Citation
Evaluating local hazard mitigation planning

Material Information

Title:
Evaluating local hazard mitigation planning quality, public participation, and soc ail vulnerability
Creator:
Janes, Enessa Holland ( author )
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English
Physical Description:
electronic file (279 pages). : ;

Subjects

Subjects / Keywords:
Hazard mitigation -- Planning -- United States ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Despite abundant amounts of data and significant advances in communications and engineering technology, global trends in hazard vulnerability and disaster losses continue to rise. Moreover, the most marginalized and disadvantaged members of society bear the brunt of these losses and impacts. The Untiled States emergency management community has responded to increasing trends in disaster losses by underscoring the importance of hazard mitigation at the local scale. Although the planning literature acknowledges that plan quality is a key factor in determining planning outcomes, and that engaging local community stakeholders in the planning process is critical to creating high-quality mitigation plans, little empirical research has been done to evaluate the nuances of the local hazard mitigation planning process. Moreover the literature has not evaluated how social vulnerability factors affect mitigation planning processes and plan quality outcomes. This study examines local hazard mitigation plan quality, models of participatory planning, and social vulnerability in the northern California Bay-Delta region. It applies a mixed methods approach to identify relationships between local social vulnerability, public participation, and plan quality outcomes in 34 communities. The results of this study show that neither public participation nor social vulnerability was associated with plan quality outcomes in the study region. Additionally, social vulnerability had no relationship to public participation levels. Planning scale emerged as an important driver of plan quality and single-jurisdiction planning efforts were positively correlated with plan quality outcomes. The case studies provide further insight into these findings and relationships and present a best-practices model for knowledge sharing and decision-making for local hazard mitigation planning.
Thesis:
Thesis (Ph.D.)--University of Colorado Denver. Design and planning
Bibliography:
Includes bibliographic references.
System Details:
System requirements: Adobe Reader.
General Note:
College of Architecture and Planning
Statement of Responsibility:
by Enessa Holland Janes.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
891588483 ( OCLC )
ocn891588483

Downloads

This item has the following downloads:


Full Text
EVALUATING LOCAL HAZARD MITIGATION PLANNING: QUALITY, PUBLIC
PARTICIPATION, AND SOCIAL VULNERABILITY
by
ENESSA HOLLAND JANES
B.S., Stanford University, 2005
M.S., Stanford University, 2006
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Design and Planning
2013
i


2013
ENESSA HOLLAND JANES
ALL RIGHTS RESERVED


This thesis for the Doctor of Philosophy degree by
Enessa Holland Janes
has been approved for
Design and Planning Program
by
Deborah Thomas, Chair
Jeremy Nemeth, Advisor
Brian Muller
Bruce Goldstein
Gregory Thomas
November 8, 2013
iii


Janes, Enessa Holland (Ph.D., Design and Planning)
Evaluating Local Hazard Mitigation Planning: Quality, Public Participation, and Social
Vulnerability
Thesis directed by Associate Professor Jeremy Nemeth
ABSTRACT
Despite abundant amounts of data and significant advances in communications
and engineering technology, global trends in hazard vulnerability and disaster losses
continue to rise. Moreover, the most marginalized and disadvantaged members of society
bear the brunt of these losses and impacts.
The United States emergency management community has responded to
increasing trends in disaster losses by underscoring the importance of hazard mitigation
at the local scale. Although the planning literature acknowledges that plan quality is a key
factor in determining planning outcomes, and that engaging local community
stakeholders in the planning process is critical to creating high-quality mitigation plans,
little empirical research has been done to evaluate the nuances of the local hazard
mitigation planning process. Moreover, the literature has not evaluated how social
vulnerability factors affect mitigation planning processes and plan quality outcomes.
This study examines local hazard mitigation plan quality, models of participatory
planning, and social vulnerability in the northern California Bay-Delta region. It applies a
mixed methods approach to identify relationships between local social vulnerability,
public participation, and plan quality outcomes in 34 communities.
The results of this study show that neither public participation nor social
vulnerability was associated with plan quality outcomes in the study region. Additionally,
IV


social vulnerability had no relationship to public participation levels. Planning scale
emerged as an important driver of plan quality and single-jurisdiction planning efforts
were positively correlated with plan quality outcomes. The case studies provide further
insight into these findings and relationships and present a best-practices model for
knowledge sharing and decision-making for local hazard mitigation planning.
The form and content of this abstract are approved. I recommend its publication.
Approved: Jeremy Nemeth


DEDICATION
I dedicate this work to my grandmother, Doris Janes, the strongest and most
inspiring woman I know. Her contagious positivity, warm advice, and late-night phone
calls made this journey a success.
vi


ACKNOWLEDGMENTS
I would like to thank my advisor, Jeremy Nemeth, for his enthusiasm, support,
and resolute dedication to the strength of my research. I would also like to thank my
committee chair, Deborah Thomas, whose selfless time and care kept me on-track and
motivated. Thank you, Deb, for sharing your intellectual passions with me and for your
exemplary teaching and mentorship. It is such an honor to call you my advisor and my
friend.
I would also like to thank the other members of my dissertation committee -
Bruce Goldstein, Brian Muller, and Gregory Simon for their ongoing encouragement
and constructive comments and criticisms.
Thank you to the National Science Foundation, whose generous support through
the IGERT Program in Sustainable Urban Infrastructure made this research possible.
I extend my thanks to Greg Kinney for his support during the final stages of my
doctoral marathon. Thank you for generously and enthusiastically trading your
motorcycle time for afternoons with me at your Anschutz office.
I am grateful to my family my father, Craig, my mother, Kitty, and my sister,
Janice for their steadfast confidence in me and for serving as examples of wise, well-
balanced individuals. Thank you for reminding me of what is truly important in life.
Finally, I would like to thank my husband, David Belsheim, for his
encouragement, humor, and unwavering patience. Thank you for making me laugh when
times were tough and for pushing me to see every challenge as an opportunity. I love you
forever.
Vll


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION..........................................................................1
The Hazards and Disasters Problem...................................................1
Local Planning for Disaster Vulnerability Reduction.................................2
Research Questions..................................................................5
Research Impacts....................................................................7
II. LITERATURE REVIEW....................................................................8
Introduction........................................................................8
Hazards, Disasters, and Vulnerability...............................................8
Measuring Social Vulnerability.....................................................13
Building Resilience into Emergency Management......................................20
Hazard Mitigation: Planning, Policy and Practice...................................29
The Comprehensive City Plan....................................................29
Hazard Mitigation in the United States.........................................36
Vulnerability Reduction through Hazard Mitigation..............................39
Assessing Vulnerability........................................................41
Engaging the Public through Participatory Planning.................................44
The Theory and Practice of Communicative Planning..............................46
Participatory Planning: Practice and Theory....................................49
Local Knowledge................................................................54
Better Plans, Safer Communities?...................................................58
The Influence of Participation on Implementation and Plan Quality..............59
Internal Implementation Factors: The Plan and Its Quality......................59
Assessing Plan Quality.........................................................62
viii


Linking Participatory Planning with Hazard Mitigation Plan Quality.............65
Placement and Contribution of this Study..........................................68
Conclusion........................................................................74
III. METHODS............................................................................76
Introduction......................................................................76
Study Area........................................................................77
Sample Selection...............................................................78
Data Collection................................................................83
Stage 1: Plan Quality Evaluation..................................................85
Methodology....................................................................85
Assessing Models of Community Participation....................................88
Stage 2: Principal Component and Regression Analysis..............................92
Stage 3: Interviews and Qualitative Analysis......................................94
Interview Sampling Strategy....................................................94
Interviewee Selection..........................................................95
Interview Data Collection and Storage..........................................97
Reflexive Journal..............................................................98
Data Analysis and Validity........................................................99
Plan Quality Analysis..........................................................99
Qualitative Interview Analysis................................................100
Validity......................................................................100
IV. EVALUATING THE RELATIONSHIPS BETWEEN PARTICIPATION,
COMMUNITY CHARACTERISTICS, AND LOCAL HAZARD MITIGATION
PLAN QUALITY.......................................................................102
Introduction.....................................................................102
Methods..........................................................................102
IX


Plan Quality Index...................................................................102
Methodology for Assessing Models of Community Participation..........................110
Data.....................................................................................113
Plan Quality and Levels of Public Participation......................................113
Community Characteristic Data........................................................117
Data Reduction Using Principal Components Analysis.......................................127
Multiple Regression Analysis.............................................................143
Results and Discussion...................................................................157
V. A QUALITATIVE EVALUATION OF LOCAL MITIGATION PLANNING.......................................162
Introduction.............................................................................162
Methods..................................................................................163
Interview Guides.....................................................................163
Data Collection and Storage..........................................................164
Reflexive Journal....................................................................165
Case Selection.......................................................................165
Interviewee Selection................................................................170
Data and Limitations.................................................................172
Planning Scale and Participation during Local Hazard Mitigation Planning: Six
Case Studies.............................................................................173
Planning Scale: Single and Multi-Jurisdiction Local Hazard Mitigation Plans..........177
Regional Scale Multi-Jurisdiction Plan: Windsor, CA..................................178
The Plan.............................................................................178
County Scale Multi-Jurisdiction Plan: Citrus Heights, CA.............................182
The Plan.............................................................................183
City Scale Single-Jurisdiction Plan: Roseville, CA...................................187
The Plan.............................................................................189
x


Results......................................................................195
Public Participation during Local Hazard Mitigation Planning.................197
Low Public Participation: Woodland and West Sacramento, CA...................198
The Plan.....................................................................199
High Public Participation: Napa, CA..........................................202
The Plan.....................................................................203
Results......................................................................207
Discussion.......................................................................210
VI. CONCLUSIONS AND DISCUSSION........................................................217
Introduction.....................................................................217
Building Disaster Resilience through Local Hazard Mitigation Planning............217
Research Questions...............................................................220
Community Case Studies...........................................................224
Contributions and Directions for Future Research.................................229
REFERENCES............................................................................235
APPENDIX..............................................................................256
xi


LIST OF TABLES
Table
1: Definitions of Resilience..............................................................23
2: The Role of the Federal Government in Disaster Loss Reduction..........................37
3: Models of U.S. Hazard Vulnerability Assessment.........................................43
4: Sample Cities..........................................................................81
5: Models of Community Participation in Hazard Mitigation Planning........................91
6: Community Characteristic Variables.....................................................93
7: Summary of Regression Models...........................................................94
8: Naturalistic Techniques for Ensuring Comprehensive Research Quality...................101
9: Plan Quality Index....................................................................105
10: Evaluation Criteria for Models of Community Participation...........................Ill
11 .Cities Ranked by Plan Quality Score.................................................114
12: Descriptive Statistics...............................................................116
13: Community Characteristic Variables...................................................118
14: Descriptive Statistics for Hazard Exposure Data......................................121
15: Summary of Multi-Hazard Exposure and Losses by City..................................122
16: Summary of Community Characteristic Variables........................................125
17: Descriptive Statistics...............................................................128
18: Summary of PCA Variables.............................................................130
19: Principal Component Analysis Runs....................................................132
20: Total variance explained.............................................................134
21: Varimax-Rotated Component Matrix.....................................................135
22: Summary of PCA results...............................................................136
23: Sample City Factor Loadings..........................................................138
xii


24: Moments..........................................................................144
25: Basic Statistical Measures.......................................................144
26: Tests for Location...............................................................145
27: Tests for Normality..............................................................145
28: Summary of Regression Models.....................................................147
29: Model 1..........................................................................148
30: Model 2..........................................................................149
31: Model 3..........................................................................150
32: Model 4..........................................................................150
33: Model 5. Analysis of Maximum Likelihood Estimates................................151
34: Model 6..........................................................................152
35: Model 7..........................................................................153
36: Model 8..........................................................................154
37: Model 9..........................................................................154
38: Model 10.........................................................................155
39: Does Planning Scale Affect Fact Base Scores?.....................................156
40: Fact Base........................................................................156
41: Case Study Sampling Based on Fact Base Scores....................................167
42: Final Case Study Selections......................................................168
43: Summary Characteristics of Selected Case Study Communities.......................170
44: Distribution of Participant Characteristics by Community.........................175
xiii


LIST OF FIGURES
Figure
1: The Hazards of Place Model.....................................................................16
2: The Planner's Triangle..........................................................................28
3: Place Making and Conflict Resolution through Planning..........................................33
5: Amstein's Ladder of Citizen Participation......................................................52
6: Framework for Stakeholder Participation During Local Hazard Mitigation Planning................72
7: Map of California Counties Included in Study Area..............................................79
8: Methodological Framework........................................................................83
9: Models of Stakeholder Participation During Local Hazard Mitigation Planning....................96
10: California State Hazard Exposure Map..........................................................121
11: Scree Plot of Eigenvalues of Varimax Rotation.................................................133
12: Normal Probability Plot.......................................................................146
13: Map of Case Study Communities.................................................................169
14: Models of Stakeholder Participation During Local Hazard Mitigation Planning...................171
15: Revised Framework for Local Hazard Mitigation Planning........................................226
xiv


CHAPTERI
INTRODUCTION
The Hazards and Disasters Problem
When coupled with development in high-risk areas, differential vulnerabilities can lead to
catastrophic results. [...] What good is a federal response plan when it clearly does not
work and does not alleviate the suffering of the most vulnerable within our society? What
does it say about the adequacy of preparedness when we know so little about the most
disadvantaged within the communities?
Moral Hazard, Social Catastrophe (Cutter & Emrich, 2006, 102)
Currently, we are losing the battle for a safe and sustainable society in the United
States and throughout the world. Despite abundant amounts of data and significant
advances in communications and engineering technology, global trends in hazard
vulnerability and disaster losses continue to rise. Some estimates describe a three-fold
increase in disaster events over the last thirty years with associated economic damages
increasing by a factor of six (sources from EM-DAT, The World Bank, MunichRe;
Mileti, 1999; Birkmann, 2006). Moreover, projections based on climate change and
current land use and development patterns predict that communities, especially those
living in urbanized areas, are expected to experience increased exposure and vulnerability
to hazards from floods, storms, landslides, heat waves, droughts and other hazard events
over time (Huq, 2007; Bouwer, 2011).
The problem of increasing hazard vulnerability is more than strictly an economic
concern. It is more than an issue of sustainable infrastructure or natural security. It is, at
its core, an issue of social justice. In their study of the legacy of Hurricane Katrina, Cutter
and Emrich (2006) observed that the disaster vulnerability of the U.S. population is
1


unevenly distributed among social groups and between places. They describe a
phenomenon in which some communities suffer higher risks of hazard impacts than
others due to the characteristics of the people who live and work within their boundaries.
Cutter and Emrich (2006) describe Hurricane Katrina as an example of a social
catastrophe. Not only were socially and economically marginalized residents of New
Orleans disproportionately affected by the disaster, but the ability of people to recover
and rebuild their lives also varied widely depending on their economic status, race,
gender, age and education level (Phillips et al., 2009). Ultimately, the disaster event
created by the storm was driven by deep-seated moral hazards. It is this unjust reality that
motivates and frames my dissertation research.
The challenge is clear: a different approach to disaster management and planning
is necessary if we hope to remedy the crisis of differential local and regional hazard
impacts. In the context of disasters, communities are only as strong as their weakest link.
Therefore, in order to protect those who are most vulnerable, and to bolster local
resilience and adaptive capacity, we must shift away from a one-size-fits-all approach to
planning towards one that can be tailored to fit the unique social contexts of
communities.
Local Planning for Disaster Vulnerability Reduction
The United States emergency management community has responded to
increasing trends in disaster losses by underscoring the importance of coupling disaster
preparedness with hazard mitigation planning. The passing of the Robert T. Stafford
Disaster Relief and Emergency Assistance Act (42 U.S.C. 5121) in 1988 represented a
2


sea change in national emergency management priorities when it established pre-disaster
hazard mitigation planning as the key mechanism for combating increasing disaster
losses in the United States (Robert T. Stafford Disaster Relief and Emergency Act, P.L.
100-707, 42 U.S. Code sec. 5121 et seq.).
The fundamental principle behind hazard mitigation is that most of the damage
and loss caused by hazard events can be prevented. Although most hazard events cannot
be prevented from occurring altogether, their disastrous impacts on people, property, and
infrastructure can be diminished if proactive measures are taken to reduce risk and local
vulnerability (Comfort et al., 1999; White, et al. 2001; Godschalk, 2003). Hazard
mitigation planning involves pre-event hazard identification, risk and vulnerability
analysis, and the creation of feasible mitigation strategies to reduce or eliminate hazard
risks. Together, these efforts contribute to reduced vulnerability of infrastructure and
people to the impacts of hazards and helps reduce physical, economic and social losses
(Nelson & French, 2002; Comfort et al., 1999; White, et al. 2001; Godschalk, 2003;
Wisner et al., 2004).
The adoption of hazard mitigation planning as a pillar of the United States
emergency management strategy was coupled with a turn towards state and local-level
emergency management activity. Specifically, with the creation of FEMAs Project
Impact in 1997, and the passage of the Disaster Mitigation Act in 2000 (DMA 2000),
there is now a strong emphasis on loss and risk reduction through state, county, and
municipal hazard mitigation planning efforts.
Since the passage of DMA 2000, researchers have been working to identify the
impacts of local hazard mitigation planning efforts on levels of disaster risks, loss, and
3


vulnerability. Although measuring prevention is a great challenge, empirical research has
shown that higher quality hazard mitigation plans have greater outcomes in terms of
reduced vulnerability to hazards and building human resilience to disaster impacts (Burby
et al., 1999; Brody, 2003; Godschalk et al., 2009; Beatley, 2009; Peacock et al., 2008).
Additionally, the inclusion of local knowledge in hazard mitigation plans has been shown
to contribute positively to plan quality (Burby et al., 1999; Brody, 2003; Burby, 2001,
2003; Godschalk & Brody, 2003).
To be sure, the creation of high quality plans is not the end point in the
overarching goal of reduced social vulnerability, hazard impacts and loss. Rather, high
quality hazard mitigations plans are an important intermediary step of the disaster risk
reduction process of all communities. Not only has plan quality been shown to predict
implementation outcomes, it can be used as a parameter for assessing public engagement
and the incorporation of local community knowledge into decision making and policy.
Ultimately, we are interested in an extension of these outcomes, including social
vulnerability reduction. One of the key goals of this study is to contribute to an
understanding of how mitigation planning can play a role in reducing impacts to disasters
in socially vulnerable communities.
Planners and emergency managers have started to acknowledge that plan quality
is a critical issue and that engaging community stakeholders in the planning process is
key for creating usable, relevant, high-quality mitigation plans. If planners and
emergency managers intend to create high-quality plans that guide local government
behavior towards reduced vulnerability and increased resilience to disaster impacts, a key
question emerges: How is information about local vulnerability incorporated into
4


decision-making processes? How can planning be improved to encourage the integration
of hazard mitigation planning with local land use policy and community priorities?
While DMA 2000 and Project Impact both emphasize increased community
participation in the local hazard mitigation planning process, guidelines and standards for
collecting community-derived information, and strategies for incorporating this
information into planning and decision-making, are nearly nonexistent. There is great
need for a scalable planning approach that planners and communities can use as a tool to
create contextual, implementable local hazard mitigation plans. This research study
contributes to the planning literature by explaining what this approach to planning might
look like.
Research Questions
This study examines plan quality, models of participatory planning and social
vulnerability in the context of hazard mitigation planning in the Northern California Bay-
Delta region. California state legislation has shown tremendous support for local
participatory action in hazard mitigation planning. As an example of a state that has taken
a highly progressive, proactive stance towards local hazard mitigation planning, two
research questions can be raised about the effectiveness of Californias approach:
Research Question #1
What is the quality of local hazard mitigation plans created in the Northern
California Bay-Delta region and to what extent has the public participated in the plan-
making process?
5


Aim 1: To identify social vulnerability data in plans and incorporate that data
into plan quality assessment metrics
Aim 2: To assess the involvement of community stakeholders during the planning
process
Aim 3: To identify the participatory planning models employed by each city
during the local hazard mitigation planning process
Research Question #2
How do community characteristics and levels of public participation affect
mitigation plan quality outcomes?
Aim 4: To identify key relationships between local community characteristics
(including population size, wealth, diversity, social vulnerability characteristics)
and local plan quality scores
Aim 5: To analyze how community characteristics affect levels of public
participation in local hazard mitigation planning
Aim 6: To evaluate how public participation in planning affects local hazard
mitigation plan quality
This study follows a mixed-methods approach, consisting of quantitative plan
quality evaluation and the qualitative analysis of in-depth interviews. It evaluates how
different approaches to public participation affect the content and quality of local hazard
mitigation plans from 34 California cities and expands upon earlier work in the planning
and disaster literature by refining plan quality measures to include social vulnerability
factors.
6


Research Impacts
Currently, federal and state efforts to reduce hazard vulnerability by developing
high quality, locally relevant hazard mitigation plans continue to fall short (Battista &
Baas, 2004; Schwab, 2011; Stevens et al., 2010). A fundamental aim of this research
study is to contribute to improved hazard mitigation planning outcomes (including
increased project implementation and reduced local social vulnerability) by examining
the relationship between models of participatory planning, local community
characteristics and hazard mitigation plan quality.
Relative to the disaster literature, scholarship related to participatory planning has
a long and rich history. In the past, empirical studies have analyzed barriers and drivers to
community participation in planning, the relationships between the participation of
specific stakeholders and plan quality, and the plan implementation outcomes associated
with participatory approaches to planning. Until now, however, the planning literature
has not examined the relationships between social vulnerability to disasters, public
participation and hazard mitigation plan quality. Participatory planning is no cure-all for
disaster risk. However, a critical analysis of participatory planning processes is necessary
in order to identify shortcomings and best practices and to strategically allocate local
resources in order to successfully and efficiently decrease social vulnerability to hazards
and disasters.
7


CHAPTER II
LITERATURE REVIEW
Introduction
The overarching research questions that motivate this study are framed by the
body of literature on social vulnerability, hazard mitigation and participatory planning.
Chapter two covers areas of consensus and dispute in these literatures, and evaluates the
nature and quality of support for various contentions in the discussion concerning public
participation and improved planning outcomes.
To begin, our review of the hazards and disasters literature highlights key factors
that influence mitigation plan quality outcomes, specifically those outcomes associated
with local vulnerability reduction and disaster resilience. Later, our exploration of
collaborative planning theory and practice offers a framework by which to evaluate
hazard mitigation planning and the influences that participatory approaches have on
mitigation plan quality and implementation outcomes. Ultimately, this chapter provides a
conceptual framework by which local hazard mitigation planning can be evaluated in a
new way, one that contributes to significant new understandings of the potential for
improved planning practice to contribute to reduced local social vulnerability to hazards
and improved local resilience.
Hazards. Disasters, and Vulnerability
2008 marked the first time in human history when urban centers became home to
the majority of the worlds people, economic activity, and physical infrastructure.
Currently, more than half of the worlds population lives in urban areas (United Nations,
8


2008). We are living in a unique time of rapid urban development in which high levels of
complexity characterize broad social-environmental challenges. It is increasingly clear
that current development and planning strategies are producing unsustainable outcomes
that have contributed to significant increases in vulnerability to hazards as well as
reductions in local, regional, and national quality of life.
An example of the alarming impacts of this development profile is the
dramatically increasing trend in international losses from disasters. Not only is the
phenomenon a function of increased development in high-risk places, but also of
increasing levels of social and physical vulnerability within communities. Increased
vulnerability puts communities on trajectories towards greater risk of catastrophic failure
or damage in the event of a disaster or shock (Kasperson at al., 1995; UNDP, 2004;
Pelling, 1997). Moreover, in particularly vulnerable systems even small perturbations
may lead to collapse.
Current mitigation practices often aim to maintain the status quo within
communities. However, this approach to planning tends to perpetuate the disaster-damage
cycle rather than addressing (and remedying) the root causes of hazard losses. Although
efforts to minimize the negative physical impacts of hazards are important, it is equally
important that communities implement ongoing planning strategies that take a
comprehensive, holistic approach to emergency management by including political,
social, and economic factors in their development and mitigation plans. To become truly
sustainable in the changing global environment, communities must address social
vulnerability as a root cause of hazard losses. This, however, may demand significant
changes in current mitigation planning practices as well as in the structure of society.
9


Definitions of Terms
The rapidly growing body of literature on hazard mitigation planning features a
wide, and sometimes confusing, array of terms. Because the study of human and
environmental vulnerability to climate change and hazard events is necessarily
interdisciplinary, sometimes the same term takes on a different meaning when used by
different authors in different contexts. As a result, relationships between terms are often
unclear.
In this study the term hazard describes a potentially damaging physical event,
phenomena, and/or human activity, which may cause loss of life, injury, property
damage, social economic disruption or environmental degradation (UN/ISDR, 2009).
Hazards can be characterized as slow-onset events (i.e. droughts or long term temperature
change) or rapid-onset events (i.e. floods, landslides or earthquakes). When hazards are
described in the context of this study they are described by their physical characteristics.
The term disaster incorporates the human elements of a hazard event. As defined
by the International Strategy for Disaster Reduction, disasters are serious disruptions of
the functioning of a community or a society causing widespread human, material,
economic, or environmental losses, which exceed the ability of the affected community
or society to cope by using its own resources (UN/ISDR, 2009). Disasters are triggered
by hazard events and are influenced by the properties of the human systems that are
exposed to and affected by the hazard.
Risk is an important term in the context of vulnerability assessment. It is a
fundamental component of an equation that has been used to determine the likelihood of
occurrence (or probability) of a hazard event (Cutter, 1996). Risk is the result of
10


interactions between a hazard and the vulnerable conditions of a community, place, or
individual. Below, the commonly used risk equation describes the relationship between
risk, hazards, and vulnerability (Wisner et al., 2004):
Risk = Hazard x Vulnerability
Researchers in the hazards field often focus on the concept of risk assessment,
while social scientists and climate change researchers often prefer to talk in terms of
vulnerability (Downing et al., 2001; Allen, 2003; Brooks, 2003). In general, vulnerability
means the potential for loss. Often times, however, social scientists and climate change
researchers mean different things when they speak about vulnerability. Generally,
definitions of vulnerability can be separated into two categories. One category defines
vulnerability in terms of the amount of potential damage caused to a system by a specific
hazard (Jones & Boer, 2003). In the global environmental change literature, for example,
vulnerability is most often conceptualized as the likelihood of occurrence of impacts from
weather and climate related events (Nicholls et al., 1999; Brooks, 2003).
The other category defines vulnerability as an inherent characteristic of a system
that changes continuously over time and exists before the system encounters a hazard
event (Allen, 2003; Cutter, 1993). For example, social scientists tend to describe
vulnerability as a representation of the social-economic factors that determine the ability
of communities and individuals to cope with stress or change (Allen, 2003; Brooks,
2003).
Over time, the concept of vulnerability has been borrowed, recycled, and reused
by numerous disciplines and has taken on a number of definitions and applications. It is
not within the scope of this dissertation to review them all, however two distinct concepts
11


of vulnerability are clearly evidenced in the hazards and disasters literature: biophysical
vulnerability and social vulnerability. Although these concepts are connected, they frame
vulnerability reduction discussions in very different ways and it is important to
understand the nuances between them.
The IPCC Third Assessment Report defines vulnerability as a measure of the
nature of the physical hazard to which the community is exposed, the likelihood of
occurrence, and the degree of human exposure to the impacts of the hazard (or the
systems sensitivity) (IPCC, 2001a). This three-tiered, aggregated measure of
vulnerability a function of hazard, exposure and sensitivity is sometimes referred to
as physical or biophysical vulnerability. Ultimately, biophysical vulnerability
assessments are concerned with the outcomes of hazard events (the amount of damage
sustained by a system from a hazard), rather than with indicators of the state of the
system before the event.
A second, distinct conceptual stream describes vulnerability as a state, or as an
inherent characteristic of a system that exists before it encounters a hazard event (Allen,
2003; Cutter, 1993). This view developed from a body of social science research related
to understanding the structural factors that make human societies more or less susceptible
to damage from hazard events (Mileti, 1999; White & Haas, 1975; Kasperson et al.,
2003; Cutter et al., 2003). The concept (as it relates to strictly human systems) is called
social vulnerability (Cutter et al., 2003; Adger, 1999; Adger & Kelly, 1999). One of the
significant lessons from this body of research is that social vulnerability is a unique
property of a system. It exists before a hazard event and changes continuously over time.
Social vulnerability is the product of social inequalities. Its theoretical origins lie
12


in the quality of life and livability studies in the social and behavioral sciences during the
1950s and 1960s. The 1960s and early 1970s witnessed a surge in federal government
interest in the identification and quantification of social well-being indicators (Cutter &
Emrich, 2006; U.S. Department of Health, Education, and Welfare, 1969; Ci.S. Office of
Management and Budget, 1973). Initially, this body of research tried to understand
specific characteristics of places that make them more or less desirable and suitable
places to live.
Increasingly, empirical research began to delve into the social attributes of people
and places as a way of understanding and predicting human behavior when faced with
problems of social, health and environmental nature (Cutter & Emrich, 2006; Maloney,
1973; Smith, 1973; Berry, 1977). Preliminary research in this area set the stage for the
interdisciplinary trends in inquiry that are currently found in the vulnerability science
literature (Cutter, 2003). Moreover, studies of this nature have made it clear that
differential social vulnerability is an important social justice concern.
Measuring Social Vulnerability
A number of attempts have been made to measure and quantify social
vulnerability, including national-scale hazards and disaster indicator studies that include
social characteristics in their multi-hazard exposure models. Others have included social
vulnerability indicators in their evaluations of social-ecological systems and different
scales, including U.S. watersheds (Hurd et al., 1999), counties within the U.S. Great
Plains (Polsky, 2004), and states across the U.S. (Cutter et al., 2003). The quantification
of social vulnerability has also become an important focus of field research and
13


vulnerability mapping projects concerned with identifying the most vulnerable members
of society and examining variations in vulnerability over time and between or within
geographic areas (Cutter, 2003; Cutter et al., 2008; Borden et al., 2007; Piegorsh et al.,
2007).
As foundations for research on vulnerability and risk, three conceptual models
have been particularly significant: the Risk-Hazard model (RH), the Pressure and Release
(PAR) model, and the Hazards of Place model. The RH model looks at hazard impacts as
a function of exposure to a hazard event and the sensitivity of the exposed system
(Warrick, 1980; Riebsame, 1991; Kates, 1985). Most quantitative system evaluations
using this model have worked forward, from the hazard to their impacts, and have
evaluated multiple impacts of a single perturbation (Warrick, 1980; Riebsame, 1991;
Kates et al., 1985).
Continued vulnerability research has revealed the RH models shortcomings
including its failure to incorporate the ways in which affected systems can magnify or
ease the impacts of a hazard event (Kaspersen et al., 1988; Turner et al., 2003).
Moreover, the model does not identify the ways in which different components of a
system create variations in the degree and nature of impacts. Finally, the model does not
include the role of social or economic institutions in affecting exposures or impacts
(Wisner et al., 2004).
These criticisms led to the development of the Pressure and Release model
(PAR). In this model, risk is generated at the intersection of two opposing forces: the
hazard event and the vulnerability of the exposed system (Wisner et al., 2004). The PAR
model focuses on dynamic social conditions of exposure (or pressures) as the root causes
14


of vulnerability. These pressures can be traced back to distant processes and factors,
creating a causal chain of vulnerability (Wisner et al., 2004). The release component of
the model has to do with vulnerability reduction activity (Wisner et al., 2004), and, like a
breaking wishbone, the moment at which the accumulated pressure within the system is
released and a disaster occurs.
According to the PAR model, in order to mitigate the vulnerabilities that cause
disasters, the entire causal chain must be addressed, all the way back to the root causes.
According to the PAR, focusing solely on the triggers of the hazard event or on the
physical unsafe conditions of vulnerability is ineffective at reducing the vulnerability of
the system. Although it explicitly emphasizes the importance of vulnerability reduction in
disaster management, the PAR model is insufficiently comprehensive for the broader
concerns of sustainable development. For one, it does not address the coupled human-
environment system or consider the vulnerability of biophysical subsystems (Kasperson
et al., 2003). Additionally, the model offers little insight into the structure of the hazards
causal sequence, including various scales of interactions, and it tends to underemphasize
feedbacks beyond the immediate system of analysis (Kates et al., 1985; Turner et al.,
2003).
The Hazards of Place model illustrates how the interaction between hazards and
social vulnerability produces outcomes in terms of physical or economic damage or loss
of human life. It has been an attractive model to researchers because it integrates two
perspectives on vulnerability the biophysical and the social (Messner & Meyer, 2005).
Both the RH and PAR models of vulnerability underscore how different communities
have different capacities, sensitivities, and exposures to hazards and shocks. Within
15


human systems these characteristics are closely linked to social factors including power,
basic human rights, and access to resources (Wisner et al., 2004; Sen, 2003). As it
became increasingly clear that focusing only on physical hazards and exposure was
insufficient for understanding the impacts of hazards on human systems, focused efforts
began to develop metrics for social vulnerability (Mileti, 1999; White & Haas, 1975;
Kasperson et al., 2003).
In the Hazards of Place model (Figure 1), risk interacts with mitigation activity to
produce a level of hazard potential. Eventually, this hazard potential is either moderated
or enhanced through geographic filters, such as the physical proximity to the hazard and
the social fabric of the place. In this model, biophysical and social vulnerability interact
to produce an overall level of place vulnerability (Cutter et al., 2003).
Figure 1: The Hazards of Place Model
Social vulnerability measures population sensitivity to hazards as well as the
ability of a population to respond and recover from hazard impacts. Because it is a
16


complex, multidimensional concept, researchers have aggregated a collection of variables
to help identify, capture, and quantify it. A large body of empirical evidence from field-
based research has contributed to more robust understandings of the characteristics that
make some people and social groups more sensitive to hazard impacts than others, as
well as what makes them better able to respond and recover from shocks (Cutter &
Emrich, 2006; NRC, 2006; Heinz Center, 2002).
Based on the Hazards of Place model, Cutter et al. (2003) developed an index for
the purpose of distilling out the driving factors of social vulnerability. Cutter and her
colleagues have conducted a suite of research studies quantifying the vulnerability of
people and places to natural hazards at different scales. In their 2003 paper, Social
vulnerability to environmental hazards, Cutter et al. (2003) used county-level
socioeconomic and demographic data to create an index of social vulnerability to
environmental hazards. They named their social vulnerability quantification tool the
Social Vulnerability Index (also known as SOVI).
Based on the results of myriad SOVI and cultural competency studies, socio-
economic status, gender, race and ethnicity are the most common characteristics defining
socially vulnerable populations. The age profile (children and the elderly), percentage of
limited language speakers, and housing tenure (renters) also play significant roles in the
ability of populations to absorb impacts, respond and recover in the event of a disaster
(Pastor et al., 2006; Andrulis, Siddiqui & Ganter, 2007). To illustrate this point, consider
for a moment why people with limited language skills are more vulnerable in the event of
a disaster. Previous studies have shown that the inability to understand evacuation
warnings or preparedness bulletins influences an individuals or communitys ability to
17


comply with safety measures. Additionally, the inability to communicate special needs to
emergency responders or law enforcement can influence their ability and willingness to
receive adequate health care or emergency supplies. Limited language ability also affects
an individuals or communitys ability to communicate their risks and vulnerabilities to
planners and emergency managers who organize pre-disaster mitigation efforts. As a
result, communities with populations made up of greater proportions of individuals with
limited language have a higher social vulnerability to hazard impacts and it will likely
take them longer to recover from a hazard event (Kreisberg et al., 2011; Pastor et al.,
2006).
More recent studies indicate that there are spatial and temporal patterns in social
vulnerability across the United States. In their 2008 paper, Temporal and spatial changes
in social vulnerability to natural hazards, Cutter et al. (2008) found that the factors that
consistently increase social vulnerability at the county scale for all time periods are urban
density, race/ethnicity, and socioeconomic status. At the national scale, trends show a
steady decrease in social vulnerability. However, there exists significant regional
variability, with many counties showing clear increases in social vulnerability over the
past fifty years (Cutter et al., 2008).
Although social vulnerability is not a function of hazard severity or probability of
occurrence, it is, to a certain extent, hazard specific. In other words, when thinking about
social vulnerability we must ask the question: vulnerability of whom to what? This is
because certain properties of human systems make them more or less vulnerable to
certain types of hazards. For example, quality of housing is an important determinant of a
18


communitys vulnerability to an earthquake or a windstorm. However, it is less likely to
influence the communitys vulnerability to drought.
To summarize, biophysical vulnerability depends on the severity and probability
of a given hazard event, whereas social vulnerability does not. Biophysical vulnerability
is conceptualized in terms of outcomes (damages from the hazard event), whereas social
vulnerability exists independently of a hazard event. In this dissertation, social
vulnerability encompasses all of the system properties that mediate the outcome of the
hazard event and is independent of the hazards to which it is exposed. This may include
environmental variables and measures of exposure because exposure and the state of the
environment in which a community lives and works are largely socially determined
(Brooks, 2003).
Increasingly, social vulnerability has been cited by researchers, emergency
managers and policy makers as an important element of hazard risk in both urban and
rural settings. As a result, increased attention is being turned towards social vulnerability
during policy making and development planning. While using data from previous disaster
events can identify physical vulnerability relatively easily, the social aspects of hazard
vulnerability are difficult to capture due to their complexity and spatial and temporal
variability. Despite the progress that has been made with respect to social vulnerability
research and variable identification, there is no agreed upon approach or index for
quantifying it. As a result, incorporating the concept into policy remains a great
challenge.
19


Building Resilience into Emergency Management
Resilience is another term that continues to emerge more frequently in the
disasters and global environmental change discourse. A concept traditionally grounded in
ecology and systems theory, resilience has been adopted and modified by a diversity of
disciplines, including emergency management. How communities are affected by and
recover from unexpected shocks and hazard events can be conceptualized in terms of
their resilience. More specifically, a disaster resilient community is one that can
withstand and absorb the impacts of extreme disturbances and maintain core functions,
often times after assuming an altered form (Innes & Booher, 2010; Berkes et al., 2003).
Although they are closely related concepts, it is important to distinguish between
disaster resilience and social-ecological resilience. Born out of systems theory, the
concept of social-ecological system (SES) resilience describes to what extent the
relationships between social and ecological processes can be disturbed before they
experience fundamental changes in identity, structure and/or function (Holling, 1973;
Alberti et al., 2003). A deeper understanding of social-ecological systems has shown that
not only are resilient systems more able to be flexible in the face of unpredictability and
change, but they are also better able to take advantage of future opportunities afforded by
change (Holling, 1937; Holling et al., 1995; Alberti et al., 2003).
Early perspectives on ecological and social system dynamics were highly linear.
Natural environments were conceptualized as inherently stable and infinitely resilient.
Moreover, it was believed that if human stressors were removed, nature would repair
itself and return to a state of perfect equilibrium (Holling, 1961; Lewontin, 1969;
Rosenzweig, 1971; May, 1972). Ultimately, research on interacting populations (e.g.
20


predators and prey) and their responses in relation to ecological stability theory led to the
emergence of a theory of multiple domains of stability in natural systems (Holling, 1961;
Lewontin, 1969; Rosenzweig, 1971; May, 1972). This was the beginning of resilience
thinking.
Scholarship on social-ecological system resilience originated in the 1960s and
early 1970s from a stream of ecology in which scholars were examining ecosystem
behavior in the face of shocks and environmental change. The social-ecological system
resilience framework includes humans, as well as environmental factors, as agents of
ecosystem change. Traditionally, ecological theory excluded social institutions from the
study of ecosystem dynamics and treated human actions as external to the system. As a
consequence, the interactions between ecological processes and social dynamics were
rarely taken into account in science, policy, or decision-making.
Social-ecological system resilience represents a shift in thinking about
unpredictability, as well as in the meaning of equilibrium within systems. In his seminal
paper on resilience, Resilience and stability of ecological systems, C.S. Buzz Holling
(1973) wrote that social-ecological system resilience is based on non-equilibrium
dynamics and that the concept of equilibrium is poorly suited to explain the behavior of
systems that are not currently at equilibrium. He argues that because the concept of
equilibrium represents a static state, it is incapable of providing insight into the behavior
of systems that are variable or transitional (Holling, 1973).
From the social-ecological perspective, resilience determines the persistence of
the relationships within systems and measures the ability of a system to absorb shocks
without losing its capacity to function in its current state of existence (Holling, 1973,
21


1996). Holling (1973) recognized early on that complex adaptive systems are not
infinitely resilient. Rather, he suggested that they require adaptive management in order
to persist. Adaptive management processes have been characterized by learning-by-
doing, where plans and policies are experiments from which communities and social
systems can learn and share knowledge (Walters, 1986; Gunderson, 1999).
Although disaster resilience describes a more reactive, outcome-based concept
associated with bouncing back and recovering quickly after a hazard event, disaster
resilience conceptualizes communities as systems of interacting components (i.e. social
systems, governance systems, infrastructure, planning entities and ecosystems). Similar
to the learning-by-doing approach to resilience building expressed by social ecological
resilience theory (Walters, 1986; Gunderson, 1999), communities can increase their
resilience to disasters through adaptive management activity including knowledge
sharing, learning from past experiences, and creating a common language between
different systems actors and decision makers (Paton et al., 2000; Goldstein, 2009; Innes
& Booher, 2008; Innes & Booher, 1999; Colten & Sumpter, 2009).
In recent years, expressions such as disaster-resilient communities and
building community resilience have become commonplace in emergency management
literature and disaster preparedness and recovery discussions. However, definitions of the
term resilience abound, as is highlighted by Table 1.
22


Table 1: Definitions of Resilience
Author Definition Disciplinary Lens
Holling, 1973, p.14 A measure of the persistence of systems and of their ability to absorb change and disturbance and still maintain the same relationships between populations or state variables. Ecological Science
Walker et al., 2004 Resilience is the capacity of an ecosystem to tolerate disturbance without collapsing into a qualitatively different state that is controlled by a different set of processes. A resilient ecosystem can withstand shocks and rebuild itself when necessary. Resilience in social system has the added capacity of humans to anticipate and plan for the future. Ecological Science, Social Sciences
Walker et al., 2006, p.2 The capacity of a system to experience shocks while retaining essentially the same function, structure, feedbacks, and therefore identity. Ecological Science
Adger, 2000, p.347 The ability of groups or communities to cope with external stresses and disturbances as a result of social, political and environmental change. Social Sciences
Adger et al., 2005, p.1036 The capacity of social-ecological systems to absorb recurrent disturbances [...] so as to retain essential structures, processes and feedbacks. Social-Ecological Science
Department of Homeland Security, TCL, 2007, p. 13 Resilience is key [to a determination of vulnerability] since it refers to our coping capacity to absorb events, adapt, and respond to and recover from its effects. Disaster Management
Mileti, 1999 Local resiliency with regard to disasters means that a locale is able to withstand an extreme natural event without suffering devastating losses, damage, diminished productivity, or quality of life without a large amount of assistance from outside the community. Disaster Management
Paton, Smith & Violanti, 2000 Resilience describes an active process of self- righting, learned resourcefulness and growth the ability to function psychologically at a level far greater than expected given the individuals capabilities and previous experiences. Disaster Management
United Nations International Strategy for Disaster Reduction (UNISDR, 2010) 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. This is determined by the degree to which the social system is capable of organizing itself, and the ability to increase its capacity for learning and adaptation, including the capacity to recover from a disaster. Disaster Management
23


The increasing use of the word resilience in the context of disasters has been
described by some as representative of a new paradigm of disaster response (McEntire et
al., 2002). While some scholars consider resilience thinking to be a new paradigm, others
describe it as merely an expression that combines other disaster terms (including
vulnerability, adaptive capacity, and risk). These scholars believe that, conceptually,
disaster resilience does not represent anything new. Citing the occurrence of previous
rhetorical shifts created by words like sustainability, they claim the only new thing is the
inclusion of the word resilience in the disaster and development discourse (Manyena,
2006, 435). Despite these critiques, in looking at the foundations and definitions of
resilience, and by examining points of overlap between multiple perspectives, it becomes
more clear that resilience thinking has great potential to catalyze new and improved ways
of planning and governance, particularly in the context of participatory planning and
local social vulnerability reduction.
Resilience thinking has emerged as a useful framework for understanding and
solving complex urban problems because it can be applied to both the social and physical
aspects of communities. Over the past decade, the broad concept of disaster resilience has
found both a practical and theoretical home in the disaster risk reduction discourse
(Manyena, 2006). For example, a number of international and federal level emergency
management, security, and planning entities have identified resilience building as a key
priority for guiding policy and sustainable development decisions. A desire to merge
disaster risk reduction with urban planning has gained momentum with the publication of
reports such as Reducing Disaster Risk: A Challenge to Development (UNDP, 2004) and
Disaster Risk Reduction: A Development Concern (DFID, 2005). These publications
24


underscore how disaster risk accumulates within inappropriate development paths and
that increasing disaster losses can be explained largely by the consequent increases in
human vulnerability. This phenomenon highlights the necessity of developing alternative
frameworks for urban planning practice and decision-making based on robust
understandings of the social factors of vulnerability and risk.
In the U.S., resiliency has taken on important meaning in the context of
emergency management and homeland security. However, as a process, disaster
resilience demands new applications of the risk equation. It necessitates a more holistic
approach to vulnerability analysis as it places a new emphasis on the role of people in
disasters. By focusing explicitly on resilience, disaster managers are encouraged to
consider the social components of vulnerability, including ways in which to enhance
community coping capacity and assets rather than simply reducing weaknesses
(Manyena, 2006).
Hazards, Disasters and Sustainability
Planners have responded in a diversity of ways to growing frustrations with
common features of urbanization including sprawl, pollution, inequity, and vulnerability
to natural hazards. In the United States, sustainable urban planning strategies have been
focused primarily on climate change mitigation measures. By and large, urban centers
have relied on improvements to physical infrastructure in response to the impacts of
climate change. In addition, climate change mitigation measures such as increased GHG
sequestration through wetland creation, decreased GHG emissions from oil and gas
industries, and increased investment in renewable energy are gaining greater support at
25


local and regional levels as a response to climate change and increased vulnerability to
storms and floods.
Despite the prominence of efficiency improvements and GHG reduction through
engineering advances and policy, urban approaches to green development have also
included an emphasis on sustainable planning tools such as urban growth boundaries and
the LEED-ND rating system. Sustainable development has emerged as a tool for planning
practitioners to increase local sustainability on both social and ecological levels. New
urbanism, infill development and smart growth strategies are other examples of
sustainable planning strategies that have been applied at neighborhood, city, and regional
scales. They include the design of compact, high-density neighborhoods for more
efficient use of land and transportation; the construction of resource and energy efficient
buildings; the greening of communities by incorporating trees, parks, and open space.
Moreover, by building within existing urban areas, cities conserve their environmental
resources, economic systems, and social fabric, while reclaiming marginal and
abandoned land (Beatley, 1998; Berke, 2002).
Recently, scores of cities across the U.S. have initiated comprehensive plans and
urban development projects with ambitious sustainability goals in mind. However, little
attention has being paid to the following question: Is green disaster resistant? City
planners play an important role in shaping the vulnerability and resilience of urban areas,
whether they know it or not. In many cases, projects labeled green or sustainable can
work against local hazard mitigation and vulnerability reduction efforts (Berke et al.,
2009; Stevens et al., 2010).
26


Communities that are vulnerable to catastrophic damages from hazards are, by
definition, unsustainable. A community is unsustainable if it is at risk of losing
environmental quality, its economic livelihood, or human lives during a hazard event. As
complex systems, cities that are planned with resiliency in mind are capable of
withstanding large shocks without sustaining either immediate failure or permanent
damage. They are designed and managed to anticipate, absorb, and recover from the
impacts of hazards. Moreover, they are built on principles derived from past experience
with disasters in urban areas. While resilient communities may bend, they will not break.
Despite this logical connection between sustainability and disaster resilience, however,
very limited research has been done to unpack the relationship.
Independent of the sustainability revolution, the planning profession has long
upheld a development agenda that supports strong economic growth, environmental
health and improved quality of life. In his paper Green Cities, Growing Cities, Just
Cities, Scott Campbell describes the planning discipline as defined by three (arguably
divergent) priorities: social justice, economic growth, and environmental protection
(Campbell, 1996). Reminiscent of the three Es of sustainable development
(environment, economics, and equity), these priorities lie at the historical heart of
planning. Conflicts between these priorities define contemporary struggles within human
settlements. Campbell (1996) describes the interaction between planners three priorities
as the Planners Triangle (Figure 2) with sustainable development situated at the center.
The Planners Triangle is also conceptualized as a three legged stool (Campbell, 1996).
27


Social Justice
Figure 2: The Planner's Triangle
In this model, Campbell argues that sustainable development cannot be achieved
by looking at one leg of the stool at a time. Rather, planners will reach sustainable
solutions only by confronting the conflicts between the three points of the triangle and by
coming up with resolutions over time. To be sure, the three axes of the planners triangle
do not interact solely through opposition. They also interact collaboratively (Campbell,
1996). Campbell stresses that in the planning discipline sustainability must be represented
as a fine balance between social justice, economic growth, and environmental health. He
argues that in an ideal world, planners would strive to achieve a perfect balance between
growth, preservation, and equality. In practice, however, professional, political and fiscal
constraints often limit the freedom of planners and decision makers to do so (Tobin,
1999).
28


Hazard Mitigation: Planning. Policy and Practice
The following sections describe the evolution of comprehensive planning practice
and how hazard mitigation planning fits into the comprehensive planning framework. The
discussion includes a review of the history of hazard mitigation planning in the U.S. and
provides the background for a clear understanding of how mitigation planning is used as
a tool for the achievement of community-scale vulnerability reduction goals.
The Comprehensive City Plan
In the most general sense, comprehensive plans are long-range physical plans
made up of a collection of policies. Typically, the city is the geographic scope of the
document. Because the typical time frame of a comprehensive plan is 20-30 years,
beyond the scope of current pressing issues, they open up dialogue about future
possibilities and challenges. Comprehensive plans focus on social, built environment, and
economic infrastructure that make the city work as a physical entity and that affect
physical form. Written as a statement of policy rather than a rigid program of specific
actions, comprehensive plans are designed to guide urban decision-makers as they make
decisions in the future (Black, 1968; So and Getzels, 1988; Innes, 1996).
In 1964, California Bay Area region planner Jack T.J. Kent presented a version
of the comprehensive city plan that is widely recognized as the strongest representation
of the comprehensive planning ideal. In his book, The Urban General Plan, Kent lays out
a set of principles and goals for comprehensive planning that planners and scholars
continue to refer to decades after their initial publication. Kents (1964) view of
comprehensive planning is distinct from the rational comprehensive model of planning,
29


which functions by clearly outlining all potential objectives and policy options,
comparing them, and ultimately choosing the optimal planning strategy (Lindbloom and
Braybrooke, 1963). It was Kents vision of the urban general plan that switched
comprehensive planning practice away from considering every alternative before making
a decision. Rather, Kent advocated for comprehensive plans to serve as toolboxes of
policies to be utilized in response to future circumstances to achieve city goals and
objectives (Innes, 1996).
Kent describes three necessary components for comprehensive plans:
comprehensiveness, explicit reasoning, and generality (Kent, 1964). Kent states that
comprehensive city plans must be comprehensive in that they cover the entire city. Plans
must focus not only on infrastructure but also on the relationships between the many
factors that affect the growth and development of cities including the physical and
nonphysical, local and regional, and social and environmental. Kent also stresses the
importance of considering technological, economic, and demographic forecasts while
developing comprehensive plans. Comprehensiveness is the key, he writes, to promoting
the public interest (which constitutes the community at large rather than individuals or
powerful stakeholder groups) (Kent, 1964).
Explicit reasoning requires that plans clearly identify the facts from which their
visions were developed and justified. This allows for decision-makers to make informed
final judgments about if and how to implement the policies outlined in the city plan. Kent
also stated that the principle of generality is critical to comprehensive planning. Plans
should not serve as a blueprint for the city; rather, they should be general and focused on
larger issues facing the city and big ideas for their resolution. Because cities and
30


communities are complex and constantly evolving, plans should not be used as ideal
models for development. Rather, they should be used as tools for democratic discourse
(Kent, 1964; Innes, 1996).
Long-range comprehensive planning is an exercise in anticipatory problem
solving and is a process through which community needs and goal setting objectives can
be identified in an organized way. In theory, comprehensive planning is intended to stand
up for common interests rather than those of a limited, powerful few. Ideally,
comprehensive planning is an on-going, dynamic process in which plans are revisited and
reevaluated regularly after their adoption. Because of the complex, adaptive nature of
cities and communities it is important for plans to be updated so that they remain relevant
to changing needs, priorities, and challenges.
Early U.S. city and regional planning developed in response to complex urban
problems including social inequity and environmental degradation. Over time, problems
changed and community priorities were redirected towards new, emerging urban issues.
For example, between the 18th and 19th centuries, the Industrial Revolution drastically
changed the function and form of cities. In his book, Urban and Regional Planning, Peter
Hall describes the noisy, disease-riddled cities of the Industrial Revolution as the most
important planning problem of the period. Urban economic growth was threatened by
inadequate infrastructure and unsanitary conditions and planners were motivated to find
fast and innovative solutions to the industrial city (Hall, 2002). Today, issues of disaster
resistance, social equity, and economic development have surpassed issues of sanitation
and undesirable industrial development as leading urban planning concerns.
31


The literature identifies two fundamental purposes around which planning
problems pivot: place making and conflict resolution (Healey, 1998). A clear
understanding of the priorities associated with conflict management and place making is
useful for setting the stage for conceptualizing contemporary planning theory and
practice. Place making involves the promotion of quality places. The aim of place making
is to use design as a way of achieving broad economic, environmental, and social
objectives. The second fundamental purpose of planning is to facilitate the regulation of
private land use rights and to manage conflicts having to do with land use and
development. All regulation and conflict management are part of a larger effort to serve
the public interest (Healey, 1998). Healey (1998) acknowledges that conflict management
and place making are deeply interconnected. However, continued analysis of planning
practice and structure has illustrated that the two priorities also diverge. Figure 3
illustrates how conflict management and place making can diverge and come together
through a process of effective spatial planning.
32


Conflict management
Relation of
land use rights
Planning by
negotiation and
contract
EFFECTIVE
SPATIAL
PLANNING
Strategies for
place making
Place making
Planning by
multi-stakeholder
collaboration
Figure 3: Place Making and Conflict Resolution through Planning
Generally, place making happens through professional and stakeholder debates
about ideas for achieving desired community outcomes. Place making focuses on the
transformation of both policy and practice. In contrast, the process of conflict
management is largely defined by the actions of legal and administrative entities. These
entities are primarily concerned with disputes about the legality of property rights
restrictions and focus on maintaining established structures of governance and regulatory
systems (Healey, 1998).
Since the 1960s planning practice and scholarship have transitioned from a focus
on physical design to a focus on process. Before the 1960s, planners brought their
creative ideas to the public by guiding the development of cities, towns, and regions.
Urban planning in the United States was characterized as a largely technocratic process,
dominated by independent experts who used objective scientific analysis to manage
33


resources and make infrastructure decisions. The preferred model in the profession was
decision-making free of political interference (Berke, 2002; Wondolleck and Yaffee
2000; Susskind et al., 1999; Beierle and Cayford, 2002; Gunton and Day, 2003).
During the 1960s and 70s, rational planning, a model of planning based on
systems theory and the scientific method, emerged in response to numerous social and
environmental crises (Berke, 2002). However, communities and decision-makers began
to question the efficacy of traditional planning as widespread urban problems including
racism, pollution, poverty, and financial failure continued to plague metropolitan areas
and regions (Berke, 2002). As a result, planners began to take a more comprehensive,
systematic, and analytical approach to their work.
After the 1960s, urban planning was largely dominated by incremental projects,
rather than large-scale ones (Berke, 2002). This transition was due, in part, to growing
dissatisfaction with the top-down, expert dominated approach to city building that was
often exclusionary and counterproductive to achieving larger community goals
(Altshuler, 1965; Davidoff, 1965). The Civil Rights Movement helped catalyze the
development of the participatory planning model and during this time urban areas were
important places of political expression and civic dissent. Scholars and communities were
acutely aware that early urban renewal and highway development projects
disproportionately affected low-income, minority neighborhoods. This contributed to a
significant decline in the publics trust of rational planning. In response to public
dissatisfaction over planning policies involving resource extraction, urban renewal and
transportation, planners formally acknowledged democratic decision-making as the
34


foundation of their profession (Gans, 1962; Davidoff, 1965; McLoughlin, 1969;
Chadwick, 1971; Hall, 2002; Chapin and Kaiser, 1979; Hester, 1996).
Ultimately, diverse values, competing priorities, and resource constraints sparked
the creation of two new planning paradigms in the U.S. Both mediation and advocacy
planning emerged as new ways for planners to think about their role in decision-making
and city building. Mediation, also known as alternative dispute resolution (ADR), calls
for planners to act as intermediaries between conflicting stakeholders. In this model, the
goal of the planning process is to guide decision-making in a way that produces mutually
beneficial solutions (Bacow and Wheeler, 1984; Susskind and Cruikshank, 1987).
Proponents of the mediation method argue that they are cost efficient, and more likely to
resolve conflict than other approaches (Gunton and Day, 2003). Moreover, because
planners seek win/win solutions to conflicts, the participatory method is believed to
produce more lasting, equitable solutions.
Advocacy planning developed as a new decision-making strategy intended to
serve disadvantaged and marginalized stakeholders. Like mediation, advocacy planning
acknowledges that different stakeholders often have competing interests. Planners
operating under the advocacy model promote the interests of the most underrepresented
stakeholders and cultivate community participation in an effort to simultaneously create
plans and empower the public (Davidoff, 1965). In the 1950s the dominant view of the
public good was what is good for business is good for all (Hester, 1996). However,
planners began to challenge this belief by facilitating the development of comprehensive
plans for diverse interest groups and they began advocating for development strategies
that represented multiple perceptions of the city. Some scholars argue, however, that
35


although there are clear benefits to this approach to planning, it leaves little room for the
cultivation of holistic, larger-scale visions of the city (Clavel, 1994; Neuman, 1998;
Berke, 2002).
Hazard Mitigation in the United States
Currently, much of the responsibility for reducing hazard losses in the U.S. is
shared by the federal government and numerous local governments and special
jurisdictions. Historically, U.S. disaster management was founded on civil defense and
behavioral science research (Drabek, 1991; Pearce, 2003). Emergency managers and
planners traditionally followed a bureaucratic model of hazard mitigation planning and
drew from privileged technical information to inform their risk identification and goal
setting activity. As a result, the focus of planning activity was often skewed towards
assessing and mitigating geophysical vulnerabilities rather than addressing social
dimensions of hazard risk (Colten, 1992; Colten & Sumpter, 2009; Vale & Campanella,
2005; Drabek, 1991; Pearce, 2003).
Before the 1950s a comprehensive disaster management program did not exist in
the United States and disaster response and recovery was funded by congress on an
incident-by-incident basis. It was not until the Federal Disaster Assistance Program was
enacted in 1950 that the federal government began addressing disasters in a coordinated
manner. In 1979 President Jimmy Carter signed Executive Order 12148, establishing the
Federal Emergency Management Agency (FEMA), and effectively consolidated the
nations historically diffuse emergency related programs into one organization.
36


Table 2: The Role of the Federal Government in Disaster Loss Reduction
Year Federal Program or Act Details
1950 Federal Disaster Assistance Program Grants President broad power to respond to crisis Authorizes Federal government to respond to major disasters
1970 Disaster Relief Act of 1970 Authorizes the use of Federal loans and tax assistance to individuals affected by disasters Authorizes Federal funding for repair and replacement of public facilities Introduces hazard mitigation as a Federal priority
1973 The Flood Disaster Protection Act Makes flood insurance mandatory for the protection of property located in Special Flood Hazard Areas
1974 Disaster Relief Act of 1974 Establishes process for Presidential disaster declarations
1979 Executive Order 12127 Merges separate disaster-related responsibilities into FEMA
1988 Robert T. Stafford Disaster Relief and Emergency Assistance Act Legislates cost sharing requirements for public assistance programs Requires State mitigation plan as a condition of disaster assistance Encourages pre-disaster mitigation planning at the state level through grant programs Provides funds for states and local governments to manage public assistance projects
1997 Project Impact: Building Disaster Resilient Communities Emphasizes local community efforts to mitigate hazards and reduce vulnerability Goal: to achieve community buy- in for mitigation projects
2000 Disaster Mitigation Act of 2000 Provides incentives for close coordination of mitigation planning and implementation efforts between State, local and Tribal entities Establishes mitigation planning requirements for local and Tribal governments as a condition of mitigation grant assistance
37


In the 1980s, in response to increasing losses from disasters, the U.S. government
reevaluated the nations emergency management system. Ultimately, the U.S.
government took on the responsibility of managing increasing hazard risks. This was
exemplified by the passing of the Robert T. Stafford Disaster Relief and Emergency
Assistance Act in 1988. The Stafford Act represented a sea change in national emergency
management priorities by establishing mitigation as the most important means for
combating increasing disaster losses. Congress acknowledged that special measures were
necessary to assist state and local governments in providing aid, assistance, emergency
services, and reconstruction aid to areas that have been impacted by hazard events. A key
component of their strategy involved the encouragement of pre-disaster hazard mitigation
planning at the state level as a mechanism for reducing overall losses from disasters.
Over time, the emergency management community has increasingly
acknowledged the diversity of interests and stakeholders within a community as a critical
piece of the mitigation approach. In 1997, FEMA moved forward with an innovative,
community-based mitigation initiative called Project Impact: Building Disaster Resilient
Communities. Project Impact was an attempt to reduce the impacts of natural disasters by
helping communities carry out risk assessments and needs prioritization before a disaster
event. By encouraging communities to build collaborative public/private partnerships
during mitigation planning Project Impact set the stage for the development of
appropriate, politically palatable, long-term strategies to protect communities from
disaster impacts and loss.
In 2000, the Disaster Mitigation Act (DMA 2000) amended the Stafford Act and
enacted a new set of rules requiring that state, local, and tribal entities coordinate their
38


mitigation planning and implementation efforts. Under DMA 2000, state mitigation plans
are still required as a condition of federal disaster assistance. However, the act introduced
incentives for the development of local scale mitigation plans. DMA 2000 now requires
that State and local communities have approved hazard mitigation plans in place in order
to be eligible for pre- and post-disaster hazard mitigation grant funds. The FEMA
crosswalk for local hazard mitigation plan approval also includes a public participation
component. FEMA consistently stresses that the mitigation planning process is just as
important as the mitigation plans themselves. Although local and state hazard mitigation
planning processes can happen in a diversity of ways, by offering incentives for multi-
stakeholder coordination during the planning process, DMA 2000 strongly encourages
participatory planning activity. Unfortunately, DMA 2000 planning requirements lack
adequate guidelines for achieving meaningful, collaborative public participation activity
during local hazard mitigation planning. Currently, there is a great need for an additional
review crosswalk to be established that includes guidance on how to acquire knowledge
and expertise specifically from vulnerable and marginalized community groups.
Vulnerability Reduction through Hazard Mitigation
In 2005, in response to intensifying hazard losses associated with unplanned
urbanization, poverty, environmental degradation, and population growth, the Second
World Conference on Disaster Reduction broached the subject of disaster resilience. In
January 2005, the Hyogo Framework for Action 2005-2015: Building the Resilience of
Nations and Communities to Disasters was endorsed by 168 UN member states
(UNISDR, 2005). The Hyogo Framework for Action (HFA) recognizes that global
39


disaster risks will only continue to magnify, due in part to the vulnerability implications
of unplanned urbanization. The framework encourages countries to undertake major
efforts to reduce their disaster risk by 2015 and outlines five priority areas for action.
Priority area four calls for the incorporation of disaster risk reduction into urban planning
activity as a method for reducing underlying risk factors (UNISDR, 2005).
The HFA represented an important step towards integrating disaster risk concerns
into planning practice. The framework emphasizes social vulnerability reduction through
largely participatory approaches to risk assessment. Both communities and local
authorities should be empowered to manage and reduce disaster risk by having access to
the necessary information, resources and authority to implement actions for disaster risk
reduction reads the HFA (Section IIIA, point 13.f). This statement is significant because
it explicitly linked disaster risk reduction to community empowerment strategies
including public participation and control over information, resources and authority for
the first time at the global scale.
In their publication Words Into Action: A Guide for Implementing the Hyogo
Framework (UNISDR, 2007) the United Nations International Strategy for Disaster
Reduction stresses the importance of community-based, collaborative decision-making
processes in disaster risk reduction efforts and emphasizes the need to develop a holistic
set of vulnerability indicators (both social and biophysical) as key resilience building
strategies. It also illustrates the ineffectiveness of explaining the impacts of disasters on
complex social, environmental, and economic systems through the use of strictly hazard
or environmental indicators (UNISDR, 2007).
40


Today, the HFA serves not only as a tool for helping nations prioritize their risk
reduction resources, but also as a framework for risk reduction efforts that embrace
collaborative processes and incorporate information derived from communities through
multi-methods approaches. The Hyogo Framework set the stage for the ongoing
discourse on hazard mitigation planning practice in the context of social vulnerability
reduction and citizen participation.
Assessing Vulnerability
As part of the hazard mitigation planning process, emergency managers and
planners conduct comprehensive risk and vulnerability assessments. These assessments
are used as tools for developing plans that help communities understand and manage
disaster risks. There exist a number of models and approaches to vulnerability assessment
and these assessments can be conducted at various scales including national, state,
regional, and neighborhood levels.
Fundamentally, vulnerability assessment is a process of identifying and conveying
interactions between the built environment, political structures, socio-economic
conditions, and the physical hazard risks in a community. They play an important role in
establishing a knowledge base of information about unique hazards and social contexts of
a particular place, and they are meant to reveal the interactions between humans and the
environment at a given location (Turner et al., 2003).
Vulnerability analysis emerged from an extensive body of research on risk,
hazards and global environmental change. One of the salient lessons of this body of
research is that vulnerability is a unique and inherent characteristic of a community that
41


changes continuously over time. Therefore, focusing on hazards alone is insufficient for
understanding the impacts of hazards on a social-ecological system (Mileti, 1999; White
& Haas, 1975; Cutter et al., 2003; Kasperson et al., 2003).
Currently, there is a spectrum of vulnerability assessment models that can be
characterized by the varying degrees of attention given to social vulnerability factors and
local community capacity (Phillips et al., 2009). These models also vary by the processes
through which information and data about vulnerability are collected. Some vulnerability
assessment models focus on single or multi-hazard risks (FEMA, 1997; Coburn, Spence,
& Pomonis, 1994; Phillips et al., 2009), while others include limited assessments of
social determinants of risk (Cutter, Boruff, & Shirley, 2003; Phillips et al., 2009). In
addition, there exist models that strive to capture locally-derived community data from
citizens in order to holistically assess local capacity (Phillips et al., 2009). Table 3
outlines the spectrum of current U.S. hazard vulnerability assessment models1.
1 Table derived from Phillips et al. (2009)
42


Table 3: Models of U.S. Hazard Vulnerability Assessment
Approach Outcomes Data Requirements
Single hazard assessment Documentation and/or risk model for a specific area Single hazard risk information; Historical event and impact data; Monitoring data for model generation or model output
Multi-hazard assessment Inventory of all hazards in a given location; Mapping and quantitative technical analysis of multiple hazard risks Single hazard risk information; Historical event and impact data; Monitoring data for model generation or model output
Basic CVA (community vulnerability assessment) Includes risks of multi-hazards; Explicit attention is given to understanding socioeconomic conditions with respect to hazard risks Existing secondary data sources for social analysis and potential hazards; US Census data for socioeconomic analysis; May include few locally derived datasets
Intermediate CVA Holistic treatment of vulnerability (includes additional data beyond US Census for social and economic analysis); Includes information on coping capacity Supplements U.S. Census data with local-level sources of data; Includes indicators of capacity, including community resources; Inclusion of limited primary data from local surveys or focus groups
Community-based vulnerability assessment Social vulnerability analysis expanded to include top-down and bottom-up information; Process involves capturing community knowledge and perceptions and incorporating them into the assessments Collection of primary data via collaborative techniques; Best practices involve the engagement of stakeholders of all types and community- driven data collection efforts; Collection of information about community support networks
At the top of the table, the more traditional vulnerability assessment models
include outcomes that consist solely of inventories of technical hazard risk data (Single
and Multi-hazard vulnerability assessments). The intermediate models represent
increasingly community-based approaches, which include varying degrees of contextual
43


information such as population demographics, economic factors, and environmental
conditions in their data requirements and outcomes (Basic CVAs) (Phillips et al., 2009).
Examples of more comprehensive mitigation models supplement technical
vulnerability analyses with locally derived information and community knowledge in an
attempt to create context-specific planning goals and policy decisions (Intermediate and
Community-based CVAs). Community-based approaches to vulnerability assessment are
considered the best way to develop mitigation plans that successfully incorporate
valuable community knowledge, including local sources of capacity and social
determinants of vulnerability (Comfort et al., 1999; White et al., 2001; Godschalk, 2003;
Wisner et al., 2004, Phillips et al., 2009).
Despite the range of assessment models, contemporary hazard mitigation planning
policy has exhibited a shift towards an increasingly collaborative, resilience-based
approach to risk reduction. For example, FEMA grant programs now mandate public
participation in disaster management and hazard mitigation planning. This shift is due, in
part, to increasing evidence of the benefits associated with multi-stakeholder knowledge
sharing and discourse particularly in the context of hazard mitigation planning.
Engaging the Public through Participatory Planning
Traditionally, expertise and knowledge have been understood as the products of
scientific or professional occupations. However, scientific and institutional experts do not
always know best. Context determines who the expert is and strongly contextualized
local knowledge can expand the professional radius of expertise (Yli-Pelkonen & Kohl,
2005).
44


Typically, local knowledge is not the product of systematic scientific study.
Rather, its validity stems from a lengthy series of local observations and experiences over
time. A post-Katrina study conducted by geographers Colten and Sumpter (2009)
explores the relationships between local knowledge, scientific expertise and disaster
resistance in the context of pre-Katrina New Orleans. Motivated by a key concept of
resilience thinking that societies can learn from hazard events and use their
accumulated social memory to adapt to future catastrophes their study looks at local
planning activity in New Orleans after Hurricane Betsy in 1965 and at how information
about the communitys experience with Betsy was used for hazard management planning
leading up to Hurricane Katrina in 2005.
Colten and Sumpter (2009) found that two critically important lessons from
Hurricane Betsy were omitted from post-Betsy hazard management activities. One of
those missing lessons was that the areas of greatest flood vulnerability were located in
eastern New Orleans, the lakefront, and downstream parishes. The failure to apply this
information to future comprehensive land use planning prevented the city from taking
action to guide development towards the safest locations. This, unfortunately, resulted in
a decrease in the overall resiliency of the city. Within a matter of years after Hurricane
Betsy, hurricane protection levees were constructed around the most historically
vulnerable areas and the city. Additionally, some of the most rapid growth within the city
took place in eastern New Orleans. The flood footprint had no bearing on development,
write Colten and Sumpter, and in effect the new [...] levees eclipsed memory (Colten
& Sumpter, 2009, 360). Over time, planning documents and development planners
45


continued to build and continued to ignore past flooding in eastern and lakefront New
Orleans further eroding the resilience of the city.
The previous study illustrates an example of the potential negative impacts
associated with hazard mitigation planning approaches that fail to learn from local
communities and incorporate local historical knowledge. Ultimately, the failure of New
Orleans to integrate local knowledge and memory into hazard mitigation and land use
planning eroded the resilience of the city and set the stage for the costliest, and perhaps
the most resentful disaster experience in U.S. history.
The hazards and disaster literature suggests that the adoption of local scale hazard
mitigation plans helps to reduce hazard losses (Nelson & French, 2002). Moreover,
hazards researchers have found that mitigation plans are unlikely to be effective unless
they involve the participation of a diverse range of stakeholders during plan creation
(Burby, 2003). Empirical research has also shown that both the quality of hazard
mitigation plans, and the likelihood that they will be implemented after adoption, tends to
increase with the level of public participation during plan making (Burby, 2003;
Godschalk et al., 2003). The following sections discuss the foundations of participatory
planning, explain a framework for understanding the impacts of various types of
participatory and review the existing body of literature related to identified benefits and
drivers of public participation in planning.
The Theory and Practice of Communicative Planning
Theories of communicative planning, and models of discursive, participatory and
collaborative planning all emphasize an inclusive, democratic approach to making plans.
46


Over the last several decades, planning practice has adopted increasingly interactive and
communicative decision making and knowledge building strategies. Although some of
the terminology and descriptions of the communicative planning model have changed
over time, the fundamental ideal of planning as a collaborative process has endured
(Forester, 1982; Healey, 1998; Innes, 1996; Taylor, 1998; Innes andBooher, 1999a).
Jurgen Habermas developed the theory of communicative rationality in the 1980s
as a way of describing a set of ideal conditions for discourse and knowledge building.
The theory challenged instrumental rationality by basing knowledge building and
reasoning on inter-subjective exchange. Habermas writes that communicative rationality
draws from the practical consciousness and life experiences of stakeholders. Because
physical, social, and experiential contexts overlap, it is argued that communicative
planning processes generate more robust forms of knowledge through deliberation and
interaction (Habermas 1981, 1989; Outhwaite, 1994). New forms of knowledge and
perception are created, collective values and norms evolve, and the communicative
planning process is able to create a new common language (Healey, 1998).
According the Habermas, if ideal conditions for communication and knowledge
building are met during inter-subjective exchange, then discourse will result in
emancipatory knowledge. Habermas theorizes that emancipatory knowledge allows
people to transcend conceptual limitations created by societys socio-cultural and
political power structures (Habermas, 1981, 1989; Innes and Booher, 1999b). For this
reason, he argues that emancipatory knowledge is vitally important in a rapidly changing
world. As change occurs more quickly it becomes increasingly challenging to maintain
an accurate, complete understanding of our urban, social, and environmental systems.
47


Therefore, emancipatory knowledge is a pathway towards innovative solutions and new
ways of understanding our increasingly complex world.
Habermas theories built a case for renewing planning practice based on a critique
of the highly deductive, rational foundations of the discipline. The idea of communicative
processes was introduced to the planning discipline by Forester (1982) and represented a
shift in perceptions about presumed objectivity and the use of information and knowledge
in decision making. Innes and Booher (2010) call this shift in planning and policy making
collaborative rationality. Grounded in Habermas theory of communicative rationality,
collaborative rationality has to do with dialogue and deliberation. Within the
communicative stream of planning thought, perfect truth, information, objectivity, and
collective knowledge are ideals that cannot be fully achieved. However, through
collaborative deliberation planners can most effectively aim towards these ideals and
produce better, more sustainable outcomes than those achieved through individual, expert
based analysis.
Collaborative rationality may also help remedy uneven power relations that exist
in traditional governance regimes and planning systems. Communicative planning theory
asserts that planning is deeply political and largely dictated by power structures. The
agendas of those in power are often favored and stakeholders are rarely afforded equal
participation or opportunities for decision-making (Healey, 1997; McGuirk, 2001).
Habermas (1981) argues that instrumental rationality creates a reality of communicative
distortion, in which false consensus is reached through discourse in the context of
unequal distributions of power. In theory, collaborative rationality can transform power
relations by creating deliberative arenas for stakeholders to participate and share. These
48


power-neutral forums enable temporary conditions of shared power and create forms of
knowledge and rationality in which distortion is minimized (McGuirk, 2001; Innes and
Booher, 2010).
Participatory Planning: Practice and Theory
Citizen participation in U.S. planning has an extensive and lengthy history. The
emergence of increasingly democratic planning paradigms was initially a response to the
failure of traditional planning models to manage complex social, economic, and
environmental problems. For example, transactive planning (Friedmann, 1973),
consensus building (Innes, 1996; Innes & Booher, 1999), and communicative planning
(Margerum, 2002; Healey, 1992; Innes, 1996) all emphasize inclusive, democratic
approaches to making plans. To varying degrees, each paradigm situates communities at
the core of planning practice.
In the 1920s the U.S. Department of Commerce developed state planning
legislation that required local governments to offer citizens the opportunity to review and
comment on plans being developed by practitioners. The state of Hawaiis innovative
growth management law of 1962 was the first state level legislation of its kind, and
required citizen involvement in planning (Lowry et al., 1997).
One of the most prominent examples of early federal-level citizen participation
requirements was the Economic Opportunity Act of 1964. The Act mandated community
antipoverty efforts be overseen by local action programs that were developed and
directed with the maximum feasible participation of the residents of the areas and
members of the groups to be served (Economic Opportunity Act of 1964, PL 88-452,
49


title II). The Act facilitated the establishment of over a thousand community action
programs. However, critics argued that the Office of Economic Opportunity, the entity
that oversaw the program, misinterpreted Congress intent. It was argued that the
Economic Opportunity Act spurred very little genuine participation of underrepresented
stakeholders and that The Act promoted clashes between individual local action programs
and elected city officials (Moynihan, 1969).
The 1954 Urban Renewal Program was the first federal-level citizen participation
mandate in the United States. It was expanded during the 1960s Model Cities and War on
Poverty programs. Throughout the 1970s Congress added participation requirements to a
diversity of federal regulations. For example, citizen participation became a requirement
of the Coastal Zone Management Act and the Energy Reorganization Act.
What was the government trying to achieve by mandating public participation?
The planning literature reasons that the early motivation behind these mandates was to
increase the accountability and commitment of local governments to the principles of
democracy. These principles include the right to information, the right to be consulted
with, and the opportunity to express goals, priorities, and opinions on government
decisions. At the core of democracy is the necessity of representing the interests of
marginalized, powerless, and disadvantaged stakeholders in governmental decision-
making and, in theory, public participation in planning contributes to the realization of
these broader democratic principles (Arnstein, 1969; Burke, 1979; Day, 1997; Godschalk
and Mills, 1966; Fainstein and Fainstein, 1985).
The shift towards increasingly bottom-up approaches to planning led to the
acceptance of a broad range of collaborative planning strategies. Diverse approaches to
50


community engagement including public meetings, advisory committees, charrettes,
workshops, and task forces became increasingly common during the 1960s. Since then,
researchers have evaluated the success of these approaches for generating increased
public participation, empowerment, and plan implementation outcomes (Beierle &
Cayford, 2002; Gunton & Day, 2003; Arnstein, 1969).
In her article A Ladder of Citizen Participation Sherry Arnstein (1969) presents a
theoretical ladder to illustrate how even with the emergence of new state and federal laws
requiring public participation in government decision-making, the majority of that
participation ranks low on the ladder of meaningful participation. The community
participation outcomes can be explained in part by the specific collaborative planning
activities that guide the engagement of, and interactions between, diverse participants
during the planning process. Arnsteins ladder has eight rungs that are divided into three
tiers illustrating a hierarchy of public power and citizen participation (Figure 4).
51


Citeen Power
Tokenism
Nonparticipation
Figure 4: Amstein's Ladder of Citizen Participation
The lowest rungs of the ladder are manipulation and therapy. At this level of the
ladder planning processes are non-participatory. In ascending order the middle rungs of
the ladder are informing, consultation, and placation. These rungs are described as
degrees of tokenism, characterized by one-way flows of information, shallow surveys,
and limited decision-making power by citizens.
The top three rungs of Amsteins ladder represent partnership, delegated power,
and citizen control. At this level, the rungs represent ascending degrees of citizen power.
In the partnership model, power is legitimately redistributed through processes of
negotiation and discourse between citizens and those in power. Delegated power affords
52


citizens the power to make decisions and to provide accountability to the public. Finally,
citizen control describes a level of participation in which the powerless are granted the
authority to collect information, prioritize programs, and direct planning in its entirety.
As illustrated by Amsteins ladder, different citizen engagement activities lead to
varying degrees of participation and power sharing, ranging from manipulation, to
placation, to citizen control. The knowledge systems of communities are complex and are
continuously being transformed through experience, social factors, and even by the
influence of power relations both internal and external to a community. Ultimately,
Arnsteins ladder emphasizes how power and control over decision-making are critical
for achieving meaningful and informed public participation.
To date, the nuances between public participation strategies have been evaluated
and distinguished using similar value-based scales including equitability, inclusivity,
impartiality, and transparency. Density and frequency of community participation in
planning have been shown to depend strongly on the identification of appropriate
community participants as well as on the choices planners make regarding specific
collaborative planning techniques (Brody, 2003; Brody, Godschalk & Burby, 2003).
(Godschalk et al., 2003; Innes & Gruber, 2005).
Decision-making in American policy and planning has long relied on scientific
expertise for guidance. As a consequence, the implementation of citizen participation has
often been a difficult concept for many policy makers to grasp and has frequently been
excluded. In many cases the scope of relevant stakeholders seems infinitely large. There
may exist such a wide variety of methods for involving citizens in planning that selecting
the most appropriate approach can seem like an insurmountable challenge. For this
53


reason, Arnsteins ladder is a useful framework for identifying appropriate approaches to
public participation that are based on larger community and project goals.
Local Knowledge
Santa Fe Institute founder George Cowan writes that that the science of the 21st
century will be the science of complexity (Waldrop, 1992). Traditionally, reductionism
has been the way in which scientists have attempted to understand problems. The idea
behind reductionism is that if we understand the parts, then we can understand how the
system works as a whole. In the context of planning, however, it has become increasingly
clear that the rational model of understanding systems no longer reaches desired
outcomes. Why? Some systems cannot be evaluated using the rational, reductionist
model because the way in which the components of the system interact is key to
understanding how the whole system works. Systems like these are called complex
systems. Complexity theory emerged as a framework for understanding how complex
systems function and change. A number of disciplinary perspectives apply complexity
theory from ecologists and social scientists, to economists and computer scientists.
Over time, new understandings of complex systems emerged including the
concept of complex adaptive systems. Planning systems and the social-ecological
communities functioning within them have been described as complex adaptive systems,
defined by changing environments, numerous stakeholders and values, and interacting
parts. Hopkins (2001) writes that planning is only useful to complex adaptive systems if
the process is integrated as a functional element within the system rather than as an
external means for achieving control. This is because complex adaptive systems achieve
54


higher levels of performance when innovation and dramatic shifts in activity are able to
occur. Innovation and adaptation depend, however, on adequate transfers of information
and diverse forms of knowledge between interacting stakeholders.
Innes and Booher (1999) stress that in order to understand the capacity of a
complex adaptive system to learn, build shared meaning, and be resilient to unexpected
shocks, it is critical to evaluate issues of power, networks, and self-organizing that are
inherent to the system (Innes and Booher, 1999). Network theory has merged with
planning research on the subject of hazards management and mitigation planning. For
example, increasing drought risks and uncertainty about exposure due to climate change
have propelled fire management to the forefront of U.S. emergency management
agendas. Goldsteins (2008) work on adaptive learning, collective memory, and fire risk
highlights the potential for networks to impact residential knowledge and local resilience.
Goldsteins study found that in the case of the San Diego Fire Recovery Network
(SDFRN), it was the local networks diversification of responses to environmental
change and uncertainty that contributed to increased community resilience. The informal
network provided the community with a range of alternative social-ecological
relationships and configurations. As a result, it could offer flexible options for
reconfiguration without causing catastrophic change to the system. To put it another way,
the network enhanced the systems ability to apply flexible interventions and persist at
the edge of chaos (Goldstein, 2008).
Researchers have also explored the implications of inflexible management
regimes on complex adaptive systems. In the context of post-disaster recovery planning,
research shows that state bureaucracies rarely increase post-disaster social-ecological
55


resilience (Adger et al., 2005; Folke et al. 2005). This is attributed to the governmental
tendency to focus on a return to the status-quo after a disaster event. In contrast,
informal, community networks are more likely to develop a diversity of innovative
responses to rapid change and alternative states of being, thus enhancing community
resilience (Folke et al, 2005; Goldstein, 2008). Goldstein argues that informal networks,
or skunkworks, are free from obligation, scrutiny, or pressure (2008). Therefore, they
are able to innovate and devise more creative, flexible solutions that transcend
organizational barriers and are adaptable to a diversity of unexpected scenarios.
In addition to evaluating how new forms of knowledge and flexible stakeholder
networks are developed during participatory planning, hazards and communicative
planning researchers have explored the potential of participatory processes to mobilize
social memory and increase local resilience (Folke et al., 2005; Goldstein, 2009). Social
memory is a part of every human societys cultural capital and effectively links past
experiences with present (and future) plans and policies (Folke et al., 2005). Although it
draws on experience, it also allows for innovation, creative adaptation, and
experimentation within a larger framework of accumulated community experience.
In their 2009 mitigation strategy study, Berke, Song, and Stevens found that
citizen participation rates during site plan review increased when planners employed
highly communicative participation techniques (specifically, community forums, goal-
setting workshops, subcommittees, and citizen advisory committees). In comparing the
characteristics of the relatively effective participation techniques with those that were
ineffective, Berke, Song, and Stevens (2009) concluded that the participation of
56


stakeholder groups was higher when planners chose to use discursive, interactive and
personal information transmission strategies.
Moving forward, the term participatory planning process is used to encompass
all collaborative planning and decision-making processes that seek to reach agreement
boy a broad spectrum of stakeholders on a planning process output. For example,
problem definition, goal setting, fact finding, priority setting, policy making or
implementation strategy.
Participatory planning is, at its core, the act of creating new knowledge.
Participants contribute their perspectives to the planning processes and diffuse knowledge
to others in the process (Goldstein, 2009; Innes and Booher, 2010). The inclusion of
knowledge provided by non-expert stakeholders can be beneficial to planning processes
because the perspectives of individuals outside of the professional urban planning sphere
often contribute new solutions that are based on collective memory and are closely
aligned with local contexts. Moreover, local knowledge has the potential to contribute
insights about environment and place that the planning discipline may never have known
or might have forgotten (Burby, 2003; Laurian, 2003; Colten & Sumpter, 2009; Coburn,
2003).
Expertise plays only a small part in the local resilience building process, and
formal, scientific knowledge is complimented by lay, or local, knowledge. As adapted
from Coburn (2003, 421), local knowledge is:
Knowledge of specific relationships, events, characteristics and circumstances, as
well as important understandings of their meaning, in local contexts or settings;
Generally acquired via life experience and mediated through cultural tradition;
57


Evidence of ones eyes that is tested through years/generations of experiences,
rather than a hunch or spontaneous intuition; and
Is made legitimate through community stories, public narratives, street theater and
other public forums, as opposed to professional knowledge which is tested
through peer review, through the media, or in the courts.
In a complex world facing uncertain, diverse, and catastrophic climate change
hazards, Innes and Booher (2008) make the case that adaptability and resilience require
enlisting a full range of knowledges, including those of non-experts and the seldom
heard (Innes and Booher, 2008). In reality, however, local knowledge is often
marginalized, particularly when it is communicated by the poor, politically
underrepresented, or socially disconnected.
Better Plans. Safer Communities?
In the U.S., scholars and practitioners frequently call for broad public
participation in planning as a condition for positive planning outcomes. It is frequently
argued that although it often takes more time and money to involve diverse groups of
stakeholders in the process, the long-term savings associated with communicative hazard
mitigation planning can make up for the initial investment and, overall, successful
collaboration often results in the identification and implementation of more appropriate
mitigation options (Mileti 1999). Only a few studies have empirically tested the
assumption that community representation and participation during the planning process
will lead to greater levels of project implementation and/or better plans. The following
58


section examines the body of work on the relationships between community participation
in planning and the quality and implementation of local hazard mitigation plans.
The Influence of Participation on Implementation and Plan Quality
In the U.S., increasing disaster loss and vulnerability trends over the last three
decades indicate that mitigation planning activity and public policy decisions at all levels
of government have been ineffective at protecting lives and infrastructure from hazards
and their impacts (Burby, 2006; Berke, Song, and Stevens, 2009). Mitigation planning
research shows that successful vulnerability reduction requires the integration of various
planning tools, better coordination with stakeholder agencies, and a more holistic
approach to public participation. This belief extends beyond academia and into the policy
realm. In its mitigation guide, FEMA emphasizes that mitigation planning approaches
must go beyond the scope of traditional emergency management and incorporate areas of
planning, development, economics, education, critical care, and cultural facilities. FEMA
also underscores the importance of including the entire community in local planning
activity (FEMA, 2001). The following sections highlight the empirical benefits of
including the public in hazard mitigation planning activity.
Internal Implementation Factors: The Plan and Its Quality
From the perspective of FEMA, the goal of including non-expert stakeholders in
local hazard mitigation planning is to reduce risk and vulnerability of people and property
to future hazard events. In order for risk reduction goals to be realized, knowledge
sharing and public education must occur as well as implementation of proposed
mitigation actions.
59


Plan implementation outcomes are driven by a number of factors. These factors
include planning agency (Dalton & Burby, 1994), the knowledge and commitment of
developers to plan policies (Laurian et al., 2004; McNamara & Healey, 1984), project
scale, interactions between agencies and developers (Burby et al., 1998), public
involvement and quality of the plan itself (Dalton & Burby, 1994; Berke, Song, and
Stevens, 2009). Although the literature on policy implementation is well developed in the
political science and public administration disciplines, the planning literature contains
relatively few studies related to plan quality the implementation of local plans.
Dalton & Burby (1994) conducted one of the key empirical studies on
implementation outcomes from the planning perspective. Their research investigated the
influence of states on development management. Dalton & Burby (1994) analyzed 176
communities across five states. In addition to confirming the importance of state
mandates on plan quality, the study identified plan quality as a critical factor contributing
to strong plan implementation outcomes (1994).
In an attempt to understand the relationship between hazard mitigation planning
and loss prevention, a specialized segment of planning research developed around the
notion of hazard mitigation planning documents as predictors of plan implementation
outcomes (Talen, 1996; Hoch, 1998; Brody, 2003). Within this body of research, the
concept of plan quality has been used as both an outcome variable for assessing planning
processes and as a predictor variable for assessing the implementation of planned actions
In 2003, Brody (2003) examined the impact of stakeholder participation and
representation on ecosystem management strategies. Brody (2003) tested the relationship
between community participation in the planning process and the quality of the local
60


plans, by looking at the breadth of stakeholder groups involved in planning and at the
presence of specific stakeholders. Statistical results found that the breadth of stakeholder
groups involved in planning, as well as the presence of specific stakeholders (industry,
government and NGOs), does significantly increase the quality of ecosystem
management plans (Brody, 2003).
In 2004, Laurian et al. (2004) conducted a systematic, quantitative analysis of key
determinants of plan implementation. Their analysis was based on 353 permits
implementing six local environmental plans in New Zealand, and on surveys of the
developers who obtained the permits and of the planning agencies that granted the
permits. The research team found that plan implementation in this context was largely
driven by the quality of the plans rather than by the characteristics of the planners
(Laurian et al., 2004).
In 2009, Berke, Song, and Stevens (2009) conducted a study related to plan
implementation outcomes in the context of hazard mitigation and risk reduction. The
research team looked at the incorporation of hazard mitigation strategies into New
Urbanist development projects and found that mitigation implementation was positively
correlated with high levels of community participation during the planning process.
Specifically, public participation during site plan review had a positive effect on the
incorporation of hazard mitigation strategies into future development projects. Across the
sample, the number of hazard mitigation strategies used in the proposed development
projects increased with the number of stakeholder groups involved in the site plan review.
Berke, Song, and Stevens (2009) work supports the hypothesis that greater public
participation increases the likelihood that hazard mitigation actions are implemented
61


during future development decisions. This conclusion is supported by previous research
showing that local governments are more inclined to pursue hazard mitigation efforts
when citizens call attention to hazards, and that the likelihood that citizens will call
attention to hazards increases with the number of stakeholder groups engaged in the
planning process (Burby, 2003; Godschalk et al., 2003).
Further exploration is needed to understand why stakeholder participation may
lead to stronger planning outcomes and what kinds of participation activity lead to better
planning. Considering increasing tends in risk and vulnerability, as well as the amount of
time and money spent on the development of public participation strategies and on
complying with federal and state participation mandates, the hazard mitigation
community has much to gain from this line of inquiry.
Assessing Plan Quality
The current body of literature draws strong positive connections between plan
quality and plan implementation (Laurian et al., 2004; Dalton & Burby, 1994), between
public participation and plan implementation (Berke, Song, and Stevens, 2009; Brody,
Godshcalk, and Burby, 2003; Godschalk and Stiftel 1981; Kaiser, Godschalk and Chapin
1995), and between public participation and plan quality (Brody, 2003). The content
analysis approach to evaluating plan quality has primarily been used to assess the quality
of comprehensive development plans between communities with diverse regulatory and
design approaches (Berke & French, 1994). Although emergency management concerns
and priorities are different from those of urban development projects, many
62


understandings from the analysis of comprehensive plans are relevant to the evaluation of
hazard mitigation plan quality.
Early attempts to evaluate plan quality focused on variations in decision-making
processes used to create several development plans. Hill (1968) created a goals-
achievement matrix as a method for ranking the plan alternatives based on their ability
to achieve the goals laid out in the plan. Litchfield et al. (1975) later expanded upon this
method of plan quality assessment by evaluating how different groups of community
stakeholders were affected by different development plans.
Alterman and Hill (1978) later developed a consistency measure to determine
how well comprehensive plans were being implemented across a range of detailed, site-
specific plans. They determined consistency by comparing the proportions of land that
were designated for the same use in comprehensive plans and in site-specific plans. State
mandates that require local plans to be consistent with state goals began to rely on the
consistency measure as a barometer of plan quality (Knapp & Nelson, 1992). The
consistency concept is limited, however, because jurisdiction consistency with state
mandates or comprehensive plans does not automatically guarantee high plan quality.
A number of important efforts to define the characteristics of high plan quality
emerged from the literature soon after the consistency measure. As a body of work, these
studies established consensus concerning the key planning characteristics that influence
the decisions of local governments the most and, therefore, are most likely to lead to
superior plan implementation outcomes.
Chapin and Kaisers (1979) work is cited as one of the most important studies
related to the definition of core plan quality characteristics. The three core characteristics
63


of plan quality are as follows: a strong factual basis, clearly articulated goals, and
appropriately directed policies/actions. Fact basis refers to the existing local conditions of
a place, previous historical events and existing needs related to the physical and socio-
economic development of a community. Goals represent general aspirations, problem
abatement, and needs that are premised on local values. Finally, policies are general
guides to decisions (or actions) about the location and types of development that are
based on the planning goals (Chapin & Kaiser, 1979).
Empirical studies have applied the three core characteristics of plan quality to
evaluate the ability of impact assessments to affect decision-making in local
governments. In 1978, Fishman evaluated the land use and housing elements of
comprehensive plans in 27 U.S. cities and concluded that the best plans contain specific,
well-articulated goals that are linked to local conditions. Fishmans study also
underscores the need for policy language to be action-oriented. For example, using words
like will or must, rather than might or should was characteristic of development proposals
in the case cities that were more consistent with existing comprehensive plans (Fishman,
1978).
In 1990, Connerly looked at the quality of the housing elements of plans created
ten Florida communities. His study provides several explanations for the high
implementation of proposed planning actions in guiding the development decisions of
local governments. First, he asserts that high quality plans have fact bases that are
characterized by detailed inventories of local housing stocks and assessments of current
and projected housing needs and capacities. Second, high quality plans have
comprehensive goals that address a broad spectrum of housing needs. Third, high quality
64


plans contain clearly stated policies that are specific to the achievement of each goal.
Each of the policies also articulate identified funding sources and implementation
timelines (Connerly, 1990).
The body of research on the relationship between plan quality and local
government plan implementation displays a clear pattern in support of the three plan
quality components initially identified by Chapin & Kaiser (1979). The highest quality
plans are typically those that a) have a strong factual basis that defines local needs, b)
have clear and comprehensive goals that evidence a commitment to local needs, and b)
have specific, action-oriented policies that are geared towards achieving plan goals.
Although creating high quality mitigation plans is not the end goal, understanding how to
create high quality plans is an important intermediary step towards achieving reduced
vulnerability and loss from hazards and disasters.
Linking Participatory Planning with Hazard Mitigation Plan Quality
Even in states with highly progressive hazard mitigation policies and mandates,
many local jurisdictions are currently developing low quality, cookie-cutter hazard
mitigation plans. Hazard mitigation plan quality is substantially low across the country
and, in a number of cases, quite variable even within states. Perhaps the most concerning
trend is that many plans lack meaningful site-specific information about local capacity
and social vulnerability, information that is essential to resilience-building (Schwab,
2011).
The burgeoning body of hazard mitigation planning research has refined the three
core plan quality components developed by Chapin & Kaiser (1979) specifically for use
65


in the hazard mitigation planning context. Indicators for the fact base component of
hazard mitigation plans cover the location and extent of hazard damage, including the
magnitude of hazard events, the location of critical facilities, numbers of exposed
populations, and evacuation clearance times. Indicators for the goals component of the
hazard mitigation plan quality index cover physical hazard impacts (e.g. water quality
impacts and the preservation of natural areas). The goals component includes goals
related to physical and economic impacts of hazards (e.g. any goals to equitably
distribute hazards management costs and reduce property loss), and impacts of hazards
on public interests (e.g. any public safety goals or hazard awareness initiatives). The
policy/actions component addresses actions associated with increasing community
stakeholder awareness of risks and hazards, regulatory actions, mitigation incentives,
actions taken to reduce structural losses, and actions related to post-disaster recovery.
Using the three core plan quality components, Wenger et al. (1980) assessed the
quality of 71 local emergency plans. Through empirical testing, Wenger et al. (1980)
concluded that the majority of pre-disaster emergency planning activities in their sample
were oriented largely towards the production of plans as deliverables for the acquisition
of funding rather than as tools for future development and policy-making. The highest
quality plans in their study were those plans created by frequent community-wide sharing
of information, fact-finding, and proposal generation (Wenger et al., 1980).
Hazards scholars Berke and French (1994) have also evaluated hazard mitigation
plan quality, this time by studying the relationship between quality outcomes and state
planning mandates in five states: California, Texas, Washington, North Carolina, and
Florida. The five selected states provide a range of variation in geographic features as
66


well as in state mandate local planning characteristics. In their study, the characteristics
of two mandate design features (structural features and facilitating features) were
identified in the each of the local plans. Berke and French (1994) then showed gamma
correlations between the structural and facilitating mandate design features, and local
plan quality characteristics. Their research found that both types of mandate features are
strongly related to the strength of policies within local plans, but they have weaker
relations with goals and fact basis.
Berke and French (1994) suggest that the relatively weak relationship between
planning mandate features and fact basis may be attributed to how fact base is
formulated. Federal agencies largely determine the extent to which plans use information,
more so than do state mandates. For example, federal agencies produce detailed flood
insurance rate maps that most communities use for plan making. This finding may
explain why their study found local commitment building to have the highest correlation
with fact basis of all the mandate features. Berke and French suggest that as commitment
to planning increases, communities are more likely to incorporate available mapping data
into plan making (1994).
As previously noted, planning researchers currently measure the fact base
component of hazard mitigation plans by identifying cited information including previous
hazard damages, probability of hazard events and the location of critical facilities.
However, despite the significant role that people play in the ability of a community to be
resilient in the face of disaster, researchers have failed to assess the incorporation of
social vulnerability concerns into their hazard mitigation plan quality assessments.
67


In response to this disconnect between theory and practice, this study has
modified the fact base component of the plan quality assessment tool to include citations
of social vulnerability information in mitigation plan scoring criteria. Specifically, fact
base indicators have been refined to include contextual data on local demographics,
education levels, health services, social services and economy, all of which are important
indicators of local social vulnerability and key factors affecting the disaster resilience of a
community (Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Berke et al., 1998;
Godschalk et al., 1999; Phillips et al., 2009). This methodological strategy is discussed in
detail in Chapter 3.
Placement and Contribution of this Study
This dissertation research expands upon the current body of literature by assessing
how various models of public participation, as well as local social vulnerability
characteristics, affect hazard mitigation plan quality outcomes. This study does not
attempt to identify the etiology or antecedent factors that explain public participation in
hazard mitigation planning. Rather, it takes an exploratory approach to identify emerging
themes related to variations in hazard mitigation plan quality between communities. The
following section describes the conceptual framework that guides this research.
Conceptual Framework
As previously discussed, the literature includes a variety of variables that affect
the impact of public participation on local hazard mitigation plan quality. A conceptual
framework can help integrate these variables and organize analysis by identifying general
sets of variables of interest. To this end, this study integrates three models developed by
68


Callon (1999), and adapted by Yli-Pelkonen & Kohl (2005), to describe the various
degrees of involvement of lay-person stakeholders in the creation of the knowledge and
information on which planning decisions are made.
Grounded in the social-ecological systems literature, Callons (1999) original
models focus on the diversity of possible types of non-expert participation in planning.
The first model, hereafter referred to as Ml, is called the public education model. Ml
describes the opposition between technical and experiential forms of knowledge. In this
model, non-expert stakeholders have nothing to contribute to professional knowledge of
risk, vulnerability or hazard mitigation during plan making. Here, local goal setting,
planning, and decision-making are based solely on the knowledge and agendas of
organizational stakeholders and that knowledge is derived primarily from historical data
and modeling.
Model 2 (M2) is called the public debate model. Here, local non-expert
stakeholders are perceived as having less robust levels of knowledge than professionals.
They are, however, invited to participate (to a limited extent) in goal setting, risk
assessment and mitigation prioritization. In this model, professional knowledge is
enriched by local community information sharing and together, organizational and
community stakeholders create a framework for decision-making.
Callon (1999) breaks up the M2 model into two sub-models in order to describe
different types of participation in more detail. The two sub-models are: M2a, the public
hearing model and M2b, the public representation model. The public hearing model
(M2a) describes interactions in which hazard mitigation planners collect information
directly from participants. In this model, the aim of the collaborative planning process is
69


for community knowledge to directly impact planning outcomes (Callon, 1999; Yli-
Pelkonen & Kohl, 2005).
In the public representation model (M2b) community knowledge is heard
indirectly through the use of community representatives. These representatives can be,
for example, homeowner associations or local NGOs. In this model of participation,
community knowledge can influence planning decisions; however, because planners are
channeling it through community representatives, there is a greater risk of the
information being misinterpreted.
Model 3 (M3), the co-production of knowledge model, represents circumstances
in which non-expert stakeholders are actively involved in the information collection, goal
setting, and decision-making that shapes local plans. In this model, community input is
treated as equally important to planning and decision-making as is professional input. In
some circumstances, community perspectives and experiential knowledge are even
treated as more important. This can happen in situations where professional knowledge
on a topic is scarce (i.e. when plans require information about social factors related to a
communitys risk, vulnerability, or capacity). Here, local community knowledge becomes
an essential part of information production and decision-making (Yli-Pelkonen & Kohl,
2005).
There exists a fourth model along the spectrum of participation. Model 4 (M4), or
the community construction of knowledge model, illustrates a participation model in
which hazard mitigation planners rely entirely on information, understandings, and
priorities based on knowledge from local communities to inform their mitigation plan
proposal. When hazard mitigation planners attempt to incorporate information about
70


social vulnerability and/or community capacity into local plans (i.e. social networks,
information about community gatekeepers, or adaptive behavior based on experiential
learning) professional knowledge about these social determinants of risk is largely non-
existent. Therefore, experts must defer to local stakeholders, including socially
marginalized or unseen groups, to construct an understanding of local social issues
related to hazard risk and vulnerability (Wisner, 2006; Folke et al., 2003, 2005; Folke,
2006).
For the purpose of this study, the four models described above provide a
functional framework by which to assess local hazard mitigation planning practice at the
city scale. Moving forward, the models are used to categorize individual hazard
mitigation planning schemes and to guide our evaluation of the relationships between
stakeholder participation, social vulnerability factors, and plan quality outcomes.
The framework for this study, adapted by planning and social-ecological systems
scholars Yli-Pelkonen & Kohl (2005), brings Callons four models together into a larger
conceptual model of stakeholder participation during local hazard mitigation planning.
The conceptual framework illustrates the interactions through which lay-person
stakeholders participate in information production, goal setting and decision-making
during hazard mitigation planning (Callon, 1999; Yli-Pelkonen & Kohl, 2005).
Below, Figure 5 illustrates the conceptual framework used in this study for
categorizing stakeholder participation during local hazard mitigation planning. Models
Ml through M4 are characterized by the specific participatory planning strategies used
during the local planning process. The interactions between local community/citizen
stakeholders, organizational stakeholders/technical experts and mitigation planning teams
71


flow according to the four vectors illustrated in Figure 5. The arrows A, B, C and D show
the interactions between the participants and identify the key social process of
collaborative planning. The dashed lines for interactions B, C and D illustrate that they
are conditional interactions.
Organizational
Stakeholders and
Technical Experts
Ml A strong interaction
B considered only at the opinion level and separately front A
C exists only when technical information educates local citizens
M2a.b A strong interaction
B considered moderately as complimentary to A
C exists, participation can enrich technical perspectives to some extent
JVJ3 A and B equally strong. B sometimes stronger in certain cases
C strong interaction, direction of relationship is case-dependent
M4 A some interaction
B and C strong interaction, direction of information collection largely from
community to mitigation planning team
*
A
f Local Community
^ /Citizen Stakeholders
Figure 5: Framework for Stakeholder Participation During Local Hazard Mitigation
Planning
In this study, I evaluate interactions A through D in order to attribute a
participatory planning model to each of the 34 local hazard mitigation plans included in
the sample frame. Based on which interactions are present during a citys mitigation 2
2 Adapted from Yli-Pelkonen & Kohl (2005)
72


planning process, one of the four participatory planning models Ml, M2a/M2b and
M3, M4 will be attributed to each of the 34 local hazard mitigation plans.
The framework of stakeholder participation during local hazard mitigation
planning can be connected back to the existing literature related to participatory planning
and mitigation planning outcomes. In 2009, Deyle & Slotterback (2009) found positive
planning outcomes in counties where interactions A and B were equally strong (planning
model M3). Deyle & Slotterback used a quasiexperiemental research design to observe
and evaluate the mitigation plans and processes in eight Florida counties. Using a pretest-
posttest design, their study uncovered higher levels of mutual understanding of planning
problems (and of solutions) in counties where community stakeholders were able to
directly incorporate their knowledge and ideas to the formal planning deliberations. This
finding is significant because strong mutual understanding among stakeholder groups has
been shown to generate stronger constituencies for both plan adoption and
implementation (Deyle and Slotterback, 2009; Brody, Godshcalk, and Burby 2003;
Godschalk and Stiftel 1981; Kaiser, Godschalk and Chapin 1995; May et al. 1996).
Pearce (2003) presents the concept of sustainable hazard mitigation as the next
step for the field of disaster management. In order to achieve sustainable hazard
mitigation, Pearce (2003) argues that it is necessary to closely integrate mitigation and
participatory community planning. To illustrate her point, Pearce presents a case study of
participatory mitigation planning in Californias Portola Valley. Pearce describes Portola
Valleys approach as an example of successful sustainable hazard mitigation in action,
exemplified by a high degree of public participation in planning (Pearce, 2003). Based on
73


the conceptual model of stakeholder participation presented here, Portola Valley applied
the M2b model of public participation.
It is important to note that the participation models are not always contradictory.
They can sometimes occur together and complement one another. Moreover, in all of
these models, even if participants are heard, it is possible for elements of both types of
information and knowledge received by planners through interactions A and B to be lost
in the process and not transmitted to the final plan (Innes & Booher, 2007). For example,
this can happen as the mitigation planning team filters important information from other
stakeholders during the final stages of plan making (interaction D) (Yli-Pelkonen &
Kohl, 2005). To address this element of the conceptual framework, this study includes in-
depth interviews with mitigation planning team members to assess the impact of the
filter on participation models and plan quality outcomes.
Conclusion
Under the Stafford Act and DMA 2000 decision makers are required to base
hazard mitigation planning decisions not only on adequate technical information from
risk assessment studies, but also on the priorities and knowledge of local residents,
community stakeholder groups and other stakeholders who can provide information on an
areas vulnerability and experience with previous disasters. A key theme that this study
explores is how local knowledge is integrated into planning and decision making.
One of the main propositions of this study is that the engagement of local
stakeholders in decision-making and vulnerability assessment is critical to reaching
disaster resiliency goals. Although the process of collaborative mitigation planning is
74


often associated with higher initial costs and extended project timelines, the up-front
investment is cost effective when it boils down to effective plan implementation, long-
term policy agreements, and safe, resilient communities (Godschalk et al., 1994).
Moreover, previous research has found that communicative processes including
collaboration, dialogue, and consensus building generate myriad secondary benefits.
These benefits include the creation of new forms of knowledge and political capital
(Forester, 1999; Innes and Booher, 1999b), increased levels of trust, stronger
commitments to plan and policy implementation, and increased social capital and
community resilience (Burby, 2003; Innes, 1996; Innes et al., 1994; Creighton, 1992;
Wondolleck and Yaffee, 2000; Forester, 1999; Moore, 1995).
Ultimately, the conclusions of this dissertation will have practical significance to
local, state and federal planners, emergency managers and policy makers seeking identify
ways of improving local citizen participation in hazard mitigation planning.
Understanding how participation affects local hazard mitigation plan quality, and how
social vulnerability affects participation, will provide important insights into how local
hazard mitigation plans can be strengthened to more adequately address the root causes
behind increasing vulnerability and risk.
75


CHAPTER III
METHODS
Introduction
This study examines the relationship between models of participatory planning
and local hazard mitigation plan quality and explores the factors that drive variations in
plan quality and content. The guiding questions for this research study are:
1) What is the quality of local hazard mitigation plans created in the northern
California Bay-Delta region and to what extent has the public participated in the
plan-making process?
2) What key relationships exist between local community characteristics, public
participation in planning and hazard mitigation plan quality outcomes?
The following chapter describes the methods by which the data to answer these
questions was collected and analyzed. It begins with a description of the selected study
area and explains the sample selection process in detail. It then describes the
characteristics of participants identified for the interview portion of the study and how
they were selected.
The data collection tools used for this research are also described in detail and
include the following: the plan quality assessment instrument and protocol, interview
guides, and a reflexive journal. A thorough description of the steps involved in data
collection is provided and efforts to protect the reliability of findings and the validity of
inferences are detailed.
76


Study Area
The state of California was selected as the location for this study. California is
unique in that it has a reputation as a national leader in the mitigation of hazard impacts
associated with earthquakes, fires, floods, climate change and other natural disasters.
Recently, this reputation has been validated by the U.S. Department of Homeland
Security through FEMA. In December of 2007, FEMA designated the 2007 State of
California Multi-Hazard Mitigation Plan as an Enhanced State Mitigation Plan. Having
an enhanced mitigation plan increases the states level of federal hazard mitigation grant
funding from 15% to 20% of the total individual and public assistance funds spent by
FEMA after each Presidential Disaster Declaration. At the time, only eleven other states
had mitigation plans of enhanced status. The designation officially distinguishes
Californias leadership in reducing the risk of death, injury and property loss from
hazards.
The California Emergency Management Agencys (Cal EMA) Hazard Mitigation
Program (HMP) administers the LHMP program for the state. In this role, Cal EMA
supports and assists local governments in the development of their LHMPs and heads the
review process for all plans before they are forwarded to FEMA for approval. Local
governments in California must have an approved LHMP to be eligible for certain
mitigation programs authorized under the Stafford Act and, ultimately, the goal of the
LHMP program is for all local governments in California to have FEMA-approved
LHMPS.
This study focuses specifically on the northern California Bay Area/Delta region
of California. This region of northern California provides a particularly information-rich
77


setting to study how diverse social factors affect local hazard mitigation plan quality. For
one, the region is demographically diverse and its levels of social vulnerability vary
greatly by county. Moreover, it boasts a diversity of geographic environments. As a
result, communities throughout the region are simultaneously managing and planning for
a variety of hazard risks.
Examining plans within the selected study area provides an opportunity to
evaluate the effects of public participation on plan quality while controlling for state-
specific policy factors, regional economics, and climate. Ultimately, the sample location
was selected not to ensure comparability among places in different states, but to provide
insight into how public participation affects LHMP plan quality in a state that is taking a
singularly progressive stance towards disaster risk reduction and planning.
Sample Selection
The sample frame for this study consists of all jurisdictions within the study area
with completed LHMPs as of July 2011 (including those in the FEMA approval process
as of April 2011), totaling 215 jurisdictions. This information was obtained from a list of
all local jurisdictions in the state of California with FEMA approved LHMPs (including
those plans in the approval pipeline). The list was provided by Cal EMA and is current as
of April 2011. There are twelve counties represented in the sample frame. They include
Social vulnerability is commonly calculated using the Social Vulnerability Index
(SoVI). SoVI is calculated at the county scale by scoring locations across 11 factors
(Cutter et al., 2003).
78


the nine counties represented by the Association of Bay Area Governments (ABAG) -
Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano,
and Sonoma and the three delta counties of Placer, Yolo and Sacramento.
Figure 6: Map of California Counties Included in Study Area
The first stage of the sampling strategy was to limit the sample of LHMPs to the
community scale by excluding all special districts, tribal entities, and agency
jurisdictions. This study focuses specifically on city and county scale jurisdictions
because they are the human settlements that are most vulnerable to both rapid- and slow-
onset disasters. Not only are cities characterized by high population densities and
concentrations of political, economic and social resources, but they are also characterized
79


by a strong dependency on infrastructure and services. Moreover, rapid urbanization and
population growth make it difficult for decision makers and communities to respond
quickly and plan for change. As a result, cities are the places where ecological, economic,
and social disparities manifest most notably. This research project focuses deliberately on
cities because these are the communities that offer the most to learn and the most to gain.
By limiting the sample to city-scale hazard mitigation plans, the list of LHMPs was
paired down to 98.
Next, the sample of 98 LHMPs was limited to those written for cities with
populations between 10,000 and 200,000 people. This paired the sample down further to
a total of 71 plans. This population range was chosen for the following reasons: cities
with a population of 10,000 or more provide a wider variety of functions and services
than do smaller cities and it is believed that large cities, such as San Jose (with a
population of 965,000), have very different contextual factors and may skew the sample.
The third stage of the sampling strategy was to limit the sample frame to plans
that were available online as of July 1, 2011. Transparency, including the availability of
planning documents to the public, has been identified by FEMA as important components
for successful loss and vulnerability reduction: An effective and open public involvement
process ensures that all citizens understand risks and vulnerabilities so that they will work
with the jurisdiction and support policies, actions and tools that over the long-term will
lead to a reduction in future losses. Moreover, continued public involvement in the
planning process is required by FEMA for LHMP approval. By sampling plans that are
available to the general public, this study is effectively evaluating LHMPs from cities that
80


have met the most basic requirement of the 44 CFR 201.6(c)(4)(iii): the provision of
opportunities for continued public involvement in the planning process.
After the three sampling phases, a final sample of 34 LHMPs remained. Table 4
depicts the 34 city plans that were selected for analysis based on the sampling techniques
outlined above:
Table 4: Sample Cities
County City Planning Scale* Population* *
Placer Roseville SJ 109,497
Alameda Berkeley SJ 100,877
Napa Napa SJ 74,071
San Mateo Redwood City SJ 73,099
Marin Mill Valley SJ 13,189
Marin Larkspur SJ 11,647
Sonoma Santa Rosa MJ-Region 154,017
Santa Clara Sunnyvale MJ-Region 130,256
Santa Clara Santa Clara MJ-Region 108,075
Solano Fairfield MJ-Region 103,305
Solano Vacaville MJ-Region 91,828
Santa Clara Gilroy MJ-Region 48,158
Santa Clara Campbell MJ-Region 38,067
Santa Clara Morgan Hill MJ-Region 37,099
Santa Clara Saratoga MJ-Region 29,974
Contra Costa Concord MJ-Region 120,775
San Mateo Burlingame MJ-Region 27,363
81


Table 4 (cont.)
County City Planning Scale* Population* *
Sonoma Windsor MJ-Region 25,158
Contra Costa Orinda MJ-Region 18,198
San Mateo Half Moon Bay MJ-Region 12,271
Marin San Anselmo MJ-Region 11,924
San Mateo Hillsborough MJ-Region 10,651
Sacramento Elk Grove MJ-County 130,007
Sacramento Citrus Heights MJ-County 84,269
Sacramento Folsom MJ-County 66,399
Yolo Davis MJ-County 61,866
Sacramento Rancho Cordova MJ-County 60,008
Yolo Woodland MJ-County 53,069
Placer Rocklin MJ-County 51,281
Yolo West Sacramento MJ-County 45,504
Placer Lincoln MJ-County 40,382
Sacramento Galt MJ-County 23,535
Napa American Canyon MJ-County 15,943
Placer Auburn MJ-County 13,068
* Data from city LH1V P documents
** Data from 2005-2009 American Community Survey 5-year Estimates
Each of the 34 jurisdictions selected have adhered to the participation
requirements of the state local hazard mitigation planning mandates, all jurisdictions are
vulnerable to threats from natural hazards, and all jurisdictions have prepared updated
plans since 2004.
82


The plan quality of each of the LHMPs listed above was measured using an
analysis tool that incorporates hazard mitigation and social vulnerability measures into
pre-existing conceptions of what constitutes a high quality plan. Ultimately, based on the
results of the plan quality analysis, a select number of cities were chosen for in-depth
interviews. The interview methodology and analysis strategies are discussed in detail in
Section 3.3.
Data Collection
This dissertation uses both qualitative and quantitative research methods to
answer the proposed research questions. Data collection has been broken into three
methodological stages leading up to the synthesis of results (Figure 7).
(S) Principal
^ Component
OJ)Analysis &
C3 Regression Analysis
GO
Figure 7: Methodological Framework
Stage 1 consists of a plan quality evaluation based on a quantitative hazard
mitigation plan quality index. Additionally, Stage 1 includes an assessment of public
participation models applied during planning for each of the 34 LHMPs. This assessment
is based on the four models of public participation explained in the participation
framework in Chapter 2.
Stage 2 involves the collection of local community characteristic data from U.S.
Census and SHELDUS databases including data on local social vulnerability,
rr) Participant
^ Interviews and
(^Qualitative Analysis
a
+-
h Plan Quality and
^ Participation
(^Evaluation
a
>
83


socioeconomics, and hazard exposure. The specific social vulnerability, socioeconomic,
and hazard exposure variables were selected based on findings from previous studies
related to hazard mitigation plan quality and public participation (Brody, 2003; Berke &
French, 1994; Berke, 1996; Cutter et al., 2003; Davis, Cook & Cohen, 2005; Chandra et
al., 2010; Birkmann, 2006).
Stand-alone mitigation plans, or LHMPs, are often prepared in the aftermath of a
disaster when the sense of crisis leads to a demand for immediate action (Burby, 1998).
Although none of the local hazard mitigation plans created in the sample frame were
written in direct response to a disaster, hazard exposure (as measured by previous losses
from disasters) has been shown to be an important driver of community participation in
planning. Sometimes described as the window of opportunity, the post disaster period
is a time when communities express the greatest levels of public support for hazard
mitigation and preparedness activity (Burby, 1998). Moreover, the memories of repeated
hazard events prompt some communities to prioritize mitigation planning to a greater
extent than communities that may have high hazard risks, but no history of loss.
In addition to hazard exposure, socioeconomic and social vulnerability factors
have also been identified as potential constraints (and drivers) of mitigation policies and
actions (Ollenburger & Tobin, 1998; Tobin, 1999). Researchers have theorized that
socioeconomic and cultural factors can affect attitudes towards mitigation and recovery
efforts and that they can act as filters in the hazard mitigation planning system (Tobin,
1999). A principal component analysis has been performed on the selected community
characteristic data in order to reduce the many demographic variables to a smaller
number of predictor variables (or factors) for use in the regression analysis.
84


Empirical research has not yet addressed the relationship between
socioeconomics, social vulnerability factors and plan quality. Nor has the literature
addressed how these contextual factors impact public participation during local hazard
mitigation planning. Using SAS software, Stage 2 involves running a number of
regression analyses on the community characteristic variables, public participation data
and plan quality data. The expectation is that some of these factors may constrain the
ability of local communities to participate in local hazard mitigation planning processes.
The regression analyses allow me to explore the nature and form and of the relationships
between plan quality, participation models and local community characteristics.
Stage 3 includes qualitative, in-depth interviews with organizational and local
community stakeholders from the local hazard mitigation planning teams from select
cities in the sample frame. Driven by the results of the regression analyses, the detailed
case studies were designed to provide a more robust understanding of the dynamics of
public participation, plan quality and social vulnerability than is possible from regression
analysis or plan quality assessment alone.
Stage 1: Plan Quality Evaluation
In the following sections, the methodological approaches for the three stages of
data collection and analysis are described in detail.
Methodology
Building on the methods from previous studies of plan quality, this study
measures hazard mitigation plan quality by assessing three equally weighted plan
components in a plan quality index: clearly articulated goals, appropriately directed
85


policies/actions, and a strong factual basis (Brody, 2003; Godschalk, Kaiser, & Berke,
1998; Berke et al., 1998; Godschalk et al., 1999).
Previous research has contributed to the identification of a set of qualitative
indicators for each of the three components to create a robust conception of plan quality.
In the context of hazard mitigation planning, indicators for the goals component of the
plan quality index cover physical impacts of hazards (e.g. assessment of water quality
impacts and the preservation of natural areas), economic impacts of hazards (e.g. any
goals to equitable distribute hazards management costs and reduce property loss), and
impacts on public interests (e.g. any public safety goals or hazard awareness initiatives).
The policy component of plan quality addresses actions associated with increasing
community stakeholder awareness of hazards and mitigation, explicit regulatory actions,
mitigation incentives, the reduction of structural losses, and recovery concerns.
Previous studies have measured the fact base component of plan quality by
assessing the range of background information upon which planning decisions, goals and
actions are predicated. In the context of hazard mitigation planning, the range of
information includes data about the location and extent of hazard damage, including the
magnitude of the hazard event, exposed populations and evacuation clearance times. For
the purpose of this study, the fact base component has been augmented to include social
vulnerability information. Here, the fact base component includes background data on
local demographics, education levels, health services, social services and economy all of
which are commonly used indicators of local social vulnerability (Phillips et ah, 2009).
Each plan quality indicator was measured on a 0 2 ordinal scale. Zero (0) is
not identified or mentioned, one (1) is suggested or identified but not detailed, and
86


Full Text

PAGE 1

i EVALUATING LOCAL HAZARD MITIGATION PLANNING: QUALITY, PUBLIC PARTICIPATION, AND SOCIAL VULNERABILITY by ENESSA HOLLAND JANES B.S., Stanford University 2005 M.S. Stanford University 2006 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Design and Planning 2013

PAGE 2

ii 2013 ENESSA HOLLAND JANES ALL RIGHTS RESERVE D

PAGE 3

iii This thesis for the Doctor of Philosophy degree by Enessa Holland Janes has been approved for Design and Planning Program by Deborah Thomas, Chair Jeremy Nmeth, Advisor Brian Muller Bruce Goldstein Gregory Thomas November 8, 2013

PAGE 4

iv Janes, Enessa Holland (Ph.D., Design and Planning) Evaluating Local Hazard Mitigation Planning: Quality, Public Participation and Social Vulnerability Thesis directed by Associate Professor Jeremy Nmeth ABSTRACT Despite abundant amounts of data and significant advances in communications and engineering technology, global trends in hazard vulnerability and disaster losses continue to rise. Moreover, the most marginalized and disadvantaged members of society bear th e brunt of these losses and impacts. The United States emergency management community has responded to increasing trends in disaster losses by underscoring the importance of hazard mitigation at the local scale. Although the planning literature acknowledges that plan quality is a key factor in determining planning outcomes, and that engaging local community stakeholders in the planning process is critical to creating high qual ity mitigation plans, little empirical research has been done to evaluate the nuances of the local hazard mitigation planning process. Moreover, the literature has not evaluated how social vulnerability factors affect mitigation planning processes and plan quality outcomes. This study examines local hazard mitigation plan quality, models of participatory planning, a nd social vulnerability in the n orthern California Bay Delta region. It applies a mixed methods approach to identify relationships between loc al social vulnerability, public participation, and plan quality outcomes in 34 communities. The results of this study show that neither public participati on nor social vulnerability was associated with plan quality outcomes in the study region. Additional ly,

PAGE 5

v social vulnerability had no relationship to public participation levels. Planning scale emerged as an important driver of plan quality and single jurisdiction planning efforts were positively correlated with plan quality outcomes. The case studies prov ide further insight into these findings and relationships and present a best practices model for knowledge sharing and decision making for local hazard mitigation planning. The form and content of this abstract are approved. I recommend its publication. Approved: Jeremy Nmeth

PAGE 6

vi DEDICATION I dedicate this work to my grandmother, Doris Janes, the strongest and most inspiring woman I know. Her contagious positivity, warm advice, and late night phone calls made this journey a success.

PAGE 7

vii ACKNOWLEDGMENTS I would like to thank my advisor, Jeremy Nmeth for his enthusiasm, support, and resolute dedication to the strength of my research. I would also like to thank my committee chair, Deborah Thomas whose selfless time and care kept me on track and motivated Thank you, Deb, for sharing your intellectual passions with me and for your exemplary teaching and mentorship. It is such an honor to call you my advisor and my friend. I would also like to tha nk the other members of my dissertation committee Bruce Goldstein, Brian Muller, and Gregory Simon for their ongoing encouragement and const ructive comments and criticisms Thank you to the National Science Foundation, whose generous support through the IGERT Program in Sustainable U rban Infrastructure made this research possible. I extend my thanks to Greg Kinney for his support during the final stages of my doctoral marathon Thank you for generously and enthusiastically trading your afternoons with me at your Anschutz office I am grateful to my family my father, Craig, my mothe r, Kitty, and my sister, Janice for their steadfast confidence in me and for serving as examples of wise well balanced individuals. Thank you for reminding me of what is truly important in life. F inally, I would like to t hank my husband, David Belsheim, for his encouragement humor and unwavering patience Thank you for making me laugh when times were tough and for pushing me to see every challenge as an opportunity. I love you forever.

PAGE 8

viii T ABLE OF CONTENTS CHAPTER I. INTRODUCTION ................................ ................................ ................................ ........................ 1 The Hazards and Disasters Problem ................................ ................................ ...................... 1 Local Planning for Disaster Vulnerability Reduction ................................ ............................ 2 Research Questions ................................ ................................ ................................ ................ 5 Research Impacts ................................ ................................ ................................ ................... 7 II. LITERATURE REVIEW ................................ ................................ ................................ ............ 8 Introduction ................................ ................................ ................................ ............................ 8 Hazards, Disasters, and Vulnerability ................................ ................................ .................... 8 Measuring Social Vulnerability ................................ ................................ ........................... 13 Building Resilience into Emergency Management ................................ .............................. 20 Hazard Mitigation: Planning, Policy and Practice ................................ ............................... 29 The Comprehensive City Plan ................................ ................................ ................................ .... 29 Hazard Mitigation in the United States ................................ ................................ ...................... 36 Vulnerability Reduction through Hazard Mitigation ................................ ................................ .. 39 Assessing Vulnerability ................................ ................................ ................................ .............. 41 Engaging the Public through Participatory Pla nning ................................ ........................... 44 The Theory and Practice of Communicative Planning ................................ ............................... 46 Participatory Planning: Practice and Theory ................................ ................................ .............. 49 Local Knowledge ................................ ................................ ................................ ........................ 54 Better Plans, Safer Communities? ................................ ................................ ....................... 58 The Influence of Participation on Implementation and Plan Quality ................................ ......... 59 Internal Implementation Factors: The Plan a nd Its Quality ................................ ........................ 59 Assessing Plan Quality ................................ ................................ ................................ ............... 62

PAGE 9

ix Linking Participatory Pla nning with Hazard Mitigation Plan Quality ................................ ....... 65 Placement and Contribution of this Study ................................ ................................ ........... 68 Conclusion ................................ ................................ ................................ ........................... 74 III. METHODS ................................ ................................ ................................ .............................. 76 Introduction ................................ ................................ ................................ .......................... 76 Study Area ................................ ................................ ................................ ........................... 77 Sample Sele ction ................................ ................................ ................................ ........................ 78 Data Collection ................................ ................................ ................................ ........................... 83 Stage 1: Plan Quality Evaluation ................................ ................................ ......................... 85 Methodology ................................ ................................ ................................ ............................... 85 Assessing Models of Community Participation ................................ ................................ ......... 88 Stage 2: Principal Component and Regression Analysis ................................ ..................... 92 Stage 3: Interviews and Qualitative Analysis ................................ ................................ ...... 94 Interview Sampling Strategy ................................ ................................ ................................ ...... 94 Interviewee Selection ................................ ................................ ................................ ................. 95 Interview Data Collection and Storage ................................ ................................ ....................... 97 Reflexive Journal ................................ ................................ ................................ ........................ 98 Data Analysis and Validity ................................ ................................ ................................ .. 99 Plan Quality Analysis ................................ ................................ ................................ ................. 99 Qualitative Interview Analysis ................................ ................................ ................................ 100 Validity ................................ ................................ ................................ ................................ ..... 100 IV. EVALUATING THE RELATIONSHIPS BETWEEN PARTICIPATION, COMMUNITY CHARACTERISTICS, AND LOCAL HAZARD MITIGATION PLAN QUALITY ................................ ................................ ................................ .................. 102 Introduction ................................ ................................ ................................ ........................ 102 Methods ................................ ................................ ................................ .............................. 102

PAGE 10

x Plan Quality Index ................................ ................................ ................................ .................... 102 Methodology for Assessing Models of Commu nity Participation ................................ ........... 110 Data ................................ ................................ ................................ ................................ .... 113 Plan Quality and Levels of Public Participation ................................ ................................ ....... 113 Community Characteristic Data ................................ ................................ ............................... 117 Data Reduction Using Principal Components Analysis ................................ .................... 127 Multiple Regression Analysis ................................ ................................ ............................ 143 Results and Discussion ................................ ................................ ................................ ...... 157 V. A QUALITATIVE EVALUATION OF LOCAL MITIGATION PLANNING .................... 162 Introduction ................................ ................................ ................................ ........................ 162 Methods ................................ ................................ ................................ .............................. 163 Interview Guides ................................ ................................ ................................ ....................... 163 Data Collection and Storage ................................ ................................ ................................ ..... 164 Reflexive Journal ................................ ................................ ................................ ...................... 165 Case Selection ................................ ................................ ................................ ........................... 165 Interviewee Selection ................................ ................................ ................................ ............... 170 Data and Limitations ................................ ................................ ................................ ................ 172 Planning Scale and Participation during Local Hazard Mitigation Planning: Six Case Studies ................................ ................................ ................................ ....................... 173 Planning Scale: Singl e and Multi Jurisdiction Local Hazard Mitigation Plans ....................... 177 Regional Scale Multi Jurisdiction Plan: Windsor, CA ................................ ............................. 178 The Plan ................................ ................................ ................................ ................................ .... 178 County Scale Multi Jurisdiction Plan: Citrus Heights, CA ................................ ...................... 182 The Plan ................................ ................................ ................................ ................................ .... 183 City Scale Single Jurisdiction Plan: Roseville, CA ................................ ................................ .. 187 The Plan ................................ ................................ ................................ ................................ .... 189

PAGE 11

xi Results ................................ ................................ ................................ ................................ ...... 195 Public Participation during Local Hazard Mitigation Planning ................................ ................ 197 Low Public Participation: Woodland and West Sacramento, CA ................................ ............ 198 The Plan ................................ ................................ ................................ ................................ .... 199 High Public Participation: Napa, CA ................................ ................................ ........................ 202 The Plan ................................ ................................ ................................ ................................ .... 203 Results ................................ ................................ ................................ ................................ ...... 207 Discussion ................................ ................................ ................................ .......................... 210 VI. CONCLUSIONS AND DISCUSSION ................................ ................................ ................. 217 Introduction ................................ ................................ ................................ ........................ 217 Building Disaster Resilience through Local Hazard Mitigation Planning ......................... 217 Research Questions ................................ ................................ ................................ ............ 220 Community Case Studies ................................ ................................ ................................ ... 224 Contributions and Directions for Future Research ................................ ............................ 229 REFERENCES ................................ ................................ ................................ ............................ 235 APPENDIX ................................ ................................ ................................ ................................ .. 256

PAGE 12

xii LIST OF TABLES Table 1: Definitions of Resilience ................................ ................................ ................................ ........... 23 2: The Role of the Federal Government in Disaster Loss Reduction ................................ ............ 37 3: Models of U.S. Hazard Vulnerability Assessment ................................ ................................ .... 43 4: Sample Cities ................................ ................................ ................................ ............................. 81 5: Models of Community Participation in Hazard Mitigation Planning ................................ ........ 91 6: Community Characteristic Variables ................................ ................................ ......................... 93 7: Summary of Regression Models ................................ ................................ ................................ 94 8: Naturalistic Techniques for Ensuring Comp rehensive Research Quality ................................ 101 9: Plan Quality Index ................................ ................................ ................................ ................... 105 10: Evaluation Criteria for Models of Community Participation ................................ ................. 111 11.Cities Ranked by Plan Quality Score ................................ ................................ ...................... 114 12: Descriptive Statistics ................................ ................................ ................................ .............. 116 13: Community Characteristic Variables ................................ ................................ ..................... 118 14: Descriptive Statistics for Hazard Exposure Data ................................ ................................ ... 121 15: Summary of Multi Hazard Exposure and Losses by City ................................ ..................... 122 16: Summary of Community Characteristic Variables ................................ ................................ 125 17: Descriptive Statistics ................................ ................................ ................................ .............. 128 18: Summary of PCA Variables ................................ ................................ ................................ .. 130 19: Principal Component Analysis Runs ................................ ................................ ..................... 132 20: Total variance explained ................................ ................................ ................................ ........ 134 21: Varimax Rotated Component Matrix ................................ ................................ .................... 135 22: Summary of PCA results ................................ ................................ ................................ ....... 136 23: Sample City Factor Loadings ................................ ................................ ................................ 138

PAGE 13

xii i 24: Moments ................................ ................................ ................................ ................................ 144 25: Basic Statistical Measures ................................ ................................ ................................ ..... 144 26: Tests for Location ................................ ................................ ................................ .................. 145 27: Tests for Normality ................................ ................................ ................................ ................ 145 28: Summary of Regression Models ................................ ................................ ............................ 147 29: Model 1 ................................ ................................ ................................ ................................ .. 148 30: Model 2 ................................ ................................ ................................ ................................ .. 149 31: Model 3 ................................ ................................ ................................ ................................ .. 150 32: Model 4 ................................ ................................ ................................ ................................ .. 150 33: Model 5. Analysis of Maximum Likelihood Estimates ................................ ......................... 151 34: Model 6 ................................ ................................ ................................ ................................ .. 152 35: Model 7 ................................ ................................ ................................ ................................ .. 153 36: Model 8 ................................ ................................ ................................ ................................ .. 154 37: Model 9 ................................ ................................ ................................ ................................ .. 154 38: Model 10 ................................ ................................ ................................ ................................ 155 39: Does Planning Scale Affect Fact Base Scores? ................................ ................................ ..... 156 40: Fact Base ................................ ................................ ................................ ................................ 156 41: Case Study Sampling Based on Fact Base Scores ................................ ................................ 167 42: Final Case Study Selections ................................ ................................ ................................ ... 168 43: Summary Characteristics of Selected Case Study Communities ................................ ........... 170 44: Distribution of Participant Characteristics by Community ................................ .................... 175

PAGE 14

xiv LIST OF FIGURES Figure 1: The Hazards of Place Model ................................ ................................ ................................ ................... 16 2: The Planner's Triangle ................................ ................................ ................................ ............................ 28 3: Place Making and Conflict Resolution through Planning ................................ ................................ ....... 33 5: Arnstein's Ladder of Citizen Participation ................................ ................................ .............................. 52 6: Framework for Stakeholder Participation During Local Hazard Mitigation Planning ........................... 72 7: Map of California Counties Included in Study Area ................................ ................................ .............. 79 8: Methodological Framework ................................ ................................ ................................ .................... 83 9: Models of Stakeholder Participation During Local Hazard Mitigation Plan ning ................................ ... 96 10: California State Hazard Exposure Map ................................ ................................ .............................. 121 11: Scree Plot of Eigenvalues of Varimax Rotation ................................ ................................ ................. 133 12: Normal Probability Plot ................................ ................................ ................................ ...................... 146 13: Map of Case Study Communities ................................ ................................ ................................ ....... 169 14: Models of Stakeholder Participation During Local Hazard Mitigation Planning ............................... 171 15: Revised Framework for Local Hazard Mitigation Planning ................................ ............................... 226

PAGE 15

1 CHAPTER I INTRODUCTION The Hazards and Disasters Problem When coupled with development in high risk areas, differential vulnerabilities can lead to work and does not alleviate the suffering of the most vulnerable within our society? What does it say about the adequacy of preparedness when we know so little about the most disadvantaged within the communities? Moral Hazard, Social Catastrophe (Cutter & Emrich, 2006, 102) Currently, we are losing the battle for a safe and sustainable society in the United States and throughout the world. Despite abundant amounts of data and significant advances in communications and engineering technology, global trends in hazard vulnerabil ity and disaster losses continue to rise. Some estimates describe a three fold increase in disaster events over the last thirty years with associated economic damages increasing by a factor of six (sources from EM DAT, The World Bank, MunichRe; Mileti, 199 9; Birkmann, 2006). Moreover, projections based on climate change and current land use and development patterns predict that communities, especially those living in urbanized areas, are expected to experience increased exposure and vulnerability to hazards from floods, storms, landslides, heat waves, droughts and other hazard events over time (Huq 2007; Bouwer, 2011). The problem of increasing hazard vulnerability is more than strictly an economic concern. It is more than an issue of sustainable infrastructure or natural security. It is, at its core, an issue of social justice. In their study of the legacy of Hurricane Katrina, Cutter and Emrich (2006) observed that the disaster vulnerability of the U.S. population is

PAGE 16

2 unevenly distributed among social groups and between places. They describe a phenomenon in which some communities suffer higher risks of hazard impacts tha n others due to the characteristics of the people who live and work within their boundaries. Orleans disproportionately affected by the disaster, but the ability of people to recover and rebuild their lives also varied widely depending on their economic status, race, gender, age and education level ( Phillips et al., 2009). Ultimately, the disaste r event created by the storm was driven by deep seated moral hazards. It is this unjust reality that motivates and frames my dissertation research. The challenge is clear: a different approach to disaster management and planning is necessary if we hope to remedy the crisis of differential local and regional hazard impacts. In the context of disasters, communities are only as strong as their weakest link. Therefore, in order to protect those who are most vulnerable, and to bolster local resilience and adapt ive capacity, we must shift away from a one size fits all approach to planning towards one that can be tailored to fit the unique social contexts of communities. Local Planning for Disaster Vulnerability Reduction The United States emergency management c ommunity has responded to increasing trends in disaster losses by underscoring the importance of coupling disaster preparedness with hazard mitigation planning. The passing of the Robert T. Stafford Disaster Relief and Emergency Assistance Act (42 U.S.C. 5 121) in 1988 represented a

PAGE 17

3 sea change in national emergency management priorities when it established pre disaster hazard mitigation planning as the key mechanism for combating increasing disaster losses in the United States (Robert T. Stafford Disaster Re lief and Emergency Act, P.L. 100 707, 42 U.S. Code sec. 5121 et seq.). The fundamental principle behind hazard mitigation is that most of the damage and loss caused by hazard events can be prevented. Although most hazard events cannot be prevented from oc curring altogether, their disastrous impacts on people, property, and infrastructure can be diminished if proactive measures are taken to reduce risk and local vulnerability (Comfort et al., 1999; White, et al. 2001; Godschalk, 2003). Hazard mitigation pla nning involves pre event hazard identification, risk and vulnerability analysis, and the creation of feasible mitigation strategies to reduce or eliminate hazard risks. Together, these efforts contribute to reduced vulnerability of infrastructure and peopl e to the impacts of hazards and helps reduce physical, economic and social losses (Nelson & French, 2002; Comfort et al., 1999; White, et al. 2001; Godschalk, 2003; Wisner et al., 2004). The adoption of hazard mitigation planning as a pillar of the United emergency management strategy was coupled with a turn towards state and local level Impact in 1997, and the passage of the Disaster Mitigation Act in 2000 (DMA 2000), there is now a strong emphasis on loss and risk reduction through state, county, and municipal hazard mitigation planning efforts. Since the passage of DMA 2000, researchers have been working to identify the impacts of local hazard mitigation planning ef forts on levels of disaster risks, loss, and

PAGE 18

4 vulnerability. Although measuring prevention is a great challenge, empirical research has shown that higher quality hazard mitigation plans have greater outcomes in terms of reduced vulnerability to hazards and building human resilience to disaster impacts (Burby et al., 1999; Brody, 2003; Godschalk et al., 2009; Beatley, 2009; Peacock et al., 2008). Additionally, the inclusion of local knowledge in hazard mitigation plans has been shown to contribute positively to plan quality (Burby et al., 1999; Brody, 2003; Burby, 2001, 2003; Godschalk & Brody, 2003). To be sure, the creation of high quality plans is not the end point in the overarching goal of reduced social vulnerability, hazard impacts and loss. Rather, hi gh quality hazard mitigations plans are an important intermediary step of the disaster risk reduction process of all communities. Not only has plan quality been shown to predict implementation outcomes, it can be used as a parameter for assessing public en gagement and the incorporation of local community knowledge into decision making and policy. Ultimately, we are interested in an extension of these outcomes, including social vulnerability reduction. One of the key goals of this study is to contribute to a n understanding of how mitigation planning can play a role in reducing impacts to disasters in socially vulnerable communities. Planners and emergency managers have started to acknowledge that plan quality is a critical issue and that engaging community s takeholders in the planning process is key for creating usable, relevant, high quality mitigation plans. If planners and emergency managers intend to create high quality plans that guide local government behavior towards reduced vulnerability and increased resilience to disaster impacts, a key question emerges: How is information about local vulnerability incorporated into

PAGE 19

5 decision making processes? How can planning be improved to encourage the integration of hazard mitigation planning with local land use p olicy and community priorities? While DMA 2000 and Project Impact both emphasize increased community participation in the local hazard mitigation planning process, guidelines and standards for collecting community derived information, and strategies for i ncorporating this information into planning and decision making, are nearly nonexistent. There is great need for a scalable planning approach that planners and communities can use as a tool to create contextual, implementable local hazard mitigation plans. This research study contributes to the planning literature by explaining what this approach to planning might look like. Research Questions This study examines plan quality, models of participatory planning and social vulnerability in the context of haz ard mitigation planning in the Northern California Bay Delta region. California state legislation has shown tremendous support for local participatory action in hazard mitigation planning. As an example of a state that has taken a highly progressive, proac tive stance towards local hazard mitigation planning, two Research Question #1 What is the quality of local hazard mitigation plans created in the Northern California Bay Delta region and to what extent has the public participa ted in the plan making process?

PAGE 20

6 Aim 1: To identify social vulnerability data in plans and incorporate that data into plan quality assessment metrics Aim 2: To assess the involvement of community stakehold ers during the planning process Aim 3: To identify the participatory planning models employed by each city during the local haz ard mitigation planning process Research Question #2 How do community characteristics and levels of public participation affect mitigation plan quality outcomes? Aim 4: To identify key relationships between local community characteristics (including population size, wealth, diversity, social vulnerability characteristics) and local plan quality scores Aim 5: To analyze ho w community characteristics affect levels of public participation in l ocal hazard mitigation planning Aim 6: To evaluate how public participation in planning affects local hazard mitigation plan quality This study follows a mixed methods approach, consist ing of quantitative plan quality evaluation and the qualitative analysis of in depth interviews. It evaluates how different approaches to public participation affect the content and quality of local hazard mitigation plans from 34 California cities and exp ands upon earlier work in the planning and disaster literature by refining plan quality measures to include social vulnerability factors.

PAGE 21

7 Research Impacts Currently, federal and state efforts to reduce hazard vulnerability by developing high quality, loca lly relevant hazard mitigation plans continue to fall short (Battista & Baas, 2004; Schwab, 2011 ; Stevens et al., 2010). A fundamental aim of this research study is to contribute to improved hazard mitigation planning outcomes (including increased project implementation and reduced local social vulnerability) by examining the relationship between models of participatory planning, local community characteristics and h azard mitigation plan quality. Relative to the disaster literature, scholarship related to participatory planning has a long and rich history. In the past, empirical studies have analyzed barriers and drivers to community participation in planning, the relationships between the participation of specific stakeholders and plan quality, and the pla n implementation outcomes associated with participatory approaches to planning. Until now, however, the planning literature has not examined the relationships between social vulnerability to disasters, public participation and hazard mitigation plan qualit y. Participatory planning is no cure all for disaster risk However, a critical analysis of participatory planning processes is necessary in order to identify shortcomings and best practices and to strategically allocate local resources in order to success fully and efficiently decrease social vulnerability to hazards and disasters.

PAGE 22

8 C HAPTER II LITERATURE REVIEW Introduction The overarching research questions that motivate this study are framed by the body of literature on social vulnerability, hazard mitigation and participatory planning. Chapter two covers areas of consensus and dispute in these literatures, and evaluates the nature and quality of support for various contentions in the discussion concerning public participation a nd improved planning o utcomes. To begin, our review of the hazards and disasters literature highlights key factors that influence mitigation plan quality outcomes, specifically those outcomes associated with local vulnerability reduction and disaster resilience. Later, our exp loration of collaborative planning theory and practice offers a framework by which to evaluate hazard mitigation planning and the influences that participatory approaches have on mitigation plan quality and implementation outcomes. Ultimately, this chapter provides a conceptual framework by which local hazard mitigation planning can be evaluated in a new way, one that contributes to significant new understandings of the potential for improved planning practice to contribute to reduced local social vulnerabi lity to hazards and improved local resilience. Hazards, Disasters, and Vulnerability 2008 marked the first time in human history when urban centers became home to Currently

PAGE 23

9 2008). We are living in a unique time of rapid urban development in which high levels of complexity characterize broad social environmental challenges. It is increasingly clea r that current development and planning strategies are producing unsustainable outcomes that have contributed to significant increases in vulnerability to hazards as well as reductions in local, regional, a nd national qualit y of life. An example of the alarming impacts of this development profile is the dramatically increasing trend in international losses from disasters. Not only is the phenomenon a function of increased development in high risk places, but also of increasing levels of social and physic al vu lnerability within communities Increased vulnerability puts communities on trajectories towards greater risk of catastrophic failure or damage in the event of a disaster or shock (Kasperson at al., 1995 ; UNDP, 2004; Pelling, 1997 ). Moreover, in parti cularly vulnerable systems even small pert urbations may lead to collapse. Current mitigation practices often aim to maintain the status quo within communities. However, this approach to planning tends to perpetuate the disaster damage cycle rather than add ressing (and remedying) the root causes of hazard losses. Although efforts to minimize the negative physical impacts of hazards are important, it is equally important that communities implement ongoing planning strategies that take a comprehensive, holisti c approach to emergency management by including political, social, and economic factors in their development and mitigation plans. To become truly sustainable in the changing global environment, communities must address social vulnerability as a root cause of hazard losses. This, however, may demand significant changes in current mitigation planning practices as well as in the structure of society.

PAGE 24

10 Definitions of T erms The rapidly growing body of literature on hazard mitigation planning features a wide, an d sometimes confusing, array of terms. Because the study of human and environmental vulnerability to climate change and hazard events is necessarily interdisciplinary, sometimes the same term takes on a different meaning when used by different authors in d ifferent contexts. As a result, relationships b etween terms are often unclear. In this study the term hazard describes a potentially damaging physical event, phenomena, and/or human activity, which may cause loss of life, injury, property damage, social ec onomic disruption or environmental degradation (UN/ISDR, 2009). Hazards can be characterized as slow onset events (i.e. droughts or long term temperature change) or rapid onset events (i.e. floods, landslides or earthquakes). When hazards are described in the context of this study they are described by t heir physical characteristics. The term disaster incorporates the human elements of a hazard event. As defined by the International Strategy for Disaster Reduction, disasters are serious disruptions of the functioning of a community or a society causing widespread human, material, economic, or environmental losses, which exceed the ability of the affected community or society to cope by using its own resources (UN/ISDR, 2009). Disasters are triggered by haza rd events and are influenced by the properties of the human systems that are exposed to and affected by the hazard. Risk is an important term in the context of vulnerability assessment. It is a fundamental component of an equation that has been used to de termine the likelihood of occurrence (or probability) of a hazard event (Cutter, 1996). Risk is the result of

PAGE 25

11 interactions between a hazard and the vulnerable conditions of a community, place, or individual. Below, the commonly used risk equation describes the relationship between risk, hazards and vulnerability (Wisner et al., 2004 ): Researchers in the hazards field often focus on the concept of risk assessment, while social scientists and climate change researchers often prefer to talk in terms of vulnerability (Downing et al., 2001; Allen, 2003; Brooks, 2003). In general, vulnerabili ty means the potential for loss. Often times, however, social scientists and climate change researchers mean different things when they speak about vulnerability. Generally, definitions of vulnerability can be separated into two categories. One category de fines vulnerability in terms of the amount of potential damage caused to a system by a specific hazard (Jones & Boer, 2003). In the global environmental change literature, for example, vulnerability is most often conceptualized as the likelihood of occurre nce of impacts from weather and climate related events (Nicholl s et al., 1999; Brooks, 2003). The other category defines vulnerability as an inherent characteristic of a system that changes continuously over time and exists before the system encounters a hazard event (Allen, 2003; Cutter, 1993). For example, social scientists tend to describe vulnerability as a representation of the social economic factors that determine the ability of communities and individuals to cope with stress or chang e (Allen, 2003; Brooks, 2003). Over time, the concept of vulnerability has been borrowed, recycled, and reused by numerous disciplines and has taken on a number of definitions and applications. It is not within the scope of this dissertation to review them all, however two distinct concepts

PAGE 26

12 of vulnerability are clearly evidenced in the hazards and disasters literature: biophysical vulnerability and social vulnerability. Although these concepts are connected, they frame vulnerability reduction discussions in very differen t ways and it is important to understand the nuances between them. The IPCC Third Assessment Report defines vulnerability as a measure of the nature of the physical hazard to which the community is exposed, the likelihood of occurrence, and the degree of human exposure to the impacts of the hazard (or the sensitivity ) (IPCC, 2001a). This three tiered, aggregated measure of vulnerability a function of hazard, exposure and sensitivity is sometimes referred to as physical or biophysical vulnerabi lity Ultimately, biophysical vulnerability assessments are concerned with the outcomes of hazard events (the amount of damage sustained by a system from a hazard), rather than with indicators of the state of the system before the event. A second, distinc t conceptual stream describes vulnerability as a state, or as an inherent characteristic of a system that exists before it encounters a hazard event (Allen, 2003; Cutter, 1993). This view developed from a body of social science research related to understa nding the structural factors that make human societies more or less susceptible to damage from hazard events (Mileti, 1999; White & Haas, 1975; Kasperson et al., 2003; Cutter et al., 2003 ). The concept (as it relates to strictly human systems) is called et al., 2003 ; Adger, 1999; Adger & Kelly, 1999). One of the significant lessons from this body of research is that social vulnerability is a unique property of a system. It exists before a hazard event and changes continuously over time. Social vulnerability is the product of social inequalities. Its theoretical origins lie

PAGE 27

13 in the quality of life and livability studies in the social and behavioral sciences during the 1950s and 1960s. The 1960s and early 1970s witnessed a surge in federal government interest in the identification and quantification of social well being indicators (Cutter & Emrich, 2006; U.S. Department of Health, Education, and Welfare, 1969; U.S. Office of Management and Budget, 1973). Initially, this body of re search tried to understand specific characteristics of places that make them more or less desirable and suitable places to live. Increasingly, empirical research began to delve into the social attributes of people and places as a way of understanding and predicting human behavior when faced with problems of social, health and environmental nature (Cutter & Emrich, 2006; Maloney, 1973; Smith, 1973; Berry, 1977). Preliminary research in this area set the stage for the interdisciplinary trends in inquiry that are currently found in the vulnerability science literature (Cutter, 2003). Moreover, studies of this nature have made it clear that differential social vulnerability is an imp ortant social justice concern. Measuring Social V ulnerability A number of att empts have been made to measure and quantify social vu lnerability, including national scale hazards and disaster indicator studies that include social characteristics in their multi hazard exposure models. Others have included social vulnerability indicato rs in their evaluations of social ecological systems and different scales, including U.S. watersheds (Hurd et al., 1999), counties within the U.S. Great Plains (Polsky, 2004), and states across the U.S. (Cutter et al., 2003 ). The quantification of social v ulnerability has also become an important focus of field research and

PAGE 28

14 vulnerability mapping projects concerned with identifying the most vulnerable members of society and examining variations in vulnerability over time and between or within geographic area s (Cutter, 2003; Cutter et al., 2008; Borden et al., 2007; Piegorsh et al., 2007). As foundations for research on vulnerability and risk, three conceptual models have been particularly significant: the Risk Hazard model (RH), the Pressure and Release (PAR ) model, and the Hazards of Place model. The RH model looks at hazard impacts as a function of exposure to a hazard event and the sensitivity of the exposed system (Warrick, 1980; Riebsame, 1991; Kates, 1985). Most quantitative system evaluations using thi s model have worked forward, from the hazard to their impacts, and have evaluated multiple impacts of a single perturbation (Warrick, 1980; Riebsame, 1991; Kates et al., 1985). Continued including its failure to incorporate the ways in which affected systems can magnify or ease the impacts of a hazard event (Kaspersen et al., 1988; Turner et al., 2003). Moreover, the model does not identify the ways in which different components of a syste m create variations in the degree and nature of impacts. Finally, the model does not include the role of social or economic institutions in affecting exposures or impacts (Wisner et al., 2004). These criticisms led to the development of the Pressure and Re lease model (PAR). In this model, risk is generated at the intersection of two opposing forces: the hazard event and the vulnerability of the exposed system (Wisner et al., 2004). The PAR model focuses on dynamic social conditions of exposure (or pressures ) as the root causes

PAGE 29

15 of vulnerability. These pressures can be traced back to distant processes and factors, creating a causal chain of vulnerability (Wisner et al., 2004). The release component of the model has to do with vulnerability reduction activity ( Wisner et al., 2004), and, like a breaking wishbone, the moment at which the accumulated pressure within the system is released and a disaster occurs. According to the PAR model, in order to mitigate the vulnerabilities that cause disasters, the entire ca usal chain must be addressed, all the way back to the root causes. According to the PAR, focusing solely on the triggers of the hazard event or on the physical unsafe conditions of vulnerability is ineffective at reducing the vulnerability of the system. A lthough it explicitly emphasizes the importance of vulnerability reduction in disaster management, the PAR model is insufficiently comprehensive for the broader concerns of sustainable development. For one, it does not address the coupled human environment system or consider the vulnerability of biophysical subsystems (Kasperson causal sequence, including various scales of interactions, and it tends to underempha size feedbacks beyond the immediate system of analysis (Kates et al., 1985; Turner et al., 2003). The Hazards of Place model illustrates how the interaction between hazards and social vulnerability produces outcomes in terms of physical or economic damage or loss of human life. It has been an attractive model to researchers because it integrates two perspectives on vulnerability the biophysical and the social (Messner & Meyer, 2005). Both the RH and PAR models of vulnerability underscore how different co mmunities have different capacities, sensitivities, and exposures to hazards and shocks. Within

PAGE 30

16 human systems these characteristics are closely linked to social factors including power, basic human rights, and access to resources (Wisner et al., 2004; Sen, 2003). As it became increasingly clear that focusing only on physical hazards and exposure was insufficient for understanding the impacts of hazards on human systems, focused efforts began to develop metrics for social vulnerability (Mileti, 1999; White & Haas, 1975; Kasperson et al., 2003). In the Ha zards of Place model (Figure 1 ), risk interacts with mitigation activity to produce a level of hazard potential. Eventually, this hazard potential is either moderated or enhanced through geographic filters, su ch as the physical proximity to the hazard and the social fabric of the place. In this model, biophysical and social vulnerability interact to produce an overall level of place vulnerability (Cutter et al., 2003). Figure 1 : The Hazards of Place Model Social vulnerability measures population sensitivity to hazards as well as the ability of a population to respond and recover from hazard impacts. Because it is a

PAGE 31

17 complex, multidimensional concept, researchers have aggregated a collection of variables to help identify, capture, and quantify it. A large body of empirical evidence from field based research has contributed to more robust understandings of the characteristics that make some people and social groups more sensitive to hazard impacts than others, as well as what makes them better able to respond and recover fro m shocks (Cutter & Emrich, 2006 ; NRC, 2006; Heinz Center, 2002). Based on the Hazards of Place model, Cutter et al. (2003) developed an index for the purpose of d istilling out the driving factors of social vulnerability. Cutter and her colleagues have conducted a suite of research studies quantifying the vulnerability of people and places to natural hazards at different scales. In their 2003 paper, Social vulnerabi lity to environmental hazards Cutter et al. (2003) used county level socioeconomic and demographic data to create an index of social vulnerability to environmental hazards. They named their social vulnerability quantification tool the Social Vulnerability Index (also known as SOVI). Based on the results of myriad SOVI and cultural competency studies, socio economic status, gender, race and ethnicity are the most common characteristics defining socially vulnerable populations. The age profile (children and the elderly), percentage of limited language speakers, and housing tenure (renters) also play significant roles in the ability of populations to absorb impacts, respond and recover in the event of a disaster (Pastor et al., 2006; Andrulis, Siddiqui & Gante r 2007 ). To illustrate this point, consider for a moment why people with limited language skills are more vulnerable in the event of a disaster. Previous studies have shown that the inability to understand evacuation warnings or preparedness bulletins inf

PAGE 32

18 comply with safety measures. Additionally, the inability to communicate special needs to emergency responders or law enforcement can influence their ability and willingness to receive adequate health care o r emergency supplies. Limited language ability also affects planners and emergency managers who organize pre disaster mitigation efforts. As a result, communities with populations made up of greater proportions of individuals with limited language have a higher social vulnerability to hazard impacts and it will likely take them longer to recover from a hazard event (Kreisberg et al., 2011; Pastor et al., 2006). More re cent studies indicate that there are spatial and temporal patterns in social vulnerability across the United States. In their 2008 paper, Temporal and spatial changes in social vulnerability to natural hazards, Cutter et al. (2008) found that the factors t hat consistently increase social vulnerability at the county scale for all time periods are urban density, race/ethnicity, and socioeconomic status. At the national scale, trends show a steady d ecrease in social vulnerability. H owever, there exists signifi cant regional variability, with many counties showing clear increases in social vulnerability over the past fifty years (Cutter et al., 2008). Although social vulnerability is not a function of hazard severity or probability of occurrence, it is, to a cer tain extent, hazard specific. In other words, when thinking about social vulnerability we must ask the question: vulnerability of whom to what? This is because certain properties of human systems make them more or less vulnerable to certain types of hazard s. For example, quality of housing is an important determinant of a

PAGE 33

19 To summarize, biophysical vulnerability depend s on the severity and probability of a given hazard event, whereas social vulnerability does not. Biophysical vulnerability is conceptualized in terms of outcomes (damages from the hazard event), whereas social vulnerability exists independently of a hazar d event. In this dissertation, social vulnerability encompasses all of the system properties that mediate the outcome of the hazard event and is independent of the hazards to which it is exposed. This may include environmental variables and measures of exp osure because exposure and the state of the environment in which a community lives and works are largely socially determined (Brooks, 2003). Increasingly, social vulnerability has been cited by researchers, emergency managers and policy makers as an impor tant element of hazard risk in both urban and rural settings. As a result, increased attention is being turned towards social vulnerability during policy making and development planning. While using data from previous disaster events can identify physical vulnerability relatively easily, the social aspects of hazard vulnerability are difficult to capture due to their complexity and spatial and temporal variability. Despite the progress that has been made with respect to social vulnerability research and var iable identification, there is no agreed upon approach or index for quantifying it. As a result, incorporating the concept into policy remains a great challenge.

PAGE 34

20 Building Resilience into Emergency M anagement Resilience is another term that continues to e merge more frequently in the disasters and global environmental change discourse. A concept traditionally grounded in ecology and systems theory, resilience has been adopted and modified by a diversity of disciplines, including emergency management. How co mmunities are affected by and recover from unexpected shocks and hazard events can be conceptualized in terms of their resilience. More specifically, a disaster resilient community is one that can withstand and absorb the impacts of extreme disturbances an d maintain core functions, often times after assuming an altered form (Innes & Booher, 2010; Berkes et al., 2003). Although they are closely related concepts, it is important to distinguish between disaster resilience and social ecological resilience. Bor n out of systems theory, the concept of social ecological system (SES) resilience describes to what extent the relationships between social and ecological processes can be disturbed before they experience fundamental changes in identity, structure and/or f unction (Holling, 1973; Alberti et al., 2003). A deeper understanding of social ecological systems has shown that not only are resilient systems more able to be flexible in the face of unpredictability and change, but they are also better able to take adva ntage of future opportunities afforded by change (Holling, 1937; Holling et al. 1995; Alberti et al., 2003). Early perspectives on ecological and social system dynamics were highly linear. Natural environments were conceptualized as inherently stable and infinitely resilient. Moreover, it was believed that if human stressors were removed, nature would repair itself and return to a state of perfect equilibrium (Holling, 1961; Lewontin, 1969; Rosenzweig, 1971; May, 1972). Ultimately, research on interacting populations ( e g

PAGE 35

21 predators and prey) and their responses in relation to ecological stability theory led to the emergence of a theory of multiple domains of stability in natural systems (Holling, 1961; Lewontin, 1969; Rosenzweig, 1971; May, 1972). This wa s the be ginning of resilience thinking. Scholarship on social ecological system resilience originated in the 1960s and early 1970s from a stream of ecology in which scholars were examining ecosystem behavior in the face of shocks and environmental change. The social ecological system resilience framework includes humans, as well as environmental factors, as agents of ecosystem change. Traditionally, ecological theory excluded social institutions from the study of ecosystem dynamics and treated human actions as external to the system. As a consequence, the interactions between ecological processes and social dynamics were rarely taken into account in scienc e, policy, or decision making. Social ecological system resilience represents a shift in thinking about unpredictability, as well as in the meaning of equilibrium within systems. In his seminal paper on resilience, Resilience and stability of ecological systems (1973) wrote that social ecological system resilience is based on non equili brium dynamics and that the concept of equilibrium is poorly suited to explain the behavior of systems that are not currently at equilibrium. He argues that because the concept of equilibrium represents a static state, it is incapable of providing insight into the behavior of systems that are variable o r transitional (Holling, 1973). From the social ecological perspective, resilience determines the persistence of the relationships within systems and measures the ability of a system to absorb shocks without losing its capacity to function in its current state of existence (Holling, 1973,

PAGE 36

22 1996). Holling (1973) recognized early on that complex adaptive systems are not infinitely resilient. Rather, he suggested that they require adaptive management in order to p ersist. Adaptive management processes have been characterized by learning by doing, where plans and policies are experiments from which communities and social systems can learn and share knowledge (W alters, 1986; Gunderson, 1999). Although disaster resil ience describes a more reactive, outcome based concept associated with bouncing back and recovering quickly after a hazard event, disaster resilience conceptualizes communities as systems of interacting components (i.e. social systems, governance systems, infrastructure, planning entities and ecosystems). Similar to the learning by doing approach to resilience building expressed by social ecological resilience theory (Walters, 1986; Gunderson, 1999), communities can increase their resilience to disasters th rough adaptive management activity including knowledge sharing, learning from past experiences, and creating a common language between different systems actors and decision makers (Paton et al., 2000; Goldstein, 2009; Innes & Booher, 2008; Innes & Booher, 1999; Colten & Sumpter, 2009). literature and disaster preparedness and recovery discussions. However, definitions of the term resili ence abound, as is highlighted by Table 1

PAGE 37

23 Table 1 : Definitions of Resilience Author Definition Disciplinary Lens Holling, 1973, p.14 A measure of the persistence of systems and of their ability to absorb change and disturbance and still maintain the same relationships between populations or state variables. Ecological Science Walker et al., 2004 Resilience is the capacity of an ecosystem to tolerate disturbance without collapsing into a qualitatively different state that is con trolled by a different set of processes. A resilient ecosystem can withstand shocks and rebuild itself when necessary. Resilience in social system has the added capacity of humans to anticipate and plan for the future. Ecological Science, Social Sciences Walker et al., 2006, p.2 The capacity of a system to experience shocks while retaining essentially the same function, structure, feedbacks, and therefore identity. Ecological Science Adger, 2000, p.347 The ability of groups or communities to cope with external stresses and disturbances as a result of social, political and environmental change. Social Sciences Adger et al., 2005, p.1036 The capacity of social ecological systems to absorb recurrent distu rbances [...] so as to retain essential structures, processes and feedbacks. Social Ecological Science Department of Homeland Security, TCL, 2007, p.13 Resilience is key [to a determination of vulnerability] since it refers to our coping capacity to abso rb events, adapt, and respond to and recover from its effects. Disaster Management Mileti, 1999 Local resiliency with regard to disasters means that a locale is able to withstand an extreme natural event without suffering devastating losses, damage, diminished productivity, or quality of life without a large amount of assistance from outside the comm unity. Disaster Management Paton, Smith & Violanti, 2000 Resilience describes an active process of self righting, learned resourcefulness and growth the ability to function psychologically at a level cap abilities and previous experiences. Disaster Management United Nations International Strategy for Disaster Reduction (UNISDR, 2010 ) 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. This is determined by the degree to which the social system is capable of organizing itself, and the ability to increase its capacity for learning and adaptation, including the capacity to recover from a disaster. Disaster Manag ement

PAGE 38

24 The increasing use of the word resilience in the context of disasters has been described by some as representative of a new paradigm of disaster response (McEntire et al., 2002). While some scholars consider resilience thinking to be a new paradigm, others describe it as merely an expression that combines other disaster terms (including vulnerability, adaptive capacity, and risk). These scholars believe that, concep tually, disaster resilience does not represent anything new. Citing the occurrence of previous rhetorical shifts created by words like sustainability, 2006, 435). Despite these critiques, in looking at the foundations and definitions of resilience, and by examining points of overlap between multiple perspectives, it becomes more clear that resilience thinking has great potential to c atalyze new and improved ways of planning and governance, particularly in the context of participatory planning and local social vulnerability reduction. Resilience thinking has emerged as a useful framework for understanding and solving complex urban pro blems because it can be applied to both the social and physical aspects of communities. Over the past decade, the broad concept of disaster resilience has found both a practical and theoretical home in the disaster risk reduction discourse (Manyena, 2006). For example, a number of international and federal level emergency management, security, and planning entities have identified resilience building as a key priority for guiding policy and sustainable development decisions. A desire to merge disaster risk reduction with urban planning has gained momentum with the publication of reports such as Reducing Disaster Risk: A Challenge to Development (UNDP, 2004) and Disaster Risk Reduction: A Development Concern (DFID, 2005). These publications

PAGE 39

25 underscore how dis aster risk accumulates within inappropriate development paths and that increasing disaster losses can be explained largely by the consequent increases in human vulnerability. This phenomenon highlights the necessity of developing alternative frameworks for urban planning practice and decision making based on robust understandings of the social factors of vulnerability and risk. In the U.S., resiliency has taken on important meaning in the context of emergency management and homeland security. However, as a process, disaster resilience demands new applications of the risk equation. It necessitates a more holistic approach to vulnerability analysis as it places a new emphasis on the role of people in disasters. By focusing explicitly on resilience, disaster m anagers are encouraged to consider the social components of vulnerability, including ways in which to enhance community coping capacity and assets rather than simply reducing weaknesses (Manyena, 2006). Hazards, Disasters and S ustainability Planners have responded in a diversity of ways to growing frustrations with common features of urbanization including sprawl, pollution, inequity, and vulnerability to natural hazards. In the United States, sustainable urban planning strategies have been focused primari ly on climate change mitigation measures. By and large, urban centers have relied on improvements to physical infrastructure in response to the impacts of climate change. In addition, climate change mitigation measures such as increased GHG sequestration t hrough wetland creation, decreased GHG emissions from oil and gas industries, and increased investment in renewable energy are gaining greater support at

PAGE 40

26 local and regional levels as a response to climate change and increased vulnerability to storms and fl oods. Despite the prominence of efficiency improvements and GHG reduction through engineering advances and policy, urban approaches to green development have also included an emphasis on sustainable planning tools such as urban growth boundaries and the LEED ND rating system. Sustainable development has emerged as a tool for planning practitioners to increase local sustainability on both social and ecological levels. New urbanism, infill development and smart growth strategies are other examples of susta inable planning strategies that have been applied at neighborhood, city, and regional scales. They include the design of compact, high density neighborhoods for more efficient use of land and transportation; the construction of resource and energy efficien t Moreover, by building within existing urban areas, cities conserve their environmental resources, economic systems, and social fabric, while reclaiming marginal and a bandoned land (Beatley, 1998; Berke, 2002). Recently, scores of cities across the U.S. have initiated comprehensive plans and urban development projects with ambitious sustainability goals in mind. However, little attention has being paid to the following question: Cit y planners play an important role in shaping the vulnerability and resilience of urban areas, whether they know it or not. I or sustainable can work against local hazard mitigation and vulnerability reduction efforts ( Berke et al. 2009; Stevens et al., 2010).

PAGE 41

27 Communities that are vulnerable to catastrophic damages from hazards are, by definition, unsustainable. A community is unsustainable if it is at risk of losing environmental quality, its economic livelihood, or h uman lives during a hazard event. As complex systems, cities that are planned with resiliency in mind are capable of withstanding large shocks without sustaining either immediate failure or permanent damage. They are designed and managed to anticipate, abs orb, and recover from the impacts of hazards. Moreover, they are built on principles derived from past experience with disasters in urban areas. While resilient communities may bend, they will not break. Despite this logical connection between sustainabili ty and disaster resilience, however, very limited research has been do ne to unpack the relationship. Independent of the sustainability revolution, the planning profession has long upheld a development agenda that supports strong economic growth, environme ntal health and improved quality of life. In his paper Green Cities, Growing Cities, Just Cities, Scott Campbell describes the planning discipline as defined by three (arguably divergent) priorities: social justice, economic growth, and environmental prote ction (environment, economics, and equity) these priorities lie at the historical heart of planning. Conflicts between these priorities define contemporary struggles within human as the (Figure 2 ) with sustainable development situated at the center. alized as a three legged stool (Campb ell, 1996)

PAGE 42

28 Figure 2 : The Planner's Triangle In this model, Campbell argues that sustainable development cannot be achieved by looking at one leg of the stool at a time. Rather, planners will reach sustainable solutions only by confronting the conflicts between the three points of the triangle and by do not interact solely through opposition. They also interact collaboratively (Campbell, 1996). Campbell stresses that in the planning discipline sustainability mus t be represented as a fine balance between social justice, economic growth, and environmental health. He argues that in an ideal world, planners would strive to achieve a perfect balance between growth, preservation, and equality. In practice, however, pro fessional, political and fiscal constraints often limit the freedom of planners and decision makers to do so (Tobin, 1999).

PAGE 43

29 Hazard Mitigation: Planning, Policy and Practice The following sections describe the evolution of comprehensive planning practice and how hazard mitigation planning fits into the comprehensive planning framework. The discussion includes a review of the history of hazard mitigation planning in the U.S. and provides the background for a clear understanding of how mitigation planning is used as a tool for the achievement of community scale vulnerability reduction goals. The Compr ehensive City P lan In the most general sense, comprehensive plans are long range physical plans made up of a collection of policies. Typically, the city is the geographic scope of the document. Because the typical time frame of a comprehensive plan is 20 30 years, beyond the scope of current pressing issues, they open up dialogue about future possibilities and challenges. Comprehensive plans focus on social, buil t environment, and economic infrastructure that make the city work as a physical entity and that affect physical form. Written as a statement of policy rather than a rigid program of specific actions, comprehensive plans are designed to guide urban decisio n makers as they make decisions in the future (Black, 1968; So a nd Getzels, 1988; Innes, 1996). In 1964, Californi Kent presented a version of the comprehensive city plan that is widely recognized as the strongest repre sentation of the comprehensive planning ideal. In his book, The Urban General Plan Kent lays out a set of principles and goals for comprehensive planning that planners and scholars ) view of comprehensive planning is distinct from the rational comprehensive model of planning,

PAGE 44

30 which functions by clearly outlining all potential objectives and policy options, comparing them, and ultimately choosing the optimal planning strategy (Lindblo om and comprehensive planning practice away from considering every alternative before making a decision. Rather, Kent advocated for comprehensive plans to serve as toolboxes of policies to be utilized in response to future circumstances to achieve city goals and objectives (Innes, 1996). Kent describes three necessary components for comprehensive plans: comprehensiveness, explicit reasoning, and generality (Kent, 1964). Kent st ates that comprehensive city plans must be comprehensive in that they cover the entire city. Plans must focus not only on infrastructure but also on the relationships between the many factors that affect the growth and development of cities including the p hysical and nonphysical, local and regional, and social and environmental. Kent also stresses the importance of considering technological, economic, and demographic forecasts while developing comprehensive plans. Comprehensiveness is the key, he writes, to promoting the public interest (which constitutes the community at large rather than individuals or powerful stakeholder groups) ( Kent, 1964). Explicit reasoning requires that plans clearly identify the facts from which their visions were developed and justified. This allows for decision makers to make informed final judgments about if and how to implement the policies outlined in the city plan. Kent also stated that the principle of generality is critical to comprehensive planning. Plans should not serv e as a blueprint for the city; rather, they should be general and focused on larger issues facing the city and big ideas for their resolution. Because cities and

PAGE 45

31 communities are complex and constantly evolving, plans should not be used as ideal models for development. Rather, they should be used as tools for democratic disco urse (Kent, 1964; Innes, 1996). Long range comprehensive planning is an exercise in anticipatory problem solving and is a process through which community needs and goal setting objective s can be identified in an organized way. In theory, comprehensive planning is intended to stand up for common interests rather than those of a limited, powerful few. Ideally, comprehensive planning is an on going, dynamic process in which plans are revisit ed and reevaluated regularly after their adoption. Because of the complex, adaptive nature of cities and communities it is important for plans to be update d so that they remain relevant to changing nee ds, priorities, and challenges. Early U.S. city and reg ional planning developed in response to complex urban problems including social inequity and environmental degradation. Over time, problems changed and community priorities were redirected towards new, emerging urban issues. For example, between the 18th a nd 19th centuries, the Industrial Revolution drastically changed the function and form of cities. In his book, Urban and Regional Planning Peter Hall describes the noisy, disease riddled cities of the Industrial Revolution as the most important planning p roblem of the period. Urban economic growth was threatened by inadequate infrastructure and unsanitary conditions and planners were motivated to find fast and innovative solutions to the industrial city (Hall, 2002). Today, issues of disaster resistance, s ocial equity, and economic development have surpassed issues of sanitation and undesirable industrial development as leading urba n planning concerns.

PAGE 46

32 The literature identifies two fundamental purposes around which planning problems pivot: place making and conflict resolution (Healey, 1998). A clear understanding of the priorities associated with conflict management and place making is useful for setting the stage for conceptualizing contemporary planning theory and practice. Place making involves the promot ion of quality places. The aim of place making is to use design as a way of achieving broad economic, environmental, and social objectives. The second fundamental purpose of planning is to facilitate the regulation of private land use rights and to manage conflicts having to do with land use and development. All regulation and conflict management are part of a larger effort to serve the public interest (Healey, 1998). Healey (1998) acknowledges that conflict management and place making are deeply interconne cted. However, continued analysis of planning practice and structure has illustrated that the two priorities also diverge. Figure 3 illustrates how conflict management and pl ace making can diverge and come together through a process of effective spatial pl anning.

PAGE 47

33 Figure 3 : Place Making and Conflict Resolution through Planning Generally, place making happens through professional and stakeholder debates about ideas for achieving desired community outcomes. Place making focuses on the transformation of both policy and practice. In contrast, the process of conflict management is largely defined by the actions of legal and administrative entities. These entities are primar ily concerned with disputes about the legality of property rights restrictions and focus on maintaining established structures of governance and regulatory systems (Healey, 1998). Since the 1960s planning practice and scholarship have transitioned from a focus on physical design to a focus on process. Before the 1960s, planners brought their creative ideas to the public by guiding the development of cities, towns, and regions. Urban planning in the United States was characterized as a largely technocratic process, dominated by independent experts who used objective scientific analysis to manage

PAGE 48

34 resources and make infrastructure decisions. The preferred model in the profession was decision making free of political interference (Berke, 2002; Wondolleck and Ya ffee 2000; Susskind et al., 1999; Beierle and Cayford, 2002; Gunton and Day, 2003). During the 1960s and 70s, rational planning, a model of planning based on systems theory and the scientific method, emerged in response to numerous social and environmental crises (Berke, 2002). However, communities and decision makers began to question the efficacy of traditional planning as widespread urban problems including racism, pollution, poverty, and financial failure continued to plague metropolitan areas and regio ns (Berke, 2002). As a result, planners began to take a more comprehensive, systematic, and anal ytical approach to their work. After the 1960s, urban planning was largely dominated by incremental projects, rather than large scale ones (Berke, 2002). This transition was due, in part, to growing dissatisfaction with the top down, expert dominated approach to city building that was often exclusionary and counterproductive to achieving larger community goals (Altshuler, 1965; Davidoff, 1965). The Civil Rights Movement helped catalyze the development of the participatory planning model and during this time urban areas were important places of political expression and civic dissent. Scholars and communities were acutely aware that early urban renewal and highway development projects disproportionately affected low income, minority neighborhoods. This contributed to a dissatisfaction over planning policies involving resource extra ction, urban renewal and transportation, planners formally acknowledged democratic decision making as the

PAGE 49

35 foundation of their profession (Gans, 1962; Davidoff, 1965; McLoughlin, 1969; Chadwick, 1971; Hall, 2002; Chapin and Kaiser, 1979; Hester, 1996). Ult imately, diverse values, competing priorities, and resource constraints sparked the creation of two new planning paradigms in the U.S. Both mediation and advocacy planning emerged as new ways for planners to think about their role in decision making and city building. Mediation, also known as alternative dispute resolution (ADR), calls for planners to act as intermediaries between conflicting stakeholders. In this model, the goal of the planning process is to guide decision making in a way that produces m utually beneficial solutions (Bacow and Wheeler, 1984; Susskind and Cruikshank, 1987). Proponents of the mediation method argue that they are cost efficient, and more likely to resolve conflict than other approaches (Gunton and Day, 2003). Moreover, becaus e planners seek win / win solutions to conflicts, the participatory method is believed to produce more lasting, equitable solutions. Advocacy planning developed as a new decision making strategy intended to serve disadvantaged and marginalized stakeholders. Like mediation, advocacy planning acknowledges that different stakeholders often have competing interests. Planners operating under the advocacy model promote the interests of the most underrepresented stakeholders and cultivate community participation in an effort to simultaneously create plans and empower the public (Davidoff, 1965). In the 1950s the dominant view of the public good was what is good for business is good for all (Hester, 1996). However, planners began to challenge this belief by facilitat ing the development of comprehensive plans for diverse interest groups and they began advocating for development strategies that represented multiple perceptions of the city. Some scholars argue, however, that

PAGE 50

36 although there are clear benefits to this appr oach to planning, it leaves little room for the cultivation of holistic, larger scale visions of the city (Clavel, 1994; Neuman, 1998; Berke, 2002). Hazard M itigation in the U nited States Currently, much of the responsibility for reducing hazard losses in the U.S. is shared by the federal government and numerous local governments and special jurisdictions. Historically, U.S. disaster management was founded on civil defense and behavioral science research (Drabek, 1991; Pearce, 2003). Emergency managers and planners traditionally followed a bureaucratic model of hazard mitigation planning and drew from privileged technical information to inform their risk identification and goal setting activity. As a result, the focus of planning activity was often skewed to wards assessing and mitigating geophysical vulnerabilities rather than addressing social dimensions of hazard risk (Colten, 1992; Colten & Sumpter, 2009; Vale & Campanella, 2005; Drabek, 1991; Pearce, 2003). Before the 1950s a comprehensive disaster manag ement program did not exist in the United States and disaster response and recovery was funded by congress on an incident by incident basis. It was n o t until the Federal Disaster Assistance Program was enacted in 1950 that the federal government began addr essing disasters in a coordinated manner. In 1979 President Jimmy Carter signed Executive Order 12148, establishing the Federal Emergency Management Agency (FEMA), and effectively consolidated the to one organization.

PAGE 51

37 Table 2 : The Role of the Federal Government in Disaster Loss Reduction Year Federal Program or Act Details 1950 Federal Disaster Assistance Program Grants President broad power to respond to crisis Authorizes Federal government to respond to major disasters 1970 Disaster Relief Act of 1970 Authorizes the use of Federal loans and tax assistance to individuals affected by disasters Authorizes Federal funding for repair and replacement of public facili ties Introduces hazard mitigation as a Federal priority 1973 The Flood Disaster Protection Act Makes flood insurance mandatory for the protection of property located in Special Flood Hazard Areas 1974 Disaster Relief Act of 1974 Establishes process for Presidential disaster declaration s 1979 Executive Order 12127 Merges separate disaster related responsibilities into FEMA 1988 Robert T. Stafford Disaster Relief and Emergency Assistance Act Legislates cost sharing requirements for public a ssistance programs R equires State mitigation plan as a condition of disaster assistance Encourages pre disaster mitigation planning at the state level through grant programs Provides funds for states and local governments to manage public assistance projects 1997 Project Impact: Building Disaster Resilient Communities Emphasizes local community efforts to mitigate hazards and reduce vulnerability G oal : or mitigation projects 2000 Disaster Mitigation Act of 2000 Provides incentives for close coordination of mitigation planning and implementation efforts between State, local and Tribal entities E stablishes mitigatio n planning requirements for local and Tribal governments as a condition of mitigation grant assistance

PAGE 52

38 In the 1980s, in response to increasing losses from disasters, the U.S. government government took on the responsibility of managing increasing hazard risks. This was exemplified by the passing of the Robert T. Stafford Disaster Relief and Emergency Assistance Act in 1988. The Stafford Act represented a sea change in national emergency management priorities by establishing mitigation as the most important means for combating increasi ng disaster losses. Congress acknowledged that special measures were necessary to assist state and local governments in providing aid, assistance, emergency services, and reconstruction aid to areas that have been impacted by hazard events. A key component of their strategy involved the encouragement of pre disaster hazard mitigation planning at the state level as a mechanism for reducing overall losses from disasters. Over time, the emergency management community has increasingly acknowledged the diversi ty of interests and stakeholders within a community as a critical piece of the mitigation approach. In 1997, FEMA moved forward with an innovative, community based mitigation initiative called Project Impact: Building Disaster Resilient Communities Projec t Impact was an attempt to reduce the impacts of natural disasters by helping communities carry out risk assessments and needs prioritization before a disaster event. By encouraging communities to build collaborative public/private partnerships during miti gation planning Project Impact set the stage for the development of appropriate, politically palatable, long term strategies to protect communities fr om disaster impacts and loss. In 2000, the Disaster Mitigation Act (DMA 2000) amended the Stafford Act a nd enacted a new set of rules requiring that state, local, and tribal entities coordinate their

PAGE 53

39 mitigation planning and implementation efforts. Under DMA 2000, state mitigation plans are still required as a condition of federal disaster assistance. However the act introduced incentives for the development of local scale mitigation plans. DMA 2000 now requires that State and local communities have approved hazard mitigation plans in place in order to be eligible for pre and post disaster hazard mitigation grant funds. The FEMA crosswalk for local hazard mitigation plan approval also includes a public participation component. FEMA consistently stresses that the mitigation planning process is just as important as the mitigation plans themselves. Although loca l and state hazard mitigation planning processes can happen in a diversity of ways, by offering incentives for multi stakeholder coordination during the planning process, DMA 2000 strongly encourages participatory planni ng activity. Unfortunately, DMA 2000 planning requirements lack adequate guidelines for achieving meaningful, collaborative public participation activity during local hazard mitigation planning. Currently, there is a great need for an additional review crosswalk to be established that includ es guidance on how to acquire knowledge and expertise specifically from vulnerable and marginalized community groups Vulnerability Reduction through Hazard M itigation In 2005, in response to intensifying hazard losses associated with unplanned urbanizatio n, poverty, environmental degradation, and population growth, the Second World Conference on Disaster Reduction broached the subject of disaster resilience. In January 2005, the Hyogo Framework for Action 2005 2015: Building the Resilience of Nations and C ommunities to Disasters was endorsed by 168 UN member states (UNISDR, 2005). The Hyogo Framework for Action (HFA) recognizes that global

PAGE 54

40 disaster risks will only continue to magnify, due in part to the vulnerability implications of unplanned urbanization. The framework encourages countries to undertake major efforts to reduce their disaster risk by 2015 and outlines five priority areas for action. Priority area four calls for the incorporation of disaster risk reduction into urban planning activity as a met hod for reducing underlying risk factors (UNISDR, 2005). The HFA represented an important step towards integrating disaster risk concerns into planning practice. The framework emphasizes social vulnerability reduction through largely participatory approac authorities should be empowered to manage and reduce disaster risk by having access to the necessary information, resources and authority to implement actions for disaster risk ection IIIA, point 13.f). This statement is significant because it explicitly linked disaster risk reduction to community empowerment strategies including public participation and control over information, resources and authority for the first time at the global scale. In their publication Words Into Action: A Guide for Implementing the Hyogo Framework (UNISDR, 2007) the United Nations International Strategy for Disaster Reduction stresses the importance of community based, collaborative decision making processes in disaster risk reduction efforts and emphasizes the need to develop a holistic set of vulnerability indic ators (both social and biophysical) as key resilience building strategies. It also illustrates the ineffectiveness of explaining the impacts of disasters on complex social, environmental, and economic systems through the use of strictly hazard or environme ntal indicators (UNISDR, 2007).

PAGE 55

41 Today, the HFA serves not only as a tool for helping nations prioritize their risk reduction resources, but also as a framework for risk reduction efforts that embrace collaborative processes and incorporate information der ived from communities through multi methods approaches. The Hyogo Framework set the stage for the ongoing discourse on hazard mitigation planning practice in the context of social vulnerability reduction and citizen participation. Assessing V ulnerability As part of the hazard mitigation planning process, emergency managers and planners conduct comprehensive risk and vulnerability assessments. These assessments are used as tools for developing plans that help communities understand and manage disaster risk s. There exist a number of models and approaches to vulnerability assessment and these assessments can be conducted at various scales including national, state, regional and neighborhood levels. Fundamentally, vulnerability assessment is a process of identifying and conveying interactions between the built environment, poli tical structures, socio economic conditions and the physical hazard risks in a community. They play an important role in establishing a knowledge base of information about unique ha zards and social contexts of a particular place, and they are meant to reveal the interactions between humans and the environment at a given location (Turner et al., 2003). Vulnerability analysis emerged from an extensive body of research on risk, hazard s and global environmental change. One of the salient lessons of this body of research is that vulnerability is a unique and inherent characteristic of a community that

PAGE 56

42 changes continuously over time. Therefore, focusing on hazards alone is insufficient fo r understanding the impacts of hazards on a social ecological system (Mileti, 1999; White & Haas, 1975; Cutter et al. 2003 ; Kasperson et al., 2003). Currently, there is a spectrum of vulnerability assessment models that can be characterized by the varyin g degrees of attention given to social vulnerability factors and local community capacity ( Phillips et al., 2009). These models also vary by the processes through which information and data about vulnerability are collected. Some vulnerability assessment m odels focus on single or multi hazard risks (FEMA, 1997; Coburn, Spence, & Pomonis, 1994; Phillips et al., 2009), while others include limited assessments of social determinants of risk (Cutter, Boruff, & Shirley, 2003; Phillips et al., 2009). In addition, there exist models that strive to capture locally derived community data from citizens in order to holistically assess local capacity ( Phillips et al., 2009). Table 3 outlines the spectrum of current U.S. hazard vu lnerability assessment models 1 1 Table derived from Phillips et al. (2009)

PAGE 57

43 Table 3 : Models of U.S. Hazard Vulnerability Assessment Approach Outcomes Data Requirements Single hazard assessment Documentation and/or risk model for a specific area Single hazard risk information; Historical event and impact data; Monitoring data for model generation or model output Multi hazard assessment Inventory of all hazards in a given location; Mapping and quantitative technical analysis of multiple hazard risks Single hazard risk information; Historical event and impact data; Monitoring data for model generation or model output Basic CVA (community vulnerability assessment) Includes risks of multi hazards; Explicit attention is given to understanding socioeconomic conditions with respect to hazard risks Existing secondary data sources for social analysis and potential hazards; US Census data for socioeconomic analysis; May include few locally derived datasets Intermediate CVA Holistic treatment of vulnerability (includes additional data beyond US Census for social and economic analysis); Includes information on coping capacity Supplements U.S. Census data with local level sources of data; Includes indicators of capacity, including community resources; Inclusion of limited primary data from local surveys or fo cus groups Community based vulnerability assessment Social vulnerability analysis expanded to include top down and bottom up information; Process involves capturing community knowledge and perceptions and incorporating them into the assessments Collection of primary data via collaborative techniques; Best practices involve the engagement of stakeholders of all types and community driven data collection efforts; Collection of information about community support networks At the top of the table, the more traditional vulnerability assessment models include outcomes that consist solely of inventories of technical hazard risk data (Single and Multi hazard vulnerability assessments). The intermediate models represent increasingly community based approaches, wh ich include varying degrees of contextual

PAGE 58

44 information such as population demographics, economic factors, and environmental conditions in their data requirements and outcomes (Basic CVAs) ( Phillips et al., 2009 ). Examples of more comprehensive mitigatio n models supplement technical vulnerability analyses with locally derived information and community knowledge in an attempt to create context specific planning goals and policy decisions (Intermediate and Community based CVAs). Community based approaches t o vulnerability assessment are considered the best way to develop mitigation plans that successfully incorporate valuable community knowledge, including local sources of capacity and social determinants of vulnerability (Comfort et al., 1999; White et al., 2001; Godschalk, 2003; Wisner et al., 2004 Phillips et al., 2009 ). Despite the range of assessment models, contemporary hazard mitigation planning policy has exhibited a shift towards an increasingly collaborative, resilience based approach to risk reduction. For example, FEMA grant programs now mandate public participation in disaster management and hazard mitigation planning. This shift is due, in part, t o increasing evidence of the benefits associated with multi stakeholder knowledge sharing and discourse particularly in the context of hazard mitigation planning. Engaging the Public through Participatory Planning Traditionally, expertise and knowledge have been understood as the products of s cientific or professional occupations. However, scientific and institutional experts do not always know best. Context determines who local knowledge can expand the profess ional radius of expertise (Yli Pelkonen & Kohl, 2005).

PAGE 59

45 Typically, l ocal knowledge is not the product of systematic scientific study. Rather, i ts validity stems from a lengthy series of local observati ons and experiences over time. A post Katrina study cond ucted by geographers Colten and Sumpter (2009) explores the relationships between local knowledge, scientific expertise and disaster resistance in the context of pre Katrina New Orleans. Motivated by a key co ncept of resilience thinking that societies ca n learn from hazard events and use their accumulated social memory to adapt to future catastrophe s their study looks at local planning activity in New Orleans after Hurricane Betsy in 1965 and at how information y was used for hazard management planning leading up to Hurricane Katrina in 2005. Colten and Sumpter (2009) found that two critically important lessons from Hurricane Betsy were omitted from post Betsy hazard management activities. One of those missing lessons was that the areas of greatest flood vulnerability were located in eastern New Orleans, the lakefront, and downstream p arishes The failure to apply this information to future comprehensive land use planning prevented the city from taking action t o guide development towards the safest locations. This, unfortunately, resulted in a decrease in the overall resiliency of the city. Within a matter of years after Hurricane Betsy hurricane protection levees were constructed around the most historically v ulnerable areas and the city Additionally, some of the most rapid growth within the city t had no bearing on development Colten & Sumpter, 2009, 360). Over time, planning documents and development planners

PAGE 60

46 continued to build and continued to ignore past flooding in eastern and lakefront New Orleans further erodi ng the resilience of the city. The previous study illustrates a n examp le of the potential negative impacts associated with hazard mitigation pl anning approaches that fail to learn from local communities and incorporate local historical knowledge. Ultimately, the failure of New Orleans to integrate local knowledge and memory into hazard mitigation and land use planning eroded the resilience of the city and set the stage for the costliest, and perhaps the most resentful disast er experience in U.S. history. The hazards and disaster literature suggests that the adoption o f local scale hazard mitigation plans helps to reduce hazard losses (Nelson & French, 2002). Moreover, hazards researchers have found that mitigation plans are unlikely to be effective unless they involve the participation of a diverse range of stakeholder s during plan creation (Burby, 2003). Empirical research has also shown that both the quality of hazard mitigation plans, and the likelihood that they will be implemented after adoption, tends to increase with the level of public participation during plan making (Burby, 2003; Godschalk et al., 2003). The following sections discuss the foundations of participatory planning, explain a framework for understanding the impacts of various types of participatory and review the existing body of literature related t o identified benefits and drivers of public participation in planning. The Theory and Practice of Communicative P lanning Theories of communicative planning, and models of discursive, participatory and collaborative planning all emphasize an inclusive, democratic approach to making plans.

PAGE 61

47 Over the last several decades, p lanning practice has adopted increasingly interactive and communicative decision making and knowledge building strategies Although some of the terminology and descriptions of the communicative planning model have changed over time, the fundamental ideal of plan ning as a collaborative process has e ndur ed (Forester, 1982; Healey, 1998; Innes, 1996; Taylor, 1998; Innes and Booh er, 1999a). Jrgen Habermas developed the theory of communicative rationality in the 1980s a s a way of describing a set of ideal conditions for discourse and knowledge building. The theory challenged instrumental rationality by bas ing knowledge building and reasoning on inter subjective exchange. Habermas writes that communicative rationality draws from the practical consciousness and life experiences of stakeholders. Because physical, social, and experiential contexts overlap, it i s argued that communicative planning processes generate more robust forms of knowledge through deliberation and interaction ( Habermas 1981, 1989; Outhwaite, 1994). New forms of knowledge and perception are created, collective values and norms evolve, and t he communicative planning process is able to create a new common language (Healey, 1998). According the Habermas if ideal conditions for communication and knowledge building are met during inter subjective exchange, then discourse will result in emancipat ory knowledge Habermas theorizes that emancipatory knowledge allows cultural and political power structures (Habermas, 1981, 1989; Innes an d Booher, 1999b). For this reaso n, he argues t hat emancipatory knowledge is vitally important in a rapidly changing world. As change occurs more quickly it becomes increasingly challenging to maintain an accurate, complete understanding of our urban, social, and environmental systems.

PAGE 62

48 Therefore, emanc ipatory knowledge is a pathway towards innovative solutions and new ways of understanding our increasingly complex world. of the highly deductive, rational foundations of th e discipline. The idea of communicative processes was introduced to the planning discipline by Forester (1982) and represented a shift in perceptions about presumed objectivity and the use of information and knowledge in decision making. Innes and Booher ( 2010) call this shift in planning and policy making collaborative rationality has to do with dialogue and deliberation. Within the communicative stream of planning thou ght, perfect truth, information, objectivity, and collective knowledge are ideals that cannot be fully achieved. However, through collaborative deliberation planners can most effectively aim towards these ideals and produce better, more sustainable outcome s than those achieved through indi vidual, expert based analysis. Collaborative rationality may also help remedy uneven power relations that exist in traditional governance regimes and planning systems. Communicative planning theory asserts that planning i s deeply political and largely dictated by power structures. The agendas of those in power are often favored and stakeholders are rarely afforded equal participation or opportunities for decision making (Healey, 1997; McGuirk, 2001). Habermas (1981) argues that instrumental rationality creates a reality of communicative distortion, in which false consensus is reached through discourse in the context of unequal distributions of power. In theory, collaborative rationality can transform power relations by crea ting deliberative arenas for stakeholders to participate and share. These

PAGE 63

49 power neutral forums enable temporary conditions of shared power and create forms of knowledge and rationality in which distortion is minimized (McGuirk, 2001; Innes and Booher, 2010 ). Part icipatory P lanning : Practice and T heory Citizen participation in U.S. planning has an extensive and lengthy history. The emergence of increasingly democratic planning paradigms was initially a response to the failure of traditional planning models to manage complex social, economic, and environmental problems. For example, transactive planning (Friedmann, 1973), consensu s building (Innes, 1996; Innes & Booher, 1999), and communicative planning (Margerum, 2002; Healey, 1992; Innes, 1996) all emphasize inclusive, democratic approaches to making plans. T o varying degrees, each paradigm situate s communities at the core of pla nning practice. In the 1920s the U.S. Department of Commerce developed state planning legislation that required local governments to offer citizens the opportunity to review and vative growth management law of 1962 was the first state level legislation of its kind, and required citizen involvement in planning (Lowry et al., 1997). One of the most prominent examples of early federal level citizen participation requirements was the Economic Opportunity Act of 1964. The Act mandated community antipoverty efforts be overseen by local action programs that were developed and members of the groups to be s 452,

PAGE 64

50 title II). The Act facilitated the establishment of over a thousand community action programs. However, critics argued that the Office of Economic Opportunity, the entity that oversaw the program, misint Economic Opportunity Act spurred very little genuine participation of underrepresented stakeholders and that The Act promoted clashes between individual local action programs and elected ci ty officials (Moy nihan, 1969). The 1954 Urban Renewal Program was the first federal level citizen participation mandate in the United States. It was expanded during the 1960s Model Cities and War on Poverty programs Throughout the 1970s Congress added participation requi rements to a diversity of federal regulations. For example, citizen participation became a requirement of the Coastal Zone Management Act and the Energy Reorganization Act. What was the government trying to achieve by mandating public participation? The p lanning literature reasons that the early motivation behind these mandates was to increase the accountability and commitment of local governments to the principles of democracy. These principles include the right to information, the right to be consulted w ith, and the opportunity to express goals, priorities, and opinions on government decisions. At the core of democracy is the necessity of representing the interests of marginalized, powerless, and disadvantaged stakeholders in governmental decision making and, in theory, public participation in planning contributes to the realization of these broader democratic principles (Arnstein 1969 ; Burke, 1979; Day, 1997; Godschalk and Mills, 1966; Fainstein and Fainstein 1985). The shift towards increasingly bottom up approaches to planning led to the acceptance of a broad range of collaborative planning strategies. Diverse approaches to

PAGE 65

51 community engagement including public meetings, advisory committees, charrettes, workshops, and task forces became increasingl y common during the 1960s. Since then, researchers have evaluated the success of these approaches for generating increased public participation, empowerment and plan implementation outcomes (Beierle & Cayford, 2002; Gunton & Day, 2003; Arnstein, 1969). I n her article A Ladder of Citizen Participation Sherry Arnstein (1969) presents a theoretical ladder to illustrate how even with the emergence of new state and federal laws requiring public participation in government decision making, the majority of that participation ranks low on the ladd er of meaningful participation. The communit y participation outcomes can be explained in part by the specific collaborative planning activities that guide the engagement of, and interactions between, diverse participants during the planning process. ladder has eight rungs that are divided into three tiers illustrating a hierarchy of public power and citizen participation (Figure 4 )

PAGE 66

52 Figure 4 : Arnstein's Ladder of Citizen Participation The lowest rungs of the ladder are manipulation and therapy At this level of the ladder planning processes are non participatory. In ascending order the middle rungs of the ladder are informing consultation and placation These rungs are described as degrees of tokenism, characterized by one way flows of inform ation, shallow surveys, and limited decision making po wer by citizens partnership delegated power and citizen control At this level, the rungs represent ascending degrees of citizen power. In the part nership model, power is legitimately redistributed through processes of negotiation and discourse between citizens and those in power. Delegated power affords

PAGE 67

53 citizens the power to make decisions and to provide accountability to the public. Finally, citize n control describes a level of participation in which the powerless are granted the authority to collect information, prioritize programs, and direct planning in its entirety. ad to varying degrees of participation and power sharing, ranging from manipulation, to placation, to citizen control The knowledge systems of communities are complex and are continuously being transformed through experience, social factors, and even by t he influence of power relations both internal and external to a community. Ultimately, how power and control over decision making are critical for achieving meaningful and informed public participation To date, the nuances be tween public participation strategies have been evaluated and distinguished using similar value based scales including equitability, inclusivity, impartiality, and transparency. Density and frequency of community participation in planning have been shown to depend strongly on the identification of appropriate community participants as well as on the choices planners make regarding specific collaborative planning techniques (Brody, 2003; Brody, Godschalk & Burby, 2003). (Godschalk et al ., 2003; Innes & Gruber, 2005). Decision making in American policy and planning has long relied on sci entific expertise for guidance. As a consequence, the implementation of citizen participation has often been a difficult concept for many policy makers to grasp an d has frequently been excluded. In many cases the scope of relevant stakeholders seems infinitely large. There may exist such a wide variety of methods for involving citizens in planning that selecting the most appropriate approach can seem like an insurmountable challenge. For this

PAGE 68

54 public participation that are based on larger community and project goals Local Knowledge Santa Fe Institute founder George Cowan writes that that the science of the 21st century will be the science of complexity (Waldrop, 1992). Traditionally, reductionism has been the way in which scientists have attempted to understand prob lems. The idea behind reductionism is that if we understand the parts, then we can understand how the system works as a whole. In the context of planning, however, it has become increasingly clear that the rational model of understanding systems no longer reaches desired outcomes. Why? Some systems cannot be evaluated using the rational, reductionist model because the way in which the components of the system interact is key to understanding how the whole system works. Systems like these are called complex systems C omplexity t heory emerged as a framework for understanding how complex systems function and change. A number of disciplinary perspectives apply complexity theory from ecologists and social scientists, to economists and computer scientists. Over t ime, n ew understandings of c omplex systems emerged including the concept of complex adaptive systems Planning systems and the social ecological communities functioning within them have been described as complex adaptive systems, defined by changing enviro nments, numerous stakeholders and values, and interacting parts. Hopkins (2001) writes that planning is only useful to complex adaptive systems if the process is integrated as a functional element within the system rather than as an external means for achi eving control. This is because complex adaptive systems achieve

PAGE 69

55 higher levels of performance when innovation and dramatic shifts in activity are able to occur. Innovation and adaptation depend, however, on adequate transfers of information and diverse form s of knowledge bet ween interacting stakeholders. Innes and Booher (1999) stress that in order to understand the capacity of a complex adaptive system to learn, build shared meaning, and be resilient to unexpected shocks, it is critical to evaluate issues of power, networks, and self organizing that are inherent to the system (Innes and Booher, 1999). Network theory has merged with planning research on the subject of hazards management and mitigation planning. For example, increasing drought risks and uncer tainty about exposure due to climate change have propelled fire management to the forefront of U.S. emergency manag ement ) work on adaptive learning, collective memory, and fire risk highlights the potential for networks to impact residential knowledge and local resilience. change and uncertainty that contributed to increa sed community resilience. The informal network provided the community with a range of alternative social ecological relationships and configurations. As a result, it could offer flexible options for reconfiguration without causing catastrophic change to th e system. To put it another way, ). Researchers have also explored the implications of inflexible management regimes on complex ada ptive systems. In the context of post disaster recovery planning research shows that state bureaucracies rarely increase post disaster social ecological

PAGE 70

56 resilience (Adger et al., 2005; Folke et al. 2005). This is attributed to the governmental tendency to informal, community networks are more likely to develop a diversity of innovative responses to rapid change and alternative states of being, thus enhancing community resilience (Fo lke et al, 2005; Goldstein, 2008 ). Goldstein argues that informal networks, are free from obligat ion, scrutiny, or pressure (2008 ). Therefore, they are able to innovate and devise more creative, flexible solutions that transcend organizati onal barriers and are adaptable to a div ersity of unexpected scenarios. In addition to evaluating how new forms of knowledge and flexible stakeholder networks are developed during participatory planning, hazards and communicative planning researchers have explored the potential of participatory processes to mobilize social memory and increase local resilience (Folke et al., 2005; Goldstein, 2009). Social experiences with p resent (and future) plans a nd policies (Folke et al., 2005 ). Although it draws on experience, it also allows for innovation, creative adaptation, and experimentation within a larger framework of accumulated community experience In their 2009 mitigation st rategy study, Berke, Song, and Stevens found that citizen participation rates during site plan review increased when planners employed highly communicative participation techniques (specifically, community forums, goal setting workshops, subcommittees, and citizen advisory committees). In comparing the characteristics of the relatively effective participation techniques with those that were ineffective, Berke, Song, and Stevens (2009) concluded that the participation of

PAGE 71

57 stakeholder groups was higher when pl anners chose to use discursive, interactive and personal information transmission strategies. all collaborative planning and decision making processes that seek to reach agreem ent boy a broad spectrum of stakeholders on a planning process output. For example, problem definition, goal setting, fact finding, priority setting, policy maki ng or implementation strategy. Participatory planning is, at its core, the act of creating new knowledge. Participants contribute their perspectives to the planning processes and diffuse knowledge to others in the process (Goldstein, 2009; Innes and Booher, 2010). The inclusion of knowledge provided by non expert stakeholders can be beneficial to p lanning processes because the perspectives of individuals outside of the professional urban planning sphere often contribute new solutions that are based on collective memory and are closely aligned with local contexts. Moreover, local knowledge has the potential to contribute insights about environment and place that the planning discipline may never have known or might have forgotten (Burby, 2003; Laurian, 2003; Colten & Sumpter, 2009; Coburn, 2003). Expertise plays only a small part in the local resil ience building process, and formal, scientific knowledge is complimented by lay, or local, knowledge. As adapted from Co burn (2003, 421), local knowledge is: Knowledge of specific relationships, events, characteristics and circumstances, as well as importa nt understandings of their meaning, in local contexts or settings; Generally acqu ired via life experience and mediated through cultural tradition;

PAGE 72

58 E rather than a hunch or sponta neous intuition; and Is made legitimate through community stories, public narratives, street theater and other public forums, as opposed to professional knowledge which is tested through peer review, through the media, or in the courts. In a complex world facing uncertain, diverse, and catastrophic climate change hazards, Innes and Booher (2008) make the case that adaptability and resilience require including those of non experts and the seldom heard (Innes and Boohe r, 2008). In reality, however, local knowledge is often marginalized, particularly when it is communicated by the poor, politically underrepresented, or socially disconnected. Better Plans, Safer Communities? In the U.S., s cholars and practitioners frequ ently call for broad public participation in planning as a condition for positive planning outcomes It is frequently argued that although it often takes more time and money to involve diverse groups of stakeholders in the process, the long term savings as sociated with communicative hazard mitigation planning can make up for the initial investment and, overall, successful collaboration often results in the identification and implementation of more appropriate mitigation options (Mileti 1999). Only a few stu dies have empirically tested the assumption that community representation and participation during the planning process will lead to greater levels of project implementation and/or better plans. The following

PAGE 73

59 section examines the body of work on the relati onship s between community participation in planning and the quality and implementation of local hazard mitigation plans. The Influence of P articipation on Implementation and Plan Q uality In the U.S., increasing disaster loss and vulnerability trends over the last three decades indicate that mitigation planning activity and public policy decisions at all levels of government have been ineffective at protecting lives and infrastructure from hazards and their impacts (Burby, 2006 ; Berke, Song, and Stevens 20 09). Mitigation planning research s hows that successful vulnerability reduction requires the integration of various planning tools, better coordination with stakeholder agencies, and a more holistic approach to public participation. This belief extends bey ond academia and into the policy realm. In its mitigation guide, FEMA emphasizes that mitigation planning approaches must go beyond the scope of traditional emergency management and incorporate areas of planning, development, economics, education, critical care, and cultural facilities. FEMA also underscores the importance of including the entire community in local planning activity (FEMA, 2001). The following sections highlight the empirical benefits of including the public in haza rd mitigation planning ac tivity. Internal Implementation Factors: The Plan and I ts Quality From the perspective of FEMA, the goal of including non expert stakeholders in local hazard mitigation planning is to reduce risk and vulnerability of people and property to future hazard events In order for risk reduction goals to be realized, knowledge sharing and public education must occur as well as implementation of proposed mitigation actions.

PAGE 74

60 Plan implementation outcomes are driven by a number of factors. These factors include pla nning agency ( Dalton & Burby, 1994 ), the knowledge and commitment of developers to plan policies (Laurian et al., 2004; McNamara & Healey, 1984) project scale, interactions between agencies and developers (Burby et al., 1998 ), public involvement and quali ty of the plan itself (Dalton & Burby, 1994; Berke, Song, and Stevens 2009) Although the literature on policy implementation is well developed in the political science and public administration disciplines, the planning literature contains relatively few studies related to plan quality the implementation of local plans Dalton & Burby (1994) conducted one of the key empirical studies on implementation outcomes from the planning perspective Their research investigated the influence of states on developmen t management Dalton & Burby (1994) analyzed 176 communities across five states. In addition to confirming the importance of state mandates on plan quality, the study identified plan quality as a critical factor contributing to strong plan implemen tation outcomes (1994 ). In an attempt to understand the relationship between hazard mitigation planning and loss prevention, a specialized segment of planning research developed around the notion of hazard mitigation planning documents as predictors of plan imple mentation outcomes (Talen, 1996; Hoch, 1998; Brody, 2003). Within this body of research, the concept of plan quality has been used as both an outcome variable for assessing planning processes and as a predictor variable for assessing the imp lementation of planned actions In 2003, Brody (2003) examined the impact of stakeholder participation and representation on ecosystem management strategies. Brody (2003) tested the relationship between community participation in the planning process and the quality of t he local

PAGE 75

61 plans, by looking at the breadth of stakeholder groups involved in planning and at the presence of specific stakeholders. Statistical results found that the breadth of stakeholder groups involved in planning, as well as the presence of specific st akeholders (industry, government and NGOs), does significantly increase the quality of ecosystem management plans (Brody, 2003). In 2004, Laurian et al. (2004) conducted a systemati c, quantitative analysis of key determin ants of plan implementation. Their analysis was based on 353 permits implementing six local environmental plans in New Zealand, and on surveys of the developers who obtained the permits and of the planning agencies that granted the permits. The research team found that plan implementation i n this context was largely driven by the quality of the plans rather than by the characteristics of the pla nners (Laurian et al., 2004). In 2009 Berke, Song, and Stevens (2009) conducted a study related to plan implementation outcomes in the context of hazard mitigation and risk reduction. The research team looked at the incorporation of hazard mitigation strategies into New Urbanist development projects and found that mitigation implementation was positively correlated with high levels of community part icipation during the planning process. Specifically, public participation during site plan review had a positive effect on the incorporation of hazard mitigation strategies into future development projects. Across the sample, the number of hazard mitigation strategies used in the proposed development projects increased with the number of stakeholder groups inv olved in the site plan review. Berke, Song, and Stevens (2009) work supports the hypothesis that greater public participation increases the likelihood that hazard mitigation actions are implemented

PAGE 76

62 during future development decisions This conclusion is supported by previous research showing that local governments are more inclined to pursue hazard mitigation efforts when citizens call attenti on to hazards, and that the likelihood that citizens will call attention to hazards increases with the number of stakeholder groups engaged in the planning process (Burby, 2003; Godschalk et al., 2003 ). Further exploration is needed to understand why stak eholder participation may lead to stronger planning outcomes and what kinds of participation activity lead to better planning. Cons idering increasing tends in risk and vulnerability, as well as the amount of time and money spent on the development of publi c participation strategies and on complying with federal and state participation mandate s the hazard mitigation community has much to gain from this line of inquiry. Assessing Plan Quality The current body of literature draws strong positive connections between plan quality and plan implementation (Laurian et al., 2004; Dalton & Burby, 1994), between public participation and plan implementation ( Berke, Song, and Stevens 2009; Brody, Godshcalk, and Burby 2003; Godschalk and Stiftel 1981; Ka iser, Godschal k and Chapin 1995 ), and between public participation and plan quality (Brody, 2003). The content analysis approach to evaluating plan quality has primarily been used to assess the quality of comprehensive development plans between communities with diverse regulatory and design approaches (Berke & French, 1994). Although emergency management concerns and priorities are different from those of urban development projects, many

PAGE 77

63 understandings from the analysis of comprehensive plans are relevant to the evaluati on of hazard mitigation plan quality. Early attempts to evaluate plan quality focused on variations in decision making ternatives based on their ability to achieve the goals laid out in the plan. Litchfield et al. (1975) later expanded upon this method of plan quality assessment by evaluating how different groups of community stakeholders were affected b y different develop ment plans. Alterman and Hill (1978) later developed a consistency measure to determine how well comprehensive plans were being implemented across a range of detailed, site specific plans. They determined consistency by comparing the proportions of land t hat were designated for the same use in comprehensive plans and in site specific plans. State mandates that require local plans to be consistent with state goals began to rely on the consistency measure as a barometer of plan quality (Knapp & Nelson, 1992) The consistency concept is limited, however, because jurisdiction consistency with state mandates or comprehensive plans does not automatically guarantee high plan quality. A number of important efforts to define the characteristics of high plan quality emerged from the literature soon after the consistency measure. As a body of work, these studies established consensus concerning the key planning characteristics that influence the decisions of local governments the most and, therefore, are most likely t o lead to superior plan implementation outcomes. related to the definition of core plan quality characteristics. The three core characteristics

PAGE 78

64 of plan quality are as follows: a strong factual basis, clearly articulated goals, and appropriately directed policies/actions. Fact basis refers to the existing local conditions of a place, previous historical events and existing needs related to the physical and socio economic developme nt of a community. Goals represent general aspirations, problem abatement, and needs that are premised on local values. Finally, policies are general guides to decisions (or actions) about the location and types of development that are based on the plannin g goals (Chapin & Kaiser, 1979). Empirical studies have applied the three core characteristics of plan quality to evaluate the ability of impact assessments to affect decision making in local governments. In 1978, Fishman evaluated the land use and housing elements of comprehensive plans in 27 U.S. cities and concluded that the best plans contain specific, well underscores the need for policy language to be action oriented. For exam ple, using words like will or must rather than might or should was characteristic of development proposals in the case cities that were more consistent with existing comprehensive plans (Fishman, 1978). In 1990, Connerly looked at the quality of the housing elements of plans created ten Florida communities. His study provides several explanations for the high implementation of proposed planning actions in guiding the development decisions of local governments. First, he a sserts that high quality plans have fact bases that are characterized by detailed inventories of local housing stocks and assessments of current and projected housing needs and capacities. Second, high quality plans have comprehensive goals that address a broad spectrum of housing needs. Third, high quality

PAGE 79

65 plans contain clearly stated policies that are specific to the achievement of each goal. Each of the policies also articulate identified funding sources and implementati on timelines (Connerly, 1990). Th e body of research on the relationship between plan quality and local government plan implementation displays a clear pattern in support of the three plan quality components initially identified by Chapin & Kaiser (1979). The highest quality plans are typi cally those that a) have a strong factual basis that defines local needs, b) have clear and comprehensive goals that evidence a commitment to local needs, and b) have specific, action oriented policies that are geared towards achieving plan goals. Although creating high quality mitigation plans is not the end goal, understanding how to create high quality plans is an important intermediary step towards achieving reduced vulnerability and loss from hazards and disasters. Linking Participatory Planning with H azard Mitigation Plan Quality Even in states with highly progressive hazard mitigation policies and mandates, many local jurisdictions are currently developing low quality, cookie cutter hazard mitigation plans. Hazard mitigation plan quality is substantia lly low across the country and, in a number of cases, quite variable even within states. Perhaps the most concerning trend is that many plans lack meaningful site specific information about local capacity and social vulnerability, information that is essen tial to resilience building ( Schwab, 2011 ). The burgeoning body of hazard mitigation planning research has refined the three core plan quality components developed by Chapin & Kaiser (1979) specifically for use

PAGE 80

66 in the hazard mitigation planning context. Indicators for the fact base component of hazard mitigation plans cover the location and extent of hazard damage, including the magnitude of hazard events, the location of critical facilities, numbers of exposed populations, and evacuation clearance times Indicators for the goals component of the hazard mitigation plan quality index cover physical hazard impacts (e.g water quality impacts and the preservation of natural areas). The goals component includes goals related to physical and economic impacts of hazards (e.g. any goals to equitably distribute hazards management costs an d reduce property loss), and impacts of hazards on public interests (e.g. any public safety goals or hazard awareness initiatives). The policy /actions component addresses actions associated with increasing community stakeholder awareness of risks and hazards, regulatory actions, mitigation incentives, actions taken to reduce structural losses, and actions rela ted to post disaster recovery. Using the three core plan quality compon ents We n g er et al. (1980) assessed the quality of 71 local emergency plans Through empirical testing, We n g er et al. (1980) concluded that the majority of pre disaster emergency planning activities in the ir sample were oriented largely towards the product ion of plans as deliverables for the acquisition of funding rather than as tools for future development and policy making. The highest quality plans in their study were those plans created by frequent community wide sharing of information, fact finding, an d proposal ge neration (We n g er et al., 1980). Hazards scholars Berke and French (1994) have also evaluated hazard mitigation plan quality this time by studying the relationship between quality outcomes and state planning mandates in five states: California Texas, Washington, North Carolina, and Florida. The five selected states provide a range of variation in geographic features as

PAGE 81

6 7 well as in state mandate local planning characteristics. In their study, the characteristics of two mandate design features (s tructural features and facilitating features) were identified in the each of the local plans. Berke and French (1994) then showed gamma correlations between the structural and facilitating mandate design features, and local plan quality characteristics. Th eir research found that both types of mandate features are strongly related to the strength of policies within local plans, but they have weaker relati ons with goals and fact basis. Berke and French (1994) suggest that the relatively weak relationship bet ween planning mandate features and fact basis may be attributed to how fact base is formulated. Federal agencies largely determine the extent to which plans use information, more so than do state mandates For example, federal agencies produce detailed flood insurance rate maps that most communities use for plan making. This finding may explain why their study found local commitment building to have the highest correlation with fact basis of all the mandate features. Berke and French suggest that as comm itment to planning increases, communities are more likely to incorporate available mapping data into plan making (1994 ). As previously noted, planning researchers currently measure the fact base component of hazard mitigation plans by identifying cited in formation including previous hazard damage s probability of hazard events and the l ocation of critical facilities. However, despite the significant role that people play in the ability of a community to be resilient in the face of disaster, researchers hav e failed to assess the incorporation of social vulnerability concerns into their hazard mitiga tion plan quality assessments.

PAGE 82

68 In response to this disconnect between theory and practice, this study has modified the fact base component of the plan quality assessment tool to include citations of social vulnerability information in mitigation plan scoring criteria Specifically fact base indicators have been refined to include contextual data on local demographics, education levels, healt h services, social services and economy, all of which are important indicators of local social vulnerability and key factors affecting the dis aster resilience of a community (Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Berke et al., 1998; Godschalk et a l., 1999; Phillips et al., 2009). This methodological strategy is discussed in detail in Chapter 3. Placement and Contribution of this Study This dissertation research expands upon the current body of literature by assessing how various models of public pa rticipation as well as local social vulnerability characteristics, affect hazard mitigation plan quality outcomes This study does not attempt to identify the etiology or an tecedent factors that explain public participation in hazard mitigation planning R ather, it takes an exploratory approach to identify emerging themes related to variations in hazard mitigation plan quality between communities The following section describes the conceptual framework that guides this research. Conceptual F ramework As previously discussed the literature includes a variety of variables that affect the impact of public participation on local hazard mitigation plan quality. A conceptual framework can help integrate these variable s and organize analysis by identifying gene ral sets of variables of interest. To this end, t his study integrates three models developed by

PAGE 83

69 Callon (1999) and adapted by Yli Pelkonen & Kohl (2005) to describe the various degrees of involvement of lay person stakeholders in the creation of the knowl edge and information on whi ch planning decisions are made. Grounded in the social ecological systems literature, original models focus on the diversity of possible types of non expert participation in planning. The first model, hereafter re ferred to as M1, is called the public education model. M1 describes the opposition between technical and experiential forms of knowledge. In this model, non expert stakeholders have nothing to contribute to professional knowledge of risk, vulnerability or hazard mitigation during plan making Here local goal setting, plannin g, and decision making are based solely on the knowledge and agendas of or ganizational stakeholders and that knowledge is derived primarily from historical data and modeling. Model 2 (M2) is called the public debate model. Here, local non expert stakehold ers are perceived as having less robust levels of knowledge than professionals. They are however, invited to participate (to a limited extent) i n goal setting, risk assessment and mitigation prioritization. In this model, professional knowledge is enriche d by local community information sharing and together, organizational and community stakeholders create a framework for decision making. Callon (1999) breaks up the M2 model into two sub models in order to describe different types of participation in more detail. The two sub models are: M2a, the public hearing model and M2b, the public representation model. The public hearing model (M2a) describes interactions in which hazard mitigation planners collect information directly from participants. In this model, the aim of the collaborative planning process is

PAGE 84

70 for community knowledge to directly impact planning outcomes (Callon, 199 9; Yli Pelkonen & Kohl, 2005). In the public representation model (M2b) community knowledge is heard indirectly through the use of c ommunity representatives. These representatives can be, for example, homeowner associations or local NGOs. In this model of participation, community knowledge can influence planning decisions; however, because planners are channeling it through community r epresentatives, there is a greater risk of the information bein g misinterpreted. Model 3 (M3), the co production of knowledge model, represents circumstances in which non expert stakeholders are actively involved in the information collection, goal setting and decision making that shapes local plans. In this model, community input is treated as equally important to planning and decision making as is professional input. In some circumstances, community perspectives and experiential knowledge are even treate d as more i mportant. This can happen in situations where professional knowledge on a topic is scarce (i.e. when plans require information about social factors related to a es an essential part of information production and decision mak ing (Yli Pelkonen & Kohl, 2005). There exists a fourth model along the spectrum of participation. Model 4 (M4), or the community construction of knowledge model illustrates a participation model in which hazard mitigation planners rely entirely on information, understandings, and priorities based on knowledge from local communities to inform their mitigation plan proposal. When hazard mitigation planners attempt to incorporate information about

PAGE 85

71 social vulnerability and/or community capacity into local plans (i.e. social networks, information about community gatekeepers, or adaptive behavior based on experiential learning) professional knowledge about these social determinants of risk is largely non existent must defer to local stakeholders, including socially marginalized or unseen groups, to construct an understanding of local social issues related to hazard risk and vulnerability (Wisner, 2 006; Folke et al., 2003, 2005; Folk e, 2006). For the purpose of this study, the four models described above provide a functional framework by which to assess local hazard mitigation planning practice at the city scale. Moving forward, the models are used to categorize individual hazard mitigation plan ning schemes and to guide our evaluation of the relationships between stakeholder participation, social vulnerability fact ors and plan quality outcomes. The framework for this study adapted by planning and social ecological systems scholars Yli Pelkonen & Kohl (2005), four models together into a larger conceptual model of stakeholder participation during local hazard mitigation planning The conceptual framework illustrate s the interactions through which lay person s takeholders participate in information production, goal setting and decision making during hazard mitigation planning (Callon, 19 99; Yli Pelkonen & Kohl, 2005). Below, Figure 5 illustrates the conceptual framework used in this st udy for categorizing stakeholder participation during local hazard mitigation planning. Models M1 through M4 are characterized by the specific participatory planning strategies used during the local planning process The interactions between local com munit y/citizen stakeholders, organizational stakeholders/technical experts and mitigation planning teams

PAGE 86

72 flow according to the four vec tors illustrated in Figure 5 The arrows A, B, C and D show the intera ctions between the participants and identify the key social process of collaborative planning. The dashed lines for interactions B, C and D illustrate that they are conditional interactions. Figure 5 : Framework for Stakeholder Participation During Local Hazard Mitigation Planning 2 In this study, I evaluate interactions A thro ugh D in order to attribute a participatory planning model to each of the 34 local hazard mitigation plans included in the sample frame. Based on which interactions are present 2 Adapted from Yli Pelkonen & Kohl ( 2005 )

PAGE 87

73 planning process, one of the four participatory planning models M1, M2a /M2 b and M3, M4 will be a ttributed to each of the 34 local hazard mitigation plans. The framework of stakeholder participation during local hazard mitigation planning can be connected back to the existing literature related to participatory planning and mitigation planning outcomes. In 2009, Deyle & Slotterback (2009) found positive planning outcomes i n counties where interactions A and B were equally strong (planning model M3). Deyle & Slotterback used a quasiexperiemental research design to observe and evaluate the mitigation plans and processes in eight Florida counties. Using a pretest posttest desi gn, their study uncovered higher levels of mutual understanding of planning problems (and of solutions) in counties where community stakeholders were able to directly incorporate their knowledge and ideas to the formal planning deliberations. This finding is significant because strong mutual understanding among stakeholder groups has been shown to generate stronger constituencies for both plan adoption and implementation (Deyle and Slotterback, 2009; Brody, Godshcalk, and Burby 2003; Godschalk and Stiftel 1 981; Kaiser, Godschalk and Chapin 1995; May et al. 1996). Pearce (2003) step for the field of disaster management In order to achieve sustainable hazard mitigation, Pearce (2003) argues that it is necessary to closely integrate mi tigation and participatory community planning To illustrate her point, Pearce presents a case study of participatory mitigation planning in Pearce describes Portola as an example of successful sustainable hazard mitigation in action, exemplified by a high degree of public participation in planning (Pearce, 2003). Based on

PAGE 88

74 the conceptual model of stakeholder participation pres ented here Portola Valley applied the M2b model of public participatio n. It is important to note that the participation models are not always contradictory. They can sometimes occur together and complement one another. Moreover, in all of these models, ev en if participants are heard, it is possible for elements of both types of information and knowledge received by planners through interactions A and B to be lost in the process and not transmitted to the final plan (Innes & Booher, 2007) For example, t his can happen as the mitigation planning team filters important information from other stakeholders during the final stages of plan making (interaction D) ( Yli Pelkonen & Kohl, 2005 ) To address this element of the conceptual framework, this study includes i n depth interviews with mitigation planning team m on participation models and plan quality outcomes. Conclusion Under the Stafford Act and DMA 2000 decision makers are required to base hazard mitigation planning decisions not only on adequate technical information from risk assessment studies, but also on the priorities and knowledge of local residents, community stakeholder groups and other stakehold ers who can provide information on an A key theme that this study explores is how local knowledge is integrated into planning and decision making. One of the main propositions of this study is th at the engagement of local stakeholders in decision making and vulnerability assessment is critical to reaching disaster resiliency goals. Although the process of collaborative mitigation planning is

PAGE 89

75 often associated with higher initial costs and extended project timelines the up front investment is cost effective when it boils down to effective plan implementation, long term policy agreements, and safe, resilient communities (Godschalk et al., 1994). Moreover, previous research has found that communicativ e processes including collaboration, dialogue, and consensus building generate myriad secondary benefits. These benefits include the creation of new forms of knowledge and political capital (Forester, 1999; Innes and Booher, 1999b), increased levels of tru st, stronger commitments to plan and policy implementation, and increased social capital and community resilience (Burby, 2003; Innes, 1996; Innes et al., 1994; Creighton, 1992; Wondolleck and Yaffee, 2000; Forester, 1999; Moore, 1995). Ultimately, the con clusions of this dissertation will have practical significance to local, state and federal planners, emergency managers and policy makers seeking identify ways of improving local citizen participation in hazard mitigation planning. Understanding how partic ipation affects local hazard mitigation plan quality, and how social vulnerability affects participation, will provide important insights into how local hazard mitigation plans can be strengthened to more adequately address the root causes behind increasin g vulnerability and risk.

PAGE 90

76 CHAPTER III M ETHODS Introduction This study examines the relationship between models of participatory planning and local hazard mitigation plan quality and explores the factors that drive variations in plan quality and content. The guiding questions for this research study are: 1) What i s the quality of local hazard mitigation plans created in the n orthern California Bay Delta region and to what extent has the public participated in the plan making process? 2) What key relationships exist between local community characteristics, public p articipation in planning and hazard mitigation plan quality outcomes? The following chapter describes the methods by which the data to answer these questions was collected and analyzed. It begins with a description of the selected study area and explains the sample selection process in detail. It then describes the characteristics of participants identified for the interview portion of the study and how they were selected. The data collection tools used for this research are also described in detail and include the following: the plan quality assessment instrument and protocol, interview guides, and a reflexive journal A thorough description of the steps involved in data co llection is provided and efforts to protect the reliability of findings and the validity of inferences are detailed.

PAGE 91

77 Study Area The state of California was selected as the location for this study. California is unique in that it has a reputation as a na tional leader in the mitigation of hazard impacts associated with earthquakes, fires, floods, climate change and other natural disasters. Recently, this reputation has been validated by the U.S. Department of Homeland Security through FEMA. In December of 2007, FEMA designated the 2007 State of California Multi Hazard Mitigation Plan as an Enhanced State Mitigation Plan Having funding from 15% to 20% of the total individual and public assistance funds spent by FEMA after each Pr esidential Disaster Declaration. At the time, only eleven other states the risk of death, injury and property loss from hazards. Program (HMP) administers the LHMP program for the state. In this role, Cal EMA supports and assists local governments in t he development of their LHMPs and heads the review process for all plans before they are forwarded to FEMA for approval. Local governments in California must have an approved LHMP to be eligible for certain mitigation programs authorized under the Stafford Act and, ultimately, the goal of the LHMP program is for all local governments in California to have FEMA approved LHMPS. This study focuses specifically on the northern California Bay Area/Delta region of California. This region of northern California p rovides a particularly information rich

PAGE 92

78 setting to study how diverse social factors affect local hazard mitigation plan quality. For one, the region is demographically diverse and its levels of social vulnerability vary greatly by county 3 Moreover, it boa sts a diversity of geographic environments. As a result, communities throughout the region are simultaneously managing and planning for a variety of hazard risks. Examining plans within the selected study area provides an opportunity to evaluate the effec ts of public participation on plan quality while controlling for state specific policy factors, regional economics, and climate. Ultimately, the sample location was selected not to ensure comparability among places in different states, but to provide insig ht into how public participation affects LHMP plan quality in a state that is taking a singularly progressive stance towards disaste r risk reduction and planning. Sample S election The sample frame for this study consists of all jurisdictions within the st udy area with completed LHMPs as of July 2011 (including those in the FEMA approval process as of April 2011), totaling 215 jurisdictions. This information was obtained from a list of all local jurisdictions in the state of California with FEMA approved LH MPs (including those plans in the approval pipeline). The list was provided by Cal EMA and is current as of April 2011. There are twelve counties represented in the sample frame. They include 3 Social vulnerability is commonly calculated using th e Social Vulnerability Index (SoVI). So VI is calculated at the county scale by scoring locations across 11 factors (Cutter et al., 2003).

PAGE 93

79 the nine counties represented by the Association of Bay Area Gov ernments (ABAG) Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano, and Sonoma and the three delta counties of Placer, Yolo and Sacramento. Figure 6 : Map of California Counties Included in Study Area The first stage of the sampling strategy was to limit the sample of LHMPs to the community scale by excluding all special districts tribal entities, and agency jurisdictions. This study focuses specifical ly on city and county scale jurisdictions because they are the human settlements that are most vulnerable to both rapid and slow onset disasters. Not only are cities characterized by high population densities and concentrations of political, economic and social resources, but they are also characterized

PAGE 94

80 by a strong dependency on infrastructure and services. Moreover, rapid urbanization and population growth make it difficult for decision makers and communities to respond quickly and plan for change. As a r esult, cities are the places where ecological, economic, and social disparities manifest most notably. This research project focuses deliberately on cities because these are the communities that offer the most to learn and the most to gain. By limiting the sample to city scale hazard mitigation plans, the list of LHMPs was paired down to 98. Next, the sample of 98 LHMPs was limited to those written for cities with populations between 10,000 and 200,000 people. This paired the sample down further to a total of 71 plans. This population range was chosen for the following reasons: cities with a population of 10,000 or more provide a wider variety of functions and services than do smaller cities and it is believed that large cities, such as San Jose (with a popu lation of 965,000), have very different contextual fa ctors and may skew the sample. The third stage of the sampling strategy was to limit the sample frame to plans that were available online as of July 1, 2011. Transparency, including the availability of planning documents to the public, has been identified by FEMA as important components for successful los s and vulnerability reduction: An effective and open public involvement process ensures that all citizens understand risks and vulnerabilities so that t hey will work with the jurisdiction and support policies, actions and tools that over the long term will lead to a reduction in future losses. Moreover, continued public involvement in the planning process is required by FEMA for LHMP approval. By sampling plans that are available to the general public, this study is effectively evaluating LHMPs from cities that

PAGE 95

81 have met the most basic requirement of the 44 CFR 201.6(c)(4)(iii): the provision of opportunities for continued public involvement in the planning process. After the three sampling phases, a final sample of 34 LHMPs remained. Table 4 depicts the 34 city plans that were selected for analysis based on the sampling techniques outlined above: Table 4 : Sample Cities County City Planning Scale* Population** Placer Roseville SJ 109,497 Alameda Berkeley SJ 100,877 Napa Napa SJ 74,071 San Mateo Redwood City SJ 73,099 Marin Mill Valley SJ 13,189 Marin Larkspur SJ 11,647 Sonoma Santa Rosa MJ Region 154,017 Santa Clara Sunnyvale MJ Region 130,256 Santa Clara Santa Clara MJ Region 108,075 Solano Fairfield MJ Region 103,305 Solano Vacaville MJ Region 91,828 Santa Clara Gilroy MJ Region 48,158 Santa Clara Campbell MJ Region 38,067 Santa Clara Morgan Hill MJ Region 37,099 Santa Clara Saratoga MJ Region 29,974 Contra Costa Concord MJ Region 120,775 San Mateo Burlingame MJ Region 27,363

PAGE 96

82 Table County City Planning Scale* Population** Sonoma Windsor MJ Region 25,158 Contra Costa Orinda MJ Region 18,198 San Mateo Half Moon Bay MJ Region 12,271 Marin San Anselmo MJ Region 11,924 San Mateo Hillsborough MJ Region 10,651 Sacramento Elk Grove MJ County 130,007 Sacramento Citrus Heights MJ County 84,269 Sacramento Folsom MJ County 66,399 Yolo Davis MJ County 61,866 Sacramento Rancho Cordova MJ County 60,008 Yolo Woodland MJ County 53,069 Placer Rocklin MJ County 51,281 Yolo West Sacramento MJ County 45,504 Placer Lincoln MJ County 40,382 Sacramento Galt MJ County 23,535 Napa American Canyon MJ County 15,943 Placer Auburn MJ County 13,068 Data from city LHMP documents ** Data from 2005 2009 American Community Survey 5 year Estimates Each of the 34 jurisdictions selected have adhered to the participation requirements of the state local hazard mitig ation planning mandates, all jurisdictions are vulnerable to threats from natural hazards, and all jurisdictions have pre pared updated plans since 2004.

PAGE 97

83 The plan quality of each of the LHMPs listed above was measured using an analysis tool that incorporate s hazard mitigation and social vulnerability measures into pre existing conceptions of what constitutes a high quality plan. Ultimately, based on the results of the plan quality analysis, a select number of c ities were chosen for in depth interviews The interview methodology and analysis strategies are discussed in detail in Section 3.3. Data C ollection This dissertation uses both qualitative and quantitative research methods to answer the proposed research questions. Data collection has been broken i nto three methodological stages leading up to the synthesis of results (Figure 7 ). Figure 7 : Methodological Framework Stage 1 consists of a plan quality evaluation based on a quantitative hazard mitigation plan quality index. Additionally, Stage 1 includes an assessment of public participation models applied during planning for each of the 34 LHMPs. This assessment is based on the four models of public participation explained in the participation framework in Chapter 2. Stage 2 involves the collection of local community characteristic data from U.S. Census and SHELDUS databases including data on local social vulnerability, Stage 1 Plan Quality and Participation Evaluation Stage 2 Principal Component Analysis & Regression Analysis Stage 3 Participant Interviews and Qualitative Analysis

PAGE 98

84 socioeconomics, and hazard exposure. The specific social vulnerability, socioeconomic, and hazard exposure variables were selected based on findings from previous studies related to hazard mitigation plan quality and public participation (Brody, 2003; Berke & French, 1994; Berke, 1996; Cutter et al., 2003; Davis, Cook & Cohen, 2005; Chandra et al., 2010; Birkmann, 2006). Stand alone mitigation plans, or LHMPs, are often prepared in the aftermath of a disaster when the sense of crisis leads to a demand for immediate action (Burby, 1998). Although none of the local hazar d mitigation plans created in the sam ple frame were written in direct response to a disaster, hazard exposure (as measured by previous losses from disasters) has been shown to be an important driver of community participation in e post disaster period is a time when communities express the greatest levels of public support for hazard mitigation and preparedness activity (Burby, 1998). Moreover, the memories of repeated hazard events prompt some communities to prioritize mitigation planning to a greater extent than communities that may have high hazard risks, but no history of loss. In addition to hazard exposure, socioeconomic and social vulnerability factors have also been identified as potential constraints (and drivers) of miti gation policies and actions (Ollenburger & Tobin, 1998; Tobin, 1999). Researchers have theorized that socioeconomic and cultural factors can affect attitudes towards mitigation and recovery efforts and that they can act as filters in the hazard mitigation planning system (Tobin, 1999). A principal component analysis has been performed on the selected community characteristic data in order to reduce the many demographic variables to a smaller number of predictor variables (or factors) for use in the regressi on analysis.

PAGE 99

85 Empirical research has not yet addressed the relationship between socioeconomics, social vulnerability factors and plan quality. Nor has the literature addressed how these contextual factors impact public participation during local hazard mit igation planning. Using SAS software Stage 2 involves running a number of regression analyses on the community characteristic variables, public participation data and plan quality data. The expectation is that some of these factors may constrain the abili ty of local communities to participate in local hazard mitigation planning processes. The regression analyses allow me to explore the nature and form and of the relationships between plan quality, participation models and local community characteristics. Stage 3 includes qualitative, in depth interviews with organizational and local community stakeholders from the local hazard mitigation planning teams from select cities in the sample frame. Driven by the results of the regression analyses, the detailed ca se studies were designed to provide a more robust understanding of the dynamics of public participation, plan quality and social vulnerability than is possible from regression analysis or plan quality assessment alone. Stage 1: Plan Quality E valuation In the following sections, the methodological approaches for the three stages of data collection and analysis are described in detail. Methodology Building on the methods from previous studies of plan quality, this study measures hazard mitigation plan quali ty by assessing three equally weighted plan components in a plan quality index: clearly articulated goals, appropriately directed

PAGE 100

86 policies/actions, and a strong factual basis (Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Berke et al., 1998; Godschalk et al., 1999). Previous research has contributed to the identification of a set of qualitative indicators for each of the three components to create a robust conception of plan quality. In the context of hazard mitigation planning, indicators for the goals c omponent of the plan quality index cover physical impacts of hazards (e.g. assessment of water quality impacts and the preservation of natural areas), economic impacts of hazards (e.g. any goals to equitable distribute hazards management costs and reduce p roperty loss), and impacts on public interests (e.g. any public safety goals or hazard awareness initiatives). The policy component of plan quality addresses actions associated with increasing community stakeholder awareness of hazards and mitigation, expl icit regulatory actions, mitigation incentives, the reduction of structural losses, and recovery concerns. Previous studies have measured the fact base component of plan quality by assessing the range of background information upon which planning decision s, goals and actions are predicated. In the context of hazard mitigation planning, the range of information includes data about the location and extent of hazard damage, including the magnitude of the hazard event, exposed populations and evacuation cleara nce times. For the purpose of this study, the fact base component has been augmented to include social vulnerability information Here, the fact base component includes background data on local demographics, education levels, health servic es, social servic es and economy all of which are commonly used indicators of local social vulnerability ( Phillips et al., 2009). Each plan quality indicator was measured on a 0 2 ordinal scale. 0 ) is not identified or mentioned, 1 ) is suggested or identified but not detailed, and

PAGE 101

87 2 ) is fully detailed or mandatory in the plan. The scoring methodology for this study is consistent with previous studies and involves assigning equal weights to the indicators of plan quality. In th is past this has been done to eliminate the need to make value judgments about which indicator or plan component deserves more emphasis in determining plan quality than others. Moreover, equal weighting has been successful at maintaining consistency in sta tistical results. For some components of the plan analysis tool, items have more than one indicator. For example, in the fact base component critical facilities can be either recorded in a list, mapped, or both. In these cases, scores were assessed using an index built by taking the total indicator score and dividing it by the total number of sub indicators (e.g., if an item received a 1 for mapping and a 1 for listing, the overall score for the item will be 1). This method allows for scores to remain on a scale of 0 to 2 and will favor LHMPs that support their decisions with multiple decision making tools. To ensure the validity of the plan quality analysis tool and consistency of results, one researcher was tasked with analyzing the entire sample of 34 pl ans. Three indices based on the identified plan quality components were created for each LHMP and then aggregated to yield a measure of overall plan quality. This procedure has been carried out previously by planning and hazards researchers including Berke et al. (1996, 1990), Berke & French (1994), and Brody (2003). Steps towards creating the indices for each of the three plan quality components are taken in four stages. First, scores for each indicator is summed within each individual plan component. Seco nd, the sums of these scores are divided by the total possible score for each plan component. Thir d, in order to put them on a scale of 0 to 10 the fractional plan

PAGE 102

88 component scores are each multiplied by 10. The final step involves the calculation of a to tal plan quality score for each LHMP by adding the scores of each of the three quality components. The maximum score for each LHMP is 30. The calculated plan quality component orded along with comprehensive community characteristic data for each city. Assessing Models of Community P articipation In addition to calculating plan quality scores across the three planning components, this study evaluates another key principle of pla n quality: local community participation in the planning process The planning literature broadly defines local community participation as the inclusion of diverse stakeholder interests including local business, advocacy groups, and lay people in planning and policy making. Because FEMA determines the approval requirements for local hazard mitigation plans this study looks to FEMA as a guide for defining what community participation means in this planning context. Under DMA 2000, FEMA mandates community par ticipation during local hazard mitigation planning. To help states, local jurisdictions and tribes improve their hazard mitigation planning activities, the agency has developed a set of mitigation planning ers topics related to the initiation of a hazard mitigation plan. This guide includes practical tips for assessing community su pport, building a planning team and engaging local communities.

PAGE 103

89 According to FEMA, one of the first steps towards building a planning team and engaging local communities is determining who the stakeholders are (FEMA, 2002). 3). FEMA recommen ds that stakeholder representation be included from the following groups: neighborhood groups and other non profit organizations and associations; state, regional, and local government representatives; businesses and development organizations; elected offi cials; federal agency representatives; academic institutions; and community or state planners (FEMA, 2002). In their how to guide, FEMA articulates that some of these stakeholders should be included because of their expert and technical knowledge (such as the state, regional, and local government representatives; businesses and development organizations; elected officials; federal agency representatives; and planners), while others should be included so that they can become educated about risk and mitigati on issues (neighborhood groups and other non profit organ izations and associations). For the purpose of this study, organizational stakeholders and technical experts fall under the first category of stakeholder participant, while the second category (mean t primarily for educational and consultative purposes) are the local community/citizen stakeholders who live and work wi thin the planning jurisdiction. Previous studies on participation and plan quality have measured the breadth of community stakeholders i nvolved in planning as an indicator of robust citizen participation. Others have looked at frequency of interaction between organizational and community stakeholders as a barometer for community participation (Brody, 2003; Burby, 2003). For this study, a m ethodology for assessing models of local community

PAGE 104

90 participation in planning was developed based on a review of previous research on participation and collaborative planning theory. Table 5 typology of citizen participation with the four conceptual models of community participation in hazard mitigation planning. The table provides a framework for attributing one of four models of community participation to the experiences of local community stakeholders during each of the 34 local hazard mitigation planning process.

PAGE 105

91 Table 5 : Models of Community Participation in Hazard Mitigation Planning Model of Community Participation in Hazard Mitigation Planning Ladder of Citizen Participation Collaborative Planning Approach M1 Manipulation N/A Therapy N/A M2 Informing Public service announcements Site visits Newspapers Leaflets, brochures, bulletins Unstaffed exhibits/displays Advertising Consultation Teleconferencing Surveys, interviews and questionnaires Deliberative polls Workshops Placation Staffed exhibits/displays Staffed telephone hotlines Website Public meeting Open house (including virtual) M3 Partnership Charrettes Visioning (including virtual) Consensus conferences Delegated Power Community advisory committees/liaison groups Community delegated juries M4 Community Control Community based data collection Community power in decision making and priority setting process This table is a tool for organizing LHMPs into one of four collaborative planning approaches by identifying the nature of the planning activities that were employed during the local hazard mitigation planning process. Based on the previously established de finition of citizen participation, coupled with the conceptual framework outlined above,

PAGE 106

92 mitigation planning process. The models were incorporated into the plan quality data set along with specific notes describing which community participants were involved in each pla n. Stage 2: Principal Component and Regression Analys is In order to isolate the effects of local community characteristics namely planning scale hazard exposure, and social vulnerability on fact base, plan quality and participation, a select set of variables was collected for analysis. T his stage of ana lysis is designed to identify relationships between planning contexts hazard exposure and social vulnerability factors on mitigation plan quality and community participation. Based on social vulnerability analyses conducted by Cutter et al. (2003), and t he body of literature on diverse vulnerability assessment models, 2 0 social vulnerability factors were selected and analyzed Table 6 outlines the three community characteristic categories and the 2 0 associated variables selected for this study.

PAGE 107

93 Table 6 : Community Characteristic Variables Community Characteristic C ategory Variables Socio economics Population size* Race: White* Race: Black or African American* Race: American Indian* Race: Asian* Race: Native Hawaiian or Pacific Islander* Race: Hispanic or Latino (of any race)* Poverty (% total population below poverty)* Unemployment* Median household income* Median home value* Hazard exposure Total county hazard losses (1990 2010)** Multi hazard exposure*** Social vulnerability Female headed household* % population living in group quarters* % Population >65* % Population <14* Linguistically isolated households* Educational attainment* Renter occupied housing* *Tract level data from the U.S. Census Bureau, American Community Survey 5 year Estimates. **Data from SHELDUS database. Crop damage plus property damage in 2009 inflation adjusted U.S. dollars. *** State of California Multi Hazard Mitigation Plan, 2010. The community characteristic data was gathered from the U.S. Census Bureau website portal for the American Community Survey 5 year Estimates and the SHELDUS database Using SAS, a principal component analysis was performed on community characteristic data included above in order to reduce the many demographic variables to a smaller number of predictor variables for use in the following regression analysis. Four factors (Factors 1, 2, 3, and 4) were extracted and used as independent variables in the regression analysis.

PAGE 108

94 The logistic regression analyses were also performed using SAS software T able 7 outlines the ten regression models completed for this study. Results are presented in Chapter 4. Table 7 : Summary of Regression Models Dependent Variable Independent Variables Model 1 Plan Quality Factors 1 4 Model 2 Fact Base Score Factors 1 4 Model 3 Actions Score Factors 1 4 Model 4 Goals Score Factors 1 4 Model 5 Participation Factors 1 4 Model 6 Plan Quality Participation Model 7 Fact Base Participation Model 8 Actions Score Participation Model 9 Goals Score Participation Model 10 Plan Quality Planning Scale Stage 3: Interviews and Qualitative Analysis In addition to the plan quality assessment and regression analyses, this study includes a set of qualitative, in depth interviews with participants from local ha zard mitigation planning teams. The purpose of the interviews is to gain detail rich information about local hazard mitigation planning in places where The data collected from the int erviews enable a focused investigation into the results of the previous quantitative analyses. Interview Sampling S trategy The case studies for this study focus on layers of interactions, including those between participant groups, cities, and counties. Therefore, it was necessary to make sampling decisions based on each layer of analysis. Key ques tions include: which cities should be sampled; and w hich participants should be interviewed from each city?

PAGE 109

95 Initially, the criterion purposive sampling techniqu e was used to select 34 local hazard mitigation plans for the plan quality and participation analysis based on jurisdiction type, population size, and accessibility of LHMP documents. For the interview stage of analysis, sample cities were selected based u pon a stratified purposeful sampling approach. The stratified purposeful sampling approach involves the identification of samples within samples. This means that researchers select particular cases for analysis that vary according to a key a dimension (or dimension s) (Miles & Huberman, 1994). The stratified purposeful sampling approach is meant to illustrate characteristics of particular subgroups of interest and facilitates comparisons between them (Miles & Huberman, 1994). Ultimately, the aim is to captur e major variations in a phenomeno n rather than a common theme. Interviewee S election The aim of the case studies is to identify key themes related to variations in planning activity. Specifically, the interviews are designed to draw out information rela ted to activity associated with planning actions and fact base. Interviewees were selected from a comprehensive sample of both organizational and local community stakeholders. In this study, organizational stakeholders include technical experts, policy mak ers and consulting professionals who are directly involved in the local hazard mitigation planning process, either as members of a hazard mitigation planning team or as outside consultants or i nformation providers (Figure 8 ).

PAGE 110

96 Figure 8 : Models of Stakeholde r Participation d uring Local Hazard Mitigation Planning As previously illustrated in Chapter 2, the conceptual framework that guides this study includes a second category of local hazard mitigation planning s takeholder participant: local community/citizen stakeholders. This category of stakeholders includes residents of the jurisdiction for which the plans are being developed, representatives of neighborhood and community special interest groups (school groups religious groups, environmental organizations, local historic preservation organizations, etc), local advocates, representatives and/or service providers for socially vulnerable groups. For the interview stage of this study, organizational and local com munity stakeholder participants were identified by the participant contact information recorded in each local hazard mitigation plan. The saturation sampling technique was used to select interviewees and involved contacting all recorded participants in the planning process, including members of the mitigation planning team as well as outside consultants and community participants.

PAGE 111

97 In addition to identifying the organizational and community participants recorded in the LHMP planning documents it was necessar y to identify additional participants who were not official members of the local hazard mitigation planning team, so as to incorporate broad data about local community ideology in the case study analysis. These additional local community participants were identified for interviews using the snowball or chain sampling technique. Because these participants are particularly difficult to identify through planning documents, their contact information was identified by referral from previous interview subjects. I nterview Data Collection and S torage In depth interviewees were contacted first through email. Follow up emails were sent to non respondents seven days after the initial email requesting an interview. In accordance with COMIRB human subjects policy, potent ial interviewees were not contacted more than three times to schedule an interview. A general interview guide approach was used for data collection, ensuring that the same general areas of information were collected from each interviewee. This provided a high level of focus for the interviews but also allowed for a degree of flexibility and adaptability during the conversation with the interviewee. Two interview guides were created for the c ase study phase of the study, one for organizational participants, and another fo r local community participants The interviews were designed to draw out details related to goals, values, challenges and experiences related to the local hazard mitigation planning process in addition to details about the public participation strategies employed during the planning process.

PAGE 112

98 The one on one in depth interviews were conducted via Skype and lasted an average of 30 minutes. The interviews were recorded (unless the interviewee choose not to be recorded in which case interview responses were recorded by hand) for the purpose of transcription and input into QSR NVivo 9 qualitative coding software. The information collected during the in depth interviews was neither sensitive nor potentially damaging; how ever, because the information is potentially identifiable, confidentiality was an important concern for this study. In order to assure confidentiality for study participants, this research adheres to the following confidentiality procedures: 1. Data has been reported in aggregate and no individuals are identified; 2. No names are used in this report; 3. Instead of reporting about individuals, the analysis focuses on groups of people; 4. Before each interview, participants were informed of the purpose of the research and the interviewer made certain that interviewees were comfortable during the interviews. Close observance of these procedures assured that informants were able to provide sincere, truthful information about their experiences and about their perspectives Reflexi ve J ournal The final data collection technique for this study consists of a reflexive journal made up of field notes. Reflexive journaling allows for researchers to describe their feelings and insights about conducting research while they are mov ing through the qualitative data collection process (Miles & Huberman, 1994). A reflexive journal serves

PAGE 113

99 as a research audit trail, accounting for what is done at each phase of the research process and why. It contributes to the reliability of the study as well as its internal and external validity. It also provides an additional method for organizing data and emerging ideas. Data Analysis and Validity The data collected for this study was analyzed in two stages, one quantitative and one qualitative. The following section described the data analysis procedures in detail. Plan Quality A nalysis Data collection for the plan quality analysis includes an ind ex driven scoring procedure based on LHMP content analysis. Once the fact base, goals and actions scores were calculated for each plan, overall plan quality scores were calculated based on the s ums of these three components. This study uses tract level co mmunity characteristic data from the U.S. Census Bureau. The hazard exposure data, specifically hazard losses over time and multi hazard Multi Hazard Mitigation Plan, respect ively. SAS software was used to perform the principal component analysis and reduce the community characteristic variables. The extracted community characteristic factors were then used in multiple regression models in order to validate the plan quality a ssessment tool and to examine relationships between social vulnerability, plan quality and participation.

PAGE 114

100 Qualitative Interview A nalysis The in depth, qualitative interviews were conducted over Skype so that the conversations could be recorded and carried out hands free. Some interviewees did not want their conversations recorded. In these instances, interview notes were taken on paper E ach of the interviews was transcribed and uploaded into QSR NVivo 9, a qualitative data analysis software tool for content coding and analysis. A codebook was developed for the purpose of analyzing the interview data. It is d escribed further in Chapter 5. Validity Research standards of objectivity and dependability have been maintained to ensure the validity of thi maintained by using the QSR NVivo 9 qualitative data analysis software. The software allows for systemic analysis by organizing and storing data in ways that allow researchers to keep track of int ricate relationships, searches, ideas and analysis. While the plan quality assessment methods have been designed to limit bias and maintain consistency, it is possible that another researcher conducting the same plan quality and participation assessment c ould turn out different scores. The use of subjectively determined calculations of plan quality and models of participation introduces concerns about bias. For this reason, a pilot test of my plan quality index was conducted on the city of Boulder, Colorad Additionally, I chose not to enlist additional researchers to score the sample of 34 plans

PAGE 115

101 as their efforts would introduce further bias to the analysis (Miles & Huberman 1994; Lincoln & Guba, 1985). During the dat a analysis phase of the study the use of membe r checks and data triangulation contribute to internal validity (Miles & Huberman, 1994). The data triangulation process uses data collected from three different sources: in depth interviews, plan content evalu ation, an d a reflexive journal. Table 8 provides an outline of the techniques that will be used in order to ensure the highest quality of data collection and analysis in this research study (Erlandson et al. 1993) Table 8 : Naturalistic Techniques for Ensuring Comprehensive Research Quality Criteria Conventional Term Naturalistic Term Naturalistic Techniques Truth Value Internal Validity Credibility Data triangulation Peer debriefing Member checks Reflexive journal Applicability External Validity Transferability Rich description Purposive sampling Reflexive journal Consistency Reliability Dependability Data triangulation Reflexive journal Neutrality Objectivity Confirmability Research audit trail

PAGE 116

102 C HAPTER IV E VALUATING THE RELATIONSHIPS BETWEEN PARTICIPATION, COMMUNITY CHARACTERISTICS, AND LOCAL HAZARD MITIGATION PLAN QUALITY Introduction Although the literature includes a number of studies about public participation in planning and how it affects planning outc omes, there is little empirical research on the relationship between public participation and local hazard mitigation plan quality. The aim of this study, and the analysis of 34 local hazard mitigation plans in the state of California, is to understand how community characteristics and levels of public participation affect mitigation plan quality outcomes In this chapter, principal component analysis and regression techniques are used to identify key relationships between local community characteristics a nd local plan quality scores, to analyze how community characteristics affect levels of public participation in local hazard mitigation planning, and to evaluate how public participation in planning affects local hazard mitigation plan quality. The followi ng chapter outlines the data, methods and results of the quantitative inquiry into plan quality, public participation and local community characteristics. Methods Plan Quality Index Based on a large and comprehensive body of social vulnerability literature, and on guidance provided by previous hazard mitigation plan quality studies, 30 plan quality

PAGE 117

103 evaluation index. The Plan Quality Index consists of 30 key indicato rs that correspond to each of the three principles of plan quality according to theory. The indicators under each principle are designed to assess the degree to which observed plan content corresponds to the overarching principles of plan quality: Fact Bas e, G oals, and A ctions. Local disaster resilience describes the ability of an individual or community to survive future natural disasters with minimum losses of life and property, as well as the ability to create a greater sense of place among residents, a stronger, more diverse economy, and a more economically integrated and diverse population (Berke & Campanella, 2006; Vale & Campanella, 2005). As a starting point for local resilience building, mitigation planning has great potential to direct hazard avoid ance activity and property protection, to foster environmental preservation and sound structural engineering projects, and to develop local a wareness about hazards (Schwab and Brower, 2008). Hazard mitigation activities that fail to address the inherent s ocial vulnerability of a community are unable to remedy the underlying social factors that contribute to weak economies, poor senses of place and community among residents, and meager community coping capacities. Without an adequate understanding of the so cial components of vulnerability, communities are limited in their ability to envision ways of enhancing their coping capacity and building community resilience. For this reason, it is hypothesized that local hazard mitigation plans of this kind are unable to increase local resilience to disasters because disaster resilience demands a deeper analysis of the risk

PAGE 118

104 equation, one that places a new emphasis on the role of people before, during, and after disasters. If communities aspire to be resilient in the face of disasters, a better understanding of their own social vulnerability must be developed in addition to a strong understanding of the physical hazards to which they are at risk. Until now, the body of liter ature has not assessed hazard mitigation plan quality or planning activities as they relate to the engagement with, and incorporation of, local social vulnerability concerns. Moreover, empirical plan quality studies have not addressed the extent to which i nformation about local social vulnerability is included into local hazard mitigation plans or local decision making. By entering social vulnerability into the plan quality assessment, we are able to observe whether or not local jurisdictions incorporate c ontextual social vulnerability information into their hazard mitigation planning strategies to a greater extent than others. The literature review suggests that there will be a positive relationship between highly participatory approaches to hazard mitigat ion planning and both overall plan quality and Fact Base score s In addition to the 30 indicators of plan quality, this study incorporates 20 social vulnerability variables into the Plan Quality Index that reveal the underlying contextual information used to guide hazard mitigation goal setting and policy making for in each of the 34 sample cities. Because the Plan Quality Index includes social vulnerability variables in the Fact Base component, plans that incorporate information about a social characteristics will attain a higher Fact Base score than those that do not address social dimensions of vulnerability. Table 9 provides a description of

PAGE 119

105 each principle of plan quality, the indicators for their corresponding content, and the scoring criteria used for each indicator. Table 9 : Plan Quality Index Plan quality principle Indicators Scoring criteria Fact Base (jurisdiction specific social vulnerability) Minority populations Elderly populations (over 65) Youth (14 and under) People living in group quarters Female headed households Linguistically isolated populations Adults with no HS diploma Preschool enrollment Health services Disabled population (physically and /or mentally) Households below poverty Households wit h public assistance income Rental housing Private and public social service agencies Support services to individuals and families that are relevant to disaster response and recovery Voting levels Housing units with no vehicles available Neighborhood organi zations Religious organizations Special interest groups 0 = no information on the item 1 = limited information; mentioned, but not detailed 2 = detailed information Fact Base (physical vulnerability by hazard) Description of hazard location Description of magnitude of hazard Current number of people exposed Number and total value of private infrastructure exposed Number and total value of public infrastructure exposed Number of different types of critical facilities exposed Loss estimations to private st ructures Loss estimations to public structures Emergency shelter demand and capacity data Evacuation clearance time data

PAGE 120

106 t.) Plan quality principle Indicators Scoring criteria Goals (economic impacts) Any goal to reduce property loss Any goal to minimize fiscal impacts of natural disaster Any goal to distribute hazard management costs equitably 0 = not mentioned in the plan 1 = mentioned in the plan 2 = mentioned and detailed in the plan Goals (physical impacts) Any goal to reduce damage to public property Any goal to reduce hazard impacts that also achieves preservation of natural resources Any goal to reduce hazard impacts that also achieves preservation of open space and recreational areas Any goal to reduce hazard impacts that also ac hieves maintenance of good water quality Goals (public interest) Any goal to protect population safety Any goal to promote a hazards awareness program Actions (general policy) Any action/policy to discourage development in hazardous areas Any action/policy to connect LHMP to general plan safety element 0 = not mentioned in the plan 1 = suggested in the plan 2 = mandatory in the plan Actions ( awareness ) Any action/policy to promote educational awareness Any action/policy for real estate hazard disclosure Any action/policy for disaster warning and response programs Any action/policy for signage indicating hazardous areas Any action/policy for participation in flood insurance programs Any action/policy for technical assistance to developers or pro perty owners for mitigation activity

PAGE 121

107 Plan quality principle Indicators Scoring criteria Actions (general policy) Any action/policy to discourage development in hazardous areas Any action/policy to connect LHMP to general plan safety element 0 = not mentioned in the plan 1 = suggested in the plan 2 = mandatory in the plan Actions (awareness) Any action/policy to promote educational awareness Any action/policy for real estate hazard disclosure Any action/policy for disaster warning and response programs Any action/policy for signage indicating hazardous areas Any action/policy for participation in flood insurance programs Any action/policy for technical assistance to developers or property owners for mitigation activity Actions (regulatory) Any regulatory action/policy involving permits for land use Any regulatory action/policy involving the transfer of development rights Any regulatory action/policy involving cluster development Any regulatory action/policy involving setbacks An y regulatory action/policy involving site plan review Any regulatory action/policy involving special study/impact assessment of development in hazardous areas Any regulatory action/policy involving building standards Actions (incentives) Any incentives i nvolving the retrofitting of private structures Any incentives involving land and property acquisition Any incentives involving tax abatement for mitigation Any density bonus incentives Any incentives involving low interest loans for mitigation activity Actions (public facilities and infrastructure) Any actions/policy involving capital improvement Any actions/policy involving retrofitting public structures Any actions/policy involving critical facilities Actions (control of hazards) Any actions/policy i nvolving storm water management Any actions/policy involving maintenance of structures

PAGE 122

108 Guided by the quantitative Plan Quality Index above, the content of 34 plans was analyzed in terms of their Fact Base, G oals and A ctions. Each plan quality indicator is measured on a 0 2 ordinal scale. The scoring criterion for the Fact Base component ad heres to the following guidelines: 0 = no information on the item; 1 = limited information on the item; mentioned, but not detailed; and 2 = detailed information on the item. The scoring criteria for both the goals and actions components adhere to the foll owing guidelines: 0 = not mentioned in the plan; 1 = mentioned in the plan; and 2 = mentioned and detailed in the plan. Equal weights were assigned to the variables and components in the Plan Quality Index. This eliminates the need to make value judgments about which indicator or plan Plan quality principle Indicators Scoring criteria Actions (recovery) Any actions/policy involving land use change Any actions/policy involving building design change Any actions/policy involving a moratorium on development Any actions/policy involving the development of a recovery organization Any actions/policy involving private acquisition of property Any actions/policy involving financing for recovery 0 = not mentioned in the plan 1 = suggested in the plan 2 = mandatory in the plan Actions (emergency preparedness) Any actions/policy involving evacuation Any actions/policy involving sheltering Any actions/policy requiring emergency plans

PAGE 123

109 component deserves more emphasis in determining plan quality than others. Moreover, equal weighting contributes to the maintenance of consi stency in statistical results. A few components of the Plan Quality Index have more than one indicator per variable. For example, under the Fact Base component, critical facilities can be either recorded in a list, mapped, or both. In these cases, scores were assessed using an index built by taking the total indicator score and dividing i t by the total number of sub indicators ( for example, if component received a 0 for mapping and a 2 for listing, the overall score for the component will be 2). This method allows for scores to remain on a 0 to 2 scale and favors plans that support the ir decisions with multiple decision making tools. Three plan quality component indices were created for each LHMP and then summed to produce a measure of overall plan quality. The indices were developed using a four step process. First, the scores for eac h variable were summed within each individual plan component. Second, the sums were divided by the total possible score for each plan component. Third, in order to represent each of the plan quality component scores on a 0 to 10 scale, the fractional p lan component scores were each multiplied by 10. As a final step, the scores of each of the three quality components were summed to produce a total plan quality score. The maximum plan qual ity score for each LHMP is 30. The Plan Quality Index does not eva luate the implementation outcomes of plans, reaching identified goals or reducing local vulnerability. However, empirical research suggests that these successes are positiv ely correlated with high plan quality as measured

PAGE 124

110 by the three indicators assessed by the index (Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Berke et al., 1998; Godschalk et al., 1999). Methodology for Assessing Models of Community P articipation In add ition to the plan quality evaluations, this study includes an assessment of community participation during local hazard mitigation planning. Based on the (1969) ladder of cit izen participation, specific collaborative planning approaches used by each jurisdiction have been identified and a planning model assigned to each plan ranging from M1 to M4 b ased on their characteristics. Table 10 describes of each of the collaborative planning approaches that were identified during the plan evaluation phase of this study, the situation of each planning community participation model assigned to each combinatio n of planni ng approaches.

PAGE 125

111 Table 10 : Evaluation Criteria for Models of Community Participation Model of Community Participation in Hazard Mitigation Planning Hierarchy of Public Power Collaborative Planning Approach M1 Manipulation N/A (no public engagement or outreach) Therapy M2 Informing Public service announcements Site visits Newspapers Leaflets, brochures, bulletins Unstaffed exhibits/displays Advertising Consultation Teleconferencing Surveys, interviews and questionnaires Deliberative polls Workshops Placation Staffed exhibits/displays Staffed telephone hotlines Website Public meeting Open house (including virtual) M3 Partnership Charrettes Visioning (including virtual) Consensus conferences Delegated Power Community advisory committees/liaison groups Community delegated juries M4 Community Control Community based data collection Community power in decision making and priority setting process Based on the criteria outlined above, LHMPs that recorded the use of no public participation, or employed strategies that were designed to be manipulative or strictly for therapeutic purposes, have been classified as examples of M1 community participation models. Jurisdictions that carried out informational, consultative participation strategies, including newspaper advertisements, questionnaires and websites were identified as M2 models. Local hazard mitigation planning documents that recorded that they cr eated

PAGE 126

112 community advisory committees or other forms of delegated decision making groups have been classified as M3 models. M4 models are representative of the highest levels of public power: community control. In these cases, the jurisdiction specific data collection, priority setting and decision making are processes that are fundamentally driven by public stakeholders and local residents. LHMP span more than one model of comm unity participation. For example, the city of Roseville included questionnaires about perceived hazard risks in resident energy bills (consultation). However, in addition to the surveys they established a hazard mitigation planning steering committee to wh ich members of the public were delegated (delegated power). In cases like this, a public participation model was assigned based on the collaborative planning approach that afforded citizens the highest level of power. ed the M3 model This evaluation method was chosen because it rewards jurisdictions for undertaking multiple approaches to collaboration and outreach. Moreover, it acknowledges previous empirical research that has found that the total number of community engagement techniques used during planning has a significant positive correlation with the total number of stakeholders involved in planning processes (Brody et al., 2003 ; Stevens, Berke & Song, 2010). Before the 34 selected local hazard mitigation plans w ere evaluated, the feasibility of the Plan Quality Index was tested and community participation model evaluation ding to the Plan Quality

PAGE 127

113 Index. The pilot test was primarily concerned with solidifying the scoring criteria for each variable and careful attention was given to the specific scoring choices that were made based on the wording of actions and goals. By usin g the index to determine the produce the plan, the scoring procedures for the index were refined so as to validate the consistency of results. Data The following section prese nts the data used for the quantitative portion of this study including plan quality data, data showing levels of public participation in planning and local community characteristic data. Plan Quality and Levels of Public Participation The Plan Quality Inde x discussed in Chapter 3 was used as a tool for assessing local hazard mitigation plans from 34 cities in the California bay and delta region. Each plan was individually scored, starting with Fact Base variables, followed by Goals variables, and concluding with Actions variables. The scores of the three components have been summed to provide an overall plan quality score for each of the 34 plans. At the conclusion of each plan content analysis, community participation models were identified for each city T able 11 summarizes the plan quality scores of each of the 34 cities in the sample frame. The table also provides results of the participation evaluation and information about the Planning Scale of each local hazard mitigation plan.

PAGE 128

114 Table 11 Cities Ranked by Plan Quality Score City County Plan Scale Participation Model Fact Base Score Goals Score Actions Score Plan Quality Score Roseville Placer SJ M3 4.31 5.55 3.37 13.25 Napa Napa SJ M3 3.33 5.55 4.19 13.08 Elk Grove Sacramento MJ M3 2.00 6.11 2.29 10.41 Redwood City San Mateo SJ M2 1.70 5.55 2.43 9.69 Berkeley Alameda SJ M2 1.62 4.44 2.97 9.04 Mill Valley Marin SJ M2 1.72 5.00 2.29 9.02 Woodland Yolo MJ M2 2.16 5.55 1.21 8.93 Windsor Sonoma MJ M2 0.45 5.55 2.29 8.31 Citrus Heights Sacramento MJ M2 2.00 4.44 1.75 8.20 Lincoln Placer MJ M2 0.94 5.55 1.62 8.12 Larkspur Marin SJ M2 1.37 3.88 2.70 7.96 Davis Yolo MJ M2 1.16 5.55 1.21 7.93 Rocklin Placer MJ M2 0.94 5.55 0.94 7.44 American Canyon Napa MJ M2 0.88 5.55 0.94 7.39 Folsom Sacramento MJ M3 1.55 3.33 2.43 7.32 West Sacramento Yolo MJ M2 0.16 6.66 0.40 7.23 Auburn Placer MJ M2 0.66 5.55 0.94 7.16 Orinda Contra Costa MJ M2 1.00 2.77 3.24 7.02 Santa Clara Santa Clara MJ M2 1.12 2.77 2.29 6.20 Half Moon Bay San Mateo MJ M2 0.66 2.77 2.56 6.01 Vacaville Solano MJ M2 0.66 2.77 2.56 6.01 Sunnyvale Santa Clara MJ M2 1.16 2.77 2.02 5.97

PAGE 129

115 Table 11 City County Plan Scale Participation Model Fact Base Score Goals Score Actions Score Plan Quality Score Santa Rosa Sonoma MJ M2 0.83 2.77 2.16 5.77 San Anselmo Marin MJ M2 0.50 2.77 2.29 5.57 Fairfield Solano MJ M2 0.69 2.77 2.02 5.44 Gilroy Santa Clara MJ M2 1.16 2.77 1.48 5.43 Campbell Santa Clara MJ M2 1.08 2.77 1.48 5.34 Burlingame San Mateo MJ M2 0.50 2.77 1.83 5.11 Concord Contra Costa MJ M2 0.58 2.77 1.62 4.98 Saratoga Santa Clara MJ M2 1.16 2.77 0.81 4.75 Hillsborough San Mateo MJ M2 0.50 2.77 1.35 4.62 Galt Sacramento MJ M3 0.83 2.77 0.81 4.42 Rancho Cordova Sacramento MJ M3 0.66 1.66 0.27 2.60 Morgan Hill Santa Clara MJ M1 0.66 0.00 0.13 0.80 highest overall Plan Quality Score (13.25) while the City of Morgan Hill has the lowest composite Plan Quality Score (0.802). Out of a total possible score of 10 points, the F act Base Scores for the thirty four local hazard mitigation plans range from 0.167 (the City of West Sacramento) to 4.315 (the City of Roseville). The Goals Scores range from 0 on the low end (the City of Morgan Hill articulated no goals in their local haz ard mitigation plan) and 6.67 on the high end (the City of West Sacramento). The Action Scores range

PAGE 130

116 from 0.14 (the City of Morgan Hill) to 4.19 (the City of Napa). Below, Table 12 summarizes the descriptive statistics from the plan content analysis. Table 12 : Descriptive Statistics Mean Standard Deviation Analysis N Fact Base Score 1.20 .84 34 Goals Score 3.79 1.58 34 Actions Score 1.95 1.12 34 Plan Quality Score 6.76 2.59 34 Research standards of objectivity and dependability were paramount during project design and data collection in order to ensure the validity of the plan quality analysis tool and consistency of results. Although we strive to remain objective, it is inevita ble that researchers own biases are introduced into the scoring and content analysis of each plan. For this reason, a pilot test of the Plan Quality Index was conducted on the itional researchers were not enlisted to assist with the plan quality assessment in order to avoid introducing additional bias into the analysis (Miles & Huberman, 1994; Lincoln & Guba, 1985). There are a few limitations to the Plan Quality data in this study. One limitation is the small sample size. Because there are only 34 plans included in the study the findings cannot be generalized to the broader community based on the results of this study alone. In addition, out of the sample of thirty four plans there are only seven plans that qualify as having highly participatory planning models. In a perfect world, a large number of cases would have been available to put through the principal components and regression analyses carried out in this phase of the s tudy.

PAGE 131

117 Despite these limitations, the results of the following quantitative analyses contribute to new understandings of the local hazard mitigation planning process, which is significant considering that the sample region is characterized as an example of best practices in planning and participation at both the state and local jurisdiction scale. Moreover, despite the smaller than ideal sample size, the results of the quantitative analyses provoke new questions about the relationship between community char acteristics, plan quality, and public participation in hazard mitigation planning. Community Characteristic Data In addition to the plan quality assessment, data for twenty contextual variables were collected across three core categories of hazard vulnerab ility: socio economics, hazard exposure and social vulnerability. Nineteen of those twenty community characteristics variables were selected from a comprehensive set of social vulnerability indicators previously identified by Cutter et al. (2003). In their study, Social Vulnerability to Environmental Hazards, Cutter et al (2003) identifies a series of composite factors that differentiate U.S. communities according to their relative levels of social vulnerability. Not only do these factors illustrate geogra phic variability in social vulnerability, but they also illustrate variations in underlying causes of vulnerability. Although Cutter et al. (2003) collected and analyzed social vulnerability factors at the county scale, the core indicators are still releva nt and applicable to vulnerability science with the broader hazards and disaster literature.

PAGE 132

118 Table 13 outlines the community characteristic categories and the twenty associated var iables selected for this study. Table 13 : Community Characteristic Variables Community Characteristics Variables Socio economics Population size Race: White Race: African American or African American Race: American Indian or Alaskan Native Race: Asian Race: Native Hawaiian or Pacific Islander Race: Hispanic or Latino (of any race) Poverty (% total population below poverty) Unemployment Median household income Median home value Hazard exposure Total county hazard losses (1990 2010) Multi hazard exposure Social vulnerability Female headed household % population living in group quarters % Population > 65 % Population < 14 Linguistically isolated households Educational attainment Renter occupied housing The selected socio economics variables paint a comprehensive picture of the social and economic character of each of the 34 cities included in the sample frame. The 2009 American Commun ity Survey (5 year estimates). This dataset was selected for a number of reasons. First, the ACS 5 year estimates are based on data collected between January 2005 and December 2009 and represent average community characteristics over the five year

PAGE 133

119 period o f time. This range of data is appropriate for this study because the median adoption year of the sample of thirty four local hazard mitigation plans is 2009. In addition, the 2005 2009 ACS 5 year estimates are published for small geographic areas and are available for areas with populations less than 20,000 people. The sample of cities in this study contains communities with populations between 10,000 and 200,000. Finally, although the estimates are less current than the 1 year and 3 year estimates, the 5 year estimates are b ased on a larger sample size. Although none of the local hazard mitigation plans created in the sample frame were written in direct response to a disaster, hazard exposure has been shown to be an important driver of community particip ation in planning. Catastrophes and disasters have with respect to local priorities, policy and practice related to emergency management. These changes often include gre ater levels of public support for hazard mitigation and preparedness activity (Godschalk et al., 1999; Smith, 2010; Berke & Campanella, 2006; Burby, 2008). Additionally, the memories of repeated hazard events can prompt some communities to prioritize mitig ation planning to a greater extent than communities that may have high hazard risks, but little to no history of loss. Data about previous hazard losses and exposure to hazards is included in the set of community characteristic variables in order to accou nt for the effect of previous hazards on communities within the sample frame. The local hazard exposure level of the

PAGE 134

120 thirty four sample cities is determined by coupling two distinct variables: total monetary losses from hazards by county between 1990 and 2 010 4 and multi hazard exposure The total monetary loss data can be found on the SHELDUS database. Values represent county scale crop damage plus property damage in U.S. 20 09 inflation adjusted dollars. The multi hazard exposure data was collected from Ca Hazard Mitigation Plan (SHMP). In 2007, a GIS based risk exposure analysis was conducted by California Polytechnic State University, San Luis Obispo, for the California forts. Multi hazard exposure data was collected for the three primary hazards that threaten California floods, earthquakes, and wildfires. Based on the risk exposure data, a series of GIS an (SHMP) to display unties to these three hazards. For the purpose of comparability, the relative hazard exposure of each county is ranked on a scale from high, to medium, to low. The exposure maps use three colors to sym bolize the relative vulnerability of California cou nties to hazards (see Figure 9 ). The map has been used to guide planning decisions during California Stat e mitigation planning efforts. 5 4 Source: 5 Source: State of California 2010 Hazard Mitigation Plan

PAGE 135

121 Figure 9 : California State Hazard Exposure Map Information from 2010 California State Hazard Mitigation Plan multi hazard exposure maps has been incorporated as one of the local community characteristic variab les. Table 14 outlines the descriptive statistics associated with hazard exposure data in Northern California Bay Delta region. Table 14 : Descriptive Statistics for Hazard Exposure Data Mean Standard Deviation Analysis N Multi hazard exposure 1.79 .41 34 Hazard Losses (U.S. $) $269 million $464 million 34

PAGE 136

122 Based on data from the 2010 California State Hazard Mitigation Plan, the majority of the cities in the sample frame lie in counties characterized by medium to low levels of multi hazard exposure. Although the communities are currently planning for a wide variety of hazard risks, there is little variation between communities in terms of their relative levels of hazard exposure. All of the cities in the sample frame are characterized by med ium levels of hazard exposure except for cities located in Placer and Yolo counties (Auburn, Davis, Lincoln, Rocklin, West Sacramento, Woodland and Roseville), which have low levels of hazard expos ure. Table 15 summarizes the exposure and hazard loss data by city and county. Table 15 : Summary of Multi Hazard Exposure and Losses by City City County Exposure Total Hazard Losses by County (1990 2010) Morgan Hill city, California Santa Clara 2 < $100 million American Canyon city, California Napa County 2 $100 $500 million Auburn city, California Placer County 1 $100 $500 million Berkeley city, California Alameda 2 > $1 billion Burlingame city, California San Mateo 2 $100 $500 million

PAGE 137

123 Table 15 City County Exposure Total Hazard Losses by County (1990 2010) Campbell city, California Santa Clara 2 < $100 million Concord city, California Conta Costa 2 $100 $500 million Davis city, California Yolo County 1 < $100 million Fairfield city, California Solano County 2 $100 $500 million Gilroy city, California Santa Clara 2 < $100 million Half Moon Bay city, California San Mateo 2 $100 $500 million Hillsborough town, California San Mateo 2 $100 $500 million Larkspur city, California Marin County 2 $100 $500 million Lincoln city, California Placer County 1 $100 $500 million Mill Valley city, California Marin County 2 $100 $500 million Napa city, California Napa County 2 $100 $500 million Orinda city, California Contra Costa 2 $100 $500 million Redwood City city, California San Mateo 2 $100 $500 million Rocklin city, California Placer County 1 $100 $500 million San Anselmo town, California Marin County 2 $100 $500 million Santa Clara city, California Santa Clara 2 < $100 million Santa Rosa city, California Sonoma County 2 $100 $500 million Saratoga city, California Santa Clara 2 < $100 million Sunnyvale city, California Santa Clara 2 < $100 million Vacaville city, California Solano County 2 $100 $500 million

PAGE 138

124 Table 15 ( City County Exposure Total Hazard Losses by County (1990 2010) West Sacramento city, California Yolo County 1 < $100 million Woodland city, California Yolo County 1 < $100 million Citrus Heights city, California Sacramento County 2 $100 $500 million Elk Grove city, California Sacramento County 2 $100 $500 million Folsom city, California Sacramento County 2 $100 $500 million Galt city, California Sacramento County 2 $100 $500 million Rancho Cordova city, California Sacramento County 2 $100 $500 million Roseville city, California Placer County 1 $100 $500 million Windsor town, California Sonoma County 2 $100 $500 million Because there was very little variation between communities in terms of their previous hazard losses and exposure, the data listed in the table above was removed from the principal component analysis and the regression analysis. The data listed Below, Tabl e 16 summarizes the eighteen community characteristic variables that were compiled for quantitative data analysis.

PAGE 139

125 Table 16 : Summary of Community Characteristic Variables Name Description Type Coding MEDIAN HOUSEHOLD INCOME Median household income in past 12 months* Continuous None (value in 2009 inflation adjusted dollars) POPULATION SIZE Total population size of incorporated place (city or town)* Continuous None POVERTY % of total population with income in the last 12 months below poverty level* Continuous Percentage (0.00 to 1.00) MEDIAN HOME VALUE Median home value* Continuous None (value in dollars) POPULATION IN GROUP QUARTERS % of total population living in a group living arrangement, managed/owned by an entity or organization providing housing and/or services for residents* Continuous Percentage (0.00 to 1.00) HOUSING TENURE % of total housing that is renter occupied* Continuous Percentage (0.00 to 1.00) Name Description Type Coding

PAGE 140

126 Name Description Type Coding 65 YEARS & OLDER % of total population that is 65 years old or older* Continuous Percentage (0.00 to 1.00) 14 YEARS & YOUNGER % of total population that is 14 years old or younger* Continuous Percentage (0.00 to 1.00) RACE: WHITE % of total population that is White* Continuous Percentage (0.00 to 1.00) RACE: AFRICAN AMERICAN OR AFRICAN AMERICAN % of total population that is African American or African American* Continuous Percentage (0.00 to 1.00) RACE: ASIAN % of t otal population that is Asian* Continuous Percentage (0.00 to 1.00) RACE: HISPANIC OR LATINO % of total population that is Hispanic or Latino (of any race)* Continuous Percentage (0.00 to 1.00) NATIVE HAWAIIAN AND OTHER PACIFIC ISLANDER % of total population that is Native Hawaiian or other Pacific Islander Continuous Percentage (0.00 to 1.00) AMERICAN INDIAN OR ALASKAN NATIVE % of total population that is American Indian or Alaskan Native Continuous Percentage (0.00 to 1.00) FEMALE HOUSEHOLDERS % of total population made up of females maintaining a household with no husband of the householder present* Continuous Percentage (0.00 to 1.00) UNEMPLOYMENT % unemployment rate for population 16 years and older* Continuous Percentage (0.00 to 1.00) LOW EDUCATIONAL ATTAINMENT % of total population over 25 years old with less than a high school diploma* Continuous Percentage (0.00 to 1.00) LINGUISTIC ISOLATION % of total households that are linguistically isolated (aggregate of Spanish, indo euro, Asian and pacific, and other)* Continuous Percentage (0.00 to 1.00) *Tract level data from the U.S. Census Bureau, American Community Survey 5 year estimates.

PAGE 141

127 Data Reduction Using Principal Components Analysis Research question number two asks: How do community characteristics and levels of public participation affect mitigation plan quality outcomes? This question drives the quantitative analysis portion of this study, which is designed to ad dress the following three aims: Aim 1: To identify key relati onships between local community characteristics and local plan quality scores; Aim 2: To analyze how community characteristics affect levels of public participation in local hazard mitigation planning; and Aim 3: To evaluate how public participation in pl anning affects local hazard mitigation plan quality During the process of identifying the key relationships outlined in the three aims above, we develop a stronger understanding of how community characteristics and levels of public participation affect mi tigation plan quality outcomes. The following sections describe the data normalization techniques used to prepare the community characteristic data for the principal components analysis and describes the set of factors that were extracted during the data r eduction process. Principal Components Analysis Principal components analysis (PCA) is a method used for estimating relationships among sets of variables. Although PCA does not create new information, it is useful because it organizes and summarizes inform ation in ways that uncover unseen, latent variables that are reflected in a set of observabl e variables.

PAGE 142

128 PCA is a statistical technique that is used to replace a large set of variables (such as the community characteristic variables collected for this stu dy) by a smaller set of variables which is the best representation of the larger set. The technique allows for a set of correlated explanatory variables to be transformed into a new set of uncorrelated variables called principal components (or factors). Th e principal components are all linear combinations of the original correlated variables and can be used in p lace of the original variables. Table 17 summarizes the descriptive statistics for the set of eighteen community characteristic variables included i n the PCA. Table 17 : Descriptive Statistics Mean Standard Deviation Analysis N POPULATION SIZE 58671.47 41360.68 34 UNEMPLOYMENT 6.58 1.60 34 MEDIAN HOUSEHOLD INCOME (U.S. $) $81,718 $31,365 34 POVERTY 8.74 5.15 34 LOW EDUCATIONAL ATTAINMENT 11.31 6.49 34 LINGUISTIC ISOLATION 6.46 4.07 34 14 YEARS & YOUNGER 20.09 3.25 34 65 YEARS & OLDER 13.06 4.27 34 RACE: WHITE 67.41 13.81 34 RACE: AFRICAN AMERICAN OR AFRICAN AMERICAN 3.48 3.80 34 RACE: ASIAN 13.42 11.43 34 RACE: HISPANIC OR LATINO 21.49 13.51 34 RACE: AMERICAN INDIAN OR ALASKAN NATIVE 0.37 0.50 34

PAGE 143

129 Mean Standard Deviation Analysis N RACE: NATIVE HAWAIIAN/OTHER PACIFIC ISLANDER 0.43 0.32 34 POPULATION IN GROUP QUARTERS 1.83 2.69 34 FEMALE HOUSEHOLDERS 10.64 3.01 34 HOUSING TENURE 36.87 13.33 34 MEDIAN HOME VALUE $611,876 $230,603 34 Collinearity between variables is an important concern when incorporating predictor variables into a multi ple regression analysis. Collinearity is defined as correlation among the predictor variables and it often occurs with regional data where different variables (e.g. unemployment, poverty, median income) tend to rise and fall together. Collinearity is a concern when performing a multiple regre ssion analysis because the redundancy of correlation entangles the effects of the predictors which complicates the interpretation of results. The results of a multiple regression analysis can be problematic if the independent variables are highly correlate d. The PCA technique is the most practical technique for solving this problem. In order to regulate collinearity between the community characteristic variables a number of variables were tested for collinearity based on theory. Because the economic variabl es unemployment, poverty, and median household income are likely to predict one another, the variables were eliminated from the PCA. The total variance (or factor groupings) from the PCA run without the economic variables did not

PAGE 144

130 differ from the results of the run with the variables included. Ultimately, none of the three economic variables were eliminated due to their negligible effect on the results of the PCA. Eighteen of the twenty originally identified community characteristic variables were included in the PCA analysis. Table 18 summarizes the variables included in the PCA. Table 18 : Summary of PCA Variables Name Description Type Coding MEDIAN HOUSEHOLD INCOME Median household income in past 12 months* Continuous None (value in 2009 inflation adjusted dollars) POPULATION SIZE Total population size of incorporated place (city or town)* Continuous None POVERTY % of total population with income in the last 12 months below poverty level* Continuous Percentage (0.00 to 1.00) MEDIAN HOME VALUE Median home value* Continuous None (value in dollars) POPULATION IN GROUP QUARTERS % of total population living in a group living arrangement, managed/owned by an entity or organization providing housing and/or services for residents* C ontinuous Percentage (0.00 to 1.00) HOUSING TENURE % of total housing that is renter occupied* Continuous Percentage (0.00 to 1.00)

PAGE 145

131 Name Description Type Coding 65 YEARS & OLDER % of total population that is 65 years old or older* Continuous Percentage (0.00 to 1.00) 14 YEARS & YOUNGER % of total population that is 14 years old or younger* Continuous Percentage (0.00 to 1.00) RACE: WHITE % of total population that is White* Continuous Percentage (0.00 to 1.00) RACE: AFRICAN AMERICAN % of total population that is African American* Continuous Percentage (0.00 to 1.00) RACE: ASIAN % of total population that is Asian* Continuous Percentage (0.00 to 1.00) RACE: HISPANIC OR LATINO % of total population that is Hispanic or Latino (of any race)* Continuous Percentage (0.00 to 1.00) NATIVE HAWAIIAN AND OTHER PACIFIC ISLANDER % of total population that is Native Hawaiian or other Pacific Islander Continuous Percentage (0.00 to 1.00) AMERICAN INDIAN OR ALASKAN NATI VE % of total population that is American Indian or Alaskan Native Continuous Percentage (0.00 to 1.00) FEMALE HOUSEHOLDERS % of total population made up of females maintaining a household with no husband of the householder present* Continuous Percentage (0.00 to 1.00) UNEMPLOYMENT % unemployment rate for population 16 years and older* Continuous Percentage (0.00 to 1.00) LOW EDUCATIONAL ATTAINMENT % of total population over 25 years old with less than a high school diploma* Continuous Percentage (0.00 to 1.00) LINGUISTIC ISOLATION % of total households that are linguistically isolated (aggregate of Spanish, indo euro, Asian and pacific, and other)* Continuous Percentage (0.00 to 1.00)

PAGE 146

132 As a final data management step, Z score transformations were created to normalize the community characteristic variables. By creating Z scores, all of the variables have a mean of zero an d a standard deviation of one. Before running the Principal Component extraction procedure, the SAS software was asked to re tain as many factors as there were eigenvalues greater than 1. To begin, a PCA model was run with no rotation. An additional PCA was run with a Varimax rotation. The results of the two models were quite similar with respect to the number of factors extract ed and the factor loading scores ( Table 19 ). Both runs yielded four factors with a total variance of 78%. Table 19 : P rincipal Component Analysis R uns Run Method Rotation # of Factors Total % Variance 1 Principal Components Analysis None 4 78% 2 Principal Components Analysis Varimax 4 78% Although the rotated and non rotated solutions yield similar results, rotation makes the factors as distinctive as possible (Kachigan, 1991; Pedhazur & Schmelkin, 1991; Gorsuch, 1983; Cattell 1978). Therefore, the Varimax rotation is preferable to the un rotated solution. Ultimately, the Principal components analysis with the Varimax rotation was selected for a more focused analysis. Figure 1 0 shows the scree plot of eigenvalues for the Varim ax rotated PCA model.

PAGE 147

133 Figure 10 : Scree Plot of Eigenvalues of Varimax Rotation The scree plot above shows the percentage of total variance accounted for by each of the extracted factors. The concept of the scree plot is that factors located along the tail end of the curve represent random error variance and that we should select the factor solution just prior to the leveling of the curve. This criterion suggests a 4 factor solution for the PCA analysis, reinforcing the decision based on the eigenvalues. Table 20 outlines the eigenvalues and percentage of variance for the Varima x rota ted solution to the PCA.

PAGE 148

134 Table 20 : Total variance explained Component Varimax Rotated Sums of Squared Loadings Eigenvalue % of Variance Cumulative % 1 7.264 40.36 40.36 2 2.881 16.01 56.36 3 2.431 13.50 69.87 4 1.559 8.66 78.53 Similar to the beta values in regression, the factors in the principal components analysis show individual relationships. The four extracted factors are all linear combinations of the original correlated variables and are derived in decreasing orde r of importance. In other words, the first factor accounts for as much as possible of the v ariation on the original data. The Varimax rotated component matrix (Table 21 ) summarizes the factor loadings for each of the 18 original variables. The factor load ing values represent the correlation coefficients between the variables and the four new factors. A lower bound of 0.06 was selected for meaningful loadings and the moderate to high factor loadings are starred in the table below. The values printed in the component matrix are multiplied by 100 and rounded to the nearest integer. None of the variables had high loa dings on more than one factor.

PAGE 149

135 Table 21 : Varimax Rotated Component Matrix Factor 1 Factor 2 Factor 3 Factor 4 Low Educational Attainment Z 91* 10 1 13 Median Home Value Z 65* 57 9 18 Race: American Indian/Alaskan Native Z 86* 7 2 21 Race: Hispanic Z 91* 7 5 18 Unemployment Z 65* 35 24 34 Female Householders Z 77* 51 4 6 Median Household Income Z 61* 32 58 18 Race: African American Z 4 91* 9 15 Race: Native Hawaiian/Other Pac Islander Z 34 65* 9 45 Poverty Z 43 13 70* 11 Age 14 and Younger Z 59 12 73* 9 Renters Z 14 6 89* 12 Race: Asian Z 43 15 6 82* Race: White Z 17 41 0 85* Linguistically Isolated Z 53 17 27 66* Population Size Z 13 51 37 30 Age 65 and Older Z 53 49 25 28 Population Living in Group Quarters Z 24 45 53 18 After running the PCA, the results of the component matrix were analyzed by studying the high loading variables of each factor. Table 22 summarizes the results of the PCA including the descriptive names for each of the extracted factors and the highest loading variables on each. The variables listed in parenthesis load negatively.

PAGE 150

136 Table 22 : Summary of PCA results Factor names High loading variables Factor loadings Factor 1: At Risk Positive Loadings Race: American Indian or Alaskan Native 86 Female Householders 77 Unemployment 65 Race: Hispanic 91 Low Educational Attainment 91 Negative Loadings Median home value 65 Median household income 61 Factor 2: Racial Disparity Positive Loadings Race: African American 91

PAGE 151

137 Table Factor names High loading variables Factor loadings Factor 3: College Towns Race: Native Hawaiian or Other Pacific Islander 65 Positive Loadings Poverty 70 Renter 89 Negative Loadings 14 years & younger 73 Factor 4: Cultural Disparity Positive Loadings Race: Asian 82 Linguistically Isolated 66 Negative Loadings Race: White 85 Table 23 outlines the factor loading for each of the thirty four cities in the sample frame. The bold numbers highlight cities that load higher than the rest on one of the four extracted factors.

PAGE 152

138 Table 23 : Sample City Factor Loadings Observation City Factor 1 Factor 2 Factor 3 Factor 4 1 City of American Canyon 0.27603 1.19888 1.53134 1.23402 2 City of Auburn 0.22254 0.45251 0.64465 1.77974 3 City of Berkeley 1.53506 1.22110 2.83620 0.22584 4 City of Burlingame 0.75766 0.70351 0.49449 0.61750 5 City of Campbell 0.20891 0.35714 0.51209 0.25630 6 City of Citrus Heights 0.44362 0.49049 0.42717 1.00868 7 City of Concord 0.64010 0.35267 0.55367 1.05090 8 City of Davis 0.57040 0.34678 2.45301 0.34485 9 City of Elk Grove 0.16745 2.40299 1.30852 0.90709 10 City of Fairfield 0.21676 2.29741 0.40921 0.61003 11 City of Folsom 0.82549 1.29715 0.02715 1.07186 12 City of Galt 1.72096 0.17746 0.75576 0.45595 13 City of Gilroy 2.24544 1.09159 0.08676 0.54012 14 City of Half Moon Bay 0.42244 1.57043 0.02673 0.08279 15 Town of Hillsborough 1.56887 0.84337 2.02645 0.88454 16 City of Larkspur 1.14901 1.11867 0.72595 0.82351 17 City of Lincoln 0.12806 0.36875 0.76089 1.08133 18 City of Mill Valley 1.06315 0.73064 0.45353 0.97551 19 City of Morgan Hill 0.54117 0.33695 0.74469 0.06298 20 City of Napa 0.94675 0.97017 0.56656 0.23076 21 City of Orinda 1.53595 0.80740 1.52873 0.46669 22 City of Rancho Cordova 0.93043 1.25344 0.28010 0.32015 23 City of Redwood City 0.45654 0.36345 0.44238 0.91409 24 City of Rocklin 0.18676 0.41185 0.58793 1.16222 25 City of Roseville 0.14951 0.42467 0.28253 0.87358

PAGE 153

139 Observation City Factor 1 Factor 2 Factor 3 Factor 4 26 City of San Anselmo 0.83273 0.44843 0.39068 1.19276 27 City of Santa Clara 0.44654 0.03771 0.86761 2.01538 28 City of Santa Rosa 0.76739 0.13814 0.75361 0.42786 29 City of Saratoga 1.65650 0.87100 0.95123 1.47027 30 City of Sunnyvale 0.49790 0.39645 0.50019 2.34943 31 City of Vacaville 0.26692 1.99715 0.23586 1.07831 32 City of West Sacramento 1.48097 0.37409 0.08576 0.38546 33 Town of Windsor 1.12795 0.60352 0.82468 1.07323 34 City of Woodland 1.55285 0.55877 0.14399 0.00788 What the factor analysis has uncovered are four independent community characteristic dimensions that underlie the eighteen individual input variables. These new, uncorrelated dimensions describe four community factors that are the best representation of th e larger set of community characteristic data. When interpreting the results of a principal components analysis, it can be revealing to interpret the opposite loadings on variables, largely because factor loadings represent the degree to which the variabl e correlates with the factor (Kachigan, 1991). Researchers can gain additional insight into their data by reading negative factor loading results as having a positive loading and a positive result as having a negative loading. For example, when the variabl es are re evaluated as negative factors, communities that have

PAGE 154

140 household income and high median home values and are relatively low risk communities. Evaluating negative factor loadings provides alternative ways of looking at the data that offers additional insight into the contexts of the planning processes w ithin the sample communities. Based on the results of the PCA, the variables with high factor loadings on Factor 1, At Risk, are as follows: Race: American Indian or Alaskan Native; Female Householders; Unemployment; Race: Hispanic; Low Educational Attainment; Median Home Value (negative); and Median Household Income (negative) The City of Gilroy loads high on this factor and is an example of an at risk community. In Factor 1 the variables for median home value and median household income both load negatively. A reinterpretation of the negative factor loadings on this factor d escribes affluent communities characterized by high educational attainment, low unemployment levels and lower populations of Hispanic minorities and female householders. The City of Saratoga loads negatively on Factor 1. Race can contribute to social vulne rability to hazards through limited access to resources, cultural differences and the social, economic, and political marginalization that is often associated with racial disparities (Cutter et al. 2003). The Cities of Elk Grove and Fairfield are examples of communities that load high on Factor 2, Racial Disparity. The variables loading high on this factor are Race: African American and Race: Native Hawaiian/Other Pacific Islander to hazards, Race: Afri can American is the racial variable that corresponds to the highest level of vulnerability to hazards. Cutter (2003) finds that African Americans are more likely to live in at risk housing or in multi unit structures, relative to other racial groups.

PAGE 155

141 As wa s explored in Chapter 2, minority communities are often excluded from disaster planning and preparation activity. As a result, planning activities often move forward without consideration of minority culture and life circumstances (Morrow 1999). Factor 2 Americans on local social vulnerability to disasters. The City of Half Moon Bay loads negatively on Factor 2 and is representative of a city characterized by low levels o f Racial Disparity related to hazard vulnerability. The variables loading high on Factor 3, College Towns, describe communities characterized with large residential universities or colleges. The City of Davis and the City of Berkeley load highly on Factor 3 and are home to two large university campuses (the University of California, Berkeley and the University of California Davis, respectively). The negative loading of 14 years and younger on Factor 3 is intuitive, as the majority of college age students i n have not yet had children. The Town of Hillsborough loads negatively on Factor 3. Communities like Davis and Berkeley are places where a large percentage of the residential population is made up of students. Typically, these students rent dorm rooms or t emporary housing, work on campus or in the local community, and earn small salaries. Often, the economies of college towns are closely related to university activity and in many cases the university is one of the largest employers in the community. Housing quality is an important factor to consider when evaluating local vulnerability to hazards. (Morrow 1999; De Souza 2006) Typically, housing quality is closely tied to personal wealth. Generally, people with lower incomes tend to live in poorly construct ed homes, in multifamily housing in densely populated urban areas, or in

PAGE 156

142 mobile homes. Each of these housing types has elevated vulnerability to storms, fire, earthquakes and other hazards. For example, previous studies on the impacts of tornadoes in the C arolinas (Eidson et al. 1990) and Oklahoma (Daley et al. 2005) have associated increased structural damage and injuries to people who live in mobile homes and older housing stock. Additionally, in the context of the 1995 Northridge, California earthquake, Peek Asa et al. (2003) found that people living in multiple unit housing structures are more likely to sustain injuries than those living in single family homes. The variables loading high on Factor 4, Cultural Disparity, are as follows: Race: Asian; Lingu istically Isolated; Race: White (negative) Studies carried out by the Centers for Disease Control and Prevention and the Agency for Toxic Substances and Disease Registry (CDC/ATSDR, 2008) have found that the presence of Asian populations correlates positi vely with hazard vulnerability. They attribute these results from the fact that some hazard prone coastal cities, particularly those of the western United States, have higher percentages of inhabitants of Asian descent. Moreover, their study found that mor e recent arrivals from Asia tend to live in substandard housing stock ( CDC/ATSDR, 2008 ). Additionally, in areas where there are significant populations of partially or fully non English speaking populations, communication barriers contribute significantly to elevated hazard vulnerability (Morrow 1999). The cities of Sunnyvale and Santa Clara (which both load high on Factor 4) are two of the top ten US metropolitan areas with the highest population of Asian

PAGE 157

143 Americans 6 These communities are characterized b y high levels of Cultural Disparity associated with lower relative populations of White residents and high levels of Asians and linguistically isolated individuals. The City of Auburn, a negatively loading city on Factor 4, is characterized by a low Asian population, high percentages of White residents, and little to no residents who identify as limited English speakers. In the following section, the four extracted factors At Risk, Racial Disparity, College Towns, and Cultural Disparity are used in a mu ltiple regression analysis to uncover relationships between community characteristics, plan quality and participation within the study region. The highly loading sample cities (both positive and negative) are used as contextual tools for interpreting the s tatistical results. Multiple Regression Analysis Regression analysis is a useful method for estimating relationships among sets of variables. For the purpose of this study, the multiple regression technique is applied for the purpose of uncovering relatio nships between community characteristics, public participation in planning and hazard mitigation plan quality. Rather than using values of one predictor variable to estimate values on a criterion variable, multiple regression uses values on several predic tor variables. The regression analysis in this study follows a General Linear Model (GLM) approach, which uses to 6 US Census Bureau.

PAGE 158

144 method of least squares to fit general linear models. This approach was selected because of its generality but also for its flexibility. The following statistical analyses were performed using SAS software. To begin, the univariate procedure was applied to test the Plan Quality data for normality. The four tables below summarize the res ults of the SAS normality test. Table 24 : Moments N 34 Sum Weights 34 Mean 6.95991176 Sum Observations 236.637 Std Deviation 2.52934112 Variance 6.39756651 Skewness 0.36760706 Kurtosis 1.30135145 Uncorrected SS 1858.09234 Corrected SS 211.119695 Coeff Variation 36.3415688 Std Error Mean 0.43377842 Table 25 : Basic Statistical Measures Location Variability Mean 6.959912 Std Deviation 2.52934 Median 7.094500 Variance 6.39757 Mode 6.012000 Range 12.44800 Interquartile Range 2.77000

PAGE 159

145 Table 26 : Tests for Location: Mu0 = 0 Test Statistic p Value Student's t t 16.04485 Pr > |t| <.0001 Sign M 17 Pr >= |M| <.0001 Signed Rank S 297.5 Pr >= |S| <.0001 Table 27 : Tests for Normality Test Statistic p Value Shapiro Wilk W 0.959127 Pr < W 0.2293 Kolmogorov Smirnov D 0.099013 Pr > D >0.1500 Cramer von Mises W Sq 0.069021 Pr > W Sq >0.2500 Anderson Darling A Sq 0.511194 Pr > A Sq 0.1912 Because the mean and median are relatively close to one another we assume that the data is normalized. In addition, the normal probability plot below (Figure 11 ) shows normality wi th the smooth 45 degree slope.

PAGE 160

146 Figure 11 : Normal Probability Plot The central limit theorem states that when the number of observations is over N = 31 you can use the theorem to assume normality. 7 Therefore, even though the sample size of N = 34 is small, we can still as sume normality with this g raph. The table below (Table 28 ) outlines the progression of the regression models completed for this study and the dependent and independent variables applied to each model. 7 Rice, John (1995), Mathematical Statistics and Data Analysis (Second Edition), Duxbury Press.

PAGE 161

147 Table 28 : Summary of Regression Models Dependent Variable Independent Variables Model 1 Plan Quality Factors 1 4 Model 2 Fact Base Score Factors 1 4 Model 3 Actions score Factors 1 4 Model 4 Goals score Factors 1 4 Model 5 Participation Factors 1 4 Model 6 Plan Quality Participation Model 7 Fact Base Participation Model 8 Actions score Participation Model 9 Goals score Participation Model 10 Plan Quality Planning Scale The ten regression models outlined above have been designed to address the three aims associated with research question number two. Together, the models accomplish the following aims: 1. To identify key relationships between local community characteristics and local plan quality; 2. To analyze how community characteristics affect public participation during local hazard mitigation planning; and 3. To evaluate how public participation during planning affects local hazard mitigation plan quality

PAGE 162

148 Models 1 4 address Research Aim 1: To identify key relationships between local community characteristics and local plan quality sc ores. Additionally, Models 1 4 take a General Linear Model approach (or GLM) to regression analysis. The GLM procedure uses a method of least squares to fit general linear models. It is a flexible method that allows researchers to test hypotheses suited for regression, ANOVA (analysis of variance), ANCOVA (analysis of covariance) and a number of additional analyses. Additionally, the Type III Sum of Squares option (Type III SS) tests the unique contribution of each independent variable included in the ana lysis. Model 1 uses Plan Quality Score as the dependent variable and Factors 1 4 as the independent variables. When interpreting the results of each regression model, 0.05 was used as the threshold for the P value. Based on this parameter, independent v ariables with P values < 0.05 are deemed significant. The SAS output results for Model 1 are presented below. Table 29 : Model 1 (Plan Quality Score) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 0.34643674 0.34643674 0.05 0.8211 Factor 2 1 0.82008517 0.82008517 0.12 0.7280 Factor 3 1 4.70816110 4.70816110 0.71 0.4070 Factor 4 1 12.35077761 12.35077761 1.86 0.1835 Based on the regression model there is no relationship between Factors 1 4 and local hazard mitigation Plan Quality Scores. As expected, because a number of the

PAGE 163

149 community characteristic variables included in the PCA were accounted for in the plan quality scoring criteria, we see no residual effects of the variables on Plan Qualit y Score. In other words, the community characteristic variables are located on both sides of this regression model so we see no associations in Model 1. To more closely assess key relationships between local community characteristics and Plan Quality, the following three models (Model 2 4) apply the three individual components of Plan Quality as dependent variables and Factors 1 4 as independent variables. In Model 2, the dependent variable is Fact Base Scores and Factors 1 4 are the independents. T he SAS output for Model 2 is presented below. The results indicate that community characteristics do not a ffect Fact Base Score outcomes. Table 30 : Model 2 (Fact Base Score) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 0.19414049 0.19414049 0.28 0.6039 Factor 2 1 0.21081148 0.21081148 0.30 0.5889 Factor 3 1 0.88836123 0.88836123 1.26 0.2712 Factor 4 1 0.00681187 0.00681187 0.01 0.9224 In Model 3, Actions Score is the dependent variable and Factors 1 4 are the independents. The output for Model 3 is presented below. The results of the model

PAGE 164

150 indicate that Factors 1 4 do not affect the Actions Score outcomes of local hazard mitig ation plans in the study area. Table 31 : Model 3 (Actions S core) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 2.52350213 2.52350213 2.96 0.0965 Factor 2 1 0.15686868 0.15686868 0.18 0.6713 Factor 3 1 1.01359851 1.01359851 1.19 0.2850 Factor 4 1 0.56041457 0.56041457 0.66 0.4245 In Model 4, Goals Score is the dependent variable and Factors 1 4 are the independents. The SAS output for Model 4 is shown below. Again, the results of the model indicate that Factors 1 4 do not affect the Goals Score outcomes of local hazard mitigation plans in the study area. The results of Model 4 are intuitive because mitigation planning goals are set and standardized by FEMA. Table 32 : M odel 4 (Goals Score) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 0.83052862 0.83052862 0.31 0.5819 Factor 2 1 0.70435459 0.70435459 0.26 0.6120 Factor 3 1 0.38316457 0.38316457 0.14 0.7080 Factor 4 1 3.85563711 3.85563711 1.44 0.2401

PAGE 165

151 This study addresses Research Aim 2 To analyze how community characteristics affect levels of public participation in local hazard mitigation planning by building a regression model with Participation Score as the dependent variable and Factors 1 4 as the independents. The logistic appro ach was selected for Model 5 because the participation data is expressed as a set of binary variables. The logistic procedure fits linear regression models for binary data by using the maximum likelihood method. The maximum likelihood estimation was carrie d out using the Fisher scoring algorithm. Below, the Analysis of Maximum Likelihood Estimates table lists the calculated parameter estimates of Model 5, their standard errors and the results of the Wald test for individual parameters. Table 33 : Model 5. Analysis of Maximum Likelihood Estimates (Participation Score) Parameter DF Estimate Standard Error Wald Chi Square Pr > ChiSq Intercept 1 3.7943 1.6316 5.4083 0.0200 Factor 1 1 2.1177 1.2223 3.0017 0.0832 Factor 2 1 1.9060 0.9291 4.2081 0.0402 Factor 3 1 1.4841 1.1054 1.8023 0.1794 Factor 4 1 2.3781 1.2288 3.7456 0.0529 The SAS output results identify Factor 2 as a significant predictor of participation score. If we recall, Factor 2 represents a racial dimension of local community

PAGE 166

152 characteristics. Race: African American and Race: Native Hawaiian and Other Pacific Islander have high factor loadings on Factor 2. Models 6 9 address Research Aim 3: To evaluate how public participation in planning affects local hazard mitigation plan quality. In Model 6, Plan Quality is the dependent variable and Participation Score the inde pendent. The table below summarizes the results of the model. Table 34 : Model 6 (Plan Quality Score) Source DF Type III SS Mean Square F Value Pr > F Participation Score 1 6.05084235 6.05084235 0.94 0.3385 The results indicate that there is no association between Plan Quality Scores and Participation Scores recorded from the local hazard mitigation planning process. The dependent variable in Model 7 is Fact Base Score. Factors 1 4 and Participation are th e independents. The results summarized in the table below indicate that neither Participation Score nor Factors 1 4 affect the Fact Base Scores of local hazard mitigation plans in the study area.

PAGE 167

153 Table 35 : Model 7 (Fact Base Sco re) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 0.19414049 0.19414049 0.28 0.6039 Factor 2 1 0.21081148 0.21081148 0.30 0.5889 Factor 3 1 0.88836123 0.88836123 1.26 0.2712 Factor 4 1 0.00681187 0.00681187 0.01 0.9224 Participation Score 1 2.57949103 2.57949103 3.66 0.0661 The dependent variable in Model 8 is Action Score. Factors 1 4 and Participation are the independents. Below, the summarized results indicate that neither Participation Scores nor Factors 1 4 affect the Actions Scores of local hazard mitigation plans in the study area.

PAGE 168

154 Table 36 : Model 8 (Actions Score) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 2.52350213 2.52350213 2.96 0.0965 Factor 2 1 0.15686868 0.15686868 0.18 0.6713 Factor 3 1 1.01359851 1.01359851 1.19 0.2850 Factor 4 1 0.56041457 0.56041457 0.66 0.4245 Participation Score 1 0.30273340 0.30273340 0.35 0.5562 Goals Score is the dependent variable in Model 9. Factors 1 4 and Participation Score are the independents. The SAS output of Model 9 is presented below and indicates that neither Participation Scores nor Factors 1 4 affect the Goals Scores of local hazard mitigation plans in the study area. Table 37 : Model 9 (Goals Score) Source DF Type III SS Mean Square F Value Pr > F Factor 1 1 0.83052862 0.83052862 0.31 0.5819 Factor 2 1 0.70435459 0.70435459 0.26 0.6120 Factor 3 1 0.38316457 0.38316457 0.14 0.7080 Factor 4 1 3.85563711 3.85563711 1.44 0.2401 Participation Score 1 0.00115569 0.00115569 0.00 0.9836

PAGE 169

155 During the Plan Quality evaluation phase of this study, patterns began to emerge in the data related to plan quality and the scale at which local plans were created. Under DMA 2000, local communities have the choice to create single jurisdiction or multi jurisdiction hazard mitigation plans for FEMA review. Over the course of the statistical analysis, planning scale began to emerge as a potential driver of Plan Quality results. Model 10 exp lores this relationship. Here, Plan Quality is the dependent variable and Planning Scale and Factors 1 4 are included as independents. The table below summ arizes the results of Model 10. Table 38 : Model 10 (Plan Quality Score) Sou rce DF Type III SS Mean Square F Value Pr > F Plan Scale 1 74.75372534 74.75372534 17.72 0.0002 Factor 1 1 1.53647702 1.53647702 0.36 0.5511 Factor 2 1 4.13020590 4.13020590 0.98 0.3309 Factor 3 1 0.42354710 0.42354710 0.10 0.7537 Factor 4 1 3.35684349 3.35684349 0.80 0.3800 The results of the model show that within the sample frame, Planning Scale is driving Plan Quality. Moreover, Planning Scale, independent of community characteristics (Factors 1 4), predicts improved Plan Quality Scores by quite a large amount (P = .0002). Looking at the relationship between Planning Scale and Plan Quality Scores more closely, the regression model below applies Fact Base Score as the dependent variable

PAGE 170

156 and Planning Scale as the independent. Below, Table 39 summarizes the SAS output results. Table 39 : Does Planning Scale Affect Fact Base Scores? Source DF Type III SS Mean Square F Value Pr > F Plan Scale 1 9.54602943 9.54602943 22.63 <.0001 The results confirm the hypothesis that Planning Scale strongly affects the Fact Base Scores of local hazard mitigation plans. This finding prompts the following question: What approach to planning a single or multi jurisdictional effort is associated with higher Fact Base sc ores? The multiple regression analysis below explore s this question ( Table 40 ). Table 40 : Fact Base (the GLM Procedure; Least Square Means) Planning Scale Fact Base Score LSMEAN Multi Jurisdiction Plans 0.95639286 Single Jurisdiction Plans 2.34633333 The results indicate that within the study area, single jurisdiction local hazard mitigation plans are associated with higher Fact Base score outcomes. This finding is significant because of the large amount of focus and attention given to fostering multi jurisdictional coordination during hazard mitigation planning.

PAGE 171

157 Results and Discussion This chapter assessed variations in Plan Quality scores across a sample of 34 local hazard mitigation plans. In addition, it consist s of a principal components analysis designed to identify underlying factors in the local community characteristic data. By achieving the following three aims, the regression models addressed the overarching research question of how community characteristic s and levels of public participation affect local mitigation plan quality outcomes: Aim 1: To identify key relationships between local community characteristics and local plan quality scores; Aim 2: To analyze how community characteristics affect levels o f public participation in local hazard mitigation planning; and Aim 3: To evaluate how public participation in planning affects local hazard mitigation plan quality. The results of the quantitative analysis indicate that within the study area public participation is not associated with recent local mitigation plan quality outcomes. Additionally, Models 1 4 demonstrate no associations between community characteristics and local Plan Quality scores. For example, At Risk communities do not have signifi cantly higher or lower quality plans than affluent communities, nor do communities with high levels of Cultural Disparity have lower quality hazard mitigation plans than predominantly White, English speaking communities. The finding that plan quality does not differ between communities based on social vulnerability is significant. Although plan quality is not consistently high in socially

PAGE 172

158 vulnerable communities, the fact that plan quality is not worse suggests that despite limited resources and socioeconomic hardship the most vulnerable communities in the study region have not been systematically underserved. Although the results do not suggest systematic planning failures in the region, communities within the sample area appear to have paid very little attention to the social vulnerability profiles of their planning jurisdictions when developing their local hazard mitigation plans. This is problematic from the disaster resilience perspective. Although social vulnerability is complex and often eclipsed by the risks posed by physical hazards, it is one of the largest threats to local disaster resiliency. As previously discussed, the Fact Base component of Plan Quality consists of a range of indicators related to the physical location and extent of hazard damage, including the magnitude of hazard events, exposed populations, and evacuation clearance times. It also includes indicators related to local demographics, education levels, health services, social services and the econ omy, all of which are commonly used indicators of local social vulnerability. Across the sample of 34 plans included in this study, Fact Base Scores are consistently low. Contextual understandings about social vulnerability and applications of local inform ation have been largely absent from the process of local hazard mitigation planning across the study region. Planning agencies must learn to understand social vulnerability, and the root causes specific to their communities, in order to build resiliency th rough planning. Ultimately, this approach demands that social vulnerability reduction play a larger role in driving decision making related to mitigation actions and goal setting.

PAGE 173

159 Out of the four extracted factors from the principal components analysis, Fa ctor 2, Racial Disparity, was the only factor that had a relationship to the Participation variable. Factor 2 emerged as a significant predictor of participation levels during planning and describes communities in which the dominant minority (in this case African Americans and Native Hawaiian/Other Pacific Islanders) is at the table. The cities of Fairfield and Elk Grove both have high loadings on Factor 2. Although it is not within the scope of this study to identify why certain stakeholders participate in the hazard mitigation planning process, future studies m ight explore the factors that explain increased participation in communities characterized by high levels of Racial Disparity in the context of hazard vulnerability. In addition to addressing Aims 1 and 2, the results of the quantitative analysis show that participation does not influence plan quality as previously thought. High levels of public participation in planning do not drive plan quality outcomes, nor do they affect the Fact Base Scores of l ocal plans in the study area. This finding leads to the conclusion that the results of this study, there is very little empirical proof that high quality hazard mitig ation plans lead to better results in terms of risk, vulnerability, and loss reduction outcomes. Despite the results that show no association between participation and plan quality outcomes, the literature maintains that participatory planning can result in valuable learning opportunities and risk reduction outcomes for those involved. Meaningful participation is often a positive end in itself, even if it does not result in empirically better plans. While public participation seems to have little to no ben efits when plan quality is

PAGE 174

160 the measured outcome, participation may have powerful implications for the success of plan implementation over the long run. Moreover, it is very unlikely that these beneficial outcomes are measurable through existing plan qualit y metrics. These discussions highlight important directions for future research studies to explore. A final key finding from this phase of the study is that Planning Scale drives plan quality outcomes to a greater degree than any of the other variables ad dressed in this study. Specifically, the single jurisdiction approach to local hazard mitigation planning is associated with higher Fact Base scores. These findings suggest that single jurisdiction plans have been more successful at incorporating locally r elevant information about social vulnerability and community capacity into plans than those created at the multi jurisdiction scale. In addition to cost reduction, one of the justifications of the multi jurisdictional approach to planning is that it enable s communities to simultaneously plan for regional climate change impacts and local hazards. Additionally, multi jurisdictional efforts allow communities to pool their resources and establish partnerships related to each of the four phases of the emergency management cycle. The results from this analytical phase open the door to deeper, more focused evaluations of local hazard mitigation planning processes and social vulnerability assessment. F uture research m ay further explore the relationship between plan Fact Base Scores and social vulnerability outcomes by evaluating local reductions in social vulnerability over time in cities with a wide range of participation scores Through a series of in depth stakeholder interviews and archival document analysis, th e following chapter offers insight into the planning environments in which high and low quality local hazard mitigation plans are created. The interviews enable deep,

PAGE 175

161 focused investigation into local planning processes of interest and aim to explain the re participatory planning methodologies reveals rich information about the planning participation processes, including how local social vulnerability information has been used in decision making (information that is nearly impossible to glean from document analysis alone). Additionally, the following case study analysis is designed to shed light on why single jurisdictional plans out perform multi jurisdictional plans in t he region.

PAGE 176

162 CHAPTER V A QUALITATIVE EVALUATION OF LOCAL MITIGATION PL A NNING Introduction In testing ideas about how public participation should play a role local hazard mitigation planning, we need to assess what the role of public participation currently is. Hardly ever is planning a cut and dry process. It is heavily influenced by politics, social dynamics, economics, physical geography and myr iad other factors that define the place and character of any community. The six case studies in this chapter focus on places of varying size and circumstance. Each case is an attempt to learn what factors have driven the successes and shortcomings experien ced by those communities in the context of lo cal hazard mitigation planning. This chapter consists of a series of instrumental case studies that help provide insight into the results of the previous quantitative analysis. Through a series of in depth stake holder interviews and archival document analysis, we investigate the participatory planning methods of diverse communities within the study area. In addition to examining the successes and failures of both single and multi jurisdictional planning efforts, the in depth interviews help to identify additional outcomes associated with high levels of public participation during local hazard mitigation planning. Ultimately, the goal of this chapter is to better understand the planning contexts in which the highes t and lowest quality plans are produced. In addition, this phase of the study will explore the factors that contribute to successful hazard mitigation planning, regardless of planning scale.

PAGE 177

16 3 Methods This section discusses the data collection tools used for the interview phase of this study including interview guides, reflexive journal and coding procedures. It also describes how decisions about case selection were made. Interview G uides A semi structured interview approach was used for this portion of the study. In lieu of using a sequence of identical, scripted questions for every interview, the semi structured approach allows interviews to be conducted with more flexibility. By using an interview guide, we ensure that the same general areas of informa tion are collected from each interviewee while simultaneously allowing for a degree of adaptability during each conversation. Two interview guides were created one for conversations with organizational participants and another for conversations with local community participants It was necessary to develop two interview guides because in most cases the organizational and community stakeholders had starkly different roles, responsibilities, and access to information during the planning process. Therefore, d ifferent questions and topics were relevant to participants from each stakeholder type. public participation in the context of local hazard mitigation planning. In each of the interview guides, discussion questions were designed to address the goals, successes and

PAGE 178

164 as well as details about public participation in their jurisdiction during local hazard mitigation planning and the overall planning environment. Data Collection and S torage Each interviewee was first contacted through email and asked to participate in the interview process. Follow up emails were sent to non respondents seven da ys after the initial interview request. In accordance with COMIRB Human Subjects policy, potential interviewees were not contacted more than three times to schedule an interview. After receiving participant consent, the interviews were conducted using Sky pe to allow for audio recording during each conversation. Although interviewees were given a guarantee of confidentiality, some interviewees declined to have their conversations recorded. In these cases, participant responses were recorded by hand for the purpose of transcription and analysis. Each interview was designed to last thirty minutes. Once an interview was completed, the conversation was transcribed and the text uploaded into NVivo 9, a qualitative data analysis software that enables researchers t o code qualitative data and conduct systematic analysis to find patterns, relationships, and themes. A codebook was developed as a tool for analyzing the interview data. It assisted in dissecting the interview data in meaningful ways while keeping the rela tionships between parts intact (Miles & Huberman, 1994). The codebook was made up of a set of descriptive codes that attribute types of phenomena to segments of text. Initially, the codebook was developed at the same time as the interview guides. However, as the field

PAGE 179

165 experience developed over time, the codes were revised and changed. New codes were added and others were broken down into sub codes. After the interviews were coded, case summary outlines were created for each city. The purpose of the case sum mary outlines was to present a summary of preliminary findings, to provide a review of data quality, and to outline an agenda for the next stage of analysis. The case summary outlines proved to be incredibly useful during data analysis because they provide d the first coherent, broad portrait of each case. Reflexive J ournal In addition to collecting detailed responses from interviewees, focused time was spent journaling directly after each interview. The reflexive journal included information about initial impressions of the interview including reactions to answers and questions that arose. Ultimately, the reflexive journal contributed to the reliability of the study as well as its internal and external validity. It provided an additional method for organizi ng data and emerging ideas and made it possible to return to the interview data again and again over time. Case S election Two key propositions guide the selection of the case studies presented in this chapter. First, although various factors influence the hazard mitigation plan quality, communities within the sample frame who take single jurisdiction approaches to mitigation planning produce higher quality plans than those who take multi jurisdictional approaches. Second, although the quantitative phase of this study provided no empirical evidence that participation contributes to higher quality mitigation plans, the literature

PAGE 180

166 maintains that participatory planning leads to improved planning outcomes over time. These propositions guide the qualitative phase of this study and provide limits to the scope of the case study analyses. A stratified purposeful sampling approach was used to select case study communities. The approach allows for the selection of particular cases for analysis that vary according to k ey dimensions and was selected to ensure the representation of communities with varied participation and quality outcomes. It involves the identification of samples within samples by highlighting characteristics of particular subgroups of interest. Ultimat ely, by facilitating comparisons between subgroups of interest, this sampling approach allows researchers to capture major variations in a phenomenon between cases (Miles & Huberman, 1994). Because this phase of analysis is interested in exploring why Plan ning Scale drives Fact Base scores and how participation may contribute to improved hazard mitigation planning outcomes other than Plan Quality, the case studies have been selected based on participation levels and Fact Base scores. Fact Base was selected as a dimension of interest over Plan Quality in an attempt to isolate the effects of community context and participatory planning activity on the collection and incorporation of information about social vulnerability into planning. By comparing the experie nces of communities that have been successful at incorporating information about local social vulnerability with those that have not, we are better able to identify the factors that contribute to the creation of detailed, locally relevant mitigation plans. The Participation and Fact Base data recorded during the plan quality analysis were used to guide the selection of the case studies. Initially, the thirty four cities in the

PAGE 181

167 sample were organized into the subgroups by calculating the mean Fact Base score for the sample. Out of a total possible score of 10 points, the mean Fact Base score of all of the local plans was 1.202 points. The population standard deviation was then calculated for the sample and yielded a value of 0.836. The plans with high and low Fact Base scores were selected by sampling all of the plans with Fact Base scores of one standard deviation above the mean and one standard deviation below the mean. This sampling approach yielded a sample of six case study cities (Table 41 ). Table 41 : Case Study Sampling Based on Fact Base Scores High Fact Base Score (1 standard deviation above the mean) Low Fact Base Score (1 standard deviation below the mean) City Roseville (M3) Napa (M3) Woodland (M2) Citrus Heights (M2) Windsor (M2) West Sacramento (M2) Next, the sample of six cities was stratified further to account for the participation dimension of each local hazard mitigation plan. In this study, local planning activity characterized by M1 or M2 participation models are treated as examples of plans cr eated with low levels of public participation. Local planning activity characterized by M3 or M4 participation models are examples of relatively high levels of public participation. The matrix below ( Table 42 ) shows the organization of sample cities based on the combined Fact Base and participatory planning criteria. Although there were no communities within the sample frame that fit in the low fact base score/high participation category, it does not mean that these types of communities do not exist. A valu able extension of this research study might include a number of case studies within communities that have done a great job at engaging and collaborating with public

PAGE 182

168 stakeholders but have been able to incorporate information about social vulnerability into their risk assessments and final planning decisions. Table 42 : Final Case Study Selections High Participation Low Participation High Fact Base Score (+) Roseville (M3) Napa (M3) Citrus Heights (M2) Woodland (M2) Low Fact Base Score ( ) None Identified Windsor (M2) West Sacramento (M2) For the interview phase of this study, two plans were selected with high Fact Base scores and high participation levels, two plans were selected with high Fact Base scores and low participation levels and two plans were selected with low Fact Base scores a nd low participation levels. Within the larger sample frame there were no examples communities with plans that had low Fact Base scores and high participation levels. Figure 12 shows the locations of each case study cit y selected for deeper analysis.

PAGE 183

169 Fig ure 12 : Map of Case Study Communities Summary characteristics of each of the six case study cities are presented in Table 43 The cities represent a range of Fact Base scores and participation models, as well as communities that have taken both single and multi jurisdi ctional approaches to planning.

PAGE 184

170 Table 43 : Summary Characteristics of Selected Case Study Communities City Roseville Napa Citrus Heights Woodland West Sacramento Windsor Planning Scale SJ (City) SJ (City) MJ (Sacramento County) MJ (Yolo County) MJ (Yolo County) MJ ( ABAG ) Participation model M3 M3 M2 M2 M2 M2 Roseville Napa Citrus Heights Woodland West Sacramento Windsor Plan quality score 13.25 13.08 8.20 8.94 7.24 8.31 Fact base score 4.32 3.33 2.00 2.17 0.17 0.46 Interviewee S election As discussed in Chapter 2, the conceptual framework that guides this study outlines a typology of stakeholder participants who are involved in the local hazard mitigation planning process. These stakeholders include official members of the hazard mitigatio n planning team, technical or subject matter experts, or local community/citizen volunteers. The interviewee selection methodology for the following six case studies draws from each of these core participant types. Figure 1 3 diagrams the interactions thro ugh which key types of stakeholders participate in information production, goal setting and decision making during local hazard mitigation planning. In order to identify the direction in which local hazard mitigation planning should go in the future, we mu st first understand the factors that affect the ways in which public and organizational stakeholders access, share and use

PAGE 185

171 local information. By interviewing participating stakeholders from the selected case study communities we were able to closely examine the conditions that spur various approaches to planning. In addition, we were able to identify why certain planning approaches lead to posi tive outcomes while others do not. Figure 13 : Models of Stakeholder Participation During Local Hazard Mitigation Planning For the purpose of this study, the organizational stakeholder participant group includes technical experts, emergency managers, policy makers and consulting professionals who are directly involved in the local hazard mitigation planning process. The local community/citizen stakeholder group includes residents of the jurisdiction for which the plan was developed, representatives of neighborhood and community special interest groups (e.g. school groups, religious groups, environmental organizations, local his toric preservation organizations, etc.) and local advocates, representatives and/or service providers for socially vulnerable groups. Stakeholders of this type are community

PAGE 186

172 volunteers. In each case study city, in depth interviews were also conducted with formal hazard mitigation planning team members. Interviewees were selected using the saturation sampling technique. Organizational and community stakeholders were identified from the participant contact information recorded in each local hazard mitigation plan. All recorded participants in the planning process, including members of the mitigation planning team, outside consultants, and members of the public were contacted with a meeting request. Additional participants were identified using the snowball sa mpling technique. Because the identities of many individuals were not recorded in planning documents, their contact information was more effectively obtained through referral from previous interview subjects. The snowball sampling technique was successful at identifying large numbers of individuals who were not formal participants during the local hazard mitigation planning process but who were key organizational or local community stakeholders behind the scenes during the planning process. Data and L imita tions An important limitation associated with collecting information from participant interviews is that the self reported data relies on memory, which has high potential for bias. Possible sources of bias related to memory include: Selective Memory : remem bering, or failing to remember, experiences or events that occurred in the past; Exaggeration : representing outcomes or embellishing events as more significant than suggested from other data;

PAGE 187

173 Attribution n agency while attributing negative events or outcomes to external actors; and, Telescoping : recalling events that occurred at one time as if they occurred at another time. In addition to the limitations associated with self reported data, interviews cannot determine causal effects and their results are rarely generalizable to an entire group of stakeholders. Yin (2002), however, maintains that case study research has the pow er to generate valid findings when conducted systematically. In this study, the interview data (collected methodically through the use of interview guides), combined with documentary evidence (including planning meeting minutes, planning materials, newspap er articles and websites) and the proper application of qualitative data analysis software, allow for systematic comparisons to be drawn between stakeholder experiences that are based on factual evidence. Planning Scale and Participation during Local Haza rd Mitigation Planning: Six Case Studies Between the months of June 2011 and November 2011, a total of forty one stakeholders participated in interviews for this study. The interviews were conducted across the six cities and five counties. Thirteen intervi ews were conducted with stakeholders from the city of Roseville and ten interviews were conducted with stakeholders from the city of Napa. Six interviews were conducted with stakeholders from the city of Citrus Heights. The cities of Woodland and West Sacr amento participated in a multi jurisdictional planning effort and eight interviews were conducted

PAGE 188

174 with planning committee members from these cities. Six interviews were conducted with planners and organizational stakeholders from the city of Windsor. The interviews focused on understanding the experiences of three key stakeholder types during their participation with local hazard mitigation planning efforts in their community. Planning participants were asked for details about: (1) public participation and inter agency collaboration during the local hazard mitigation planning process, (2) community contexts related to previous hazard events and mitigation planning activity, (3) successes related to the planning process, and (4) challenges related to the pla nning process Time was also spent reviewing planning documents and compiling timelines to develop a detailed picture of the planning approaches adopted by each community. Information about each of the case study cities was supplemented by additional plans and reports about those communities. These resources (either provided by planning participants of collected from public web based sources) provided important information about the histories and socio political contexts of the case study communities. Group s of interview participants from each community were made up of varied stakeholder types ( Table 44 ). In Citrus Heights (n = 6), Woodland and West Sacramento (n = 8) over half of the interview participants were members of local hazard mitigation planning te ams. In Windsor (n = 4), all of the interview participants were mitigation planning team members. In Roseville (n = 13), almost half of the interview participants were public stakeholders and in Napa (n = 10), 30 percent of interview participants were publ ic sta keholders and local volunteers.

PAGE 189

175 Table 44 : Distribution of Participant Characteristics by Community Community Citrus Heights Napa Roseville West Sacramento and Woodland Windsor Organizational affiliation Consulting Company 1 0 1 1 0 Organizational affiliation Citrus Heights Napa Roseville West Sacramento and Woodland Windsor County/City Agency 4 7 6 6 4 Business/Landowners/ Resident 1 3 4 0 0 NGO 0 0 2 1 0 Total 6 10 13 8 4 Role in Planning Process Planning Team member 5 7 6 6 4 Technical Advisor 0 0 1 1 0 Volunteer 1 3 6 1 0 Total 6 10 13 8 4

PAGE 190

176 Across the six communities, higher numbers of organizational stakeholders were identified and interviewed than were public stakeholders. This was true even after snowball sampling and collecting information about additional stakeholders from interview participants. As a result, higher numbers of organizational stakeholders and mitigation planning team members were intervi ewed than public s takeholders. As expected, it was more challenging to identify public stakeholders from communities with plans characterized by low levels of public participation. Although the interviews did not include questions about plan quality scores or the results o f the participation analysis, stakeholders from communities with low quality plans were less willing to participate in interviews than stakeholders from communities with high quality plans. A number of public stakeholders were frustrated with their experie nces with the planning process and did not want to spend additional time talking about the plan. In the communities with the lowest quality plans, organizational stakeholders and mitigation planning team members were reluctant to share any details about th eir process at all. NVivo 9 software was used to draw out themes and identify patterns in the interview data. Two conclusions emerged from the results of the previous quantitative analysis and prompted a deeper exploration: 1) Planning scale is a major drive r of Fact Base score outcomes. Moreover, single jurisdiction planning efforts produce higher quality plans than multi jurisdictional efforts; and 2) Public participation in planning does not contribute to improved local hazard mitigation Plan Quality outcomes

PAGE 191

177 The qualitative phase of this study answers the why for the two statistically driven propositions above. To facilitate this process, the six case studies were organized into themes. The thematic organization allows for the close evaluation of patterns in community contexts and experiences that lead to specific planning outcomes. Moreover, it facilitates comparison between cases. The first theme Planning Scale explores why planning scale emerged as a major factor responsible for the quality outcomes of local hazard mitigation plans. Based on the data and insights derived from the community interviews, a comparison of the Citrus Heights, Windsor, and Roseville case studies were selected as a lens through which to unpack the relationship planning scale an d Fact Base scores within the study area. The second theme Public Participation considers the beneficial planning outcomes associated with public participation that were not captured by Plan Quality metrics. Here, the case studies from the cities of Woodland, West Sacramento and Napa offer insight into the positive outcomes associated with collaborative pu blic participation in planning. Planning Scale: Single and Multi Ju risdiction Local Hazard Mitigation Plans The following analysis presents three case studies from the California communities of Windsor, Citrus Heights, and Roseville. The case studies include evaluations of a regional scale multi jurisdiction hazard mitiga tion planning environment, a county scale multi jurisdictional planning environment, and a city scale single jurisdiction planning environment. Together, the case studies help to clarify why

PAGE 192

178 planning scale emerged as a major factor in determining the quali ty outcomes of local hazard mitigation plans. Regional Scale Multi Jurisdiction Plan: Windsor, CA The Town of Windsor is located in central Sonoma County, twenty miles east of the Pacific coast. Windsor was incorporated in 1992 and at the time that the to local hazard mitigation plan was approved by FEMA, the population was approximately 25,158. Windsor has experienced a number of costly flooding events over the last twenty years that have damaged property and crops. Between 1990 and 2010 Sonoma C ounty lost $356 million due to flood related hazards. Additionally, because Windsor is located along the Healdsburg Fault, earthquakes present a considerable hazard concern for the community. The Plan After DMA 2000 was passed, Windsor created their first local hazard mitigation plan. In 2005, the Town Council adopted a resolution to partner with the Association of Bay Area Governments (ABAG) to update their local mitigation plan every five years. ABAG is a regional planning agency and Council of Governmen ts that is governed by a 38 member Board. The Board, which is comprised of locally elected officials based on regional population, makes decisions, appoints committee members, authorizes expenditures and recommends policy for the region. All nine counties and 101 cities and towns within the Bay Area are voluntary members of ABAG, representing over seven million people.

PAGE 193

179 Although ABAG members are not required to partner with the Council of Governments on planning projects, many communities, including Windsor elected to develop their FEMA jurisdictional planning effort. Counties with previously secured FEMA mitigation planning grants (Contra Costa, for example) typically choose to hire planning consultants a nd create multi on information collected from planners from ABAG and Windsor, one of the key benefits of partnering with ABAG on their regional approach to planning was the opportunity to remain aware of what other communities in the region were doing to reduce their hazard risks. Additionally, by pooling their resources (including data, outreach materials, technical assistance, etc.) and holding planning meetings at the regional scale, loc al communities were able to save both time and money on their mitigation planning efforts. 2009. Windsor submitted a letter of commitment to ABAG to participate in the regional hazard mitigation planning process in May. Soon after, representatives from participating ju risdictions across the region met at an ABAG led mitigation workshop where they and discussed regional priorities for each strategy. During the regional workshop, mitigat ion strategies were added, modified, and deleted based on participant recommendations. Sixteen months later, The Town of Windsor reviewed the draft plan and submitted any comments related to the content of the plan to ABAG. At the same time, the Windsor S

PAGE 194

180 various departments (and met twice a month as a communication team for development, construction and other Town related projects) oversaw the local review process for annex to the ABAG plan. A member of the SRC described their role in the planning process this way: at new development projects without any public participation. It is an open forum for experts to talk and then their feedback goes back to the applicant and community with comments. The advantage is being able to talk without the public Once the draft mitigation strategies were agreed upon by the SRC, the Windsor annex wa s presented at a public Town of Windsor Planning Commission meeting and advertised on the Town of Windsor website. The majority of organizational stakeholders effort was to meet FEMA requirements and to educate the community about general hazard risks. There were no attendees at the town meeting and no comments were received about the electronic draft of the plan. In response to further questioning about the lack of publ ic participation in the planning process, a member of the Windsor SRC mentioned that despite strong public interest and involvement in other local development projects, momentum for disaster related issues was nonexistent: downtown revitalization project is getting lots of public involvement and they have started thinking about the broader impacts of redevelopment projects on the community. But planners are not having reall y in depth discussions with people planning model that is low on the ladder of public participation. At the start of the plan

PAGE 195

181 update process, public outrea ch was focused towards engaging local decision makers and organizational stakeholders through policy and planning workshops. Later, local jurisdictions were directed to engage their communities. Windsor and the other communities involved in the 2010 ABAG m itigation plan were informed that they must provide the public with at least two opportunities to comment before submitting the plan to FEMA and that ultimately it was up to them as to how they wanted to engage their community members. One planning team m ember described local mitigation plans a waste of resources. Without mitigation planning, local resources could be reallocated to more meaningful projects, ones with greater potential to reduce risk and more appropriately address hazards in the community. the Sustainable Community Strategy lately with Cal EMA and I believe that hazards can be ptation have become really salient issues. People are interested, but not so much in earthquakes. According to three organizational stakeholders, personnel changes in the local planning depa get the plan approved. One Windsor SRC member acknowledged that because the community leaned on other agencies to complete their local plan annex, there were many aspects of the pla n development process that the community was not involved in. Three of the four interviewed planning team members also expressed that the local mitigation planning requirement established by DMA 2000 was redundant.

PAGE 196

182 their hazards Out of the sample of thirty 2010 Mitigation Plan was the one with the lowest Plan Quality score. The plan scored lo wer than average on all three elements of Plan Quality and did not include any information about socially vulnerable communities in the hazard risk and vulnerability analys is Additionally, the planning effort garnered limited public input and support. Alt hough the interviews did not include a discussion about plan quality, very few members of the SRC and Town Council were willing to be interviewed about the planning effort and its outcomes. As part of the plan update process, the Town of Windsor committed to updating their multi agency planning effort. In the event that ABAG is unwilling or unable to act as the lead agency in the multi jurisdictional planning effort, the Town plans to contact other agencies in order to identify another regional forum for developing their local hazard mitigation plan. County Scale Multi Jurisdiction Plan: Citrus Heights, CA Located in northeast Sacramento County, Citrus Heights is described as an economically significant suburb of the larger Sacramento region due to its thriving base

PAGE 197

183 Incorporated in 1997, Citrus Heights is surrounded by unincorporated industri al and agricultural lands and unincorpo rated residential communities. Citrus Heights sits in a low lying delta with large floodplain areas. As a result, the community faces a variety of flood related hazards. Although the city is situated within several d rainage basins, the majority of streams in within the city limits remain unaltered and contribute to the natural riparian corridor. Two creeks in particular, Arcade and Cripple Creek, have relatively limited hydraulic capacities and are quickly exceeded du ring severe run off events. In the past, this has led to overflow of stream banks and temporary inundation of the floodplain and adjacent low lying areas. In April of 1998, Federal Disaster Declaration was issued in the Citrus Heights dues to flooding. The Plan facilitated by an outside consulting agency. Development of the multi jurisdictional plan brou ght together the mitigation efforts of the county and eight municipalities. Rather than developing individual mitigation plans, the risk and vulnerability profiles of each of the eight participating communities were included in the larger county plan throu gh chapters The primary purpose of the community specific elements in the plan was to address the DMA 2000 regulatory requirement that hazards with different degrees of risk across an entire planning area be evaluated separately Additionally, the community

PAGE 198

184 elements also separate flood and repetitive loss related problems and solutions for City and County Community Rating System purposes. The community elements included short community profiles that summarized the capabilities, critical infrastructure and general plan elements that were relevant to local flood risk. The community elements also included lists of community specific flood hazard mitigation recommendations that were identified by the multi jurisdictional Hazard Mitig include any contextual information about human vulnerability to hazards nor about ways in which the mitigation plan was tied into existing community regulations and programs. The Hazard M itigation Planning Committee, which was responsible for leading the multi jurisdictional planning meetings, setting mitigation priorities and evaluating data, included representatives from: City of Citrus Heights City of Elk Grove City of Folsom City of Galt City of Isleton City of Rancho Cordova Sacramento County and 69 additional special districts and local government organizations within Sacramento County Twenty one citizens from jurisdictions across the county participated as members of Hazard Mitiga tion Planning Committee (HMPC). Despite the high number of public

PAGE 199

185 stakeholders who were recorded as members of the HMPC, none of the stakeholders listed on the HMPC roster were residents (or employed within the city limits) of Citrus Heights. During the e arly stages of the plan development process, the multi jurisdictional HMPC held three meetings to discuss risk, vulnerability and hazard prioritization at the regional scale. Over the course of the planning project, five additional meetings were organized for the public. These meetings were designed to solicit feedback on multiple drafts of the plan. Prior to the public meetings, a draft of the hazard mitigation plan was posted on the Sacramento County website to provide the general public with an opportuni ty to review the document. Hard copies of the draft plan were made available for review at libraries, County Municipal Services Agency public counters, and at Sacramento City Hall. These public outreach strategies met the public participation requirements established by DMA 2000, however, no one attended One public meeting held in the City of Sacramento drummed up a number of comments concerning emergency response in the event of a levee breach and changing risk due to re certification of the American River levee system. The levee related comments were addressed in the Mitigation Action section of the hazard mitigation plan. No public comments were made at the remaining public meetings, including during the meeting held in Citrus Heights. Organizational stak eholders across all planning roles expressed dissatisfaction in the lack of local community participation during the multi jurisdictional planning effort. Two organizational stakeholder participants expressed that hazard mitigation was a difficult

PAGE 200

186 topic to have with lay people at the table. Moreover, it was challenging for them to gain adequate public interest when communicating solely about hazards. knowledge of the general p The same stakeholders attributed the lack of public participation during planning to risk perception issues, the technical nature of hazard mitigation terminology, and the timing of the public meetings In addition to achieving minimal levels of public participation in the planning process, the county scale planning process did little to boost the knowledge of the public about hazards or to drum up public support for mitigation action. The lone attende e of the two public meetings held near Citrus Heights, agreed to be interviewed for this study. The public stakeholder expressed extreme frustration with the content and quality of the meetings he attended. public me one person from the County, and a planning consultant. I asked the consultant to f or a hard copy of the plan or for a URL to an electronic version. She stated that it was in draft form and not yet available to the public. Then she showed a Power Point presentation that I did not understand. Why were they asking for public comments when the public was still being kep t in the dark about The sparsely attended public meetings were criticized for being poorly organized and the topics presented by planning facilitators and agency representatives were described as jargon heavy and difficult to understand. Despite his interest in participating in the planning process, the public stakeholder felt as though his community was poorly

PAGE 201

187 notified about the public meetings and that he was provided with inadequate information about the relevance of the mitigation plan to his community. much Sacramento County was paying the consultant for the public meetings, but in my opinion, $1 would have been too much for what was presented at the t The public stakeholder had not been informed about the 2011 Plan update process or about opportunities to participate further. hazard mitigation plan and I do not understand why Sacramento County could not do the same for all of the time I volunteere d with them on the pre paration of Although only one public stakeholder interview was possible for the Citrus Heights case study, his responses and anecdotes highlight some of the shortcomings of the multi jurisdictional approach to plan ning, especially as it relates to public engagement. City Scale Single Jurisdiction Plan: Roseville, CA The City of Roseville is situated in the western foothills of the Sierra Nevada Range. Located within Placer County, approximately 16 miles northeast o f downtown Sacramento, Roseville is the largest city within the county. Over the past two decades Roseville has experienced significant residential and commercial growth. Moreover, the community has attracted large amounts of industrial development and tec h companie s. and office space, and industrial development. Analysts anticipate that Roseville will continue to be the focus of significant regional development in the future. The focus of new development will be located primarily along the western, eastern and northern areas

PAGE 202

188 of the community. At the time that the 2011 Roseville Multi Hazard Mitigation Plan (RHMP) was approved, the city had an estimated residential population o f 115,781. The Roseville community is no stranger to disasters. The city lies within the boundaries of the Sacramento Metropolitan Area and borders on the Sacramento San Joaquin Delta region, which has long history of flooding. Although the majority of previous losses to property and crops have been due to wildfire and flooding events, Roseville is also susceptible to seismic activity. On April 28, 1973, a massive explosion erupted in the Southern Pacific Railroad yard near Roseville. A 21 car munitions train, carrying more than 7,000 bombs heading for Vietnam, caught fire and sent debris for miles into surrounding neighborhoods. The explosions began at 8:00 am and continued through the following day. The majority of the Roseville rail yard was destroyed in addition to surrounding buildings and homes. More than 350 people were injured. Although DMA 2000 does not require communities to assess their risk related to human caused hazards, the City of Roseville includes human caused hazards in their mitigation plan in part because of their experience managing the rail yard explosion in 1973. In 1983, Roseville began participating in the National Flood Insurance Program (NFIP). The NFIP requires participating communities to follow two key guidelines: 1) all new b uildings and major improvements to development must be elevated to or above the 100 year flood level, and 2) new developments within the floodplain are prohibited from increasing flood problems or damaging other properties. Later, in 1991, Roseville joine d the national Community Rating System (CRS), a voluntary incentive program for NFIP participating communities. The objective of CRS

PAGE 203

189 is to reward communities that go above and beyond the minimum NFIP requirements to help their citizens prevent or reduce fl ood losses. The CRS provides incentives in the form of flood insurance premium rate reductions that increase in value according to the community receives for its eligible floo dplain management activities. Class 1 is the Class 1 community and as of May 2012, it was the only Class 1 CRS community in the nation. Previous hazard events have gre mitigation. In addition to acquiring federal funds for local flood control improvements (including property buy outs and freeboard projects), the 1995 flooding event generated substantial public support for taki ng an aggressive positive stance towards hazard mitigation activity. As a result, Roseville was able to swiftly and successfully link their land use and development planning with hazard mitigation projects. As explained by one Roseville Mitigation Planning Team member: President Clinton and the city manager walked through our town and vowed that this would never happen again. After the flood, all future land use was changed. No new development was allowed in the floodplain. When planning, they start by identifying all flood plains and now, before any new land is annexed by the city, The 1995 flood, along with a collection of sig nificant flood events in following years, context of flood risk red uction and land use planning. The Plan

PAGE 204

190 In addition to being the highest quality plan in the sam ple frame, the 2011 Roseville Hazard Mitigation Plan had the highest Fact Base score out of the 34 plans public participation during pl anning within the sample frame technical expertise of a Planning Team made up of internal City staff and staff from a p rivate consulting firm. In addition to the Planning Team, Roseville established a fifteen member Steering Committee, made up of local citizens and staff from the city planning office. Generally, it was the responsibility of the planning team to produce the planning document and deliver it to the Steering Committee for feedback and approval. The Steering Committee was responsible for supervising the planning team and offering them with guidance throughout the various stages of the planning process. The City of Roseville took the Steering Committee approach towards planning and decision making in order to meet CRS requirements related to the integration of Hazard Mitigation Planning with local flood risk reduction. Both the DMA 2000 and CRS are highly prescrip tive and provide communities with crosswalks to facilitate compliance with program requirements and benchmarks. When asked about why the Roseville Planning Team approached the plan development process in the way they did, a Steering Committee member descri bed CRS as a critical guide: It calls for the Steering Committee approach in which fifty percent of the committee must be made up of local citizens. Because DMA 2000 planning requirements are very similar to CRS it is easy for CRS cities like us to meet DMA requirements so long as we create a multi

PAGE 205

191 numbers of local stakeholders in the hazard mitigation planning proc ess. The structure of the Steering Committee (as required by CRS) facilitated collaboration between planners, emergency management experts, and local stakeholders. All steering committee meetings were open to the public. The role of the Steering Committe e was not to identify mitigation actions. Rather, it was to develop and approve appropriate community outreach strategies, develop planning goals and objectives, to come up with a working definition of critical facilities in the community (in place of thos e defined by FEMA), and to serve as the local community spokes body for the project. Two Steering Committee members described their main role during the goal setting and hazard prioritization process as critical listeners. Although not all of the hazard re lated information was new to them, a majority of respondents expressed satisfaction with the amount of new information they learned about hazards and disaster risks. As equal members of the Steering Committee, citizen participants provided input to gauge public concern for specific hazards and vulnerable areas. Citizen members of the Steering Committee commonly expressed their lack of technical knowledge on the subjects of hazards and disasters. However, none of the six public stakeholder participants inte rviewed for this study described their lack of technical knowledge as a barrier to meaningful participation during the planning process. Respondents described community hazards education as an important, and necessary, goal of the participatory planning pr ocess. Public stakeholders spoke about their renewed sense of empowerment

PAGE 206

192 and responsibility for turning a local focus towards issues of flood control, reducing risk, and even towards changing local land use decisions. lnerability. There were experts there who did. The simple fact that I was made aware of the risks, contingencies, and been a great deal of turnover in the planning staff and committee membership, and the focus of mitigation projects has shifted to things other than flood control. I have taken it upon myself to keep steerin In Roseville, all citizen members of the Steering Committee described some degree of personal and/or local empowerment as a product of the planning process. One participant described the plan a powerful tool for affecting local change: be to gauge the public concern and to protect the public from risks in various ways. The long term goal is certainly to ur plans as weapons to fire politicians, to cite their bad development decisions and the Although Roseville is an example of a city that has been relatively successful at addressing the human aspects of disasters in their hazar d risk and vulnerability assessments, half of the public stakeholders felt as through social vulnerability concerns were inadequately addressed during planning. Some felt as though concerns about vulnerable groups were completely left out of the conversati on. Community stakeholders increased protection of vulnerable groups, particularly if the outreach activity occurs well before the start of the next plan update.

PAGE 207

193 When res ponding to a question concerning their role on the Steering Committee, one of the organizational stakeholders described the planning professionals and emergency Committee they were slightly removed from the planning process, looking down from outside, and overseeing the development of public outreach strategies, mitigation catalogues, and community goals and objectives. They provided historical and institutional knowledge of the community and its neighborhoods from the land use perspective. They also helped maintain general plan consistency with b roader city plans and policies. In addition to forming a mitigation planning Steering Committee made up of fifty percent local c itizen volunteers, Roseville took a number of focused approaches to public outreach that were designed to foster communication and information sharing within the series together stakeholders from adjacent neighborhoods with similar hazard risk profiles. At each Open House multiple members of the Planning Team and Steering Committee led a review o f the risk assessment results and mitigation priorities with the public. Multiple organizational stakeholders were available to answer individual questions and to take advantage of impromptu educational opportunities with local community members. The Stee ring Committee also developed a phone based survey to gather information from the public about their perceptions of risk, vulnerability and mitigation priorities. The existing Roseville early warning auto dial notification system was used to invite

PAGE 208

194 subscri bers to participate in the mitigation planning survey. The Planning Team and Steering Committee had an 18 19 percent re sponse rate for the population. In addition to meeting local CRS requirements and aligning mitigation goals with community priorities, Committee mentioned that public participation during mitigation planning was necessary to educate the community an d to drive plan implementation. get community buy in and to provide education about emergency preparedness, but not so much that it The majority of Steering Committee members expressed that it was difficult to compel the public to participate in mitigation planni ng, particularly when the community has a number of other things to worry about. However, the same respondents were adamant that the benefits of public participation outweighed the costs associated with extending planning timelines and using resources for public outreach. In addition to the strong public and political support for mitigation in Roseville, over half of the planning participants interviewed for the study spoke to the importance of local leadership and experienced consulting in determining the success of the city scale planning effort. stress the businesses on the committe e t

PAGE 209

195 The vision of the planning consultant kept the wind in the sails of the Planning deal to do with the access to resources, inter agency relations hips, and flexibility afforded by the single jur isdiction approach to planning. Results The planning experiences of Citrus Heights, Windsor and Roseville offer considerable insight into why planning scale can affect the Fact Base scores of local hazard mit igation plans. In addition to establishing a smaller planning scope than multi jurisdictional plans, single jurisdiction planning affords communities greater levels of control over all aspects of the planning process, including data collection, team buildi ng, and public outreach. In Roseville, the city scale planning scope allowed for the planning team to build a steering committee that consisted of a representative sample of the comes of the plan because they were able to bring information and details about the context and character of the com munity directly into planning. One Steering Committee member from the City of Roseville offered advice to other jurisdictions that are in the early stages of developing or updating their local hazard mitigation plans: not going to participate that much, so it is vitally important to get a representative committee were on the emergency response team; others were hospital representatives from Kaiser and Sutter. These members are in the trenches and offer a great persp

PAGE 210

196 Citrus Heights and Windsor had very little control over the creation of the multi jurisdictional planning teams or over the logistics of the planning processes. Therefore, they were unsuccessful at engaging local leaders and stakeholders who could provide local perspectives on vulnerability and risk for use in the broader planning effort. In addition to having limited control over the makeup of their multi jurisdictional planning teams, Windsor and Citrus Heights were beholden to pre determined planning schedules agreed upon by their regional planning bodies and co uncil of governments. Roseville, in contrast, had complete control over their local planning effort and was able to align mitigation planning milestones and activities with ongoing development programs and community events. This level of control was respo nsible for maintaining consistency of participation across planning meetings and contributed, in part, for the higher than average plan quality outcomes. Based on the interview data and plan quality analysis, neither Windsor nor Citrus Heights was able to achieve high levels of public interest in hazard mitigation issues, nor did their plans include robust, contextual information related to the social vulnerability profiles of their communities. Despite these results, it is important to keep in mind that mi tigation planning does not happen in a vacuum. Multi jurisdiction planning unifies much of the planning effort (and distributes the cost) that goes into mitigation planning. Sometimes it makes more sense for a region, or a collection of communities, to joi n their planning efforts to carry out mutually beneficial mitigation planning activities at a lower cost. Although the case studies suggest that single jurisdiction mitigation planning leads to preferable planning outcomes, single jurisdiction planning ma y not be the appropriate

PAGE 211

197 choice for every community. Communities can focus instead on the relevant and scalable best practices of communities that have been successful at creating high quality, implementable mitigation plans. Single jurisdiction planning p layed a role in the success strategy that contributed to their higher quality plan and increased community buy in. ategic in aligning their mitigation strategies with other municipal programs (i.e. Community Rating System). By highlighting the intersections between the steps required for CRS certification and hazard mitigation planning process Roseville wa s able to leverage previously agreed upon and engrained community goals to achieve improved mitigation planning and action strategies. In addition to integrating mitigation planning with existing local policies or projects, Roseville was able to bring the right local participants to the table and enlist strong leadership to reach local planning goals. This is another best practice that has the potential to improve multi jurisdictional mitigation planning outcomes. The mitigation planning effort in Rosevill e was led by an experienced planning consultant and a committee made up of local leaders in the health, emergency response, and land use sectors. Multiple organizational and public stakeholders from the City of Roseville spoke about the benefits of working with an experienced planner and organizational stakeholders with years of experience working in the community. Strong leadership was paramount to determining what types of questions the Steering Committee asked in order to paint a clear picture of risk an d vulnerability in the community. Public Participation during Local Hazard Mitigation Planning

PAGE 212

198 The results of the quantitative analysis indicate that public participation in the hazard mitigation planning process has no relationship to plan quality outcome s. The following case studies offer insight into the why of participation does not affect plan quality and they explore the unquantified effects of public participation on loc al hazard mitigation planning. Although public participation does not affect plan quality scores, the literature supports the hypothesis that public participation in hazard mitigation planning is beneficial to plan implementation outcomes. Here we review the participatory processes of communities within the study area that have both hi gh and low participation levels and investigate the effects of public participation (or the lack thereof) on hazard mitigation in those communities. The experiences of the following communities are examined in detail: Woodland, West Sacramento, and Napa. L ow Public Participation: Woodland and West Sacramento, CA Sacramento and is the county seat of Yolo County. Situated in an important agricultural region of the state, Woodland is surrounded on all sides by farm land. One of the fastest 80 has subjected the city to significant growth pressures. To accommodate long term projected growth, Woodland has plans in place to make addit ional land available for urban development, to continue infill development, and to encourage the re use of underutilized lands. The majority of Woodland lies within the existing FEMA 100 year floodplain. Valuable agricultural land and frequently inundated floodplains present very real physical

PAGE 213

199 winds have caused the greatest damage to property while flooding and winter weather has been costly to both crops and property a cross the county. West Sacramento is located at the eastern boundary of Yolo County in the Sacramento Valley. The city is bounded on its eastern, western and northern borders by water bodies, including the Sacramento River, the American River, and the Sacr amento River Deep Water Channel and Yolo Bypass. The entire area of West Sacramento lies within the natural floodplain of the Sacramento River. The land is currently protected in part by levees and by the Yolo and Sacramento Bypasses, which divert floodwat ers west around the city boundary. However, revised FEMA flood maps are expected to be released at the end of 2013 that will more accurately illustrate the flood risks in areas previously deemed safe due to the levee system. In addition to flooding, impact s from wildland and agricultural fire s also pose a risk to the city. The Plan The cities of Woodland and West Sacramento both participated in the 2006 multi jurisdictional Yolo County Multi Hazard Mitigation Plan. A wide range of organizational stakeholde rs were involved in the development of the Yolo County Multi Hazard Mitigation Plan. They included staff from public works, County emergency management officials, and local fire department staff. An outside consultant was hired by the county to collect vu lnerability data and evaluate risk for the county and the majority of the data collected for the local risk and vulnerability assessments was collected from federal sources. Information about local

PAGE 214

200 history, including disaster declarations, damages and miti gation priorities were provided by each local jurisdiction and incorporated into the individual community elements of the plan. distinct efforts: 1) a risk perception questionnaire and 2) a public meeting. First, the cities distributed hard copy questionnaires to residents to gather information about local priorities, preparedness and perceptions of risk. The questionnaires asked residents to rank their concerns about specific hazar ds and about their individual levels of preparedness. The questionnaires also provided residents with an opportunity to rank potential disaster mitigation policies and risk reduction actions by priority. The two cities took slightly different approaches t o distributing the questionnaires. West Sacramento posted the questionnaire on the city website. Members of the public were invited to print out the questions and send their responses to the hazard mitigation planning team. Rather than posting the question naires online, Woodland included copies of the questionnaire in residential utility bills. West Sacramento received 15 responses and Woodland received 889 responses. The data from the questionnaires section of the plan. Although the questionnaires were able to draw out valuable contextual information from local communities, the data was not incorporated into the local risk and vulnerability assessments of the plans, nor wa s the information used to inform local mitigation goals and actions. In addition to the surveys, Woodland and West Sacramento both held public information meeting to provide residents with an opportunity to review and comment on

PAGE 215

201 the draft hazard mitigation plan. Details about the meetings were advertised in city newspapers and online. However, no public stakeholders attended the public meetings in either city. Almost half of the organizational stakeholder participants expressed dissatisfaction with the out comes of the public outreach efforts. Those who did cited a lack of resources as the key barrier to public participation during mitigation planning: power, or local attention. Other issues are m portion of every budget to be spent on one on one meetings with communities within each jurisdiction. As it stands, there is not e nough money for m ore than assessment of emerging threats including socio economic and political risk s but Both cities took a consultative approach to public outreach rather than a collaborative approach. Moreover, neither city used the information they gathered from the community to improve their plans. The information collected and shared from the public outreach efforts did no t inform broader planning goals or mitigation actions in either city. Although social vulnerability to hazards and disasters was not addressed explicitly in the Yolo County Hazard Mitigation Plan nor at the local hazard mitigation planning meetings, two or ganizational stakeholders from the planning team expressed concerns about the exclusion of social vulnerability from the planning process: Especially becau se they are changing s

PAGE 216

202 Russian and Ukrainian immigrants concentrated in West Sacramento. We also have an older demographic of white European farmers. There is generally a large disinterest [in ha zard mitigation] In contrast, two organizational stakeholders felt that despite rapidly changing demographics, socially vulnerable populations within the community were already very good at taking care of themselves. Additio nally, these stakeholders expressed that social vulnerability analysis was outside of the scope of loca l mitigation planning activity: They are survivors. They can take care of themse lves and are good at working the system. They are not interested in learning from us about hazard risks and them. We need to give them an understanding of their risks, and we need to empower them to take care of themselves. Mitigation planning does not need to address these Additionally, one organizational stakeholder commented that public participation was not necessary to contribute to a complete assessment of risk and vulnerability in a local community: fully understand led to a High Public Participation: Napa, CA Incorporated in 1872, the city of Napa is located in Napa County the city straddles the Napa River in the southern portion of the Napa Valley. The city of Napa is roughly 50 miles northeast of San Francisco and 60 miles west of Sacramento. To the south of the

PAGE 217

203 lands and vineyards. Best known as a destination for w ine lovers, the city of Napa became the primary business and economic center for the Napa Valley during the 20th century. A 1975 currently limits its outward growth. Napa is currently the largest city in Napa County and has a population of approximately 74,071 people. Between 1990 and 2010, Napa County lost $436 million due to hazard events. Flooding, earthquakes and landslides have been the most prevalent (and costly) disasters in the County. By and large the largest losses to crops and property in the county over the last 20 years in have been due to flooding. The negative economic impacts of repetitive flooding led the residents of Napa to pass small tax incr eases to pay fo r flood mitigation activities. The Plan mitigation plan adopted in 2004. The City of Napa Hazard Mitigation Planning Team was tasked with updating the plan and de veloping an appropriate public outreach strategy. The Hazard Mitigation Planning Team was made up of representatives from the local fire department, Public Works, the Office of City Planning, the Water Department, and the Community Development Department.

PAGE 218

204 plans in the sample frame that included information related to social vulnerability, including deta iled information about populations living in group quarters, housing tenure, unemployment, per capita income, and tourist populations. Although this information importance of these factors as potential determinants of risk were not articulated in the plan. participatory model of planning. A number of mitigation goals previously developed by local special interest groups (primarily the local fire prevention and flood mitigation groups) were integrated into the larger multi hazard mitigation planning effort. The planning team actively sought input from these local stakeholders and special interest groups, an d cultivated relationships built during previous planning efforts, to create continuity between the 2004 and 2009 mitigation plans. Moreover, information from local stakeholders was used to inform mitigation actions and goals for the plan update. Additiona lly, Napa was one of the only communities in the sample frame that achieved significant levels of attendance at community planning meetings and workshops. Way is a grassroots sp ecial interest group that was created in early 2006 after a citizens. As stated by r

PAGE 219

205 When they were informed about the local hazard mitigation plan update, and about the call for public related planning efforts were years behind schedule and the group resolved to accelerate the completion of the plan through active importa nt conduit through which community perspectives and priorities were brought to and every flood board meeting. y recalled a city employee and Hazard Mitigation Planning Team member. A prevailing sentiment expressed by a number of community stakeholders was that the participation of local communities in hazard mitigation planning was fundamental to both the quality of the planning process as wel l as the outcomes of the plan. county was non existent. As an invested group of locals we were responsible for getting the col laborative process going, getting communication between the city and county started, and ge According to resident participants in the planning process, the Mayor of Napa stepped in to encourage consistent dialogue between community members and the er.

PAGE 220

206 Hazard Mitigation Planning Team rep orted that engaging the public in hazard mitigation planning was still a challenge, while bringing professionals together to talk about hazard mitigation was relatively easy. The organizational stakeholders and members of the Mitigation Planning Team were quick to acknowledge the significance of the strong public participation that they experienced during the local mitigation planning effort. significantly affected the prioritiza tion of a few more months but it turned out ahead of schedule. Why? Because of a neighborhood group who spoke out and was very vocal. They provide d very vocal persuasion and good arguments at city council meetings, and they definitely With respect to the ability of community stakeholders to provide the Planning Team with data and information related to vulnerab le populations, the organizational stakeholders saw this as a core intention of fostering meaningful public participation. In Napa, the planning team solicited information about vulnerable populations during the early stages of plan development by reaching out to nonprofit groups and local social and about local losses and at risk areas exposed by those events. itigation plan that they could hold themselves accountable to and that would be meaningful to the community. In their interviews, the organizational stakeholders described public participation as critical to making those goals a reality.

PAGE 221

207 The importance of public participation as an educational opportunity also emerged as strong theme among organizational stakeholders. Since the last major flood on New public about t heir risks and about evacuation. One of the community planners remarked: Citizen participants also acknowledged the importance of their participation in the process based on new knowledge and capacity building. A number of public stakeholders expressed that they had learned a great deal from being brought into the fold of the local mitigation planning effort. Once community member co mmented: Although public participation results in positive planning outcomes (including local flood risk reduction and increased preparedness), current Plan Quality scoring methodologies are unable to capture the myriad intangible benefits of participatory hazard mitigation planning including knowledge sharing, public empowerment, and trust buildin g. This study prompts us to question the usefulness of Plan Quality as a metric for assessing the success of local mitigation planning efforts. Results The West Sacramento, Woodland and Napa case studies provide a lens through which to examine variations in public participation in the context of local hazard mitigation planning. As indicated by the quantitative results of this study, high levels of public participation do not necessarily lead to higher quality hazard mitigation plans.

PAGE 222

208 However, the Napa cas e study shows that public participation can play a critical role in driving public and political support for plan implementation and can turn the focus of mitigation actions towards vulnerable populations and at risk communities. From the very start of the data collection process, there was a stark contrast in the ability to identify, contact and engage with stakeholders in cities with low community participation levels versus those with highly participatory models. As was expected, fewer community stakehol ders participated in planning within jurisdictions that carried out minimal public outreach and engagement efforts. Additionally, the enthusiasm of stakeholders about being interviewed, as well as their willingness to discuss the details of the mitigation planning process, was much more pronounced in cities with high levels of public participation. In highly participatory planning environments, enthusiasm about the planning process and its outcomes was expressed by interviewees from all three stakeholder pa rticipant groups. Woodland and West Sacramento, both communities with low levels of public participation during planning compared to the other communities included in the sample frame, took a consultative approach to public outreach. Their efforts were cha racterized by low attendance at community meetings, communication breakdowns, and only mildly contextualized local mitigation planning documents. In Napa, where public participation was driven largely by the efforts of local special interest groups, not on ly was the planning team able to attract significant public interest in the planning process, but the mitigation plan is high quality as measured by the fact base, actions, and goals articulated by the plan.

PAGE 223

209 In the city of Napa, the high level of public p articipation in mitigation planning activity was due in part to the existing social infrastructure built by the efforts of previous community based planning initiatives. The most active community participants in the hazard mitigation planning process never thought of their activity as hazard mitigation. Instead, they described their participation as a product of their deep investment in Napa Creek related issues. These community participants asserted that it was their personal responsibility to do their res earch and engage with professional planners and decision makers throughout the entire planning process. Ultimately, these actions led to increased political support for mitigation activity among local decision makers and spurred the rapid completion of a l ong awaited flood control project. In addition to driving positive implementation outcomes and local support for mitigation activity, the case studies highlight the potential for public participation to turn the focus increased attention towards mitigation actions that benefit vulnerable populations and at risk communities. In Woodland and West Sacramento, both organizational and community stakeholders acknowledged the changing demographic profiles of their communities as well as the difficulty associated w ith adequately assessing vulnerability issues related to language barriers and poverty. Public participation in these communities incredibly limited, and in some cases, planners articulated that vulnerable groups and at risk populations (including minoriti es, increased interaction with at risk communities, it is unlikely that future mitigation plans will include accurate interpretations about local coping capacity and soc i al vulnerability.

PAGE 224

210 In Napa, consistent interaction between the planning team, local representatives of social service organizations, and Napa residents prompted conversations about at risk populations that occurred throughout all stages of the planning proc ess. Information about vulnerable neighborhoods made its way into the risk assessment of the plan and, ultimately, into the mitigation strategies. The Napa case study indicates that without public participation, understandings about local social vulnerabil ity are framed exclusively by the perspectives of organizational stakeholders. Communities that plan in the absence of collaborative, public engagement are at risk of establishing long term mitigation priorities that are based on misconceptions about local social vulnerability and capacity. Discussion The six case studies presented in this chapter examined an understudied area of planning related to public participation, planning scale and local hazard mitigation plan quality. By analyzing the nature of pu blic participation during local hazard mitigation planning, and by providing an account of stakeholder experiences in relation to single and multi jurisdiction planning environments, the case studies verify the importance of planning context in determining hazard mitigation plan quality outcomes. Hazard mitigation planning is a dynamic process. Regardless of the jurisdictional scale at which local mitigation plans are developed, stakeholders with diverse perspectives and agendas can act together to incorpor ate local risk and vulnerability data into long term planning decisions. Findings from this study suggest that the engagement of public stakeholders in collaborative, participatory planning activity may not contribute

PAGE 225

211 to higher quality mitigation plans. Ho wever, this study shows that in some cases the richness and diversity of public stakeholder perspectives engaged during local mitigation planning affects local preparedness levels, public buy in, and political support for mitigation activity. Similar to the way in which planners often act as information filters during the comprehensive planning process, local hazard mitigation planning teams filter information and knowledge collected from technical, public, and organizational stakeholders alike Regardless of the amount of public participation and knowledge sharing that happens between the public and planning team members, the collected information and priorities still has to be passed through and approved by the planning team. Ultimately, the p lanning team decides on which hazards will be included in the plan. The planning team sets priorities for mitigation actions and establishes definitions for mitigation planning boundaries including critical facilities, community assets, and vulnerable popu lations. T his reality may explain why participation does not relate to hazard mitigation plan quality or social vulnerability reduction Moreover, it strengthens the conclusion that the planning team matters. In communities where public stakeholders have been brought on as formal members of the mitigation planning team, mitigation priorities are closely aligned with local development priorities. Addition ally, the planning efforts are associated with higher levels of public parti cipation throughout the plan development process. In Roseville, the integrated planning team conceptualized and rolled out customized community engagement strategies that were dramatically more successful at eliciting jurisdiction specific information about community needs, prioritie s, and vulnerabilities In Napa, the

PAGE 226

212 planning team leveraged existing community networks to reach out to and engage with members of the public who were not previously interested in or concerned about hazards Not only did these efforts increase communicati ve and collaborative participation activity between local community stakeholders and the planning team, they strengthened the political palatability of local mitigation actions and led to increased individual preparedness. Communicative planning processes include discourse, knowledge sharing and power sharing As an alternative to strictly technical approaches to planning, communicative planning fosters mutual understanding s of under represented priorities and stakeholder concerns. In the context of hazard risk reduction, communicative planning provides opportunities for resilience building and adaptive management to occur. However, c ommunicative planning becomes increasingly challenging (and logistically infeasible) as the collectio n of planning stakeholders increases. Regional planning processes, for instance, must consider the needs and constraints of multiple jurisdictions, special interest groups, and publics. Very often, ambitious communicative planning strategies are scrapped f or the sake of time and money. As seen from the communities profiled in this study, the hazard mitigation planning team is the heartbeat of the local hazard mitigation planning process. The mitigation planning team sets priorities, filters information, and designs community engagement strategies. Moreover, the planning team serves as a forum for communicative planning to occur regularly and often. By including public stakeholders and community leaders as formal members of mitigation planning teams manageab le boundaries are set around the communicative

PAGE 227

213 process. What is more, this practice can be implemented in both single and multi jurisdiction planning environments. Variations in the planning outcomes of the case study communities were apparent in terms o f the local relevance of the plans (i.e. integration with current land use policy and inclusion of information about local social vulnerability) and community support for mitigation projects. Participants involved in single jurisdiction planning efforts ar ticulated positive outcomes in the community including the timely implementation of mitigation actions and renewed local commitment to mitigation in areas most vulnerable to hazards. It was evident that the planning approaches in these communities had a si gnificant positive effect on the local acceptance of hazard mitigation projects. Moreover, planning participants in these communities spoke about the importance of consulting with residential stakeholders when identifying socially vulnerable populations or at risk infrastructure. The public participation strategies and plan quality outcomes in cities that took multi jurisdictional approaches to planning were very similar. Based on the interviews with community and organizational stakeholders, the cities of Citrus Heights, Woodland, West Sacramento, and Windsor all took a consultative approach to public participation. The outcomes were low public involvement, high dissatisfaction among community stakeholders with respect to the plan, and low Plan Quality sco res. Although Citrus Heights, Woodland, West Sacramento, and Windsor all achieved FEMA approval for their plans, the case studies indicated that their plans offered limited functionality with respect to their lack of integration with local planning priorit ies. Although not articulating positive outcomes may not necessarily indicate a lack of positive planning outcomes

PAGE 228

214 (participants might have known about successes but did not discuss them) very few of the participants involved in multi jurisdiction planning efforts spoke positively about their experience (or about their plans) during the case study interviews. In addition to planning scale, public participation was heavily influenced by previous local experiences with hazards and disasters. The case studies indicate that previous experiences with disasters and loss promote higher levels of engagement between organizational and community stakeholders during local hazard mitigation planning. This is consistent with other studies that have identified previous l osses from disasters as an important driver of community participation in planning (Burby, 1998). Disasters open a window of opportunity in communities when, for a limited time, emergency management activity is propelled to the forefront of local prioritie s. During these windows of opportunity, the affected community is fertile ground for ambitious goal setting as priorities swing towards mitigation and prevention. Personal experiences provided additional motivation for community members to engage in discu ssions about mitigation planning, as was evident in the communities of Roseville and Napa. Napa, for example, experienced its eighth flood in thirteen years on to the disast hazard mitigation activities, primarily by getting local residents involved in planning. In Napa, the high level of public participation during all phases of planning was due in part to the existing social infrastructure built by the efforts of previous community based planning initiatives to reduce flood risk. Moreover, the most active community participants in the hazard mitigation planning process never thought of their activity as hazard mitigation. Ins tead,

PAGE 229

215 they described their participation in the planning process as a product of their deep investment in Napa Creek related issues. These community participants asserted that it was their personal responsibility to do their research and engage with profes sional planners and decision makers throughout the entire planning process. L ocal planning context does influence hazard mitigation plan quality. Although local social vulnerability factors did not affect plan quality outcomes, planning scale influenced the ability of local communities to customize their mitigation planning activity, engage local leadership, and, ultimately, create high quality plans. This study supports previous research that suggests that local mitigation plans prepared as part of multi jurisdictional planning efforts often reflect minimal effort from participating jurisdictions in establishing unique vulnerability assessments or locally relevant mitigation measures ( Schwab, 2011 ). Comprehensive planning requires a strong correspondence between popular priorities, local needs, and government policies. Ultimately, planning for disaster resilience is no different. While the case studies presented here are unrepresentative of other contexts, the themes and observations highlight new directi ons in which mitigation planning research must go in order to clarify a framework for resilience building. These questions include: Does the incorporation of social vulnerability data into risk and vulnerability assessments contribute to reduced local haz ard risk? What can we learn from multi jurisdiction mitigation planning environments where plan quality outcomes are high? and, Is plan quality an ideal benchmark for successful hazard mitigation planning? If not, then what metric is?

PAGE 230

216 These questions, among others, are discussed in greater detail in the following chapter.

PAGE 231

217 C HAPTER VI CONCLUSIONS AND DISCUSSION Introduction This chapter synthesizes the key findings of this study and presents their implications for local hazard miti gation planning under the Disaster Mitigation Act of 2000. The strengths and limitations of this study are considered as well as key contributions to the literature. The chapter concludes with a reflection on the achievements related to the objectives of t he study and presents d irections for future research. Building Disaster Resilience through Local Hazard Mitigation Planning term strategy to reduce disaster losses and hazard risk. Not on ly is their purpose to identify policies and actions that can be implemented to reduce risk and future loss, but the planning process itself creates a foundation for risk based decision making and knowledge sharing. The adoption of hazard mitigation plann ing as a cornerstone of U.S. emergency management was coupled with a turn towards local level risk reduction strategies. Disaster Mitigation Act in 2000 (DMA 2000), a strong emphasis was put on reducing loss and risk through state, county, and municipal hazar d mitigation planning efforts. Since the passage of DMA 2000, researchers have been working to identify the impacts of local hazard mitigation planning efforts on disaster risk, property loss, and

PAGE 232

218 vulnerability. Although measuring disaster prevention is a great challenge, empirical research has shown that high quality hazard mitigation plans have superior outcomes in terms of plan implementation and reduced vulnerab ility to hazards than do low quality plans (Burby et al., 1999; Brody, 2003; Godschalk et al., 2009; Beatley, 2009; Peacock et al., 2008). Additionally, plan quality studies grounded in the communicative planning literature have found that planning efforts that engage local stakeholders in participatory planning activities are the most effective at producing high quality plans and implementation outcomes (Bates & Peacock, 1 989; McEntire, 2005; Smith & We n g er, 2006; Godschalk et al., 1999; Innes & Booher 2010; Berke, Song, and Stevens 2009). The purpose of this study was to contribute to a better understanding of the role of public stakeholders during local hazard mitigation planning. An additional objective of this study was to examine how community ch aracteristics (including demographics and planning scale) affect both plan qua lity and public participation. This study evaluated the quality of thirty four local hazard mitigation plans in the Northern California Bay Delta region. It focused on assessing levels of public participation during local plan development and on identifying social vulnerability data in plans. This study measured hazard mitigation plan quality by scoring three equally weighted plan components included in a plan quality index. By i ncorporating social vulnerability metrics into hazard mitigation plan quality analysis, this study expands upon current plan quality assessment methodology to gauge how well local plans address the most at risk members of their communities As a method for assessing and categorizing levels of public participation during planning, this study applied a conceptual framework grounded in the social ecological

PAGE 233

219 systems literature. The framework refined by planning and social ecological systems sch olars Yli Pelkonen & Kohl (2005) brings together four planning models to illustrate a larger conceptual model of stakeholder participation during local hazard mitigation planning. This conceptual lens highlights the interactions through which stakeholder s participate in information production, goal setting and decision making during hazard mitigation planning (Callon, 199 9; Yli Pelkonen & Kohl, 2005). Ultimately, the results of this study offer new insight into the effects of public participation and loc al community characteristics on mitigation plan quality outcomes. Across the sample of thirty four plans assessed in this study, plan quality was consistently low with Fact Base scores scoring lowest across all three plan quality elements. The results of t he quantitative analysis also found that community characteristics did not affect plan quality outcomes, nor did they affect levels of public participation during planning. In addition to these results, planning scale emerged as a key dr iver of plan qualit y outcomes. This study is consistent with previous research and found that local mitigation plans prepared as part of multi jurisdictional planning efforts struggle to establish customized vulnerability assessments and mitigation strategies that are integ rated with local priorities and land use policies ( Schwab, 2011 ). Moreover, the findings from the qualitative phase of this study are consistent with previous research showing that higher quality plans are created in planning environments where there are s tronger interactions between public stakeholders, mitigation planning teams, and organizational stakeholders (Deyle & Slotterback, 2009). These results open the door to deeper, more focused

PAGE 234

220 evaluations of local hazard mitigation planning conventions, a num ber of which are discussed at the end of this chapter. Research Questions In response to steadily increasing hazard vulnerability and losses from disasters, mitigation planning in the U.S. has shifted towards a local approach to goal setting and action dev elopment. This shift is due, in part, to increasing recognition of the benefits associated with multi stakeholder knowledge sharing and with the integration of mitigation with broader community priorities and contexts. Very few empirical studies have exam ined hazard mitigation planning in detail to identify key factors associated with superior plan quality outcomes. This study investigated the nature of public participation during local hazard mitigation planning in thirty four California communities. It a lso identified key process related factors that drive variations in plan quality and content. Data collection and analysis was organized into three methodological stages: a quantitative evaluation of plan quality; a factor analysis of community characteris tics and plan quality; and a qualitative case study analysis. The key findings have been organi zed below by research question. Research Question 1 What is the quality of local hazard mitigation plans created in the Northern California Bay Delta region and to what extent has the public participated in the plan making process? Aim 1: To identify social vulnerability data in plans and incorporate that data into plan quality assessment metrics Aim 2: To assess the involvement of community stakeholders during the planning process

PAGE 235

221 Aim 3: To identify the participatory planning models employed by each city during the local hazard mitigation planning process Based on the results of the plan quality analysis, hazard mitigation plan quality was consistently low acr oss the sample of thirty four plans. On a plan quality scale of 0 30 points, the mean plan quality score of the sample was 6.76 points. The highest quality plan in the sample, created by Roseville, CA, had a score of 13.25 points. The lowest quality plan created by Morgan Hill, CA, had a plan quality score of 0.80 points. Additionally, the fact base scores were the lowest scoring plan quality components across the sample of thirty four plans included in this study. Very little information about local so cial vulnerability was included in the plans (as reflected by the results of the mitigation plan Fact Base scores). This suggests that contextual understandings and applications of local information were largely absent from the process of local hazard miti gat ion planning in the study area. Levels of public participation during planning were low across the sample frame. Most communities took consultative approaches to engaging their public stakeholder in mitigation planning and their public outreach activiti es were characterized by one way flows of information, flimsy surveys, and limited decision making power by citizens. Over seventy public participation mandate by holding one to two public meetings and/or by posting an electronic draft of their mitigation plan online for public review. Over half of those communities reported having no attendance at their meetings and received no feedback from the commun ity concerning the draft plan.

PAGE 236

222 Resea rch Question 2 How do community characteristics and levels of public participation affect mitigation plan quality outcomes? Aim 4: To identify key relationships between local community characteristics (including population size, wealth, diversity, social vulnerability characteristics) and local plan quality scores Aim 5: To analyze how community characteristics affect levels of public participation in l ocal hazard mitigation planning Aim 6: To evaluate how public participation in planning affects local hazard mitigation plan quality The quantitative phase of this study produced four key findings: 1) no relationship was found between community characteristics and plan quality scores, 2) a factor named Racial Disparity, which loads highly on Race: African American and Race: Native Hawai ian or Other Pacific Islander, was associated with higher levels of public participation during hazard mitigation planning, 3) no relationship was found between public participation and plan quality, and 4) planning scale was a key driver of plan quality a mong local communities. The results of the regression analysis demonstrated no associations between community characteristics and local plan quality scores (Aim 4). For example, gher (or lower) quality plans than communities with low levels of social vulnerability. The finding that plan quality did not differ between communities based on social vulnerability is important because it suggests that, despite limited planning resources and potentially divergent development proprieties, the most vulnerable communities in the study area were not systematically underser ved by local planning efforts.

PAGE 237

223 Although community characteristics did not influence plan quality as previously thought, m itigation planning activities in communities characterized by Racial Disparity (as represented by the two variables Race: African American and Race: Native Hawaiian and Other Pacific Islander) were associated with higher levels of public participation (Aim 5). It was not within the scope of this study to identify why certain stakeholders participate in hazard mitigation planning. However, future studies m ight explore the factors that explain increased participation in communities characterized by high level s of Racial Disparity in the c ontext of hazard vulnerability. In addition to satisfying Aims 4 and 5, the regression analysis also revealed that high levels of public participation did not drive mitigation plan quality outcomes in the study area, nor did t hey affect the Fact Base scores of local plans (Aim 6). In other words, public participation in planning did not lead to mitigation plans with greater amounts of social vulnerability data included in their risk and vulnerability assessments. These results are inconsistent with the initial hypotheses presented in this study. Further research is needed to clarify if/how communities can leverage local knowledge and social capital to reduce local hazard risk and vulnerability. Ultimately, Planning Scale emerged as a key driver of plan quality outcomes and was found to have the strongest relationship to plan quality than the other variables included in this study. Specifically, the single jurisdiction approach to local hazard mitigation planning was associated wi th higher plan quality scores. These findings (coupled with the results from the case studies) suggest that single jurisdiction planning environments have been better able to produce customized vulnerability assessments, to develop mitigation strategies th at are closely integrated with local priorities and land use policies, and to

PAGE 238

224 engage the local public in education, goal setting and risk assessment. Not only do these results contribute new information to the existing body of hazard mitigation research, t hey spur new questions related to practice and policy surrounding regional and multi jurisdiction approaches to lo cal hazard mitigation planning. Community Case Studies The quantitative phase of this study contradicted an initial hypothesis that public par ticipation contributes to higher quality mitigation plans. Additionally, the regression analysis exposed planning scale as a key driver of plan quality. The community case studies were conducted, in part, to clarify these results. Interviews and analysis w ere focused primarily on two understudied elements of local hazard mitigation planning: planning scale and public participation. In addition to evaluating the successes and failures of single and multi jurisdictional planning efforts, the stakeholder inter views helped to identify unobserved outcomes associated with high levels of public participation during local hazard mitigation planning. Ultimately, the evaluation of the six community case studies drew out a number of key themes related to local hazard m itigation planning in the study region. In case study cities with recent histories of hazard losses from disasters, higher levels of interaction and knowledge sharing occurred between organizational and community stakeholders during the hazard mitigation p lanning process. This is consistent with the existing literature and the factor analysis results, and can be explained by the theory of the post opportunity, mitigation activity is driven to the for efront of local priorities. Affected

PAGE 239

225 communities are more inclined to engage in ambitious goal setting and information sharing. Within the case study communities, personal experiences motivated members of the public to engage in discussions about mitigati on planning, as was evide nt in both Roseville and Napa. Although the quantitative results of this study suggest otherwise, t he models of participation that are utilized during local hazard mitigation planning matter to mitigation outcomes in a number of ways The case studies reveal that t he ways in which planning teams and orga nizational stakeholders engage with public stakeholders and share information play an important role in the ability of decision makers to garner public and political support for mi ti gation action. Additional ly, public stakeholders who are engaged in communicative planning activities have more readily implemented individual preparedness measures and helped to maintained local momentum for hazard mitigation and local risk reduction ac tivity. One of the key conclusions of this study is that the composition of the local hazard m itigation planning team matters. Based on the results of the quantitative analysis, local public participation did not affect plan quality scores. However, the c ase studies revealed superior planning outcomes in communities where there were strong interactions between the mitigation planning team, organizational stakeholders, and local community which attributed positive planning outcomes to environments where interactions A and B in the Framework of Stakeholder Participation During Planning were equally strong the Model 3 (M3) planning communities included in this study exhibited stronger loca l

PAGE 240

226 support for the implementation of risk reduction actions, increased community preparedness, and increased risk awareness. Figure 14 : Revised Framework for Local Hazard Mitigation Planning In communities with high levels of public participation and high levels of social v ulnerability information in their hazard risk assessments not only were a broad range of stakeholder groups invited to participate in the planning process, public stakeholders and community leaders were del egated as formal members of the mitigation planning team. In some cases, fifty percent of the local mitigation planning team was made up of members of the public. Figure 14, a revised conceptual framework for local hazard mitigation planning, offers an alt ernative approach to public participation one that supports communicative planning activity during local hazard mitigation planning regardless of planning scale. In both Roseville and Napa, mitigation planning was characterized by continuous interaction between the Mitigation Planning Team, local community stakeholders, and

PAGE 241

227 organizational stakeholders. Additionally, both communities took a delegated power approach to planning in which local community stakeholders were included as formal members of the Mit igation Planning Team. Unlike the conceptual framework presented in Chapter 2, this revised conceptual framework, interactions between local community stakeholders (vector B) are not conditional interactions. Rather they occur throughout the life of the planning project and consist of two way flows of information and knowledge sharing Additionally, the revised conceptual model situates local community stakeholders at the center of mitigation priority setting and decision makin g. As was exhibited in Roseville and Napa, when local community stakeholders represent their communities as formal members of the hazard mitigation planning team they are directly involved in decision making, priority setting, and definition development th rough a process of collaborative planning and knowledge sharing The revised conceptual framework provides local communities and planning regions with a useful approach to organizing their local hazard mitigation planning projects. Regardless of the juris dictional scale of the planning effort, Roseville and evidenced in the other case study communities. These effects included increased public support for mitigation pro jects, increased political salience of hazards, and the close integration of mitigation goals and actions with local priorities and land use policy. Ultimately, the richness and diversity of public stakeholders engaged in local mitigation planning plays an important, yet little understood role in determining local preparedness levels, public buy in, and political support for mitigation.

PAGE 242

228 The case studies also offered insight into why single jurisdiction planning environments led to higher quality plans than those produced in multi jurisdictional environments. First, those communities who took single jurisdiction approaches to mitigation planning were afforded significantly more control over the planning process, including goal setting, public outreach, and de velopment of their mitigation planning teams. This enabled the single jurisdiction communities to organize planning working groups that included local leaders and leveraged existing community relationships, to design community appropriate public outreach s trategies, and to align mitigation goals with existing com munity development priorities. The large majority of the multi jurisdictional planning efforts in the study region took a consultative approach towards public engagement. The outcomes of this plann ing approach included low public involvement, high dissatisfaction among the public with respect to the mitigation plan, and low Plan Quality scores. Although all of the multi jurisdictional plans in the study area met the minimum requirements for FEMA app roval, the case studies indicated that these plans were limited in their functionality due to their lack of integration with local planning priorities and a broader absence of publ ic support for implementation. When the Disaster Mitigation Act was passed in 2000, it represented a new approach to mitigation. Proponents emphasized its potential to boost local disaster resilience, mitigation capacity, and commitment through locally driven approaches to planning. Additionally, DMA 2000 was designed to facilita te coordination between State, Tribal, and local entities to better link their unique mitigation planning and implementation efforts. Although the multi jurisdiction plans analyzed in this study met

PAGE 243

229 FEMA approval requirements, they missed the mark in terms of meeting the goals of DMA 2000. These results have important implications for communities who are planning for future catastrophes and want to foster local disaster resilience through mitigation p lanning and capacity building. Contributions and Directi ons for Future Research One of the key goals of this study was to contribute to our understanding of how mitigation planning can play a role in reducing hazard impacts and social vulnerability. The findings from this study add to a growing body of literat ure on hazard mitigation plan quality and local disaster resilience building. This study confirms previous findings and contributes additional evidence that suggests public participation is beneficial to local hazard mitigation planning. Although public pa rticipation and local social vulnerability factors did not affect mitigation plan quality in the northern California Bay Delta region, the relevance of public engagement during planning is clearly supported by the results of this study. By taking a mixed methods approach, this study expands upon the existing body of literature and challenges conventional methodolo gical approach es to plan quality assessment While the quantitative phase of this study highlighted unseen relationships betwee n variables of interest, the case studies provided rich details about those relationships and clarified the reasons why those relationships emerged in the study region In some cases, the findings from the two phases of inquiry were closely aligned, in others they were not. For example, the quantitative analysis helped to identify a strong,

PAGE 244

230 positive relationship between plan quality and planning scale. Additionally, the q uantitative analysis revealed that models of participation did not play a role in determining plan quality outcomes. The case studies allowed for the scrutinization of these findings and revealed that although participation may not affect plan quality scor es, collaborative public engagement is incredibly valuable to mitigation planning in other ways. Additionally, the nuances between single and multi jurisdiction planning approaches were fleshed out by multiple conversations with planners, community member s, and technical experts in communities of interest. The insights and findings related to planning scale would not have been possible without the addition of a quantitative approach. By coupling quantitative document analysis with qualitative case studies, this study (and its results) underscores the importance of context and process on the relative success of local planning efforts. As a result, t his study brings in to question the exclusive use of plan quality as a barometer for successful planning, parti cularly in the context of hazard mitigation. A limitation of this study is that the numbers of local hazard mitigation plans and community stakeholders interviewed were relatively small. Notwithstanding this limitation the mixed methods approach may be a pplied to other communities and regions elsewhere in the United States. T he sample size is within the minimum ratio of cases to independent variables for a regression analys is. Therefore, similar results can be expected, even if the sample size were scaled up significantly. Despite its exploratory nature, this study contributes to new understandings of how to frame and measure the success of hazard mitigation plans. This work expands upon current plan quality assessment methodology to gauge how well local plans address

PAGE 245

231 the most at risk members of their communities. Despite acknowledgement of the importance of social vulnerability in the hazards and disaster literature and theory, previous research does not include social vulnerability in their hazard mitiga tion plan quality metrics. By incorporating social vulnerability concerns into the plan quality assessment, this study provides a new method for evaluating hazard mitigation plans that integrates theory r elated to disaster resilience. For the first time, this study extracted key factors that represent underlying relationships between local demographics and hazard mitigation plan quality. This study found that the local social vulnerability profiles of communities do not play a role in d etermining the quality outcomes of local hazard mitigation plans. Unexpectedly, socially vulnerable communities do not necessarily have lower quality plans. This is a bit of good news, as it contradicts hypothesis that suggest that poor, marginalized co mmu nities have a lower capacities for planning. Another result of this study shows that local social vulnerability is not statistically related to levels of participation during planning. Th is result has important implications for the potential for plan impl ementation in at risk communities It is critical that mitigation activities are implemented regularly in communities with elevated risk By definition, socially vulnerable communities are more at risk to hazards than others. Plan implementation outcomes a re positively associated with participation levels during planning. Therefore, it is important that socially vulnerable communities engage their publics in the mitigation planning process. This study indicates that public participation is low, regardless o f local social vulnerability, and highlights an area for concern and focus

PAGE 246

232 for state and counties looking to reduce the vulnerability of their most at risk jurisdictions. Ultimately, this study illustrates that communities in the study area had very limit ed understandings of their local social vulnerability to hazards, even while they were developing their mitigation priorities and actions. This is problematic from the disaster resilience perspective. Although social vulnerability is complex and often ecli psed by the risks posed by physical hazards, it is one of the largest threat s to local disaster resiliency. The body of research would benefit from additional analysis of plan quality and participation, specifically in socially vulnerable regions of the co untry. While this study d oes not confirm that public participation during mitigation planning contributes to higher quality plans, it d oes partially substantiate the hypothesis that public participation contributes to local resilience through capacity building and increased local commitment to mitigation. Ultimately, this study brings in to question the mentioned previously, the creation of high quality plans is not the end point in the mission towards reduced social vulnerability and losses from hazards. High quality hazard mitigations plans are an important intermediary step in the disaster risk reduction process for all communities. Ultimately, we are interested in an exte nsion of these outcomes, including social vulnerability reduction. The findings of this study bring to light sev eral areas for future research It was not within the scope of this study to test the relationship between the inclusion of social vulnerabilit y information in planning risk assessments and changes in local hazard risk over time. However a longitudinal study exploring the effects of social vulnerability

PAGE 247

233 assessments on reduced local hazard risk would contribute greatly to current evaluations Similarly, a l ongitudinal study of how public participation during hazard mitigation planning affects local social vulnerability over time would offer yet another way to evaluate planning success. Another future research direction has to do with regional versus single jurisdiction approaches to local mitigation planning. Hazards do no t adhere to jurisdictional boundaries especially those associated with climate change R egional planning efforts have great potenti al to reduce hazard risks, especially risks from hazards that have large geographic extents. Moving forward, it is important that region al planning environments contribute to reduced local hazard risk to the same degree as local planning efforts. Future re search may con sist of case studies within multi jurisdiction mitigation planning environments where plan quality o utcomes are high to better understand the processes inv olved and to identify best practices. If the opportunity arose to conduct this study ag ain, I would situate this research study somewhere in the southern United States. Although California is an example of a progressi ve planning environment with a history of ambitious approach es to hazard mitigation, it is far more affluent and less socially vulnerable than most other parts of the country. More study of the relationships between plan quality, social vulnerability, and participation is needed in areas of the country that are faced with significant social risks and cultural disparities. Perhaps the relationships identified in this study would be more pronounced within the most disadvantaged and marginalized areas of the United States. This study enhances our understanding of the current challenges facing local communities, planners and emergency managers related to local mitigation planning.

PAGE 248

234 Taken together, the findings suggest a role for public participation in promoting hazard risk reduction in local communities. This research will serve as a base for future studies as they explore mi tigation planning in the context of vulnerability reduction and resilience building. It also provides practitioners and local communities with a framework for managing and evaluating their ongoing mitigation planning activity.

PAGE 249

235 REFERENCES Adger, W.N. (2000). Soc ial and ecological resilience: a re they related? Progress in Human Geography. Vol. 24(3): 347 364. Adger, W.N. (2006). Vulnerability. Global Environmental Change, 16(3): 268 281. Adger, W.N., Hughes, T.P., Folke, C., Carpenter, S.R., Rockstrom J. (2005). Social ecological resilience to coastal disasters. Science, 309(5737):1036 1039. Alberti, M., Marzluff, J., Shulenberger, E., Bradley, G., Ryan, C., Zumbrunnen, C. (2003). Integrating humans into ecosystems: opportunities and challenges for u rban ecology. BioScience, 53(12): 1169 1179. Alterman, R., Hill M. (1978) Implementation of urban land use plans. Journal of the American Institute of Planners, 44(3): 274 285. Altshuler, A. (1965). The goals of comprehensive planning. Journal of the A merican Planning Association, 31: 186 95. Andrulis, D.P., Siddiqui N.J., Ganter, J.L. ( 2007 ). Preparing racially and ethnically diverse communities for public heath emergencies. Heath Affairs, 26(5): 1269 1279. Armitage, D., Berkes, F., Doubleday, N. ( 2007). Adaptive co management: c ollaboration, learning, and multi level governance. Vancouver: UBC Press. Arnstein, S. (1969). A ladder of citizen participation. Journal of the American Institute of Planners. 35: 216 224. Bacow, L.S., Wheeler, M. (1984). Environmental dispute resolution. New York, NY: Plenum Pub lishing Corp oration Bak, P. (1996). How nature works: The science of self organized criticality. New York, NY: Copernicus Press. Battista, F., Baas, S. ( 2004 ). The role of local institutions in reducing vulnerability to recurrent natural disasters and in sustainable livelihoods development. Consolidated report on case studies and workshop findings and recommendations. Rome: FAO, Rural Institutions and Participation Service. Beatley, T. (1998 ). The vision of sustainable communities: cooperating with n ature. Washington, DC: National Academy Press, Joseph Henry Press. Beatley, T., Manning, K. (1997). The ecology of place: p lanning for environment, economy, and community. Washington, DC: Island Press.

PAGE 250

236 Beatley, T. (2009). Pl anning for coastal resilience: b est practices for calamitous times Washington, DC: Island Press, Beierle, T., Cayford, J. (2002). Democracy in practice: public participation in environmental decisions. Washington, D.C.: Reso urces for the Future. Bengtsson, J., Angelstam, P., Elmqvist, T., Emanualsson, U., Fo lke, C., Ihse, M., Moberg, F., Nystron, M. (2003). Reserves, resilience and dynamic landscapes. Ambio, 32(6): 389 396. Berke, P.R. (2002). Does sustainable development offer a new direction for planning? Journal of Planning Literature, 17(1), 21 36. Berke, P.R., French, S. (1994). The influence of state planning mandates on local plan quality. Journal of P lanning Education and Research, 1 3(4), 237 250. Berke, P.R., Roenigk, D.J., Kaiser, E.J., & Burby, R.J. ( 1996). Enhancing plan quality: e valuating the role of state planning mandates for hazard mitigation. Journal of Environmental Planning and Management, 37(2), 155 169. Berke, P.R., Conroy, M. (2000). Are we plan ning for sustainable development? An evaluation of 30 comprehensive plans. Journal of the American Plan ning Association, 66(1): 21 33. Berke, P.R., Campanella, T.J. (2006). Planning for post disaster resiliency. The Annals of the American Academy of Poli tical and Social Science, 604:192. Berke, P., Song, Y., S tevens, M. (2009). Integrating hazard mitigation into new u rban and conventional developments. Journal of Planning Education and Research, 28: 441 455. Berkes F., Turner, N.J. (2006). Knowledge, learning, and the evolution of conservation practice for social ecological system resilience. Human Ecology, 34, 479 494. Berkes, F., Folke, C. (2002). Back to the future: e cosystem change, institutions and local knowledge. In L. H. Gunderson and C S. Holling (Eds.), Panarchy: understanding yransformations in human and natural s ystems. Washington, D.C.: Island Press. Berkes, F., Colding, J. (1998). Ecological practices and social mechanisms for building resilience and sustainability. In F. Berkes and C. Folke (Eds.), Linking social and ecological systems: management practices and social mechanisms for building resilience (pp. 414 436). Cambridge: Cambridge University Press.

PAGE 251

237 Ber kes, F., Colding, J., Folke, C. (E ds. ) (2003). Navigating social ecologica l systems: Building resilience for complexity and change. Cambridge, UK: Cambridge University Press. Berkes, F., Folke, C. (2001). Back to the future: ecosystem dynamics and local knowledge. In: Gunderson, L., Holling, C.S. (eds.). Panarchy. Washington, DC: Island Press. Berkes, F., Folke, C. (eds) (1998). Linking sociological and ecological systems: management practices and social mechanisms for building resilience. New York, New York: Cambridge University Press. Birkmann, J. (2006). Measuring vulner ability to natural hazards. New York, NY: United Nations University Press, USA. Black, A. (1968). The comprehensive plan: principles and practice of urban p lanning. Washington, DC: International City Management Association. 349 78. Blaikie, P., Brookfield, H.C (1987). Land degradation and s ociety. Methuen, London. Borden, K., Emrich, C., Schmidtlein, M., Cutter, S., Piegorsch, W. (2007). Vulnerability of US cities to environmental hazards. Journal of Homeland Security and Emergency Management 4(2), 5. Bouwer, L.M. ( 2011 ) Have disaster losses increased due to anthropogenic climate change? Bulletin of the American Meteorological Society, 92(1): 39 46. Brody, S.D. (2003). Measuring the effects of stakeholder participation on the quality of loca l plans based on the principles of collaborative ecosystems management. Journal of P lanning Education and Research, 22: 407 419. Brody S D Godschalk D R Burby R J (2003) Mandating citizen participation in plan making: six strategic planning choices. Journal of the American Planning Association, 69(3):245 264. Bronfenbr enner, U. (1979). The ecology of human development: experiments by nature and d esign. Cambridge, MA: Harvard University Press. rke, T.D., Reinhorn, A.M., Shinozuka, M., Tierney, K., Wallace, W.A., and von Winterfeldt, D. (2003). A framework to quantitatively assess and enhance the seismic resilience of communities. Earthquake Spectra, 19(4): 733 752. Bullock, R., Armitage, D., Mi tchell, B. (2008). Refocusing on learning for resilient forest based communities. Paper presented at The Virginia Tech Symposium on

PAGE 252

238 Enhancing Resilience to Catastrophic Events through Communicative Planning. Blacksburg, VA, November, 18 20. Burby, R. J., Beatley, T, Berke, P.R., Deyle, R.E, French, S.P., Godschalk, D.R., Kaiser, E.J., Kartez, J.D., May, P.J., Olshansky, R., Paterson, R.G. and Plat t, R.H. (1999). Unleashing the power of planning to create disaster resistant c ommunities. Journal of the Ame rican Planning Association 65 (3): 247 258. Burby, R.J. (2003). Making plans that matter: Citizen involvement and government action. Journal of American Planning Association, 69, 33 49. Burby, R.J., May, P., Berke, P.R., Dalton, L., French, S., & Kaiser, E. (1 997). Making government plans: s tate experiments in managing land use. Baltimore: Johns Hopkins University Press. Burke, E.M. (1979). A participatory approach to urban planning. New York, NY: Human Science Press. Bates, F.L., Peacock, W.G. (1989) Long term recovery. International Journal of Mass Emergencies and Disaster 7(3): 349 365. Berke, P., Ericksen, N., Dixon, J. (1997). Coercive and cooperative intergovernmental mandates: A comparative analysis of Florida and New Zealand environmental pl ans. Environment and Planning B: Planning and Design, 24: 451 468. Berke, P.R., Godschalk, D. (2009). Searching for the good plan: A meta analysis of plan quality studies. Journal of Planning Literature, 23(3): 227 240. Berke, P.R., Smith, G.P. (2009). Hazard mitigation planning and disaster resiliency: Challenges and strategic choices for the 21 st Century, in Planning the risk: Spatial planning as a strategy for mitigation and adaptation to natural hazards. Amsterdam, NL: ISO Press. Berekes, F., J. Co lding, Folke, C. (2003). Navigating social ecological systems: Building resilience for complexity and change. Cambridge, UK. Cambridge University Press. Borden, K., Emrich, C., Schmidtlein, M., Cutter, S., Piegorsch, W. (2007). Vulnerability of US cities to environmental hazards. Journal of Homeland Security and Emergency Management 4(2), 5. Burby, R.J. (2005). Have state comprehensive planning mandates reduced insured losses from natural hazards? Natural Hazards Review 6(2): 67 81.

PAGE 253

239 Burby, R.J. (2006) Hurricane Katrina and the paradoxes of government disaster policy: Brining about wise governmental decisions for hazardous areas. Annals of the American Academy of Political and Social Sciences 604(1): 171 191. Burby, R.J., Arthur, C.N., Thomas, W.S. (2006). The problems of containment and the promise of planning. In: Birch, E.L., Wachter, S.M. (eds) Rebuilding urban places after disaster: Lessons from hurricane Katrina. University of Pennsylvania Press, Philadelphia. Burby, R.J., Beatley, T., Berke, P.R., Deyle, R.E., French, S.P., Godschalk, D.R., Kaiser, E.J., Kartez, J.D., May, P.J., Olshansky, R., Paterson, R.G., Platt, R.H. (1999). Unleashing the power of planning to create disaster resistant communities. Journal of the American Planning Associa tion 65(3): 274 258. Burby, R.J., French, S.P., Nelson, A.C. (1998). Plans, code enforcement, and damage reduction: Evidence from the Northridge earthquake. Ear thquake Spectra 14(1): 59 74. Burby, R., May, P., Berke, P., Dalton, L., French, S., Kaiser, E. (1997). Making governments plan: State experiments in managing land use Baltimore: Johns Burton, I., Huq, S., Lim, B., Pilifosova, O., Schipper, E.L. (2002). From impacts assessment to adaptation priorities: the shaping of adaptation policy. Climate Policy, 2(2 3): 145 159. Callon, M. (1999). The role of lay people in the production and dissemination of scientific knowledge. Science, Technology and Society, 4: 81 94. Campbell, S. (1996). Green cities growing cities, just cities? u rban planning and the contradictions of sustainable development. Journal of the American Planning Association, 62. Carafano, J (2008). Risk and resiliency: d eveloping the right homeland security public policies for the post Bush era. Testimony before the Sub Committee on Homeland Security United States House of Representatives. June 24, 2008. Carbonell, A ., Meffert, D. (2009). Climate change and the resilience of New Orleans: the adaptation of deltaic urban f orm. The Lincoln Institute of Land Policy. Carpe rnter, S.R., Walker, B., Anderies, J.M., Abel, N. (2001). From metaphor to measurement: r esilience of what to what? Ecosystems, 4: 765 781. Carr, D.S., Selin, S.W., Schuett M.A. (1998). Managing public forests: understanding the role of collaborative planning. Environmental Management, 22: 767 776.

PAGE 254

240 Cattell, R.B. (1978). The scientific use of factor analysis in behavioral and life sciences. New York: Plenum. Centers for Disease Control and Prevention and Agency for Toxic Substances an d Disease Registry (CDC/ATSDR) (2008) The CDC/ATSDR Public Health Vulnerability Mapping System. National Center for Environmental Health (NCEH) Office of Terrorism Preparedness and Emergency Response (OTPER). Chadwick, G. (1971). A systems view of planning: toward a theory of the urban and regional planning process. New York: Pergamon. Chapin, F.S., Kaiser, E.J. (1979). Urban land use planning. Urbana: University of Illinois. Clavel, P. ( 1994). The evolution of advocacy planning. Journal of the American Planning Association. 60(2): 146 149. Coburn, J. ( 2003). Bringing local knowledge into environmental decision making: improving urban planning for communities at risk. Journal of Planning Education and Research, 22(4): 420 433. Cohen, M.D., March, J.G., Olsen, J.P. (1972). A garbage can model of organizational choice. Administrative Science Quarterly, 17(1), 1 25. Colten, E.C., Sumpter, A.R. (2009). Social memory and resilience in New Or leans. Natural Hazards, 48(3): 355 364. Comfort, L., Wisner, B., Cutter, S., Pulwarty, R., Hewitt, K., Oliver Smith, A., Wiener, J., Fordham, M., Peacock, W., Krimgold, F. (1999). Reframing disaster policy: the global evolution of vulnerable communities. Environmental Hazards. 1: 39 44. Creighton, J.L. (1992). Involving citizens in community decision making: A guidebook. Washington, DC: Program for Community Problem Solving. Cutter, S.L. (1996) Vulnerability to environmental h azards. Progress in Human Geography 20(4): 529 539. Cutter, S.L., Emrich, C.T. ( 2006 ). Moral hazard, social c atastrophe : the changing face of vulnerability along the hurricane coasts. The Annals of the American Academy of Political and Social Science, 604(1): 102 112. Cutter, S.L., Barnes, L., Berry, M., Burton, C., Evans, E., Tate, E., Webb, J. (2008). A place based model for understanding community resilience to natural disasters. Global Environmental Change, 18: 598 606.

PAGE 255

241 Cutter, S.L., Boruff B. J., Shirley, W.L. (2003). So cial vulnerability to environmental hazards. Social Science Quarterly, 84(2): 242 261. Daley WR, Brown S, Archer P, Kruger E, Jordan F, Batts D, and Mallonee S. 2005. Risk of tornado related death and injury in Oklahoma, May 3, 1999. Am J Epidemiol 161(1 2):1144 50. Dalton, L.C, Burby, R.T. (1994). Plans can matter! The role of land use plans and state planning mandates in limiting development of hazardous areas. Public Administration Review. 54(3): 229 238. Danter, K.J., Griest, D.L., Mullins, G.W., Norland, E. (2000). Organizational change as a component of ecosystem management. Society and Natural Resources, 13: 537 547. Davidoff, P. (1965). Advocacy and pluralism in planning. Journal of the American Institute of Planners. 31(4): 331 338. Day, D. (1997). Citizen participation in the planning process: An essentially contested concept? Journal of Planning Literature, 11(3), 421 434. Day, J.C., Thomas, G., Tanis M.F. (2003). Toward environmental sustainability in British Columbia: The Role of collabo rative planning. Environments, 31(2): 5 19. De Souza, R M. ( 2004 ) change. Population Reference Bureau. Available at: http://www.prb.org/Articles/2004/InHarmsWayHurricanesPopulationTrendsandE nvironmentalChange.aspx. Department of Homeland Security (2010). Quadrennial Homeland Security Review report: A strategic framework for a secure homeland. Washington, DC. Deyle, R., Slotterback, C.S. (2009). Group learning in participatory planning processes: An exploratory quasi experimental analysis of local mitigation planning in Florida. Journal of Planning Education and Research, 29, 2 3 38. Dovers, S. (2001). Informing institutions and policy. In Venning, J., and Higgins, J., (eds). Toward sustainability: emerging systems for informing sustainable development. Sydney: University of NSW Press. Drabek, T.E. (1991). The evolution of eme rgency management, in Drabek, T.E, Hoetmer, G.J. (eds), Emergency management: p rinciples and practice for local government. Washington, DC: International City Management Association. Dresner S. (2002). The principles of s ustainability. London: Earthscan Publications Ltd.

PAGE 256

242 Duffy, D.M., Doselend, M., Gunton, T.I. (1996). A preliminary assessment of shared decision making in land use and natural resource planning. Environments, 23(2): 1 16. Economic Opportunity Act of 1964, United States Public Law 88 452 title II. Erlandson, D., Harris, E., Skipper, B., Allen, S. (1993). Doing naturalistic inquiry: A guide to methods. Newbury Park, CA: Sage Publications. Eidson MJ, Lybarger A, Parsons JE, MacCormack JN, Freeman JI. 1990. Risk factors for tornado inju ries. International Journal of Epidemiology, 19(4):1051 56. Fainstein, N.I., Fainstein, S.S. (1985). Citizen participation in local government. In D.R. Judd (Ed.), Public policy across states and communities (pp. 223 238). Greenwich, CT: Jai Press. Fainstein, S.S. (2002). New directions in planning theory. Urban Affairs Review, 35: 451. F ederal E mergency M anagement A gency, FEMA (2001). Concept papers for section 322 of the Stafford Act as Amended by the Disaster Mitigation act of 2000. Washington, D.C. Fisher, R., Ury, E. (1981). Getting to Yes: Negotiating agreement without giving in. 1st ed. New York: Penguin Books. Folke, C., Carpenter, S.R., Elmqvist, T., Gunderson, L., Holling, C.S., Walker, B. (2002). Resilience and sustainable development: building adaptive capacity in a world of transformations. Ambio, 31: 437 40. Folke, C., Hahn, T., Olsson, P., Norberg, J. (2005). Adaptive governance of social ecological systems. Annual Review of Environmental Resources, 30: 441 73. Folke, C. (2006). Re silience: the emergence of a perspective for social ecological systems analyses. Global Environmental Change, 16(3): 253 267. Forester, J. (1982). Planning in the face of power. Journal of the American Planning Association, 48(1): 67 80. Forester, J. (19 99). Challenges of mediation and deliberation in the design professions: Practice stories from Israel and Norway. Journal of Architectural Planning and Research, 16(2), 116 132. _______. (1973). Retracking America: A theory of transactive planning. New Y ork: Anchor Press.

PAGE 257

243 Foster, H. (1995). Disaster mitigation: the role of resilience. In D. Etckin (ed.) Proceedings of a Tri Lateral Workshop on Natural Hazards, Merrickwille, Ontario, Canada, Feb 11 14, p 93 108. Fraser, E. (2003). Social vulnerability a nd ecological fragility: building bridges between social and natural sciences using the Irish Potato Famines and as a case study. Conservation Ecology, 7. Gallopin, G.C. (2006). Linkages between vulnerability, resilience, and adaptive capacity. Global Env ironmental Change, 16: 293 303. Gans, H.J. (1962). The urban villagers: Groups and class in the life if Italian Americans. New York: Free Press of Glencoe. 2nd Ed. New York, NY: McGraw Hill Companies Inc. Godschalk, D.R. (2003). Urban hazard mitigation: c reating resilient cities. Natural Hazards Review, August 2003. Godschalk, D.R., Mills, W. (1966). A collaborative approach to planning through urban activities. Journal of the American Institute of Planners, 32, 86 95. Godschalk, D.R., Stiftel, B. (1981). Making waves: public participation in state water planning. Journal of Applied Behavioral Science, 17(4): 597 614. Godschalk, D.R., Parham, D.W., Porter, D.R., Potapchuck, W.R., Schukraft, S.W. (1994). Pulling together: A planning and development consensus building manual. Washington, DC: Urban Land Institute. Godschalk, D.R., Kaiser, E.J., Berke, P.R. (1998). Hazard assessment: the factual basis for planning a nd mitigation, In Burby, R. (ed.), Cooperating with na ture: c onfronting natural hazards with land use planning for sustainable communities. Washington, DC: Joseph Henry. Godschalk, D., Kaiser, E., Berke, P. (1998). Integrating hazard mitigation and local land use planning. C ooperating with nature. Washington, DC: National Academy Press, Henry Joseph Press. Godschalk, D., Beatley, T., Berke, P., Brower, D., Kaiser, E. (1999). Natural hazard mitigation: r ecasting disaster policy and planning Washington, D.C.: Island Press. Godschalk, D.R., Brody, S., Burby, R. (2003). Public participation in natural hazard mitigation policy formation: Challenges for comprehensive planning. Journal of Environmental Planning and Management, 46(5), 733 754.

PAGE 258

244 Godschalk D., Rose., A., Mittler, E., Porter, K., and Taylor, C.W. (2009). Estimating the value of foresight: Aggregate analysis of natural hazard mitigation benefits and costs. J. Environ, Plann. Mgmt, 52(6), 739 756. Goldstein, B.E. (2008). Skunkworks in the embers of the cedar fire: enhancing resilience in the aftermath of disaster. Human Ecology, 36: 15 28. Goldstein, B.E. (2009). Resilience to surprises through communicative planning. Ecology and Society, 14(2): 33. Gorsuch, R.L. (1983). Factor anal ysis (2nd ed.). Hillsdale, NJ: Erlbaum. Guenno, G., Tizzi, S. (1998). The index of sustainable economic welfare (ISEW) for Italy. Nota Di Lavoro vol. 5.98. Gunderson, L.H., Holling, C.S., Light, S.S. (1995). Barriers and bridges to the renewal of ecosyst ems and institutions. New York: Columbia University Press. Gunderson, L. (1999). Resilience, flexibility and adaptive management: antidotes for spurious certitude? Conservation Ecology, 3(1): 7. Gunderson, L., Holling, C.S. (2001). Panarchy; understandi ng transformations in systems of humans and nature. Washington, DC: Island Press. Gunton, T., Day, J.C. (2003). The theory and practice of collaborative planning in resource and environmental management. Environments, 31(2): 5 19. Habermas, J. (1970). To ward a rational society. Boston, MA: Beacon. ________. (1979). Communication and the evolution of society. Boston, MA: Beacon. ________. (1981). The theory of communicative action: Reason and the rationalization of society (T. McCarthy, Trans.). Boston, MA: Beacon Press. ________. (1989). The theory of communicative action. Lifeworld and system: A critique of functionalist reason (T. McCarthy, Trans .). Boston, MA: Beacon Press. Hall, P. (2002). Urban and regional planning. Third Edition. New York, NY: Routledge. Hanna, S., Folke, C., Mler, K G. (E ds.) (1996). Rights to nature. Washington D.C.: Island Press. Hart, M., Mazzotta, M., Kellman, K. (1999). Measuring sustainability: An assessment of criteria for defining and selecting sustainability indicators. Montpelier, VT: Green Mountain Institute for Environmental Democracy.

PAGE 259

245 Hawley, A. (1950). Human Ecology: A Theory of Community Structure. New York: Ronald Press. Healey, P. (1998). Collaborative planning in a stakeholder society. The Town Planning Review, 69(1), 1 21. Healey, P. (2002). Planning through debate: The communicative turn in planning theory. Town Planning Review, 63(2), 143 162. _______. (1997). Collaborative planning: Shaping places in fragmented societies. London: Macmillan. Hester, R. (1996). Wanted: p articipation with a view. In Public and private places, Jack Nasar and Barbara Brown, eds. Edmond, OK: Environmental Design Res earch Association. Holden, M. (2008). The tough minded and the tender minded: a pragmatic turn for sustainable development planning and policy. Planning Theory and Practice, 9:475 96. Holling, C.S. (1961). Principles of insect predation. Annual Review o f Entomology, 6: 163 182. Holling, C.S. (1973). Resilience and stability of ecological systems. Annual Review of Ecology and Systematics. 4: 1 23. Holling, C.S. (1986). The resilie nce of terrestrial ecosystems: l ocal surprise and global change, in Clark, W.C., Munn R.E. (eds.) Sustainable development of the biosphere. Cambridge, U.K.: Cambridge University Press. Holling, C.S., Sanderson, S. (1996). Dynamics of (dis)harmony in ecological and social systems. In: H anna, S., Folke, C., Maler, K. (eds) Rights to nature: Institutions for humans and the environment. Washington, DC: Island Press. Holling, C.S. (2001). Understanding the complexity of economic, ecological, and social systems. Ecosystems, 4: 390 405. Hom eland Security, (2010). Quadrennial homeland security review: A strategic framework for a secure homeland. Department of Homeland Security, Washington, D.C. February, 2010. Hopkins, L.D. (2001). Urban development: The logic of making plans. Washington, D C: Island Press.

PAGE 260

246 Houlihan, A.F. (2007). Achieving sustainability through disaster resistance: the fourth component. Paper presented at the annual meeting of the Midwest Political Science Association, Palmer House Hotel, Chicago, IL. (2010, July 20). Huq S. (2007). Community based adaptation: An IIED Briefing Note. London: International Institute for Environment and Development. Innes, J.E. (1996). Planning through consensus building: A new view of the comprehensive planning ideal. Journal of the Ameri can Planning Association, 62, 460 472. Innes, J.E. (1998). Information in communicative planning. Journal of the American Association of Planners, 64, 52 63. Innes, J.E., Gruber, J., Nueman, M., Thompson, R. (1994). Coordinating growth and environmental management through consensus building. CPS Brief, 6(4), 1 8. Innes, J.E., Booher, D.E. (1999 a ). Consensus building as role playing and bricolage: Toward a theory of collaborative planning. Journal of the American Planning Association, 62(4), 460 472. Innes, J.E., Booher, D.E. (1999b). Consensus building and complex adaptive systems. Journal of the American Planning Association, 65(4), 412 423. Innes, J., Booher, D. (20 10). Planning with complexity: a n introduction to collaborative rationality for pub lic policy. New York: Routledge. Jackson, T., Marks, N., Ralls, J., Strymne, S. (1997). An index of sustainable economic welfare for the UK 1950 1996. Centre for Environmental Strategy, University of Surrey: Guildford. Jepson, E.J. (2004). Human natur e and sustainable development: a strategic challenge for planners. Journal of Planning Literature, 19(1): 3 15. Jepson, E.J. (2001). Sustainability and planning: diverse concepts and close associations. Journal of Planning Literature, 15(4): 499 510. Kach igan, S.K. (1991). Multiv ariate statistical analysis: a conceptual i ntroduction, 2nd Edition. New York, New York. Radius Press. K aiser, E.J., Godschalk, D.R., Chapin, S.F. (1995). Urban land use planning. 4th ed. Urbana, IL: University of Illinois Press. Kasperson, R. E., Renn, O., Slovic, P., Brown, H., Emel, J., Gob le, R., Kasperson, J. X., Ratick, S. (1988). The social amplification of risk: a conceptual framework. Risk Analysis, 8: 177 187.

PAGE 261

247 Kasperson, J.X., Kasp erson, R.E., Turner II, B.L., (E ds.) (1995). Regions at risk: Comparisons of threatened environments. Tokyo: United Nations University Press. Kasperson, J.X., Kasperson, R.E. (2001). International workshop on Vulnerability and Global Environmental Change, May 17 19, 2001. Stockholm Env ironmental Institute (SEI), Stockholm, Sweden: A workshop summary. SEI Risk and Vulnerability Report 2001 01. Stockholm Environment Institute, Stockholm, Sweden. Kasperson J.X., Kasperson, R.E., Turner, B.L., Schiller, A., Hsieh, W. (2003). The human dimensions of global environmental change Diekmann, A., Dietz, T., Jaeger, C., Rosa, E.S. (E ds. ). Cambridge, MA: MIT Press. Kates, R.W., Ausubel, J.H., Berberian, M. (E ds.) (1985). Climate impact assessment: Studies of the interaction of climate and soc iety. New York, NY: John Wiley. Kates, R.W., Clark, W.C., Corell, R., Hall, J.M., Jaeger, C.C., Lowe, I., McCArthy, J.J., Schellnhuber, H.J., Bolin, B., Dickson, N.M., Faucheux, S., Gallopin, G.C., Grubler, A., Huntley, B., Jager, J., Jodha, N.S., Kaspers on, R.E., Mabogunje, A., Sustainability science. Science, 292: 641 642. Keiner, M. (2005). History, definition(s) and models of sustainable development. ETH e Collection. (2009, De c. 10). Kent, T.J. (1964). The urban general p lan. San Francisco: Chandler Publishing Company. 1991. Second Edition. Chicago, IL: Planners Press, American Planning Association. Klein, R.J.T., Nicholls, R.J., Thomalla, F. (2003). Resilience to natural h azards: how useful is this concept? Environmental Hazards, 5: 35 45. Knapp G., Nelson A.C. (1992) The regulated landscape: lessons on state land use planning from Oregon. Lincoln Institute of Land Policy, Cambridge, MA. Leach, W.D., Pelkey, N., Sabatier, P. (2002). Stakeholder partnerships as collaborative policymaking: evaluation criteria applied to watershed management in California and Washington. Journal of Policy Analysis and Management, 21(4): 645 670. Laurian, L., Day, M., Backhurst, M., Berke, P., Ericksen, N., Dixon, J., Chapman, S. (2004). What Drives Plan Implementation? Plans, Planning Agencies and Developers. Journal of Environmental Planning and Management, 47 (4): 555 577.

PAGE 262

248 Levin, S.A. (1999). Towards a science of ecological manag ement. Conservation Ecology, 3(2): 6. Lewis Beck, M., Bryman, A., Liao, T.F. (2004). Analysis of Covariance (ANCOVA). The SAGE Encyclopedia of Social Science Research Methods, Retrieved October 17, 2011, from http://srmo.sagepub.com/view/the sage encyclopedia of social science research methods/n14.xml Lewontin, R.C. (1969). The meaning of stability : d iversity and stability of ecological systems. B ro okhaven Symposia in Biology no. 22. Brookhaven, New York. Lincoln, Y.S., Guba E.G. ( 1985 ). Naturalistic inquiry. Beverly Hills, CA: Sage Lindblom, C.E. (1956). The science of muddling through. Public Affairs Review, 19: 79 88. Lindblom, C.E., Braybrooke, D. (1963). A strategy of decision: p olicy evaluation as a social process. Litchfield, N.P., Kettle, P., Whitebread, M. (1975). Evaluation of the planning process. Oxford, UK: Pergamon. Locke, C., Adger, W.N., Kelly, P.M. (2000). Changing environmental consequences. Environment, 42(7): 24 35. Lowry, K., Adler, P., & Milner, N. (1997). Participating the public: Group process, politics, and planning. Journal of Planning and Education Research, 16(3), 177 187. Manyena, S.B. (2006). The concept of resilience revisited. Disasters, 30(4): 433 450. Margerum, R.D. ( 2002). Collaborative planning: b uilding consensus and building a distinct model for practice. Journal of Planning Education and Research, 21, 237 253. Margerum, R.D., Whitall, D. (2004). The challenges and implications of collaborative management on a river basin scale. Journal of Environmental Planning and Management, 47(3), 407 427. Margerum, R.D. (2008). A typology of collaboration efforts in environmental management. Environmental Management, 41, 487 500. May, R.M. (1972). Will a large complex ecosystem be stable? Nature, 238: 413 414. McEntire, D.A. (2005). Why vulnerability matters. D isaster Prevention and Management 14(2): 206 222.

PAGE 263

249 McE ntire D.A., Fuller, C., Johnston, C.W., Weber, R. (2002). A comparison of disaster paradigms: the search for a holistic policy guide. Public Administration Review, 62(3): 267 281. McGuirk, P.M. (2001). Situating co mmunicative planning theory: c ontext, power, and knowledge. Environment and Planning, 33, 195 217. McLoughlin, J.B. (1969). Urban and regional planning: a systems approach. London: Faber. Merkhofer, M.W., Conway, R., Anderson, R.G. (1997). Multiattribute utility analysis as a framework for public participation in sit ing a hazardous waste facility. Environmental Management, 21(6), 831 839. Mileti, D (1999). Disasters by design: a r eassessment of natural hazard mitigation and land use planning. Washington, DC: Joseph Henry Press. Miles M. B., Huberman A.M. ( 1994 ). Qualitative data analysis, 2 nd Ed. Newbury Park, CA: Sage. Moore, C.N. (1995). Participation tools for better land use planning: t echniques and case studies. Sacramento, CA: Local Government Commission. Moynihan, D.P. (1969). Maximum feasible misunderstanding: c ommunity action in the war on poverty. New York: Free Press. Morrow BH. ( 1999 ) Identifying and mapping community vulnerability. Disasters 23(1)1 18. Nelson, A.C., French S. (2002). Plan quality and mitigating damage from natural disaster. Journal of the American Planning Association, 68(2): 194 207. Neuman, M. (1998a). Does planning need the plan? Journal of the American Planning Association, 64(2): 208 20. ., Venema, H., Aandahl, G., Tompkins, H., Javed, A., Bhadwal, S., Barg, S., Nygaard, L., West, J. (2004). Mapping vulnerability to multiple stressors: climate change and globalization in India. Global Environmental Change, 14(4): 303 313. Ollenburger J.C., Tobin G.A. (1998). Women on postdisaster stress. E. En arson, B.H. yes, Greemwood Press, Westport, CT, 95 107.

PAGE 264

250 Ostrom, E. (1990). Governing the commons: t he evolution of institutions for collective action. Cambridge, UK: Cambridge University Press. O uthwaite, W. (1994). Habermas: a critical introduction. Cambridge: Polity Press. Paton, D., Johnston, D. (2001). Disasters and communities: vulnerabilities, resilience, and preparedness. Disaster Prevention and Management, 10(4): 270 277. Paton, D., Smith, L., and Violanti J. ( 2000 ). Disaster response: risk, vulnerability and resilience. Disaster Prevention and Management, 9(3): 173 180. Paton, D., Johnston, D (2006). Disaster resilience: a n integrated approach. Springfield, IL Pastor, M., Bullard, R., Boyce, J., Fothergill, A., Morello Frosch, R., Wright, B. (2006). In the wake of the storm: environment, disaster and race after Katrina. New York: Russell Sage Foundation. Peacock, W.G., Kunreuther, H., Hooke, W.H., Cutter, S.L., Chang, S.E., and Berke, P.R. (2008). Toward a resiliency and vulnerabil ity observatory network: RAVON. HRRC reports: 08 02R. Pearce, L. (2003). Disaster management and community plann ing, and publi c participation: h ow to achieve sustainable hazard mitigation. Natural Hazards, 28: 211 228. Pedhazur, E.J., Schmelkin, L.P. (1991). Measurement design, and analysis: a n integrated approach. Hillsdale, NJ: Erlbaum. Peek Asa C., Ramirez, M., Seligson, H., Shoaf, K. ( 2003 ) Seismic, structural, and individual factors associated with earthquake related injury. Inj ury Prev ention, 9(1):62 66. Pelling, M. (2003). T he Vulnerability of cities: n atural disasters and social resilience. London: Earthscan. Pelling, M. ( 1997 ). Learning from others: scope and challenges for participatory disaster risk assessment. Disaters, 31(4): 373 385 Peterson, G.D., Allen, C.R., Holling, C.S. (1998). Diversity, ecological function, and scale: resilience within and across scales. Ecosystems, 1:6 18. Phillips, B., Thomas, D.S.K., Fothergill, A., Blinn Pike, L. (2009). Social Vulnerability to Disasters, CRC Press.

PAGE 265

251 Piegorsch, W., Cu tter, S., Hardisty, F. (2007). Benchmark analysis for quantifying urban vulner ability to terro rist incidents. Risk Analysis, 27: 1411 1425. Plummer, R., Armitage, D.R. (2007). Charting the new territory of adaptive co management: a Delphi study. Ecology and Society, 12(2): 10. Prescott Allen, R. (20 01). The Wellbeing of Nations: a country by country index of quality of life and the environment. Island Press. Quarantelli, E.L. (1980). Evacuation behavior and problems: Findings an d implications from the research literature. Disaster Research Center Miscellaneous Report, 27. Riebs ame, W.E., Changnon Jr., S.A., Carl, T.R. (1991). Drought and natural resource ma nagement in the United States: i mpacts and implications of the 1987 89 drought. Boulder, Colorado: Westview Press. Roling, N.G., Wagemakers, M.A.E. (E ds.) (1998). Facilitati ng sustainable agriculture: participatory learning and adaptive management in times of environmental uncertainty. Cambridge: Cambridge University Press. Rose, A., Liao, S.Y. (2005). Modeling regional ec onomic resilience to disaster: a computable general equilibrium model of water service disruptions. Journal of Regional Science, 48(1): 75 112. Rosenzweig, M.L. (1971). Paradox of enrichment: destabilization of exploitation ecosystems in ecological time. Science, 171: 385 387. Scheffer, M., Brock, W., We stley, F. (2000). Socioeconomic mechanisms preventing optimum use of ecosystem services: an interdisciplinary theoretical analysis. Ecosystems, 3:451 471. Scheffer, M., Carpenter, S.R., Foley, J., Folke, C., Walker, B. (2001). Catastrophic shifts in ecosy stems. Nature, 413:591 596. Schwab A.K., and Brower, D.J. (2008). Increasing resilience to natural hazards: o bstacles and opportunities for local governments under the disaster mitigation act of 2000. Environmental Law Review, News and Analysis, 38:10171. Schwab, J.C. (Ed.) (2011 ). Hazard mitigation: integrating best practices into planning. American Planning Association. Planning Advisory Service, Report Number 560. Scruggs, P. (1993). Chapter 1: Definitions and principles. G uidelines for state level sustainable development, S. Park. (Ed.) Chapel Hill: Center for Policy Alternatives and Environmental resource Program, University of North Carolina. Sen, A. (1981). Poverty and Famines. Oxford University Press. Oxford.

PAGE 266

252 Singh, R.K., Murty, H.R., Gupta, S.K., and Dikshit, A.K. (2009). An overview of sustainability assessment methodologies. Ecological Indicators, 9(2): 189 212. Smit, B., Burton, I., Klein, R.J.T., Street, R. (1999). The science of adaptation: a framework for assessment. Mitigation and A daptation Strategies for Global Change, 4: 199 213. Smit, B., Wandel, J., (2006). Adaptation, adaptive capacity and vulnerability. Global Environmental Change, 16(3): 282 292. Smith, G. (2010). Lessons from the United States: Planning for post disaster r ecovery and reconstruction. The Australasian Journal of Disaster and Trauma Studies, 2010:1. Smith, G. (2008). Planning for sustainable and disaster resilient communities, in Natural hazards analysis: reducing the impact of disasters, Pine, J. (E d ). Boc a Raton: CRC Press. Smith, G., We n g er, D. (2006). Sustainable disaster recovery: Operationalizing an existing agenda, in Handbook of disaster research, Rodriguez, H ., Quarantelli, E., Dynes, R. (E ds). New York: Springer. So, F.S., Getzels, J. (1988). The practice of local government planning. Washington, DC: International City Management Association, Municipal Management Series. Spangenberg, J., Bonniot, O. ( 1998). Sustainable indicators: a compass on the road towards sustainability. Wuppertal paper no. 81. Wuppertal. Stevens, M., Berke, P., Song, Y. (2010). Creating disaster resilient communities: evaluating the promise and performance of new u rbanism. Landscape and Urban Planning, 94(2): 105 115. Stevens, M., Berke P., Song, Y. (2010). Public participation in local government review of development pro posals in hazardous locations: d oes it matter, and what do local government planners have to do with it? Environmental Management, 45: 320 335. Straus, D.A. (1999). Designing a consensus b uilding process using a graphic road map, in: L. Susskind, S. McKearnan & J.T. Larner The Consensus Building Handbook. Thousand Oaks, CA: Sage Publications. Susskind, L., McKearnan, S., Larner, J.T. (1999). The consensus building handbook. Thousand Oaks CA: Sage Publications.

PAGE 267

253 Tainter, J.A. (1988). The collapse of complex societies. Cambridge University Press. Cambridge. Taylor, N. (1998). Mistaken interests and the discourse model of planning. Journal of the American Planning Association, 64(1), 64 7 5 Thomas, D.K., Furuseth, O.J. (1997). The realities of incorporating sustainable developm ent into local level planning: a case study of Davidson, North Carolina. Cities, 14(4): 219 226. Tierney, K, Bruneau, M. (2007). Conceptualizing and measuring resi lience: a key to disaster loss reduction. TR News May June: 14 17. Tobin, G.A. (1999). Sustainability and community resilience: the holy grain of hazards planning? Global Environmental Change Part B: Environmental Hazards. 1(1): 13 25. Turner, B.L., Cla rk, W.C., Kates, R.W., Richards, J.F ., Mathews, J.T., Meyer, R.B. (E ds.) (1990). The Earth as transformed by human action. Cambridge, Cambridge, U.K.: Cambridge University Press. Turner, B.L., Kasperson, R.E., Matson, P.A., McCarthy, J.J., Corell, R.W., C hrinstensen, L., Eckley, N., Kasperson, J.X., Luers, A., Martello, M.L., Plasky, C., Pulsipher, A., Schiller, A. (2003). A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Science of the United States o f America. The National Academy of Sciences. UNDP (2004). Reducing disaster risk: a c hallen ge to development Bureau for C risis Prevention and Recovery, United Nations Development Project (BCPR UNDP). UNISDR (2005). Hyogo f ramework for action 2005 2015: b uilding the resilience of nations and communities to disasters. World Conference on Disaster Reduction, 18 22 January 2005, Kobe, Hyogo. UNISDR (2007). Words Into Action: A guide for implementing the Hyogo framework. United Nations Office for Disaster Risk Reduction. DDR10382. United Nations Development Programme (1999). Human development report 1999. New York: Oxford University Press. United Nations International Strategy for Disaster Reduction (UN/ISDR) (2010). Making cities resilient: My city is getting ready. (2010, May 10). U nited S tates Government Accountability Office (GAO) (2010). Critical infrastructure protec tion: u pdate to national infrastructure protection plan includes increased

PAGE 268

254 emphasis on risk management and resilience. Repo rt to Congressional Requesters, March 2010. GAS 10 296. Vale, L., Campanella, T.J. (2005). The resilient city: How modern cities recover from disaster. Oxford: Oxford University Press. Vogt, B.M. (1990). Evacuation of institutionalized and specialized po pulations. Oak Ridge National Lab Technical Report. ORNL/Sub 7685/1 T23. Von Bertalanffy, L. 1968. General system theory: foundations, development, applications. Revised edition. New York: George Braziller. Wackernagel, M., Rees, W.E. (1997). Perceptual and structural barriers to investing in natural capital: e conomics from an ecological footprint perspective. Ecological Economics, 20:3 24. Wal drop, M.M. (1992). Complexity: t he emerging science at the edge of chaos. New York: Simon and Schuster. Wallac e, D., Wallace, R. (2008). Urban s ystems during disasters: factors for resilience. Ecology and Society, 13(1): 18. Walters, C.J. (1986). Adaptive management of renewable resources. New York: McGraw Hill. Warrick, R.A. (1980). Drought in the Great Plains: a case study of research on climate and society in the USA. In: Ausubel, J., Biswas, A.K. (E ds ) Climatic constraints on human activities. Pergamon Press: Oxford. Wen ger, D.E., James, T.F., Faupel, C.F. ( 1980 ) Disaster Beliefs and Emergency Planning. Newark, Delaware: Disaster Research Project, University of Delaware. Westley, F., Carpenter, S.R., Brock, W.A., Holling, C.S., Gunderson, L.H. (2002). Why systems of people and nature are not just social a nd ecolog ical systems. In C.S. Holling (E d ). Panarchy. Washington, DC: Island Press. White G.F. Ha as, J.E. (1975). Assessment of research on natural h azards. Cambridge, MA: MIT Press. White, G.F. (1974). Natural h azards. Oxford, New York White, G.F., Kates, R.W., Burton, I. (2001). Knowing better and losing even more: the use of knowledge in hazards management. Global Environmental Change Part B: Environmental Hazards, 3(3 4): 81 92.

PAGE 269

255 Wisner, B., Blaikie, P., Cannon, T. and I. Davis. (2004). At risk: n vulnerability and d isasters. 2nd edition. London: Routledge. Wondo l leck, J., Yaffee, S. (200 0). Making collaboration work: l essons from innovation in natural resource management. Washington, DC: Island Press. Wondolleck, J., Yaf fee, S. (2003). Collaborative ecosystem planning pr ocesses in the United States. Environments, 31(2), 59 73. Yli Pelkonen, V., Kohl, J. ( 2005 ). The role of local ecological knowledge in sustainable urban planning: perspectives from Finland. Sustainability: Science, Practice, and Policy, 1(1). Yin, R.K. (2002). Case study evaluations: a decade of progr ess? Evaluation Models, 49: 185 193.

PAGE 270

256 APPENDIX INTERVIEW GUIDES Section A (Used for organizational stakeholders) development process: 1. What was your role during the local hazard mitigation planning process? a. What were t he main tasks, responsibilities or duties that you undertook in that role? b. What other participants did you frequently interact with? 2. greatest potential contribution) to the development of the local hazard mitigation plan? 3. Who would you describe as the participant group most responsible for determining the direction of local hazard mitigation planning in your city? ith public participation during the local hazard mitigation planning process: 1. FEMA requires a public participation element to be included in the development of the local hazard mitigation plan. Do you feel that your jurisdiction adequately met this requirement? a. In your opinion, what level of public participation is most appropriate in the context of local hazard mitigation planning? b. Were there challenges or barriers to achieving this desired level of public participation? 2. What were your jurisdictio a. Was there any specific information that the planning committee was hoping to glean from public participants? 3. Which members of the public were invited to participate and why? a. Please describe the opportunities that were made available for the public to participate in the planning process. b. Was anyone from the public not involved who you think might have contributed significantly to the planning process? omponent of plan quality: 1. Were there unique benefits to including community stakeholders in the local hazard mitigation planning process? a. If yes, what are they? b. If no, why not?

PAGE 271

257 2. Did any information related to the social or economic conditions in the local community emerge as a direct result of public participation? a. If yes, was this information used later in the planning process? i. If yes, how so? ii. If no, why not? 3. Did any information about local hazard risks or vulnerable populations emerge as a direct result of the public participation process? a. If yes, was this information used later in the planning process? i. If yes, how so? ii. If no, why not? 4. In your opinion, how did the public partici pation process affect the following stages of the hazard mitigation planning process?: a. Vulnerability and risk assessment? b. Goal setting? c. Identification of jurisdiction appropriate mitigation strategies? d. Prioritization of mitigation actions? 5. In your opinion hazard mitigation plan? a. If yes, in what ways? Questions to allow the interviewee to provide additional information: 1. What improvements can be made to the public participation process? 2. Do you have any other comments that you would like to add with regard to your experience as a participant in the local hazard mitigation planning process or about the interactions between you and other participants during the process? Section B (Used for community stakeholder participants) Questions aimed at discovering how the interviewee was involved in the development of 1. What was your role during the local hazard mitigation planning process? a. What were the main tasks, responsibilities or duties that you undertook in that role? b. What other participants did you frequently interact with? 2. to the development of the local hazard mitigation p lan? a. Was there anything you were unable to contribute? Please describe. 3. Who would you describe as the participants most responsible for determining the direction of local hazard mitigation planning in your city? Questions aimed at discovering the intervie planning:

PAGE 272

258 1. As a participating stakeholder in the hazard mitigation planning process, what motivated you to participate? c. What were your goals or desired outcomes for the hazard mitigation planning process? i. Were those desires realized? 1. Why or why not? d. Was there any specific information that you hoped to share? 2. Please describe the public participation opportunities that were made available during the planning process. 3. Were there challenges or barriers to reaching y our desired or expected level of participation in the planning process? 4. FEMA requires a public participation element to be included in the development of the local hazard mitigation plan. Do you feel that your jurisdiction adequately met this requirement ? a. In your opinion, what level of public participation is most appropriate in the context of local hazard mitigation planning? plan quality: 1. From your experience, are t here unique benefits to including the public in the local hazard mitigation planning process? a. If yes, what are they? 2. Did any information related to the social or economic conditions of the local community emerge as a direct result of public participation? a. If yes, was this information used later in the planning process? i. If yes, how so? ii. If no, why not? 3. Did any information about local hazard risks or vulnerable populations emerge as a direct result of the public participation process? a. If yes, was this inform ation used later in the planning process? i. If yes, how so? ii. If no, why not? 4. In your opinion, how did the public participation process affect the following stages of the hazard mitigation planning process?: a. Vulnerability and risk assessment? b. Goal setting? c. Ide ntification of jurisdiction appropriate mitigation strategies? d. Prioritization of mitigation actions? 5. In your opinion, did public participation affect overall quality of your a. If yes, in what ways? Questions to allow the interviewee to provide additional information:

PAGE 273

259 5. What improvements can be made to the public participation process? 6. Do you have any other comments that you would like to add with regard to local capacity, your experience as a participant in the local hazard mitigation planning process or about the interactions between you and other participants during the process? INTERVIEW CODEBOOK Node 1: Roseville or Citrus Heights (+,+) I. Respondent Type 1: Organizational Stakeholder a. On HMPT b. Not on HMPT I. Leve l of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated power d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriat e mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participation? a. public b. resources

PAGE 274

260 c. institutional structure d. time e. other VI. Did public participation affect overall quality? a. yes b. no VII. Lessons for improvement? 2. Respondent Type 2: Community Stakeholder a. On HMPT b. Not on HMPT I. Level of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated power d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriate mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participation? a. public b. resources c. institutional structure

PAGE 275

261 d. time e. other VI. Did public participation affect overall quality? c. yes d. no VII. Lessons for improvement? Node 2: (+, ): Napa or Woodland I. Respondent Type 1: Organizational Stakeholder a. On HMPT b. Not on HMPT I. Level of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated power d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriate mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participati on? a. public b. resources

PAGE 276

262 c. institutional structure d. time e. other VI. Did public participation affect overall quality? a. yes b. no VII. Lessons for improvement? 2. Respondent Type 2: Community Stakeholder c. On HMPT d. Not on HMPT I. Level of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated power d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriate mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participation? a. public b. resources c. institutional structure

PAGE 277

263 d. time e. other VI. Did public participation affect overall quality? i. yes ii. no V. Lessons for improvement? Node 3 ( ): Windsor or West Sacramento I. Respondent Type 1: Organizational Stakeholder c. On HMPT d. Not on HMPT I. Level of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated powe r d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriate mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participation? f. public g. resources

PAGE 278

264 h. institutional structure i. time j. other VI. Did public participation affect overall quality? e. yes f. no VII. Lessons for improvement? 2. Responde nt Type 2: Community Stakeholder e. On HMPT f. Not on HMPT I. Level of public participation most appropriate a. high b. med c. low II. Public participation strategies? a. Public stakeholders on committees b. surveys c. town hall meetings d. websites e. newspaper articles f. PSAs III. Goals of Participation a. M1: manipulation, therapy b. M2: informing, consultation, placation c. M3: partnership, delegated power d. M4: community control IV. Unique benefits/value of participation? a. vulnerability and risk assessment b. goal setting c. identification of jurisdiction appropriate mitigation strategies d. prioritization of mitigation actions e. public buy in d. education f. incorporation of local contextual information V. Challenges for public participation? a. p ublic b. resources c. institutional structure

PAGE 279

265 d. time e. other VI. Did public participation affect overall quality? g. yes h. no VII. Lessons for improvement ?