilience cannot do anything to reduce the property damage after the event, but can reduce the BI by using remaining resources as effectively as possible and recovering as quickly as possible. When economists and policy-makers talk about indicators of societal well-being, they focus on flow indices such as the BI, which in the grander sense is really just a lay term for a decrease in gross national/regional product.
Also, resilience can be defined narrowly or holistically. System resilience is usually a good example of the former because it focuses on the maintenance of the service flow. Economic resilience is more encompassing because it focuses on the contribution these services make to the economy. It includes not just the supply but also demand (i.e., both the provision of a good or service and its utilization, and not just to the first line of customers but to successive ones down the customer chain). An example of this dichotomy would be transportation resilience in the aftermath of a natural disaster or terrorist attack. It could begin with consideration of resilient actions by providers of transportation services and then proceed to the resilience of its customers through the alternative modes, telecommuting, and greater reliance on existing inventories (as opposed to new shipments). In the latter case, it is not only the number of trips that is important but also the contribution they make to transportation customers’ production levels
or well-being. This way, telecommuting, would be viewed as a resilient strategy, because it maintains production (reduces BI) even with fewer trips; otherwise, its contribution might be overlooked (e.g., Cox et al., 2011).
In a similar vein, a recent study of the resilience of the New York City Metropolitan Area economy in the aftermath of 9/11 found its resilience to be very high—72% according to one if the definitions noted in the previous section, because 95% of the 1,100 firms located in the World Trade Center area were able to move to other locations, primarily in the metropolitan area (72% is lower than 95% because of the lost production caused by delays in relocation) (Rose et al., 2009). Thus, temporary locations, often becoming permanent, saved more than $40 billion of gross regional product. To use all of society’s resources effectively, such flexibility to use excess building stock (if available) before reconstruction could take place needs to be factored into programs such as the San Francisco example.
Resilience and Post-Earthquake Recovery
The Evansville and SPUR examples described above focused on aspects of the built environment and on advance planning for recovery, but they do not illustrate actions that can be taken after the event to promote resilience in terms of maintaining function of the broad set of societal attributes and hastening recovery. Table 2.1 provides examples of resilient actions at various stages of recovery and reconstruction in relation to a broader set of societal attributes and indicators. The details of the table provide only some of many examples of such actions. We illustrate their usefulness and importance with respect to the last column “Economic Resilience.”
• Immediate (< 72 hours)—It is important to maintain a supply of critical goods and services such as water, power, and food to support the economy and social system.
• Emergency (3-7 days)—It is necessary for businesses, households, government, and nongovernment organizations to prioritize the use of resources, such as by the use of rationing. In many instances it is important to find substitutes for key inputs and to conserve them as well.
• Very Short-run (7-30 days)—The marketplace is an important inherent resource in addressing resilience. Prices reflect value and act as indicators of the scarcity of goods and services. When markets are working effectively, these price signals need to be considered in making decisions regarding the allocation of resources. When markets are not working effectively, as when market institutions are destroyed or prone to various types of market failure (including price gouging), it may be necessary for authorities to override market signals and make decisions with other
TABLE 2.1 Resilience Applications to Social, Ecological, Physical, and Economic Recovery by Time Period
|Timescale||Emergency Response||Health & Safety||Utilities||Buildings||Environmental/ Ecological||Economic|
|Immediate < 72 hours||Tactical emergency response||Deal with casualties/ Reunite families||Use of emergency backup systems||Remove debris||Limit further ecological damage||Maintain supply of critical goods and services|
|Emergency 3-7 days||Strategic emergency response||Provide mass care||Begin service restoration||Provide shelter for homeless||Remove debris||Prioritize use of resources/ substitute inputs/ conserve|
|Very short 7-30 days||Selective response||Fight infectious outbreaks||Continue restoration||Provide shelter for homeless||Protect sensitive ecosystems||Shore up or override markets|
|Short 1-6 months||Assist in recovery||Deal with post-traumatic stress||Complete service restoration||Provide temporary housing and business sites||Deal with ensuing problems||Cope with small business strain|
|Medium 6 months–1 year||Reassess for future emergencies||Deal with post-traumatic stress||Reassess for future emergencies||Provide temporary housing and business sites||Initiate remediation||Cope with large business strain/ recapture lost production|
|Long >1 year||N/A||Reassess for future emergencies||Mitigation for future events||Rebuild and mitigate||Mitigate for future events||Cope with business failures/mitigation|
approaches. This may be the case especially where equity, or fairness, is concerned. Free markets are known to lead to the efficient allocation of resources but are effectively blind to equity concerns.
• Short-run (1-6 months)—Small businesses are especially vulnerable in the immediate aftermath of a major disaster, and require special attention.
• Medium-run (6 months-1 year)—One of the major sources of resilience is the ability to recapture lost revenue after the event; many businesses have standing orders for their product production, and these can be filled by working overtime or extra shifts at the relatively low cost of overtime pay.
• Long-run (> l year)—It is important that mitigation be integrated into the reconstruction effort to reduce losses from future events.
Many of the points of this chapter can be reiterated by summarizing the many dimensions of resilience:
1. Multi-scale dimension. The concept of resilience is applicable at multiple scales, from the resilience of an individual person (e.g., psychological, financial) to that of an organization, neighborhood, city, or nation.
2. Multi-hazard dimension. Resilience pertains to all hazards and not just earthquakes. Moreover, resilience to other hazards can in many cases be applied to earthquakes.
3. Stock (property damage) and flow (production of goods and services) dimensions of assets, systems, economies, and communities. Property damage takes place at a given point in time, but the service flows (to which maintaining function applies) are disrupted until recovery is completed, and are thus more central to the idea of rebounding after a disaster.
4. Behavioral and policy dimensions. The length of the recovery following disasters is not some constant that can be known beforehand, but an outcome that depends critically on decisions and activities undertaken by private- and public-sector decision-makers.
5. Geophysical dimension. Resilience generally varies inversely to the size of the shock to the system.
6. Bifurcation of temporal dimensions. Static resilience refers to the ability of an entity or system to maintain function when shocked and relates to how to efficiently allocate the resources remaining after the disaster. Dynamic resilience refers to the speed at which an entity or system recovers from a shock and is a relatively more complex problem because it involves a long-term investment associated with repair and reconstruction.
7. Contextual dimension. The level of function of the system at a point in time has to be compared to the level that would have existed had the ability been absent, requiring that a reference point or worst-case outcome be established first.
8. Capacity dimension. Inherent resilience refers to the ordinary ability already in place to deal with crises. Adaptive resilience refers to ability in crisis situations to maintain function on the basis of ingenuity or extra effort.
9. Market dimension. This refers to the need to consider both the providers and customers of building and infrastructure services in moving toward a holistic definition of resilience.
10. Cost dimension. Resilience essentially represents a measure of benefits of various actions. However, the cost side cannot be neglected in policy decisions.
11. Process dimension. Resilience is not just about actions and targets, but the manner in which these are achieved is a critical aspect. This refers to developing and applying a set of adaptive capacities.
12. Fairness dimension. Resilience should be applied in an equitable manner, to be sensitive to the needs of the most disadvantaged groups in society with care being taken to try to avoid having any group adversely affected by its implementation.