Background Paper on Disaster Resilient Cities

PDF Version (Size: 434 KB)

Help on accessing alternative formats, such as PDF, PPT and ZIP files, can be obtained in the alternate format help section.

Report prepared for Infrastructure Canada by:

Dan Henstra, Ph.D. candidate, Department of Political Science, University of Western Ontario

Paul Kovacs, M.A., Executive Director, Institute for Catastrophic Loss Reduction and Adjunct Research Professor, Department of Economics, University of Western Ontario

Dr. Gordon McBean, Professor, Departments of Geography and Political Science and Research Chair in Policy, Institute for Catastrophic Loss Reduction, University of Western Ontario

Rob Sweeting, Manager of Research, Institute for Catastrophic Loss Reduction

---

Table of Contents

• Abstract
• Context
• Dealing with Disasters
  • Hazards
  • Vulnerability
• The Disaster Resilient Community: A Conceptual Analysis
  • Disaster Resistance
  • Resilience and Sustainability
  • Invulnerable Development
  • Comprehensive Vulnerability Management
  • The Resilient City
  • Proactive / Anticipatory Adaptation
• The Disaster Resilient Community: Core Concepts
• Infrastructure
• Disasters and Infrastructure: Threats and Vulnerabilities
  • Power Systems
  • Communications and Information
  • Transportation
  • Water and Wastewater
• Resilience and Infrastructure: Countermeasures
  • Robustness
  • Redundancy
  • Adaptation
• Future Research: What Do We Need to Know?
  • Disaster Resilience
  • Disaster Resilience and Infrastructure
• Closing Remarks
• Sources
• Acknowledgements

---

Abstract

Canada is a highly urbanized country - roughly 80 per cent of Canadians live in urban areas. Canadian cities play an important social and economic role as centres of commerce, trade, transportation and communication. At the same time, they face a wide range of natural, technological and human-induced hazards which threaten the safety of residents and the continuity of urban systems. Past and recent disaster impacts illustrate the need to address the risks that environmental hazards pose to cities. In light of the uncertainty we face regarding hazards, it is suggested herein that we should focus our efforts on reducing vulnerabilities, and to enhance the capacity of Canadian cities to adapt to and recover quickly from hazard impacts.

The paper establishes a research context for strategies to build disaster resilient cities, by examining the interplay between hazards and vulnerabilities, developing a conceptual model of a disaster resilient city and placing a specific emphasis on the resilience of infrastructure. Recommendations for future research are outlined.

Context

Throughout the 20th century, the world's cities have grown in size and importance. It is estimated that about half of all people live in urban areas, a figure that is expected to rise to at least 60 per cent by 2030 (UNHSP, 2002). In Canada, roughly 80 per cent of the population lives in urban areas, one of the highest rates of urbanization in the world today. The Canadian economy is inextricably linked to the health of its cities; for example, the seven largest cities in Canada generate almost 45 per cent of the national GDP and large cities like Vancouver and Montreal account for over half of provincial GDP (Bradford, 2003).

Ironically, while cities are of vital importance as economic engines, hubs for transportation and communications, cultural centres and homes to the majority of the earth's people, they are particularly vulnerable to hazard impacts. For example, high population density in urban areas means more people at risk, large concrete expanses absorb summer sun and exasperate heat waves, while the use of sealed asphalt for city streets prevents ground absorption, contributing to a risk of flooding when drainage systems become overwhelmed (MunichRe, 1999). City functions are increasingly dependant on the continuity of urban infrastructures, but in many cases these systems are old and deteriorated and their minimum design standards are no longer appropriate for our climate.

Environmental hazards (natural, technological and human-induced) pose a significant threat to cities around the world, carrying the potential to disrupt economic and social activities, cause substantial damage to property and even kill people. Moreover, it is predicted that the frequency and intensity of weather and weather-related hazards will increase as the global climate changes (McBean, 2003). During the 1990s, losses from natural disasters were more than four times greater than during the 1950s, with more than 500,000 fatalities and over C$1 trillion in damages (Walter, 2003). Most disaster damage has occurred in large urban centres, where losses have been doubling every five to seven years since the 1950s.

Every year, disasters pose a serious threat to Canadian cities, disrupting economic and social activities, taxing resources, causing substantial damage to property and endangering the lives of residents. Some more notable disasters in recent history include flooding in the Saguenay Region of Québec in 1996, which resulted in losses of over $1.5 billion; the 1997 Red River flood, which caused nearly $1 billion in damages; and the ice storm which struck Ontario, Québec and New Brunswick in 1998, which was Canada's most costly natural disaster, with losses exceeding $5 billion (OCIPEP, 2003). In 2003, wildfires were responsible for incredible losses in British Columbia, Alberta and Ontario; Toronto was greatly impacted by an infectious disease (SARS) outbreak, and Hurricane Juan wreaked havoc in communities along the East Coast. These losses, while significant, pale in comparison to the predicted impact of an earthquake which will one day strike Vancouver or Montreal, causing thousands of fatalities and more than C$30 billion in damages.

In light of the diversity of hazards that Canadians face, the many points of urban vulnerability that contribute to losses and in order to reduce the impact of disasters in the future, we must investigate new approaches to the design and operation of Canada's cities. The purpose of this paper is to establish a research context for strategies to build disaster resilient cities. It will discuss the concept of resilience in general, develop a conceptual model of a disaster resilient city, analyze the role of infrastructure (particularly critical infrastructure) and identify areas where further research is required.

Dealing with Disasters

In Canada, the United States and many other countries, the tremendous cost of disasters has prompted interest in finding ways to counteract their destructive potential. Disaster-related research conducted over the last several decades has provided the basis for a better understanding of these events. Where we once thought of disasters as "Acts of God" - random, unfortunate calamities beyond our control - we have come to understand that they actually stem from interaction between two variables: hazards (i.e., triggering agents stemming from nature, as well as from human activity) and vulnerability (i.e., susceptibility to injury or loss influenced by physical, social, economic and cultural factors) (e.g., McEntire, 2001).

From this perspective, disasters are not inevitable; if comprehensive efforts are made before a hazard event, disasters can be prevented or their impacts significantly reduced. While the hazards themselves may not be preventable, our decisions and actions can significantly minimize vulnerability and enhance our ability to recover quickly from disaster impacts. This is the primary impetus for mitigation - actions specifically designed and undertaken to reduce the impacts of hazards on people and property.

Hazards

A hazard can be defined as "any potential threat to something that people value, including one's life, health, environment or lifestyle (Mills et al, 2001)." Canadian cities must contend with a wide variety of natural hazards, including weather hazards (tornadoes, hailstorms, winter storms, heat waves), weather-related hazards (drought, wildfires, storm surges, floods) and geophysical hazards (earthquakes, landslides). The incidence of weather and weather-related hazards varies across Canada: for example, tornadoes and hailstorms are most common on the Prairies and in southern Ontario (Etkin et al, 2002; Etkin and Brun, 2001), storm surges are common on the East Coast (OCIPEP, 2001), and winter storms can occur across the country. While the number of geophysical hazards has remained approximately constant, the number of weather-related hazards has increased from 2-4 per year in past decades to about 12 per year in the last decade (with considerable year to year variability) (OCIPEP, 2001).

Cities are also affected by technological hazards, such as transportation accidents, chemical spills and fires. While these events rarely escalate to catastrophic proportions, they occur frequently and have a large compounded effect on cities and urban systems. Examples from 2003 include a freight train derailment in February, which sparked a propane fire and prompted the evacuation of 300 people in Melrose, Ontario; and the massive August blackout, which left 50 million people in Canada and the United States without power (U.S.-Canada Power System Outage Task Force, 2003).

A major challenge in dealing with hazards is how to make decisions under conditions of considerable uncertainty. The problem is particularly acute in the context of natural hazards, where uncertainty stems from varying degrees of predictability in the evolution of nature. Hilborn (1987) describes three types of uncertainty in natural systems:

• Noise - year-to-year variability makes it impossible to accurately predict the timing and magnitude of a particular event. However, if these events occur often enough, we can effectively manage the hazard, using statistical analysis to design coping methods. For weather-related events, our predictive skill is improving; we are now able to predict the probability of anomalous events up to a year in advance, especially those related to long­term variability of the ocean-atmosphere system, such as El Niño.
• Uncertain states of nature - this form of uncertainty stems from the general unpredictability of nature, where changes in natural systems occur which are unexpected and for which there is little previous experience on which to base a response. Difficulty in responding to changes in weather patterns as a result of climate change is an example of this. Climate change models project changes in statistics of extreme and other events in response to changing atmospheric concentrations of greenhouse gases. Uncertainties in projecting future human-caused emissions are then added to the uncertainty in the mode projections. The key to managing uncertain states of nature is flexibility. A strategy that works well for natural resource managers is adaptive management, which accommodates uncertainty by allowing for learning and policy adjustment as information develops.
• Surprise - this refers to unanticipated and unprecedented events for which our previous experience is insufficient to adequately respond. In the context of disaster management, surprise events would be those that occur with little or no warning and would have truly catastrophic outcomes. While specific planning to deal with these events may be impossible, there are steps that we can take to enhance our ability to adapt to and recover from them.

How can we reduce the impact of disasters if we cannot accurately know the probability, magnitude, intensity or duration of future hazard events? Ideally, we could predict the occurrence of natural hazards in order to take action before a community is impacted, but we are limited in this predictive capacity. There are two types of predictions: deterministic predictions - predicted occurrence of a particular hazard, with specific characteristics and within a specific, short time interval; and statistical predictions - probabilistic predictions of the likely occurrence of events within a particular time period, such as a season (McBean, 2000; Sarewitz et al, 2000). Our skill in predicting natural hazards such as volcanoes and earthquakes is exceedingly limited in terms of deterministic predictions, but is stronger for statistical predictions (i.e., we can predict that they will happen, but not when they will occur). For weather-related hazards, our skill in developing deterministic predictions is good for a few days and decreases to zero by about two weeks. Relatively strong probabilistic predictions can be made for seasons and in response to a changing climate (which will change the probability of extreme events).

Because we do not know the full range of possible hazard outcomes, or we cannot accurately define the probability of outcomes because of a lack of data, only a limited range can be hypothesized, limiting the scope of action we can undertake to address the hazard variable.

Vulnerability

Apart from, and in addition to, the hazard variable, the vulnerability of a community is a key factor in determining the impacts of a disaster. A comprehensive definition offered by the United Nations International Strategy for Disaster Reduction (UNISDR) captures many of the different elements of vulnerability that have been discussed in disaster literature:

Vulnerability to disasters is a function of human action and behaviour. It describes the degree to which a socio-economic system or physical assets are either susceptible or resilient to the impact of natural hazards. It is determined by a combination of several factors, including awareness of hazards, the condition of human settlements and infrastructure, public policy and administration, the wealth of a given society and organized abilities in all fields of disaster and risk management. The specific dimensions of social, economic and political vulnerability are also related to inequalities, often related to gender relations, economic patterns, and ethnical or racial divisions. It is also largely dependent on development practices that do not take into account the susceptibility to natural hazards (UNISDR, 2002).

There are several important points that should be highlighted in this definition, namely that vulnerability is influenced by a wide range of social, economic, political and cultural factors, that it varies among different groups within a community and that it can be reduced through human decisions and actions. In fact, in many cases vulnerability and resilience are seen as inversely correlated; that is, as resilience is enhanced, vulnerability is reduced. For example, Timmerman (1981) describes vulnerability as a measure of the adverse effects of a hazardous event on a system, the magnitude of which is influenced by the system's resilience.

Synthesizing various perspectives on vulnerability, McEntire (2001) makes this link more explicit, by incorporating resilience as one of four variables that determine vulnerability, including:

• risk - proximity or exposure to hazards, which affects the probability of adverse impact;
• susceptibility - proneness of individuals to adverse impacts of disasters, based on social, economic, political and cultural variables;
• resistance - the ability of community systems (e.g., buildings and infrastructure) to withstand the stress exerted by triggering agents;
• resilience - the coping capacity and ability to recover quickly from adverse impacts of disasters.

The Disaster Resilient Community: A Conceptual Analysis

The word 'resilience' is used in a variety of contexts and has been the subject of significant definitional debate since the 1970s (Klein et al, 2002). In light of this, the following section explores some of the contexts within which the term has been used and discusses some of the various models that have been proposed for disaster resilient communities.

The Merriam-Webster's Dictionary provides a definition of resilience as "an ability to recover from or adjust easily to misfortune or change." Writing in the context of climate change adaptation, Timmerman (1981) refers to resilience as the ability of systems to resist and recover from hazards induced by a changing climate. More specifically related to disaster management, the United Nations International Strategy for Disaster Reduction defines resilience as "the capacity of a system, community or society to resist or to change in order that it may obtain an acceptable level in functioning and structure (UNISDR, 2002b)." These definitions are all consistent with the oft-cited phrase "bend without breaking", referring to the ability of a system or object to accommodate or adapt to environmental stresses.

Others refer to resilience in a broader sense. Mileti (1999) relates the concept with community sustainability, defining a disaster resilient community as one that is "able to withstand an extreme natural event without suffering devastating losses, damage, diminished productivity, or quality of life and without a large amount of assistance from outside the community." Burby et al (2000) discuss resilience in the context of efforts to reduce losses from hazard impacts, suggesting a parallel with mitigation.

Resilience has also been connected with recovery capacity; for example, Pimm (1984) suggests it can be measured as the speed at which a system returns to its original state following an interruption, while Emergency Management Australia (1998) calls it "a measure of how quickly a system recovers from failures."

While we hope to engage in a wider workshop discussion in order to fully develop these concepts in the Canadian context, for the purposes of this paper we will define disaster resilience as the capacity to adapt to stress from hazards and the ability to recover quickly from their impacts.

Disaster Resistance

Geis (2000) argues that traditional approaches to hazard mitigation (structural and non­structural) are no longer sufficient to counter the rising toll of disasters and he suggests that planning, design and construction practices must be changed in order to make communities resistant to natural hazards. Geis chooses the term resistance over resilience, because the former suggests that a community can be built to resist or withstand a hazard in order to prevent it from reaching disastrous proportions, an idea which he argues is more marketable than the latter, which refers to the ability of a community to adapt to or recover quickly from a disaster and thus "insinuates that one has already occurred (p. 152)."

Geis posits that building disaster resistance is part of a wider goal of achieving a sustainable "quality-of-life community", meaning "the most human/socially, environmentally, and economically viable community possible, one that first optimizes the safety, health, and general well-being of the community and its residents (p. 154)." He offers a conceptual model of a disaster resistant community based on ten founding principles, ten guidelines for planning and development and twelve design guidelines for infrastructure systems and facilities.

Despite his use of the term "resistance", Geis' model does not prescribe exclusively structural approaches and the principles he outlines would also enhance community resilience. However, as McEntire et al (2002) point out, there are a number of limitations that detract from the applicability of his model. First, the concept is designed largely to address natural hazards and makes no reference to technological or human-induced triggering agents that can also pose a threat to communities. Secondly, while the model provides a strong basis for engineering and planning approaches, it does not sufficiently account for social, political and cultural variables that must be considered in assessing vulnerability (McEntire et al, 2002).

Resilience and Sustainability

The goal of building resilient communities shares much with the principles of intergenerational equity espoused under the rubric of sustainable development, commonly defined as "development which meets the needs of the present without compromising the ability of future generations to meet their own needs (WCED, 1987)." In many ways, the decisions we make regarding the siting, design and construction of communities will affect their sustainability over the long term. Moreover, decisions made today may augment vulnerability in the future, creating problems for future generations.

Tobin (1999) focuses on the role of sustainability in hazard mitigation and suggests that connections among a wide range of variables must be considered in the design of local community resilience. He defines sustainable, resilient communities as "societies which are structurally organized to minimize the effects of disasters, and, at the same time, have the ability to recover quickly by restoring the socio-economic vitality of the community (p. 13)."

Sustainability was also the focus of the Second National Assessment of Natural and Related Technological Hazards in the United States, which suggested that all policies and programs related to hazards and sustainability should be integrated into a holistic framework of sustainable hazard mitigation, wherein "actions to reduce losses would be taken only when they are consistent with the other principles of sustainability (Mileti, 1999: 30-31)." In this model, disaster resilience is seen as one element among a broad set of variables that define a sustainable community, including environmental quality, quality of life, disaster resiliency, economic vitality, inter- and intragenerational equity and a participatory process (Mileti, 1999: 31).

Invulnerable Development

While other models advocate changes to existing development practices in order to reduce current and future vulnerability, McEntire (2000) proposes a more proactive approach called "invulnerable development", which involves "decisions and activities that are intentionally designed and implemented to reduce risk and susceptibility, and also raise resistance and resilience to disaster (p. 58)." This concept "attempts to decrease the quantity (or frequency) and quality (or severity) of emergencies and disasters through liability reduction and building capacity (McEntire et al, 2002)."

The invulnerable development concept is meant to be a holistic approach, drawing on the strengths of a wide range of community actors to foster disaster resilience. The primary focus of the model is vulnerability, because, while humans have little control over natural hazards, they can manage their level of exposure or susceptibility to hazards. The concept strongly emphasizes the need to change cultural perceptions regarding hazards and disasters, identifies specific links between development practices and vulnerability, accounts for a wider range of triggering agents and stresses the importance of strengthening all components of emergency management (McEntire, 2001).

Comprehensive Vulnerability Management

Recognizing the potential for definitional confusion over the terms "invulnerable" and "development", McEntire et al (2002) propose a paradigm called "comprehensive vulnerability management", which they define as "holistic and integrated activities directed toward the reduction of emergencies and disasters by diminishing risk and susceptibility and building resistance and resilience." Comprehensive vulnerability management can be seen as a sort of meta-paradigm, drawing on the strengths of the other concepts discussed above and addressing their perceived shortfalls. Specific elements include:

• an inclusive, holistic approach - policies for comprehensive vulnerability management would be based on an consideration for risks and vulnerability in the physical, social and organizational environments;
• a primary focus on vulnerability - the approach would involve a "concerted effort to identify and reduce all types of disaster vulnerabilities";
• an all-hazards approach - the concept recognizes the need to address all types of triggering agents, natural or otherwise;
• incorporation of comprehensive emergency management - the paradigm incorporates and steps beyond the four elements of comprehensive emergency management - mitigation, preparedness, response and recovery;
• participation by a wide range of actors - the concept requires participation of and collaboration among a diverse set of actors, including public sector organizations, citizens, businesses and nonprofit organizations.

Among the various models discussed herein, comprehensive vulnerability management seems to be the most wide-ranging and holistic. By incorporating the many positive elements of other models and strongly emphasizing the need to reduce all forms of vulnerability, the paradigm provides a framework for developing proactive, tangible strategies to create disaster resilient communities.

The Resilient City

Godschalk (2003) argues that, since general hazard mitigation guidelines do not sufficiently accommodate the particular vulnerabilities of "cities under stress", a special emphasis must be placed on "urban hazard mitigation." The specific goal of urban hazard mitigation would be to develop resilient cities, which he suggests would be "capable of withstanding severe shock without either immediate chaos or permanent harm. Designed in advance to anticipate, weather, and recover from the impacts of natural or terrorist hazards, resilient cities would be built on principles derived from past experience with disasters in urban areas. While they might bend from hazard forces, they would not break."

Godschalk draws on other authors to identify a series of characteristics of resilient systems that can be applied to physical and social systems to create disaster-resilient cities, including:

• redundancy - systems designed with multiple nodes to ensure that failure of one component does not cause the entire system to fail
• diversity - multiple components or nodes versus a central node, to protect against a site­specific threat
• efficiency - positive ratio of energy supplied to energy delivered by a dynamic system
• autonomy - capability to operate independent of outside control
• strength - power to resist a hazard force or attack
• interdependence - integrated system components to support eachother
• adaptability - capacity to learn from experience and the flexibility to change
• collaboration - multiple opportunities and incentives for broad stakeholder participation

Godschalk's model emphasizes resilience as a way to cope with uncertainty. Because we can rarely predict the frequency and magnitude of hazard agents, and because the vulnerability of community systems cannot be fully known before a hazard event, cities must be designed with the strength to resist hazards, the flexibility to accommodate extremes without failure and the robustness to rebound quickly from disaster impacts.

Proactive / Anticipatory Adaptation

Conceived primarily as a model for adaptation of coastal cities to climate change, proactive or anticipatory adaptation aims to build resilience and reduce vulnerability by minimizing risk and maximizing adaptive capacity (Klein et al, 2002). In this model, there are five main objectives:

• increase robustness - improving the ability of systems to withstand stresses or increasing its tolerance for loss or failure (e.g., through insurance);
• increase flexibility - change operation or location of vulnerable facilities;
• enhance adaptability of natural systems - reducing stresses on natural systems or removing barriers to natural processes (e.g., water flow);
• reverse and prevent vulnerability - implement regulations for new development and strengthen or relocate existing development in hazardous areas;
• improve societal awareness and preparedness - inform public of risks and establish early warning systems (adapted from Klein and Tol, 1997; Klein, 2002).

As with other models, the proactive/anticipatory adaptation approach seeks to reduce existing vulnerability and mitigate future susceptibility to hazards. It differs slightly from the others in that it suggests we may need to remove barriers to natural processes that have been erected by humans.

The Disaster Resilient Community: Core Concepts

Humans are naturally resilient beings, capable of absorbing environmental stresses (stemming from nature and other humans) and persisting - the ability of people to respond to and recover from a disaster is a testament to this. However, because of the way we have built cities - artificial communities in many ways disconnected from our natural environment - we have augmented human vulnerability and increased our susceptibility to negative impacts of natural and human­induced hazards.

Fortunately, the fact that disasters are the outcome of interplay between hazards and vulnerability suggests that the way we design and build our communities can have a significant influence on the outcome of unanticipated hazard events. As illustrated above, there has been recent interest in the idea of making our communities more resilient to disasters and there have been several proposed models for a disaster resilient community. While there are many differences between the various models, this concept has sparked some interesting debate, from which we can extract the following set of core elements that could be used to design and develop a disaster resilient community.

Cultural attitudes must accommodate resilience

Adopting a strategy of resilience requires a fundamental change in our cultural perceptions regarding the management of risk stemming from hazards, natural or otherwise. It is increasingly clear that relying on structural approaches to control or contain hazards is ineffective and that these efforts are, at best, temporary solutions. Moreover, there is a great deal of uncertainty regarding the frequency, timing, duration and magnitude of hazards, as well as what we can expect from the natural environment in the future, given the evidence that our climate is changing. In light of this, we may have to recognize and accept that we cannot control many aspects of the hazard variable.

Instead, our efforts to build disaster resilience should focus on elements we can control, such as mitigating our vulnerability and susceptibility to hazard stresses, reducing our potential for disaster losses and planning for swift recovery following a hazard event. In this way, resilience accommodates uncertainty, permitting us to create more sustainable communities without having to "know what we do not know".

Disaster resilience is a philosophy, a process and a condition

Because the hazard environment is dynamic and because decisions are made regularly that affect vulnerability, disaster resilience must be seen as an ongoing process and not just an ideal condition that can be achieved and forgotten. For this reason, a holistic approach is required, to incorporate input from a diverse set of community actors in order to develop a workable and lasting strategy that can be integrated into long-term community planning and implemented with minimal resistance. In addition, a priority should be placed on building local adaptive capacity, to facilitate learning from hazard events and encourage responsive policy-making in the aftermath of a disaster.

Resilience requires an all-hazards approach

One of the key elements of hazard mitigation that must be incorporated into the disaster resilient community concept is an all-hazards approach, which addresses natural hazards as well as technological hazards (e.g., train derailment, industrial accidents) and human-induced threats (e.g., terrorism). In order to facilitate this, an important first step is to identify potential hazards in the community and assess the level of risk associated with them.

Resilience requires an all-vulnerabilities approach

In order to reduce the potential for loss from hazards, points of community vulnerability must be identified and addressed. As discussed above, vulnerability takes many forms, including physical vulnerability (e.g., susceptibility of systems or structures to failure from external stresses), social vulnerability (e.g., stemming from poor gender relations or racial and ethnic tension), economic vulnerability (e.g., availability of financial resources for preparedness efforts, mitigation or recovery) and cultural vulnerability (e.g., inaccurate risk perception, beliefs regarding the relationship between humans and nature). Moreover, while it is important to take steps to reduce existing vulnerability, it is equally important that current decisions, policies and practices do not augment future vulnerability (e.g., land use planning, siting of hazardous facilities). Reduction of vulnerability is a theme that is consistent across the various models and it is clear that vulnerability - in all its forms - must be reduced in order to facilitate disaster resilience.

Communities require greater resistance to hazard stresses

While the word "resistance" often triggers debate, it is clear from the various models that a disaster resilient community must incorporate a greater ability to resist or withstand stress imposed by hazards. To increase the overall resistance of a community, existing buildings and infrastructures may need to be "hardened" to withstand hazard forces, while regulations may need to be imposed on new construction to reduce the probability of failure in the face of future events. The "resistance" element is one area where further research is clearly required.

Community systems must be flexible

Disaster resilient communities require flexibility, to be able to absorb hazard stresses without failure. Flexibility of systems can be enhanced by designing for uncertainty (e.g., greater capacity) and incorporating redundancy and diversity (e.g., multiple nodes to ensure system continuity) to reduce susceptibility to site-specific threats. Flexibility can also be applied in a metaphorical sense to policy-making; for example, land use regulations and building codes should be flexible enough to allow for adjustment and adaptation based on disaster experiences.

Recovery capacity must be enhanced

Because it is impossible to anticipate or plan for every hazard scenario, an essential component of a disaster resilient community is its ability to recover quickly following a disaster. The recovery process is very complex and occurs at many different levels (e.g., recovery from individual psychosocial impacts, recovery of local businesses). The capacity of a community to recover quickly from a disaster is influenced by many variables, including the individual recovery capacity of households and businesses, autonomy in decision-making, financial resources, intergovernmental relations, community participation, integration in regional and national networks and many other factors (Mileti, 1999).

In order to be sustainable, disaster recovery should include rebuilding to reduce future potential losses, achieved through stronger tolerances, limiting redevelopment and adapting land use policies that recognize natural hazards as recurrent cycles. Planning for recovery can foster flexibility in post-disaster decision-making and can help to minimize disagreements over policy objectives (e.g. quickly return to normal or build safer) that are common following a disaster.

Communities must develop an adaptive capacity

While it is important to plan for anticipated risks, there will always be uncertainty regarding hazards in our environment. We can counteract this uncertainty by developing our adaptive capacity and the flexibility to cope with unanticipated events. Unlike traditional emergency management, the disaster resilient community concept incorporates an adaptation element; rather than simply returning to a state of normalcy following a disaster, we must adapt policies and practices based on lessons learned during the event. People are often more supportive of mitigation policies in the post-disaster period, providing an opportunity to implement stronger disaster management policies. Disasters also expose community vulnerabilities, which can be noted and addressed in anticipation of a future hazard event.

Infrastructure

Over many decades, we have developed complex urban infrastructures - physical facilities and systems that support human activity and provide services that improve our quality of life. The term 'infrastructure' refers collectively to transportation and communication networks, water and wastewater systems, energy and utilities and public institutions such as schools, hospitals, public housing and criminal justice facilities. Together, these vital systems are required for our economy, government and society to function.

Infrastructure is one of many instruments used by government to advance collective social, environmental, and economic objectives. Infrastructure planning, innovation, maintenance and protection are important to all industry sectors. The importance of infrastructure for the long­term economic growth and quality of life of Canadians has been reflected in the strong role governments have played in infrastructure development; the construction of railways, canals, ports, highways, seaways and airports has been central to nation-building. Many view the provision, maintenance, and protection of public (and often, private) infrastructure as a major responsibility of government.

While all urban infrastructure systems can be considered important, special attention is often given to critical infrastructures; that is, systems that are "critical to the health, safety, security and economic well-being of Canadians and to the effective functioning of governments", including "physical structures (bridges, canals and pipelines), as well as information technology­based networks and services (in the financial, telecommunications and energy sectors, for example) (PSEPC, 2003)." The services provided by these systems and our dependence on their continuity demand that they be able to absorb stresses and recover quickly from interruption.

Unfortunately, Canadian public investment in infrastructure has declined significantly since the 1960s. Measured as a proportion of Gross Domestic Product (GDP), public investment peaked at almost 5.0% in 1966 and fell to 2.6% by 2002 (the latest year for which data are available). Mirza and Haider (2003) assert that infrastructure maintenance and rehabilitation have been deferred considerably in the past few decades in Canada. This has had significant consequences, including "the failure and closing down of some facilities, such as roads, bridges, and water supply and sewage disposal lines (p. 3)." Recently, Infrastructure Canada reviewed 11 Canadian publications dating from 1996 to 2003 that were relevant to the collection and analysis of data on the condition and needs of infrastructure. While the reports are not consistent in terms of definitions, terminology, timeframes, scope and methodologies (e.g., life expectancy of the assets, and actual and future needs), they do provide an insight into the estimated size of the municipal infrastructure gap. In this regard, estimates ranged from $44 billion (total municipal infrastructure shortfall) to more than $125 billion (total municipal deficit including costs to upgrade existing infrastructure).

Disasters and Infrastructure: Threats and Vulnerabilities

Despite our high dependence on the services they provide, urban infrastructure systems are vulnerable to a wide range of hazards, stemming from nature (earthquakes, floods, winter storms), technological error (faulty design, component failure) and human activity (operator error, sabotage).

The impact of a particular hazard may be indirect, exasperated by weaknesses in infrastructure systems. For example, in the event of an earthquake, few properties are destroyed by shaking but many are destroyed by fire (e.g., earthquake ruptures gas line - leaking gas causes fire - fire destroys property). This example illustrates how an interdependent system of linked relationships connects a hazard event with its ultimate outcome (Little, 2002). When analyzed separately, the impacts of one disrupted infrastructure system on service delivery can be fairly accurately estimated; however, interdependence introduces an added layer of uncertainty, in that it is difficult to foresee the interdependent effects that one system's failure may induce. Little (2002) examines the nature of interdependence and distinguishes between three classes of infrastructure systems failure, including cascading failure, where disruption of one infrastructure causes disruption of another; escalating failure, where a disrupted infrastructure prohibits the recovery of another, dependent infrastructure which failed earlier; and common cause failure, where two or more infrastructure systems fail as a result of a common cause, such as a natural disaster (p. 3).

The Office of Critical Infrastructure Protection and Emergency Preparedness (OCIPEP) - now part of Public Safety and Emergency Preparedness Canada (PSEPC) - has funded studies on various components of critical infrastructure, such as communications and transportation. PSEPC recently announced a partnership with the Natural Sciences and Engineering Research Council (NSERC) to form the Joint Infrastructure Interdependencies Research Program (JIIRP), a program which will provide funding to study infrastructure interdependencies.

In our initial consultations with researchers, an idea emerged to study the "urban environment" as a complete system, to observe and analyze the interconnections and interdependencies among urban infrastructures and identify points of physical and social vulnerability. A collaborative research effort could be organized to "map" urban vulnerability and resilience.

Power Systems

Power systems are perhaps the most important component of critical infrastructure, since most other systems require a continuous flow of energy to operate. Canadians have become very dependent on electrical power and, while we have learned to cope with small local outages caused by storms or accidents, larger interruptions can have significant impacts on our social and economic activities. While large scale, sustained power outages are rare in Canada, some notable recent events have illustrated the vulnerability of power systems and our dependence on them. For example, the August 2003 blackout was the product of a site-specific failure which, because of the vulnerable interdependence of the system, cascaded into a wider regional failure, leaving 50 million people in Canada and the United States without power (U.S.-Canada Power System Outage Task Force, 2003). In 1998, ice storms in Ontario, Québec and New Brunswick felled thousands of hydro towers, disrupting the lives of millions of residents for weeks (Environment Canada, 2001).

Beyond these well-documented hazards, there is also evidence that Canadian power systems could be affected by other, lesser known hazards. Boteler (2001) suggests that major geomagnetic storms, which stem from eruptions on the surface of the sun and occur twice per year on average, could create unexpected surges in power systems which could damage equipment and threaten continuity. In March 1989, for example, a geomagnetic storm caused a major blackout on the Hydro-Québec system, which interrupted the power supply of six million people for nine hours (Boteler, 2003).

Communications and Information

Communications and information infrastructure includes linkages which move data from point to point, networks that deal with addressing, routing and transport coordination, and data manipulation components such as computing systems (Hunteman et al, 1997). Combined, these elements support a wide range of services for Canadians.

In the context of disaster resilience, one of the most important elements of communications and information infrastructure is emergency communications. A conclusion that is commonly found in post-disaster incident reports is that the quality, interoperability and continuity of communications among emergency responders are essential to effectively manage a hazard impact and to facilitate swift recovery. Despite this, Moore et al (2002) suggest that Canadian municipal mobile radio networks are highly vulnerable to hazard-induced disruption. Vulnerability of local emergency communications stems mainly from a lack of common standards and interoperability, coupled with the fact that system maintenance and improvement tends to be a low priority among municipal governments (Moore et al, 2002). Moreover, in developing their emergency communications, many Canadian municipalities have come to rely on commercial mobile telecommunications, which are vulnerable to a loss of performance or disruption from hazards (Anderson and Gow, 1999; 2003). In both of these cases, the authors suggest that further research is required to fully assess the vulnerabilities of emergency communications and identify mitigation measures.

The continuity of communication systems is largely dependent on a continuous flow of electrical power. During the August 2003 blackout, for example, it was observed that thousands of significant Internet communications portals were interrupted, leaving millions of individual users and organizations such as banks, educational institutions, government units and hospitals without connectivity for hours or days (Cowie et al, 2003).

Transportation

Transportation - particularly the flow of goods in and out of urban areas - is an important component of urban infrastructure. As more Canadian businesses have adopted the doctrine of "just-in-time delivery", the demand for daily shipping via rail and truck has increased substantially; in 2000, about 45 percent of domestic goods were shipped by rail, followed closely by trucking at 43 percent (Transport Canada, 2000), but both modes of transportation are vulnerable to hazards. Two major 1999 snowstorms in southern Ontario dumped 118 centimetres of snow within two weeks on the City of Toronto, virtually prohibiting transportation (including air traffic) in and around the city (OCIPEP, 2003). Near Sackville, New Brunswick, the only rail line connecting Halifax with the rest of Canada is located within two metres of the normal spring high tide line, where it is vulnerable to severance in the event of a storm surge (McBean, 2000).

Water and Wastewater

Canada's buried urban infrastructure networks have declined rapidly over the last 15 years, due to aging, poor construction practices, lack of quality control, little or no maintenance, misuse (e.g., discharge of chemicals) and operation at capacities higher than design intended (Allouche, 2003). Quoting from various studies, Allouche suggests that the replacement cost of Canada's water and sewer infrastructure could exceed $140 billion (in 1996 dollars), and that even upgrading systems to appropriate standards over the next 15 years could cost around $75 billion. Currently, spending from all sources adds to only about half of the investment needed to upgrade these systems. Specific hazard impacts that stem from or are exasperated by decaying buried infrastructure include health-related costs and losses due to inadequate potable water distribution system; flooded roads, which creates risk for drivers; flooded basements, which can lead to costly property damage and potential health hazards, and contamination of groundwater due to breached sanitary sewer systems (Allouche, 2003).

Resilience and Infrastructure: Countermeasures

Given the uncertainty surrounding the hazard variable (e.g., location, frequency and magnitude of hazard events), we cannot anticipate and prevent all disasters. However, we can ensure greater reliability in the continuity of infrastructure systems through robustness, redundancy and adaptation.

Robustness

As discussed earlier, resilience clearly must involve a structural dimension, for it is the way systems are designed and built that will ultimately determine their ability to absorb environmental stresses and recover quickly. In this regard, robustness (i.e., the strength of construction and capacity to absorb internal and external hazard stresses) can be seen as one of the "ends" of resilience - a condition specifically designed to enhance resilience (Bruneau et al, 2003).

Redundancy

The word "redundancy" is often used in a derogatory way, to refer to instances of "wasteful" overlap or duplication. Indeed, under normal circumstances, duplication of a particular function seems like an inefficient use of scarce resources and easily becomes the target of efforts to "streamline" operations. As several authors have pointed out, however, this conventional wisdom produces a false sense of organizational efficiency (Landau, 1969) and leaves an organization or system more vulnerable to failure in the event of unanticipated stress (Lerner, 1986). Without a backup, failure of a single component can bring about the failure of the system.

Lerner (1986) develops several theoretical models of redundancy through duplication - where all functions of unit A are also assigned to unit B (to be distinguished from overlap - where unit A and unit B are each assigned a range of functions, some of which are identical). He assesses each arrangement on the basis of (1) costs and benefits of redundancy, (2) predictability of breakdown and (3) the ability of the system to contain breakdowns. In his examples, he uses a theoretical desired output of 6.

1. Enlightened Waste. In this model, two parallel units perform the same function simultaneously, both producing the required level of output (i.e., each produces a load of 6). In the event that one should fail, the other will continue to produce the desired output, thus there will be no drop in service. This configuration is the most reliable for a number or reasons: first, it does not rely on a triggering mechanism, as both units are in operation at all times; second, it does not require complex prediction as to which unit is more likely to fail, thus it is not necessary to designate one as primary and the other as secondary. However, this approach is also costly and inefficient, as excess production capacity (i.e., >6) is wasted. Secondly, this configuration requires that the system have the ability to absorb the excess output under normal circumstances (through storage or "spill off") and adjust to one unit's production in the event that the other fails.

2. Stress the Survivor. In this model, two parallel units operate simultaneously, but each produces at a level below its capacity. For example, if the required level of output is 6, each unit would produce 3, leaving a reserve capacity. In the event that one fails, the other is boosted to full capacity and is "stressed" until the first unit can be brought back online. The primary advantage that this configuration has over the enlightened waste model is that the units work together to produce the required level of output, hence there is no waste and no requirement that the system absorb excess output. On the other hand, this model requires a mechanism within the system to distinguish which unit failed, in order to trigger the surviving unit to full production. Secondly, it requires that each unit is capable of performing at full capacity to compensate for the failure of the other. The stress of operating at full capacity, moreover, increases the risk of error or breakdown than can be expected under normal circumstances, requiring sensitive monitoring and warning systems.

Lerner emphasizes that the continuity of a system requires people with authority to make difficult decisions, including assigning blame to the failed unit, switching the surviving unit to full capacity and switching it back to normal capacity after the offending unit is repaired.

3. Mobilizing Reserves. This approach also employs two parallel units, but in this case the first unit produces at maximum capacity and the second is kept in reserve, ready to be mobilized in the event that the first unit fails. This approach would be more economical than either of the preceding models, as it involves only one unit in operation (i.e., requiring maintenance, monitoring) at a time. The major weakness of this model is that it requires the reserve unit to be maintained at a high state of readiness, such that it can be quickly activated. Since many failures are gradual, a monitoring system must be employed on the first unit, attached to a triggering mechanism which activates the backup unit before complete failure. These factors can involve significant costs. In addition, as with the second model, this configuration requires that the reserve unit be able to absorb the stress of operating at full capacity. Unlike the second model, however, which permits testing of the reserve unit under normal circumstances (at least at half­capacity), this model relies on a reserve unit that must leap from a dormant state to full production.

Adaptation

The uncertainty which surrounds future hazards and their impacts complicates structural approaches to building resilience in two ways: first, since we cannot predict every possible hazard scenario we will face, clearly-defined and specifically-targeted strategies are difficult to justify, particularly when measured against more immediate, visible concerns. Second, even if we could accurately pinpoint the nature, probability and impact of future hazard events, the investment required to prevent their occurrence may not be feasible or even desirable, relative to other uses of scarce resources.

As such, we must also ensure that infrastructure systems have a strong adaptive capacity, meaning "the ability or capacity of a system to modify or change its characteristics or behaviour so as to cope better with existing or anticipated external stresses (Brooks, 2003)." Over time, as new knowledge reduces uncertainty, policies and systems should be flexible and able to adjust. An explicit strategy of adaptation accommodates uncertainty in that efforts to enhance adaptive capacity can be made without precise knowledge of the nature and timing of future hazard events. Combined with greater robustness and forms of redundancy, adaptation permits learning from failures and allows flexible responses to unanticipated disruptions.

Future Research: What do we need to know?

As outlined herein, a growing body of research is developing regarding the resilience of systems in general and the disaster resilience of cities in particular. In reviewing this literature and in initial consultations with other researchers, we have identified a number of areas where further research will help to operationalize the concept of disaster resilience. The following section introduces some questions for future research and provides some initial discussion.

Disaster Resilience

How can we measure resilience?

In order to effectively build urban resilience, we must be able to identify where investment is most urgently needed and to gauge the effectiveness of initiatives to enhance resilience. Some have already begun this research; for example, Bruneau et al (2003) suggest that resilience incorporates four interrelated dimensions: a technical dimension - the ability of physical systems to perform at an acceptable level under hazard stress; an organizational dimension - the capacity of organizations to manage critical facilities and have the responsibility and authority to carry out disaster-related functions needed to achieve resilience; a social dimension - measures specifically designed to mitigate the negative social consequences of failed critical services following a disaster; and an economic dimension - the capacity to mitigate direct and indirect economic losses from a disaster event. Because of this complexity, it is difficult to identify a single measure of community resilience, but the authors have developed a "conceptual framework and a set of measures" with which to assess the resilience of a community. By quantifying system performance criteria (e.g., reduced probability of failure, reduced consequences from failures), the resilience of a community can be measured and improvement can be monitored. Though developed as a measure of seismic resilience, the model is applicable in the context of others hazards as well. Further study should build on this research to develop a more sophisticated understanding of resilience and methods with which we can measure progress in this area.

What are the barriers to creating disaster resilient cities?

In May 2003, ICLR hosted a workshop attended by municipal officials, provincial government representatives and academics to discuss risks posed to municipal water and sewer systems by hazards, particularly in the context of climate change. The purpose of the workshop was to engage participants in a discussion about how vulnerability can be reduced within this specific sector of municipal critical infrastructure. The workshop involved presentations on a variety of topics, followed by a group discussion to obtain ideas and feedback from participants. Participants were asked to identify barriers within their community that would hamper the incorporation of mitigation into sewer and water infrastructure - these points are also useful to discuss within the context of disaster resilient cities.

One of the primary barriers identified by participants is a lack of political commitment. Building disaster resilience requires action and investment in anticipation of a future event, so its benefits are not immediately tangible and may never be realized. Though the initial costs of building disaster resilience may be far less than the savings following a disaster event, they are immediately visible and are often rejected in favour of other pressing policy priorities.

One reason for this subdued interest is a lack of pressure from the public. People have a tendency to minimize the risk they perceive from environmental hazards (Drabek, 1991); they are often apathetic towards the risks posed by hazards or take a fatalistic view that little can be done to avoid them (Berke, 1998). Faced with this public apathy, local elected officials have little to gain politically from advocating disaster resilience; on the other hand, they may face significant political fallout if large investments or restrictions are required which may negatively impact local interests.

A second barrier to disaster resilience is a lack of resources. First, municipalities lack personnel resources - individuals who are aware of hazards and who can focus on reducing vulnerability and developing strategies for resilience in municipal infrastructure. Second, municipal managers require data resources, to better understand how resilience can be incorporated into design and operations, but also to support requests for funding from decision-makers. Data resources required by municipal managers include specific information on hazards, predictions of the changing risk due to climate change and examples of resilience from other communities, illustrated through case studies and success stories. The third and perhaps most obvious problem is a lack of financial resources. Without a significant financial commitment, municipal infrastructure will continue to deteriorate and system failures will threaten the sustainability of Ontario communities.

Further research should explore the incentives and disincentives which influence decision­making regarding disaster resilience. It would also be useful to identify strategies used by municipal managers to circumvent the barriers to resilience in the urban environment.

How can we better communicate the value of disaster resilience?

A commitment to building disaster resilient cities begins with a recognition that (1) disasters pose a serious threat to the sustainability of our cities, (2) because of uncertainty, prevention of disasters is difficult or impossible, and (3) that there are concrete actions we can take to reduce vulnerability and improve our capacity to absorb environmental stresses and recover quickly. While it is relatively easy to accept the value of disaster resilience when it is presented in the abstract, it may be more difficult to envision the application of its principles to a specific city. Though Canadian cities face a wide range of differing environmental conditions, they have in common most components of urban infrastructure. As such, it would be useful to design and develop a series of case studies, which could be used to:

• Demonstrate resilience. In order to foster a better understanding and deeper appreciation for the value of resilience as a strategy, case studies could be designed which illustrate the concept in practice. For example, a study could document the interruption of aninfrastructure system, identifying the causes of failure, the system's response and preventative measures that were (or were not) implemented after the system was restored.
• Showcase investments in resilience. A case study which showcases a city's successful efforts in promoting and implementing the principles of disaster resilience would be useful to encourage action in other communities. All too often, failed systems receive undue negative attention; instead of being used to facilitate learning, these stories discourage innovation.
• Demonstrate costs and benefits. While there are substantial benefits associated with disaster resilience, the implementation of the principles outlined herein would inevitably entail certain direct or indirect costs or would require compromise with other policy objectives. A case study which analyzes the costs and benefits associated with the choices of a profiled community would be useful to support decision-making in other communities.
• Draw lessons from past events. One particularly useful case study would be an analysis of how a community was impacted by and dealt with a disaster event. This could involve a normative component (what could we do?), a descriptive component (what did we do?) and a prescriptive component (what should we do?). Such an approach would better contextualize the principles of disaster resilience and provide a useful decision support tool for other cities.
• What role should the various levels of government play in building disaster resilience?

Though this paper has focused on the disaster resilience of cities, this is not meant to imply that local governments should bear sole responsibility for its implementation. The disaster resilience of Canada's cities is clearly an issue of local, provincial and national importance and should thus involve a collaborative effort among the three levels of government. But what role should the various governments play in building disaster resilient cities?

Traditionally, the government role in disaster management has largely been reactive in nature, with little emphasis on risk reduction or disaster prevention and a heavy reliance on post-disaster recovery assistance. For the most part, this strategy has worked, as disaster events have been sporadic and could be absorbed in budgetary appropriations. Since 1970, Ottawa has administered the Disaster Financial Assistance Arrangements (DFAA), a program established to provide recovery funds in the event that damages from a disaster exceed the capabilities of a provincial or territorial government (OCIPEP, 2003b). Several provinces also have formal disaster assistance programs, such as the British Columbia Disaster Financial Assistance Program and the Ontario Disaster Relief Assistance Program.

In light of the hazards we will face as the climate changes, however, passive reliance on post­disaster relief is becoming an unsustainable approach. The public cost of disaster recovery is already massive, totaling over $1.4 billion since 1970 (OCIPEP, 2003b), and has risen sharply over the past decade: "since 1994, government disaster finance assistance payments are averaging $300 million per year, compared to an annual average of only $25 million for the period 1970-1993. This is money that had to be diverted from other public policy and spending priorities (ICLR and EPC, 1998)".

The primary objective of disaster relief is to rebuild communities as quickly as possible, restoring buildings and infrastructure to pre-disaster conditions. Unfortunately, this approach fails to address the underlying factors (i.e., hazards and vulnerability) that contributed to the disaster in the first place, thus setting the stage for repeat losses and perpetuation of a disaster­rebuild-disaster cycle. Moreover, paying for disaster losses without addressing the root causes can unwittingly reinforce high-risk decisions and behaviour. Wright and Rossi (1981) argue:

…such policies pose perverse and counterproductive incentives: in brief, postdisaster relief provisions punish risk-averters and reward risk-takers; the wise and cautious, that is, are made to pay for the folly, shortsightedness, and simple bad luck of others. Thus these policies encourage the rehabitation of hazardous areas after disaster has struck, because they absolve individuals from any responsibility for the risk (p. 50).

There are positive indications, however, that the federal government and some provincial governments are shifting to a more proactive approach to disaster management. In Québec and Ontario, for example, recent legislation now requires all local governments to participate in comprehensive disaster planning, including hazard assessment and mitigation (Government of Québec, 2000; Government of Ontario, 2003). The federal government has also taken some preliminary steps, including extensive public consultation on the idea of a National Disaster Mitigation Strategy. In 1998, Emergency Preparedness Canada partnered with the Institute for Catastrophic Loss Reduction to hold a series of regional workshops, bringing together over 400 people, drawn from a broad range of stakeholder groups, to discuss disaster management in Canada and propose future policy directions (ICLR and EPC, 1998). In 2001, a second round of consultations was initiated, to engage stakeholders in dialogue, which would be used to further, develop a national model for disaster mitigation (OCIPEP, 2002). These discussions revealed a broad base of support for greater efforts in mitigation as an investment against future disaster impacts.

Further research should examine the nature of intergovernmental relationships in the context of disaster management and analyze how federalism influences the outcome of disaster policies. Specifically, research is needed to assess alternatives to the current approach to disaster management and propose alternative arrangements that would provide for adaptation and enhance the disaster resilience of Canada's cities.

How can disaster resilience be coupled with other objectives?

While disaster resilience can be envisioned as a valuable end in itself, it cannot be pursued in isolation, but rather must be incorporated as an element within a larger strategy for long-term urban sustainability. Future research in this area should assess how strategies for disaster resilience can be coupled with other important policy goals such as green infrastructure, climate change mitigation, pollution control and energy efficiency.

The vulnerability of a community, influenced by a wide range of social, economic and cultural factors, is a primary determinant of its capacity to adapt to and recover quickly from disaster impacts. As such, collaborative efforts are required to identify and analyze factors that contribute to vulnerability and explore the role of social capital and social support networks in responding to disasters and facilitating quick recovery. Current efforts to reduce vulnerability (e.g., poverty reduction, gender equity) should be recognized and incorporated into a more holistic strategy for community disaster resilience.

Disaster Resilience and Infrastructure

Where should infrastructure funds be allocated and why?

Investment in infrastructure can play a major role in urban disaster resilience but, given scarce resources, it is necessary to identify and prioritize target areas that need more urgent attention. This requires a framework with which to compare and evaluate investments, to ensure funds are best allocated to improve resilience. Robinson, Woodward and Varnado (1998) suggest that the importance of critical infrastructure systems should be evaluated using a consequence-based assessment procedure, which they describe as follows:

It begins by defining the consequences of disruptions, then by identifying critical nodes - elements that are so important that severe consequences would result if they could not operate. Finally, it outlines protection mechanisms and associated costs of protecting those nodes...It permits the costs and benefits of each protection option to be assessed realistically and is particularly attractive in situations in which the threat is difficult to quantify, because it allows the costs of disruptions to be defined independently of what causes the disruptions (p. 64).

In another recent study, Chang (2003) uses life cycle cost analysis, a method widely used for municipal infrastructure management, to develop a framework for evaluating the costs and benefits of investment in urban infrastructure systems to mitigate disaster impacts. These and other studies could be used to develop a decision-support tool for identifying priority areas and allocating funds.

How should decisions be made and who should be involved?

Construction and maintenance of public urban infrastructure is a responsibility shared among the three levels of Canadian government. As such, decision-making regarding infrastructure investment should incorporate municipal, provincial and federal representatives. Moreover, much of Canada's critical infrastructure is privately owned, so it is important to involve representatives from the private sector.

Multilevel decision-making in other areas has demonstrated that differing interests and priorities of stakeholders can sometimes stymie collaborative efforts. In order to minimize conflict, one approach is to create advisory groups of government managers and private sector representatives who share common interests and have expertise in a particular area of infrastructure. Expert knowledge on particular areas of urban infrastructure can be compiled into a set of "recommended practices" that could be used by municipal officials to develop better, more resilient infrastructure systems. This is the approach used for the National Guide to Sustainable Municipal Infrastructure, created in 2001 through a partnership between Infrastructure Canada, the National Research Council and the Federation of Canadian Municipalities.

Closing Remarks

Canadian cities face a wide range of natural, technological and human-induced hazards. While there is considerable uncertainty as to the nature, frequency, intensity and magnitude of hazards in the future, this should not be used to justify inaction. The principles of resilience specifically accommodate uncertainty by focusing on the reduction of vulnerability, a goal which offers societal benefits far beyond our capacity to deal with disasters. While there are many different models for creating disaster resilient cities, they all share a common goal of improving our capacity to adapt to hazards and recover quickly from their impacts.

This paper was written to provide a background context for further research on disaster resilient Canadian cities. We feel it would be useful to organize a larger workshop (approx. 30-40 people) to better develop the concept in the Canadian context and to establish an agenda for future research. We invite comments on the concepts and principles outlined herein.

Sources

• Allouche, Erez N. (2003). «Buried Municipal Infrastructure: Vulnerability and Mitigation". Presented at workshop Mitigating Municipal Disaster Losses: Reducing Vulnerability in Water and Sewer Infrastructure, 29 May, London, Ontario.
• Anderson, Peter S. and Gordon A. Gow. (1999). Commercial Mobile Telephone Services and the Canadian Emergency Management Community: Prospects and Challenges for the Coming Decade. Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness.
• Anderson, Peter S. and Gordon A. Gow. (2003). "A General Framework for Mitigation-Oriented Planning Assessments of Mobile Telecommunications Lifelines (MTL)," Natural Hazards 28 (2): 305-318.
• Berke, P. R., J. Kartez and D. Wenger. (1993). "Recovery After a Disaster: Achieving Sustainable Development, Mitigation, and Equity," Disasters 17: 93-109.
• Boteler, D. H. (2001). "Assessment of Geomagnetic Hazard to Power Systems in Canada," Natural Hazards 23: 101-120.
• Boteler, D. H. (2003). "Geomagnetic Hazards to Conducting Networks," Natural Hazards 28 (2): 537-561.
• Bradford, Neil. (2003). Why Cities Matter: Policy Research Perspectives for Canada. Discussion Paper No. F23, Canadian Policy Research Network, Ottawa.
• Brooks, Nick. (2003). Vulnerability, Risk and Adaptation: A Conceptual Framework. Working Paper 38, Norwich, U.K: Tyndall Centre for Climate Change Research.
• Bruneau, Michel, et al. (2003). "A Framework to Quantitatively Assess and Enhance the Seismic Resilience of Communities," Earthquake Spectra 19 (4): 733-752.
• Burby, Raymond J., et al. (2000). "Creating Hazard Resilient Co mmunities Through Land-Use Planning," Natural Hazards Review 1 (2): 99-106.
• Canada. Environment Canada. (2001). Ice Storm 1998. Ottawa: Environment Canada.
• Chang, Stephanie E. (2003). "Evaluating Disaster Mitigations: Methodology for Urban Infrastructure Systems," Natural Hazards Review 4 (4): 186-196.
• Cowie, James H., et al. (2003). Impact of the 2003 Blackouts on Internet Communications. Hanover, NH: Renesys Corporation.
• Drabek, T. E. (1991). "The Evolution of Disaster Management," in Drabek, T.E. and G.J. Hoetmer (eds.), Disaster Management: Principles and Practice for Local Government. Washington, D.C.: International City Management Association: 3-29.
• Emergency Management Australia. (1998). Australian Emergency Management Glossary. Canberra: Emergency Management Australia.
• Etkin, David and Soren Erik Brun. (2001). Canada's Hail Climatology: 1977-1993. Toronto: Institute for Catastrophic Loss Reduction. Available: <www.iclr.org>.
• Etkin, David A., et al. (2002). A Tornado Scenario for Barrie, Ontario. Toronto: Institute for Catastrophic Loss Reduction. Available: <www.iclr.org>.
• Geis, Donald E. (2000). "By Design: The Disaster Resistant And Quality-Of-Life Community," Natural Hazards Review 1(3): 151-160.
• Godschalk, David R. (2003). "Urban Hazard Mitigation: Creating Resilient Cities," Natural Hazards Review 4(3): 136-143.
• Government of Ontario. (2002). Bill 148: Emergency Readiness Act. 3rd Session, 37th Legislature. Chapter 14, Statutes of Ontario. Available : <www.ontla.on.ca/documents/Bills/37_Parliament/Session3/b148ra.pdf>.
• Government of Québec. (2000). Civil Protection Act. 1st Session, 36th Legislature. Statutes of Québec. Available: <www.assnat.qc.ca/archives-36leg1se/eng/Publications/Projets-loi/publics/00-a173.htm>.
• Hilborn, Ray. (1987). "Living with Uncertainty in Resource Management," North American Journal of Fisheries Management 7 (1): 1-5.
• Hunteman, William J., et al. (1997). Research & Development Priorities for Communications and Information Infrastructure Assurance. Berkeley, CA: Distributed Systems Department, Ernest Orlando Lawrence Berkeley National Laboratory, University of California.
• Institute for Catastrophic Loss Reduction (ICLR) and Emergency Preparedness Canada (EPC). (1998). A National Mitigation Policy. Ottawa: Emergency Preparedness Canada. Available: <ww3.psepc-sppcc.gc.ca/NDMS/Files/nmrdspolicy_e.doc>.
• Klein, Richard J.T. (2002). "Adaptation to Climate Variability and Change: What is Optimal and Appropriate?," in Giupponi, Carlo and Mordechai Shechter (eds), Climate Change in the Mediterranean: Socio-Economic Perspectives of Impacts, Vulnerability and Adaptation. Cheltenham, UK: Edward Elgar.
• Klein, Richard J.T. and Richard S.J. Tol. (1997). Adaptation to Climate Change: Options and Technologies: An Overview Paper. Technical Paper FCCC/TP/1997/3. Bonn, Germany: United Nations Framework Convention on Climate Change Secretariat.
• Klein, Richard J.T., Robert J. Nicholls and Frank Thomalla. (2002). The Resilience of Coastal Megacities to Weather-Related Hazards: A Review. Presented at conference "The Future of Disaster Risk: Building Safer Cities", World Bank Disaster Management Facility, December 2002.
• Landau, Martin. (1969). "Redundancy, Rationality, and the Problem of Duplication and Overlap," Public Administration Review 29 (4): 346-358.
• Lerner, Allan W. (1986). "There is More Than One Way to Be Redundant," Administration and Society 18 (3): 334­-359.
• Little, Richard G. (2002). Toward More Robust Infrastructure: Observations on Improving the Resilience and Reliability of Critical Systems. Proceedings of the 36th Hawaii International Conference on System Sciences: University of Hawaii at Mãnoa.
• McBean, Gordon. (2000). Climate Change and Coastal Zones. Presentation to the New Brunswick Roundtable on the Environment and Economy.
• McBean, Gordon. (2000). Forecasting in the 21st Century. 9th IMO Lecture, World Meteorological Organization, Geneva, WMO-No. 916.
• McBean, Gordon. (2003). "Climate Change and Severe Weather: A Basis for Action", Natural Hazards, forthcoming.
• McEntire, David A. (2000). "Sustainability or Invulnerable Development? Proposals for the Current Shift in Paradigms," The Australian Journal of Emergency Management Autumn: 58-61.
• McEntire, David A. (2001). "Triggering Agents, Vulnerabilities and Disaster Reduction: Towards a Holistic Paradigm," Disaster Prevention and Management 10 (3): 189-196.
• McEntire, David A., et al. (2002). "A Comparison of Disaster Paradigms: The Search for a Holistic Policy Guide," Public Administration Review 62 (3): 267-281.
• Mileti, Dennis S. (1999). Disasters by Design: A Reassessment of Natural Hazards in the United States. Washington, D.C.: Joseph Henry Press.
• Mills, Brian, Jean Andrey, Jennifer Yessis and Dwight Boyd. (2001). "The Urban Environment as Hazard Source and Sink," Environments 29 (1): 17-38.
• Mirza, M. Saeed and Murtaza Haider. (2003). The State of Infrastructure in Canada: Implications for Infrastructure Planning and Policy. Montreal: McGill University.
• Moore, Alexandra, Ken Hancock, Roger Stacey and Paul Stacey. (2002). The Vulnerability of Mobile Telecommunications to Natural Hazards. Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness.
• Munich Reinsurance Company (MunicRe). (1999). Topics 2000. Natural Catastrophes: The Current Position. Munich: Munich Reinsurance Company.
• Office of Critical Infrastructure Protection and Emergency Preparedness (OCIPEP). (2001). Natural Hazards in Canada. Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness.
• Office of Critical Infrastructure Protection and Emergency Preparedness (OCIPEP). (2002). NDMS: Final Report of the Stakeholder Workshops. Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness. Available: <ww3.psepc-sppcc.gc.ca/NDMS/syn_sum_e.asp>.
• Office of Critical Infrastructure Protection and Emergency Preparedness (OCIPEP). (2003). Canadian Disaster Database. Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness. Available: <www.ocipep.gc.ca/res/em/cdd/index-en.asp>
• Office of Critical Infrastructure Protection and Emergency Preparedness (OCIPEP). (2003b). Disaster Financial Assistance Arrangements (DFAA). Ottawa: Office of Critical Infrastructure Protection and Emergency Preparedness. Available: <www.ocipep.gc.ca/info_pro/fact_sheets/general/FA_df_assist_e.asp>.
• Pimm, Stuart L. (1984). "The Complexity and Stability of Ecosystems," Nature 307: 321-326.
• Public Safety and Emergency Preparedness Canada. (NSEPC). (2003). National Critical Infrastructure Assurance Program - Synopsis . Ottawa: Public Safety and Emergency Preparedness Canada. Available: <www.ocipep.gc.ca/critical/nciap/synopsis_e.asp>.
• Robinson, C. Paul, Joan B. Woodward and Samuel G. Varnado. (1998). "Critical Infrastructure: Interlinked and Vulnerable," Issues in Science and Technology (Fall): 61-67.
• Sarewitz, Daniel, Roger A. Pielke, Jr. and Radford Byerly, Jr. (2000). Prediction: Science, Decision Making and the Future of Nature. Washington, D.C.: Island Press.
• Statistics Canada. 2002. A Profile of the Canadian Population: Where We Live. Ottawa: Statistics Canada. Available: <http://geodepot2.statcan.ca/Diss/Highlights/Index_e.cfm>
• Timmerman, Peter. (1981). Vulnerability, Resilience and the Collapse of Society: A Review of Models and Possible Climatic Applications. Toronto: Institute for Environmental Studies, University of Toronto.
• Tobin, Graham A. (1999). "Sustainability And Community Resilience: The Holy Grail Of Hazards Planning?," Environmental Hazards 1: 13-25.
• Transport Canada. (2000). Transportation in Canada 2001. Available: <www.tc.gc.ca/pol/en/report/anre2001/tc0107be.htm>
• United Nations Human Settlement Programme (UNHSP). (2002). Urbanization: Facts and Figures. Available: <www.unchs.org/downloads/docs/3160_99185_backgrounder5.doc>.
• United Nations International Strategy for Disaster Reduction (ISDR). (2002). Natural Disasters and Sustainable Development: Understanding the Links Between Development, Environment and Natural Disasters. Background document for the World Summit on Sustainable Development (WSSD) No. 5., Geneva.
• United Nations International Strategy for Disaster Reduction (ISDR). (2002b). Living with Risk: A Global Review of Disaster Reduction Initiatives. Geneva: International Strategy for Disaster Reduction Secretariat.
• United States-Canada Power System Outage Task Force. (2003). Interim Report: Causes of the August 14th Blackout in the United States and Canada. U.S.-Canada Power System Outage Task Force.
• Walter, Jonathan (ed.) (2003). World Disasters Report 2003: Focus on Ethics and Aid. Geneva: International Federation of Red Cross and Red Crescent Societies.
• World Commission on Environment and Development (WCED). (1987). Our Common Future. Oxford: Oxford University Press.
• Wright, James D. and Peter H. Rossi. (1981). "The Politics of Natural Disaster: State and Local Elites," in Wright, James D. and Peter H. Rossi (eds.), Social Science and Natural Hazards. Cambridge, MA: ABT Associates Inc.: 45­67.

Acknowledgements

• The authors wish to thank all who contributed thoughts and ideas to the development of this paper. Special thanks are extended to the following individuals for their participation in a mini-workshop to discuss concepts and ideas related to disaster resilient cities:
• Dr. Alan G. Davenport, Founding Director and Professor Emeritus, Boundary Layer Wind Tunnel Laboratory, Department of Civil and Environmental Engineering, The University of Western Ontario
• Dr. Slobodan Simonovic, Professor and ICLR Chair in Engineering, University of Western Ontario
• Dr. Andrew Sancton, Professor and Chair, Department of Political Science, University of Western Ontario
• Dr. Robert Young, Professor and Canada Research Chair in Multi-Level Governance, Department of Political Science, University of Western Ontario
• Dr. Dan Shrubsole, Associate Professor, Department of Geography, University of Western Ontario
• Dr. Michael Bartlett, Associate Professor, Department of Civil and Environmental Engineering, University of Western Ontario