Content uploaded by Christopher G. Burton
Author content
All content in this area was uploaded by Christopher G. Burton on Jul 14, 2014
Content may be subject to copyright.
Journal of Homeland Security and
Emergency Management
Volume 7, Issue 1 2010 Article 51
Disaster Resilience Indicators for
Benchmarking Baseline Conditions
Susan L. Cutter
∗
Christopher G. Burton
†
Christopher T. Emrich
‡
∗
University of South Carolina, scutter@sc.edu
†
University of South Carolina, burton@mailbox.sc.edu
‡
University of South Carolina, emrich@mailbox.sc.edu
Copyright
c
2010 The Berkeley Electronic Press. All rights reserved.
Disaster Resilience Indicators for
Benchmarking Baseline Conditions
∗
Susan L. Cutter, Christopher G. Burton, and Christopher T. Emrich
Abstract
There is considerable federal interest in disaster resilience as a mechanism for mitigating
the impacts to local communities, yet the identification of metrics and standards for measuring
resilience remain a challenge. This paper provides a methodology and a set of indicators for
measuring baseline characteristics of communities that foster resilience. By establishing baseline
conditions, it becomes possible to monitor changes in resilience over time in particular places and
to compare one place to another. We apply our methodology to counties within the Southeastern
United States as a proof of concept. The results show that spatial variations in disaster resilience
exist and are especially evident in the rural/urban divide, where metropolitan areas have higher
levels of resilience than rural counties. However, the individual drivers of the disaster resilience
(or lack thereof)—social, economic, institutional, infrastructure, and community capacities—vary
widely.
KEYWORDS: disaster resilience, indicators, Southeastern U.S.
∗
This research was funded by the Community and Regional Resilience Institute (CARRI) through
a grant from the Oak Ridge National Laboratory.
Introduction
Within federal circles, there continues to be considerable interest in the subject of
disaster resilience. The Subcommittee on Disaster Reduction’s (SDR) Grand
Challenges report (SDR 2005), which provided a blueprint for characterizing and
fostering disaster resilient communities stimulated the initial interest in disaster
resilience. Now with the formal establishment of the Office of Resilience within
the National Security Council in the White House, the policy community has
adopted resilience as one of the guiding principles for making the nation safer.
The policy goal is clear and pragmatic—if communities can increase their
resilience then they are in a much better position to withstand adversity and to
recover more quickly than would be the case if there were few or no investments
in building community resilience.
Interestingly, the policy community is slightly ahead of the research
community in pushing resilience as a means of mitigating disaster impacts.
Lingering concerns from the research community focus on disagreements as to
the definition of resilience, whether resilience is an outcome or a process, what
type of resilience is being addressed (economic systems, infrastructure systems,
ecological systems, or community systems), and which policy realm
(counterterrorism; climate change; emergency management; long-term disaster
recovery; environmental restoration) it should target. Some of these issues have
been discussed elsewhere (Cutter 2008a, b; Kahan et al. 2009; Klein et al. 2003;
Manyena 2006; Norris et al. 2008; Rose 2007; also see the CARRI research
reports at www.resilientus.org/publications).
This article provides a methodology and a set of indicators to measure the
present conditions influencing disaster resilience within communities. It then
applies this methodology to the Southeastern U.S. One key question drives the
analysis: How can we identify changes (either positive or negative) in community
resilience to disasters if we do not first have an understanding of the existing
conditions? The resilience indicators proposed in this paper serve as the baseline
set of conditions, from which to measure the effectiveness of programs, policies,
and interventions specifically designed to improve disaster resilience. While not
exhaustive, this set of baseline indicators provides one of the first empirically
based efforts to benchmark the pre-existing conditions that foster community
resilience.
Divergent Views on Community Resilience
The application of the resilience concept to natural hazards was initially the focal
argument in the assessment of natural hazards (Mileti 1999), which suggested that
resilience was the ability of a community to recover by means of its own
1Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
resources. Norris et al. (2008) also focus on community resilience and view it as a
process linking the myriad of adaptive capacities (such as social capital and
economic development) to responses and changes after adverse events. Here
resilience is as a set of capacities that can be fostered through interventions and
policies, which in turn help build and enhance a community’s ability to respond
and recover from disasters.
A very different conceptualization of disaster resilience comes from the
engineering sciences, with an emphasis on buildings and critical infrastructure
resilience. Using seismic risks as the exemplar, Bruneau et al. (2003) proposed a
resilience framework with an emphasis on structural mitigation, especially the
engineered systems concepts of robustness, redundancy, resourcefulness, and
rapidity. More recent research on resilience from a homeland security perspective
(primarily protecting critical infrastructure from terrorism) (Kahan et al. 2009)
also focuses on critical infrastructure resilience assuming that resilience is an
outcome measure with an end goal of limiting damage to infrastructure (termed
resistance); mitigating the consequences (called absorption); and recovery to the
pre-event state (termed restoration). While perhaps useful for counterterrorism
and protection of critical infrastructure, this operational framework ignores the
dynamic social nature of communities and the process of enhancing and fostering
resilience within and between communities.
Composite Indicators for Disaster Resilience
Not only is it vital to evaluate and benchmark the baseline conditions that lead to
community resilience, but it is equally important to measure the factors
contributing to adverse impacts and the diminished capacity of a community to
respond to and rebound from an event (Cutter et al. 2008a). Just as companies
have identified areas of opportunity and benchmarked their performance against
industry standards, governments are finding it useful to evaluate the performance
of communities in terms of their comparative resilience. While the latter is
partially to attract public interest in disaster loss reduction, it also provides
metrics to set priorities, measure progress, and aid in decision-making processes.
Composite indicators (often referred to as indices) are useful tools to accomplish
this task.
We use the term “composite indicator” to designate a manipulation of
individual variables to produce an aggregate measure of disaster resilience. An
indicator is a quantitative or qualitative measure derived from observed facts that
simplify and communicate the reality of a complex situation (Freudenberg 2003).
Indicators reveal the relative position of the phenomena being measured and when
evaluated over time, can illustrate the magnitude of change (a little or a lot) as
well as direction of change (up or down; increasing or decreasing). A composite
2 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
indicator is the mathematical combination of individual variables or thematic sets
of variables that represent different dimensions of a concept that cannot be fully
captured by any individual indicator alone (Nardo et al. 2008).
Composite indicators are increasingly recognized as useful tools for policy
making and public communication because they convey information that may be
utilized as performance measures (Saisana and Cartwright 2007). Numerous
approaches for measuring composite indicators at both global and local scales
have emerged. Many of these are central to the environmental hazards and natural
disasters community as they were developed to capture a society’s vulnerability to
social and/or environmental change. Among these are metrics created to assess
the susceptibility of small states to fluctuations within international economies
(Briguglio 1995; Easter 1999) and indicators designed to measure national well-
being (Millennium Change Corporation (MCC) 2007; Neumayer 2001; Prescott-
Allen 2001).
Also significant are composite indicators of social vulnerability to natural
or technological hazards. Cutter et al.’s Social Vulnerability Index (SoVI) is
perhaps the most well-known and widespread example (Cutter et al. 2003).
Additional indices that focus explicitly on aspects of social vulnerability include
the Prevalent Vulnerability Index (Cardona 2005), the Index of Social
Vulnerability to Climate Change for Africa (Vincent 2004), the Disaster Risk
Index (United Nations Development Program 2004), and the Predictive Indicator
of Vulnerability (Adger et al. 2004).
Similar metrics provide global assessments of quality of life and
sustainable development. These include the Human Development Index (United
Nations Development Program 1990, 2005) and the Environmental Sustainability
Index (Esty et al. 2005). Several indices have also been constructed to evaluate
the vulnerability of natural environments (Kaly et al. 2003; Kaly et al. 2004),
ecological health and environmental sustainability (Heinz Center 2008; National
Research Council 2000; Organization for Economic Cooperation and
Development (OECD) 2001) at a sub-national level. Lastly, composite indicators
have been utilized to determine the physical and social vulnerability of coastal
environments to sea level rise and its impacts (Boruff et al. 2005; Gommes et al.
1998; Pethick and Crooks 2000).
3Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Study Area Description
This study focuses on counties within the U.S. Federal Emergency Management
Agency’s (FEMA) Region IV (Figure 1). Region IV serves the southeastern states
Figure 1: Study area
of Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South
Carolina, and Tennessee. The impacts of natural disasters within this region are
widespread and vary extensively. Abundant rainfall and frequent thunderstorms
mean that flooding is a common problem for the region. Since 2000, for instance,
there have been more than 900 flood events in the state of Mississippi alone,
resulting in $410 million in property damage (Oxfam America 2009). The region
is also vulnerable to hurricanes that bring damaging winds, coastal and inland
flooding, catastrophic storm surge, and coastal erosion. Most of North America’s
well-known and most destructive hurricanes have affected this region--Hurricane
Hugo in 1989, Hurricane Andrew in 1992, Hurricane Ivan in 2004, and Hurricane
Katrina in 2005, the costliest disaster in U.S. history. Natural events such as
tornados, earthquakes, drought, and sea level rise also threaten the region.
4 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
national and international importance. Many rural and peripheral areas remain
poor, however, benefitting little from the new prosperity enjoyed by the
burgeoning metropolitan areas. These counties continue to have a high degree of
racial inequality and health disparities, where education and job skills are limited,
where the age composition of the existing population is elderly, and where a high
rate of outmigration exists.
Data and Methods
Theoretical framework for measuring disaster resilience
The literature on composite indicators is vast, and contains many methodological
approaches for index construction and validation. Most of the literature highlights
the need for a process of indicator construction that entails a number of specific
steps (Freudenberg 2003; Nardo et al. 2008). The first step involves the
development or application of a theoretical framework to provide the basis for
variable selection, weighting, and aggregation. This paper utilizes the inherent
resilience portion of the disaster resilience of place (DROP) model (Cutter et al.
2008b) as its conceptual basis. The DROP model presented the relationship
between vulnerability and resilience in a manner that is theoretically grounded
and amenable to empirical testing. Furthermore, the DROP framework explicitly
focused on antecedent conditions, specifically those related to inherent resilience.
Antecedent conditions are the product of place specific, multi-scale processes that
occur within and between natural systems, the built environment, and social
systems. Most of the scientific literature points to resilience within natural
systems (e.g., keeping wetlands intact or controlling development), yet the
resilience of social and organizational systems is equally significant. Disaster
impacts may be reduced through improved social and organizational factors such
as increased wealth, the widespread provision of disaster insurance, the
improvement of social networks, increased community engagement and
participation, and the local understanding of risk (Cutter et al. 2008a), as well as
through improvements in resilience within natural systems.
Historically, the states that make up FEMA Region IV have been at the
periphery of the U.S. economy, yet in recent decades, the region has become one
of significant growth with massive transformations in urbanization and
industrialization, in cultural and societal viewpoints, in agriculture, and in politics
(Wheeler 1999). The region was predominantly rural, with relatively few major
cities. This has changed over the past several decades and cities such as Atlanta,
Miami, Charlotte, Birmingham, Nashville, and Memphis continue to grow in
5Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
within the research community that resilience is a multifaceted concept, which
includes social, economic, institutional, infrastructural, ecological, and
community elements (Bruneau et al. 2003; Cutter et al. 2008a, b; Gunderson
2009; NRC 2010; Norris et al. 2008). Based on these findings, our index
comprises these subcomponents that were then further defined for analytic and
comparative purposes. Since it is often difficult to measure resilience in absolute
terms, we use a comparative approach and employ variables as proxies for
resilience (Cutter et al. 2008b; Schneiderbauer and Ehrlich 2006). The variable
selection had two considerations: 1) justification based on the extant literature on
its relevance to resilience; and 2) availability of consistent quality data from
national data sources.
We purposefully excluded ecological (or natural systems) resilience in this
present formulation. This is primarily due to data inconsistency and relevancy
when developing proxies for ecological systems resilience for large and diverse
study areas. For example, the inclusion of variables (in coastal areas) that account
for the prevalence or loss of wetlands and dunes is essential because they provide
buffers against storm surges. However, the use of such variables in regions far
removed from the coast or where wetlands and dunes are non-existent would
improperly skew results by implying a reduction in disaster resilience based on
the lack of these particular attributes.
Before the construction of the sub-indices could occur, a third step toward
creating a suitable composite index took place. All raw data values were
transformed into comparable scales utilizing percentages, per capita, and density
functions. These forms of standardization were essential to avoid problems
inherent when mixing measurement units since our variables were delineated in a
number of statistical units, ranges, and scales. The variables were then analyzed
for significantly high correlations between individual variables and when such
high correlations (e.g. Pearson’s R>0.70) were found, the variable was eliminated
from further consideration. Additionally, the internal consistency (reliability) of
the composite indicators was assessed utilizing a Chronbach’s Alpha
Reliability/Item analysis. We coupled a correlation analysis with a test for internal
consistency to decide whether the nested structure of the index was well defined
from a statistical perspective and if the available sub-indicators were sufficient
and appropriate to describe the disaster resilience phenomenon from a theoretical
Variable selection
Another crucial step in the creation of composite indicators is the identification of
variables that are relevant, robust, and representative, since the strengths and
weaknesses of composite indicators are based on the quality of the variables
chosen. Criteria for assuring the quality of variables are widespread within the
indicators literature, yet to date there is no single set of established indicators or
frameworks for quantifying disaster resilience. However, there is consensus
6 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Table 1: Variables used to construct disaster resilience index by subcomponent
Category Variable Effect on
Resilience
Justification Data Source
Social Resilience
Educational
equity
Ratio of the pct. population with
college education to the pct.
population with no high school
diploma
Negative Norris et al. 2008
Morrow 2008
U.S. Census 2000
Age Percent non-elderly population Positive Morrow 2008
U.S. Census 2000
Transportation
access
Percent population with a vehicle Positive Tierney 2009
U.S. Census 2000
Communication
capacity
Percent population with a telephone Positive Colten et al. 2008 U.S. Census 2000
Language
competency
Percent population not speaking
English as a second language
Positive Morrow 2008 U.S. Census 2000
Special needs Percent population without a
sensory, physical, or mental
disability
Positive Heinz Center 2002 U.S. Census 2000
Health coverage Percent population with health
insurance coverage
Positive Heinz Center 2002 U.S. Census 2000
Economic Resilience
Housing capital Percent homeownership Positive Norris et al. 2008
Cutter et al. 2008a
U.S. Census2000
Employment Percent employed Positive Tierney et al. 2001 U.S. Census 2000
Income and
equality
GINI coefficient Positive Norris et al. 2008 Computed from
U.S. Census 2000
Single sector
employment
dependence
Percent population not employed in
farming, fishing, forestry, and
extractive industries
Positive Berke &
Campanella 2006
Adger 2000
U.S. Census 2000
Employment Percent female labor force
participation
Positive NRC 2006 U.S. Census 2000
Business size Ratio of large to small businesses Positive Norris et al. 2008 County Business
Patterns (NAICS)
2006
Health Access Number of physicians per 10,000
population
Positive Norris et al. 2008 U.S. Census 2000
Institutional Resilience
Mitigation Percent population covered by a
recent hazard mitigation plan
Positive Burby et al. 2000
Godschalk 2007
FEMA.gov
Flood coverage Percent housing units covered by
NFIP policies
Positive Burby et al. 2000 bsa.nfipstat.com
Municipal
services
Percent municipal expenditures for
fire, police, and EMS
Positive Sylves 2007 USA Counties
2000
Mitigation Percent population participating in
Community Rating System for Flood
(CRS)
Positive Godshalk 2003
FEMA.gov
Political
fragmentation
Number of governments and special
districts
Negative Norris et al. 2008 U.S. Census 2002
Previous
disaster
experience
Number of paid disaster declarations Positive Cutter et al.
2008a
FEMA.gov
Mitigation and
social
connectivity
Percent population covered by
Citizen Corps programs
Positive Godshalk 2003 citizen.corps.gov
Mitigation Percent population in Storm Ready
communities
Positive Godshalk 2003 stormready.noaa.
gov
Infrastructure Resilience
Housing type Percent housing units that are not
mobile homes
Positive Cutter et al. 2003 U.S. Census 2000
Shelter capacity Percent vacant rental units Positive Tierney 2009 U.S. Census 2000
Medical
capacity
Number of hospital beds per 10,000
population
Positive Auf de Heide and
Scanlon 2007
American Hospital
Directory
www.ahd.com
Access/
evacuation
potential
Principle arterial miles per square
mile
Positive NRC 2006 GIS derived from
National Atlas.gov
Housing age Percent housing units not built
before 1970 and after 1994
Positive Mileti 1999
City and County
Databook 2007
Sheltering needs Number of hotels/motels per square
mile
Positive Tierney 2009 County Business
Patterns (NAICS)
2006
Recovery Number of public schools per square
mile
Positive Ronan and
Johnston 2005
Gnis.usgs.gov
Community Capital
Place
attachment
Net international migration Negative Morrow 2008 census.gov
Place
attachment
Percent population born in a state
that still resides in that state
Positive Vale &
Campanella 2005
U.S. Census 2000
Political
engagement
Percent voter participation in the
2004 election
Positive Morrow 2008 City and County
Databook 2007
Social capital-
religion
Number of religious adherents per
10,000 population
Positive Morrow 2008
Murphy 2007
Assn. of Religion
Data Archives
Social capital –
civic
involvement
Number of civic organizations per
10,000 population
Positive Morrow 2008
Murphy 2007
County Business
Patterns (NAICS)
2006
Social capital –
advocacy
Number of social advocacy
organizations per 10,000 population
Positive Murphy 2007 County Business
Patterns (NAICS)
2006
Innovation Percent population employed in
creative class occupations
Positive Norris et al. 2008 USDA Economic
Research Service
ers.usda.gov
7Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Components of disaster resilience
Our first subcomponent, social resilience, captured the differential social capacity
within and between communities. Linking demographic attributes to social
capacity (see Table 1) suggests that communities with higher levels of educational
equality, and those with fewer elderly, disabled residents, and non-native English
speaking residents likely exhibit greater resilience than places without these
characteristics. Similarly, communities that have high percentages of inhabitants
with vehicle access, telephone access, and health insurance also may demonstrate
higher levels of disaster resilience.
Economic resilience, the second subcomponent measures the economic
vitality of communities including housing capital, equitable incomes,
employment, business size, and physician access. Variables within this
component include percent employment, percent homeownership, business size,
female labor force participation, and a proxy for single sector employment
dependence. This variable provides a measure whether the local economic base is
diversified (more resilient) or largely based on a single sector such as agriculture
or fishing, which makes the community less resilient. These indicators allow the
examination of links that enhance or diminish economic stability at the
community level, particularly the stability of livelihoods. Dependency on a
narrow range of natural resources is an example of an economic factor directly
related to the stability of livelihoods. A reduction in resilience occurs, for
example, due to the boom and bust nature of single sector markets (Adger 2000)
and due to the threat of losses that may occur in a single sector economy (such as
fishing or agriculture) from an extreme event.
From a natural hazards perspective our third component, institutional
resilience, contains characteristics related to mitigation, planning, and prior
disaster experience (Table 1). Here, resilience is affected by the capacity of
communities to reduce risk, to engage local residents in mitigation, to create
organizational linkages, and to enhance and protect the social systems within a
community (Norris et al. 2008). Federal, state, and local governments within the
U.S. are slowly beginning to comprehend that the long term benefits of planning
and mitigation are important tools for increasing resilience and reducing losses
perspective. More than 50 variables were originally collected for this analysis.
However, after removing all highly correlated variables and achieving a level of
internal consistency that is generally accepted within the literature (Chronbach’s
Alpha = 0.700) (Nardo et al. 2008), thirty-six variables were employed in our
analysis. Each of the subcomponents contains seven to eight variables, culled
from publically available data sources (Table 1).
8 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
The fourth subcomponent, infrastructural resilience, is mainly an appraisal
of community response and recovery capacity (e.g. sheltering, vacant rental
housing units, and healthcare facilities). These indicators also provide an overall
assessment of the amount of private property that may be particularly vulnerable
to sustaining damage and likely economic losses. Vulnerable infrastructure
includes mobile homes that are particularly susceptible to catastrophic loss during
an event, and houses built prior to the enactment of mandatory building codes.
Critical infrastructure variables such as the amount principle arterial miles within
an area are also included since this type of infrastructure not only provides a
means for pre-event evacuations, but also acts as conduits for vital supplies, post-
disaster. A coastal community only accessible by a two-lane bridge may be more
vulnerable and less resilient than one with multiple ingress and egress routes.
Such a community would remain isolated and dependent upon costly airlifts and
boatlifts for supplies until an alternate route or temporary bridge was constructed.
The final sub-index, community capital, captures the relationships that
exist between individuals and their larger neighborhoods and communities. The
community capital sub-index embodies what many refer to as social capital. We
attempt to capture three key dimensions of social capital: sense of community,
place attachment, and citizen participation. We do this through proxies such as the
number of religious adherents (per 10,000 people), the number of civic and social
advocacy organizations (per 10,000 people), and the percentage of the population
employed in creative class occupations (knowledge-based workers, science,
engineering, arts, design, and the media) (Florida 2002), which is used as a
surrogate for social innovation. A sense of community is directly related to
bonding and is characterized by high concern for community issues, respect for
and service to others, and a sense of connection (Goodman et al. 1998; Norris et
al. 2008). Place attachment refers to one’s sense of community and often
underlies citizens’ efforts to revitalize a community (Perkins et al. 2002), and
citizen participation is the engagement of community members in formal
organizations, including religious congregations and self-help groups.
following natural disasters since no two areas are alike in their capacities to
sustain and recover from future disasters (Burby et al. 2000). Institutional
resilience variables include the percentage of the population covered by a recent
hazard mitigation plan, the percent of the population residing in Storm Ready
communities, and the number of governments and special districts per county (a
measure of political fragmentation).
9Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
involved reversing the order of their contribution to the overall resilience index
before the rescaling process could take place. This was done by taking the inverse
of the observation and then rescaling the variable, so that zero equals low
resilience and one represents high resilience. Our net international migration
variable provides an example of this form of inverse scaling since a high net
international migration reduces resilience. When scaled by taking the inverse of
the value, the highest value for international migration within the dataset receives
a score of zero and the lowest value for that variable receives a score of one. In
other words, communities with a large influx of recent international immigrants
are less resilient than those without.
After normalizing the variables for cross county comparisons, we
employed a method of aggregation in which our final disaster resilience score
represents the summation of the equally weighted average sub-index scores. In
other words, the variable scores in each sub-index were averaged to reduce the
influence of the different number of variables in each sub-index. These arithmetic
mean scores resulted in a sub-index score for each county, and then these sub-
index scores were summed to produce a final composite resilience score. Since
there are five sub-indices, scores range between zero and five (0 being the least
and 5 being the most resilient). We chose an equally weighted index at both the
sub-index and composite indicator level for two reasons. First, this simple
method of aggregation is transparent and easy to understand, a criteria we deemed
important for potential users. Second, we find no theoretical or practical
justification for the differential allocation of importance across indicators. While
methods exist for determining weights that are subjective or data reliant, such
weighting schemes do not always reflect the priorities of decision makers (Esty et
al. 2005).
Data aggregation and weighting
Once selected, the variables were normalized using a Min-Max rescaling scheme
to create a set of indicators on a similar measurement scale. Min-Max rescaling is
a method in which each variable is decomposed into an identical range between
zero and one (a score of 0 being the worst rank for a specific indicator and a score
of 1 being the best). All other values were scaled in between the minimum and
maximum values. This scaling procedure subtracted the minimum value and
divided by the range of the indicator values. For some variables in which high
values corresponded to low levels of resilience, our rescaling process also
10 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Figure 2: Spatial distribution of disaster resilience for FEMA region IV
Results
The disaster resilience scores provide a comparative assessment of community
resilience for the 736 counties within FEMA Region IV. We provide two
approaches to the results. The first is a spatial assessment and the second
provides an empirical ranking of the most and least resilient counties. Figure 2 is
a spatial representation of the disaster resilience within the study region. Our
scores, mapped as standard deviations from the mean, highlight those counties
that are ranking exceptionally well or poor in terms of their disaster resilience.
Counties symbolized in dark blue are highly resilient whereas counties
symbolized in red are the least resilient.
11Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
When visualized in the form of a regional map, geographic variations are
evident. The results illustrate an urban-rural bias, where metropolitan areas such
as Louisville, Nashville, Atlanta, Birmingham, Charlotte, and Tampa-St.
Petersburg show comparatively high levels of resilience. The notable exception to
this pattern is the absence of Miami, which shows only moderate levels of
resilience. The rural counties within our analysis typically demonstrate moderate
to low levels of resilience. This pattern is not only evident across the entire
Southeast, but is also present within each state individually.
To determine some of the underlying driving factors that contribute to this
trend, we delineated the social, economic, institutional, infrastructure, and
community capital components of the disaster resilience index (Figure 3). Several
spatial patterns are noteworthy. First, counties that rank high in social resilience
(Figure 3a) tend to cluster within or very near to medium/large metropolitan
areas, with the exceptions of central and south Florida cities. The economic
resilience component (Figure 3b) displays a slightly different distribution with
higher levels of economic resilience concentrated in inland counties (Ft.
Lauderdale and Jacksonville are exceptions), particularly along the I-85 and I-20
corridors stretching from Birmingham to Raleigh. The economic resilience found
in the capital cities (Jackson, MS) and in major industrial and tourist hubs
(Memphis, Nashville) are clear as well. The institutional component of disaster
resilience (Figure 3c) diverges from the urban-rural pattern, however. Here,
nearly all counties within Florida are highly resilient based on this measure,
which reflects prior disaster experience and the adoption of mitigation measures
such as flood insurance or community participation in Storm Ready programs.
Most of the coastal counties in the region also display moderate to high levels of
institutional resilience. Also notable is the low level of institutional resilience in
the Appalachian region of western Tennessee.
12 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Figure 3: Subcomponents of disaster resilience for FEMA Region IV: A) social
resilience, B) economic resilience, C) institutional resilience, D) infrastructure
resilience, E) community capital
13Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Table 2 provides a ranking of the 10 most and least resilient counties. The
places with the highest disaster resilience are in four different states and include
counties within the metropolitan statistical areas of Louisville, Birmingham,
Nashville, and Tampa-St. Petersburg (Table 2). The resilience scores within the
counties ranking highest in resilience are primarily the result of comparatively
high rankings within the social, economic, and institutional subcomponents of the
resilience index. A high degree of social homogeneity, diverse economies with
elevated levels of property ownership, high employment rates, and the
institutional capacities to mitigate the effects of natural disasters are all attributes
found in these urban areas.
Hancock, Chattahoochee, Quitman, and Echols County, Georgia; as well
as Issaquena County, Mississippi are the five least resilient counties within FEMA
Region IV. With the exception of Chattahoochee (which is part of the Columbus,
GA-AL metropolitan area, these counties are all rural areas. The low resilience
scores here are a function of lower than average infrastructure and institutional
resilience, as well as lower scores on the remaining sub-indices (Table 2).
Table 2: Most and Least Resilient Counties
Rank County
Resilience
Score Social Economic
Insti-
tutional
Infra-
structure
Community
Capital
Most Resilient
1 Fayette, KY 3.086 0.751 0.637 0.711 0.474 0.513
2 Jefferson, AL 3.069 0.765 0.625 0.706 0.445 0.528
3 Davidson, TN 3.063 0.765 0.619 0.700 0.487 0.491
4 Pinellas, FL 3.034 0.715 0.567 0.711 0.646 0.395
5 Williamson, TN 3.023 0.835 0.640 0.671 0.316 0.560
6 Durham, NC 2.995 0.733 0.661 0.712 0.405 0.484
7 Fulton, GA 2.983 0.681 0.587 0.667 0.562 0.487
8 Forsyth, NC 2.968 0.774 0.618 0.702 0.395 0.480
9 Franklin, KY 2.957 0.788 0.534 0.667 0.350 0.619
10 Daviess, KY 2.949 0.787 0.587 0.698 0.353 0.524
Least Resilient
1 Hancock, GA 1.608 0.485 0.453 0.154 0.173 0.343
2 Chattahoochee, GA 1.630 0.749 0.304 0.149 0.213 0.215
3 Quitman, GA 1.672 0.493 0.344 0.421 0.086 0.328
4 Echols, GA 1.696 0.620 0.378 0.277 0.123 0.298
5 Issaquena, MS 1.737 0.463 0.385 0.274 0.243 0.371
6 Taliaferro, GA 1.737 0.487 0.457 0.164 0.220 0.410
7 DeSoto, FL 1.755 0.539 0.300 0.484 0.138 0.293
8 Noxubee, MS 1.771 0.476 0.477 0.173 0.201 0.444
9 Sharkey, MS 1.799 0.490 0.465 0.199 0.203 0.441
10 Grundy, TN 1.808 0.575 0.456 0.185 0.208 0.383
A different pattern emerges with the infrastructure component (Figure 3d)
which shows a north-south bias in which a large percentage of counties within
Kentucky, Tennessee, and North Carolina have moderate to high levels of
infrastructural resilience. This is partially the result fewer mobile homes, and
greater medical and sheltering capacity. It is also a function of the availability of
evacuation routes (highways). Finally, the community capital component shows a
slight western bias within the region. For example, a large number of counties in
central and western Kentucky have high levels of community capital (Figure 3e),
as do the western portions of Tennessee and most of Mississippi. Much of this is
attributed to place attachments and the role of civic organizations, social advocacy
organization, and religious adherents within these counties.
14 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Community Case Studies: Gulfport-Biloxi, Charleston, Memphis
To illustrate the scalability of the resilience metric and to articulate the
significance of a component-based approach for objective measurement, we
employed a case study by scaling our index from the entire Southeast region to
counties within three metropolitan statistical areas: Gulfport-Biloxi MS,
Charleston-North Charleston, SC, and Memphis, TN-MS. These three
metropolitan areas span thirteen counties, and they are highly diverse in terms of
population, income, racial and ethnic identity, and age. We chose these particular
urban areas for a case study since they are primary test beds of the Community
and Regional Resilience Institute (CARRI). CARRI is a major interdisciplinary
research initiative supported by the U.S. Department of Homeland Security and
operated by the U.S. Department of Energy’s Oak Ridge National Laboratory (see
http://www.resilientus.org).
Table 3 presents the delineation of resilience scores for the three
metropolitan statistical areas, and Figure 4 shows their spatial distribution. When
averaging the composite scores for each test bed, Charleston has the highest
overall resilience closely followed by Memphis and Gulfport-Biloxi. However,
when you disaggregate the scores by individual county (Figure 4 and Table 3), the
driving forces of the composite scores become clear. The primacy of Charleston
as the most resilient of the three, for example, is a function of moderate to high
levels of resilience among all of its counties, while for Gulfport-Biloxi, only one
county, Harrison, has an overall score within the moderate to high level. There is
a mixed pattern for the Memphis metro area, which includes counties with very
high resilience (Shelby) as well as very low resilience (Tunica), and everything in
between.
In examining the individual county profiles across the test beds some
interesting findings also appear. Shelby County, TN; Charleston County, SC;
DeSoto and Harrison Counties, MS rank highest in terms of their overall disaster
resilience. Within this context, the scores for these counties do not diverge
significantly, yet the contributions of each of the subcomponents vary
considerably between these places. DeSoto County’s overall score, for instance, is
primarily the product of its high rank for social resilience (social resilience score
= 0.913) whereas Shelby County’s score is primarily a function of infrastructural
and economic resilience (infrastructure resilience = 0.845; economic resilience =
0.716).
To increase the resilience ranking of DeSoto County to a position similar
to that of Shelby County, disaster planners and decision makers could concentrate
on components other than social, which is already quite good. Striving to increase
15Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Table 3: Resilience of Gulfport-Biloxi, Charleston-North Charleston, and
Memphis Metropolitan Statistical Areas
Resilience Type
Social Economic Institu-
tional
Infra-
structure
Community
Capital
Resilience
Score
Gulfport-Biloxi MSA
2.271
Hancock County, MS 0.498 0.409 0.499 0.203 0.499 2.108
Harrison County, MS 0.527 0.552 0.579 0.504 0.660 2.821
Stone County, MS 0.440 0.324 0.324 0.258 0.537 1.883
Charleston-North
Charleston MSA
2.583
Berkeley County, SC 0.657 0.495 0.554 0.073 0.446 2.216
Charleston County, SC 0.500 0.648 0.702 0.519 0.722 3.091
Dorchester County, SC 0.743 0.549 0.423 0.141 0.586 2.442
Memphis MSA
2.330
DeSoto County, MS 0.913 0.557 0.456 0.328 0.501 2.754
Marshall County, MS 0.356 0.537 0.325 0.185 0.351 1.753
Tate County, MS 0.573 0.471 0.287 0.197 0.502 2.031
Tunica County, MS 0.252 0.400 0.401 0.167 0.308 1.528
Fayette, County TN 0.622 0.469 0.429 0.289 0.618 2.427
Shelby County, TN 0.582 0.716 0.499 0.845 0.534 3.175
Tipton County, TN 0.750 0.536 0.527 0.300 0.532 2.646
Figure 4: Spatial patterns of disaster resilience indicators for
Gulfport-Biloxi, Charleston, and Memphis metro areas
16 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
DeSoto County’s rank in infrastructural resilience provides the most fruitful
opportunity, yet increases in institutional resilience and community capital may
also provide valid starting points. Utilizing such an approach may not only
provide officials in the county with a means to benchmark their scores against
other counties, it also provides a basis for investment prioritization, and a method
for tracking the resilience of a county relative to other counties’ scores over time.
Discussion and Conclusion
This paper provides a first-attempt in developing replicable and robust baseline
indicators for measuring and monitoring the disaster resilience of places. Because
the science of resilience is still in its infancy, incremental empirical developments
such as these are necessary to 1) advance our understanding of the multi-
dimensional nature of resilience and its constituent parts, but more importantly to
2) provide metrics that are easily understood and applicable to the decision
making process. Once established, the baseline resilience indicators for
communities (or BRIC for short) provide a useful way to examining not only the
composite score when compared to other places such as the comparisons between
the three metro areas (inter-metro), but it also enables comparisons within each
metro area (intra-metro). While regional authorities may be more interested in the
overall patterns and look for intervention strategies that improve the entire region,
local leadership (e.g. county level officials) might be more interested in
intervention opportunities at the county scale, so they would be more inclined to
only examine county level scores.
The efficacy of the baseline resilience indicators to scale up from the
county, to the metro area, to the state, to the region, and to the nation is one of the
greatest strengths of this approach from a public policy perspective. The other
beneficial outcome of this baseline resilience index (BRIC) is the visualization of
the results, which provides a quick comparative overview of where improvements
in baseline indicators of resilience are most needed. More significantly, these
baseline indicators (thanks to their sub-index type of construction methodology)
identify which category of intervention (social, economic, infrastructure,
institutional, community capacity) would provide overall improvement in the
score.
Since BRIC is one of the first empirically based disaster resilience indices,
it is not without shortcomings. The most significant is the reliance on national
data sources, which are often out of date and inadequate to the task of
characterizing local circumstances. While local data could be used, such data
would not be comparable or always available across regions. However, if a
national effort to obtain resilience indicators data were undertaken, we would
suggest a number of important items for inclusion. The most underdeveloped
17Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
sub-index is community capacity, so measures of volunteerism at the county level
(such as number of people involved in parent-teacher organizations, youth groups,
etc.); number of community based organizations; level of innovation within the
community; composition and activity level of Community Emergency Response
Teams ( CERTs) would be useful. In addition, a consistent metric for employment
in the tourism sector (by county) and building permits and building code data at
the county scale would also be useful.
We suggest that the baseline indicators provide the first “broad brush” of
the patterns of disaster resilience within and between places and the underlying
factors contributing to it. A second step is a more detailed analysis within
jurisdictions to assess place-specific capacities in each of these areas (social,
economic, institutional, infrastructure, community) and the development of fine-
tuned and local appropriate mechanisms for enhancing disaster resilience. This
baseline resilience index for communities (BRIC) can help in initiating research
interest, community discussions, and for attracting public interest and local
concern for fostering disaster resilient communities.
References
Adger, W.N. 2000. "Social and Ecological Resilience: Are They Related?"
Progress in Human Geography 24: 347-63.
Adger, W.N., N. Brooks, G. Bentham, M. Agnew and S. Eriksen. 2004. New
Indicators of Vulnerability and Adaptive Capacity. Norwich, UK:
Tyndall Centre for Climate Change Research, Technical Report 7.
Auf der Heide, E. and J. Scanlon. 2007. "Health and Medical Preparedness and
Response, " In Emergency Management: Principles and Practice for
Local Government, edited by W.L. Waugh and K. Tierney. Washington,
D.C.: International City Managers Association, pp. 183-206.
Berke, P.R. and T.J. Campanella. 2006. "Planning for Postdisaster Resiliency."
Annals of the American Academy of Political and Social Science 604: 192-
207.
Boruff, B.J., C.T. Emrich and S.L. Cutter. 2005. "Erosion Hazard Vulnerability of
US Coastal Counties." Journal of Coastal Research 21(5): 932-42.
Briguglio, L. 1995. "Small Island State and Their Vulnerable Economies." World
Development 23: 1615-32.
Bruneau, M., S.E. Chang, R.T. Eguchi, G.C. Lee, T.D. O'Rourke, A. M.
Reinhorn, M. Shinozuka, K.T. Tierney, W.A. Wallace and D. von
Winterfeldt. 2003. "A Framework to Quantitatively Assess and Enhance
the Seismic Resilience of Communities." Earthquake Spectra 19(4): 733-
52.
18 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Burby, R.J., R.E. Deyle, D.R. Godschalk and R.B. Olshansky. 2000. "Creating
Hazard Resilient Communities through Land-Use Planning." Natural
Hazards Review 2(1): 99-106.
Cardona, O.D. 2005. Indicators of Disaster Risk and Risk Management.
Summary Report. Washington D.C.: Inter-American Development Bank.
Colten, C.E., R.W. Kates and S.B. Laska. 2008. "Community Resilience: Lessons
from New Orleans and Hurricane Katrina." CARRI Research Report 3.
Oak Ridge: Community and Regional Resilience Institute.
http://www.resilientus.org/library/FINAL_COLTEN_9-25-
08_1223482263.pdf
Cutter, S.L., L. Barnes, M. Berry, C.G. Burton, E. Evans, E.C. Tate and J. Webb.
2008a. "Community and Regional Resilience: Perspectives from Hazards,
Disasters, and Emergency Management." CARRI Research Report 1. Oak
Ridge: Community and Regional Resilience Institute,
http://www.resilientus.org/library/FINAL_CUTTER_9-25-
08_1223482309.pdf
———. 2008b. "A Place-Based Model for Understanding Community Resilience
to Natural Disasters." Global Environmental Change 18: 598-606.
Cutter, S.L., B.J Boruff and W.L. Shirley. 2003. "Social Vulnerability to
Environmental Hazards." Social Science Quarterly 84(1): 242-61.
Easter, C. 1999. "Small States Development: A Commonwealth Vulnerability
Index." The Round Table 351: 403-22.
Esty, D.C., M. Levy, T. Srebotnjak and A. de Sherbinin. 2005. Environmental
Sustainability Index: Benchmarking National Environmental
Stewardship. New Haven: Yale Center for Environmental Law & Policy.
Florida, R. 2002. "The Economic Geography of Talent." Annals of the
Association of American Geographers 92(4): 743-55.
Freudenberg, M. 2003. Composite Indicators of Country Performance: A Critical
Assessment. Paris, France: OECD Publishing.
Godschalk, D.R. 2007. "Functions and Phases of Emergency Management." In
Emergency Management: Principles and Practice for Local Government,
edited by W.L. Waugh and K. Tierney. Washington D.C.: International
City Managers Association, pp. 87-112.
———. 2003. "Urban Hazard Mitigation: Creating Resilient Cities." Natural
Hazards Review 4: 136-43.
Gommes, R., J. Du Guerney, F. Nachtergaele and R. Brinkman. 1998. Potential
Impacts of Sea-Level Rise on Agriculture. Rome, Italy: Food and
Agricultural Organization of the United Nations.
Goodman, R., M.A. Speers, K. McLeroy, S. Fawcett, M. Kegler, E. Parker, S. R.
Smith, T.D. Sterling and N. Wallerstein. 1998. "Identifying and Defining
the Dimensions of Community Capacity to Provide a Basis for
Measurement." Health Education & Behavior 25: 258-78.
19Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Gunderson, L. 2009. “Comparing Ecological and Human Community Resilience.”
CARRI Research Report 5. Oak Ridge: Community and Regional
Resilience Institute.
http://www.resilientus.org/library/Final_Gunderson_1-12-
09_1231774754.pdf
H. John Heinz III Center for Science Economics, and the Environment. 2002.
Human Links to Coastal Disasters. Washington, D.C.: H. John Heinz
Center.
———. 2008. The State of the Nation's Ecosystems: Measuring the Land, Waters,
and Living Resources of the United States. Washington D. C.: Island
Press.
Kahan, J.H., A.C. Allen, and J.K. George. 2009. "An Operational Framework for
Resilience." Journal of Homeland Security and Emergency Management
6(1): Article 83.
Kaly, U., C. Pratt, J. Mitchell and R. Howorth. 2003. The Demonstration
Environmental Vulnerability Index (EVI). Suva, Fiji: South Pacific
Applied Geoscience Commission (SOPAC).
http://www.sidsnet.org/docshare/other/20030910165813_Demo_EVI_Rep
ort_2003.pdf
Kaly, U., C. Pratt, and J. Mitchell. 2004. The Environmental Vulnerability Index
(EVI). Suva, Fiji: South Pacific Applied Geoscience Commission
(SOPAC).
http://www.vulnerabilityindex.net/Files/EVI%202004%20Technical%20R
eport.pdf
Klein, R., J.T. Robert, J. Nicholls and F. Thomalla. 2003. "Resilience to Natural
Hazards: How Useful Is This Concept?" Environmental Hazards 5(1-2):
35-45.
Manyena, S.B. 2006. "The Concept of Resilience Revisited." Disasters 30(4):
433-50.
Mileti, Dennis S. 1999. Disasters by Design: A Reassessment of Natural Hazards
in the United States, Natural Hazards and Disasters. Washington, D.C.:
Joseph Henry Press.
Millennium Change Corporation (MCC). 2007. "Indicators."
http://www.mcc.gov/mcc/selection/indicators/index.shtml
.
Murphy, B. 2007. “Locating Social Capital in Resilient Community-Level
Emergency Management.” Natural Hazards 41: 297-315.
Morrow, B. 2008. "Community Resilience: A Social Justice Perspective." CARRI
Research Report 4. Oak Ridge: Community and Regional Resilience
Institute. http://www.resilientus.org/library/FINAL_MORROW_9-25-
08_1223482348.pdf
20 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
Nardo, M., M. Saisana, A. Saltelli and S. Tarantola. 2008. Handbook on
Constructing Composite Indicators: Methodology and User Guide. Paris,
France: OECD Publishing.
National Research Council (NRC). 2000. Ecological Indicators for the Nation.
Washington D.C.: National Academy Press.
———. 2006. Facing Hazards and Disasters: Understanding Human
Dimensions. Washington, DC: National Academy Press.
______. 2010. Private-Public Sector Collaboration to Enhance Community
Disaster Resilience: A Workshop Report. Washington, D.C.: National
Academy Press.
Neumayer, E. 2001. "The Human Development Index and Sustainability: A
Constructive Proposal." Ecological Economics 39: 101-14.
Norris, F.H., S.P. Stevens, B. Pfefferbaum, K.F. Wyche and R. L. Pfefferbaum.
2008. "Community Resilience as a Metaphor, Theory, Set of Capacities,
and Strategy for Disaster Readiness." American Journal of Community
Psychology 41: 127-150.
Organization for Economic Cooperation and Development (OECD). 2001. OECD
Environmental Indicators. Paris, France: OECD Publishing.
Oxfam America. 2009. Exposed: Social Vulnerability and Climate Change in the
US Southeast. Boston, MA: Oxfam America Inc.
Perkins, D., J. Hughey and P. Speer. 2002. "Community Psychology Perspectives
on Social Capital Theory and Community Development Practice." Journal
of the Community Development Society 33: 33-52.
Pethick, J. and S. Crooks. 2000. "Development of a Coastal Vulnerability Index:
A Geomorphological Perspective." Environmental Conservation 27: 359-
67.
Prescott-Allen, R. 2001. The Wellbeing of Nations: A Country-by-Country Index
of Quality of Life and the Environment. Washington D.C.: Island Press.
Ronan, K.R. and D.M. Johnston. 2005. Promoting Community Resilience in
Disaster: The Role for Schools, Youth, and Families. New York:
Springer.
Rose, A. 2007. "Economic Resilience to Natural and Man-Made Disasters:
Multidisciplinary Origins and Contextual Dimensions." Environmental
Hazards 7(4): 383-95.
Saisana, M. and F. Cartwright. 2007. “Composite Indicators: Science or
Artifacts?" 2007 Biannual Conference, European Survey Research
Association. Prague, Czech Republic.
http://composite-indicators.jrc.ec.europa.eu/Document/Composite%20indi
cators-%20Science%20or%20Artifacts.pdf
21Cutter et al.: Disaster Resilience Indicators
Published by The Berkeley Electronic Press, 2010
Schneiderbauer, S. and D. Ehrlich. 2006. "Social Levels and Hazard (in)-
Dependence.” In Measuring Vulnerability to Natural Hazards: Towards
Disaster Resilient Societies, edited by J. Birkmann. Tokyo, Japan: United
Nations University Press, pp. 78-102.
Subcommittee on Disaster Reduction (SDR). 2005. Grand Challenges for
Disaster Reduction. Washington D.C.: National Science and Technology
Council.
Sylves, R. 2007. "Budgeting for Local Emergency Management and Homeland
Security.” In Emergency Management: Principles and Practice for Local
Government, edited by W.L. Waugh and K. Tierney. Washington, D.C.:
International City Managers Association, pp. 183-206.
Tierney, K. 2009. "Disaster Response: Research Findings and Their Implications
for Resilience Measures." CARRI Research Report 6. Oak Ridge:
Community and Regional Resilience Institute.
http://www.resilientus.org/library/Final_Tierney2_dpsbjs_1238179110.pd
f
Tierney, K.J., M.K. Lindell and R.W. Perry. 2001. Facing the Unexpected:
Disaster Preparedness and Response in the United States, Natural
Hazards and Disasters. Washington, D.C.: Joseph Henry Press.
United Nations Development Programme (UNDP). 1990. Human Development
Report 1990. New York: Oxford University Press.
———. 2005. Human Development Report 2005. New York: United Nations
Development Programme.
———. 2004. Reducing Disaster Risk: A Challenge for Development. New York:
John S. Swift Co.
Vale, L.J. and T.J. Campanella. 2005. The Resilient City: How Modern Cities
Recover from Disaster. New York: Oxford University Press.
Vincent, K. 2004. Creating an Index of Social Vulnerability to Climate Change
for Africa. Norwich, UK: Tyndall Centre for Climatic Change Research.
Wheeler, J.O. 1999. "The American South." In North America: A Geographical
Mosaic, edited by F.W. Boal and S.A. Royle. London, UK: Arnold, pp.
265-276.
22 JHSEM: Vol. 7 [2010], No. 1, Article 51
http://www.bepress.com/jhsem/vol7/iss1/51
