Citation
Evaluating the integration of CWPP and county governance wildfire risk reduction best practices across the American West : a plan quality review

Material Information

Title:
Evaluating the integration of CWPP and county governance wildfire risk reduction best practices across the American West : a plan quality review
Creator:
Flohr, Travis Lee
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of Philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
College of Architecture and Planning, CU Denver
Degree Disciplines:
Design and Planning
Committee Chair:
Moreno, Rafael
Committee Members:
Troy, Austin
Simon, Gregory
Goldstein, Bruce
Beck, Jody

Notes

Abstract:
In 2003, President George W. Bush enacted the Healthy Forest Restoration Act (P.L. 108-148) (HFRA), expediting the preparation and implementation of hazardous fuels reduction efforts, if communities created community wildfire protection plans (CWPPs). CWPPs contextually define the wildland-urban interface (WUI) and evaluate risk. Additionally, CWPPs identify and prioritize both public and private land mitigation projects. However, CWPPs are only one strategy in reducing wildfire risk. Research also suggests codifying wildfire risk reduction efforts into land use regulations, such as comprehensive planning, building and zoning codes, and subdivision guidelines. This study uses document analysis and cumulative odds ordinal logistic regressions to answer the following questions: 1) how well are CWPP wildfire planning best practices integrated, 2) how well are wildfire risk reduction best practices incorporated into land use regulations, and 3) what social, economic, demographic and geographic factors predict the level of best practice integration of CWPP inputs and outputs? While the regression results proved to statistically insignificant, the study found several interesting trends. First, the WUI is still growing geographically. Second, the WUI is also increasing in population. Third, there is a small positive correlation between the increase in WUI seasonal home growth and increased composite scores (CWPP wildfire planning score + local governance wildfire land use regulations), suggesting it may be a leverage point for increased wildfire risk reduction planning. Fourth, there is a small negative correlation between longer homeowner tenure and the composite score, suggesting that wildfire risk reduction perceptions among long-time residents are complex. The results suggest that counties are not fully integrating CWPP or land use best practices. Both CWPPs and land use regulations can both improve by expanding their frames of reference from single development protectionism frames to include ecosystem health and incorporating more than one frame at a time. Additionally, counties must ensure frequent and timely update cycles, to better include emerging best practices and provide a sense of urgency to the process. Expanding frames requires a broader collective of participation. Noticeably absent from both the CWPP and comprehensive planning process were licensed land use professionals (e.g., architects, landscape architects, planners, surveyors, and civil engineers) who perform development work within each county. Counties should engage with licensing bodies to ensure professionals are adequately prepared to address health, safety, and welfare best practices to mitigate wildfire. As such professional licensure for health, safety, and welfare could be an expanded wildfire frame. When incorporating expanded frames, goals and objectives should be edited to be specific, measurable, achievable, relevant and timed (SMART). Finally, this study suggests the integration of the CDC's social vulnerability index (SVI) into risk mapping efforts and an expanded anticipatory development risk evaluation process.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Travis Lee Flohr. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
EVALUATING THE INTEGRATION OF CWPP AND COUNTY GOVERNANCE
WILDFIRE RISK REDUCTION BEST PRACTICES ACROSS THE AMERICAN WEST: A
PLAN QUALITY REVIEW by
TRAVIS LEE FLOHR
B.L.A., The Pennsylvania State University, 2002 M.S.L.A., The Pennsylvania State University, 2011
A dissertation 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 Program
2019


©2019
TRAVIS LEE FLOHR
ALL RIGHTS RESERVED


This thesis for the Doctor of Philosophy degree by Travis Lee Flohr has been approved for the Design and Planning Program by
Rafael Moreno, Chair Austin Troy, Advisor Gregory Simon Bruce Goldstein Jody Beck
Date: August 3, 2019
m


Flohr, Travis Lee (PhD, Design and Planning Program)
EVALUATING THE INTEGRATION OF CWPP AND COUNTY GOVERNANCE WILDFIRE RISK REDUCTION BEST PRACTICES ACROSS THE AMERICAN WEST: A PLAN QUALITY REVIEW Thesis directed by Professor Austin Troy
ABSTRACT
In 2003, President George W. Bush enacted the Healthy Forest Restoration Act (P.L. 108-148) (HFRA), expediting the preparation and implementation of hazardous fuels reduction efforts, if communities created community wildfire protection plans (CWPPs). CWPPs contextually define the wildland-urban interface (WUI) and evaluate risk. Additionally, CWPPs identify and prioritize both public and private land mitigation projects. However, CWPPs are only one strategy in reducing wildfire risk. Research also suggests codifying wildfire risk reduction efforts into land use regulations, such as comprehensive planning, building and zoning codes, and subdivision guidelines. This study uses document analysis and cumulative odds ordinal logistic regressions to answer the following questions: 1) how well are CWPP wildfire planning best practices integrated, 2) how well are wildfire risk reduction best practices incorporated into land use regulations, and 3) what social, economic, demographic and geographic factors predict the level of best practice integration of CWPP inputs and outputs?
While the regression results proved to statistically insignificant, the study found several interesting trends. First, the WUI is still growing geographically. Second, the WUI is also increasing in population. Third, there is a small positive correlation between the increase in WUI seasonal home growth and increased composite scores (CWPP wildfire planning score + local
IV


governance wildfire land use regulations), suggesting it may be a leverage point for increased wildfire risk reduction planning. Fourth, there is a small negative correlation between longer homeowner tenure and the composite score, suggesting that wildfire risk reduction perceptions among long-time residents are complex.
The results suggest that counties are not fully integrating CWPP or land use best practices. Both CWPPs and land use regulations can both improve by expanding their frames of reference from single development protectionism frames to include ecosystem health and incorporating more than one frame at a time. Additionally, counties must ensure frequent and timely update cycles, to better include emerging best practices and provide a sense of urgency to the process. Expanding frames requires a broader collective of participation. Noticeably absent from both the CWPP and comprehensive planning process were licensed land use professionals (e.g., architects, landscape architects, planners, surveyors, and civil engineers) who perform development work within each county. Counties should engage with licensing bodies to ensure professionals are adequately prepared to address health, safety, and welfare best practices to mitigate wildfire. As such professional licensure for health, safety, and welfare could be an expanded wildfire frame. When incorporating expanded frames, goals and objectives should be edited to be specific, measurable, achievable, relevant and timed (SMART). Finally, this study suggests the integration of the CDC's social vulnerability index (SVI) into risk mapping efforts and an expanded anticipatory development risk evaluation process.
The form and content of this abstract are approved. I recommend its publication.
Approved: Austin Troy


ACKNOWLEDGEMENTS
This study was made possible by the diligent efforts of my faculty advisor, Dr. Austin Troy. I am grateful for his years of guidance and support. Additional thanks go out to the other committee members of my dissertation committee for their assistance on this project: Drs. Rafael Moreno (defense chair), Gregory Simon, Bruce Goldstein, and Jody Beck. Your help and support will forever be remembered and appreciated.
I want to thank my wife, Stephanie, and daughter, Ruth. Thank you for your patience, smiles, laughs, and the balance you brought to my life. Without it, this project would be meaningless and incomplete. My final thanks go to my family, the Ph.D. cohort of Alessandro Rigolon and Mehdi Heris, and friends and colleagues, Mike Hinke, Suzy Anderson, Ed Russell, Heidi Ochis, and Jim Robb. Thank you for being there; your collective intelligence and patience inspires me.
VI


TABLE OF CONTENTS
CHAPTER
I. WHY STUDY THE INTEGRATION OF BEST PRACTICES IN COMMUNITY WILDFIRE
PROTECTION PLANS (CWPP)?........................................................1
Research Aims, Questions, and Significance...................................5
Dissertation Organization...................................................11
II. FEDERAL WILDFIRE RISK REDUCTION EFFORTS....................................12
Introduction...............................................................12
Federal Legislative and Administrative Risk Reduction Actions...............12
Fire Suppression Policies: 1871-2003....................................12
The Healthy Forest Initiative and the Health Forest Restoration Act: 2002 to Present.... 15 CWPP Legislative Incentives, Content, and Process.......................18
III. CWPP BEST PRACTICES AND LOCAL GOVERNMENT WILDFIRE RISK
REDUCTION EFFORTS..............................................................22
Introduction...............................................................22
CWPP best practices.........................................................22
CWPP Context............................................................23
Goal and objectives.....................................................25
Community capacity......................................................28
Partnerships and collaboration..........................................28
Base map................................................................29
Risk assessment.........................................................36
Hazardous fuels reduction...............................................50
vii


Reducing structural ignitibility...........................................60
Education and outreach.....................................................61
Emergency management capacity..............................................62
Long-term success..........................................................63
Local government wildfire risk reduction efforts...............................63
Comprehensive Plan.........................................................64
Zoning Codes and Development Standards.....................................66
Fire-Resistant Materials...................................................68
IV. RESEARCH DESIGN AND METHODOLOGY...............................................70
Research Permissions and Ethical Considerations................................70
Pilot Study Results............................................................70
Study Area.....................................................................71
Research Design................................................................73
Independent and Dependent Variables........................................73
Independent Variable Data..................................................74
Study Methodology and Sample...............................................75
Data Analysis..................................................................97
V. STATISTICAL RESULTS OF CWPP PLANNING AND IMPLEMENTATION......................103
Integration and Implementation................................................103
CWPP Integration Scores...................................................103
CWPP Implementation Scores................................................112
CWPP Composite Scores (integration + implementation)......................119
Conditions of Effective CWPPs and Risk Reduction Implementation...............120
viii


CWPP Process and Plan Evaluation Inferential Statistics.........................121
Spearman’s correlation coefficient results....................................121
Ordinal logistic regressions results..........................................121
CWPP Implementation Local Governance Evaluation Inferential Statistics..........122
Spearman’s correlation coefficient results....................................122
Ordinal logistic regressions results..........................................122
Composite Evaluation Inferential Statistics......................................124
Spearman’s correlation coefficient results....................................124
Ordinal logistic regressions results..........................................124
VI. CONCLUSIONS ON CWPP AND LOCAL GOVERNANCE INTEGRATION IN
REDUCING WIIDFIRE RISK...............................................................126
Regression Results...............................................................126
CWPP Process Document Analysis Scores and Results................................130
CWPP Document Analysis Scores and Results.....................................130
Local Governance Document Analysis Scores and Results.........................178
Final Thoughts: Theory, Practice, and Policy Implications for HFRA...............194
Implications for Planning Theory..............................................194
Implications for Practice.....................................................204
Implications for HFRA: Flaws and Improvements.................................215
Limitations and Future Research...............................................218
Conclusion....................................................................220
REFERENCES...........................................................................224
IX


APPENDIX
A. CWPP Process and Plan Content Evaluation Instrument.............................249
B. Local Governance Instrument.....................................................253
C. Pilot Study Write-Up............................................................255
D. CWPP Bibliography...............................................................279
E. Local Governance Bibliography...................................................285
F. Gis Data Bibliography...........................................................293
G. Crook County, Wyoming’s Risk Field Assessment...................................294
H. Mohave County, Arizona Fuel Treatment and Modification Plan.....................296
I. Cal Fire’s Wildfire Evacuation Guide.............................................298
J. Cal Fire’s Evacuation Planning Guide.............................................306
K. Model Community Widlfire Evacuation Guidelines..................................312
L. Calfire’s Home Ignition Zone Inspection Form....................................321
M. WFAP Home Assessment Checklist..................................................323
X


LIST OF TABLES
TABLE
3.1. Fire Danger Rating System adjective class ratings.............................42
3. 2. NFPA Firewise Community Recommended Defensible Space Requirements............59
4. 1. Number of sampled CWPPs per strata per American West state...................83
4. 2. Sample strata one WUI area change (Km2) from 2000-2010 descriptive statistics, summarized per state...............................................................84
4. 3. Sample strata two WUI area change (Km2) from 2000-2010 descriptive statistics,
summarized per state............................................................................85
4. 4. Sample strata three WUI area change (Km2) from 2000-2010 descriptive statistics,
summarized per state............................................................................85
4. 5. Sample strata one population change from 2000-2010 descriptive statistics, summarized per
state...........................................................................................86
4. 6. Sample strata two population change from 2000-2010 descriptive statistics, summarized
per state.......................................................................................87
4. 7. Sample strata three population change from 2000-2010 descriptive statistics, summarized
per state.......................................................................................87
4. 8. Sample strata one housing unit change from 2000-2010 descriptive statistics, summarized
per state.......................................................................................88
4. 9. Sample strata two housing unit change from 2000-2010 descriptive statistics, summarized
per state.......................................................................................89
4. 10. Sample strata three housing unit change from 2000-2010 descriptive statistics, summarized
per state
89


4. 11. Sample strata one seasonal housing unit change from 2000-2010 descriptive statistics,
summarized per state...........................................................................90
4. 12. Sample strata two seasonal housing unit change from 2000-2010 descriptive statistics,
summarized per state...........................................................................91
4. 13. Sample strata three seasonal housing unit change from 2000-2010 descriptive statistics,
summarized per state...........................................................................91
4. 14. Sample strata one median household income change from 2000-2010 descriptive statistics,
summarized per state...........................................................................92
4. 15. Sample strata two median household income change from 2000-2010 descriptive
statistics, summarized per state..............................................................93
4. 16. Sample strata three median household income change from 2000-2010 descriptive
statistics, summarized per state..............................................................93
4. 17. Sample strata one median age change from 2000-2010 descriptive statistics, summarized
per State......................................................................................94
4. 18. Sample strata two median age change from 2000-2010 descriptive statistics, summarized
per state......................................................................................95
4. 19. Sample strata three median age change from 2000-2010 descriptive statistics, summarized
per state......................................................................................95
4. 20. Sample strata one length of homeowner tenure descriptive statistics summarized per state.
...............................................................................................96
4. 21. Sample strata two length of homeowner tenure descriptive statistics summarized per state. ...............................................................................................97
Xll


4. 22. Sample strata three length of homeowner tenure descriptive statistics summarized per
state..............................................................................................97
5. 1. Cohen’s Kappa levels of reliability interpretation..........................................104
5. 2. Average of categorical scores CWPP “Process and Plan Evaluation Instrument” document
analysis score results for all strata.................................................105
5.3. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument”
document analysis score results for strata one........................................108
5. 4. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument”
document analysis score results for strata two........................................109
5.5. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument”
document analysis score results for strata three.................................110
5. 6. Score to letter grade conversion chart.....................................112
5. 7. Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local
Governance Evaluation Instrument” document analysis score results for all strata......113
5. 8. Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local Governance Evaluation Instrument” document analysis score results for strata 1........116
5.9 Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local
Governance Evaluation Instrument” document analysis score results for strata 2........117
5.10 Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local
Governance Evaluation Instrument” document analysis score results for strata 3......118
5.11 Minimum, maximum, and average composite score results for all strata...........120
CWPP Implementation Local Governance Evaluation proportional odds test results......123
XUl


6. 1. Siskiyou County, CA’s Fire Safe Council Projects in Siskiyou County...................131
6. 2. Siskiyou County, CA’s list of critical stakeholders..................................142
6. 3. Excerpt (not a complete listing of participants) from Montrose County, CO CWPP
development team’s roles and responsibilities................................................148
6. 4. Boulder County, CO community values at risk and weights of importance................161
6. 5. Benewah County, ID’s legal and regulatory resources available for wildfire mitigation
efforts......................................................................................176
6. 6. Base map requirements.................................................................212
6. 7 HFRA CWPP sliding scale incentives.....................................................217
XIV


LIST OF FIGURES
FIGURE
1.1. CWPP inputs and outputs................................................................8
3.1. Observed Fire Danger Class............................................................43
3. 2. LANDFIRE fuels data acquisition process for FARSITE..................................46
4. 1. Study states, defined as the American West and associated county boundaries..........72
4. 2. Percent increase in large fire activity in the American West.........................73
4. 3. Simple, stratified random sampling framework.........................................76
4. 4. County 2010 median income stratification without normalization (natural breaks)......78
4. 5. County 2010 median income stratification without normalization (quantiles)...........79
4. 6. County 2010 median income z-score quantile stratification............................80
4. 7. Final county sample population (counties that contain WUI and a county level CWPP).... 81
4. 8. Final county sample data set stratified per state-normalized z-scores................82
4. 9. Change in WUI area (Km2) per sampled county from 2000-2010, summarized per state. 84
4. 10. WUI population change per sampled county from 2000-2010, summarized per state.......86
4. 11. Amount of WUI housing unit change per sampled county from 2000-2010, summarized
per state...................................................................................88
4. 12. WUI seasonal housing unit change per sampled county 2000- 2010, summarized per state.
............................................................................................90
4. 13. WUI median household income per sampled county from 2000-2010, summarized per
state.......................................................................................92
4. 14. WUI median age per sampled county from 2000-2010, summarized per state...............94
XV


4. 15. WUI average length of homeowner tenure per sampled county from 2000-2010,
summarized per state...................................................................96
4. 16. Model one - CWPP process and content index....................................98
4. 17. Model two - CWPP implementation index.........................................99
4. 18. Model three - CWPP process and content + implementation index.................99
5.1. Minimum, maximum, and average of “Process and Plan Evaluation Instrument” document analysis total score results..........................................................106
5. 2. Average of “Process and Plan Evaluation Instrument” document analysis total score results
per strata, summarized per state......................................................107
5.3. Minimum, maximum, and of the “CWPP Implementation: Local Governance Evaluation
Instrument” document analysis total score results....................................114
5. 4. Average of the “CWPP Implementation: Local Governance Evaluation Instrument” document analysis total score results per strata, summarized per state...............115
5. 5. Final composite score results per strata, summarized per state..................120
6. 1. Boulder County, Colorado CWPP goals.............................................136
6. 2. Boulder County, Colorado CWPP goals.............................................136
6. 3. Asotin County, Washington public meeting announcement..........................145
6. 4. El Paso County, Colorado steering team responsibilities........................147
6. 5. Fremont County, WY proposed project area priorities and timelines, excerpt......150
6. 6. Benewah County, Idaho Forest Owners Field Day Announcement......................153
6. 7. Gunnison County, Colorado letter soliciting participation as a community wildfire mitigation advocate (WMA).............................................................154
XVI


6. 8. Harney County, Oregon CWPP base map.............................................156
6. 9. Bingham County, Idaho CWPP base map.............................................157
6. 10. Cochise County, Arizona CWPP base map..........................................158
6. 11. Boulder County, CO wildfire risk assessment map results........................161
6. 12. Campbell County CWPP industrial development and land ownerships................164
6. 13. Okanogan County, WA priority fuels reduction projects and map locations........166
6. 14. 6153 Laurel Dr. Paradise, CA, taken July, 2012.................................190
6. 15. 565 Valley View Dr. Paradise, CA, taken July, 2012.............................191
6. 16. 5915 Pine View Dr. Paradise, CA, taken July, 2012.............................192
6. 17. Anticipatory modeling process.................................................214
6. 18. Embedded comprehensive planning and local regulatory CWPP process.............222
xvii


CHAPTER I
WHY STUDY THE INTEGRATION OF BEST PRACTICES IN COMMUNITY WILDFIRE PROTECTION PLANS (CWPP)?
The Healthy Forest Restoration Act (P.L. 108-148) (HFRA), first national legislation that expedited the preparation and implementation of hazardous fuels reduction efforts, provides a useful instrument for analyzing whether and under what conditions local governments’ are implementation wildfire best practices. This study evaluates Community Wildfire Protection Plans (CWPP) and local governments policy implementation supporting the reduction of wildfire risk (e.g., zoning, building codes). The study also determines whether and how adoption of best practices vary by median age, length of homeowner tenure, full-time/part-time residency, and income.
While the role of CWPP and enforceable code best practices implementation is important in exploring risk and theory, it is of considerable importance to policy makers as well. First and foremost, the evaluation indices help quantify the degree to which each local government has implemented scientific best practices for wildfire risk reduction, elucidating gaps in CWPP plans and local codes and guidelines. Secondly, understanding what social, economic, demographic and geographic factors predict the integration of best practices in CWPPs is paramount to how decision-makers engage community members.
Wildfires have been increasing in severity and cost over the last several decades due to increasing fuel loading and ex-urban development in the wildland-urban interface (WUI). WUIs are where wildland uses meet urban land uses. Radeloff, Hammer, Stewert, Fried, Holcomb, and McKeefry (2005) estimated that the United States (U.S.) WUI covers 719,156 km2 and contains 44.8 million housing units. The expansion of homes and associated commercial development in
1


the WUI places property, assets, and human lives at risk from wildfires (Bhandary & Muller, 2009; Reams, Haines, Renner, Wascom, & Kingre, 2005). At the same time federal costs to suppress wildfires are drastically increasing; in 1985 costs were $239,943,000, and costs in 2012 were $1,902,446,000 (National Interagency Fire Center, 2018). The total Federal cost for wildfire suppression between 1985 and 2012 was $25,370,157,000 (National Interagency Fire Center, 2018). While the land use change drivers of WUI development may differ in other parts of the world, the expansion of the WUI into high-risk wildfire zones is not unique to the U.S. (Brummel, Nelson, & Jakes, 2012; Carmo, Moreira, Casimiro, & Vaz, 2011; Dondo Biihler, Curth, & Garibaldi, 2013; Harris, McGee, & McFarlane, 2011; Holland, March, Yu, & Jenkins, 2013).
Initial responses to wildfire in the United States WUI almost exclusively emphasized fire suppression (Steelman & Burke, 2007). The fire suppression policies implemented between 1905 through 1911 failed to address several key issues related to wildfire in the WUI: 1) the necessity of fire for ecosystem health; 2) the ability of fire to moderate fuel load buildup; and 3) the altering the public’s perception of wildland aesthetics and processes (Busenberg, 2004; Veblen, Kitzberger, & Donnegan, 2000). Despite the failings of fire suppression policies, these policies governed wildfire practices for almost 100 years.
WUI development conflicts were initially raised by Vaux (1982), who classified the interface of urban development and forestry as a significant area of concern and cautioned foresters to not underestimate its political and policy significance. Additionally, Bradley (1984) also spoke of the significance of the urban/forest interface, defined as two traditional land uses occurring in proximity to each other (i.e. forestry and urban development). Early views of the urban/forest interface defined the problems as resource conflicts that begin as spatial conflicts
2


and quickly become socio-political, often pitting a set of values against one another (Bradley, 1984; Vaux, 1982). These early conflicts often revolved around forest managers’ concern over forest resources as a commodity, such as mineral extraction and timber harvests, and urban edge residents’ concern in forests as land commodity, such as aesthetic amenities, recreation, and wildlife (Bradley, 1984). However, land managers tackling these concerns failed to address the growing wildfire issues in the WUI.
Recently, U.S. wildfire management policy has shifted from fire suppression to an integrated program of fire suppression, preparedness, mitigation, and community assistance (Gonzalez-Caban, Haynes, McCaffrey, Mercer, & Watson, 2007). This shift is recognizable in the wildfire research literature. Researchers, designers, and planners have been working to understand the increasing conflicts between WUI landowners and natural wildfire regimes by using a wider multitude of research designs, methods, and theories, in addition to continuing wildfire-modeling research (Bhandary & Muller, 2009; Brown, Agee, & Franklin, 2004; R. Burby & Deyle, 2000; Heyerdahl, Brubaker, & Agee, 2001; Muller & Schulte, 2011; Paveglio, Jakes, Carroll, & Williams, 2009; Reams et al., 2005). For example, research and practitioners have developed best management practices for community wildfire protection planning processes (Jakes et al., 2012; Society of American Foresters, 2004), identified Firewise development and land management practices (Headwaters Economics, 2014; M. A. Moritz et al., 2014; Paterson, 2007; Winter, McCaffrey, & Vogt, 2009), identified socio-economic barriers to implementing Firewise best management practices (Chuvieco, Martinez, Roman, Hantson, & Pettinari, 2014; T. W. Collins, 2008a; Gardner, Cortner, & Widaman, 1987; Kousky, Olmstead, & Sedjo, 2011; Poudyal, Johnson-Gaither, Goodrick, Bowker, & Gan, 2012), and created better models of wildfire risk and behavior (Ager, Vaillant, & Finney, 2011; Kramer, Collins, Kelly, &
3


Stephens, 2014; LANDFIRE, 2010). Practitioners have attempted to implement these best practices to mediate increasing fire risk with policies limiting development in high-risk areas, requiring fire rated building materials and sprinkler systems, implementing community wildfire protection plans, creating defensible space around buildings, and reducing fuel loads (Bhandary & Muller, 2009; Headwaters Economics, 2014) with varying levels of success.
Current wildfire planning efforts, particularly those related to community wildfire protection plans, are the result of The Healthy Forests Restoration Act of 2003 (HFRA). HFRA was the culmination of a decade of changing wildfire mitigation research and wildland fire policy reforms that were in response to the growth of WUI development, danger from catastrophic WUI wildfires, and decline in WUI ecosystem health (Steelman, 2008). HFRA called for communities to implement Community Wildfire Protection Plans (CWPP) (Grayzeck-Souter, Nelson, Brummel, Jakes, & Williams, 2009). CWPPs are often implemented at the county or community (i.e. municipality, borough, town, city or local fire district) scale, hence the term ‘County’ or ‘Community Wildfire Protection Plan.’ Increasingly, CWPPs are also implemented at a neighborhood or subdivision scale. CWPPs encourages collaboration between local fire departments, the state agency responsible for forest management and relevant local government, in consultation with adjacent federal land management agencies and surrounding community residents (Grayzeck-Souter et al., 2009).
CWPPs have several key benefits and objectives for achieving a more effective wildfire mitigation strategy. The development of CWPPs should include priority areas for fuel reduction and provide ignitability assessment throughout the community ("The Healthy Forest Restoration Act of 2003," 2003). Communities benefit from having CWPPs because it allows for a flexible and contextually defined, localized WUI boundary; localized fuel treatment prioritization;
4


prioritization of funding; and integration into local land use policies (Jakes et al., 2011; Steelman & Burke, 2007). Robustly implemented CWPPs allow land managers to reestablish natural fire regimes while minimizing the risk to people, thus rehabilitating and restoring fire-adapted ecosystems and minimizing ongoing wildfire risks over the long-term (Steelman & Burke, 2007). Researchers have only begun to tackle the integration of CWPP best practices.
While research is expanding into effective integration of CWPP, there are still numerous gaps in the literature. Often CWPP and local wildfire research is not generalizable because studies have used small n or single case study examples, which make it difficult to understand specific contexts that facilitate effective CWPP processes and implementation. As is discussed in the literature review, the vast majority of studies on CWPPs tend to focus on the social outcomes of CWPP processes, including capacity building and the interrelationships of participants, while giving little attention to the content of CWPP products, policies or the physical implementation of CWPP goals and objectives. This oversight creates a disconnect in understanding how processes lead to deliverables as well as how deliverables lead to implementation and the reduction of risk. Finally, CWPP literature has understudied the links between CWPP processes, documents, and the integration of wildfire safe policies into enforceable local zoning and land use codes. As a result, the question of how well CWPP processes and CWPP implementation are at integrating best practices and achieving their goals remains.
Research Aims, Questions, and Significance
The purpose of this research is to evaluate whether CWPPs and county governance wildfire risk reduction best practices meet minimum process and outcome criteria across the American West, which requires an understanding of the process that created the CWPP as well as CWPP implementation. Counties were chosen as a unit of analysis because, as discussed in
5


the literature review: 1) counties are the logical scale to mitigate landscape-scale wildfire risk and 2) county governments are implementing CWPPs for broad expanses of the WUI and have not been widely studied. Local CWPPs were excluded from this study for two reasons: 1) feasibility and limitations on scope and time and 2) the incongruent boundaries between CWPPs and socio-demographic and economic data introduced too much error and uncertainty in such an extensive study (see Appendix C for a more detailed discussion on the pilot study). For the purposes of this project, minimum process and outcome criteria includes 1) the process of developing the CWPP is inclusive of the diverse stakeholders in the area, involving participatory actions of those stakeholders; 2) the CWPP clearly articulates and documents the extent of the wildfire hazard; 3) the CWPP plans adhere to principles of defensible space and other evidence-based wildfire mitigation best practices, stating implementable and measurable goals; and 4) the implementation of the CWPP goals result in observable reduction in WUI development and implementation of building codes and policies that reflect CWPP goals and objectives. In order to evaluate CWPP planning in terms of integration, indices were created to examine the CWPP process and the resulting documents as well as the CWPP implementation. These indices served as the dependent variables for this research while community contextual variables, such as socioeconomic and demographic variables were independent variables because they are known to impact planning programs and their implementation.
Specifically, my research objectives/questions are:
• Create an index that defines the level of best practice integration of CWPP process (inputs);
• Create an index that defines the level of best practice integration of CWPP implementation (outputs); and
6


• What social, economic, demographic and geographic factors predict the level of
best practice integration of CWPP inputs and outputs? Inputs and outputs are further clarified in Figure 1.1.
Inputs
Outputs
Minimum HFRA Requirements
Bio-physical Interventions
Partnerships and
Collaboration
Reducing Hazardous Fuels
Reducing Structure
Ignitability
Agreement -Three
Land Managers
V________________y
CWPP Best Practices
Risk
Assessment
Education/
Outreach
Emergency
Management
->c
CWPP
WUI Expansion and Densification
Fuel Reduction Treatments*
<"â–  â– >
Defensible Space Implementation*
*"â–  ""s
Fire Code Compliant Building Construction*
Local Governance
Comprehensive
Plan
Zoning Code, Standards, and Design Guidelines
s________________>â– 
Building
Codes
^ >
Plan Review and Inspection Procedures
Note. *Denotes outputs that are not measurable across large geographic scales at this time, thus they are excluded from this project.
7


Figure 1. 1. CWPP inputs and outputs.
Addressing these research objectives/questions is important to researchers for four main reasons. First, little research has been conducted to evaluate the integration of CWPP processes best practices and to evaluate the integration of CWPPs on implementing wildfire mitigation strategies. Second, this study will methodologically broaden the scope of wildfire research through a larger n sample, which allows for greater generalizability, enables the pinpointing of outliers more easily, and creates the potential for a better margin of error. These are not possible with the current small n or case study wildfire research. Third, this study introduces the spatial analysis of the implementation of CWPP mitigation objectives, such as limiting WUI expansion into known wildfire hazard areas, which is paramount in assessing CWPP effectiveness. Fourth, this study will empirically evaluate the socio-economic, demographic, and biophysical conditions that are associated with inhibit CWPP processes and implementation best practices.
For this study, the CWPP process includes both the actual CWPP plan, as a product of the process, and meeting minutes, as artifacts of the engagement in the process. Implementation is a composite of both policy and biophysical interventions. CWPP policy implementation are the integration of wildfire policies into building and zoning codes; comprehensive plans; subdivision ordinances, and HOA guidelines. The biophysical components of CWPP implementation include no net change in WUI expansion, densification, and health WUI forests and development.
Details on each index and measurement protocols are discussed at length in Chapter II and located in Appendix A and B.
The wildfire and hazard planning literature has identified a number of socio-economic and demographic variables that can influence CWPP processes and implementation, including age, length of homeowner tenure, full-time/part-time residency, and income. Survey research has
8


shown that these four variables can play a statistically significant role in effective public participation processes, and CWPP implementation (T. W. Collins, 2008b; Crow et al., 2015; J. W. Smith, Leahy, Anderson, & Davenport, 2013; Wolters, Steel, Weston, & Brunson, 2017). Based on this previous research, the hypothesis for this study is: CWPP geographies with older populations, short homeowner tenure, high part-time residency status, and lower income will have lower integration in the dependent variables. Other variables considered were: educational attainment, owner occupied housing, retirement status, slope, percent WUI, percent wildland cover, aspect, proximity to recent fire, hazard exposure, and single egress points; however, these have proved statistically insignificant or less significant in previous studies (T. W. Collins, 2008b; Crow et al., 2015; J. W. Smith et al., 2013; Wolters et al., 2017).
The purpose of indices, in general, is to allow comparisons across time and space (Ebert & Welsch, 2004). The proposed process index and implementation index are critical to wildfire mitigation in the CWPP process because they aid in decision-making surrounding the prioritization of planning processes and implementation strategies. This is significant in the daily practices of wildfire mitigation and the CWPP process for both the decision-makers and the community members. In regards to decision-makers, an integration process index is important for the following reasons: 1) to compare a community’s process with other communities in order to validate localized best practices of community engagement, 2) to prioritize the utilization of staff-time and resources, and 3) to facilitate community participation without overburdening community members. For community members, there are two main benefit of a process index:
1) the promotion of more meaningful engagement in the planning process; by focusing community participation in areas where community members can actively contribute, the community members will not be fatigued and frustrated by diluting their participation across too
9


many initiatives and 2) legitimizes the CWPP planning process by providing sustained appraisals on planning products, procedures and results.
An integration implementation index also benefits decision-makers and community members during the CWPP process. To decision-makers, an integration implementation index facilitates 1) the comparison of community implementation in order to validate localized best practices of defensible space, zone sizing, forest thinning practices, and so forth; 2) the prioritization of implementation strategies in regards to economic restraints and location throughout the community; 3) the utilization of incentives for implementation, particularly on private property; 4) the application of implementation policy mechanisms, including fines for unmitigated WUI development and requirements of disclosure of guidelines during property transfer; and 5) as a performance indicator. For community members, an implementation index has two benefits: 1) knowing and understanding the levels of wildfire safety in their community as well as the progress being made to reach the community safety goals, and 2) receiving targeted guidance in how to best reduce risk on their private property.
Understanding what social, economic, demographic and geographic factors predict CWPP best practice integration is paramount to how decision-makers engage community members. These underlying structural contexts of the community members’ ability to participate in wildfire mitigation efforts regardless of how much wildfire mitigation education has been presented. For example, members of a community in abject poverty may understand the importance of wildfire mitigation but may also need to choose between landscape alteration and food. Furthermore, these same community members may not be able to attend community participation events because they work multiple jobs to provide for themselves and/or their family. Therefore, the use of these predictive factors by decision-makers is 1) to gauge whether
10


the lack of environmental outcomes (i.e. wildfire safe communities) is due to these community factors or due to the CWPP process and/or implementation, 2) to identify and allocate necessary resources and time to address other community barriers and issues, such as poverty or language barriers, in order to produce more effective processes and implementation, and 3) to understand the limit of a community’s capacity to participate in various implementation processes and strategies.
Dissertation Organization
This document is organized into six chapters. The first chapter discussed the importance of studying the integration of the CWPP process and content best practices. Additionally, I discussed the need to study the integration of wildfire risk reduction best practices into local land use regulations, also known as county or local governance. Chapter Two discusses federal wildfire risk reduction efforts and how they have culminated in the Healthy Forest Initiative, which led to the Healthy Forest Restoration Act and CWPPs. Chapter Three outlines CWPP and local governance best practices. Chapter Four outlines my research design and methodology. Chapter Five presents my statistical results. In Chapter Six, I discuss my conclusions and their implications for planning practice and theory.
11


CHAPTER II
FEDERAL WILDFIRE RISK REDUCTION EFFORTS
Introduction
The Healthy Forest Restoration Act (HFRA) was enacted due to the inadequacies of preexisting wildfire risk reduction efforts in the wildland urban interface. In order to understand the conditions which led to the creation of HFRA, I first discuss a brief history of federal legislative and administrative risk reduction actions prior to HFRA. Second, I discuss HFRA and other present wildfire policies. Finally, I discuss the key components of HFRA and CWPPs.
Federal Legislative and Administrative Risk Reduction Actions Fire Suppression Policies: 1871-2003
The current understanding of wildfire management needing to address a complex mix of natural environment, economic, and cultural factors follows over 100 years of fire suppression as the wildfire policy. It is important to understand the background on the origination and evolution of suppression policies to address the effects of these policies on current development and attitudes and the resulting challenges in modern wildfire mitigation in the WUI. Complex regional and national economics, westward expansion, agricultural resources, energy and mineral resource extraction, urban growth, and recreation policies and attitudes have all contributed to the United States’ development of the WUI (Headley, 1916; Riebsame, Gosnell, & Theobald, 1996; Theobald & Romme, 2007; Travis, 2007). Initial WUI development of the 1800s was driven by resource economies, such as timber, agricultural activities, and mining (Riebsame et al., 1996; Theobald & Romme, 2007; Travis, 2007), and wildfire was viewed as a threat to commercial timber activities and watershed protection (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910). Legendary fires, such as Peshtigo Fire of 1871, bolstered the
12


conservationist argument that wildfires are a threat to forest commodities, leading to the creation of national forests as protected timber reserves (Fire History Society; Headley, 1916; Muir,
1941; Silcox, 1910).
The U.S. Forest Service (USFS) was established in 1905 and had managerial control of national forests (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910). During its early years, catastrophic forest fires burnt over three million acres in Montana, Idaho, and Washington (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910). Administrators were convinced that catastrophic fire events could be prevented (Fire History Society; Headley, 1916; Muir,
1941; Silcox, 1910). The USFS convinced lawmakers and the American public that fire suppression was the only way to prevent such events from ruining timber economies and communities (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910).
These early USFS initiatives culminated in the passage of the Weeks Act of 1911, which established a framework between the federal and state governments for cooperative firefighting, which would later include private forest associations and landowners (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910). The Weeks Act offered financial incentives to states for suppressing fires under the direction of the USFS, a framework that is still in place today (Southard, 2011). The grant support efforts of the Weeks Act were expanded in 1924 through the Clark-McNary Act, to further support state efforts in fire protection (Southard, 2011). Following several severe fire seasons in the early 1930s, fire suppression took on a greater urgency. The Civilian Conservation Corps supplied thousands of men to work building fire breaks and fighting fires (Fire History Society). In 1935, the USFS established the 10 a.m. policy, which stated that every fire should be suppressed by 10 a.m. the day following its initial report (Fire History Society).
13


The fire suppression policies implemented between 1905 and 1935 continued through 1970 and exacerbated the fire problem; fires have continued to grow in size and intensity. Suppression policies failed to acknowledge that fire is necessary to moderate fuel load buildup and altered the public’s perception of wildland aesthetics and processes, leading to increased WUI development and preferred unnatural aesthetics (Busenberg, 2004; M. A. Moritz et al., 2014; Nielsen-pincus, Ribe, & Johnson, 2015; Steelman & Burke, 2007; Scott L. Stephens & Collins, 2007). During the 1960s, research began to show the positive role fire played in forest ecology: restoring vegetation by releasing seeds, controlling diseases and invasive species, and managing wildfire fuel loads. Despite this emerging knowledge, the federal government further expanded its influence in fire suppression policies and management procedures through the National Wildfire Coordinating Group in 1976 and expanded their grant funds to include administrative resources and surplus federal equipment to rural fire departments through the Cooperative Forestry Assistance Act in 1978 ("Cooperative Forestry Assistance Act of 1978," 1978). However, with the support of emerging fire science, the USFS instituted a let-burn policy (Fire History Society), but it fell out of favor as early as the 1980s due to dangerous fire conditions in the WUI where development had increased - driven by resort development, urban expansion, low-density homeowner development (Fire History Society; Manning, 2012).
In 1986, the National Wildland/Urban Interface Fire Protection Initiative brought together representatives from federal land management, fire protection agencies, and the National Associate of State Foresters to address the emerging problem of fire in the WUI (J. Cohen, 2008). As a result, the home destruction problem became nationally recognized and the U.S. Forest Service and National Fire Protection Association held conference - the 1986 Wildfire Strikes Home conference - which spawned the current Firewise program (J. Cohen,
14


2008). These initiatives marked the change in the wildfire narrative from resource protection to WUI management, highlighted by the fact that the WUI was identified in as a principle concern in the 2000 National Fire Plan, 2001 Federal Wildland Fire Management Policy, and Healthy Forest Restoration Act. Despite the evolution and changing attitudes of wildfire and the WUI within federal policy, these approaches were flawed because they failed to address three pivotal issues: 1) subsidizing the financial risk of living in the WUI through the allocation of fire suppression funds, 2) engaging private land owners inadequately, if at all, and 3) failing to acknowledge the influence of municipal policies, such as zoning role in wildfire risk reduction.
The Healthy Forest Initiative and the Health Forest Restoration Act: 2002 to Present
Healthy Forest Initiative
From 2001 to 2003 the U.S. experienced 147,049 fires that burned approximately 11 million acres (The White House, 2003). The fires in 2002 claimed the lives of 28 firefighters (The White House, 2003). The 2003 California fires alone accounted for approximately $250 million in fire suppression costs and claimed 22 civilian lives (The White House, 2003). These staggering figures reinforced the recognition that the catastrophic fires of the American West were burning hotter and faster than most ordinary wildland fires. In August 2002, President George W. Bush established the Health Forests Initiative (HFI), which directed the Departments of Agriculture and Interior, and the Council of Environmental Quality to improve regulatory processes to ensure more timely decisions, greater efficiency, and better results in reducing the risk of catastrophic wildland fires (One Hundred Eighth Congress of the United States of America, 2003). HFI is momentous because for the first time the initiative strove to engage the public and all levels of government while caring for forests and rangelands to reduce wildfire risk, save lives, and protect endangered and threatened species.
15


The Healthy Forest Restoration Act
The Healthy Forest Restoration Act (HFRA), passed in 2003, is the result of wildland fire policy reforms that emerged from the HFI, which was designed to improve the capacities of land-management agencies to protect communities, watersheds, and other at-risk lands from catastrophic wildland fires (Jakes et al., 2011). The legislation provisions are to expedite the preparation and implementation of the reduction of hazardous fuels on federal lands and assisting rural communities, states, and landowners in restoring healthy forest and watershed conditions. HFRA—without providing uniform guidelines or standards—encourages communities to collaboratively develop Community Wildfire Protection Plans (Jakes et al., 2011).
The components and process of the Community Wildfire Protection Plan (CWPP) is structured loosely by HFRA. Within that structure is built the flexibility of local communities to, ideally, create a CWPP that reflects community values and area-specific environmental conditions in order to effectively mitigate wildfire in the WUI. CWPP guidelines are designed to be flexible in order to address local contexts, though each plan should have the following components: description of the WUI and associated resources at risk, documentation of community preparedness, community risk analysis that prioritizes fuel treatment priorities and methods of treatment, ways to reduce structure ignitability, an implementation plan, and collaboration of stakeholders (Society of American Foresters, 2004). Hazardous fuel reduction area treatments must be identified and prioritized, and the methods of fuel reduction treatments must also be recommended (Jakes et al., 2011; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; Steelman & Burke, 2007). CWPPs are should identify at least two zones of defensible space surrounding building structures.
16


CWPPs should be collaboratively developed by local, state, and federal stakeholders, and the local government (e.g. counties or cities); local fire departments); and state entity responsible for forest management must agree on the final contents (Society of American Foresters, 2004). HFRA also recommends at least three entities must agree to the final CWPP: applicable local governments (i.e. counties or cities); local fire departments); and state entity responsible for forest management (Society of American Foresters, 2004). While not required, HFRA does outline and recommend an eight-step process for CWPP development. The outline is as follows: convene decision-makers; involve federal agencies; engage interested parties; establish a community base map; develop a community risk assessment; establish community priorities and recommendations; develop an action plan and assessment strategy; and finalize community wildfire protection plan.
HFRA provides the authorization to expedite environmental assessment, administrative appeals, and legal review for hazardous fuels projects on federal land (Society of American Foresters, 2004). HFRA also emphasizes the need for federal agencies to collaborate with communities in developing hazardous fuel reduction projects by placing priority on treatment areas identified by CWPPs (Society of American Foresters, 2004). It also distributes financial resources across federal and non-federal projects according to CWPP objectives. The HFRA also disincentives the absence of a CWPP through the financial policy of requiring at least 50 percent of all funds to be used within the WUI. Without a CWPP, the WUI is defined within U mile of a community’s boundary or within 1 V2 miles when mitigating circumstances exists, e.g. steep slopes, which are less than most locally developed CWPP definitions.
HFRA is a step in the right direction, but it does present several issues. HFRA’s biggest flaw is that it does not address continued development in the WUI. The spirit of individualism
17


and the resentment of government intervention in private property rights hinder such legislative actions surrounding wildfire risk and mitigation (T. W. Collins, 2008b), which explains why only half of state legislatures address land use planning and only 11 of these encourage local governments to plan for hazard mitigation (R. Burby & Deyle, 2000). The result of continued WUI growth is that wildfire funds are continually allocated suppress fires in order to protect homes rather than towards risk reduction efforts. Exacerbating the issue is that forest mitigation work - such as Healthy Forest initiatives - has been consistently underfunded (Steelman & Burke, 2007; Trego, 2014).
The goal of these efforts is to maintain healthy forests that reduce risk for WUI residents. Healthy forests (HF) within the context of HFI and HFRA consists of low forest density, low fuel loads, and more variability in age of vegetation and forest structure. A healthy forest structure, in effect, is of pre-fire suppression activities. Healthy WUI forests and WUI development include the observable implementation of fire-resistant building materials, defensible space, and reduction in overall fuel loads within the WUI. While HF are theoretically important to evaluating the integration of CWPP implementation, they are currently not able to be evaluated across large, disparate geographies. HF measures are currently not feasible for this study because datasets do not exist or would require expensive high-resolution remote sensing multi-spectral imagery. Additionally, such datasets require time-intensive on-the-ground calibration assessments. As such, the current study will not contain measures of HF and future research should expand in these areas.
CWPP Legislative Incentives, Content, and Process
As previously mentioned, CWPPs are locally generated wildfire protection plans, a HFRA policy incentive requirement. It is worth reiterating the core components, incentives and
18


disincentives. Encouragement of CWPPs primarily comes in the form of facilitating agencies and collaboration. HFRA provides the authorization to expedite environmental assessment, administrative appeals, and legal review for hazardous fuels projects on federal land (Society of American Foresters, 2004). HFRA also places priority on treatment areas identified by community CWPPs, emphasizing the need for federal agencies to collaborate with communities in developing hazardous fuel reduction projects (Society of American Foresters, 2004). Furthermore, financial resources can be distributed across federal and non-federal projects according to CWPP objectives.
The components and process of the CWPP is structured loosely by HFRA. HFRA suggests a minimum of the following: prioritized fuel reduction, treatment of structural ignitability, and collaboration, (Society of American Foresters, 2004). The CWPP process consists of eight steps: 1) convene decision-makers; 2) involve federal agencies; 3) engage interested parties; 4) establish a community base map; 5) develop a community risk assessment; 6) establish community priorities and recommendations; 7) develop an action plan and assessment strategy; and 8) finalize community wildfire protection plan (Jakes et al., 2007; Jakes et al., 2012; Society of American Foresters, 2004). Ideally, within that structure is built the flexibility for local communities to create a CWPP that reflects community values and area-specific environmental conditions in order to effectively mitigate wildfire in the WUI. For example, HFRA allows for communities to define their own WUI interface areas, valuing local knowledge of area contexts. In fact, if communities do not define and designate WUI areas, then the WUI interface area defaults to National Register’s WUI classification and the community is not prioritized in federal and state funding designations. Additionally, communities should prioritize project and treatment areas and provide baseline information for monitoring of long-
19


term effects on the CWPP in wildfire risk reduction. While CWPP guidelines are designed to be flexible to address local contexts, each plan should have the suggested components, and HFRA recommends following the eight-step process for CWPP development as stated above.
The initial Federal Registrar definition of an urban wildland interface community were derived from “A Report to the Council of Western State Foresters—Fire in the West—The Wildland/Urban Interface Fire Problem” (Teie & Weatherford, 2000). Per this definition, wildland interface communities exist where humans and their associated development meet or intermix with wildland fuel. The Federal Registrar (C. N. Thompson, 2001) provides a further categorization of the WUI using three sub-classifications: interface, intermix, and occluded communities. An interface community is where three or more structures per square mile exist, but there is also a clear line of demarcation between residential, business, and public structures and wildland fuels. Alternatively, interface communities emphasize a population density of 250 or more people per square mile. An intermix community is where close together structures to 40 structures per acre intermix with wildland fuels and are continuous outside and within the developed area. Intermix communities have population densities of 28-250 people per square mile. Occluded communities exist when wildland fuels exist in parks and open spaces (<1,000 acres) located within a city, with a clear demarcation between fuels and structures. Federal agencies focus on interface and intermix communities only.
These three sub-classifications all have significant limitations. First, the definitions do not define wildland vegetation communities or patterns. Because the Federal Register does not set a minimum threshold of wildland vegetation type or density, small areas of urban parks may be erroneously included. Second, the Federal Register does not standardize the WUI interface thresholds for the maximum housing or population density measures. Third, the distance to
20


development from the interface line is not systematic. The Healthy Forest Restoration Act (HFRA) has subsequently implemented definitions of WUI that supplement the Federal Register categorization of the WUF HFRA seeks to empower local communities to develop a CWPP that is accepted by the community, fits community-specific goals and objectives, and addresses the practical needs of wildfire mitigation for local WUI context. To further this goal of CWPP development, HFRA defines the WUI and thresholds, which includes the establishment of a buffer-zone around the town, civic infrastructure, and evacuation routes (Stewart, Radeloff, Hammer, & Hawbaker, 2007). I will elaborate on how CWPP best practices have attempted to contextualize the operational measurement of the WUI and wildfire risk in Chapter III.
The CWPP should identify and prioritize hazardous fuel reduction area treatments, and the methods of fuel reduction treatments must also be recommended (Jakes et al., 2012; Jakes et al., 2011; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; Steelman & Burke, 2007). Additionally, these plans are should identify at least two zones of defensible space surrounding building structures, as previously defined. CWPPs are meant to be collaboratively developed by local, state, and federal stakeholders; and the local government (e.g. counties or cities), local fire departments), and state entity responsible for forest management must agree on the final contents (Society of American Foresters, 2004). HFRA encourages at least three entities agree to the final CWPP: applicable local governments (i.e. counties or cities); local fire departments); and state entity responsible for forest management (Society of American Foresters, 2004).
21


CHAPTER III
CWPP BEST PRACTICES AND LOCAL GOVERNMENT WILDFIRE RISK
REDUCTION EFFORTS Introduction
A thorough discussion of CWPP best practices and local government wildfire risk reduction efforts is paramount to assessing the CWPP process as well as how CWPPs penetrate local government policies. First, best practices will be discussed, which were used to structure the CWPP process instrument. Second, local government wildfire risk reduction efforts and reduction options will be discussed.
CWPP best practices
Research and practitioner experience has identified best practices for the CWPP process and content, and the implementation of these best practices is often a coordinated effort between local, state, and federal agencies, just as HFRA intended these three to be the key signatories of CWPPs (Busenberg, 2004; Jakes et al., 2011; One Hundred Eighth Congress of the United States of America, 2003; Scott L. Stephens & Collins, 2007). These best practices are categorized and presented according to the following order: CWPP context, goals and objectives, community capacity, partnerships and collaboration, base mapping, risk assessment, hazardous fuel reduction, reducing structural ignitability, education and outreach, emergency management capacity and long-term success. These 11 categories have been deemed necessary to create an effective CWPP (Jakes et al., 2007; Jakes et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; Williams et al., 2012), and are the groupings utilized in the CWPP process analysis instrument, detailed in Chapter IV and in Appendix A.
22


In this first section of this chapter, a systematic discussion will step through each of these eleven groups recommended for CWPP development. In sub-section one, the need for an adequate CWPP context will be discussed as well as what should be included in that context. In sub-section two, goals and objectives will be defined and suggestions to creating effective goals and objectives will be provided. In sub-section three, the need to document, and augment, community capacity through the CWPP will be presented. In sub-section four, partnerships and collaboration will be exemplified. In sub-section five, the need for a base map will be discussed and what should be included on the base map will be identified, such as human presence and development, wildland vegetation, and wildfire risk. In sub-section six, risk assessment will be explored by reviewing LANDFIRE, wildfire exposure modelling, and behavioral modelling, and finishing with CWPP wildfire risk modeling to highlight some best practices. In sub-section seven, hazardous fuels reduction will be presented by discussing fuel management basics and types of fuel treatments, rounding out with details on defensible space. Finally, in sections eight through eleven, reducing structural ignitability, education and outreach, emergency management capacity, and long-term success will be discussed, respectively.
CWPP Context
Adequate context for the community is paramount in developing a robust CWPP, and this context should be created through the efforts of all key stakeholders, including the local state, and federal governments. Jakes et al (2012) outlines five components to community context, that are critical to CWPP success: 1) remind community members of how they handled past challenges, such as a wildfire or environmental disaster in order to help the community understand how it is vulnerable and create a sense of urgency for developing a CWPP; 2) study previous collaborative efforts in the community, whether wildfire planning or other projects, to
23


identify how they were successful and use lessons from those experiences to lay the groundwork for doing a CWPP; 3) identify people who were involved in earlier collaborative or wildfire planning efforts and bring their experience to developing a CWPP; 4) find ways to overcome the challenge of inexperienced communities in collaboration or wildfire planning; and 5) address disagreements within a community early, related to wildlife or not, that could threaten the CWPP process (Jakes et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004). Each of these elements, while not part of the core HFRA legislation, are necessary because community history provides context for the CWPP and can determine the integration of the CWPP process.
Understanding how communities handled past adversities provides a window of opportunity to inspire change. Indeed, if these events were recent they serve as a leverage point to spur action, if you act quickly and before the urgency of desired action diminishes (A. M. S. Smith et al., 2016; Williams et al., 2012). Identifying and documenting past collaborative efforts can identify existing formal and informal community group structures that can support the collective actions of CWPP processes (Fleeger, 2008; Goldstein & Butler, 2012; Innes &
Booher, 2014). These groups can be wildfire specific, such as Fire Safe Councils or non-wildfire groups, such as homeowners’ associations or neighbourhood groups. These groups have the capacity to be organized for taking collective action in support CWPP efforts in addition to becoming peer advocates in implementing CWPP outcomes.
CWPPs should also identify past wildfire risk reduction participants as they can serve as vital experience in working collaboratively and convey its benefits to new team members (Fleeger & Becker, 2010; Headwaters Economics, 2016b; Homey, Nguyen, Salvesen, Tomasco, & Berke, 2016; Jakes & Sturtevant, 2013; K. C. Nelson, Souter, Jakes, & Williams, 2010).
24


Indeed, it also provides continuity of ongoing efforts and preserves the institutional knowledge of community wildfire efforts. It is also just as helpful to identify if the community has little collaborative or wildfire planning experience. This helps to direct community’s efforts in identifying outside support and collaborators such as, public agencies, nongovernmental organizations or consultants to lead the necessary collaborative processes. If a community lacks collaborative skills, it may require investments in training or new personnel to ensure collaboration readiness. Finally, it is imperative that disagreements within the community are documented because earlier conflicts can often become barriers and derail CWPP collaborative efforts (Fleeger, 2008; Goldstein & Butler, 2012; Innes & Booher, 2014). In some cases, it may be necessary to openly discuss past disagreements to determine lessons learned and bridge past differences. In extreme cases this may require external, professional facilitation because of broken community trust.
Goal and objectives
The purported genius of the HFRA is that it is vague and allows communities to craft goals and objectives specific to their values in relation to wildfire risk reduction efforts (Jakes et al., 2011). However, goals and objectives should go beyond vague terms of reducing wildfire risk, reducing fuel loads and structural ignitibility. Goals and objectives provide a community framework to minimize wildfire risks, but more importantly these should articulate implementable action items across all aspects of the CWPP and best practices. The goals, objectives, and action items serve as the baseline for evaluating the plan’s performance. Per Oregon’s Resource Innovations Institute for a Sustainable Environment (2008), goals and objectives should include the following issues: partnerships and collaboration; risk assessment;
25


fuels reduction; structural vulnerability reduction; emergency management; and education and
outreach.
Goals and objectives will vary depending on the scale at which the CWPP is created. The issue of the appropriate geographic scale of the CWPP is neglected by the HFRA and the Resource Innovations Institute for a Sustainable Environment’s guidelines. Scale choice should be driven by the motivations and goals articulated in the CWPP (Jakes et al., 2007; Jakes et al., 2012; Rodman & Stram, 2008; Society of American Foresters, 2004). For example, CWPPs should be developed at a small scale, e.g. neighborhood or community level, if the goals are to motivate homeowners to reduce hazards on their properties. However, CWPPs should be developed at a larger scale, e.g. counties, municipalities, and fire districts if the goals are to reduce regional landscape wildfire risk. This research effort evaluates county level CWPPs, and therefore the CWPPs should address the following goals and objectives at a minimum: definition of the WUI, risk mapping, plans to reduce structural ignitibility and hazardous fuels, and connection to multiple frames.
Goals and objectives should clearly articulate and reflect community values and connect to multiple contexts and frames. Framing is important because CWPPs can be framed in a multitude of ways: fuels management, life safety, and ecosystem health. Jakes et. al. (2012) provides the following example, if someone frames wildfire management as a life safety concern, they will be more interested in a CWPP that frames goals and objectives in terms of evacuation and response times rather than fuels management at a landscape scale for ecosystem health. As a result, it is critical to know which frame(s) are being considered and used in the CWPP process because it will often determine who chooses to participate in the CWPP process and identify potential conflicts that may arise determining which goals and objectives are
26


priorities. CWPPs at the county scale should rarely use a single frame. It is necessary to consider multiple frames in order to broaden community engagement by persuading community members participation and implementation is in their own self-interest. Landscape scale risk reduction efforts are complex and require a multitude of frames to be effective. Additionally, multiple frames broaden diversity, skills, and access to additional resources. For example, a frame of watershed and ecosystem health provides access to additional funding resources, such as The National Forests Foundation Grant Programs (National Forest Foundation, 2019), that can also reduce fuel loads.
Goals and objectives should be specific, measurable, achievable, relevant, and time-bound (SMART). Following this SMART framework is beneficial because goals need to be specific enough to provide a link to objectives, which need to be measurable. Measuring success in implementing goals is critical to the long-term success of projects in order to maintain community involvement and engagement in implementation activities. Setting goals that are not achievable or not measurable are counterproductive to maintaining risk reduction efforts. Irrelevant goals are detrimental because it raises questions about the veracity of the CWPP process and the other goals and objectives. Time-sensitive goals help to create early and repeatable successes that are critical to maintaining momentum for long-term success, helping to build trust among the participants. Short-term goals, depending on the county’s past wildfire efforts, may be as simple as documenting the number of structures that are not wildfire code compliant in order to provide a better assessment of wildfire risk. Long-term goals may include ensuring all existing and new construction meets current code standards or reducing landscape scale fuel reduction loads to a more natural condition.
27


Community capacity
CWPPs should document community capacity. Community capacity includes characteristics such as community norms and values, economic diversity, growth trends, land ownership practices, and financial and physical resource availability. Tapping into resources that help CWPP participants successfully work together is paramount because effectively functioning collaborative groups can overcome financial and resource obstacles (Jakes et al., 2012; Rodman & Stram, 2008). The key benefits of this include building community capacity to be used during the CWPP process and other activities beyond wildfire planning. While the expansion of the network can provide more people, technology, and funding to the CWPP process, additional conflicts about CWPP goals can arise. The outcomes of a CWPP go beyond the document itself, but also the expanding capacity for action it builds, opportunities and networks it creates, the knowledge it advances, and the connections among people and organizations it develops. Documenting and expanding community capacity also mobilizes individuals to participate and lend legitimacy to the CWPP, secure funding, and shepherd the process. These leaders might include federal, state, or local government representatives, community residents, or activists. Expanded capacity can help achieve outcomes beyond wildfire preparedness such as watershed and forest health because the community’s knowledge of local ecological issues and role of fire ecology is expanded and they have built a stronger sense of community. Communities who do not identify and assess their capacity will often find the CWPP process is stalled because of inadequate funding, lack of physical capacity, and a lack of community and agency support. Partnerships and collaboration
The success of the CWPP will hinge on the core team effectively engaging a broad range of stakeholders. Substantive input from a diversity of interests ensures the final document
28


reflects the highest priorities of the community and helps to facilitate the implementation of recommended projects. Identifying leaders, within or outside the community, who can help mobilize others will help to expand the creation of core CWPP member and community participants (Jakes et al., 2012). These leaders can also serve as catalysts for action and recruit others to the CWPP process. Stakeholders can include, but is not limited to forest management groups, city council members, resource advisory committees, HO As, Division of Wildfire/Fish and Game, Department of Transportation, local and state emergency management agencies, water districts, utilities, recreation organizations, environmental organizations, forest products interests, local chambers of commerce, and watershed councils. To solicit additional input, the core team may choose to hold public meetings.
The partnerships and collaboration should also outline which entities are responsible for which pieces of the CWPP and its implementation, with key timelines. Such efforts ensure accountability and are a mechanism of trust building among the community and CWPP signatories. Expansive partnerships help bring new ideas and resources to CWPPs and ensure that vulnerable communities are not overlooked. The partners should identify the shelf life of the CWPP as it is a living and guiding document, it will need semi-regular updates and maintenance efforts.
Base map
The core team, agencies, and stakeholders should collaboratively develop a community basemap that includes: 1) inhabited areas at potential risk to wildland fire; 2) areas containing critical infrastructure that are at risk to fire disturbance events, e.g. escape routes, municipal water supply structures, and major power or communication lines; and 3) a preliminary designation of the community’s WUI zone (Jakes et al., 2012; Society of American Foresters,
29


2004). This map will be used to facilitate a community risk assessment, which helps the core team and community members to prioritize areas for treatment and to identify the highest priority uses for available financial and human resources. A successful and meaningful community assessment should be co-developed considering the following risk factors: fuel hazards; risk of wildfire occurrence; homes, businesses, and essential infrastructure at risk; other community values at risk; and local preparedness and firefighting capacity (Jakes et al., 2012; Society of American Foresters, 2004). Ranking systems typically consist of adjective rating systems, e.g. high, medium, and low. The key objective of these discussions is to develop the community’s prioritized recommendations for fuel treatment projects on federal and nonfederal lands in the WUI, including preferred treatment methods for each project. These decisions result in the action plan (identifies the roles and responsibilities, funding needs, and timetables for carrying out the highest priority projects) and the assessment strategy (ensures that the document maintains longterm relevance and effectiveness) (Rodman & Stram, 2008).
WUI designations vary widely from CWPP to CWPP. The literature reports on a variety of definitions including the use of the Federal Registrar (C. N. Thompson, 2001). Since the CWPP will be frequently updated it is important to identify and track how a community identifies the WUI, to see how it grows and changes in the associated levels of risk are assessed. Several operational methods of defining the WUI have been proposed and are outlined below. Operationalizing the WUI requires defining the three integral components of the WUI: human presence, wildland vegetation, and wildfire risk, which are discussed below.
Human Presence and Development
Operationalizing the measurement of human presence in the WUI involves either zonal or point based measurements (Bar-Massada et al., 2013). The zonal approach requires an areal unit,
30


and traditionally, the areal units or zones are census blocks or parcels. The Federal Register density measurement works very well with census blocks because housing data is available across the United States, is consistently collected, and can serve as the unit of analysis for which housing and vegetation are evaluated (Bar-Massada et al., 2013).
Radeloff et al (2005) operationalized an integrated WUI definition using the Federal Register’s WUI definition as a starting point and combining this with detailed housing density data, high resolution vegetation data, and spatial adjacencies to wildland vegetation1. Their WUI measurements are unique because they use thresholds of wildland vegetation requirements as core parts of interface and intermix classifications. Additionally, they use a 2.4 km community fire planning zone distance because it represents the estimated distance a firebrand can fly ahead of a fire front. With this schema, if a census block were only partially within the 2.4 km distance, the census block would be split so that only the portion within the 2.4 km would be included as interface. However, simply splitting census blocks is problematic, as the distribution and locations of housing density within the census block is unknown. A more conservative measure would be to include the entire census block, particularly as Radeloff et al (2005) report that census blocks ranged from 0.001 km2 to 2,700 km2.
Theobald and Romme’s (2007) WUI definition and measurement expands Radeloff et al’s (2005) research in three ways: 1) using a differing definition of WUI that adjusts census blocks to reflect development patterns; 2) integrating a more detailed approach to classifying wildland vegetation; and 3) adding wildfire risk categorizations. Through ‘ad hoc analysis’ Theobald and Romme (2007) have determined the Federal Register’s WUI definition of >3 units
1 See Radeloff et al (2005) for specific details.
31


per acre does not adequately address all WUI development types at risk; however, it is unknown if they have empirically evaluated this assumption. Future research should confirm Theobald and Romme’s ‘ad hoc analysis’ or further refine a more nuanced definition of the WUI that is synonymous with wildfire risk. Finally, Theobald and Romme (2007) removed the portion of each census block that overlapped with protected lands because private development typically cannot be built on public or protected lands, and they also removed blocks that overlapped water features. This approach eliminates the error Radeloff et al (2005) encountered by splitting a census block based on a given distance, as they are unaware as to the distribution of housing within the census block. Approximately 1/3 of Colorado is public land, and by removing these public lands from the calculations, the area average was reduced nearly 40%, resulting in identifying over 131,600 additional hectares (-18%) of WUI.
The WUI intermix lands in need of wildfire mitigation defined by the Federal Registrar are different from those identified by researcher definitions. While the Federal Registrar determines the density of WUI intermix to be 1 housing unit per 0.30 - 40 acres, Theobald & Romme (2007) calculate a WUI intermix density of 2.40 - 40 acres for wildfire mitigation. This difference is a result of Theobald & Romme (2007) accounting for a community protection zone to surround the traditionally-defined WUI intermix area. This approach incorporates gradients of housing density in the intermix zone. However, the thresholds need to be validated and refined in future research because they were based off the researchers’ assumptions rather than data analysis.
Both Radeloff et al’s (2005) and Theobald and Romme’s (2007) approaches to measuring WUI suffer from the modifiable areal unit problem (MUAP), which is a bias that is the result of using point-based measurements aggregated into areas for which summary statistics are
32


calculated (Bar-Massada et al., 2013). As previously mentioned, census block geometry can vary in size and there is no way of knowing exactly where each person lives. Comparing census blocks to each other is problematic because the population distribution unknown density and distribution. MUAP issues contain two parts: scale and zonal issues. WUI researchers have written extensively on zonal MUAP issues while little is written on WUI mapping scale MUAP issues. The point-based approach to operationalizing the WUI for measurement is exemplified through the work of Bar-Massada et al (2013), who refined the measurement of the WUI by using housing location data. Bar-Massada et al (2013) used a moving window analysis with various window sizes to represent neighborhood sizes in order to calculate housing and wildland vegetation. The results showed similar area results from previous zonal measures (Radeloff et al., 2005; Stewart et al., 2007) but produced more precise spatial location results (Arganaraz et al., 2017; Bar-Massada et al., 2013). This process, however, can only be used in areas where housing location data is available. Additionally, while the neighborhood radiuses of the moving window analysis are scalable, sensitivity analysis shows that different urban forms across the United States produce varying results; each urban location has a different best-fit radius.
Wildland Vegetation
U.S. Geological Survey (USGS) National Land Cover Data (NLCD) is often used to classify wildland vegetation, which is comprised of coniferous, deciduous, and mixed forest; shrubland; grasslands/herbaceous; transitional; and woody and emergent herbaceous wetlands (Radeloff et al., 2005; Theobald & Romme, 2007b; Bar-Massada et al., 2013). This vegetation classification is adequate for national, state or landscape scale WUI assessments; however, among the three forest types, there are numerous types of plant communities with differing fire
33


regimes and risks. In an effort to confront this issue, Theobald and Romme (2007) augmented
the NLCD data set with higher resolution vegetation data to map wildfire risk.
Wildfire Risk
Wildfire risk is addressed inconsistently in measurement of the WUI.2 Radeloff et al
(2005) did not explicitly address wildfire risk other than utilizing the travel distance of a fire brand (2.4 km) in their WUI interface measurements. HFRA typical distances (800, 1600 and 3200m) for treatments in community protection zones (CPZ) were used by Theobald and Romme (2007). These distances capture the range of wildfire fighting objectives: structure protection, safe fire-fighting zone based on the maximum sustained flame length of a crown fire, and avoidance of flying embers (Theobald & Romme, 2007). Under extreme conditions, however, these distances are not sufficient. Additionally, these distances vary depending on the differing vegetation communities, slope, wind, and types of fires. Theobald & Romme (2007) used a variable-width buffering technique that utilizes cost-distance computations, but the weights used are arbitrary and do not reflect the role that topography, wind, stand age/structure, and fire types play in wildfire behavior within different vegetation communities. The greater oversight is integrating wildfire risk into a model without considering defensible space, which can reduce a home’s vulnerability (Theobald & Romme, 2007).
Wildfire risk categories, classified using an aggregated synthesis of raster data sets, were defined as 1) high risk or crown fires, 2) low risk or ground fires, 3) variable risk fires, which are
2 As a reminder, the working definitions for hazard, risk, and vulnerability are: 1) hazard is the potential threat to be exposed to a wildfire event (Ager, Day, McHugh, et al., 2014; Ager et al., 2012); 2) risk is the impact that a wildfire event could have on community infrastructure and populations, assuming the community population has an equal ability to respond to wildfire events (Bryant & Westerling, 2014); vulnerability is an individual or household’s ability to anticipate, respond to, and recover from a wildfire event, in addition to broader hazard and risk elements (Collins, 2008a).
34


generally low risk fires that can occasionally become high risk fires due to natural cycles and 4) high risk fires that were historically low or variable fire regimes but are now high risk due to high fuel loads due to fire suppression policy (Theobald & Romme, 2007). High severity fire regimes involve fires burning at a high intensity through crowns and are often difficult to contain or suppress, posing the greatest threat to structures (Theobald & Romme, 2007). Public perception of wildfire generally involves images of these crown fires with the large flames and damage to infrastructure. Low severity fire regimes are fires that burn at a relatively low intensity through surface fuels. These fires minimally spread into tree or shrub crowns, making them relatively easy to contain or suppress (Theobald & Romme, 2007). Both the variable fire regimes and high risk fires that were historically low fires have significantly increased in severity due to the fuel load accumulation caused by the fire suppression policies of the 20th century (Theobald & Romme, 2007). This fuel load accumulation becomes ladder fuel, which allows these fires to turn into crown fires. While many risk modeling and reduction efforts have focused on high intensity crown fires because of their threat to structure ignition, the ground fire and variable fire regimes is underestimated in terms of wildfire risk and, therefore, in wildfire mitigation.
Due to differences in operationalization, the estimates of WUI lands varies significantly. According to Radeloff et al (2005),719,156 km2 of WUI exists in the United States. Theobald and Romme (2007b) estimate WUI interface to be 465,614 km2 nationwide. And while these WUI area estimation methods and results are significantly different, it is important to note that both sets of results illustrate a significant growth in the WUI and project continued growth into the future. This growth is in existing WUI areas as well as significant grown in new areas, particularly the Intermountain West (Radeloff et al., 2005; Stewart et al., 2007; Theobald &
35


Romme, 2007). Therefore, wildfire mitigation efforts in the WUI are increasingly important to reducing threats to losses of life and of economic interests. As a result CWPPs should be very clear in how they are defining the frame of their goals and objectives because it will often influence how they define and map the WUI, which can have very clear biases in how risk is assessed.
Risk assessment
The WUI definitions are largely driven by how communities perceive wildfire, as a hazard, risk, or vulnerability, as a result how a community defines and maps its WUI will often determine the means with which it assess risk. It is important to distinguish the difference between wildfire hazard, risk, and vulnerability since these terms are often used interchangeably in the literature and media. Wildfire hazard relates to factors affecting the fire environment and likely fire behavior, including fuel and vegetation properties, topography, climate and weather variables, and ignition characteristics (Bryant & Westerling, 2014; M. P. Thompson, Ager, Calkin, Finney, & Vaillant, 2012). Wildfire risk characterizes the potential for wildfire to harm human life and safety or damage highly valued resources and assets (HVRAs) (Ager et al., 2011; Keane, Drury, Karau, Hessburg, & Reynolds, 2010; M. P. Thompson et al., 2012). While WUI definitions mention and imply wildfire risk, they give no specific criteria to measure or calculate risk (Haas, Calkin, & Thompson, 2013). As a result, there are varying means and approaches to how wildfire risk is addressed in the WUI (Ager, Vaillant, Finney, & Preisler, 2012; Bryant & Westerling, 2014; Elia, Lafortezza, Colangelo, & Sanesi, 2014; Haas et al., 2013). For example, wildfire forest resource management literature defines risk as an element of hazard exposure, but does not include wildfire impacts (Ager et al., 2012; Carmel, Paz, Jahashan, & Shoshany, 2009; Finney, McHugh, Grenfell, Riley, & Short, 2011). Land use planning defines wildfire risk as a
36


combination of fire exposure and impacts (Abukhater, 2011; Bryant & Westerling, 2014; Haas et al., 2013; Paveglio, Prato, & Hardy, 2013). Finally, political ecologists include concepts of socio-economic vulnerability (T. W. Collins, 2008a, 2011). Wildfire vulnerability is an individual or household’s ability to anticipate, respond to, and recover from a wildfire event, in addition to broader hazard and risk elements (T. W. Collins, 2008a).
Vulnerability involves three realms: 1) root causes, 2) dynamic pressures, and 3) unsafe conditions. The first realm—root causes—refers to a wide range of historical, political, economic, demographic, and environmental factors that produce unequal distributions of resources (T. W. Collins, 2008b). Previous wildfire risk studies have operated under the assumption that households are exercising free residential choice, particularly in the face of resort-centric or affluent WUI development (T. W. Collins, 2008b). The integration of LUCC modeling and vulnerability approaches to wildfire risk could further our understanding of future WUI development and risk.
The second realm of vulnerability involves dynamic pressures. Dynamic pressures are processes and activities, such as rapid population growth, urbanization, environmental degradation, global economic pressures, political conflict, (T. W. Collins, 2008b). These processes, together, create unsafe conditions under which some people in a given place and time must live.
Finally, the third realm of vulnerability—unsafe conditions—includes both spatial location and other characteristics of the built environment and socio-economic barriers. Built environment vulnerability factors include architectural and landscape architectural design choices, as well as the institutionalization of real estate market and planning decisions that determine residential settings, such as single family subdivision, mobile home park, apartment
37


complex, condominium, gated enclave, and isolated dwellings (T. W. Collins, 2008b). Additionally, socio-economic vulnerability factors include fragile livelihoods, resource dependency, inadequate incomes, education, legal and political inequalities, and a lack of preparedness for emergencies (T. W. Collins, 2008b).
Political ecologists have contributed to a broader understanding of wildfire risk by utilizing the concept of vulnerability to provide understandings of wildfire mitigation perceptions, and barriers to implementing wildfire safe mitigations practices. Unfortunately, their work has been largely ignored in modeling efforts due to the complex and contextual nature of vulnerability factors (T. W. Collins, 2008b). This is of particular concern because a third of the 13 million WUI residents of the western US lack incomes sufficient to meet basic economic needs, much less the cost of basic wildfire protection (Lynn, 2003). Additionally, due to the pervasive use of modeling in decision-making processes (Papadopoulos & Pavlidou, 2011), the omission of vulnerability in these models becomes increasingly important.
Modeling efforts, while omitting vulnerability, focus on hazard and risk. Hazard modeling is generally split into two approaches: 1) wildfire exposure and 2) wildfire behavior. Wildfire exposure models explore the predicted scale and spatiotemporal relationships of causative risk factors. Therefore, they evaluate the likelihood of a wildfire event based on current climate and weather, vegetation, slope, and fuel-loads (Stratton, 2006; M. P. Thompson et al., 2012). Wildfire behavior models simulate the behavior of the fire itself, producing bum probabilities for fire intensity, rate of spread, flame length, and crown fire activity (M. P. Thompson et al., 2012). Wildfire risk modeling combines the two types of hazard models— exposure and behavior—and efforts to model the likelihood of wildfire interacting with valued resources loads (Stratton, 2006; M. P. Thompson et al., 2012). In order to adequately model
38


wildfire risk, multiple iterations of the wildfire behavior model are calculated for a variety of conditions in order to produce a comprehensive burn probability (M. P. Thompson et al., 2012). Therefore, differing definitions and approaches to risk lead to differing wildfire risk and behavior models. Disagreement about wildfire processes, topography, environmental factors and their interactions produces variable levels of uncertainty and error, which complicates planning efforts to reduce risk (Cochrane et al., 2012; Finney, 2002; Green, Finney, Campbell, Weinstein, & Landrum, 1995; Keane, Mincemoyer, Schmidt, & Garner, 2000). The different types of wildfire modeling, the benefits and drawbacks of each model, and demonstrates the need for vulnerability in wildfire risk modeling efforts for effective planning policy and mitigation efforts will now be discussed. First, a review of Landscape Fire and Resource Management Planning Tools (LANDFIRE) is needed because the tools of this program are often used in wildfire modeling efforts.
LANDFIRE
Landscape Fire and Resource Management Planning Tools (LANDFIRE), is a program provided by the wildfire fire management programs of the U.S. Department of Agriculture Forest Service and U.S. Department of the Interior. LANDFIRE is a collection of the most prominent nationwide fire hazard mapping products and used in a wide-range of wildfire modeling efforts. LANDFIRE products are designed and developed to be used at the landscape level in order to facilitate national and regional strategic planning as well as the reporting of wild land fire. LANDFIRE provides landscape scale geo-spatial vegetation, fire regime, topographic, fuel disturbance, and reference database products at 30-meter pixels (Stratton, 2009). Therefore, the adaptation of these products can support a variety of local management applications (Stratton,
2009).
39


The benefits of using LANDFIRE are that it is: 1) consistent landscape-scale and crossboundary geospatial products; 2) adaptable to support local planning, management, and monitoring activities requiring consistent vegetation data; 3) aid in strategic and tactical planning for fire operations where other necessary data are unavailable; and 4) usable by Federal and State agencies and private organizations to collaborate regarding fire and other natural resource management issues. Additionally, LANDFIRE receives comprehensive updates roughly every 10 years, with incremental updates as available.
Wildfire Exposure Modeling
Wildfire exposure modeling, also known as fire danger mapping or fire potential index, includes mapping and modeling various biophysical elements of risk, but does not include the expected impacts of wildfire or concepts of social and community vulnerability. Exposure modeling efforts focus on causative risk factors, such as flame length and fire size to quantify risk as burn probability (BP)—the likelihood of a given location experiencing a wildfire during a specified period of time (Ager et al., 2012). Wildfire exposure literature acknowledges BPs are best calculated with a composite of fuels, topography, and weather conditions (Ager et al., 2014; Finney, 2006; Stratton, 2006), yet many risk assessments preclude one or more of these critical variables because operationalizing these factors across larges scales is difficult. The difficulty is due to incompatible data resolutions, gaps in data availability, daily fluctuations of datasets, or expertise and time required to gather adequate fuel load data (Keane et al., 1998; Keane et al., 2000; Kramer et al., 2014; Papadopoulos & Pavlidou, 2011; Stratton, 2009). These issues are highlighted in the two predominate US exposure models: The Fire Danger Rating System (NFDRS) and the National Fire Danger Rating’s (NFDR) Fire Potential Index (FPI). These models provide the basis for state and national risk mapping and are used to allocate financial
40


resources for mitigation efforts and suppression activities. They also provide large national trend datasets for national forest and WUI policy decision-making.
The NFDRS uses current and antecedent weather, fuel types, and both live and dead fuel moisture (Stratton, 2004). NFDRS ratings are based on an adjective class rating system (Table 3.1), which is calculated by normalizing rating classes across different fuel models, indexes, and station locations to map daily fire danger (Figure 3.1). Classes are based on the following data collected at fire stations: primary fuel model, staffing levels, and climatological class breakpoints (Arroyo, Pascual, & Manzanera, 2008; Haas et al., 2013; Keane et al., 2010; Keane et al., 2000; Papadopoulos & Pavlidou, 2011). Most of these stations use the Burning Index (BI)—a measure of fire intensity (Geographic Area Coordination Centers, n.d.); however, a few use the Energy Release Component (ERC) (Keane et al., 2000). BI combines the Spread Component (SC) and ERC to relate to the contribution of fire behavior to the effort of containing a fire. BI has no units, but in general it is 10 times the flame length of a fire. SC is a rating for the forward rate of spread at the fire head (Geographic Area Coordination Centers, n.d.). SC integrates the effects of wind, slope, fuel bed, and fuel particle properties. The daily variations are caused by the changes in the wind and moisture contents of the live fuels and the dead fuel time-lag classes of 1, 10, and 100 hr. ERC is an estimate of the potential available energy released per unit area in the flaming zone of a fire (Geographic Area Coordination Centers, n.d.) and is dependent upon the same fuel characteristics as the SC. ERC—expressed in BTU’s per square foot—is derived from predictions of the rate of heat release per unit area during flaming combustion and the duration of the burning. Day-to-day variations of the ERC are caused by changes in the moisture contents of the various fuel classes, including the 1000-hour time lag class.
41


Table 3. 1. Fire Danger Rating System adjective class ratings
Fire Danger Rating and Color Code Description
Low (L) (Dark Green) Fuels do not ignite readily from small firebrands although a more intense heat source, such as lightning, may start fires in duff or punky wood. Fires in open cured grasslands may bum freely a few hours after rain, but woods fires spread slowly by creeping or smoldering, and burn in irregular fingers. There is little danger of spotting.
Moderate (M) (Light Green or Blue) Fires can start from most accidental causes, but with the exception of lightning fires in some areas, the number of starts is generally low. Fires in open cured grasslands will burn briskly and spread rapidly on windy days. Timber fires spread slowly to moderately fast. The average fire is of moderate intensity, although heavy concentrations of fuel, especially draped fuel, may burn hot. Short-distance spotting may occur, but is not persistent. Fires are not likely to become serious and control is relatively easy.
High (H) (Yellow) All fine dead fuels ignite readily and fires start easily from most causes. Unattended brush and campfires are likely to escape. Fires spread rapidly and short-distance spotting is common. High-intensity burning may develop on slopes or in concentrations of fine fuels. Fires may become serious and their control difficult unless they are attacked successfully while small.
Very High (VH) (Orange) Fires start easily from all causes and, immediately after ignition, spread rapidly and increase quickly in intensity. Spot fires are a constant danger. Fires burning in light fuels may quickly develop high intensity characteristics such as long-distance spotting and fire whirlwinds when they bum into heavier fuels.
Extreme (E) (Red) Fires start quickly, spread furiously, and burn intensely. All fires are potentially serious. Development into high intensity burning will usually be faster and occur from smaller fires than in the very high fire danger class. Direct attack is rarely possible and may be dangerous except immediately after ignition. Fires that develop headway in heavy slash or in conifer stands may be unmanageable while the extreme burning condition lasts. Under these conditions the only effective and safe control action is on the flanks until the weather changes or the fuel supply lessens.
Note: Source (United States Forest Service Rocky Mountain Research Station, 2016)
42


LEGEM D
. Reporting Weather Station i
â–¡ Low â–¡ Very High
â–¡ Moderate 1 Extreme
â–¡ High â–¡ Water
WFAS-MAP5 Graphics
(Inv. Diet. Interp.J FIRE BEHAVIOR RESEARCH
MISSOULA, MT
Observed Fire Danger Class: 22-Aug-16
Note: Source (United States Forest Service Rocky Mountain Research Station, 2016). Figure 3. 1. Observed Fire Danger Class
The NFDRS and Fire Potential Index (FPI) use fuel data sets that are based on remotely sensed vegetation classes. These classes are refined through a process called ground truthing, during which various vegetative locations are sampled to further classify typical vegetation fuel characteristics. Vegetative characteristics include percent cover, height, and diameter on the four major tree and shrub species; and percent cover and depth of subshrubs, forbs, mosses, and grasses; and shrub and grass morphology and density classes. Additional key data inputs for these models are relative greenness (RG) of vegetation and time lag live and dead fuel moisture (Keane et al., 2000; Ottmar, Blake, & Crolly, 2012; Papadopoulos & Pavlidou, 2011). The FPI
43


adequately addresses the computational complexity of the NFDRS by eliminating wind and increasing resolution of fuel modeling, NDVI derived composites of RG and RG trends, and using a 10-hour time lag fuel moisture model (Keane et al., 2000; Stratton, 2006). While US national coverage of digital elevations models (DEM) at 30 and 10 meter resolutions, and in some cases 3 meter resolutions, are available, neither FPI nor NFDRS use topography as a risk variable because topography datasets, such as slope and aspect are computationally expensive to compute and integrate across large regions. Yet, topography is critical at local scales to understand wildfire risk and behavior.
While the NFDRS is used in state and national efforts, local fire managers use historic fire weather climatology to set current and future staffing class breakpoints (Hayes, Ager, & Barbour, 2004). Using historic fire weather climatology is problematic because values between stations are estimated with an inverse distance-squared technique on a 10-km grid, which works well for areas with high station density, but does not work well in low-density areas like the WUF Additionally, research has shown that historic data is no longer an accurate predictor of future wildfire risk due to climate change (Bryant & Westerling, 2014). However, two issues preclude using the NFDRS in wildfire land use planning efforts: 1) the modeling process is computationally complex and resource intensive and 2) it is not scalable for local decisionmaking because of the reliance on fire station data and ignores critical fire risk data sets such as topography.
Behavior Modeling
Behavioral modeling, including simulator models, are used to fill the gaps of exposure modeling efforts. These models are more appropriate for local decision-making because they provide a greater level of detail and use a composite simulated approach to measuring risk.
44


Behavior models measure risk as a composite of exposure or probability of fire likelihood, fire intensity, fire effects and impacts, and fire event simulations or rate of spread (ROS) (Haas et al., 2013). The key benefits of behavioral modeling include: the ability to understand and prepare for wildfire behavior during fire suppression activities, understand the future impacts of wildfires, provide enough detail for local decision makers to identify various contextual risk factors, and evaluate the effectiveness of various mitigation scenarios (Ager et al., 2014; Finney, 2001;
Keane et al., 2000; Stratton, 2004; van Wagtendonk, 1996). Over 23 behavioral models have been identified in the literature (Papadopoulos & Pavlidou, 2011). However, two simulation models stand out: FARSITE and FlamMAP. The FARSITE simulator model stands out because FARSITE: 1) is a mature modeling environment and in use by many levels of governmental agencies, 2) supports a variety of input data and modeling parameters, 3) uses input data based on spatial data, 4) utilizes input and output data that are very detailed and multiparametric, making it more reliable and accurate, 5) handles multiple fire fronts and ignitions, and 6) can accommodate customized, high resolution input variables (Arroyo et al., 2008; Keane et al., 2000; Papadopoulos & Pavlidou, 2011; Stratton, 2006, 2009).
FARSITE simulates wildfire growth and behavior for long time periods across a variety of terrain, fuels, and weather (Papadopoulos & Pavlidou, 2011). It is a deterministic modeling system, meaning that simulation results can be compared to all model inputs, allowing for evaluation of uncertainty, error, and understanding of contextually significant variables (Carmel et al., 2009; Stratton, 2009). FARSITE uses a variety of mathematical fire models to adequately capture the complex dynamics and behavior of fire. Models include Rothermel’s (1972) surface fire spread model, Van Wagner’s (1977, 1993) crown fire initiation model, Rothermel’s (1972) crown fire spread model, Albini’s (1976), spotting model, and Nelson’s (2000) dead fuel
45


moisture model. The value that each of these models brings, is that FARSITE can be used to simulate air and ground suppression actions and land use and mitigation actions, including fuel treatments (P. Berke et al., 2015; Gebru et al., 2017; Sang, Zhang, Yang, Zhu, & Yun, 2011; Tian, Ouyang, Quan, & Wu, 2011; Yun, Chen, Li, & Tang, 2011). The process for FARSITE data acquisition is outlined in Figure 3.2.
tAHdire Tjbuioi fe
Lmdiupi lilu 1 *â–  G>op*i.al Critique
{UCPJ LLANBMRfcj of Inputs
* »DJLT Wit tM
M'lcmAI HAp LCR Cntlqu*
* WM>53
â– k- SSSSSSSoSsSSsSir
Ancillary CIS
Sv^trira
1
* HHrid Date tniimy (ntw . USFfi ArrfJlS Imijj Sirw
* ySFS Chsprlnqhnuw
* NP* Dnlr*
* NP*-U5 Mapping Prr.| Hilling Hndaat
Oang« tn it*
iC*
. fabsjtt
Calculator . WFW5
Unri-trnpa Minr
Changes to l -nj'poji^jl LCP L»r«rfc
I
â–  Paccar Chang* Toni
â–  Ara* Chang* Tnnl
â–  ArrHip or ArtWtVi
+ HTT [flairHap) + rusitr
Gaoapabal Data Maint«rj'i:i A B a: urn in till on
Note: Source (Stratton, 2009),_________________________________________
Figure 3. 2. LANDFIRE fuels data acquisition process for FARSITE.
Despite the many benefits of FARSITE, the model has three technical drawbacks: 1) the highly complex needs to customize contextual input models and variables; 2) the computational expense of high resolution inputs and outputs; and 3) the lack of updates to the model (Papadopoulos & Pavlidou, 2011). Additionally, there is a critical flaw in using behavioral models in risk mapping efforts. Behavioral models address what-ifs of particular wildfire events but have not traditionally been used to integrate the broader landscape trends of risk due to the needed computational complexity (Papadopoulos & Pavlidou, 2011; Stratton, 2009).
46


Advancements, however, have been made by Thompson et al. (2015) in their exploration of using the FSim wildfire behavior model in the development of a probabilistic method for wildfire suppression cost modeling. While their implementation of FSim facilitated the efforts of evaluating California’s level of risk at a state scale, it would be of limited use for local decisionmaking due to the simplification of risk outputs in FSim compared to FlamMAP.
FlamMAP—another behavior model—is compatible with FARSITE, and they are often used together in comprehensive modeling processes. FlamMAP uses the same datasets as FARSITE, which allows for modeling continuity, but FlamMAP focuses on the spatial variability in fire behavior, (Finney, 2006). FlamMAP fire behavior calculations are performed independently for each cell on the gridded landscape (Finney, 2006). Original spatial dataset layers are often differing resolutions; however, they must be processed into identical resolution, extent, and co-registered in order to be used in FlamMAP (Finney, 2006), a process that can introduce uncertainty and error. Additionally, the structure and types of datasets are not dynamic. Indeed fuel moisture, wind speed, and wind direction are constant in time (Finney, 2006). As a result, they cannot adequately model live fire events with changing weather patterns. Basic model outputs include: fireline intensity, flame length, rate of spread, heat per unit area, horizontal movement rate, midflame windspeed, spread vectors, crown fire activity, solar radiation, 1-hr dead fuel moisture, and 10-hr dead fuel moisture for each pixel at a given point in time (Finney, 2006). Additionally, a different suite of inputs is generated for minimal travel time calculations on basic FlamMAP outputs, including: rate of spread, influence grid, arrival time grid, fireline intensity grid, flow paths, major paths, arrival time contour, and burn probabilities (Finney, 2006). An emerging benefit of FlamMAP is its ability to evaluate pre- and postmitigation fuel treatments effects on wildfire behavior, allowing insights into mitigations effects
47


on future risk (Finney, 2006). However, this use of FlamMAP is still limited to large fires and mitigation treatments are targeted to work under a very specific set of weather and fuel moisture conditions (Finney, 2006).
CWPP Wildfire Risk Modeling
As mentioned above, wildfire behavior models can be utilized as wildfire risk models by running multiple simulations of different conditions in order to calculate risk. FARSITE and other behavioral models are utilized in this way in order to create risk models, but still do not specifically address issues of social vulnerability or community values—issues that help in prioritizing wildfire land use and mitigation decision-making, which is an important omission in risk modeling and the subsequent development of mitigation efforts and priorities. However, these models can be used in conjunction with other tools to integrate some aspects of vulnerability. For example, these models could estimate areas of high burn probability, which could then be overlaid with maps of specific types of vulnerability. Maps of low economic status, minority ethnicities and races, and affordable housing can add the missing vulnerability aspect of these models in order for planners to make decisions based on a more comprehensive context of the area. Initial efforts to overcome this gap in the literature have been undertaken by Elia et al (2014) and Paveglio et al (2013). Elia et al (2014) modeled risk using population densities to streamline the spatial allocation of fuel removals. Paveglio et al (2013) used current and projected future land use scenarios and total economic value to document and prioritize areas of higher and lower risk for mitigation efforts. Still, neither of these studies fully address issues of vulnerability, which are needed to assist in prioritizing mitigation locations.
Wildfire risk models are critical to identifying and prioritizing municipal wildfire treatment areas (Keane et al., 2010). However, current risk models lack the appropriate
48


resolution or scale of datasets, integration of vulnerability information, or community values for decision-makers to adequately evaluate and prioritize land use and mitigation decisions (Miller & Ager, 2013). Additionally, many modeling efforts do not include uncertainty and error assessments. Of the models and research reviewed, only one, FARSITE, clearly articulated and evaluated uncertainty and error (Stratton, 2009). Indeed, many modeling efforts of the last decade have largely ignored uncertainty and errors in data model inputs (Alexander & Cruz, 2013; Cruz & Alexander, 2013; Miller & Ager, 2013). While all models mentioned climate changed induced increases in risk, only one model sought to address it (Bryant & Westerling, 2014). Bryant and Westerling (2014) did account for projected weather related climate change, changing demographics, and development patterns, but they did not account for changes in vegetative land cover migration. Additionally, models need to adequately address the temporal dynamics of wildfire risk (Miller & Ager, 2013). As a result, there is need for a broader wildfire risk evaluation framework to support decision-making.
CWPPs should use scientifically appropriate modeling methods to evaluate risk, specific model choices will depend on their frame and the associated goals and objectives; however, the process should be well documented so it can be updated and reproduced in the future. Additionally, risk mapping efforts can help quantify existing and changing conditions related to the population; age; percentage of youth; percentage of elderly; number of housing units; percent of owner and renter occupied housing units; percentage of people in the labor force; percentage of families below the federal poverty line; unemployment rate; length of homeowner tenure; full-time/part-time residency status; and income at risk. These efforts will help prioritize specific goals, objectives, and project locations with the community and whether the community is meeting their target risk reductions over time.
49


Hazardous fuels reduction
The goal of fuel treatment is to modify potential fire behaviour, thus minimizing the negative effects of wildfire. Specific to CWPPs, the goal is to reduce risks to human lives and communities and improving ecosystem health. Inherent in these broad goals are the goals of reducing fire intensity, rate of spread, severity of fire effects, and restoring historic fuel quantity and structure (Husari, Nichols, Sugihara, & Stephens, 2006; Society of American Foresters, 2004). HFRA suggests CWPPs document fuel reduction priorities and projects. These priorities should be for both federal and non-federal land in the WUI, including the preferred treatment methods for each project location. Additionally, a list of best practices should be provided, so individuals and the community can further reduce risk. Hazardous fuels reduction goals should be clearly articulated and indicate wither priority projects serve to protect the community and its essential infrastructure or are geared toward reducing risks to other community values (Society of American Foresters, 2004). Priority project lists should also have timelines and reoccurring evaluation periods because fuel reduction efforts are not a static, one-time undertaking (Bums & Cheng, 2007; A. M. S. Smith et al., 2016). CWPPs should fully articulate fuel management basics and treatment types, which are outlined below.
Fuel Management Basics
Fuel
Fuel is defined as live and dead plant biomass (Ager et al., 2011; Arroyo et al., 2008; Husari et al., 2006). Fuel moisture, chemical composition, surface area to volume ration, size, and structural arrangement of the fuel in the stand and on the landscape influence the conditions under which fuel will bum (Husari et al., 2006). Additionally, these characteristics will also determine the characteristics and nature of the resulting fire. Fuel management is the intentional
50


act of manipulating the amount, composition, and structure of fuel within wildland ecosystems for the express purposed of modifying potential fire behaviour and effects (Agee et al., 2000; Ager et al., 2014; Ager et al., 2011; Arroyo et al., 2008; B. M. Collins et al., 2013). Fuel has 15 characteristics that can be modified to influence its potential to burn and the characteristics of the result wildfire, including: total fuel quantity, fuel size, packing ratio of surface fuel, surface fuel continuity, crown fuel continuity, surface fuel, crown fuel, horizontal fuel continuity, vertical fuel continuity, ladder fuel, potential for surface fire, and potential for crown fire (Husari et al., 2006).
Fuel Quantity
The quantity of fuel in an ecosystem is an important factor to determining the character and impact of fires (Stratton, 2004). The metric used to describe the amount of fuel is dry weight per unit area (tons/acre) (Ager et al., 2012; Arroyo et al., 2008; Husari et al., 2006). This dry weight is divided into size classes. Each size class is based on the time the fuel takes to reach equilibrium with moisture in the air (Arroyo et al., 2008; Husari et al., 2006). For example, small fuels (e.g., pine needles) respond to changes in relative humidity more rapidly than large dense logs. Fuel can be removed from a site by a variety of means, thus reducing fuel quantity.
Fuel Size
Fuel particle size is important to determining the likelihood of ignition and determining the resulting fires behaviour and effects (Jack D. Cohen & Finney, 2010). Fine fuels, fuels less than a quarter inch in diameter, have the greatest influence on the ignition and spread of fires (Husari et al., 2006). Since fine fuels are integral in fire ignition the removal of fine fuels is often a primary focus of fuel management projects (Jack D. Cohen & Finney, 2010).
51


Packing Ratio
Packing ratio is the measure of how densely packed fuel particles are (Husari et al.,
2006). Fuel may be compacted through a variety of mechanical treatments including mastication, chipping, and shredding. Compacted fuel bums more slowly because the oxygen required for combustion is not readily available to the fuel away from the surface (Husari et al., 2006).
Surface Fuel
Surface fuel is composed of small shrubs, grasses, and plant detritus lying on the ground (Agee & Skinner, 2005). Surface fuel is necessary for fire to spread continuously across landscapes. Fuel management goals are achieved by disrupting the continuity of surface fuels (Husari et al., 2006).
Crown Fuel
Tree branches and foliage and shrubs over six feet in height are considered crown fuel (Cruz, Alexander, & Wakimoto, 2003). Continuous crown fuel is required to spread fire through the tree canopy as a crown fire. Crown fires can spread in discontinuous stands of trees if supported by surface fire. Wind speed and foliar moisture are important to the spread of crown fires. Removing trees and ladder fuels and treating surface fuels reduces crown fire risks because they reduce the continuity and bulk density of crown fuels, while increasing the separation between crown and surface fuels (Agee & Skinner, 2005; Ager et al., 2014; Husari et al., 2006).
Horizontal Fuel Continuity
Horizontal fuel continuity is necessary for a surface or crown fire to spread laterally across the landscape (Husari et al., 2006). Surface discontinuities act as barriers to fire spread under most conditions; however, under extreme conditions fires can spot across bare areas. Fuel
52


treatments that aim to disrupt fuel continuity include: fuel breaks and strategically placed area treatments (Finney, 2001).
Vertical Fuel Continuity
Vertical fuel continuity is necessary for surface fires to spread into the crown or canopy of trees (Finney, 2001; Husari et al., 2006). Vertical fuel continuity can be reduced to increase the separation between surface and crown fuels, thus reducing the probability of crown fires (Agee & Skinner, 2005; Finney, 2001; Husari et al., 2006).
Ladder Fuel
Fuel, such as intermediate sized trees or shrubs, that provide a fuel conduit that allows a surface fire to ‘climb’ into crown fuel (Cruz et al., 2003; Husari et al., 2006; Riccardi et al., 2007). Fuel treatments should remove shrubs, small trees, and lower branches to reduce ladder fuels (Cruz et al., 2003; Husari et al., 2006; Riccardi et al., 2007).
Types of Fuel Treatments
Fuel treatments take on a variety of forms, but are generally divided into two treatment categories: fire and mechanical (Husari et al., 2006). The use of fire as a treatment method is beneficial because it is a key process in many plant ecosystems in the American West and also modifies fuels (North, Collins, & Stephens, 2012; Oldham, 2016; Vaillant & Reinhardt, 2017). Mechanical treatments, including forest thinning, mastication, and grazing, also modify fuels. Mechanical treatments are often used to restore fuel conditions where fire can be used to maintain the desired range of conditions over a longer period of time. HFRA directs half of federal fuel reduction funds to be used in the WUI even though fuel reduction treatments in the WUI is more expensive and limits the amount of fuel treatments elsewhere (Headwaters Economics, 2014). Headwaters argues that prescribed fires should be used more extensively in
53


mitigating wildfire (Gorte, 2013). It is estimated that 230 million acres of non-WUI Forest Service and Department of Interior lands are in need of treatment because they are at risk from ecological damage from wildfire (Headwaters Economics, 2014). However, less than three million acres are treated per year (Headwaters Economics, 2014), which is insufficient to reduce wildfire risk. It is unlikely there will be a dramatically increased acreage receiving fuel treatments in the near future due to budgetary and political constraints.
Fire Treatments
Prescribed Fire
Prescribed fire is an intentionally ignited fire allowed to bum in desired locations under certain conditions to modify fuels (Husari et al., 2006). Prescribed fire has been a supplement or in some cases, a replacement to natural sources of ignitions. It is important to distinguish between restoration and maintenance burns. Restoration bums modify the current ecological condition to a preferred state, while maintenance bums maintain ecological conditions within a specified range. Modifications may include the decrease of hazardous quantities of dead and downed fuel, the stimulation of fire-dependent species, improvement of range conditions, or the creation of wildlife habitat (Husari et al., 2006; Parks et al., 2015).
It is critical to consider the variables that influence a fires behaviour, the ecological role of fire, and the ability to control the fire - minimizing potential for escapes before a prescribed burn is initiated. Specific site considerations include: slope, aspect, topographic position, and role of fire in the project area (Husari et al., 2006; Reinhardt, Keane, Calkin, & Cohen, 2008). Specific conditions conducive to the use of prescribed fire include: season, weather, fuel conditions, and the availability of qualified personnel. It is also critical to establish to establish measurable objectives and means of monitoring them. The value of prescribed fire to land
54


managers decreases significantly with the inability to quantify the purpose of the fire and its accomplishments because prescribed fire within an adaptive management context is critical for each land manager.
Wildland Fire
Wildland fires, under the correct, conditions can serve the same functions as prescribed fires. While wildland fire has been recognized as beneficial, as previously state in Chapter 2’s discussion of the USFS let-bum policy, it is often only used under narrow conditions (Fire History Society; Manning, 2012). The fires that are allowed to burn are often categorized as within historic or natural ranges of variability and are of minimal risk to human settlement and lives. To allow a wildland fire to bum fires need to meet additional planning approvals, and implementation requirements (Husari et al., 2006). The primary issue of concern is air quality and minimizing its impact on public safety.
Mechanical Treatments
Mechanical treatments remove, rearrange, or modify biomass; however, the effectiveness of differing mechanical treatments reducing wildfire risk are contentious (Agee et al., 2000). Equipment such as feller bunchers, skidders, and grapplers are used to thin the forest to various densities, thus removing live and dead woody fuel. The characteristics of this fuel are changed by crushing, chipping, shredding, or chopping it. The material is then either removed from the site or piled and burned under safer, localized conditions Mechanical methods can be more precise; however, they have two large drawbacks. First, the use of mechanical treatments removes organic material which reduces the amount of carbon and nutrients on site (S L Stephens et al., 2012). Second, mechanical treatments often still need the follow application of fire to maintain fire adapted ecosystem health (Husari et al., 2006; S L Stephens et al., 2012).
55


Forest Thinning
Thinning modifies the fuel structure of forests by reducing the quantity and density of vegetation. This vegetation modification serves to reduce the fuel load within wildland vegetation for potential wildfires as well as to defensible space around dwellings. Thinning can be effective at moderating crown fire behaviour and often has some economic return on investment; however, thinning is only effective when fine fuels are also reduced. Thinning activities can remove trees to create forests with specific stand densities, patterns, distributions, and species compositions. Treatments should specify target densities by tree diameter classes of trees. Wildfire modeling has shown in simulations that thinning impacts the wildfire risk via reduction of wildfire fuel loads (Cochrane et al., 2012; Paveglio et al., 2013; Stratton, 2004). These efforts are effective because thinning often limits the ability of surface fires to transition to crown fires by breaking up vertical and horizontal fuel continuity.
Mastication
Mastication is the mechanical chopping, chipping, grinding, crushing, and shredding of fuels to reduce fireline intensity and rate of fire spread (Scott L. Stephens & Moghaddas, 2005). Mastication reduces potential fire behaviour by reducing the fuelbed depth, thus increasing the fuels packing ratio (Husari et al., 2006; Scott L. Stephens & Moghaddas, 2005). Due to the mechanical equipment used, mastication can be very precise; however, these techniques are only recommended during ecosystem restoration efforts because the presence of heavy equipment is more damaging and less beneficial than the use of fire in ongoing wildland ecosystem maintenance (Kane, Varner, Knapp, & Powers, 2010; Kreye, Kobziar, & Zipperer, 2013).
56


Grazing
Livestock grazing has been effective at reducing surface fuels and modifying the rate of fire spread (Scott L. Stephens & Moghaddas, 2005); however, grazing is limited in its scale of application. Still, its use as a fuel break maintenance tool and other linear fuel reduction projects has proven significant. Specifically, the use of animals grazing of fine fuels can shorten the fire season and reduce fire potential. Additionally, private landowners who own grazing lands are strong advocates for using fire to promote grazing land ecosystem health, which provides additional fuel modification benefits.
Defensible Space
Defensible space uses a combination of the above fuel treatment types to create a perimeter around buildings and structures with modified vegetation cover to reduce fuel for potential wildfires. It also provides firefighters a clear environment in which to maneuver to protect structures. Additionally, defensible space reduces the chance that an initial structure fire will spread into the surrounding area to create a wildfire (Gill & Stephens, 2009). Defensible space is typically separated into prescribed zones; whose definitions vary by agency and state. The HFRA CWPP documentation do not specify defensible space zone, distance, or vegetation requirements. As a result, implementations of defensible space are various and inconsistent. However, the National Fire Protection Association (NFPA) is promoting a set of Firewise community defensible space standards, in which defensible space is created in the home ignition zone - the area within 200 feet of the house (National Fire Protection Association, 2016a). In fact, the term “home ignition zone” is now used as a replacement for “defensible space.”
The home ignition zone is often subdivided into three zones. The following zone measurements are for flat ground only; zone distances should increase on steeper slopes, but
57


specific guidelines for slopes are often not provided in local ordinances. It is worth reiterating that CWPPs typically only include two defensible space zones. Zone 1 is within 30 feet of a structure vegetation should be carefully spaced, low-growing and free of resins, oils and waxes that bum easily. Zone 2 is from 30 to 100 feet of the structure, plants in this zone should be low-growing, well irrigated and less flammable. Zone 3 is from 100 to 200 feet, vegetation in this zone should be thinned. Each zone has specific vegetation spacing and composition requirements (Table 3.2). It is important to note that any defensible space zone can be truncated at a private landowner’s property line. Whether homeowners engage in defensible space mitigation is tied into their concerns about privacy, desired natural aesthetics, wildlife and recreational values, physical and economic capacity, and the individual’s knowledge (Brzuszek & Walker, 2008; Winter et al., 2009).
58


Table 3. 2. NFPA Firewise Community Recommended Defensible Space Requirements.
Defensible space zone Distance Vegetation, fuel thinning, and reduction requirements
Zone 1 0-30’ • Plants should be carefully spaced, low-growing and free of resins, oils and waxes. • Lawns should be mown regularly. • Trees should not overhang structures, be pruned up six to ten feet off the ground, and conifers should be spaced 30 feet between tree crowns. • Create a ‘fire-free’ area within five feet of the home, using non-flammable landscaping • materials and/or high-moisture-content annuals and perennials. • Remove dead vegetation from under deck and within 10 feet of house. • Water plants, trees and mulch regularly. • Consider xeriscaping if you are affected by water-use restrictions.
Zone 2 30-100’ • Leave 30 feet between clusters of two to three trees, or 20 feet between individual trees, and prune trees up six to ten feet off the ground. • Use a mixture of deciduous and coniferous trees. • Create ‘fuel breaks’, like driveways, gravel walkways and lawns.
Zone 3 100-200’ • Thin this area, reducing the density of tree canopy. • Remove smaller conifers that are growing between taller trees. • Remove heavy accumulation of woody debris.
Note: Source - (National Fire Protection Association, 2016a).
Regardless, of the mitigation technic used, local, state and federal fuel reduction efforts have failed because they are treating too few acres per year (Vaillant & Reinhardt, 2017). In all regions of the U.S. annual acres disturbed by wildfire and treatment are much lower than historically burned (Parks et al., 2015; Vaillant & Reinhardt, 2017). These efforts are costly, and both federal and state mitigation budgets have been consumed by fire suppression efforts, with little political will to expand the budget to what is necessary for significant risk reduction efforts (A. Berry, 2007). The levels of fuel reduction on federal lands has been insufficient, the federal
59


government have only managed fuel reduction on 3 million of the 230 million acres that need fuel load reduction, this number does not address the millions of acres of private land that need mitigated (Gorte, 2013). Budgets and political will are further eroded due to several out of control high profile prescribed burns lawsuits. This is concerning considering the use of prescribed fire, in combination with mechanical thinning techniques has proven very effective in reducing catastrophic wildfires (Agee et al., 2000; Forthofer, Butler, & Wagenbrenner, 2014; Kalies & Yocom Kent, 2016; North et al., 2012; S L Stephens et al., 2012; Wales, Suring, & Hemstrom, 2007).
Reducing structural ignitibility
WUI fires ignite homes in two principal ways: direct flame heating or firebrand ignition. The principal approach to reducing structural ignitions is to lessen the ignitibility of the home ignition zone and increase survivability (Jack D Cohen, 2001). The home ignition zone includes the home and an area surrounding the home within 100 to 200 feet (Jack D Cohen, 2001). CWPPs should identify regulatory and non-regulatory strategies to reduce structural ignitibility (Jakes et al., 2007; Jakes et al., 2012; Jakes et al., 2011; Society of American Foresters, 2004). These efforts should include actions communities and individuals can take. Local governments play a pivotal role in reducing structural ignitibility because in the United States, local governments are responsible for land development regulations. However, some state governments have developed minimal regulations because in some situations local codes are non-existent or insufficient. Regulatory structural ignitibility mechanisms include: zoning regulations, development standards, building codes, fire prevention codes, and fire response. Local government actions are discussed in the subsection below. Individual responsibility suggestions should include: fire safe construction practices, private property forest thinning, and
60


defensible space. While these are individual efforts, they can also be regulatory, as such each of these will be described in great detail in the subsection below.
Education and outreach
Education and outreach are crucial to broadening CWPP support and implementation. State and local agencies play a pivotal role in wildfire outreach and risk reduction education efforts. Each state has multiple agencies that manage land where wildfire can occur; however, usually a single agency is responsible for supporting and guiding state wildfire efforts in the wildland and WUI. Education efforts emphasizes raising risk awareness and preventive measures. Risk awareness efforts refers to fire weather watches and risk mapping. Education efforts refers to various forms of multimedia campaign efforts, workshops, field trips, reports, and how-to guides. Fire weather watches, which are issued by the National Weather Service to alert all fire agencies of the onset, or possible onset of critical weather and dry conditions that could lead to rapid or dramatic increases in wildfire activity. The ratings are shown in Table 3.1 and broadcast to the public via television, radio, and public postings outside of fire stations and other state and federal agency offices.
Numerous studies show that being aware of risk is a necessary condition of mitigation decision making; however, it needs to be partnered with additional regulations and incentives to reduce risk (Nielsen-pincus et al., 2015; Scott L. Stephens & Collins, 2007; M. P. Thompson et al., 2012; Austin R. Troy, 2001). Education, in combination with risk mapping, plays a pivotal role in alerting the public about the dangers of living in wildfire-prone areas, and the importance of mitigation efforts. While each state is a key facilitator in creating and disseminating education material, they increasingly partner with universities, local government, and other nongovernmental fire risk reduction programs, such as Firewise. Firewise is a National Fire
61


Protection Association outreach effort that intends to teach people how to adapt to living with wildfire and encourages neighbors to work together and act to prevent loss (National Fire Protection Association, 2016b). Collectively each of these groups relay publications that are tailored to a geographic area to promote hazard reduction efforts, including fire protection and safety, landscaping and defensible space guidelines, lists of recommended fire-resistant plant species, and residential building guidelines. Published material is disseminated through websites, brochures, pamphlets, community events, mailings, news outlets, blogs, videos, and podcasts. Classroom and teacher efforts are also part of the education. In several states, a fire science is a part of the science curriculum in K-12 education, using multimedia to educate students on wildfire ecology, safety, and protection. Fire protection officials have developed their own educational programs. These programs include hands on defensible space and fire safety courses for grade school students. Those targeting high schools have involved fuel removal around schools and field exercises, such as risk assessments and community fire-risk mapping. Emergency management capacity
While the majority of CWPP content is geared towards pre-fire planning efforts, suppression activities are still critical to preserving community values, property, and lives. As such, CWPPs should assess local preparedness for wildfire and firefighting capabilities. These efforts should assess the community’s emergency preparedness, including evacuation planning (people and agricultural livestock), safety zones, and fire assistance agreements as well as the response capability of community and cooperative fire protection forces (Jakes et al.,
2007; Jakes et al., 2012). Documentation should include gaps in and training for incident command and the number and percentage of homes in each fire district. Emergency management capacity should also document the number and per percent of trained and/or certified fire fighters
62


and crews; fire suppression equipment; and response times. Local preparedness information should be incorporated into the base map as appropriate (Society of American Foresters, 2004). Long-term success
CWPPs should not be static, as such it is critical to ensure their long-term success by sustaining community engagement and support. Jakes et al (2012) suggest four strategies to ensure long-term success: 1) incorporate projects into the CWPP that can be accomplished quickly to foster homeowner buy-in and broaden long-term support; 2) nest local CWPPs within broader plans to augment resources, support, and implementation; 3) incorporate CWPP into formal government structure and processes; and 4) quickly identify changes that affect the CWPP and adapt accordingly. Additionally, CWPPs should suggest update timelines because new and aging developments and evolving fuel loads alter wildfire risk overtime. Assessing CWPPs is critical to maintaining their relevance and effectiveness over the long-term, especially under evolving conditions (Jakes et al., 2012; Society of American Foresters, 2004). CWPPs incorporated into local ordinances and codes gain efficiencies and relevance, thus ensuring longer term support and enforcement. Additionally, incorporating CWPPs will also better leverage community resources, achieve multiple adjectives, increase political acceptance of mitigation activities, and provide a consistent message (H. Brenkert-smith, Meldrum, Champ, & Barth, 2017; Clark & Stankey, 2007; T. W. Collins, 2008b; A. M. S. Smith et al., 2016). Mechanisms for incorporating CWPPs into local governance are discussed in the following section.
Local government wildfire risk reduction efforts
Local government wildfire risk reduction efforts, or options for potential reduction efforts, will be presented in this section. Local governments are critical partners in the CWPP
63


process. Research has documented land developments contributions to elevated wildfire risk and shown that well integrated comprehensive and hazard planning can reduce said risk (Bhandary & Muller, 2009; Brzuszek & Walker, 2008; Butsic, Kelly, & Moritz, 2015; Headwaters Economics, 2014, 2016b; Syphard, Bar Massada, Butsic, & Keeley, 2013). Additionally, zoning codes, development standards, subdivision design guidelines, building codes, and plan review and inspection procedures can create safer communities (Greiving, Fleischhauer, & Liickenkotter, 2006; Headwaters Economics, 2014, 2016b; Jakes et al., 2012; Kocher & Butsic, 2017; National Fire Protection Association, 2016a, 2016b; Paveglio et al., 2013; Reams et al., 2005; Scott L. Stephens & Collins, 2007; Syphard et al., 2013). Since CWPPs are not regulatory documents or policies, it is imperative to deeply integrate CWPPs and local government land development controls. This chapter describes local government best practices for planning and land development policy efforts to reduce wildfire risk. These efforts are described below and include the following sections: 1. comprehensive plan, 2. zoning codes and development standards and 3. fire-resistant materials.
Comprehensive Plan
Comprehensive plans are the result of a process that assists communities in planning pleasant, liveable, safe, and well-ordered urban environments (Mandelker, 1976). Due to their anticipatory approach to future land development, comprehensive plans, have the capacity to steer growth and development away from hazard-prone areas, restrict land uses in sensitive areas, locate public infrastructures away from hazard areas, and impose building standards that reduce the vulnerability of structures (Schwab, Meek, & Simone, 2005; Srivastava & Laurian, 2006). However, the framing of comprehensive plans is critical to whether hazards are considered a key driver to the planning effort. Indeed, research has shown that comprehensive
64


plans have framed growth within the context of flood and droughts; however, they have not equally done so for wildfire (Schwab et al., 2005; Srivastava & Laurian, 2006). Additionally, plans that have framed the process solely in the interest of economic development have often downplayed all hazard risks and environmental concerns (P. Berke & Godschalk, 2009; Samuel D Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Srivastava & Laurian, 2006). This is concerning because hazards that are onset quickly and have localized impact, such as fire, can be most effectively and sustainably addressed through land use controls and polices that discourage or regulate development in high-risk areas (P. Berke & Godschalk, 2009; Samuel D Brody,
2003; Butsic et al., 2015; Deyle & Smith, 2000; Godschalk et al., 1998; Greiving et al., 2006; Headwaters Economics, 2014, 2016b; Paveglio et al., 2013; Srivastava & Laurian, 2006). Additionally, comprehensive plans can also prepare for effective post-disaster recovery and reconstruction to minimize future losses from repeated events (Srivastava & Laurian, 2006). Finally, they can promote risk awareness, education, and community capacity building (National Fire Protection Association, 2016b; Schwab et al., 2005; Srivastava & Laurian, 2006). As such, comprehensive plans should be sophisticated enough to represent multiple frames, such as economic growth, environmental concerns, and hazard risks, particularly wildfire.
Comprehensive plans should include strategies to reduce the extent and severity of fires impact on communities by reducing structural ignitibility and facilitating responses to fires (Headwaters Economics, 2014; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006; Scott L. Stephens & Collins, 2007). Broadly these strategies employed in comprehensive plans should include development controls for wildfire prevention, open space preservation, critical resource (e.g. water supply, hospitals, fire stations, and elder care facilities) identification and protection, and enforcement of zoning and building codes process
65


development (Headwaters Economics, 2014, 2016b; National Fire Protection Association, 2016b; Srivastava& Laurian, 2006).
Specific wildfire interventions that comprehensive plans should discuss fall into three categories: wildfire prevention, reducing wildfire impacts, and facilitating emergency response. Wildfire prevention includes: best forestry management practices, forest fire fuel reduction efforts, vegetation management, urban forestry management, and wetlands protection (Headwaters Economics, 2014, 2016b; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006). Reducing wildfire impacts includes development controls (e.g. open space preservation, building codes, performance standards, density controls, design review guidelines, environmental review standards, hillside development standards, and subdivision guidelines), property protection, and public awareness (Headwaters Economics, 2014, 2016b; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006; Scott L. Stephens & Collins, 2007). Facilitating emergency responses should include hazard recognition, warning systems, emergency response services, and post-disaster mitigation (Srivastava & Laurian, 2006).
Zoning Codes and Development Standards
Zoning is used to regulate land uses in order to prevent incompatible adjacent land uses, undue density and traffic congestion, restrict height and size/bulk of buildings, provide setbacks to lessen fire hazard and promote aesthetic value (Metzenbaum, 1957; Whitnall, 1931).. Zoning regulations can dictate development density, size of parcels, open space requirements, provision of adequate light and air, and efficient public infrastructure (e.g., water, sewer, schools, police, fire, and transportation) (Metzenbaum, 1957; Whitnall, 1931). Additionally, local zoning ordinances have traditionally attempted to guide development away from hazardous areas, minimize sprawl, or to minimize land use adjacency conflicts (Butsic et al., 2015; Headwaters
66


Economics, 2016b; Kocher & Butsic, 2017; Syphard et al., 2013). Zoning is administered through three components: 1. the zoning map, 2. zoning text, and 3. the comprehensive plan as discussed above.
The zoning map is where the zoning becomes applicable to individual property owners. The map is often color-coded to identify what types of land uses are allowable at what locations. To reduce wildfire risk, the zoning map should direct development away from high-risk wildfire areas (Headwaters Economics, 2016b; Hughes & Mercer, 2009; Syphard et al., 2013). Overlay zoning provides a set of standards that apply to properties within a specific geographic area, superseding the underlying base standards of a given zoning district. This is an instrumental tool in avoiding potential conflicts between 1. resource protection and 2. forest thinning or defensible space requirements. For example, wildfire intensity is dictated by fuel loads and topography, so a zoning overlay could disallow developments on steep, fuel loaded slopes.
The zoning text describes the exact regulations being implemented within a given land use classification (Metzenbaum, 1957; Whitnall, 1931). This document is adopted as law by a local governing body, in the case of this research it is implemented through county governments. At a minimum, the text establishes the different zone classifications within the county and the uses allowable within each zone either by right or with a conditional use permit (Metzenbaum, 1957; Whitnall, 1931). Zoning text also defines key definitions, various requirements for building setbacks, open space, landscaping, parking, height restrictions, and procedures for zoning processes. The zoning text should coordinate with the CWPP by providing defensible space requirements, use of open-space as community fuel breaks, and provide documentation of safe development patterns and densities within the applicable zoning classifications (Headwaters Economics, 2014, 2016b; Hughes & Mercer, 2009). Appropriate zoning in the WUI can also
67


ensure each home has the ability to adequately implement defensible space zones (Headwaters
Economics, 2014, 2016b; Hughes & Mercer, 2009).
Development standards are land use regulations that determine the quality of development. For wildfire, development standards include adequate water supply, defensible space, resource protection, and ongoing maintenance (Colburn, 2007; Headwaters Economics, 2016b). Subdivision regulations determine how lots are created and divided, including layout standards for new subdivision developments. For wildfire these include adequate water access, water supply, and mitigation requirements (Colburn, 2007; Headwaters Economics, 2016b). Some WUI communities are also implementing WUI specific codes that enforce required codes for buildings, landscapes, and lot development.
Fire-Resistant Materials
Very few comprehensive laws, statutes, or building codes exist that address combustion-resistant building materials (Scott L. Stephens & Collins, 2007) even though using combustion-resistant building materials is key in the survival of structures during wildfires. Possibly due to enhanced public perception and increasingly stringent building codes, the construction industry and homeowners are adapting fire-resistant materials (T. W. Collins, 2008b). Fire-resistant building materials and practices include fire-rated roofing, siding, and decking; closed eaves and soffits; and protecting vents and windows (Brzuszek & Walker, 2008; Meldrum et al., 2015; Reams et al., 2005). An effective wildfire mitigation response should include all of these elements within local building codes. However, while fire-resistant material is being required in new construction, these efforts do not address the thousands of existing non-fire-resistant homes. Currently, homes without appropriate fire-resistant construction materials are not required to be brought up to code. Accurately accounting for how many homes that are not fire resistant is
68


difficult because collecting the data requires in the field assessments and to-date cannot be done remotely. Fire-resistant construction does not justify continued development in the WUI and does not address the subsequent suppression costs (Headwaters Economics, 2014).
A number of tools exist for local governments to integrate CWPPs in order to reduce wildfire risk. Comprehensive plans, zoning codes and development standards are all all tools that would be utilized if best practices were followed in wildfire reduction efforts. In the following chapters, results of analyzing CWPPs and local governance structures will be discussed, questioning to what extent best practices are being followed across the American West.
69


CHAPTER IV
RESEARCH DESIGN AND METHODOLOGY
The research design and methodology that are used in this research effort is described throughout this chapter. Sections in the chapter include a short description of the ethical implications, pilot study results, a description of the study setting, and the research design. The research design is split into the following subsections: dependent and independent variables, study sample, and sample descriptive statistics. This chapter concludes with a brief discussion of the study’s data analysis methodology.
Research Permissions and Ethical Considerations
Ethical issues were considered and addressed at each phase of the study. This research study falls under the classification of non-human subject research because this research uses secondary, publicly available data and records: CWPP reports, county government ordinances and policies, census, satellite remote sensed data, and open access GIS data.
Pilot Study Results
The initial impetus of this research was to study municipal or neighborhood level CWPPs within CO. Before embarking on the full study, a pilot study was conducted for Boulder County, CO to document the process, documents needed, test both document coding instruments, and evaluate error and uncertainty. While the process and intercoder reliability achieved acceptable intercoder reliability, the uncertainty and error introduced by the mismatch between CWPP geographic boundaries and census block groups proved unacceptable. The full pilot study writeup can be found in Appendix C. As such, the scope of the project evolved to focus on County level CWPPs across the American West. This proved beneficial because counties are better match to the CWPP frame of interest and the landscape scale risk reduction (Jakes et al., 2007;
70


Jakes et al., 2012) while also potentially providing a greater diversity in instrument - the pilot study results suggested regional homogeneity in constructing sub-county level CWPP documents.
Study Area
The National Association of Foresters (NASF), estimates that 76,934 communities in the United States are at risk of wildfire with more than 17,655 communities covered by CWPPs (National Association of State Foresters, 2017). The NASF western region is estimates that 7,587 communities are at risk of wildfire with 6,357 communities covered by CWPPs (National Association of State Foresters, 2017). This study focuses on wildfire mitigation in the American West and utilizes a representative sample to understand the socio-economic, demographic, and biophysical relationships to CWPP objective implementation. The American West is defined as containing the following states (number of counties for each state are located in parentheses): Arizona (15), California (58), Colorado (64), Idaho (44), Montana (56), Nevada (17), New Mexico (33), Oregon (36), Utah (29), Washington (39), and Wyoming (23) for a total of 414 counties (Figure 4.1).
71


American West Counties Non-American We.-t States States defined as American West
Spatial Reference PCS: USA Contiguous Albers Equal Area Conic
0 500 1,000 1000
Kilometers
1 - Washington 4 - Idaho
2 - Oregon 5 - Nevada
3- California 0 - Montana
7 - Wyoming 10 - Colorado
8- L'tali 11 - New Mexico
9 - Arizona
Figure 4. 1. Study states, defined as the American West and associated county boundaries.
The American West was chosen because it has experienced drastically changing wildfire conditions due to climate change, fuel build-up, and residential development. Indeed, large fire activity has increased dramatically since the 1970s. Since the 1970s, the frequency of large fires has increased most dramatically in the forests of the Northwest, Northern Rocky Mountains, Southwest, Southern Rockies, and Sierra Nevada (Figure 4.2) (Schoennagel et al., 2017; Westerling Anthony, 2016). Specifically, climate change has caused earlier snow melt, rising
72


temperatures, and increased drought across the American West, causing an average increase of the fire season by 2.8 months since 1970 (Westerling Anthony, 2016). These changing conditions in addition to a large estimate of communities with CWPPs provide a robust study area for evaluating CWPPs.
Northwest Northern Rocky Southwest Southern Rocky Sierra Nevada Mountains Mountains
American West Forest Geographies
Figure 4. 2. Percent increase in large fire activity in the American West.
Research Design
Independent and Dependent Variables
Dependent Variable Data
The dependent variables consist of two indices “CWPP Process and Plan Evaluation Instrument” and “CWPP Implementation: Local Governance Evaluation Instrument,” and a composite of the two. The instruments used to code CWPPs were created through an extensive literature review, as discussed in Chapter III. The instruments are located in Appendix A and B.
73


The instalment scores are ordinal, which are converted into letter grade scores ranging between 0 and 100 percent (see Table 5.6 for letter grade conversion details). Indexing CWPP processes required the following document data: CWPP reports and CWPP process meeting minutes. County CWPP reports and process meeting minutes were downloaded from the internet using a Google search on a sampled county by county basis. Appendix D contains a CWPP and meeting minutes bibliography. Indexing CWPP implementation is twofold: policy and physical environment interventions. The policy interventions required document data in the form of: local building and zoning codes, HOA guidelines, subdivision design guidelines, design review processes, and comprehensive plans. Again, each sampled county was searched using Google search for all relevant documents and downloaded. Appendix E contains a bibliography for local governance CWPP intervention documents.
The physical intervention data was evaluated by change in WUI area and the change in average WUI population density between 2000 and 2010 per each county. This dataset was obtained from Radeloff et. al.’s, (2017) dataset. This dataset was chosen because it is the only dataset that provides uniform and comprehensive coverage of the American West. The processing of this dataset is described in the sampling section of this chapter. However, additional processing was required to calculate the change in median population density within areas designated WUI. This included spatially joining the WUI dataset to the county sample dataset and summarizing each type of WUI (WUI intermix and interface) for 2000 and 2010. Once these values were calculated, the delta change for each were calculated.
Independent Variable Data
The independent variable data consists of US Census data attributes at the county level for the 2000 (pre-HFRA) and 2010 census. These variables include age, length of homeowner
74


tenure, full-time/part-time residency, and income. The first task of aggregating the independent variables was to obtain 2000 and 2010 census attribute tables for each variable from the IPUMS National Historical Geographic Information System initiative (NHGIS). Metadata for attribute tables and map layers are given in Appendix F. Second, each variable’s attribute table was joined to the sampled county boundary file, using each county’s census unique identifier. Non-sampled county attribute information was removed, by using the keep only matching records join option. This data was exported to create a new permanent joined dataset. Due to the large number of fields within the attribute table, all extraneous, non-variables fields were deleted. New fields were added, and the delta change values of each variable - age, length of homeowner tenure, full-time/part-time residency, and income - min, median, and max were calculated. The delta change values for each county were used for statistical data analysis.
Study Methodology and Sample Sampling Strategy
To understand the integration of CWPP best practices as well as the socio-demographic and biophysical relationships in the CWPP process and implementation, this study selected a representative sample of 1) economic status of the American West population and 2) CWPP coverage. Obtaining a representative sample was difficult given the size and heterogeneity of the American West. To accomplish this goal of representation, a cluster sampling methodology was utilized involving 1) filtering of counties with WUI and CWPPs, 2) a stratification of counties based on 2010 median household income levels, and 3) a simple random sample of cases from each stratum (Figure 4.3). The sampling strategy was designed in this way to reduce selection bias as well as standard errors of the population.
75


Filter for WUI and CWTP Filter counties by the jncsence of WUI raid a CWPP
iM>n-VVI ] aiKL'ni' \ i CWPP County J l with a C WFP J
Stratifirailan Slnili ly liy iinmiMtilled cruuily riiedifsu lioiLsehciltL ttkxhiib (2010 Camix)
l > f i
f X f \
Struts 1 Simla 2 J f Simla 3 J
Simple Rimdom Sample_______________________
Forendl tOhIh a simple miL'.l'.'ju suupk
tvas used to select ease CWTFS uoiu cadi suara.
â– v V
f Strut* I V / Strain 2 X f Stratu 3 X
V Sutuple CWFFs J ( Sample CWFFs J ^ Sample CWFFs )
Figure 4. 3. Simple, stratified random sampling framework.
Input of Data Needed for Sampling
The first task of the sampling phase was to input map layers that would be needed for obtaining a representative sample. Metadata for map layers is given in Appendix F. First, state boundary files were obtained, from the 2010 US Census TIGER data set compiled by the IPUMS National Historical Geographic Information System initiative (NHGIS). The next set of data obtained was the 2010 US Census US county boundaries. To obtain only the American West counties required a two-step process. First, the American West states were selected, followed by a spatial selection of all counties that are within American West states and a new file was created. The next set of data obtained were the counties with CWPPs. There is no national list of CWPPs. This data set was created by searching state-wide CWPP lists and searching Google for specific county CWPPs. The presence or absence of a CWPP was coded as 1 or 0, respectively.
WUI classifications were obtained from Radeloff et. al.'s, (2017). The WUI data was clipped to the American West states and areas were re-calculated. Because of the prohibitively
76


large size of this file - it mapped list everything it mapped here - a new file was created with only the WUI classification polygons and relevant fields in the attribute table. The amount of WUI area, interface and intermix, for 2000 and 2010 were calculated and aggregated to each county, including the delta change in WUI. However, it was decided that the mere presence of any kind of WUI, interface or intermix, was sufficient to be considered WUI, as either designation could be integrated into a CWPP.
Next the needed county level income information was acquired. County median income data was gathered for the 2000 and 2010 census using the IPUMS NHGIS database (Manson, Schroeder, Riper, & Ruggles, 2018). These attribute tables were joined to the existing American West county database. It was determined to use the 2010 census income for sample stratification because it was the most current. However, to control for differing cost of living per state, a new field was added to normalize median county incomes by z-value per median state incomes, using the following formula {county median income/state median income) (Figure 4.4 and 4.5). Normalizing the median income values avoided clusters of high and low incomes contained within particular states as well as accounted for state variability in higher costs of living (Figure 4.6).
77


1
I | Stales defined ,is’ Animcni West
| Non-Aliin ituu WchI Slales
Spntinl Reference H(\S: I SA (’onttsjims Albers Hqii.il Area Conic
2010 Median House Hold lneiuiie (N'afiiral Breaks)
tJSl-SBO it! .382.002 <£3,21^.889
0 500 1.000
Kilometers
Figure 4. 4. County 2010 median income stratification without normalization (natural breaks).
78


I | States defined as American West
Noii-Aniciuaii West States 2010 Median Household Income (Quantile) I I <$4,149
””| <$19,183 ■$3,217,880
500
Kilometers
1,000
Spatial Reference PCS: LSA Contiguous Albers Equal Area Conic
Figure 4. 5. County 2010 median income stratification without normalization (quantiles).
79


Sampling Strata Design and Procedure
Categorization intervals were then chosen for the income attributes by which counties would be stratified. First, a sampling population was defined as the subset of counties where data existed, specifically, counties that contained WUI lands and CWPPs. This proved to be 326 counties, or about 79% of American West counties3 (Figure 4.7). For these 326 counties, the
3 While some counties have no CWPP at all, other counties, particularly counties in Utah and Arizona, have multi-county or local CWPPs. However, as these were not within the scale of concern, county, they were filtered as having no county CWPP (Figure 4.7).
80


intervals were determined that broke them into roughly three strata using quantiles. With the income of counties stratified, a simple random sample was used to select 40 counties from each stratum for a total project N of 1204 * &.
Figure 4. 7. Final county sample population (counties that contain WUI and a county level CWPP).
4 Power analysis for an ordinal multiple regression was conducted to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, and a large effect size (f=0.35), based on the
aforementioned assumptions, the desired sample size is 40 per strata (Faul, Erdfelder, Buchner,
& Lang, 2009).
81


Remedial Measures for Sampling
Once these 120 samples were chosen, it was necessary to see if some of these counties did not adequately fit the criteria needed for the study. For example, it was hoped to avoid counties that were overly clustered geographically or an over representation of a state. In the few cases where counties were overly clustered, one or more counties were randomly discarded and replaced with another county from the same strata. This was done until there were no significant spatial clusters. Finally, a list of counties was ready for which additional census variables, CWPPs, and local governance documents could be downloaded (Figure 4.8).
0 500
Kilometer*
1,000
Spatial Reiei'ciioe PCS: USA ContiguousAlbas Equal Area Come
Non-Sampled Counties Counties Without a CWPF
Figure 4. 8. Final county sample data set stratified per state-normalized z-scores.
82


Key Sample Descriptive Statistics and Demographics
The final sample consists of three CWPPs from Arizona, fifteen from California, eighteen from Colorado, nineteen from Idaho, twenty from Montana, five from Nevada, eight from New Mexico, thirteen from Oregon, eleven from Washington, and eight from Wyoming. Table 4.1 documents the number of CWPPs from each state in their corresponding income strata, following income z-score normalization.
Table 4. 1. Number of sampled CWPPs per strata per American West state.
State Number of CWPPs sampled Strata 1 Strata 2 Strata 3
Arizona 3 0 2 1
California 15 5 5 5
Colorado 18 4 7 7
Idaho 19 4 9 6
Montana 20 11 4 5
Nevada 5 0 3 2
New Mexico 8 2 4 2
Oregon 13 8 0 7
Washington 11 4 6 1
Wyoming 8 4 0 4
Total sample 120 40 40 40
Collectively, when the sampled counties were aggregated by state, the WUI progressively grew in each state between 2000 and 2010 by a cumulative total of 4,216 Km2 (Figure 4.9). However, when disaggregated 109 counties experienced positive WUI growth, two counties experienced no significant WUI growth, and 9 counties experienced negative WUI growth.
Strata three experienced the most growth (689 Km2), followed by strata two (235 Km2), and strata three (47 Km2). The minimum WUI growth for strata one was a decrease of 1.3 Km2 while the maximum WUI change was an increase of 11.2 Km2 with an average of 1.2 Km2 (Table 4.2). The minimum WUI growth for strata two was a decrease of 5.1 Km2 while the maximum WUI change was an increase of 46.6 Km2 with an average of 6.2 Km2 (Table 4.3). The minimum WUI
83


Full Text

PAGE 1

EVALUATING THE INTEGRATION OF CWPP AND COUNTY GOVERNANCE WILDFIRE RISK REDUCTION BEST PRACTICES ACROSS THE AMERICAN WEST: A PLAN QUALITY REVIEW by TRAVIS LEE FLOHR B.L.A. , The Pennsylvania State University, 2002 M.S.L.A ., The Pennsylvania State University, 2011 A dissertation 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 Program 2019

PAGE 2

ii 2019 TRAVIS LEE FLOHR ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Doctor of Philosophy degree by Travis Lee Flohr has been approved for the Design and Planning Program by Raf ael Moreno , Chair Austin Troy, Advisor Gregory Simon Bruce Goldstein Jody Beck Date: August 3, 2019

PAGE 4

iv Flohr, Travis Lee ( PhD, Design and Planning Program) EVALUATING THE INTEGRATION OF CWPP AND COUNTY GOVERNANCE WILDFIRE RISK REDUCTION BEST PRACTICES ACROSS THE AMERICAN WEST: A PLAN QUALITY REVIEW Thesis directed by Professor Austin Troy ABSTRACT In 2003, President George W. Bush enacted the Healthy Forest Restoration Act (P.L. 108148) (HFRA), expediting the preparation and implementation of hazardous fuels reduction efforts, if communities created community wildfire protection plans (CWPPs). CWPPs contextually define the wildland urban interface (WUI) and evaluate risk. Additionally, CWPPs identify and prioritize both public and private land mitigation projects. However, CWPPs are only one strategy in reducing wildfire risk. Research also suggests codifying wildfire risk reduct ion efforts into land use regulations, such as comprehensive planning, building and zoning codes, and subdivision guidelines. This study uses document analysis and cumulative odds ordinal logistic regressions to answer the following questions: 1) how well are CWPP wildfire planning best practices integrated, 2) how well are wildfire risk reduction best practices incorporated into land use regulations, and 3) what social, economic, demographic and geographic factors predict the level of best practice integration of CWPP inputs and outputs? While the regression results proved to statistically insignificant, the study found several interesting trends. First, the WUI is still growing geographically. Second, the WUI is also increasing in population. Third, there is a small positive correlation between the increase in WUI seasonal home growth and increased composite scores (CWPP wildfire planning score + l ocal

PAGE 5

v governance wildfire land use regulations ), suggesting it may be a leverage point for increased wildfire risk reduction planning. Fourth, there is a small negative correlation between longer homeowner tenure and the composite score, suggesting that wildfire risk reduction perceptions among longtime residents are complex. The results suggest that counties are not fully integrating CWPP or land use best practices. Both CWPPs and land use regulations can both improve by expanding their frames of reference from single development protectionism frames to include ecosystem health and incorporating more than one frame at a time. Additionally, counties must ensure frequent and timely update cycles, to better include emerging best practices and provide a sense of urgency to the process. Expanding frames requires a broader collective of participation. Noticeably absent from both the CWPP and comprehensive planning process were licensed land use professionals (e.g., architects, landscape architects, planners, surveyors, and civil engineers) who perform development work within each count y. Counties should engage with licensing bodies to ensure professionals are adequately prepared to address health, safety, and welfare best practices to mitigate wildfire. As such professional licensure for health, safety, and welfare could be an expanded wildfire frame. When incorporating expanded frames, goals and objectives should be edited to be specific, measurable, achievable, relevant and timed (SMART). Finally, this study suggests the integration of the CDC's social vulnerability index (SVI) into risk mapping efforts and an expanded anticipatory development risk evaluation process. The form and content of this abstract are approved. I recommend its publication. Approved: Austin Troy

PAGE 6

vi ACKNOWLEDGEMENTS This study was made possible by the diligent e fforts of my faculty advisor, Dr. Austin Troy. I am grateful for his years of guidance and support. Additional thanks go out to the other committee members of my dissertation committee for their assistance on this project: Drs. Rafael Moreno (defense chair ), Gregory Simon, Bruce Goldstein, and Jody Beck. Your help and support will forever be remembered and appreciated. I want to thank my wife, Stephanie, and daughter, Ruth. Thank you for your patience, smiles, laughs, and the balance you brought to my lif e. Without it, this project would be meaningless and incomplete. My final thanks go to my family, the Ph.D. cohort of Alessandro Rigolon and Mehdi Heris, and friends and colleagues, Mike Hinke, Suzy Anderson, Ed Russell, Heidi Ochis, and Jim Robb. Thank you for being there; your collective intelligence and patience inspires me.

PAGE 7

vii TABLE OF CONTENTS CHAPTER I. WHY STUDY THE INTEGRATION OF BEST PRACTICES IN COMMUNITY WILDFIRE PROTECTION PLANS (CWPP)? ...................................................................................................1 Research Aims, Questions, and Significance ........................................................................... 5 Dissertation Organization ....................................................................................................... 11 II. FEDERAL WILDFIRE RISK REDUCTION EFFORTS .......................................................12 Introduction ............................................................................................................................. 12 Federal Legislative and Administrative Risk Reduction Actions ........................................... 12 Fire Suppression Policies: 18712003 .............................................................................. 12 The Healthy Forest Initiative and the Health Forest Restoration Act: 2002 to Present .... 15 CWPP Legislative Incentives, Content, and Process .............................................................. 18 III. CWPP BEST PRACTICES AND LOCAL GOVERNMENT WILDFIRE RISK REDUCTION EFFORTS ..............................................................................................................22 Introduction ............................................................................................................................. 22 CWPP best practices ............................................................................................................... 22 CWPP Context .................................................................................................................. 23 Goal and objectives ........................................................................................................... 25 Community capacity ......................................................................................................... 28 Partnerships a nd collaboration .......................................................................................... 28 Base map ........................................................................................................................... 29 Risk assessment ................................................................................................................ 36 Hazardous fuels reduction ................................................................................................. 50

PAGE 8

viii Reducing structural ignitibility ......................................................................................... 60 Education and outreach ..................................................................................................... 61 Emergency management capacity ..................................................................................... 62 Longterm success ............................................................................................................ 63 Local government wildfire risk reduction efforts ................................................................... 63 Comprehensive Plan ......................................................................................................... 64 Zoning Codes and Development Standards ...................................................................... 66 Fire Resistant Materials .................................................................................................... 68 IV . RESEARCH DESIGN AND METHODOLOGY ..................................................................70 Research Permissions and Ethical Considerations ................................................................. 70 Pilot Study Results .................................................................................................................. 70 Study Area .............................................................................................................................. 71 Research Design ...................................................................................................................... 73 Independent and Dependent Variables ............................................................................. 73 Independent Variable Data ............................................................................................... 74 Study Methodology and Sample ....................................................................................... 75 Data Analysis .......................................................................................................................... 97 V . STATISTICAL RESULTS OF CWPP PLANNING AND IMPLEMENTATION ..............103 Integration and Implementation ............................................................................................ 103 CWPP Integration Scores ............................................................................................... 103 CWPP Imp lementation Scores ........................................................................................ 112 CWPP Composite Scores (integration + implementation) ............................................. 119 Conditions of Effective CWPPs and Risk Reduction Implementation ................................. 120

PAGE 9

ix CWPP Process and Plan Evaluation Inferential Statistics .................................................... 121 Spearman’s correlation coefficient results ...................................................................... 121 Ordinal logistic regressions results ................................................................................. 121 CWPP Implementation Local Governance Evaluation Inferential Statistics ....................... 122 Spearman’s correlation coefficient results ...................................................................... 122 Ordinal logistic regressions results ................................................................................. 122 Composite Evaluation Inferential Statistics .......................................................................... 124 Spearman’s correlation coefficient results ...................................................................... 124 Ordinal logistic regressions results ................................................................................. 124 VI . CONCLUSIONS ON CWPP AND LOCAL GOVERNANCE INTEGRATION IN REDUCING WILDFIRE RISK ...................................................................................................126 Regression Results ................................................................................................................ 126 CWPP Process Document Analysis Scores and Results ....................................................... 130 CWPP Document Analysis Scores and Results .............................................................. 130 Local Governance Document Analysis Scores and Results ........................................... 178 Final Thoughts: Theory, Practice, and Policy Implications for HFRA ................................ 194 Implications for Planning Theory ................................................................................... 194 Implications for Practice ................................................................................................. 204 Implic ations for HFRA: Flaws and Improvements ......................................................... 215 Limitations and Future Research .................................................................................... 218 Conclusion ...................................................................................................................... 220 REFERENCES ............................................................................................................................224

PAGE 10

x APPENDIX A . CWPP Pro cess a nd Plan Content Evaluation Instrument ......................................................249 B . Local Governance Instrument ................................................................................................253 C . Pilot Study Write Up ..............................................................................................................255 D . CWPP Bibliography ...............................................................................................................279 E . Local Governance Bibliography .............................................................................................285 F . Gis Data Bibliography ............................................................................................................293 G . Crook County, Wyoming’s Risk Field Assessment ...............................................................294 H . Mohave County, Arizona Fuel Treatment a nd Modification Plan.........................................296 I . Cal Fire’s Wildfire Evacuation Guide . ....................................................................................298 J . Cal Fire’s Evacuation Planning Guide ....................................................................................306 K . Model Community Widlfire Evacuation Guidelines .............................................................312 L . Calfire’s Home Ignition Zone Inspection Form. ....................................................................321 M . WFAP Home Assessment Checklist . ....................................................................................323

PAGE 11

xi LIST OF TABLES TABLE 3. 1. Fire Danger Rating System adjective class ratings ............................................................... 42 3. 2. NFPA Firewise Community Recommended Defensible Space Requirements. ................... 59 4. 1. Number of sampled CWPPs per strata per American West state. ........................................ 83 4. 2. Sample strata one WUI area change (Km2) from 20002010 descriptive statistics, summarized per state ..................................................................................................................... 84 4. 3. Sample strata two WUI area change (Km2) from 2000 2010 descriptive statistics, summarized per state. .................................................................................................................... 85 4. 4. Sample strata three WUI area change (Km2) from 2000 2010 descriptive statistics, summarized per state. .................................................................................................................... 85 4. 5. Sample strata one population change from 20002010 descriptive statistics, summarized per state. .............................................................................................................................................. 86 4. 6. Sample strata two population change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 87 4. 7. Sample strata three population change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 87 4. 8. Sample strata one hous ing unit change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 88 4. 9. Sample strata two housing unit change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 89 4. 10. Sample strata three housing unit change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 89

PAGE 12

xii 4. 11. Sample strata one seasonal housing unit change from 20002010 descriptive statistics, summarized per state. .................................................................................................................... 90 4. 12. Sample strata two seasonal housing unit change from 20002010 descriptive statistics, summarized per state. .................................................................................................................... 91 4. 13. Sample stra ta three seasonal housing unit change from 20002010 descriptive statistics, summarized per state. .................................................................................................................... 91 4. 14. Sample strata one median household income change from 20002010 descriptive statistics, summarized per state. .................................................................................................................... 92 4. 15. Sa mple strata two median household income change from 20002010 descriptive statistics, summarized per state. .................................................................................................... 93 4. 16. Sample strata three median household income change from 20002010 descriptive statistics, summarized per state. .................................................................................................... 93 4. 17. Sample strata o ne median age change from 20002010 descriptive statistics, summarized per State. ....................................................................................................................................... 94 4. 18. Sample strata two median age change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 95 4. 19. Sample strata three median age change from 20002010 descriptive statistics, summarized per state. ........................................................................................................................................ 95 4. 20. Sample strata one length of homeowner tenure descriptive statistics summarized per state. ....................................................................................................................................................... 96 4. 21. Sample strata two length of homeowner tenure descriptive statistics summarized per state. ....................................................................................................................................................... 97

PAGE 13

xiii 4. 22. Sample strata three length of homeowner tenure descripti ve statistics summarized per state. .............................................................................................................................................. 97 5. 1. Cohen’s Kappa levels of reliability interpretation. ............................................................. 104 5. 2. Average of categorical scores CWPP “Process and Plan Evaluation Instrument” document analysis score result s for all strata. ............................................................................................. 105 5. 3. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” document analysis scor e results for strata one. ........................................................................... 108 5. 4. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” document analysis score results for strata two. ........................................................................... 109 5. 5. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” docume nt analysis score results for strata three. ......................................................................... 110 5. 6. Score to letter grade conversion chart. ................................................................................ 112 5. 7. Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for all strata. .................. 113 5. 8. Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for strata 1. .................... 116 5.9 Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local Governance Evaluation Instrument ” docume nt analysis score results for strata 2. .................... 117 5.10 Minimum, maximum, and average categorical scores of the “CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for strata 3. .................... 118 5.11 Minimum, maximum, and average composite score results for all strata. ......................... 120 CWPP Implementation Local Governance Evaluation proportional odds test results. .............. 123

PAGE 14

xiv 6. 1. Siskiyou County, CA’s Fire Safe Council Projects in Siskiyou County ............................ 131 6. 2. Siskiyou County, CA’s list of critical stakeholders ............................................................ 142 6. 3. Excerpt (not a complete listing of participants) from Montrose County, CO CWPP development team’s roles and responsibilities. .......................................................................... 148 6. 4. Boulder County, CO community values at risk and weights of importance. ..................... 161 6. 5. Benewah County, ID’s legal and regulatory resources available for wildfire mitigation efforts .......................................................................................................................................... 176 6. 6. Base map requirements ....................................................................................................... 212 6. 7 HFRA CWPP sliding scale incentives. ................................................................................ 217

PAGE 15

xv LIST OF FIGURES FIGURE 1. 1. CWPP inputs and outputs. ...................................................................................................... 8 3. 1. Observed Fire Danger Class ................................................................................................. 43 3. 2. LANDFIRE fuels data acquisition p rocess for FARSITE. .................................................. 46 4. 1. Study states, defined as the American West and associated county boundaries. ................. 72 4. 2. Percent increase in large fire activity in the American West. ............................................... 73 4. 3. Simple, stratified random sampling framework. ................................................................... 76 4. 4. County 2010 median income stratification without normalization (natural breaks). ........... 78 4. 5. County 2010 median income stratification without normalization (quantiles). .................. 79 4. 6. County 2010 median income z score quantile stratification. ................................................ 80 4. 7. Final county sample population ( counties that contain WUI and a county level CWPP). ... 81 4. 8. Final county sample data set stratified per state normalized zscores. ................................. 82 4. 9. Change in WUI area (Km2) per sampled county from 20002010, summarized per state. 84 4. 10. WUI population change per sampled county from 2000 2010, summarized per state. ...... 86 4. 11. Amount of WUI housing unit change per sampled county from 20002010, summarized per state . ........................................................................................................................................ 88 4. 12. WUI seasonal housing unit change per sampled county 20002010, summarized per state. ....................................................................................................................................................... 90 4. 13. WUI median household income per sampled county from 20002010, summarized per state. .............................................................................................................................................. 92 4. 14. WUI median age per sampled county from 20002010, summarized per stat e. ................. 94

PAGE 16

xvi 4. 15. WUI average length of homeowner tenure per sampled county from 20002010, summarized per state. .................................................................................................................... 96 4. 16. Model one – CWPP process and content index .................................................................. 98 4. 17. Model two – CWPP implementation index. ...................................................................... 99 4. 18. Model three – CWPP process and content + implementation index. ................................. 99 5. 1. Minimum, maximum, and average of “Process and Plan Evaluation Instrument” document analysis total score results. .......................................................................................................... 106 5. 2. Average of “Process and Plan Evaluation Instrument” document analysis total score results per strata, summarized per state. ................................................................................................. 107 5. 3. Minimum, maximum, and of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis total score results. ..................................................................... 114 5. 4. Average of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis total score results per strata, summarized per state. ..................................... 115 5. 5. Final composite score results per strata, summarized per state. ......................................... 120 6. 1. Boulder County, Colorado CWPP goals ............................................................................. 136 6. 2. Boulder County, Colorado CWPP goals ............................................................................. 136 6. 3. Asotin County, Washington public meeting announcement. .............................................. 145 6. 4. El Paso County, Colorado steering team responsibilit ies. .................................................. 147 6. 5. Fremont County, WY proposed project area priorities and timelines, excerpt. ................. 150 6. 6. Benewah County, Idaho Forest Owners Field Day Announcement ................................... 153 6. 7. Gunnison County, Colorado letter soliciting participation as a community wildfire mitigation advocate (WMA). ...................................................................................................... 154

PAGE 17

xvii 6. 8. Harney County, Oregon CWPP base map. ......................................................................... 156 6. 9. Bingham County, Idaho CWPP base map. ......................................................................... 157 6. 10. Cochise County, Arizona CWPP base map. ..................................................................... 158 6. 11. Boulder County, CO wildfire risk assessment map results ............................................... 161 6. 12. Campbell County CWPP industrial development and land ownerships. .......................... 164 6. 13. Okanogan County, WA priority fuels reduction projects and map locations. .................. 166 6. 14. 6153 Laurel Dr. Paradise, CA, taken July, 2012. ............................................................. 190 6. 15. 565 Valley View Dr. Paradise, CA, taken July, 2012. ...................................................... 191 6. 16. 5915 Pine View Dr. Paradise, CA, taken July, 2012. ....................................................... 192 6. 17. Anticipatory modeling process. ........................................................................................ 214 6. 18. Embedded comprehensive planning and local regulatory CWPP process. ...................... 222

PAGE 18

1 CHAPTER I WHY STUDY THE INTEGR ATION OF BEST PRACTI CES IN COMMUNITY WILDFIRE PROTECTION PLANS (CW PP)? The Healthy Forest Restoration Act (P.L. 108148) (HFRA), first national legislation that expedited the preparation and implementation of hazardous fuels reduction efforts, provides a useful instrument for analyzing whether and und er what conditions local governments’ are implementation wildfire best practices . This study evaluates Community Wildfire Protection Plans (CWPP) and local governments policy implementation supporting the reduction of wildfire risk (e.g., zoning, building codes). The study also determines whether and how adoption of best practices vary by median age, length of homeowner tenure, full time/part time residency, and income. While the role of CWPP and enforceable code best practices implementation is important in exploring risk and theory, it is of considerable importance to policy makers as well. First and foremost, the evaluation indices help quantify the degree to which each local government has implemented scientific best practices for wildfire risk reductio n, elucidating gaps in CWPP plans and local codes and guidelines. Secondly, understanding what social, economic, demographic and geographic factors predict the integration of best practices in CWPP s is paramount to how decisionmakers engage community members. Wildfires have been increasing in severity and cost over the last several decades due to increasing fuel loading and ex urban development in the wildland urban interface (WUI). WUIs are where wildland uses meet urban land uses. Radeloff , Hammer, Stew ert, Fried, Holcomb, and McKeefry (2005) estimated that the United States (U.S.) WUI covers 719,156 km2 and contains 44.8 million housing units. The expansion of homes and associated commercial development in

PAGE 19

2 the WUI places property, assets, and human lives at risk from wildfires (Bhandary & Muller, 2009; Reams, Haines, Renner, Wascom, & Kingre, 2005) . At the same time federal costs to suppress wildfires are drastically increasing; in 1985 costs were $239,943,000, and costs in 2012 were $1,902,446,000 (National Interagency Fire Center, 2018) . The total Federal cost for wildfire suppression between 1985 and 2012 was $25,370,157,000 (National Interagency Fire Center, 2018) . While the land use change drivers of WUI development may differ in other parts of the world, the expansion of the WUI into high risk w ildfire zones is not unique to the U.S. (Brummel, Nelson, & Jakes, 2012; Carmo, Moreira, Casimiro, & Vaz, 2011; Dondo Bhler, Curth, & Garibaldi, 2013; Harris, McGee, & McFarlane, 2011; Holland, March, Yu, & Jenkins, 2013) . Initial responses to wildfire in the United States WUI almost exclusively emphasized fire suppression (Steelman & Burke, 2007) . The fire suppression policies implemented between 1905 through 1911 failed to address several key issues rel ated to wildfire in the WUI: 1) the necessity of fire for ecosystem health; 2) the ability of fire to moderate fuel load buildup; and 3) the altering the public’s perception of wildland aesthetics and processes (Busenberg, 2004; Veblen, Kitzberger, & Donnegan, 2000) . Despite the failings of fire suppression policies, these policies governed wildfire practices for almost 100 years. WUI development conflicts were initi ally raised by Vaux (1982) , who classified the interface of urban development and forestry as a signifi cant area of concern and cautioned foresters to not underestimate its political and policy significance. Additionally, Bradley ( 1984) also spoke of the significance of the urban/forest interface, defined as two traditional land uses occurring in proximity to each other (i.e. forestry and urban development). Early views of the urban/forest interface defined the problems as resour ce conflicts that begin as spatial conflicts

PAGE 20

3 and quickly become sociopolitical, often pitting a set of values against one another (Bradley, 1984; Vaux, 1982) . These early conflicts often revolved around forest managers’ concern over forest resources as a commodity, such as mineral extraction and timber harvests, and urban edge residents’ concern in forests as land commodity, such as aesthetic amenities, recrea tion, and wildlife (Bradley, 1984) . However, land managers tackling these concerns failed to address the growing wildfire issues in the WUI. Recent ly, U.S. wildfire management policy has shifted from fire suppression to an integrated program of fire suppression, preparedness, mitigation, and community assistance (Gonzlez Cabn, Haynes, McCaffrey, Mercer, & Watson, 2007) . This shift is recognizable in the wildfire research literature . R esearchers, designers, and planners have been working to understand the increasing conflicts between WUI landowners and natural wildfire regimes by using a wider multitude of research designs, methods, and theories, in addition to continuing wildfire modeling research (Bhandary & Muller, 2009; Brown, Agee, & Franklin, 2004; R. Burby & Deyle, 2000; Heyerdahl, Brubaker, & Agee, 2001; Muller & Schulte, 2011; Paveglio, Jakes, Carroll, & Williams, 2009; Reams et al., 2005) . For example, research and practitioners have developed best management practices for community wildfire protection planning processes (Jakes et al., 2012; Society of American Foresters, 2004) , identified Firewise development and land management practices (Headwaters Economics, 2014; M. A. Moritz et al., 2014; Paterson, 2007; Wi nter, McCaffrey, & Vogt, 2009) , identified socio economic barriers to implementing Firewise best management practices (Chuvieco, Martnez, Romn, Hantson, & Pettinari, 2014; T. W. Collins, 2008a; Gardner, Cortner, & Widaman, 1987; Kousky, Olmstead, & Sedjo, 2011; Poudyal, JohnsonGaither, Goodrick, Bowker, & Gan, 2012) , and created better mo dels of wildfire risk and behavior (Ager, Vaillant, & Finney, 2011; Kramer, Coll ins, Kelly, &

PAGE 21

4 Stephens, 2014; LANDFIRE, 2010) . Practitioners have attempted to implement these best practices to m ediate increasing fire risk with policies limiting development in high risk areas, requiring fire rated building materials and sprinkler systems, implementing community wildfire protection plans, creating defensible space around buildings, and reducing fue l loads (Bhandary & Muller, 2009; Headwaters Economics, 2014) w ith varying levels of success. Current wildfire planning efforts, particularly those related to community wildfire protection plans, are the result of The Healthy Forests Restoration Act of 2003 (HFRA). HFRA was the culmination of a decade of changing wil dfire mitigation research and wildland fire policy reforms that were in response to the growth of WUI development, danger from catastrophic WUI wildfires, and decline in WUI ecosystem health (Steelman, 2008) . HFRA called for communities to implement Community Wildfire Protection Plans (CWPP) (Grayzeck Souter, Nelson, Brummel, Jakes, & Williams, 2009) . CWPPs are often implemented at the county or community (i.e. municipality, borough, town, city or local fire district) scale, hence the term ‘County’ or ‘Community Wildfire Protection Plan.’ Increasingly, CWPPs are also implemented at a neighborhood or subdivision scale. CW P P s encoura ges collaboration between local fire department s , the state agency responsible for forest management and relevant local government, in consultation with adjacent federal land management agencies and surrounding community residents (Grayzeck Souter et al., 2009) . CWPPs have several key benefits and objectives for achieving a more effective wildfire mitigation strategy. The development of CWPPs should include pr iority areas for fuel reduction and provide ignitability assessment throughout the community ("The Healthy Forest Restoration Act of 2003," 2003) . Communities benefit from having CWPPs because it allows for a flexible and contextually defined, localized WUI boundary; localized fuel treatment prioritization;

PAGE 22

5 prioritization of funding; and integration into local land use policies (Jakes et al., 2011; Steelman & Burke, 2007) . Robustly implemented CWPPs allow land managers to reestablish natural fire regimes while minimizing the risk to people, thus rehabilitating and resto ring fire adapted ecosystems and minimizing ongoing wildfire risks over the longterm (Steelman & Burke, 2007) . Researchers have only begun to tackle the integration of CWPP best practices. While res earch is expanding into effective integration of CWPP, there are still numerous gaps in the literature. Often CWPP and local wildfire research is not generalizable because studies have used small n or single case study examples, which make it difficult to understand specific contexts that facilitate effective CWPP processes and implementation. As is discussed in the literature review, the vast majority of studies on CWPPs tend to f ocus on the social outcomes of CWPP processes, including capacity building and the interrelationships of participants, while giving little attention to the content of CWPP products, policies or the physical implementation of CWPP goals and objectives. This oversight creates a disconnect in understanding how processes lead to deliverables as well as how deliverables lead to implementation and the reduction of risk. Finally, CWPP literature has understudied the links between CWPP processes, documents, and the integration of wildfire safe policies into enforceable local zoning and land use codes. As a result, the question of how well CWPP processes and CWPP implementation are at integrating best practices and achieving their goals remains. Research Aims, Questions, and Significance The purpose of this research is to evaluate whether CWPPs and county governance wildfire risk reduction best practices meet minimum process and outcome criteria across the American West , which requires an understanding of the process that created the CWPP as well as CWPP implementation. Counties were chosen as a unit of analysis because, as discussed in

PAGE 23

6 the literature review: 1) counties are the logical scale to mitigate landscape scale wildfire risk and 2) county governments are implementing CWPPs for broad expanses of the WUI and have not been widely stud ied. Local CWPPs were excluded from this study for two reasons: 1) feasibility and limitations on scope and time and 2) the incongruent boundaries between CWPPs and sociodemographic and economic data introduced too much error and uncertainty in such an ex tensive study (see Appendix C for a more detailed discussion on the pilot study). For the purposes of this project, minimum process and outcome criteria includes 1) the process of developing the CWPP is inclusive of the diverse stakeholders in the area, involving participatory actions of those stakeholders; 2) the CWPP clearly articulates and documents the extent of the wildfire hazard; 3) the CWPP plans adhere to principles of defensible space and other evidence based wildfire mitigation best practices, st ating implementable and measurable goals; and 4) the implementation of the CWPP goals result in observable reduction in WUI development and implementation of building codes and policies that reflect CWPP goals and objectives . In order to evaluate CWPP plan ning in terms of integration , indices were created to examine the CWPP process and the resulting documents as well as the CWPP implementation. These indices served as the dependent variables for this research while community contextual variables, such as socio economic and demographic variables were independent variables because they are known to impact planning programs and their implementation. Specifically, my research objectives/ questions are: Create an index that defines the level of best practice i ntegration of CWPP process (inputs); Create an index that defines the level of best practice integration of CWPP implementation (outputs); and

PAGE 24

7 What social, economic, demographic and geographic factors predict the level of best practice integration of CWPP inputs and outputs? Inputs and outputs are further clarified in Figure 1.1. Note. * Denotes outputs that are not measurable across large geographic scales at this time, thus they are excluded from this project.

PAGE 25

8 Figure 1. 1. CWPP inputs and outputs. Addressing these research objectives/questions is important to researchers for four main reasons. First, little research has been conducted to evaluate the integration of CWPP processes best practices and to evaluate the integration of CWPPs on implementing wildfire mitigation stra tegies. Second, this study will methodologically broaden the scope of wildfire research through a larger n sample, which allows for greater generalizability, enables the pinpointing of outliers more easily , and creates the potential for a better margin of error. These are not possible with the current small n or case study wildfire research. Third, this study introduces the spatial analysis of the implementation of CWPP mitigation objectives, such as limiting WUI expansion into known wildfire hazard areas , which is paramount in assessing CWPP effectiveness. Fourth, this study will empirically evaluate the socio economic, demographic, and biophysical conditions that are associated with inhibit CWPP process es and implementation best practices. For this stud y, the CWPP process includes both the actual CWPP plan, as a product of the process, and meeting minutes, as artifacts of the engagement in the process. I mplementation is a composite of both policy and biophysical interventions. CWPP policy implementation are the integration of wildfire policies into building and zoning codes; comprehensive plans; subdivision ordinances, and HOA guidelines. The biophysical components of CWPP implementation include no net change in WUI expansion, densification, and health WUI forests and development . Details on each index and measurement protocols are discussed at length in Chapter II and located in A ppendix A and B . The wildfire and hazard planning literature has identified a number of socioeconomic and demographic variables that can influence CWPP processes and implementation, including age, length of homeowner tenure, full time/part time residency, and income. Survey research has

PAGE 26

9 shown that these four variables can play a statistically significant role in effective public participation processes, and CWPP implementation (T. W. Collins, 2008b; Crow et al., 2015; J. W. Smith, Leahy, Anderson, & Davenport, 2013; Wolters, Steel, Weston, & Brunson, 2017) . Based on this previous research, the hypothesis for this study is: CWPP ge ographies with older populations, short homeowner tenure, high part time residency status, and lower income will have lower integration in the dependent variables. Other variables considered were: educational attainment, owner occupied housing, retirement status, slope, percent WUI, percent wildland cover, aspect, proximity to recent fire, hazard exposure, and single egress points; however, these have proved statistically insignificant or less significant in previous studies (T. W. Collins, 2008b; Crow et al., 2015; J. W. Smith et al., 2013; Wolters et al., 2017) . The purpose of indices, in general, is to allow comparisons across time and space (Ebert & Welsch, 2004) . The propos ed process index and implementation index are critical to wildfire mitigation in the CWPP process because they aid in decision making surrounding the prioritization of planning processes and implementation strategies. This is significant in the daily pract ices of wildfire mitigation and the CWPP process for both the decision makers and the community members. In regards to decisionmakers, an integration process index is important for the following reasons: 1) to compare a community’s process with other communities in order to validate localized best practices of community engagement, 2) to prioritize the utilization of staff time and resources, and 3) to facilitate community participation without overburdening community members. For community members, there are two main benefit of a process index : 1) the promotion of more meaningful engagement in the planning process; by focusing community participation in areas where community members can actively contribute, the community members will not be fatigued and frustrated by diluting their participation across too

PAGE 27

10 many initiatives and 2) legitimizes the CWPP planning process by providing sustained appraisals on planning products, procedures and results . An integration implementation index also benefits decision makers and community members during the CWPP process. To decision makers, an integration implementation index facilitates 1) the comparison of communi ty implementation in order to validate localized best practices of defensible space, zone sizing, forest thinning practices, and so forth; 2) the prioritization of implementation strategies in regards to economic restraints and location throughout the comm unity; 3) the utilization of incentives for implementation, particularly on private property; 4) the application of implementation policy mechanisms, including fines for unmitigated WUI development and requirements of disclosure of guidelines during proper ty transfer ; and 5) as a performance indicator . For community members, an implementation index has two benefits: 1) knowing and understanding the levels of wildfire safety in their community as well as the progress being made to reach the community safety goals, and 2) receiving targeted guidance in how to best reduce risk on their private property. Understanding what social, economic, demographic and geographic factors predict CWPP best practice integration is paramount to how decision makers engage community members. These underlying structural contexts of the community members’ ability to participate in wildfire mitigation efforts regardless of how much wildfire mitigation education has been presented. For example, members of a community in abject pover ty may understand the importance of wildfire mitigation but may also need to choose between landscape alteration and food. Furthermore, these same community members may not be able to attend community participation events because they work multiple jobs t o provide for themselves and/or their family. Therefore, the use of these predictive factors by decision makers is 1) to gauge whether

PAGE 28

11 the lack of environmental outcomes (i.e. wildfire safe communities) is due to these community factors or due to the CWPP process and/or implementation, 2) to identify and allocate necessary resources and time to address other community barriers and issues, such as poverty or language barriers, in order to produce more effective processes and implementation, and 3) to underst and the limit of a community’s capacity to participate in various implementation processes and strategies. Dissertation Organization This document is organized into six chapters. The first chapter discussed the importance of studying the integration of th e CWPP process and content best practices. Additionally, I discussed the need to study the integration of wildfire risk reduction best practices into local land use regulations, also known as county or local governance. Chapter Two discusses federal wildfire risk reduction efforts and how they have culminated in the Health y Forest Initiative, which led to the Healthy Forest Restoration Act and CWPPs. Chapter Three outlines CWPP and local governance best practices. Chapter Four outlines my research design an d methodology. Chapter Five presents my statistical results. In Chapter Six , I discuss my conclusions and their implications for planning practice and theory.

PAGE 29

12 CHAPTER II FEDERAL WILDFIRE RIS K REDUCTION EFFORTS Introduction The Healthy Forest Restoration Act ( HFRA) was enacted due to the inadequacies of preexisting wildfire risk reduction efforts in the wildland urban interface. In order to understand the conditions which led to the creation of HFRA, I first discuss a brief history of federal legislative and administrative risk reduction actions prior to HFRA . Second, I discuss HFRA and other present wildfire policies. Finally, I discuss the key components of HFRA and CWPPs. Federal Legislative and Administrative Risk Reduction Actions Fire Suppression Policies: 1871 2003 The current understanding of wildfire management needing to address a complex mix of natural environment, economic, and cultural factors follows over 100 years of fire suppression as the wildfire policy. It is important to unde rstand the background on the origination and evolution of suppression policies to address the effects of these policies on current development and attitudes and the resulting challenges in modern wildfire mitigation in the WUI. Complex regional and nationa l economics, westward expansion, agricultural resources, energy and mineral resource extraction, urban growth, and recreation policies and attitudes have all contributed to the United States’ development of the WUI (Headley, 1916; Riebsame, Gosnell, & Theobald, 1996; Theobald & Romme, 2007; Travis, 2007) . Initial WUI development of the 1800s was driven by resource economies, such as timber, agricultural activities, and mining (Riebsame et al., 1996; Theobald & Romme, 2007; Travis, 2007) , and wildfire was viewed as a threat to commercial timber activities and watershed protection (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . Legendary fires, such as Peshtigo Fire of 1871, bolstered the

PAGE 30

13 conservationist argument that wildfires are a threat to forest commodities, leading to the creation of national forests as protected timber reserves (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . The U.S. Forest Service (USFS) was established in 1905 and had managerial control of national forests (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . During its early years, catastrophic forest fires burnt over three million acres in Montana, Idaho, and Washington (Fire History Society; Head ley, 1916; Muir, 1941; Silcox, 1910) . Administrators were convinced that catastrophic fire events could be prevented (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . The USFS convinced lawm akers and the American public that fire suppression was the only way to prevent such events from ruining timber economies and communities (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . These early USFS initiatives culminated in the passage of the Weeks Act of 1911, which established a framework between the federal and state governments for cooperativ e firefighting, which would later include private forest associations and landowners (Fire History Society; Headley, 1916; Muir, 1941; Silcox, 1910) . The Weeks Act offered financial incentives to states for suppressing fires under the direction of the USFS, a framework that is still in place toda y (Southard, 2011) . The grant support efforts of the Weeks Act were expanded in 1924 through the Clark McNary Act, to further support state efforts in fire protection (Southard, 2011) . F ollowing several severe fire seasons in the early 1930s, fire suppression took on a greater urgency. The Civilian Conservation Corps supplied thousands of men to work building fire breaks and fighting fires (Fire Histo ry Society) . In 1935, the USFS established the 10 a.m. policy, which stated that every fire should be suppressed by 10 a.m. the day following its initial report (Fire History Society) .

PAGE 31

14 The fire suppression policies implemented between 1905 and 1935 continued through 1970 and exacerbated the fire problem; fires have continued to grow in size and intensity. Suppression polic ies failed to acknowledge that fire is necessary to moderate fuel load buildup and altered the public’s perception of wildland aesthetics and processes, leading to increased WUI development and preferred unnatural aesthetics (Busenberg, 2004; M. A. Moritz et al., 2014; Nielsenpincus, Ribe, & Johnson, 2015; Steelman & Burke, 2007; Scott L. Stephens & Collins, 2007) . During the 1960s, research began to show the positive role fire played in forest ecology: restoring vegetation by releasing seeds, controlling diseases and invasive species, and managing wildfire fuel loads. Despite this emerging knowledge, the federal government further expanded its influence in fire suppression policies and management procedures through the National Wildfire Coordinating Group in 1976 and expanded their grant funds to include administrative resources and surplus federal equipment to rural fire departments through the Cooperative Forestry Assistance Act in 1978 ("Cooperative Forestry Assistance Act of 1978," 1978) . However, with the support of emerging fire science, the USFS instituted a let burn policy (Fire History Society), but it fell out of favor as early as the 1980s due to dangerous fire conditions in the WUI where development had increased – driven by resort development, urban expansion, low density homeowner development (Fire History Society; Manning, 2012) . In 1986, the National Wildland/Urban Interface Fire Protection Initiative brought together representatives from federal land management, fire protection agencies, and the National Associate of State Foresters to address the emerging problem of fire in the WUI (J. Cohen, 2008) . As a result, the home destruction problem became nationally recognized and the U.S. Forest Service and National Fire Protection Association held confe rence – the 1986 Wildfire Strikes Home conference – which spawned the current Firewise program (J. Cohen,

PAGE 32

15 2008) . These initiatives marked the change in the wildfire narrative from resource protection to WUI management, highlighted by the fact that the WUI was identified in as a principle concern in the 2000 National Fire Plan, 2001 Federal Wildland Fire Management Policy, and Healthy Forest Restoration Act. Despite the evolu tion and changing attitudes of wildfire and the WUI within federal policy, these approaches were flawed because they failed to address three pivotal issues: 1) subsidizing the financial risk of living in the WUI through the allocation of fire suppression f unds, 2) engaging private land owners inadequately, if at all, and 3) failing to acknowledge the influence of municipal policies, such as zoning role in wildfire risk reduction. The Healthy Forest Initiative and the Health Forest Restoration Act: 2002 to Present Healthy Forest Initiative From 2001 to 2003 the U.S. experienced 147,049 fires that burned approximately 11 million acres (The White House, 2003) . The fires in 2002 claimed the lives of 28 firefighters (The White House, 2003) . The 2003 California fires alone accounted for approximately $250 million in fire suppression costs and claimed 22 civilian lives (The White House, 2003) . These staggering figures reinforced the recognition that the catastrophic fires of the American West were burning hotter and faster than most ordinary wildland fires. In August 2002, President George W. Bush established the Health Forests Initiative (HFI), which directed the Departments of Agriculture and Interior, and the Council of Environmental Q uality to improve regulatory processes to ensure more timely decisions, greater efficiency, and better results in reducing the risk of catastrophic wildland fires (One Hundred Eighth Congress of the United States of America, 2003) . HFI is momentous because for the first time the initiative strove to engage the public and all levels of government while caring for forests and rangelands to reduce wildfire risk, save lives, and protect endangered and threatened species.

PAGE 33

16 The Healthy Forest Restoration Act The Healthy Forest Rest oration Act ( HFRA ) , passed in 2003, is the result of wildland fire policy reforms that emerged from the HFI, which was designed to improve the capacities of landmanagement agencies to protect communities, watersheds, and other at risk lands from catastrophic wildland fires (Jakes et al., 2011) . The legislation provisions are to expedite the pr eparation and implementation of the reduction of hazardous fuels on federal lands and assisting rural communities, states, and landowners in restoring healthy forest and watershed conditions. HFRA— without providing uniform guidelines or standards —encourage s communities to collaboratively develop Community Wildfire Protection Plans (Jakes et al., 2011) . The components and process of the Community Wildfire Protection Plan (CWPP) is structured loosely by HFRA. Within that structure is built the flexibility of local communities to, ideally, create a CWPP that reflects community values and areaspecific environmental conditions in order to effectively mitigate wildfire in the WUI. CWPP guidelines are designed to be flexible in order t o address local contexts, though each plan should have the following components: description of the WUI and associated resources at risk, documentation of community preparedness, community risk analysis that prioritizes fuel treatment priorities and methods of treatment, ways to reduce structure ignitability, an implementation plan, and collaboration of stakeholders (Society of American Foresters, 2004) . Hazardous fuel reduction area treatments must be id entified and prioritized, and the methods of fuel reduction treatments must also be recommended (Jakes et al., 2011; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; Steelman & Burke, 2007) . CWPPs are should identify at least two zones of defensible space surrounding building structures.

PAGE 34

17 CWPPs should be collaboratively developed by local, state, and federal stakeholders, and the local government (e.g. counties or cities); local fire department(s); and state entity responsible for forest management must agree on the final contents (Society of American Foresters, 2004) . HFRA also recommends at least three entities must agree to the final CWPP: applicable local governments (i.e. counties or cities); local fire department(s); and state entity responsible for forest management (Society of American Foresters, 2004) . While not required, HFRA does outline and recommend an eight step process for CWPP development. The outline is as follows: convene decisionmakers; involve federal agencies; engage interested parties; establish a community base map; develop a community risk assessment; establish community priorities and recommendations; develop an action plan and assessment strategy; and finalize community wildfire protection plan. HFRA provides the authorizat ion to expedite environmental assessment, administrative appeals, and legal review for hazardous fuels projects on federal land (Society of American Foresters, 2004) . HFRA also emphasizes the need for federal agencies to collaborate with communities in developing hazardous fuel reduction projects by placing priority on treatment areas identified by CWPPs (Society of American Foresters, 2004) . It also distributes financial resources across federal and non federal projects according to CWPP objectives. The HFRA also disincentives the absence of a CWPP through the financial policy of requiring at least 50 percent of all funds to be used within the WU I. Without a CWPP, the WUI is defined within mile of a community’s boundary or within 1 miles when mitigating circumstances exists, e.g. steep slopes , which are less than most locally developed CWPP definitions. HFRA is a step in the right direction, but it does present several issues. HFRA’s biggest flaw is that it does not address continued development in the WUI. The spirit of individualism

PAGE 35

18 and the resentment of government intervention in private property rights hinder such legislative actions surr ounding wildfire risk and mitigation (T. W. Collins, 2008b) , which explains why only half of state legislatures address land use planning and only 11 of these encourage loca l governments to plan for hazard mitigation (R. Burby & Deyle, 2000) . The result of continued WUI growth is that wildfire funds are continually allocated suppress fires in order to protect homes rather than towards risk reduction efforts. Exacerbati ng the issue is that forest mitigation work – such as Healthy Forest initiatives – has been consistently underfunded (Steelman & Burke, 2007; Trego, 2014) . The goal of these efforts is to maintain healthy forests that reduce risk for WUI residents. Healthy forests (HF) within the context of HFI and HFRA consists of low forest density, low fuel loads, and more variability in age of vegetation and forest structure. A healthy forest structure, in effect, is of pre fire suppression activities. Healthy WUI forests and WUI development i nclude the observable implementation of fire resistant building materials, defensible space, and reduction in overall fuel loads within the WUI. While HF are theoretically important to evaluating the integration of CWPP implementation, they are currently not able to be evaluated across large, disparate geographies. HF measures are currently not feasible for this study because datasets do not exist or would require expensive highresolution remote sensing multispectral imagery. Additionally, such datasets r equire time intensive onthe ground calibration assessments. As such, the current study will not contain measures of HF and future research should expand in these areas. CWPP Legislative Incentives, Content, and Process As previously mentioned, CWPPs are locally generated wildfire protection plans, a HFRA policy incentive requirement . It is worth reiterating the core components, incentives and

PAGE 36

19 disincentives. Encouragement of CWPPs primarily comes in the form of facilitating agencies and collaboration. HFRA provides the authorization to expedite environmental assessment, administrative appeals, and legal review for hazardous fuels projects on federal land (Society of American Foresters, 2004) . HFRA also pl aces priority on treatment areas identified by community CWPPs, emphasizing the need for federal agencies to collaborate with communities in developing hazardous fuel reduction projects (Society of American Foresters, 2004) . Furthermore, financial resources can be distributed across federal and non federal projects according to CWPP objectives. The components and process of the CWPP is structured loosely by HFRA. HFRA suggests a minimum of the fo llowing: prioritized fuel reduction, treatment of structural ignitability, and collaboration, (Society of American Foresters, 2004) . The CWPP process consists of eight steps: 1) convene decision makers; 2) involve federal agencies; 3) engage interested parties; 4) establish a community base map; 5) develop a community risk assessment; 6) establish community priorities and recommendations; 7) develop an action plan and assessment strategy; and 8) finalize community wildfire protection plan (Jakes et al., 2007; Jakes et al., 2012; Society of American Foresters, 2004) . Ideally, within that structure is built the fle xibility for local communities to create a CWPP that reflects community values and area specific environmental conditions in order to effectively mitigate wildfire in the WUI. For example, HFRA allows for communities to define their own WUI interface areas, valuing local knowledge of area contexts. In fact, if communities do not define and designate WUI areas, then the WUI interface area defaults to National Register’s WUI classification and the community is not prioritized in federal and state funding desi gnations. Additionally, communities should prioritize project and treatment areas and provide baseline information for monitoring of long-

PAGE 37

20 term effects on the CWPP in wildfire risk reduction. While CWPP guidelines are designed to be flexible to address local contexts, each plan should have the suggested components, and HFRA recommends following the eight step process for CWPP development as stated above. The initial Federal Registrar definition of an urban wildland interface community were derived from “A Re port to the Council of Western State Foresters —Fire in the West —The Wildland/Urban Interface Fire Problem” (Teie & Weatherford, 2000) . Per this definition, wildland interface communities exist where humans and their associated development meet or intermix with wildland fuel. The Federal Registrar (C. N. Thompson, 2001) provides a further categorization of the WUI using three subclassifications: interface, intermix, and occluded communities. An interface community is where three or more structures per square mile exist, but there is also a clear line of demarcation between residential, business, and public structures and wildland fuels. Alternatively, interface communities emphasize a population density of 250 or more people per square mile. An intermix community is where close together structures to 40 structures per acre intermix with wildland fuels and are continuous outside and within the developed area. Intermix communities have population densities of 2825 0 people per square mile. Occluded communities exist when wildland fuels exist in parks and open spaces (<1,000 acres) located within a city, with a clear demarcation between fuels and structures. Federal agencies focus on interface and intermix communities only. These three sub classifications all have significant limitations. First, the definitions do not define wildland vegetation communities or patterns. Because the Federal Register does not set a minimum threshold of wildland vegetation type or densi ty, small areas of urban parks may be erroneously included. Second, the Federal Register does not standardize the WUI interface thresholds for the maximum housing or population density measures. Third, the distance to

PAGE 38

21 development from the interface line is not systematic. The Healthy Forest Restoration Act (HFRA) has subsequently implemented definitions of WUI that supplement the Federal Register categorization of the WUI. HFRA seeks to empower local communities to develop a CWPP that is accepted by the com munity, fits communityspecific goals and objectives, and addresses the practical needs of wildfire mitigation for local WUI context. To further this goal of CWPP development, HFRA defines the WUI and thresholds, which includes the establishment of a buffe r zone around the town, civic infrastructure, and evacuation routes (Stewart, Radeloff, Hammer, & Hawbaker, 2007) . I will elab orate on how CWPP best practices have attempted to contextualize the operational measurement of the WUI and wildfire risk in Chapter III. The CWPP should identify and prioritize hazardous fuel reduction area treatments, and the methods of fuel reduction treatments must also be recommended (Jakes et al., 2012; Jakes et al., 2011; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 20 04; Steelman & Burke, 2007) . Addi tionally, these plans are should identify at least two zones of defensible space surrounding building structures, as previously defined. CWPPs are meant to be collaboratively developed by local, state, and federal stakeholders; and the local government (e.g. counties or cities), local fire department(s), and state entity responsible for forest management must agree on the final contents (Society of American Foresters, 2004) . HFRA encourages at least three entities agree to the final CWPP: applicable local governments (i.e. counties or cities); local fire department(s); and state entity responsible for forest management (Society of American Foresters, 2004) .

PAGE 39

22 CHAPTER III CWPP BEST PRACTICES AND LOCAL GOVERNMENT WILDFIRE RISK REDUCTION EFFORTS Introduction A thorough discussion of CWPP best practices and local government wildfire risk reduction efforts is paramount to assessing the CWPP process as well as how CWPPs penetrate local government policies. First, best practices will be discussed, which were used to structure the CWPP process instrument. Second, local government wildfire risk reduction efforts and reduction options will be discussed. CWPP bes t practices Research and practi ti oner experience has identified best practices for the CWPP process and content, and the implementation of these best practices is often a coordinated effort between local, state, and federal agencies, just as HFRA intended these three to be the key signatories of CWPPs (Busenberg, 2004; Jakes et al., 2011; One Hundred Eighth Congress of the United States of America, 2003; Scott L. Stephens & Collins, 2007) . These best practices are categorized and presented according to the following order : CWPP context, goals and objectives, community capacity, partnerships and collaboration, base mapping, risk assessment, hazardous fuel reduction, reducing structural ignitability, education and outreach, emergency management capacity and longterm success. These 11 categories have been deemed necessary to create an effective CWPP (Jakes et al., 2007; Jakes et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; Williams et al ., 2012) , and are the groupings utilized in the CWPP process analysis instrument, detailed in Chapter IV and in Appendix A .

PAGE 40

23 In this first section of this chapter, a systematic discussion will step through each of these eleven groups recommended for CWPP developm ent. In subsection one, the need for an adequate CWPP context will be discussed as well as what should be included in that context. In subsection two, goals and objectives will be defined and suggestions to creating effective goals and objectives will be provided. In subsection three, the need to document, and augment, community capacity through the CWPP will be presented. In subsection four, partnerships and collaboration will be exemplified. In sub section five, the need for a base map will be di scussed and what should be included on the base map will be identified, such as human presence and development, wildland vegetation, and wildfire risk. In subsection six, risk assessment will be explored by reviewing LANDFIRE, wildfire exposure modelling, and behavioral modelling, and finishing with CWPP wildfire risk modeling to highlight some best practices. In sub section seven, hazardous fuels reduction will be presented by discussing fuel management basics and types of fuel treatments, rounding out with details on defensible space. Finally, in sections eight through eleven, reducing structural ignitability, education and outreach, emergency management capacity, and longterm success will be discussed, respectively. CWPP Context Adequate context for the community is paramount in developing a robust CWPP, and this context should be created through the efforts of all key stakeholders, including the local state, and federal governments. Jakes et al (2012) outlines five components to community context, that are critical to CWPP success: 1) remind community members of how they handled past challenges, such as a wildfire or environmental disaster in order to help the community understand how it is vulnerable and create a sense of urgency for developing a CWPP; 2) study previous collaborative efforts in the community, whethe r wildfire planning or other projects, to

PAGE 41

24 identify how they were successful and use lessons from those experiences to lay the groundwork for doing a CWPP; 3) identify people who were involved in earlier collaborative or wildfire planning efforts and bring their experience to developing a CWPP; 4) find ways to overcome the challenge of inexperienced communities in collaboration or wildfire planning; and 5) address disagreements within a community early, related to wildlife or not, that could threaten the CWP P process (Jakes et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004) . Each of these elements, while not part of the core HFRA legislation , are necessary because community histor y provides context for the CWPP and can determine the integration of the CWPP process. Understanding how communities handled past adversities provides a window of opportunity to inspire change. Indeed, if these events were recent they serve as a leverage point to spur action, if you act quickly and before the urgency of desired action diminishes (A. M. S. Smith et al., 2016; Williams et al., 2012) . Identifying and documenting past collaborative eff orts can identify existing formal and informal community group structures that can support the collective actions of CWPP processes (Fleeger, 2008; Goldstein & Butler, 2012; Innes & Booher, 2014) . These groups can be wildfire specific, such as Fire Safe Councils or nonwildfire groups, such as homeowners’ associations or neighbourhood groups. These groups have the capacity to be organized for taking collective a ction in support CWPP efforts in addition to becoming peer advocates in implementing CWPP outcomes. CWPPs should also identify past wildfire risk reduction participants as they can serve as vital experience in working collaboratively and convey its benefi ts to new team members (Fleeger & Becker, 2010; Headwaters Eco nomics, 2016b; Horney, Nguyen, Salvesen, Tomasco, & Berke, 2016; Jakes & Sturtevant, 2013; K. C. Nelson, Souter, Jakes, & Williams, 2010) .

PAGE 42

25 Indeed, it also provides continuity of ongoing efforts and preserves the institutional knowledge of community wildfire efforts. It is also just as helpful to identify if the community has little collaborative or w ildfire planning experience. This helps to direct community’s efforts in identifying outside support and collaborators such as, public agencies, nongovernmental organizations or consultants to lead the necessary collaborative processes. If a community lack s collaborative skills, it may require investments in training or new personnel to ensure collaboration readiness. Finally, it is imperative that disagreements within the community are documented because earlier conflicts can often become barriers and derail CWPP collaborative efforts (Fleeger, 2008; Goldstein & Butler, 2012; Innes & Booher, 2014) . In some cases, it may be necessary to openly discuss past disagr eements to determine lessons learned and bridge past differences. In extreme cases this may require external, professional facilitation because of broken community trust. Goal and objectives The purported genius of the HFRA is that it is vague and allows communities to craft goals and objectives specific to their values in relation to wildfire risk reduction efforts (Jakes et al., 2011) . However, goals and objective s should go beyond vague terms of reducing wildfire risk, reducing fuel loads and structural ignitibility. Goals and objectives provide a community framework to minimize wildfire risks, but more importantly these should articulate implementable action item s across all aspects of the CWPP and best practices. The goals, objectives, and action items serve as the baseline for evaluating the plan’s performance. Per Oregon’s Resource Innovations Institute for a Sustainable Environment (2008) , goal s and objectives should include the following issues: partnerships and collaboration; risk assessment;

PAGE 43

26 fuels reduction; structural vulnerability reduction; emergency management; and education and outreach. Goals and objectives will vary depending on the s cale at which the CWPP is created. The issue of the appropriate geographic scale of the CWPP is neglected by the HFRA and the Resource Innovations Institute for a Sustainable Environment’s guidelines. Scale choice should be driven by the motivations and goals articulated in the CWPP (Jakes et al., 2007; Jakes et al., 2012; Rodman & Stram, 2008; Society of American Foresters, 2004) . For example, CWPPs should be developed at a small scale, e.g. neighborhood or community level, if the goals are to motivate homeowners to reduce hazards on their properties. However, CWPPs should be developed at a larger scale, e.g. counties, municipalities, and fire districts if the goals are to reduce regional landscape wildfire risk. This research effort evaluates county level CWPPs, and therefore the CWPPs should address the following goals and objectives at a minimum: defini tion of the WUI, risk mapping, plans to reduce structural ignitibility and hazardous fuels, and connection to multiple frames. Goals and objectives should clearly articulate and reflect community values and connect to multiple contexts and frames. Framing is important because CWPPs can be framed in a multitude of ways: fuels management, life safety, and ecosystem health. Jakes et. al. (2012) provides the following example, if someone frames wildfire management as a life safety concern, they will be more interested in a CWPP that frames goals and objectives in terms of evacu ation and response times rather than fuels management at a landscape scale for ecosystem health. As a result, it is critical to know which frame(s) are being considered and used in the CWPP process because it will often determine who chooses to participate in the CWPP process and identify potential conflicts that may arise determining which goals and objectives are

PAGE 44

27 priorities. CWPPs at the county scale should rarely use a single frame. It is necessary to consider multiple frames in order to broaden communit y engagement by persuading community members participation and implementation is in their own self interest. Landscape scale risk reduction efforts are complex and require a multitude of frames to be effective. Additionally, multiple frames broaden diversi ty, skills, and access to additional resources. For example, a frame of watershed and ecosystem health provides access to additional funding resources, such as The National Forests Foundation Grant Programs (National Forest Foundation, 2019) , that can also reduce fuel loads. Goals and objectives should be specific, measurable, achievable, relevant, and timebound (SMART). Following this SMART framework is beneficial because goals need to be specific enough to provide a link to objectives, which need to be measurable. Measuring success in implementing goals is critical to the long term success of projects in order to maintain community involvement and engagement in implementation activi ties. Setting goals that are not achievable or not measurable are counterproductive to maintaining risk reduction efforts. Irrelevant goals are detrimental because it raises questions about the veracity of the CWPP process and the other goals and objective s. Time sensitive goals help to create early and repeatable successes that are critical to maintaining momentum for long term success, helping to build trust among the participants. Short term goals, depending on the county’s past wildfire efforts, may be as simple as documenting the number of structures that are not wildfire code compliant in order to provide a better assessment of wildfire risk. Longterm goals may include ensuring all existing and new construction meets current code standards or reducing landscape scale fuel reduction loads to a more natural condition.

PAGE 45

28 Community capacity CWPPs should document community capacity. Community capa city includes characteristics such as community norms and values, economic diversity, growth trends, land ownership practices, and financial and physical resource availability. Tapping into resources that help CWPP participants successfully work together i s paramount because effectively functioning collaborative groups can overcome financial and resource obstacles (Jakes et al., 2012; Rodman & Stram, 2008) . The key benefits of this include building community capacity to be used during the CWPP process and other activities beyond wildfire planning. While the expansion of the network can provide more people, technology, and funding to the CWPP process, additional conflicts about CWPP goals can arise. The out comes of a CWPP go beyond the document itself, but also the expanding capacity for action it builds, opportunities and networks it creates, the knowledge it advances, and the connections among people and organizations it develops. Documenting and expanding community capacity also mobilizes individuals to participate and lend legitimacy to the CWPP, secure funding, and shepherd the process. These leaders might include federal, state, or local government representatives, community residents, or activists. Exp anded capacity can help achieve outcomes beyond wildfire preparedness such as watershed and forest health because the community’s knowledge of local ecological issues and role of fire ecology is expanded and they have built a stronger sense of community. C ommunities who do not identify and assess their capacity will often find the CWPP process is stalled because of inadequate funding, lack of physical capacity, and a lack of community and agency support. Partnerships and collaboration The success of the C WPP will hinge on the core team effectively engaging a broad range of stakeholders. Substantive input from a diversity of interests ensures the final document

PAGE 46

29 reflects the highest priorities of the community and helps to facilitate the implementation of re commended projects. Identifying leaders, within or outside the community, who can help mobilize others will help to expand the creation of core CWPP member and community participants (Jakes et al., 2012) . These leaders can also serve as catalysts for action and recruit others to the CWPP process. Stakeholders can include, but is not limited to forest management groups, city council members, resource advisory committees, HOAs, Division of Wildfire/Fish and Game, Department of Transportation, local and state emergency management agencies, water districts, utilities, recreation organizations, environmental organizations, forest products interests, local chambers of commerce, and watershed councils. To solicit additional input, the core team may choose to hold public meetings. The partnerships and collaboration should also outlin e which entities are responsible for which pieces of the CWPP and its implementation, with key timelines. Such efforts ensure accountability and are a mechanism of trust building among the community and CWPP signatories. Expansive partnerships help bring n ew ideas and resources to CWPPs and ensure that vulnerable communities are not overlooked. The partners should identify the shelf life of the CWPP as it is a living and guiding document, it will need semi regular updates and maintenance efforts. Base map The core team, agencies, and stakeholders should collaboratively develop a community basemap that includes: 1) inhabited areas at potential risk to wildland fire; 2) areas containing critical infrastructure that are at risk to fire disturbance events, e.g. escape routes, municipal water supply structures, and major power or communication lines; and 3) a preliminary designation of the community’s WUI zone (Jakes et al., 2012; Society of American Foresters,

PAGE 47

30 2004) . This map will be used to facilitate a community risk assessment, which helps the core team and community members to prioritize areas for treatment and to identify the highest priority uses for available financial and human resources. A successful and meaningful community assessment should be co developed considering the following risk factors: fuel hazards; risk of wildfire occurrence; homes, businesses, and essential infrastructure at risk; other community values at risk; and local preparedness and firefighting capacity (Jakes et al., 2012; Society of American Foresters, 2004) . Ranking systems typically consist of adjective rating systems, e.g. high, medium, and low. The key objective of these discussions is to develop the community’s prioritized recommendations for fuel treatment projects on federal and nonfederal lands in the WUI, including preferred treatment methods for each project. These decisions result in the action plan (identifies the roles and respo nsibilities, funding needs, and timetables for carrying out the highest priority projects) and the assessment strategy (ensures that the document maintains long term relevance and effectiveness) (Rodman & Stram, 2008) . WUI designations vary widely from CWPP to CWPP . The literature reports on a variety of definitions including the use of the Federal Registrar (C. N. Thompson, 2001) . Since the CWPP will be frequently updated it is important to identify and track how a commu nity identifies the WUI, to see how it grows and changes in the associated levels of risk are assessed. Several operational methods of defining the WUI have been proposed and are outlined below. Operationalizing the WUI requires defining the three integral components of the WUI: human presence, wildland vegetation, and wildfire risk, which are discussed below. Human Presence and Development Operationalizing the measurement of human presence in the WUI involves either zonal or point based measurements (Bar Massada et al., 2013) . The zonal approach requires an areal unit,

PAGE 48

31 and traditionally, the areal units or zones are census blocks or parcels. The Federal Register density measurement works very well with census blocks because housing data is avail able across the United States, is consistently collected, and can serve as the unit of analysis for which housing and vegetation are evaluated (Bar Massada et al., 2013) . Radeloff et al (2005) operationalized an integrated WUI definition using the Federal Register’s WUI definition as a starting point and combining this with detailed housing density data, high resolution vegetation data, and spatial adjacencies to wildland vegetation1. Their WUI measurements are unique because they use thresholds of wildland vegetation requirements as core parts of interface and intermix classifications. Addition ally, they use a 2.4 km community fire planning zone distance because it represents the estimated distance a firebrand can fly ahead of a fire front. With this schema, if a census block were only partially within the 2.4 km distance, the census block would be split so that only the portion within the 2.4 km would be included as interface. However, simply splitting census blocks is problematic, as the distribution and locations of housing density within the census block is unknown. A more conservative measur e would be to include the entire census block, particularly as Radeloff et al (2005) report that census blocks ranged from 0.001 km2 to 2,700 km2. Theobald and Romme’s (2007) WUI definition and measurement expands Radeloff et al’s (2005) research in three ways: 1) using a differing definition of WUI that adjusts census blocks to refle ct development patterns; 2) integrating a more detailed approach to classifying wildland vegetation; and 3) adding wildfire risk categorizations. Through ‘ad hoc analysis’ Theobald and Romme (2007) have determined the Federal Register’s WUI definition of 1 See Radeloff et al (2005) for specific details.

PAGE 49

32 per acre does not adequately address all WUI development types at risk; however, it is unknown if they have empirically evaluated th is assumption. Future research should confirm Theobald and Romme’s ‘ad hoc analysis’ or further refine a more nuanced definition of the WUI that is synonymous with wildfire risk. Finally, Theobald and Romme (2007) removed the portion of each census block that overlapped with protected lands because private development typically cann ot be built on public or protected lands, and they also removed blocks that overlapped water features. This approach eliminates the error Radeloff et al (2005) encountered by splitting a census block based on a given distance, as they are unaware as to the distribution of housing within the census block. Approximately 1/3 of Colorado is public land, and by removing these public lands from the calculations, the area average was reduced nearly 40%, resulting in identifying over 131,600 additional hectares (~18%) of WUI. The WUI intermix lands in need of wildfire mitigation defined by the Federal Registrar are different from those identified by researcher definitions. While the Federal Registrar determines the density of WUI intermix to be 1 housing unit per 0.30 – 40 acres, Theobald & Romme (2007) calculate a WUI intermix density of 2.40 – 40 acres for wildfire mitigation. This difference is a result of Theobald & Romme (2007) accounting for a community protection zone to surround the traditionallydefined WUI intermix area. This approach incorporates gradients of housing density in the intermix zone. However, the thresholds need to be validated and refined in future research because they were based off the researchers’ assumptions rather than data analysis. B oth Radeloff et al’s (2005) and Theobald and Romme’s (2007) approaches to measuring WUI su ffer from the modifiable areal unit problem (MUAP), which is a bias that is the result of using point based measurements aggregated into areas for which summary statistics are

PAGE 50

33 calculated (Bar Massada et al., 2013) . As previously mentioned, cens us block geometry can vary in size and there is no way of knowing exactly where each person lives. Comparing census blocks to each other is problematic because the population distribution unknown density and distribution. MUAP issues contain two parts: sca le and zonal issues. WUI researchers have written extensively on zonal MUAP issues while little is written on WUI mapping scale MUAP issues. The point based approach to operationalizing the WUI for measurement is exemplified through the work of Bar Massada et al (2013) , who refine d the measurement of the WUI by using housing location data. Bar Massada et al (2013) used a moving window analysis with various window sizes to represent neighborhood sizes in order to calculate housing and wildland vegetation. The results showed similar area results from previous zonal measures (Radeloff et al., 2005; Stewart et al., 2007) but produced more precise spatial location results (Arganaraz et al., 2017; Bar Massada et al., 2013) . This process, however, can only be used in areas where housing location data is available. Additionally, while the neighborhood radiuses of the moving window analysis are scalable, sensitivity analysis shows that different urban forms across the United States produce varying results; each urban location has a different best fit radius. Wildland Vegetation U.S. Geological Survey (USGS) National Land Cover Data (NLCD) is often used to classify wildland vegetation, which is comprised of coniferous, deciduous, and mixed forest; shrubland; grasslands/herbaceous; transitional; and woody and emergent herbaceous wetlands (Radeloff et al., 2005; Theobald & Romme, 2007b; Bar Massada et al., 2013) . This vegetation classification is adequate for national, state or landscape scale WUI assessments; however, among the three forest types, there are numerous types of plant communities with differing fire

PAGE 51

34 regimes and risks. In an effort to confront this issue, Theobald and Romme (2007) augmented the NLCD data set with higher resolution vegetation data to map wildfire risk. Wildfire Risk Wildfire risk is addressed inconsistently in measurement of the WUI.2 Radeloff et al (2005) did not explicitly address wildfire risk o ther than utilizing the travel distance of a fire brand (2.4 km) in their WUI interface mea surements. HFRA typical distances (800, 1600 and 3200m) for treatments in community protection zones (CPZ) were used by Theobald and Romme (2007) . These distances capture the range of wildfire fighting objectives: structure protection, safe fire fighting zone based on the maximum sustained flame length of a crown fire, and avoidance of flying embers (Theobald & Romme, 2007) . Under extreme conditions, however, these distances are not sufficient. Additionally, these distances vary depending on the differing vegetation communities, slope, wind, and types of fires. Theobald & Romme (2007) used a variable width buffering technique that utiliz es cost distance computations, but the weights used are arbitrary and do not reflect the role that topography, wind, stand age/structure, and fire types play in wildfire behavior within different vegetation communities. The greater oversight is integrating wildfire risk into a model without considering defensible space, which can reduce a home’s vulnerability (Theobald & Romme, 2007) . Wildfire risk categories, classified using an aggregated synthesis of raster data sets, were defined as 1) high risk or crown fires, 2) low risk or ground fires, 3) variable risk fires, which are 2 As a reminder, the working definitions for hazard, risk, and vulnerability are: 1) hazard is the potential threat to be exposed to a wildfire event (Ager, Day, McHugh, et al., 2014; Ager et al., 2012) ; 2) risk is the impact that a wildfire event could have on community infrastructure and populations, assuming the community population has an equal ability to respond to wildfire events (Bryant & Westerling, 2014) ; vulnerability is an individual or household’s ability to anticipate, respond to, and recover from a wildfire event, in addition to broader hazard and risk elements (Collins, 2008a).

PAGE 52

35 generally low risk fires that c an occasionally become high risk fires due to natural cycles and 4) high risk fires that were historically low or variable fire regimes but are now high risk due to high fuel loads due to fire suppression policy (Theobald & Romme, 2007) . High severity fire regimes involve fires burning at a high intensity through crowns and are often difficult to contain or suppr ess, posing the greatest threat to structures (Theobald & Romme, 2007) . Public perception of wildfire general ly involves images of these crown fires with the large flames and damage to infrastructure. Low severity fire regimes are fires that burn at a relatively low intensity through surface fuels. These fires minimally spread into tree or shrub crowns, making th em relatively easy to contain or suppress (Theobald & Romme, 2007) . Both the variable fire regimes and high r isk fires that were historically low fires have significantly increased in severity due to the fuel load accumulation caused by the fire suppression policies of the 20th century (Theobald & Romme, 2007) . This fuel load accumulation becomes ladder fuel, which allows these fires to turn into crown fires. While many risk modeling and reduction efforts have focused on high intensity crown fires because of their threat to structure ignition, the ground fire and variable fire regimes is underestimated in terms of wildfire risk and, therefore, in wildfire mitigation. Due to differences in operationalization, the estimates of WUI lands varies significantly. According to Radeloff et al (2005) ,719,156 km2 of WUI exists in the United States. Theobald and Romme ( 2007b) estimate WUI interface to be 465,614 km2 nationwide . And while these WUI area estimation methods and results are significantly different, it is important to n ote that both sets of results illustrate a significant growth in the WUI and project continued growth into the future. This growth is in existing WUI areas as well as significant grown in new areas, particularly the Intermountain West (Radeloff et al., 2005; Stewart et al., 2007; Theobald &

PAGE 53

36 Romme, 2007) . Therefore, wildfire mitigation efforts in the WUI are increasingly important to reducing threats to losses of life and of economic interests. As a result CWPPs should be very clear in how they are defining the frame of their goals and objectives because it will often influence how they define and map the WUI, which can have very clear biases in how risk is assessed. Risk assessment The WUI definitions are largely driven by how communities perceive wildfire, as a hazard, risk, or vulnerability, as a result how a community defines and maps its WUI will often determine the means with which it assess risk. I t is important to distinguish the difference between wildfir e hazard, risk, and vulnerability since these terms are often used interchangeably in the literature and media. Wildfire hazard relates to factors affecting the fire environment and likely fire behavior, including fuel and vegetation properties, topography, climate and weather variables, and ignition characteristics (Bryant & Weste rling, 2014; M. P. Thompson, Ager, Calkin, Finney, & Vaillant, 2012) . Wildfire risk characterizes the potential for wildfire to harm human life and safety or damage highly valued resources and assets (HVRAs) (Ager et al., 2011; Keane, Drury, Karau, Hessburg, & Reynolds, 2010; M. P. Thompson et al., 2012) . While WUI definitions mention and imply wildfire risk, they give no specific criteria to meas ure or calculate ris k (Haas, Calkin, & Thompson, 2013) . As a result, there are varying means and approaches to how wildfire risk is addressed in the WUI (Ager, Vaillant, Finney, & Preisler, 2012; Bryant & Westerling, 2014; Elia, Lafortezza, Colangelo, & Sanesi, 2014; Haas et al., 2013) . For example, wildfire forest resource management literature defines risk as an element of hazard exposure, but does not include wildfire impacts (Ager et al., 2012; Carmel, Paz, Jahashan, & Shoshany, 2009; Finney, McHugh, Grenfell, Riley, & Short, 2011) . Land use planning defines wildfire risk as a

PAGE 54

37 combination of fire exposure and impacts (Abukhater, 2011; Bryant & Westerling, 2014; Haas et al., 2013; Paveglio, Prato, & Hardy, 2013) . Finally, political ecologists include concepts of socio economic vulnerability (T. W. Collins, 2008a, 2011) . Wildfire vulnerability is an individual or household’s ability to anticipate, respond to, and recover f rom a wildfire event, in addition to broader hazard and risk elements (T. W. Collins, 2008a) . Vulne rability involves three realms: 1) root causes, 2) dynamic pressures, and 3) unsafe conditions. The first realm —root causes—refers to a wide range of historical, political, economic, demographic, and environmental factors that produce unequal distributions of resources (T. W. Collins, 2008b) . Previous wildfire risk studies have operated under the assumption that households are exercising free residential choice, particularly in the face of resort centric or affluent WUI development (T. W. Collins, 2008b) . The integration of LUCC modeling and vulnerability approaches to wildfire risk could furthe r our understanding of future WUI development and risk. The second realm of vulnerability involves dynamic pressures. Dynamic pressures are processes and activities, such as rapid population growth, urbanization, environmental degradation, global economic pressures, political conflict, (T. W. Collins, 2008b) . These processes, together, create unsafe conditions under which some people in a given place and time must live. Fin ally, the third realm of vulnerability —unsafe conditions —includes both spatial location and other characteristics of the built environment and socioeconomic barriers. Built environment vulnerability factors include architectural and landscape architectura l design choices, as well as the institutionalization of real estate market and planning decisions that determine residential settings, such as single family subdivision, mobile home park, apartment

PAGE 55

38 complex, condominium, gated enclave, and isolated dwellings (T. W. Collins, 2008b) . Additionally, socioeconomic vulnerability factors include fragile livelihoods, resource dependency, inadequate incomes, education, legal and poli tical inequalities, and a lack of preparedness for emergencies (T. W. Collins, 2008b) . P olitical ecologist s have contributed to a broader understanding of wildfire risk by utilizing the concept of vulnerability to provide understandings of wildfire mitigation perceptions, and barriers to implementing wildfire safe mitigations practices. Unfortunately, their work has been largely ignored in modeling efforts due to the complex and contextual nature of vulnerability factors (T. W. Collins, 2008b) . This is of particular concern because a third of the 13 million WUI residents of the western US lack incomes sufficient to meet basic economic needs, much less the cost of basic wildfire protection (Lynn, 2003) . Additionally, due to the pervasive use of modeling in decision making processes (Papadopoulos & Pavlidou, 2011) , the omission of vulnerability in these models becomes increasingly important. Modeling efforts, while omitting vulnerability, focus on hazard and risk. Hazard modeling is generally split into two approaches: 1) wildfire exposure and 2) wildfire behavior. Wildfire exposure models explore the predicted scale and spatiotemporal relationships of cau sative risk factors. Therefore, they evaluate the likelihood of a wildfire event based on current climate and weather, vegetation, slope, and fuel loads (Stratton, 2006; M. P. Thompson et al., 2012) . Wildfire behavior models simulate the behavior of the fire itself, producing burn probabilities for fire intensity, rate of spread, flame length, and crown fire activity (M. P. Thompson et al., 2012) . Wildfire risk modeling combines the two types of hazard models — exposure and behavior —and efforts to model the likelihood of wildfire interacting with valued resources loads (Stratton, 2006; M. P. Thompson et al., 2012) . In order to adequately model

PAGE 56

39 wildfire risk, multiple iterations of the wildfire behavior model are calculated for a variety of conditions in order to produce a comprehensive burn probability (M. P. Thompson et al., 2012) . Therefore, differing definition s and approaches to risk lead to differing wildfire risk and behavior models. Disagreement about wildfire processes, topography, environmental factors and their interactions produces variable levels of uncertainty and error, which complicates planning effo rts to reduce risk (Cochrane et al., 2012; Finney, 2002; Green, Finney, Campbell, Weinstein, & Landrum, 1995; Keane, Mincemoyer, Schmidt, & Garner, 2000) . The different types of wildfire modeling, the benefits and drawbacks of each model, and demonstrates the need for vulnerability in w ildfire risk modeling efforts for effective planning policy and mitigation efforts will now be discussed . First, a review of Landscape Fire and Resource Management Planning Tools (LANDFIRE) is needed because the tools of this program are often used in wildfire modeling efforts. LANDFIRE Landscape Fire and Resource Management Planning Tools (LANDFIRE), is a program provided by the wildfire fire management programs of the U.S. Department of Agriculture Forest Service and U.S. Department of the Interior. LAND FIRE is a collection of the most prominent nationwide fire hazard mapping products and used in a wide range of wildfire modeling efforts. LANDFIRE products are designed and devel oped to be used at the landscape level in order to facilitate national and reg ional strategic planning as well as the reporting of wild land fire . LANDFIRE provides landscape scale geo spatial vegetation, fire regime, topographic, fuel disturbance, and reference database products at 30 meter pixels (Stratton, 2009) . Therefore, the adaptation of these products can su pport a variety of local manage ment applications (Stratton, 2009) .

PAGE 57

40 The benefits of using L ANDFIRE are that it is: 1) consistent landscapescale and cros s boundary geospatial products; 2) adaptable to support local planning, management, and monitoring activities requiring co nsistent vegetation data; 3) aid in strategic and tactical planning for fire operations where other necessary data are unavailable; and 4) usable by Federal and State agencies and pr ivate organizations to collaborate regarding fire and other natural resource management issues. Additionally, LANDFIRE receives comprehensive upda tes roughly every 10 years, with incremental updates as available. Wildfire Exposure Modeling Wildfire exposure modeling, also known as fire danger mapping or fire potential index, includes mapping and modeling various biophysical elements of r isk, but doe s not include the expected impacts of wildfire or concepts of social and community vulnerability. Exposure modeling efforts focus on causative risk factors, such as flame length and fire size to quantify risk as burn probability (BP) —the likelihood of a gi ven location experiencing a wildfire during a specified period of time (Ag er et al., 2012) . Wildfire exposure literature acknowledges BPs are best calculated with a composite of fuels, topography, and weather conditions (Ager et al., 2014; Finney, 2006; Stratton, 2006) , yet many risk assessments preclude one or more of these critical variables because operationalizing these factors across larges scales is difficult. The difficulty is due to incompatible data resolutions, gaps in data availability, daily fluctuations of datase ts, or expertise and time required to gather adequate fuel load data (Keane et al., 1998; Keane et al., 2000; Kramer et al., 2014; Papadopoulos & Pavlidou, 2011; Stratton, 2009) . These issues are highlighted in the two predominate US exposure models: The Fire Danger Rating System (NFDRS) and the National Fire Danger Rating’s (NFDR) Fire Potential Index (FPI). These models provide the basis for state and national risk mapping and are used to allocate financial

PAGE 58

41 resources for mitigation efforts and suppression activities. They also provide large national trend datasets for national forest and WUI policy decisionm aking. The NFDRS uses current and antecedent weather, fuel types, and both live and dead fuel moisture (Stratton, 2004) . NFDRS ratings are based on an adjective class rating system (Table 3.1), which is calculated by normalizing rating classes across different fuel models, indexes, and station locations to map daily fire danger (Figure 3.1). Classes are based on the following dat a collected at fire stations: primary fuel model, staffing levels, and climatological class breakpoints (Arroyo, Pascual, & Manzanera, 2008; Haas et al., 2013; Keane et al., 2010; Keane et al., 2000; Papadopoulos & Pavlidou, 2011) . Most of these stations use the Burning Index (BI) —a measure of fire intensity (Geographic Area Coordination Centers, n.d.) ; however, a few use the Energy Release Component (ERC) (Keane et al., 2000) . BI combines the Spread C omponent (SC) and ER C to relate to the contribution of fire behavior to the effort of containing a fire. BI has no units, but in general it is 10 times the flame length of a fire. SC is a rating for the forward rate of spread at the fire head (Geographic Area Coordination Centers, n.d.) . SC integrates the effects of wind, slope, fuel bed, and fuel particle properties. The daily variations are caused by the changes in the wind and moisture contents of the live fuels and the dead fuel time lag classes of 1, 10, and 100 hr. ERC is an estimate of the potential available energy released per unit are a in the flaming zone of a fire (Geographic Area Coordination Centers, n.d.) and is dependent upon the same fuel characteristics as the SC . E RC—expressed in BTU’s per square foot —is derived from predictions of the rate of heat release per unit area during flaming combustion and the duration of the burning. D ay to day variations of the ERC are caused by changes in the moisture contents of the va rious fuel classes, including the 1000hour time lag class.

PAGE 59

42 Table 3. 1. Fire Danger Rating System adjective class ratings Fire Danger Rating and Color Code Description Low (L) (Dark Green) Fuels do not ignite readily from small firebrands although a more intense heat source, such as lightning, may start fires in duff or punky wood. Fires in open cured grasslands may bum freely a few hours after rain, but woods fires spread slowly by creeping or smoldering, and burn in irregular fingers. There is little danger of spotting. Moderate (M) (Light Green or Blue) Fires can start from most accidental causes, but with the exception of lightning fires in some areas, the number of starts is generally l ow. Fires in open cured grasslands will burn briskly and spread rapidly on windy days. Timber fires spread slowly to moderately fast. The average fire is of moderate intensity, although heavy concentrations of fuel, especially draped fuel, may burn hot. Short distance spotting may occur, but is not persistent. Fires are not likely to become serious and control is relatively easy. High (H) (Yellow) All fine dead fuels ignite readily and fires start easily from most causes. Unattended brush and campfires are likely to escape. Fires spread rapidly and short distance spotting is common. High intensity burning may develop on slopes or in concentrations of fine fuels. Fires may become serious and their control difficult unless they are attacked successfully while small. Very High (VH) (Orange) Fires start easily from all causes and, immediately after ignition, spread rapidly and increase quickly in intensity. Spot fires are a constant danger. Fires burning in light fuels may quickly develop high intensity characteristics such as long distance spotting and fire whirlwinds when they burn into heavier fuels. Extreme (E) (Red) Fires start quickly, spread furiously, and burn intensely. All fires are potentially serious. Development into high intensity burning will usually be faster and occur from smaller fires than in the very high fire danger class. Direct attack is rarely possible and may be dangerous except immediately after ignition. Fires that develop headway in heavy slash or in conifer stands ma y be unmanageable while the extreme burning condition lasts. Under these conditions the only effective and safe control action is on the flanks until the weather changes or the fuel supply lessens. Note: Source (United States Forest Service Rocky Mountain Research Station, 2016) .

PAGE 60

43 Note: Source (United States Forest Service Rocky Mountain Research Station, 2016) . Figure 3. 1. Observed Fire Danger Class The NFDRS and Fire Potential Index (FPI) use fuel data sets that are based on remotely sensed vegetation classes. These classes are refined through a process called ground truthing, during which various vegetative locations are sampled to further classify typi cal vegetation fuel characteristics. Vegetative characteristics include percent cover, height, and diameter on the four major tree and shrub species; and percent cover and depth of subshrubs, forbs, mosses, and grasses; and shrub and grass morphology and density classes. Additional key data inputs for these models are relative greenness (RG) of vegetation and time lag live and dead fuel moisture (Keane et al., 2000; Ottmar, Blake, & Crolly, 2012; Papadopoulos & Pavlidou, 2011) . The FPI

PAGE 61

44 adequately addresses the computational complexity of the NFDRS by eliminating wind and increasing resolution of fuel modeling, NDVI derived composites of RG and RG trends, and using a 10hour time lag fuel moisture model (Keane et al ., 2000; Stratton, 2006) . While US national coverage of digital elevations models (DEM) at 30 and 10 meter resolutions, and in some cases 3 meter resolutions, are available, neither FPI nor NFDRS use topography as a risk variable because topography datasets, such as slope and aspect are computationally expensive to compute and integrate across large regions. Yet, topography is critical at local scales to understand wildfire risk and behavior. While the NFDRS is used in state and national efforts, local fire managers use historic fire weather climatology to set current and future staffing class breakpoints (Hayes, Ager, & Barbour, 2004) . Using historic fire weather climatology is problematic because v alues between stations are estimated with an inverse distancesquared techn ique on a 10km grid , which works well for areas with high station density, but does not work well in low density areas like the WUI. Additionally, research has shown that historic data is no longer an accurate predictor of future wildfire risk due to clim ate change (Bryant & Westerling, 2014) . However, two issues preclude using the NFDRS in wildfire land use planning efforts: 1) the modeling process is computationally complex and resource intensive and 2) it is not scal able for local decisionmaking because of the reliance on fire station data and ignores critical fire risk data sets such as topography. Behavior Modeling Behavioral modeling, including simulator models, are used to fill the gaps of exposure modeling efforts. These models are more appropriate for local decision making because they provide a greater level of detail and use a composite simulated approach to measuring risk.

PAGE 62

45 Behavior models measure risk as a composite of exposure or probability of fire likelihood, fire intensity, fire effects and impacts, and fire event simulations or rate of spread (ROS) (Haas et al., 2013) . The key benefits of behavioral modeling include: the ability to understand and prepare for wildfire behavior during fire suppression activities, understand the future impacts of wildfires, provide enough detail for local decision m akers to identify various contextual risk factors, and evaluate the effectiveness of various mitigation scenarios (Ager et al., 2014; Finney, 2001; Keane et al., 2000; Stratton, 2004; van Wagtendonk, 1996) . Over 23 behavioral models have been identified in the literature (Papadopoulos & Pavlidou, 2011) . However, two sim ulation models stand out: FARSITE and FlamMAP. The FARSITE simulator model stands out because FARSITE: 1) is a mature modeling environment and in use by many levels of governmental agencies, 2) supports a variety of input data and modeling parameters, 3) uses input data based on spatial data, 4) utilizes input and output data that are very detailed and multiparametric, making it more reliable and accurate, 5) handles multiple fire fronts and ignitions, and 6) can accommodate customized, high resolution inpu t variables (Arroyo et al., 2008; Keane et al., 2000; Papadopoulos & Pavlidou, 2011; Stratton, 2006, 2009) . FARSITE simulates wildfire growth and be havior for long time periods across a variety of terrain, fuels, and weather (Papadopoulos & Pavlidou, 2011) . It is a deterministic modeling s ystem, meaning that simulation results can be compared to all model inputs, allowing for evaluation of uncertainty, error, and understanding of contextually significant variables (Carmel et al., 2009; Stratton, 2009) . FARSITE uses a variety of mathematical fire models to adequately capture the complex dynamics and behavior of fire. Models include Rothermel’s (1972) surface fire spread model, Van Wagner’s (1977, 1993) crown fire initiation model, Rothermel’s (1972) crown fire spread model, Albini’s (1976) , spotting model, and Nelson’s (2000) dead fuel

PAGE 63

46 moisture model. The value that each of these models brings, is that FARSITE can be used to simulate air and ground suppression actions and land use and mitigation actions, including fuel treatments (P. Berke et al., 2015; Gebru et al., 2017; Sang, Zhang, Yang, Zhu, & Yun, 2011; Tian, Ouyang, Quan, & Wu, 2011; Yun, Chen, Li, & Tang, 2011) . The process for FARSITE data acquisition is outlined in Figure 3.2. Note: Source (Stratton, 2009) . Figure 3. 2. LANDFIRE fuels data acquisition process for FARSITE. Despite the many benefits of FARSITE, the model has three technical drawbacks: 1) the highly complex needs to customize contextual input models and variables; 2) the computational expense of high resolution inputs and outputs; and 3) the lack of updates to the model (Papadopoulos & Pavlidou, 2011) . Additionally, there is a critical flaw in using behavioral models in risk mapping e fforts. Behavioral models address what ifs of particular wildfire events but have not traditionally been used to integrate the broader landscape trends of risk due to the needed computational complexity (Papadopoulos & Pavlidou, 2011; Stratton, 2009) .

PAGE 64

47 Advancements, however, have been made by Thompson et al. (2015) in their exploration of using the FSim wildfire behavior model in the development of a probabilistic method for wildfire suppression cost modeling. While their implementation of FSim facilitated the efforts of evaluating California’s level of risk at a state scale, it would be of limited use for local decision making due to the simplification of risk outputs in FSim compared to FlamMAP. FlamMAP —another behavior model —is compatible with FARSITE, and they are often used together in comprehensive modeling processes. FlamMAP uses the same datasets as FARSITE, which allows for modeling continuity, but FlamMAP focuses on the spatial variability in fire behavior, (Finney, 2006) . FlamMAP fire behavior calculations are performed independently for each cell on the gridded landscape (Finney, 2006) . Original spatial dataset layers are often differing resolutions; however, they must be processed into identical resolution, extent, and coregistered in order to be used in FlamMAP (Finney, 2006) , a process that can introduce uncertainty an d error. Additionally, the structure and types of datasets are not dynamic. Indeed fuel moisture, wind speed, and wind direction are constant in time (Finney, 2006) . As a result, they cannot adequately model live fire events with changing weather patterns. Basic model outputs include: fireline intensity, flame length, rate of spread, heat per unit area, horizontal movement rate, midflame windspeed, spread vectors, cr own fire activity, solar radiation, 1 hr dead fuel moisture, and 10hr dead fuel moisture for each pixel at a given point in time (Finney, 2006) . Additionally, a different suite of inputs is generated for minimal travel time calculations on basic FlamMAP outputs, including: rate of spread, influence grid, arrival time grid, fireline intensity grid, flow paths, major paths, arrival time contour, and burn probabilities (Finney, 2006) . An emerging benefit of FlamMAP is its ability to evaluate pre and post mitigation fuel treatments effects on wildfire behavior, allowing insights into mitigations effects

PAGE 65

48 on future risk (Fi nney, 2006) . However, this use of FlamMAP is still limited to large fires and mitigation treatments are targeted to work under a very specific set of weather and fuel moisture conditions (Finney, 2006) . CW PP Wildfire Risk Modeling As mentioned above , wildfire behavior models can be utilized as wildfire risk models by running multiple simulations of different conditions in order to calculate risk. FARSITE and other behavioral models are utilized in this way in order to create risk models, but still do not specifically address issues of social vulnerability or community values—issues that help in prioritizing wildfire land use and mitigation decision making, which is an important omission in risk modeling and the subsequent development of mitigation efforts and priorities. However, these models can be used in conjunction with other tools to integrate some aspects of vulnerability. For example, these models could estimate areas of high burn probability, which co uld then be overlaid with maps of specific types of vulnerability. Maps of low economic status, minority ethnicities and races, and affordable housing can add the missing vulnerability aspect of these models in order for planners to make decisions based on a more comprehensive context of the area. Initial efforts to overcome this gap in the literature have been undertaken by Elia et al (2014) and Paveglio et al (2013) . Elia et al (2014) modeled risk using population densities to streamline the spatial allocation of fuel removals. Paveglio et al (2013) used current and projected future land use scenarios and total economic value to document and prioritize areas of higher and lower risk for mitigation efforts. Still, neither of these studies fully addres s issues of vulnerability, which are needed to assist in prioritizing mitigation locations. Wildfire risk models are critical to identifying and prioritizing municipal wildfire treatment areas (Keane et al., 2010) . However, current risk models lack the appropriate

PAGE 66

49 resolution or scale of datasets, integration of vulnerability information, or community values for decisionmakers to adequately evaluate and prioritize land use and mitigation decisions (Miller & Ager, 2013) . Additionally, many modeling efforts do not include uncertainty and error assessments. Of the models and research reviewed, only one, FARSITE, clearly articulated and evaluated uncertainty and error (Stratton, 2009) . Indeed, many modeling efforts of the last decade have larg ely ignored uncertainty and errors in data model inputs (Alexander & Cruz, 2013; Cruz & Alexander, 2013; Miller & Ager, 2013) . While all models mentioned climate changed induced increases in risk, only one model sought to address it (Bryant & Westerling, 2014) . Bryant and Westerling (2014) did account for projected weather related climate change, changing demographics, and development patterns, but they did not account for changes in vegetative land cover migration. Additionally, models need to adequately address the temporal dynamics of wildfire risk (Miller & Ager, 2013) . As a result, there is need for a broader wildfire risk evaluation framework to support decisionmaking. CWPPs should use scientifically appropriate modeling methods to evaluate risk, specific model choices will depend on their frame and the associated goals and objectives; however, the process should be well documented so it can be updated and reproduced in the future. Additionally, risk mapping efforts can help quantify existing and changing conditions related to the population; age; percentage of youth; percentage of elderly; number of housing units; percent of owner and renter occupied housing units; percentage of people in the labor force; percentage of fa milies below the federal poverty line; unemployment rate; length of homeowner tenure; full time/parttime residency status; and income at risk. These efforts will help prioritize specific goals, objectives, and project locations with the community and whet her the community is meeting their target risk reductions over time.

PAGE 67

50 Hazardous fuels reduction The goal of fuel treatment is to modify potential fire behaviour, thus minimizing the negative effects of wildfire. Specific to CWPPs, the goal is to reduce risks to human lives and communities and improving ecosystem health. Inherent in these broad goals are the goals of reducing fire intensity, rate of spread, severity of fire effects, and restoring historic fuel quantity and structure (Husari, Nichols, Sugihara, & Stephens, 2006; Society of American Foresters, 2004) . HFRA suggests C WPPs document fuel reduction priorities and projects. These priorities should be for both federal and nonfederal land in the WUI, including the preferred treatment methods for each project location. Additionally, a list of best practices should be provide d, so individuals and the community can further reduce risk. Hazardous fuels reduction goals should be clearly articulated and indicate wither priority projects serve to protect the community and its essential infrastructure or are geared toward reducing r isks to other community values (Society of American Foresters, 2004) . Priority project lists should also have timelines and reoccurring evaluation periods because fuel reduction efforts are not a static, one time undertaking (Burns & Cheng, 2007; A. M. S. Smith et al., 2016) . CWPPs should fully articulate fuel management basics and treatment types, which are outlined below. Fuel Management Basics Fuel Fuel is defined as live and dead plant biomass (Ager et al., 2011; Arroyo et al., 2008; Husari et al., 2006) . Fuel moisture, chemical composition, surface area to volume ration, size, and structural arrangement of the fuel in the stand and on the landscape influence the conditions under which fuel will burn (Husari et al., 2006) . Additionally, these characteristics will also determine the characte ristics and nature of the resulting fire. Fuel management is the intentional

PAGE 68

51 act of manipulating the amount, composition, and structure of fuel within wildland ecosystems for the express purposed of modifying potential fire behaviour and effects (Agee et al., 2000; Ager et al., 2014; Ager et al., 2011; Arroyo et al., 2008; B. M. Collins et al., 2013) . Fuel has 15 characteristics that can be modified to influence it s potential to burn and the characteristics of the result wildfire, including: total fuel quantity, fuel size, packing ratio of surface fuel, surface fuel continuity, crown fuel continuity, surface fuel, crown fuel, horizontal fuel continuity, vertical fue l continuity, ladder fuel, potential for surface fire, and potential for crown fire (Husari et al., 2006) . Fuel Quantity The quantity of fuel in an ecosystem is an important factor to determining the character and impact of fires (Stratton, 2004) . The metric used to describe the amount of fuel is dry weight per unit area (tons/acre) (Ager et al., 2012; Arroyo et al., 2008; Husari et al., 2006) . This dry weight is divided into size classes. Each size class is based on the time the fuel takes to reach equilibrium with moisture in the air (Arroyo et al., 2008; Husari et al., 2006) . For example, small fuels (e.g., pine needles) respond to changes in relative humidity more rapidly than large dense logs. Fuel can be removed from a site by a variety of means, thus reducing fuel quantity. Fuel Size Fuel particle size is important to determining the likelihood of ignition and determining the resulting fires behaviour and effects (Jack D. Cohen & Finney, 2010) . Fine fuels, fuels less than a quarter inch in diameter, have the greatest influence on the ignition and spr ead of fires (Husari et al., 2006) . Since fine fuels are integral in fire ignition the remo val of fine fuels is often a primary focus of fuel management projects (Jack D. Cohen & Finney, 2010) .

PAGE 69

52 Packing Ratio Packing ratio is the measure of how densely packed fuel particles are (Husari et al., 2006) . Fuel may be compacted through a variety of mechanical treatments including mastication, chipping, and shredding. Compacted fuel burns more slowly because the oxygen required for combustion is not readily available to the fuel away from the surface (Husari et al., 2006) . Surface Fuel Surface fuel is composed of small shrubs, grasses, and plant detritus lying on the ground (Agee & Skinner, 2005) . Surface fuel is necessary for fire to spread continuously across landscapes. Fuel management goals are achieved by disrupting the continuity of surface fuels (Husari et al., 2006) . Crown F uel Tree branches and foliage and shrubs over six feet in height are considered crown fuel (Cruz, Alexander, & Wakimoto, 2003) . Continuous crown fuel is required to spread fire through the tree canopy as a crown fire. Crown fires can spread in discontinuous stands of trees if su pported by surface fire. Wind speed and foliar moisture are important to the spread of crown fires. Removing trees and ladder fuels and treating surface fuels reduces crown fire risks because they reduce the continuity and bulk density of crown fuels, whil e increasing the separation between crown and surface fuels (Agee & Skinner, 2005; Ager et al., 2014; Husari et al., 2006) . Horizontal Fuel Continuity Horizontal fuel continuity is nec essary for a surface or crown fire to spread laterally across the landscape (Husari et al., 2006) . Surface discontinuities act as barriers to fire spread under most conditions; however, under extreme conditions fires can spot across bare areas. Fuel

PAGE 70

53 treatments that aim to disrupt fuel continuity include: fuel breaks and strategically placed area treatments (Finney, 2001) . Vertical Fuel Continuity Vertical fuel continuity is necessary for surface fires to spread into the crown or canopy of trees (Finney, 2001; Husari et al., 2006) . Vertical fuel continuity can be reduced to increase the separation between surface and crown fuels, thus reducing the probability of crown fires (Agee & Skinner, 2005; Finney, 2001; Husari et al., 2006) . Ladder Fuel Fuel, such as intermediate sized trees or shrubs, th at provide a fuel conduit that allows a surface fire to ‘climb’ into crown fuel (Cruz et al., 2003; Hus ari et al., 2006; Riccardi et al., 2007) . Fuel treatments should remove shrubs, small trees, and lower branches to reduce ladder fuels (Cruz et al., 2003; Husari et al., 2006; Riccardi et al., 2007) . Types of Fuel Treatments Fuel treatments take on a variety of forms, but are generally divided into two treatment categories: fire and mechanical (Husari et al., 2006) . The use of fire as a treatment method is beneficial because it is a key process in many plant ecosystems in the American West and also modifies fuels (North, Collins, & Stephens, 2012; Oldham, 2016; Vaillant & Reinhardt, 2017) . Mechanical treatments, including forest thinning, mastication, and grazing, also modify fuels. Mechanical treatments are often used to restore fuel conditions where fire can be used to maintain the desired range of conditions over a longer period of time. HFRA directs half of federal fuel reduction funds to be used in the WUI even though fuel reduction treatments in the WUI is more expen sive and limits the amount of fuel treatments elsewhere (Headwaters Economics, 2014) . Headwaters argues that prescribed fires should be used more extensively in

PAGE 71

54 mitigating wildfire (Gorte, 2013) . It is estimated that 230 million acres of nonWUI Forest Service and Department of Interior lands are in need of treatment because they are at risk from ecological damage from wildfire (Headwaters Economics, 2014) . However, less than three million acres are treated per year (Headwaters Economics, 2014) , which is insuffi cient to reduce wildfire risk. It is unlikely there will be a dramatically increased acreage receiving fuel treatments in the near future due to budgetary and political constraints. Fire Treatments Prescribed Fire Prescribed fire is an intentionally ignited fire allowed to burn in desired locations under certain conditions to modify fuels (Husari et al., 2006) . Prescribed fire has been a supplement or in some cases, a replacement to natural sources of ignitions. It is important to distinguish between restoration and maintenance burns. Restoration burns modify the current ecol ogical condition to a preferred state, while maintenance burns maintain ecological conditions within a specified range. Modifications may include the decrease of hazardous quantities of dead and downed fuel, the stimulation of fire dependent species, impro vement of range conditions, or the creation of wildlife habitat (Husari et al., 2006; Parks et al., 2015) . It is critical to consider the variables that influence a fires behaviour, the ecological role of fire, and the ability to control the fire – minimizing potential for escapes before a prescribed burn is initiated. Specific site considerations include: slope, aspect, topographic position, and role of fire in the project area (Husari et al., 2006; Reinhardt, Keane, Calkin, & Cohen, 2008) . Specific conditions conducive to the use of prescribed fire include: season, weather, fuel conditions, and the availability of qualified personnel. It is also critical to establish to establish mea surable objectives and means of monitoring them. The value of prescribed fire to land

PAGE 72

55 managers decreases significantly with the inability to quantify the purpose of the fire and its accomplishments because prescribed fire within an adaptive management cont ext is critical for each land manager. Wildland Fire Wildland fires, under the correct, conditions can serve the same functions as prescribed fires. While wildland fire has been recognized as beneficial, as previously state in Chapter 2’s discussion of t he USFS let burn policy, it is often only used under narrow conditions (Fire History Society; Manning, 2012) . The fires that are allowed to burn are often categorized as within historic or natural ranges of variability and are of minimal risk to human settlement and lives. To allow a wildland fire to burn fires need to meet additional planning approvals, and implementation requirements (Husari et al., 2006) . The primary issue of concern is air quality and minimizing its impact on public safety. Mechanical Treatments Mechanical treatments remove, rearrange, or modify biomass; however, the effectiveness of differing mechanical treatments reducing wildfire risk are contentious (Agee et al., 2000) . Equipment such as feller bunchers, skidders, and grapplers are used to thin the forest to various densities, thus removing live and dead woody fuel. The characteristics of this fuel are changed by crushing, chipping, shredding, or chopping it. The material is then either removed from the site or piled and burned under safer, localized conditions Mechanical methods can be more precise; however, they have two large drawbacks. Fi rst, the use of mechanical treatments removes organic material which reduces the amount of carbon and nutrients on site (S L Stephens et al., 2012) . Second, mechanical treatments often still need the follow application of fire to maintain fire adapted ecosystem health (Husari et al., 2006; S L Stephens et al., 2012) .

PAGE 73

56 Forest Thinning Thinning modifies the fuel structure of forests by reducing the quantity and density of vegetation. This vegetation modification serves to reduce the fuel load within wildland vegetation for potential wildfires as well as to defensible space around dwellings. Thinning can be effective at moderating crown fire behaviour and often has some economic return on investment; however, thinning is only effective when fine fuels are also reduced. Thinning activities can remove trees to create forests with specific stand densities, patterns, distributions, and species compositions. Treatments should specify target densities by tree diameter classes of trees. Wildfire modeling has shown in simulations that thinning impacts the wildfire risk via reduction of wildfire fuel loads (Cochrane et al., 2012; Paveglio et al., 2013; Stratton, 2004) . These efforts are effective because thinning often limits th e ability of surface fires to transition to crown fires by breaking up vertical and horizontal fuel continuity. Mastication Mastication is the mechanical chopping, chipping, grinding, crushing, and shredding of fuels to reduce fireline intensity and rate of fire spread (Scott L. Stephens & Moghaddas, 2005) . Mastication reduces potential fire behaviour by reducing the fuelbed depth, thus increasing the fuels packing ratio (Husari et al., 2006; Scott L. Stephens & Moghaddas, 2005) . Due to the mechanical equi pment used, mastication can be very precise; however, these techniques are only recommended during ecosystem restoration efforts because the presence of heavy equipment is more damaging and less beneficial than the use of fire in ongoing wildland ecosystem maintenance (Kane, Varner, Knapp, & Powers, 2010; Kreye, Kobziar, & Zipperer, 2013) .

PAGE 74

57 Grazing Livestock grazing has been effective at reducing surface fuels and modifying the rate of fire spread (Scott L. Stephens & Moghaddas, 2005) ; however, grazing is limited in its scale of application. Still, its use as a fuel break maintenance tool and other linear fuel reduction projects has proven significant. Specifically, the use of animals grazing of fine fuels can shorten the fire season and reduce fire potential. Additionally, private landowners who own grazing lands are strong advo cates for using fire to promote grazing land ecosystem health, which provides additional fuel modification benefits. Defensible Space Defensible space uses a combination of the above fuel treatment types to create a perimeter around buildings and structures with modified vegetation cover to reduce fuel for potential wildfires. It also provides firefighters a clear environment in which to maneuver to prot ect structures. Additionally, defensible space reduces the chance that an initial structure fire will spread into the surrounding area to create a wildfire (Gill & Stephens, 2009) . Defensible space is typically separated into prescribed zones; whose definitions vary by agency and state. The HFRA CWPP documentation do not specify defensible space zone , distance, or vegetation requirements. As a result, implementations of defensible space are various and inconsistent. However, the National Fire Protection Association (NFPA) is promoting a set of Firewise community defensible space standards, in which de fensible space is created in the home ignition zone – the area within 200 feet of the house (National Fire Protection Association, 2016a) . In fact, the term “home ignition zone” is now used as a replacement for “defensible space.” The home ignition zone is often subdivided into three zones. The following zone measurements are for flat ground only; zone distances should increase on steeper slopes, but

PAGE 75

58 specific guidelines for slopes are often not prov ided in local ordinances . It is worth reiterating that CWPPs typically only include two defensible space zones. Zone 1 is within 30 feet of a structure vegetation should be carefully spaced, low growing and free of resins, oils and waxes that burn easily. Zone 2 is from 30 to 100 feet of the structure, plants in this zone should be low growing, well irrigated and less flammable. Zone 3 is from 100 to 200 feet, vegetation in this zone should be thinned. Each zone has specific vegetation spacing and compositi on requirements (Table 3.2). It is important to note that any defensible space zone can be truncated at a private landowner’s property line. Whether homeowners engage in defensible space mitigation is tied into their concerns about privacy, desired natural aesthetics, wildlife and recreational values, physical and economic capacity, and the individual’s knowledge (Brzuszek & Walker, 2008; Winter et al., 2009) .

PAGE 76

59 Table 3. 2. NFPA Firewise Community Recommended Defensible Space Requirements. Defensible space zone Distance Vegetation, fuel thinning, and reduction requirements Zone 1 030’ Plants should be carefully spaced, low growing and free of resins, oils and waxes. Lawns should be mown regularly. Trees should not overhang structures, be pruned up six to ten feet off the ground, and conifers should be spaced 30 feet between tree crowns. Create a ‘firefree’ area within five feet of the home, using nonflammable landscaping materials and/or high moisture content annuals and perennials. Remove dead vegetation from under deck and within 10 feet of house. Water plants, trees and mulch regularly. Consider xeriscaping if you are affected by water use restrictions. Zone 2 30100’ Leave 30 feet between clusters of two to three trees, or 20 feet between individual trees, and prune trees up six to ten feet off the ground. Use a mixture of deciduous and coniferous trees. Create ‘fuel breaks’, like driveways, gravel walkways and lawns. Zone 3 100200’ Thin this area, reducing the density of tree canopy. Remove smaller conifers that are growing between taller trees. Remove heavy accumulation of woody debris. Note: Source (National Fire Protection Association, 2016a) . Regardless, of the mitigation technic used, local, state and federal fuel reduction efforts have failed because they are treating too few acres per year (Vaillant & Reinhardt, 2017) . In all regions of the U.S. annual acres disturbed by wildfire and treatment are much lower than historically burned (Parks et al., 2015; Vaillant & Reinhardt, 2017) . These efforts are costly, and both federal and state mitigation budgets have been consumed by fire suppression e fforts, with little political will to expand the budget to what is necessary for significant risk reduction efforts (A. Berry, 2007) . The levels of fuel reduction on federal lands has been insufficient, the federal

PAGE 77

60 government have only managed fuel reduction on 3 million of the 230 million acres that need fuel load reduction, this number does not address the millions of acres of private land that need mitigated (Gorte, 2013) . Budgets and political will are further eroded due to several out of control high profile prescribed burns lawsuits. This is concerning considering the use of prescribed fire, in combination with mechanical thinning techniques has proven very effective in reducing catastrophic wildfires (Agee et al., 2000; Forthofer, Butler, & Wagenbrenner, 2014; Kalies & Yocom Kent, 2016; North et al., 2012; S L Stephens et al., 2012; Wales, Suring, & Hemstrom, 2007) . Reducing structural ignitibility WUI fires ignite homes in two principal ways: direct flame heating or firebrand ignition. The principal approach to reducing structural ignitions is to lessen the ignitibility of the home ignition zone and increase survivability (Jack D Cohen, 2001) . The home ignition zone includes the home and an area surrounding the home within 100 to 200 feet (Jack D Cohen, 2001) . CWPPs should identify regulatory and nonregulatory strategies to reduce structural ignitibility (Jakes et al., 2007; Jakes et al., 2012; Jakes et al., 2011; Society of American Foresters, 2004) . These efforts should include actions communities and individuals can take. Local governments play a pivotal role in reducing structura l ignitibility because in the United States, local governments are responsible for land development regulations. However, some state governments have developed minimal regulations because in some situations local codes are nonexistent or insufficient. Regulatory structural ignitibility mechanisms include: zoning regulations, development standards, building codes, fire prevention codes, and fire response. Local government actions are discussed in the subsection below. Individual responsibility suggestions s hould include: fire safe construction practices, private property forest thinning, and

PAGE 78

61 defensible space. While these are individual efforts, they can also be regulatory, as such each of these will be described in great detail in the subsection below. Educ ation and outreach Education and outreach are crucial to broadening CWPP support and implementation. State and local agencies play a pivotal role in wildfire outreach and risk reduction education efforts. Each state has multiple agencies that manage land w here wildfire can occur; however, usually a single agency is responsible for supporting and guiding state wildfire efforts in the wildland and WUI. Education efforts emphasizes raising risk awareness and preventive measures. Risk awareness efforts refers t o fire weather watches and risk mapping. Education efforts refers to various forms of multimedia campaign efforts, workshops, field trips, reports, and how to guides. Fire weather watches, which are issued by the National Weather Service to alert all fire agencies of the onset, or possible onset of critical weather and dry conditions that could lead to rapid or dramatic increases in wildfire activity. The ratings are shown in Table 3.1 and broadcast to the public via television, radio, and public postings outside of fire stations and other state and federal agency offices. Numerous studies show that being aware of risk is a necessary condition of mitigation decision making; however, it needs to be partnered with additional regulations and incentives to reduce risk (Nielsen pincus et al., 2015; Scott L. Stephens & Collins, 2007; M. P. Thompson et al., 2012; Austin R. Troy, 2001) . Education, in combination with risk mapping, plays a pivotal role in alerting the public about the dangers of livi ng in wildfire prone areas, and the importance of mitigation efforts. While each state is a key facilitator in creating and disseminating education material, they increasingly partner with universities, local government, and other nongovernmental fire ris k reduction programs, such as Firewise . Firewise is a National Fire

PAGE 79

62 Protection Association outreach effort that intends to teach people how to adapt to living with wildfire and encourages neighbors to work together and act to prevent loss (N ational Fire Protection Association, 2016b) . Collectively each of these groups relay publications that are tailored to a geographic area to promote hazard reduction efforts, including fire protection and safety, landscaping and defensible space guidelin es, lists of recommended fire resistant plant species, and residential building guidelines. Published material is disseminated through websites, brochures, pamphlets, community events, mailings, news outlets, blogs, videos, and podcasts. Classroom and teacher efforts are also part of the education. In several states, a fire science is a part of the science curriculum in K 12 education, using multimedia to educate students on wildfire ecology, safety, and protection. Fire protection officials have developed their own educational programs. These programs include hands on defensible space and fire safety courses for grade school students. Those targeting high schools have involved fuel removal around schools and field exercises, such as risk assessments and com munity fire risk mapping. Emergency management capacity While the majority of CWPP content is geared towards pre fire planning efforts, suppression activities are still critical to preserving community values, property, and lives. As such, CWPPs should assess local preparedness for wildfire and firefighting cap abilities. These efforts should assess the community’s emergency preparedness, including evacuation planning (people and agricultural livestock) , safety zones, and fire assistance agreements as well as the response capability of community and cooperative f ire protection forces (Jakes et al., 2007; Jakes et al., 2012) . Documentation should include gaps in and training for incident command and the numbe r and percentage of homes in each fire district. Emergency management capacity should also document the number and per percent of trained and/or certified fire fighters

PAGE 80

63 and crews; fire suppression equipment; and response times. L ocal preparedness informati on should be incorporated into the base map as appropriate (Society of American Foresters, 2004) . Long term success CWPPs should not be static, as such it is critical to ensure their long term success b y sustaining community engagement and support. Jakes et al (2012) su ggest four strategies to ensure long term success: 1) incorporate projects into the CWPP that can be accomplished quickly to foster homeowner buyin and broaden longterm support; 2) nest local CWPPs within broader plans to augment resources, support, and implementation; 3) incorporate CWPP into formal government structure and processes; and 4) quickly identify changes that affect the CWPP and adapt accordingly. Additionally, CWPPs should suggest update timelines because new and aging developments and evolving fuel loads alter wildfire risk overtime. Assessing CWPPs is critical to maintaining their relevance and effectiveness over the long term, especially under evolving conditions (Jakes et al., 2012; Society of American Foresters, 2004) . CWPPs incorporated into local ordinances and codes gain efficiencies and relevance, thus ensuring longer term support and enforcement. Additionally, incorporating CWPPs will also better leverage community resources, achieve multiple adjectives, increase political acceptance of mitigation activities, and provide a consistent message (H. Brenkert smith, Meldrum, Champ, & Barth, 2017; Clark & Stankey, 2007; T. W. Collins, 2008b; A. M. S. Smith et al., 2016) . Mechanisms for incorporating CWPPs into local governance are discussed in the following section. Local government wildfire risk reduction efforts Local government wildfire risk reduction efforts, or options for potential reduction efforts, will be presented in this section. Local governments are critical partners in the CWPP

PAGE 81

64 process. Research has documented land developments contributions to elevated wildfire risk and shown that well integrated comprehensive and hazard planning can reduce said risk (Bhandary & Muller, 2009; Brzuszek & Walker, 2008; Butsic, Kelly, & Moritz, 2015; Headwaters Economics, 2014, 2016b; Syphard, Bar Massada, Butsic, & Keeley, 2013) . Additionally, zoning codes, development standards, subdivision design guidelines, building codes, and plan review and inspection procedures can create safer com munities (Greiving, Fleischhauer, & Lckenktter, 2006; Headwaters Economics, 2014, 2016b; Jakes et al., 2012; Kocher & Butsic, 2017; National Fire Protection Association, 2016a, 2016b; Paveglio et al., 2013; Reams et al., 2005; Scott L. Stephens & Collins, 2007; Syphard et al., 2013) . Since CWPPs are not regulatory documents or policies, it is imperative t o deeply integrate CWPPs and local government land development controls. This chapter describes local government best practices for planning and land development policy efforts to reduce wildfire risk. These efforts are described below and include the foll owing sections: 1. comprehensive plan, 2. zoning codes and development standards and 3. fire resistant materials. Comprehensive Plan Comprehensive plans are the result of a process that assists communities in planning pleasant, liveable, safe, and well o rdered urban environments (Mandelker, 1976) . Due to their anticipatory approach to future land development, comprehensive plans, have the c apacity to steer growth and development away from hazard prone areas, restrict land uses in sensitive areas, locate public infrastructures away from hazard areas, and impose building standards that reduce the vulnerability of structures (Schwab, Meck, & Simone, 2005; Srivastava & Laurian, 2006) . However, the framing of comprehensive plans is cri tical to whether hazards are considered a key driver to the planning effort. Indeed, research has shown that comprehensive

PAGE 82

65 plans have framed growth within the context of flood and droughts; however, they have not equally done so for wildfire (Schwab et al., 2005; Srivastava & Laurian, 2006) . Additionally, plans that have framed the process solely in the interest of economic development have often downplayed all hazard risks and environmental concerns (P. Berke & Godschalk, 2009; Samuel D Brody, 2003; Godschalk, Kaiser, & Berke, 1998; Srivastava & Laurian, 2006) . This is concerning becaus e hazards that are onset quickly and have localized impact, such as fire, can be most effectively and sustainably addressed through land use controls and polices that discourage or regulate development in highrisk areas (P. Berke & Godschalk, 2009; Samuel D Brody, 2003; Butsic et al., 2015; Deyle & Smith, 2000; Godschalk et al., 1998; Greiving et al., 2006; Headwaters Economics, 2014, 2016b; Paveglio et al., 2013; Srivastava & Laurian, 2006) . Additionally, comprehensive plans can also prepare for effective post disaster recovery and r econstruction to minimize future losses from repeated events (Srivastava & Laurian, 2006) . Finally, they can promote risk awareness, education, and community capacity building (National Fire Protection Association, 2016b; Schwab et al., 2005; Srivastava & Laurian, 2006) . As such, comprehensive plans should be sophisticated enough to represent multiple frames, such as economic growth, environmental concerns, and hazard risks, particularly wildfire. Comprehensive plans should include strategies to reduce the extent and severity of fires impact on communit ies by reducing structural ignitibility and facilitating responses to fires (Headwaters Economics, 2014; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006; Scott L. Stephens & Collins, 2007) . Broadly these strategies employed in comprehensive plans should include development controls for wildfire prevention, open s pace preservation, critical resource (e.g. water supply, hospitals, fire stations, and elder care facilities) identification and protection, and enforcement of zoning and building codes process

PAGE 83

66 development (Headwaters Economics, 2014, 2016b; National Fire P rotection Association, 2016b; Srivastava & Laurian, 2006) . Specific wildfire interventions that comprehensive plans should discuss fall into three categories: wildfire prevention, reducing wildfire impacts, and facilitating emergency response. Wildfire prevention includes: best forestry management practices, fores t fire fuel reduction efforts, vegetation management, urban forestry management, and wetlands protection (Headwaters Economics, 2014, 2016b; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006) . Reducing wildfire impacts includes development controls (e.g. open space preser vation, building codes, performance standards, density controls, design review guidelines, environmental review standards, hillside development standards, and subdivision guidelines), property protection, and public awareness (Headwaters Economics, 2014, 2016b; National Fire Protection Association, 2016b; Srivastava & Laurian, 2006; Scott L. Stephens & Collins, 2007) . Facilitating emergency responses should include hazard recognition, warning systems, emergency response services, and post disaster mitigation (Srivastava & Laurian, 2006) . Zoning Codes and Development Standards Zoning is used to regulate land uses in order to prevent incompatible adjacent land uses, undue densit y and traffic congestion, restrict height and size/bulk of buildings, provide setbacks to lessen fire hazard and promote aesthetic value (Metzenbaum, 1957; Whitnall, 1931) .. Zoning regulations can dictate development density, size of parcels, open space requirements, provision of adequate light and air, and efficient public infrastructure (e.g., water, sewer, schools, police, fire, and transportation) (Metzenbaum, 1957; Whitnall, 1931) . Additionally, local zoning ordinances have traditionally attempted to guide development away from hazardous areas, mini mize sprawl, or to minimize land use adjacency conflicts (Butsic et al., 2015; Headwaters

PAGE 84

67 Economics, 2016b; Kocher & Butsic, 2017; Syphard et al., 2013) . Zoning is administered through three components: 1. the zoning map, 2. zoning text, and 3. the comprehensive plan as discussed above. The zoning map is where the zoning becomes applicable to individual property owners. The map is often color code d to identify what types of land uses are allowable at what locations. To reduce wildfire risk, the zoning map should direct development away from highrisk wildfire areas (Headwaters Economics, 2016b; Hughes & Mercer, 2009; Syphard et al., 2013) . Overlay zoning provides a set of standards that apply to properties within a specific geographic area, superseding the underlying base standards of a given zoning district. This is an instrumental tool in avoiding potential conflicts between 1. resource protection and 2. forest thinning or defensible space requirements. For example, wildfire intensity is dictated by fuel loads and topography, so a zoning overlay could disallow developments on st eep, fuel loaded slopes. The zoning text describes the exact regulations being implemented within a given land use classification (Metzenbaum, 1957; Whitnall, 1931) . This document is adopted as law by a local governing body, in the case of this research it is implemented through county governments. At a minimum, the text establishes the different zone classifications within the county and the uses allowable within each zone either by right or with a conditional use permit (Metzenbaum, 1957; Whitnall, 1931) . Zoning text also defines key definitions, vari ous requirements for building setbacks, open space, landscaping, parking, height restrictions, and procedures for zoning processes. The zoning text should coordinate with the CWPP by providing defensible space requirements, use of open space as community f uel breaks, and provide documentation of safe development patterns and densities within the applicable zoning classifications (Headwaters Economics, 2014, 2016b; Hughes & Mercer, 2009) . Appropriate zoning in the WUI can also

PAGE 85

68 ensure each home has the ability to adequately implement defensible space zones (Headwaters Economics, 2014, 2016b; Hughes & Mercer, 2009) . Development standards are land use regulations that determine the quality of development. For wildfire, development standards include adequate water supply, defensible space, resource protection, and ongoing maintenance (Colburn, 2007; Headwaters Economics, 2016b) . Subdivision regulations determine how lots are created and divided, including layout standards for new subdivisi on developments. For wildfire these include adequate water access, water supply, and mitigation requirements (Colburn, 2007; Headwaters Economics, 2016b) . Some WUI communit ies are also implementing WUI specific codes that enforce required codes for buildings, landscapes, and lot development. Fire Resistant Materials Very few comprehensive laws, statutes, or building codes exist that address combustionresistant building ma terials (Scott L. Stephens & Collins, 2007) even though using combustionresistan t building materials is key in the survival of structures during wildfires. Possibly due to enhanced public perception and increasingly stringent building codes , the construction industry and homeowners are adapting fire resistant materials (T. W. Collins, 2008b) . Fire resistant building materials and practices include fire rated roofing, siding, and decking; closed eaves and soffits; and protecting vents and windows (Brzuszek & Walker, 2008; Meldrum et al., 2015; Reams et al., 2005) . An effective wildfire mitigation response should include all of these elements within local building codes. However, while fire resistant material is bei ng required in new construction, these efforts do not address the thousands of existing nonfire resistant homes. Currently, h omes without appropriate fire resistant construction materials are not required to be brought up to code. Accurately accounting for how many homes that are not f ire resistant is

PAGE 86

69 difficult because collecting the data requires in the field assessments and to date cannot be done remotely. Fire resistant construction does not justify continued development in the WUI and does not address the subsequent suppression cost s (Headwaters Economics, 2014) . A number of tools exist for local governments to integrate CWPPs in order to reduce wildfire risk. Comprehensive plans, zoning codes and development s tandards are all all tools that would be utilized if best practices were followed in wildfire reduction efforts. In the following chapters, results of analyzing CWPPs and local governance structures will be discussed, questioning to what extent best practi ces are being followed across the American West.

PAGE 87

70 CHAPTER IV RESEARCH DESIGN AND METHODOLOGY The research design and methodology that are used in this research effort is described throughout this chapter. Sections in the chapter include a short description of the ethical implications, pilot study results, a description of the study setting, and the research design. The research design is split into the following subsections: dependent and independent variables, study sample, and sample descripti ve statistics. This chapter concludes with a brief discussion of the study’s data analysis methodology. Research Permissions and Ethical Considerations Ethical issues were considered and addressed at each phase of the study. This research study falls unde r the classification of nonhuman subject research because this research uses secondary, publicly available data and records: CWPP reports, county government ordinances and policies, census, satellite remote sensed data, and open access GIS data. Pilot Study Results The initial impetus of this research was to study municipal or neighborhood level CWPPs within CO. Before embarking on the full study, a pilot study was conducted for Boulder County, CO to document the process, documents needed, test both document coding instruments, and evaluate error and uncertainty. While the process and intercoder reliability achieved acceptable intercoder reliability, the uncertainty and error introduced by the mismatch between CWPP geographic boundaries and census block groups proved unacceptable. The full pilot study write up can be found in Appendix C . As such, the scope of the project evolved to focus on County level CWPPs across the American West. This proved beneficial because counties are better match to the CWPP fra me of interest and the landscape scale risk reduction (Jakes et al., 2007;

PAGE 88

71 Jakes et al., 2012) while also potentially providing a greater diversity in instrument – the pilot study results suggested regional homogeneity in constructing subcounty level CWPP documents. Study Area The National Association of Foresters (NASF), estimates that 76,934 com munities in the United States are at risk of wildfire with more than 17,655 communities covered by CWPPs (National Association of State Foresters, 2017) . The NASF western region is estimates that 7,587 communities are at risk of wildfire with 6,357 communities covered by CWPPs (National Association of State Foresters, 2017) . This study focuses on wildfir e mitigation in the American West and utilizes a representative sample to understand the socio economic, demographic, and biophysical relationships to CWPP objective implementation. The American West is defined as containing the following states (number of counties for each state are located in parentheses): Arizona (15), California (58) , Colorado (64), Idaho ( 44) , Montana (56), Nevada (17), New Mexico (33), Oregon (36), Utah (29), Washington (39), and Wyoming (23) for a total of 414 counties (Figure 4.1).

PAGE 89

72 Figure 4. 1. Study states, defined as the American West and associated county boundaries . The American West was chosen because it has experienced drastically changing wildfire conditions due to climate change, fuel buildup, and residential development. Indeed, large fire activity has increased dramatically since the 1970s. Since the 1970s, the frequency of large fires has increased most dramatically in the forests of the Northwest, Northern Rocky Mountains, Southwes t, Southern Rockies, and Sierra Ne vada ( Figure 4.2) (Schoennagel et al., 2017; Westerling Anthony, 2016) . Specifically, climate change has caused earlier snow melt, rising

PAGE 90

73 temperatures, and increased drought across the American West, causing an average increase of the fire season by 2.8 months since 1970 (Westerling Anthony, 2016) . These changing conditions in addition to a large estimate of communities with CWPPs provide a robust study area for evaluating CWPP s . Figure 4. 2. Percent increase in large fire activity in the American West. Research Design Independent and Dependent Variables Dependent Variable Data The dependent variables consist of two indices “CWPP Process and Plan Evaluation Instrument” and “CWPP Implementation: Local Governance Evaluation Instrument ,” and a composite of the two. The instruments used to code CWPPs were created through an extensive literature review, as discussed in Chapter III . The instruments are located in Appendix A and B . 1000% 889% 462% 274% 256%Northwest Northern Rocky Mountains Southwest Southern Rocky Mountains Sierra NevadaPercent IncreaseAmerican West Forest Geographies

PAGE 91

74 The i nstrument scores are ordinal , which are converted into letter grade scores ranging between 0 and 100 percent (see Table 5.6 for letter grade conversion details). Indexing CWPP processes required the following document data: CWPP reports and CWPP process me eting minutes. County CWPP reports and process meeting minutes were downloaded from the internet using a Google search on a sampled county by county basis. Appendix D contains a CWPP and meeting minutes bibliography. Indexing CWPP implementation is twofold: policy and physical environment interventions. The policy interventions required document data in the form of: local building and zoning codes, HOA guidelines, subdivision design guidelines, design review processes, and comprehensive plans. Again, each s ampled county was searched using Google search for all relevant documents and downloaded. Appendix E contains a bibliography for local governance CWPP intervention documents. The physical intervention data was evaluated by change in WUI area and the change in average WUI population density between 2000 and 2010 per each county. This dataset was obtained from Radeloff et. al.’s, (2017) dataset. This dataset was chosen because it is the only dataset that provides uniform and comprehensive coverage of the American West. The processing of this dataset is described in the sampling section of this chapter. However, additional processing was required to calculate the change in median population density within areas designated WUI. This included spatially joini ng the WUI dataset to the county sample dataset and summarizing each type of WUI (WUI intermix and interface) for 2000 and 2010. Once these values were calculated, the delta change for each were calculated. Independent Variable Data The independent varia ble data consists of US Census data attributes at the county level for the 2000 (pre HFRA) and 2010 census. These variables include age, length of homeowner

PAGE 92

75 tenure, full time/parttime residency, and income. The first task of aggregating the independent va riables was to obtain 2000 and 2010 census attribute tables for each variable from the IPUMS National Historical Geographic Information System initiative (NHGIS). Metadata for attribute tables and map layers are given in Appendix F . Second, each variable’s attribute table was joined to the sampled county boundary file, using each county’s census unique identifier. Nonsampled county attribute information was removed, by using the keep only matching records join option. This data was exported to create a new permanent joined dataset. Due to the large number of fields within the attribute table, all extraneous, non variables fields were deleted. New fields were added, and the delta change values of each variable – age, length of homeowner tenure, full time/par t time residency, and income – min, median, and max were calculated. The delta change values for each county were used for statistical data analysis. Study Methodology and Sample Sampling Strategy To understand the integration of CWPP best practices as well as the socio demographic and biophysical relationships in the CWPP process and implementation, t his study select ed a representative sample of 1) economic status of the American West population and 2) CWPP coverage. Obtaining a repre sentative sample was difficult given the size and heterogeneity of the American West . To accomplish this goal of representation , a cluster sampling methodology was utilized involving 1) filtering of counties with WUI and CWPPs, 2) a stratification of count ies based on 2010 median household income levels, and 3) a simple random sample of cases from each stratum (Figure 4.3). The sampling strategy was designed in this way to reduce selection bias as well as standard errors of the population.

PAGE 93

76 Figure 4. 3. Simple, stratified random sampling framework. Input of Data Needed for Sampling The first task of the sampling phase was to input map layers that would be needed for obtaining a representative sample. Metadata for map layers is given in Appendix F . First, state boundary files were obtained, from the 2010 US Census TIGER data set compiled by the IPUMS National Historical Geographic Information System initiative (NHGIS). The next set of data obtained was the 2010 US Census US county boundaries. To obtain only the American West counties required a two step process. First, the American West states were selected, followed by a spatial selection of all counties that are within American West states and a new file was created. The next set of data obtained were the count ies with CWPPs. There is no national list of CWPPs. This data set was created by searching statewide CWPP lists and searching Google for specific county CWPPs. The presence or absence of a CWPP was coded as 1 or 0, respectively. WUI classifications were obtained from Radeloff et. al.'s, (2017) . The WUI data was clipped to the American West states and areas were recalculated. Because of the prohibitively

PAGE 94

77 large size of this file – it mapped list everything it mapped here – a new file was created with only the WUI classification polygons and relevant fields in the attribute table. The amount of WUI area, interface and intermix, for 2000 and 2010 were calculated and aggregated to each county, including the delta change in WUI. However, it was decided that the mere presence of any kind of WUI, interface or intermix, was sufficient to be considered WUI, as either designation could be integrated into a CWPP. Next the needed county level income information was acquired. County median income data was gathered for the 2000 and 2010 census using the IPUMS NHGIS database (Manson, Sch roeder, Riper, & Ruggles, 2018) . These attribute tables were joined to the existing American West county database. It was determined to use the 2010 census income for sample stratification because it was the most current. However, to control for differing cost of living per state, a new field was added to normalize median county incomes by z value per median state incomes, using the following formula ( county median income/state median income ) (Figure 4.4 and 4.5). Normalizing the median income values avo ided clusters of high and low incomes contained within particular states as well as accounted for state variability in higher costs of living (Figure 4.6).

PAGE 95

78 Figure 4. 4. County 2010 median income stratification without norm alization (natural breaks).

PAGE 96

79 Figure 4. 5. County 2010 median income stratification without normalization (quantiles).

PAGE 97

80 Figure 4. 6. County 2010 median income z score quantile stratification. Sampling Strata Design and Procedure Categorization intervals were then chosen for the income attributes by which counties would be stratified. First, a sampling population was defined as the subset of counties where data existed, specifi cally, counties that contained WUI lands and CWPPs. This proved to be 326 counties, or about 79% of American West counties3 (Figure 4.7). For these 326 counties, the 3 While some counties have no CWPP at all, other counties, particularly counties in Utah and Arizona, have multi county or local CWPPs. However, as these were not within the scale of concern, county, they were filtered as having no county CWPP (Figure 4.7).

PAGE 98

81 intervals were determined that broke them into roughly three strata using quantiles. With the income of counties stratified, a simple random sample was used to select 40 counties from each stratum for a total project N of 1204. Figure 4. 7. Final county sample population (counties that contain WUI and a county level CWPP). 4 Power analysis fo r an ordinal multiple regression was conducted to determine a sufficient sample size using an alpha of 0.05, a power of 0.80, and a large effect size (f=0.35), based on the aforementioned assumptions, the desired sample size is 40 per strata (Faul, Erdfelder, Buchner, & Lang, 2009) .

PAGE 99

82 Remedial Measures for Sampling Once these 120 samples were chosen, it was necessary to see if some of these counties did not adequately fit the criteria needed for the study. For example, it was hoped to avoid counties that were ove rly clustered geographically or an over representation of a state. In the few cases where counties were overly clustered, one or more counties were randomly discarded and replaced with another county from the same strata. This was done until there were no significant spatial clusters. Finally, a list of counties was ready for which additional census variables, CWPPs, and local governance documents could be downloaded (Figure 4.8). Figure 4. 8. Final county sample data set stratified per state normalized zscores.

PAGE 100

83 Key Sample Descriptive Statistics and Demographics The final sample consists of three CWPPs from Arizona, fifteen from California, eighteen from Colorado, nineteen from Idaho, twenty from Montana, five from Neva da, eight from New Mexico, thirteen from Oregon, eleven from Washington, and eight from Wyoming. Table 4.1 documents the number of CWPPs from each state in their corresponding income strata, following income z score normalization. Table 4. 1. Number of sampled CWPPs per strata per American West state. State Number of CWPPs sampled Strata 1 Strata 2 Strata 3 Arizona 3 0 2 1 California 15 5 5 5 Colorado 18 4 7 7 Idaho 19 4 9 6 Montana 20 11 4 5 Nevada 5 0 3 2 New Mexico 8 2 4 2 Oregon 13 8 0 7 Washington 11 4 6 1 Wyoming 8 4 0 4 Total sample 120 40 40 40 Collectively, when the sampled counties were aggregated by state, the WUI progressively grew in each state between 2000 and 2010 by a cumulative total of 4,216 Km2 (Figure 4.9). However, when disaggregated 109 counties experienced positive WUI growth, two counties experienced no significant WUI growth, and 9 counties experienced negative WUI growth. Strata three experienced the most growth ( 689 Km2), followed by strata two ( 235 Km2), and strata three ( 47 Km2). The minimum WUI growth for strata one was a decrease of 1.3 Km2 while the maximum WUI change was an increase of 11.2 Km2 with an average of 1.2 Km2 (Table 4.2 ). The minimum WUI growth for strata two was a decrease of 5.1 Km2 while the maximum WUI change was an increase of 46.6 Km2 with an average of 6.2 Km2 (Table 4.3). The minimum WUI

PAGE 101

84 growth for strata three was a decrease of 1.2 Km2 while the maximum WUI change was an inc rease of 172.5 Km2 with an average of 19.2 Km2 (Table 4.4). Figure 4. 9. Change in WUI area (Km2) per sampled county from 2000 2010, summarized per state. Table 4. 2. Sample strata one WUI area change (Km2) from 20002010 descriptive statistics, summarized per state State Minimum Maximum Average Arizona California 0.04 11.19 2.93 Colorado 1.13 3.50 0.43 Idaho 1.07 4.38 1.27 Montana 1.22 0.34 0.14 Nevada New Mexico 0.06 2.19 1.13 Oregon 0.27 3.08 0.89 Washington 0.00 5.90 1.59 Wyoming 1.30 7.32 3.17 Total Sample 1.30 11.19 1.16 500 1,500 2,500 3,500 4,500 5,500 6,500 7,500 8,500 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Square Kilometers of WUIState 2000 2010

PAGE 102

85 Table 4. 3. Sample strata two WUI area change (Km2) from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 10.16 16.79 13.48 California 0.43 23.75 8.09 Colorado 0.40 10.44 3.16 Idaho 0.27 16.99 5.63 Montana 0.05 13.70 4.75 Nevada 5.10 5.57 0.98 New Mexico 5.09 8.66 1.28 Oregon Washington 0.24 46.59 11.41 Wyoming Total Sample 5.10 46.59 5.89 Table 4. 4. Sample strata three WUI area change (Km2) from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 13.41 13.41 13.41 California 8.27 172.51 56.64 Colorado 1.23 52.14 16.74 Idaho 0.58 13.33 7.90 Montana 0.76 34.16 10.30 Nevada 4.83 43.33 24.08 New Mexico 0.21 0.87 0.54 Oregon 0.12 19.06 9.07 Washington 27.88 27.88 27.88 Wyoming 0.63 20.89 9.03 Total Sample 1.23 172.51 17.24 This WUI expansion exposed an additional 1,927,384 people, 897,171 housing units, and 79,981 seasonal homes to wildfire. Figure 4. 10 documents the change in population for the sampled counties within each state from 2000 to 2010. The minimum population change for strata one was a loss of 607 people while the maximum was an increase of 1,781 people with an average gain of 375 (Table 4.5). The minimum population change for strata two was a loss of 1,932 people while the maximum was an increase of 32,725 people with an average increase of 3,543 people (Table 4.6 ). The minimum population change for strata three was a decrease of 406

PAGE 103

86 people while the maximum was an increase of 644,254 people with an average of 44,266 (Table 4.7). Figure 4. 10 . WUI population change per sampled county from 20002010, summarized per state. Table 4. 5. Sample strata one population change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona California 237 1,669 875 Colorado 139 680 316 Idaho 26 500 233 Montana 348 181 70 Nevada New Mexico 218 375 79 Oregon 607 1,227 191 Washington 131 1,692 804 Wyoming 207 1,781 1,170 Total Sample 607 1,781 375 0 1 2 3 4 5 6 7 8 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Total Population MillionsState 2000 2010

PAGE 104

87 Table 4. 6. Sample strata two population change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 13,591 35,725 24,658 California 599 6,731 2,959 Colorado 1,932 3,819 805 Idaho 1,006 4,171 837 Montana 518 922 243 Nevada 895 3,527 1,867 New Mexico 733 1,618 440 Oregon Washington 1,072 28,816 9,353 Wyoming Total Sample 1,932 35,725 3,543 Table 4. 7. Sample strata three population change fro m 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 45,154 45,154 45,154 California 13,295 644,254 222,679 Colorado 679 105,334 29,052 Idaho 2,385 21,712 9,196 Montana 406 13,497 3,226 Nevada 2,817 81,921 42,369 New Mexico 1,499 2,958 2,229 Oregon 2,605 84,378 29,161 Washington 20,650 20,650 20,650 Wyoming 2,419 12,435 5,865 Total Sample 406 644,254 44,266 All sampled counties, when aggregated by state, exhibited WUI housing unit growth (Figure 4.11 ). However, this was not equally distributed across counties or strata. The minimum change in housing units for strata one was a decrease of 167 units while the m aximum was an increase of 3,561 units with an average gain of 453 units (Table 4.8). The minimum housing unit change for strata two was a decrease of 175 units while the maximum was an increase of 13,710 units with an average increase of 2,574 units (Table 4.9). The minimum housing unit change for

PAGE 105

88 strata three was an increase of 541 units while the maximum was an increase of 216,034 units with an average of 19,402 units (Table 4.10). Figure 4. 11 . Amount of WUI housing unit change per sampled county from 20002010, summarized per state. Table 4. 8. Sample strata one housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona California 385 796 614 Colorado 124 454 254 Idaho 100 654 276 Montana 167 337 35 Nevada New Mexico 37 80 59 Oregon 17 2,453 737 Washington 55 3,561 1,328 Wyoming 274 1,050 680 Total Sample 167 3,561 453 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,000 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Total Housing UnitsState 2000 2010

PAGE 106

89 Table 4. 9. Sample strata two housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 7,915 13,710 10,813 California 1,963 8,308 4,309 Colorado 35 5,086 1,846 Idaho 4 3,705 990 Montana 175 2,269 712 Nevada 552 1,110 919 New Mexico 251 2,221 1,147 Oregon Washington 761 8,339 4,625 Wyoming Total Sample 175 13,710 2,574 Table 4. 10. Sample strata three housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 30,849 30,849 30,849 California 7,340 216,034 83,151 Colorado 2,097 50,425 17,537 Idaho 1,838 9,247 4,325 Montana 541 8,787 3,108 Nevada 2,251 40,933 21,592 New Mexico 1,720 2,861 2,291 Oregon 3,891 33,541 14,119 Washington 6,300 6,300 6,300 Wyoming 1,693 5,667 3,085 Total Sample 541 216,034 19,402 All sampled counties, when aggregated by state, exhibited WUI seasonal housing unit growth (Figure 4.12). However, this was also not equally distributed across counties or strata. The minimum change in seasonal housing units for strata one was a decrease o f 47 units while the maximum was an increase of 1,972 units with an average gain of 183 units (Table 4.11). The minimum seasonal housing unit change for strata two was a decrease of 121 units while the maximum was an increase of 4,497 units with an average increase of 727 units (Table 4.12). The

PAGE 107

90 minimum housing unit change for strata three was a n increase of 19 units while the maximum was an increase of 12,297 units with an average of 1,090 units (Table 4.13 ). Figure 4. 12 . WUI seasonal housing unit change per sampled county 2000 2010, summarized per state. Table 4. 11. Sample strata one seasonal housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona California 68 308 171 Colorado 10 173 95 Idaho 8 314 143 Montana 28 141 26 Nevada New Mexico 3 37 17 Oregon 17 1,612 394 Washington 47 1,972 666 Wyoming 39 70 38 Total Sample 47 1,972 183 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Total Seasonal Housing UnitsState 2000 2010

PAGE 108

91 Table 4. 12. Sample strata two seasonal housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 121 4,497 2,188 California 742 2,418 1,366 Colorado 12 2,959 847 Idaho 4 2,072 396 Montana 7 1,755 497 Nevada 94 104 11 New Mexico 93 300 61 Oregon Washington 12 2,069 1,020 Wyoming Total Sample 121 4,497 727 Table 4. 13. Sample strata three seasonal housing unit change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona 6,338 6,338 6,338 California 174 12,297 4,114 Colorado 196 1,895 658 Idaho 20 1,043 273 Montana 69 1,273 685 Nevada 49 1,401 725 New Mexico 196 828 512 Oregon 93 1,055 544 Washington 19 19 19 Wyoming 78 350 178 Total Sample 19 12,297 1,090 All sampled counties, when aggregated by state , exhibited significant median household income increases (Figure 4.1 3), and this was not equally distributed across counties or strata. The minimum change in median household income for strata one was an increase of $ 5,778 while the maximum was an increase of $ 23,163 with an average gain of $11,940 (Table 4.14). The

PAGE 109

92 minimum change in median household income for strata two was an increase of $2,122 while the maximum was an increase of $ 22,028 with an average increase of $ 10,384 (Table 4.15 ). The minimum change in median household income for strata three was an increase of $ 5,686 while the maximum was an increase of $ 27,554 with an average of $10,985 (Table 4.16). Figure 4. 13 . WUI median household income per sampled county from 20002010, summarized per state. Table 4. 14. Sample strata one median household income change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona California $8,858 $10,534 $9,866 Colorado $5,778 $20,682 $11,384 Idaho $6,348 $9,724 $8,232 Montana $9,424 $19,788 $14,101 Nevada New Mexico $7,113 $12,427 $9,770 Oregon $7,600 $9,981 $8,605 Washington $5,905 $17,927 $10,840 Wyoming $15,718 $23,163 $20,038 Total Sample $5,778 $23,163 $11,940 $0 $10,000 $20,000 $30,000 $40,000 $50,000 $60,000 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Median Household IncomeState 2000 2010

PAGE 110

93 Table 4. 15. Sa mple strata two median household income change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona $8,974 $13,400 $11,187 California $7,725 $13,664 $10,154 Colorado $2,122 $13,175 $8,855 Idaho $5,287 $15,026 $9,441 Montana $8,759 $18,367 $13,527 Nevada $7,314 $22,028 $14,357 New Mexico $3,975 $10,898 $8,259 Oregon Washington $8,781 $13,821 $10,844 Wyoming Total Sample $2,122 $22,028 $10,384 Table 4. 16. Sample strata three median household income change from 20002010 descriptive statistics, summarized per state. State Minimum Maximum Average Arizona $7,862 $7,862 $7,862 California $9,820 $16,306 $13,337 Colorado $6,743 $17,248 $11,258 Idaho $8,133 $15,243 $10,343 Montana $6,480 $10,199 $8,600 Nevada $8,179 $12,178 $10,179 New Mexico $7,073 $8,193 $7,633 Oregon $5,686 $12,253 $8,944 Washington $9,428 $9,428 $9,428 Wyoming $14,092 $27,554 $18,332 Total Sample $5,686 $27,554 $10,985 All sampled counties, when aggregated by state , exhibited an increase in median age (Figure 4.14 ), and like the above, this was not equally distributed across counties or strata. The minimum change in median age for strata one was a n increase of one year while the maximum was an increase of seven years with an average gain of four years (Table 4.1 7). The minimum change in median age for strata two was a decrease of one year while the maximum was an increase of seven years with an average increase of four years (Table 4. 18 ). The minimum

PAGE 111

94 change in median age for strata three was a decrease of one year while the maximum was an increase of seven years with an average increase of two years (Table 4. 19 ). Figure 4. 14 . WUI median age per sampled county from 20002010, summarized per state. Table 4. 17. Sample strata one median age change from 2000 2010 descriptive statistics , summarized per State. State Minimum Maxim um Average Arizona California 2 5 3 Colorado 1 7 3 Idaho 2 7 5 Montana 2 7 5 Nevada New Mexico 1 4 3 Oregon 4 6 5 Washington 5 6 6 Wyoming 2 3 2 Total Sample 1 7 4 0 10 20 30 40 50 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Median AgeState 2000 2010

PAGE 112

95 Table 4. 18. Sample strata two median age change from 2000 2010 descriptive statistics , summarized per state. State Minimum Maximum Average Arizona 0 3 1 California 3 5 4 Colorado 0 7 4 Idaho 1 7 4 Montana 1 6 4 Nevada 1 4 3 New Mexico 4 6 5 Oregon Washington 0 5 3 Wyoming Total Sample 1 7 4 Table 4. 19. Sample strata three median age change from 20002010 descriptive statistics , summarized per state. State Minimum Maximum Average Arizona 5 5 5 California 0 2 1 Colorado 1 5 3 Idaho 1 3 1 Montana 1 7 4 Nevada 1 3 2 New Mexico 3 6 4 Oregon 1 5 3 Washington 1 1 1 Wyoming 0 4 1 Total Sample 1 7 2 All sampled counties, when aggregated by state , exhibited significant length of homeowner tenure (Figure 4.15), which was not equally distributed across counties or strata. The minimum length of homeowner tenure for strata one was six years while the maximum was 19 years with an average of 11 years (Table 4. 20). The minimum length of homeowner tenure for strata two was five years while the maximum was 14 years with an average nine years (Table 4.21). The minimum length of homeowner tenure for strata three was five years while the maximum was 12 years with an average of eight years (Table 4.22).

PAGE 113

96 Figure 4. 15 . WUI average length of homeowner tenure per sampled county from 2000 2010, summarized per state. Table 4. 20. Sample strata one length of homeowner tenure descriptive statistics summarized per state. State Minimum Maximum Average Arizona California 8 10 9 Colorado 8 15 11 Idaho 6 10 9 Montana 8 19 14 Nevada New Mexico 12 15 14 Oregon 8 15 10 Washington 9 15 11 Wyoming 8 10 9 Total Sample 6 19 11 0 2 4 6 8 10 12 14 Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Length of Home Owner TenureState 2000 2010

PAGE 114

97 Table 4. 21. Sample strata two length of homeowner tenure descriptive statistics summarized per state. State Minimum Maximum Average Arizona 7 8 8 California 9 9 9 Colorado 6 9 8 Idaho 5 13 10 Montana 9 14 11 Nevada 7 8 7 New Mexico 9 14 12 Oregon Washington 6 10 8 Wyoming Total Sample 5 14 9 Table 4. 22. Sample strata three length of homeowner tenure descriptive statistics summarized per state. State Minimum Maximum Average Arizona 7 7 7 California 7 8 7 Colorado 6 9 7 Idaho 7 10 8 Montana 6 10 9 Nevada 6 7 7 New Mexico 8 12 10 Oregon 6 9 7 Washington 8 8 8 Wyoming 5 9 7 Total Sample 5 12 8 Data Analysis Four types of data analysis were used for this study. The first data analysis was the creation of indices for both the CWPP process and implementation using the instruments that were introduced earlier and can be found in Appendix A and B . These instruments were used to create the CWPP process index and the CWPP implementation index, and this process is described in greater detail in the following two subsections. Second, descriptive statistics and delta changes for each independent variab le – age, length of homeowner tenure, full time/part-

PAGE 115

98 time residency, and income – were provided at each time period. The observational unit for the independent variables is the geographic extent of each CWPP. For each independent variable the min, max, and average were calculated for 2000 and 2010. The delta – or change – for the min, max, and average of each independent variable were also calculated for each CWPP extent. Third, relationships between the independent and dependent variables were determined using Pearson product moment correlation coefficients. Fourth, in order to identify associations between the independent variables and the dependent variable – individually and as a composite index of overall CWPP integration – inferential statistics consisting of ordinal logistic regression were used because: 1) the dependent variable is ordinal; 2) there is one dependent and several independent variables; and 3) the independent variables are interval, continuous, or ratio. All assumptions were tested and met before statistical analysis. GIS data tables were exported and imported into SPSS. SPSS was used for both the descriptive and inferential statistics. Figures 4.16, 4.17, and 4.18 outline the statistical models that were performed for the data analysis. Figure 4. 16 . Model one – CWPP process and content index

PAGE 116

99 Figure 4. 17 . Model two – CWPP implementation index. Figure 4. 18 . Model three – CWPP process and content + implementation index.

PAGE 117

100 CWPP Process Index To systematicall y and quantitatively evaluate CWPP processes, I employed a protocol designed specifically for evaluation of wildfire in the WUI based on the literature presented in Chapters II and III. The evaluation protocol was buil t on the best management practice literature published by Society of American Foresters (2004) , Jakes et al (2007) , Resource Innovations Institute for a Sustainable Environment University of Oregon (2008) , Rodman and Stram (2008) , and Jakes et al (2012) . This study was also buil t on protocols previously developed for the evaluation of flood, hurricanes, tsunamis, and earthquake mitigation in comprehensive plans and planning processes (Anderssonsko, 2016; P. Berke et al., 2015; Samuel D Brody, 2003; Frazier, Walker, Kumari, & Thompson, 2013; Horney et al., 2017; Johansen, Horney, & Tien, 2017; Lyles, Berke, & Smith, 2014; Stevens & Shoubridge, 2015) . This protocol was based o n the assumption that plans and meeting minutes identify local hazards, specify hazard mitigation goals and objectives, and promote the use of best CWPP process, reporting, and implementation practices (Anderssonsko, 2016; Lyles et al., 2014; Srivastava & Laurian, 2006) . Additionally, plans should use the correct frame, be internally consistent, demonstrate adequacy of content, procedural valid, adequacy of scope, guidance for implementation, technically and scientifically sound, and exhibit quality communication (P. Berke & Godschalk, 2009; Jakes et al., 2007; Sabatier & Mazmanian, 1980) . The evaluation protocol relied on 56 indicators, organized into 11 themes: context, goals and objectives, community capacity, partnerships and collaboration, base map, risk assessment, hazardous fuels reduction, reducing structural ignitability, education and outreach, emergency management capacity, and longterm success (see Appendix A ). The context of the planning process was evaluated through five indicators. Plan goals and objectives were evaluated through

PAGE 118

101 eight indicators: vulnerability reduction goals, environmental quality goals, and goals related to the protectio n of public interest. Community capacity was evaluated through two indicators. Public participation was evaluated through the theme of partnerships and collaboration; it included 11 indicators. The factual basis of plan was evaluated through two themes: ba se map and risk assessment, containing one and nine indicators, respectively. Wildfire hazard mitigation strategies were evaluated through three themes: hazardous fuels reduction, reducing structural ignitability, and education and outreach, containing one , five, and four indicators, respectively. Emergency management capacity contained six indicators and long term success contains four indicators. The full evaluation protocol instrument can be found in Appendix A . Plans were coded using the protocol to as sess the presence or absence of each of the 56 indicators. Each indicator was coded as “1” if each indicator definition was present in the plan, and ” if the indicator were absent. Meeting minutes5, when available, were utilized to understand the context in which the CWPPs were developed to further understand the perceptions of the quality of plans, necessary gaps and improvements, and potential barriers to CWPP implementation. Because there are 56 indicators spread unevenly across 11 themes, each indicator was normalized, using z scores, in order to proportionally weight them. A sensitivity analysis was used to determine if a theme, or indicator within a theme, impacted the final results. Future models can use the sensitivity analysis to place more importance on those indicators and themes that are statistically more important than others. CWPP Implementation Index Successful CWPP implementation requires both policy and biophysical interventions, thus the CWPP implementation index created is a composite score of policy and observable 5 Meeting minutes themselves will not be scored according to the instruments.

PAGE 119

102 interventions. Policy interventions included wildfire risk reduction best management practices integrated into building and zoning codes; comprehensive plans; subdivision ordinances, and HOA guidelines. To systematicall y and quantitatively evaluate CWPP implementation into local governance structure, a second protocol instrument was used to code the policy interventions. That instrument contained the following four categor ies: comprehensive plan; zoning codes, development standards, subdivision design guidelines, and HOA ordinances ; building codes; and plan review and inspection procedures. There was a total of 17 total indicators. For further description of the analysis pr ocess, see above. The evaluation protocol instrument can be found in Appendix B . Biophysical changes are the quantified aggregate of WUI expansion, densification, and healthy WUI forests and development. WUI boundaries were calculated using Radeloff et al .’s (2005) WUI work, as outlined in “ H uman Presence and Development ” subsection of C hapter III. WUI boundaries in the sampled counties were calculated for the years 2000 and 2010. Percent change in the WUI’s geographic extent were calculated for each sampled CWPP’s county, geographic boundary using the following: ((WUI Year 2WUI Year 1)/ WUI Year 1), calculations were completed using QGIS. WUI densification were calculated using population data for each CWPP county, geographic region, which were compiled using census blocks for the 2000 and 2010 US Census in QGIS. Percent change was comput ed by: ((Population Year 2Population Year 1)/Population Year 1). CWPP and county ordinances promoting wildfire wise land use practices should eliminate or reduce WUI expansion and densification in highrisk wildfire zones. The composite index for CWPP implementation was calculated using the following formula: ((WUI Year 2 WUI Year 1)/WUI Year 1) + ((Population Year 2 Population Year 1)/Population Year 1).

PAGE 120

103 CHAPTER V STATISTICAL RESULTS OF CWPP PLANNING AND IMPLEMENTATION Statistical results for the CWPP planning and CWPP implementation are presented through this chapter. The first section is the descriptive statistic results for the integration of the CWPP planning process and the CWPP implementation in local governance. The second section discusses t he conditions of effective CWPPs and risk reduction integration through the use of inferential statistics of the scores in relation to independent variables. Integration and Implementation This section presents the results for the “CWPP Process and Plan E valuation Instrument ,” “CWPP Implementation: Local Governance Evaluation Instrument ,” and composite index scores. This section is subdivided into the following: CWPP integration scores, CWPP implementation scores, and CWPP composite scores. These scores are translated into a grade – A through F – and variation across states and strata are discussed. CWPP Integration Scores Each CWPP was given a unique identifier to facilitate coding and analysis. The pilot study, previously mentioned in Cha pter IV , was used to e valuate the reliability of the “CWPP Process and Plan Evaluation Instrument (Appendix A) and to align the coders’ understanding of the instrument. The pilot study’s interrater reliability of coder one and coder two was computed using Cohen’s Kappa to rate each coder’s question scoring. The coders targeted a “moderate” to “perfect” level of agreement (Kappa values of 0.60 – 1.0) as seen in the associated Cohen’s Kappa values of Table 5.1. Two rounds of coding were needed to achieve acce ptable Kappa values for all coding questions. The final results are: 42 questions (75%) have almost perfect interrater reliability, two questions (4%) have strong interrater reliability, and 12 questions (21%)

PAGE 121

104 have moderate reliability. For further details on interrater reliability see the full pilot write up in Appendix C . A total of 120 CWPP documents were coded and their metadata is provided in Appendix D . Table 5. 1. Cohen’s Kappa levels of reliability interpretation (McHugh, 2012) . Value of Kappa Level of Agreement % of Data that are Reliable 0 .20 None 0 – 3% .21 .39 Minimal 4 14% .40 .59 Weak 15 34% .60 .79 Moderate 35 63% .80 .90 Strong 64 81% Above .90 Almost Perfect 82 100% A detailed summary of average “Process and Plan evaluation instrument” categorical scores and total scores are summarized by state and presented in Table 5.2. Figure 5.1 presents a summary of the average total “Process and Plan evaluation instrument” score per state, per sample strata. The raw “Process and Plan evaluation instrument” scoring data is presented in Appendix B . The categorical or theme scores were created by taking the average of a particular category’s individual indicator scores. For example: theme one, “Context” has five indicators, as such, the categorical score would be Category 1 score = (Indicator 1 + Indicator 2 + Indicator 3 + Indicator 4 + Indicator 5)/5. The total score is an average of all the categorical scores. The minimum scores f or each category, one through eleven, were 0%. The maximum scores for each category exhibited greater variation, ranging from 33.3% for category three to 100% for categories two, five, seven, and eleven, and the average score was 62.6%. Category one (conte xt) maximum scores ranged from 0 40% with an average of 9%. Category two (goals and objectives) maximum scores ranged from 0 100% with an average of 52.3%. Category three (community capacity) maximum scores ranged from 0 33.3% with an average of 2.8%. Cate gory

PAGE 122

105 four (partnerships and collaboration) maximum scores ranged from 081.8% with an average of 37.9%. Category five (base map) maximum scores ranged from 0100% with an average of 2.5%. Category six (risk assessment) maximum scores ranged from 0 77.8% wi th an average of 42.5%. Category seven (hazardous fuels reduction) maximum scores ranged from 0100% with an average of 30.0%. Category eight (reducing structural ignitability) maximum scores ranged from 0 80% with an average of 21.5%. Category nine (educa tion and outreach) maximum scores ranged from 075% with an average of 24.4%. Category ten (emergency management capacity) maximum scores ranged from 0 66.7% with an average of 21.7%. Category eleven (longterm success) maximum scores ranged from 0 100% wi th an average of 44.0%. Table 5. 2. Average of categorical scores CWPP “Process and Plan Evaluation Instrument” document analysis score results for all strata (results are presented in percentage of total categorical score). State Category 1 2 3 4 5 6 7 8 9 10 11 Tota l Arizona Avg 6.7 66.7 22.2 42.4 0.0 63.0 66.7 13.3 33.3 16.7 83.3 37.7 Min 0.0 50.0 0.0 36.4 0.0 55.6 0.0 0.0 25.0 0.0 75.0 28.9 Max 20.0 87.5 33.3 45.5 0.0 77.8 100 20.0 50.0 33.3 100 47.0 California Avg 18.7 45.0 13.3 43.0 6.7 50.4 73.3 24.0 45.0 32.2 58.3 37.3 Min 0.0 0.0 0.0 9.1 0.0 22.2 0.0 0.0 25.0 0.0 0.0 23.5 Max 40.0 87.5 33.3 72.7 100 66.7 100 40.0 75.0 66.7 100 62.6 Colorado Avg 4.4 38.9 1.9 32.8 5.6 38.3 16.7 15.6 23.6 19.4 52.8 22.7 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 25.0 10.3 Max 40.0 87.5 33.3 72.7 100 77.8 100 60.0 75.0 50.0 75.0 46.2 Idaho Avg 3.2 52.0 1.8 41.6 0.0 39.8 26.3 29.5 22.4 21.9 52.6 26.5 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.3 Max 20.0 75.0 33.3 72.7 0.0 66.7 100 80.0 50.0 50.0 75.0 40.7 Montana Avg 9.0 54.4 0.0 38.6 0.0 39.4 15.0 24.0 20.0 24.2 31.3 23.3 Min 0.0 0.0 0.0 9.1 0.0 11.1 0.0 0.0 0.0 16.7 0.0 11.2 Max 20.0 75.0 0.0 63.6 0.0 55.6 100 60.0 75.0 50.0 75.0 40.1 Nevada Avg 20.0 80.0 0.0 29.1 0.0 46.7 20.0 20.0 5.0 3.3 25.0 22.6 Min 20.0 50.0 0.0 27.3 0.0 44.4 0.0 20.0 0.0 0.0 25.0 20.4 Max 20.0 87.5 0.0 36.4 0.0 55.6 100 20.0 25.0 16.7 25.0 31.7

PAGE 123

106 New Mexico Avg 10.0 60.9 0.0 30.7 12. 5 48.6 0.0 17.5 31.3 16.7 37.5 24.1 Min 0.0 0.0 0.0 9.1 0.0 33.3 0.0 0.0 0.0 0.0 25.0 6.1 Max 20.0 100 0.0 54.5 100 77.8 0.0 40.0 50.0 50.0 75.0 41.4 Oregon Avg 7.7 51.9 0.0 37.8 0.0 38.5 23.1 24.6 21.2 20.5 38.5 24.0 Min 0.0 37.5 0.0 18.2 0.0 11.1 0.0 0.0 0.0 16.7 0.0 12.0 Max 40.0 75.0 0.0 81.8 0.0 66.7 100 40.0 50.0 33.3 75.0 44.4 Washington Avg 10.9 56.8 0.0 52.1 0.0 46.5 45.5 25.5 20.5 27.3 40.9 29.6 Min 0.0 0.0 0.0 0.0 0.0 22.2 0.0 20.0 0.0 0.0 0.0 9.7 Max 40.0 75.0 0.0 72.7 0.0 55.6 100 60.0 25.0 33.3 75.0 46.1 Wyoming Avg 7.5 54.7 0.0 20.5 0.0 36.1 37.5 2.5 15.6 12.5 25.0 19.3 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.9 Max 20.0 100 0.0 72.7 0.0 55.6 100 20.0 25.0 33.3 50.0 39.9 Total Avg 9.0 52.3 2.8 37.9 2.5 42.5 30.0 21.5 24.4 21.7 44.0 26.2 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Max 40.0 100 33.3 81.8 100 77.8 100 80.0 75.0 66.7 100 62.6 Figure 5. 1. Minimum, maximum, and average of “Process and Plan Evaluation Instrument” document analysis total score results (results are presented in percentage of total categorical score). 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% 1 2 3 4 5 6 7 8 9 10 11 Total PercentCategory Minimum Maximum Average

PAGE 124

107 As shown in Figure 5.2, average total scores per state were moderately variable across strata . However, the average total scores were only minimally variable across states, except for the relatively higher scores in strata two in Arizona as well as strata one and two in California. Strata one’s total s core ranged between a minimum of 3.3% and a maximum of 49.5% with an average of 24.3%. Figure 5. 2. Average of “Process and Plan Evaluation Instrument” document analysis total score results per strata, summarized per state (results are presented in percentage of total categorical score). Table 5.3 summarizes the minimum, maximum, and average, per state for strata one. Strata two’s total score ranged between a minimum of 6.1% and a maximum of 62.6% with an average of 2 8.4%. Table 5.4 summarizes the minimum, maximum, and average, per state for strata two. Strata three’s total score ranged between a minimum of 8.9% and a maximum of 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Average total "Process and Plan Evaluation Instrument" scoreState Strata 1 Strata 2 Strata 3

PAGE 125

108 46.2%, with an average of 25.9%. Table 5.5 summarizes the minimum, maximum, and average, pe r state for strata three. Table 5. 3. M inimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” document analysis score results for strata one (results are presented in percentage of total categorical sco re). State Category 1 2 3 4 5 6 7 8 9 10 11 Tota l Arizona Avg Min Max California Avg 36.0 50.0 13.3 47.3 0.0 51.1 100 28.0 45.0 33.3 65.0 42.6 Min 20.0 25.0 0.0 18.2 0.0 33.3 100 20.0 25.0 16.7 50.0 33.6 Max 40.0 87.5 33.3 63.6 0.0 66.7 100 40.0 75.0 50.0 75.0 33.6 Colorado Avg 0.0 15.6 0.0 27.3 0.0 30.6 0.0 15.0 18.8 8.3 43.8 14.5 Min 0.0 0.0 0.0 9.1 0.0 11.1 0.0 0.0 0.0 0.0 25.0 10.3 Max 0.0 62.5 0.0 45.5 0.0 44.4 0.0 20.0 25.0 33.3 50.0 10.3 Idaho Avg 0.0 37.5 0.0 31.8 0.0 27.8 0.0 15.0 18.8 4.2 25.0 14.5 Min 0.0 0.0 0.0 0.0 0.0 11.1 0.0 0.0 0.0 0.0 0.0 3.3 Max 0.0 75.0 0.0 72.7 0.0 66.7 0.0 40.0 25.0 16.7 50.0 3.3 Montana Avg 12.7 52.3 0.0 39.7 0.0 37.4 27.3 21.8 15.9 25.8 34.1 24.3 Min 0.0 25.0 0.0 18.2 0.0 11.1 0.0 0.0 0.0 16.7 0.0 11.5 Max 20.0 62.5 0.0 63.6 0.0 55.6 100 40.0 50.0 50.0 75.0 11.5 Nevada Avg Min Max New Mexico Avg 10.0 50.0 0.0 40.9 50.0 55.6 0.0 20.0 37.5 16.7 37.5 28.9 Min 0.0 50.0 0.0 27.3 0.0 33.3 0.0 20.0 25.0 0.0 25.0 16.4 Max 20.0 50.0 0.0 54.5 100 77.8 0.0 20.0 50.0 33.3 50.0 16.4 Oregon Avg 3.3 54.2 0.0 37.9 0.0 40.7 0.0 23.3 25.0 19.4 33.3 21.6 Min 0.0 37.5 0.0 18.2 0.0 22.2 0.0 0.0 0.0 16.7 25.0 12.0 Max 20.0 75.0 0.0 54.5 0.0 66.7 0.0 40.0 50.0 33.3 50.0 12.0 Washington Avg 10.0 59.4 0.0 52.3 0.0 50.0 50.0 20.0 18.8 25.0 37.5 29.4 Min 0.0 50.0 0.0 9.1 0.0 33.3 0.0 20.0 0.0 16.7 0.0 12.9 Max 20.0 75.0 0.0 72.7 0.0 55.6 100 20.0 25.0 33.3 75.0 12.9

PAGE 126

109 Wyoming Avg 5.0 56.3 0.0 11.4 0.0 30.6 50.0 0.0 18.8 8.3 18.8 18.1 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 9.1 Max 20.0 100 0.0 27.3 0.0 44.4 100 0.0 25.0 16.7 25.0 9.1 Total Avg 10.5 48.1 1.7 36.8 2.5 39.4 30.0 19.0 23.1 19.6 36.9 24.3 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.3 Max 40.0 100 33.3 72.7 100 77.8 100 40.0 75.0 50.0 75.0 49.5 Table 5. 4. Mi nimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” document analysis score results for strata two (results are presented in percentage of total categorical score). State Category 1 2 3 4 5 6 7 8 9 10 11 Tota l Arizona Avg 10.0 75.0 16.7 40.9 0.0 66.7 100 20.0 37.5 8.3 87.5 42.1 Min 0.0 62.5 0.0 36.4 0.0 55.6 100 20.0 25.0 0.0 75.0 37.1 Max 20.0 87.5 33.3 45.5 0.0 77.8 100 20.0 50.0 16.7 100 47.0 California Avg 20.0 40.0 26.7 45.5 20.0 46.7 60.0 28.0 50.0 43.3 60.0 40.0 Min 0.0 0.0 0.0 27.3 0.0 22.2 0.0 0.0 25.0 33.3 25.0 29.3 Max 40.0 62.5 33.3 72.7 100 66.7 100 40.0 75.0 66.7 100 62.6 Colorado Avg 2.9 37.5 4.8 32.5 14.3 38.1 14.3 8.6 14.3 23.8 53.6 22.2 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 25.0 12.3 Max 20.0 75.0 33.3 72.7 100 55.6 100 40.0 25.0 50.0 75.0 38.3 Idaho Avg 4.4 51.4 3.7 47.5 0.0 49.4 33.3 37.8 25.0 29.6 58.3 31.0 Min 0.0 0.0 0.0 0.0 0.0 33.3 0.0 20.0 0.0 16.7 50.0 17.3 Max 20.0 75.0 33.3 72.7 0.0 66.7 100 80.0 50.0 50.0 75.0 40.7 Montana Avg 5.0 65.6 0.0 34.1 0.0 41.7 0.0 25.0 25.0 25.0 31.3 23.0 Min 0.0 50.0 0.0 9.1 0.0 11.1 0.0 20.0 0.0 16.7 0.0 11.2 Max 20.0 75.0 0.0 63.6 0.0 55.6 0.0 40.0 75.0 33.3 50.0 30.2 Nevada Avg 20.0 87.5 0.0 27.3 0.0 44.4 0.0 20.0 0.0 0.0 25.0 20.4 Min 20.0 87.5 0.0 27.3 0.0 44.4 0.0 20.0 0.0 0.0 25.0 20.4 Max 20.0 87.5 0.0 27.3 0.0 44.4 0.0 20.0 0.0 0.0 25.0 20.4 New Mexico Avg 10.0 53.1 0.0 27.3 0.0 44.4 0.0 15.0 25.0 20.8 37.5 21.2 Min 0.0 0.0 0.0 9.1 0.0 33.3 0.0 0.0 0.0 0.0 25.0 6.1 Max 20.0 75.0 0.0 54.5 0.0 55.6 0.0 40.0 50.0 50.0 75.0 32.9 Oregon Avg

PAGE 127

110 Min Max Washington Avg 10.0 56.3 0.0 48.5 0.0 42.6 50.0 30.0 20.8 27.8 45.8 30.2 Min 0.0 0.0 0.0 0.0 0.0 22.2 0.0 20.0 0.0 0.0 25.0 9.7 Max 40.0 75.0 0.0 63.6 0.0 55.6 100 60.0 25.0 33.3 75.0 46.1 Wyoming Avg Min Max Total Avg 9.0 53.8 5.8 39.5 5.0 45.3 30.0 24.5 24.4 25.4 50.0 28.4 Min 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.1 Max 40.0 87.5 33.3 72.7 100 77.8 100 80.0 75.0 66.7 100 62.6 Table 5. 5. Minimum, Maximum and Average of the CWPP “Process and Plan Evaluation Instrument” document analysis score results for strata three (results are presented in percentage of total categorical score). State Category 1 2 3 4 5 6 7 8 9 10 11 Tota l Arizona Avg 0.0 50.0 33.3 45.5 0.0 55.6 0.0 0.0 25.0 33.3 75.0 28.9 Min 0.0 50.0 33.3 45.5 0.0 55.6 0.0 0.0 25.0 33.3 75.0 28.9 Max 0.0 50.0 33.3 45.5 0.0 55.6 0.0 0.0 25.0 33.3 75.0 28.9 California Avg 0.0 45.0 0.0 36.4 0.0 53.3 60.0 16.0 40.0 20.0 50.0 29.2 Min 0.0 0.0 0.0 9.1 0.0 33.3 0.0 0.0 25.0 0.0 0.0 23.5 Max 0.0 75.0 0.0 54.5 0.0 66.7 100 20.0 50.0 50.0 75.0 32.5 Colorado Avg 8.6 53.6 0.0 36.4 0.0 42.9 28.6 22.9 35.7 21.4 57.1 27.9 Min 0.0 0.0 0.0 9.1 0.0 22.2 0.0 0.0 25.0 0.0 25.0 15.3 Max 40.0 87.5 0.0 63.6 0.0 77.8 100 60.0 75.0 50.0 75.0 46.2 Idaho Avg 3.3 62.5 0.0 39.4 0.0 33.3 33.3 26.7 20.8 22.2 62.5 27.6 Min 0.0 50.0 0.0 18.2 0.0 0.0 0.0 20.0 0.0 0.0 50.0 16.9 Max 20.0 75.0 0.0 54.5 0.0 55.6 100 40.0 25.0 50.0 75.0 37.6 Montana Avg 4.0 50.0 0.0 40.0 0.0 42.2 0.0 28.0 25.0 20.0 25.0 21.3 Min 0.0 0.0 0.0 18.2 0.0 22.2 0.0 0.0 0.0 16.7 0.0 11.3 Max 20.0 75.0 0.0 54.5 0.0 55.6 0.0 60.0 50.0 33.3 50.0 30.2 Nevada Avg 20.0 68.8 0.0 31.8 0.0 50.0 50.0 20.0 12.5 8.3 25.0 26.0 Min 20.0 50.0 0.0 27.3 0.0 44.4 0.0 20.0 0.0 0.0 25.0 20.4 Max 20.0 87.5 0.0 36.4 0.0 55.6 100 20.0 25.0 16.7 25.0 31.7

PAGE 128

111 New Mexico Avg 10.0 87.5 0.0 27.3 0.0 50.0 0.0 20.0 37.5 8.3 37.5 25.3 Min 0.0 75.0 0.0 18.2 0.0 44.4 0.0 20.0 25.0 0.0 25.0 21.1 Max 20.0 100 0.0 36.4 0.0 55.6 0.0 20.0 50.0 16.7 50.0 29.4 Oregon Avg 11.4 50.0 0.0 37.7 0.0 36.5 42.9 25.7 17.9 21.4 42.9 26.0 Min 0.0 37.5 0.0 18.2 0.0 11.1 0.0 20.0 0.0 16.7 0.0 14.0 Max 40.0 62.5 0.0 81.8 0.0 55.6 100 40.0 25.0 33.3 75.0 44.4 Washington Avg 20.0 50.0 0.0 72.7 0.0 55.6 0.0 20.0 25.0 33.3 25.0 27.4 Min 20.0 50.0 0.0 72.7 0.0 55.6 0.0 20.0 25.0 33.3 25.0 27.4 Max 20.0 50.0 0.0 72.7 0.0 55.6 0.0 20.0 25.0 33.3 25.0 27.4 Wyoming Avg 10.0 53.1 0.0 29.5 0.0 41.7 25.0 5.0 12.5 16.7 31.3 20.4 Min 0.0 25.0 0.0 9.1 0.0 22.2 0.0 0.0 0.0 0.0 25.0 8.9 Max 20.0 62.5 0.0 72.7 0.0 55.6 100 20.0 25.0 33.3 50.0 39.9 Total Avg 7.5 55.0 0.8 37.3 0.0 42.8 30.0 21.0 25.6 20.0 45.0 25.9 Min 0.0 0.0 0.0 9.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.9 Max 40.0 100 33.3 81.8 0.0 77.8 100 60.0 75.0 50.0 75.0 46.2 Translating the CWPP scores into letter grades resulted in 119 CWPPs earning an F (<59%) and one county earning a D (62.6%). As a result, it proved necessary to subdivide the Fs into high, middle, and low Fs (Table 5.6) to provide a more distributed datase t for correlation and ordinal regressions. It was decided not to logarithmically curve the grades due to the potential misconception that CWPPs that earned logarithmically curved As or Bs need little to know improvement. Indeed, there is a social value to earning a failing grade. Emerging research into wildfire mitigation and reducing structural ignitibility suggests implementing a few risk reduction measures, but not all best practices, does not adequately minimize risk. Therefore, researchers and practit ioners are beginning to advocate for an all or nothing approach (Graham et al., 2012; Lasky, 2018a) . The resulting reclassification resulted in one CWPP earning a D, 18 CWPPs earning a high F, 67 CWPPs earning a middle F, and 34 CWP Ps earning a low F.

PAGE 129

112 Table 5. 6. Score to letter grade conversion chart. Percentage Letter Grade 90 100% A 80 89% B 70 79% C 60 69% D 39 59% F – High 19 38% F – Middle 0 18% F – Low CWPP Implementation Scores A single coder, coded all relevant implementation documents using the “CWPP Implementation: Local Governance Evaluation Instrument” (Appendix A). As a result, there was no need for an interrater reliability measure. A total of 307 local government documents were coded, i ncluding: area, general and comprehensive plans, land use plans, zoning and building codes, multi hazard plans, subdivision regulations, development ordinances, growth policies, and fire codes. Every county had a different number of plans, policies, codes, and ordinances. Seven counties had no local documents of any kind because they have no planning, zoning, codes, or ordinances of any kind. In these cases, the counties fall under state minimum guidelines, so an additional 22 state statutes as well as buil ding and WUI codes were analyzed. A full listing of coded documents are listed in Appendix E . The minimum score across all strata was 0.0% while the max was 92.9% with an average of 21.2%. A summary of “C WPP Implementation: Local Governance Evaluation Inst rument” categorical scores and total scores are presented in Table 5.7. Figure 5.3 documents the minimum, maximum, and average scores across the four document analysis categories.

PAGE 130

113 Table 5. 7. Minimum, maximum, and average cat egorical scores of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for all strata (results are presented in percentage of total categorical score). Category State 1 2 3 4 Total Minimum Arizona 0.0 14.3 0.0 0.0 3.6 California 0.0 0.0 0.0 20.0 13.6 Colorado 0.0 0.0 0.0 0.0 0.0 Idaho 0.0 14.3 0.0 20.0 8.6 Montana 0.0 0.0 0.0 0.0 0.0 Nevada 0.0 14.3 0.0 20.0 8.6 New Mexico 0.0 0.0 0.0 0.0 0.0 Oregon 0.0 0.0 0.0 0.0 0.0 Washington 0.0 0.0 0.0 0.0 0.0 Wyoming 0.0 0.0 0.0 0.0 0.0 Total 0.0 0.0 0.0 0.0 0.0 Maximum Arizona 33.3 28.6 0.0 0.0 11.9 California 100 85.7 100 80.0 84.3 Colorado 100 85.7 100 80.0 78.9 Idaho 100 71.4 100 100 92.9 Montana 66.7 85.7 50.0 80.0 70.6 Nevada 100 42.9 50.0 40.0 54.6 New Mexico 66.7 28.6 0.0 40.0 20.2 Oregon 33.3 71.4 50.0 60.0 46.5 Washington 100 42.9 50.0 40.0 58.2 Wyoming 33.3 42.9 50.0 20.0 28.2 Total 100 85.7 100 100 92.9 Average Arizona 11.1 19.0 0.0 0.0 7.5 California 57.8 40.0 43.3 37.3 44.6 Colorado 18.5 25.4 16.7 31.1 22.9 Idaho 21.1 24.1 15.8 31.6 23.1 Montana 16.7 20.0 2.5 29.0 17.0 Nevada 20.0 22.9 20.0 28.0 22.7 New Mexico 12.5 8.9 0.0 10.0 7.9 Oregon 5.1 23.1 11.5 32.3 18.0 Washington 21.2 16.9 9.1 23.6 17.7 Wyoming 8.3 7.1 6.3 5.0 6.7 Total 21.1 22.5 14.2 27.0 21.2

PAGE 131

114 Figure 5. 3. Minimum, maximum, and of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis total score results (results are presented in percentage of total categorical score). As shown in Figure 5.4 there is a wide degree of variability in local governance implementation of wildfire risk reduction best practices by state and strata. Strata one’s total score ranged between a minimum of 0.0% and a maximum of 65.4%, with an average of 12.0%. Table 5.8 summarizes strata one’s minimum, maximum, and average, per state. Strata two’s total score ranged between a minimum of 0.0% and a maximum of 62.9% with an average of 22.0%. Table 5.9 summarizes strata two’s minimum, maximum, and average, per state. Strata three’s total score ranged between a minimum of 0.0% and a maximum of 92.9%, with an average of 29.6%. Table 5.10 summarizes strata three’ s minimum, maximum, and average, per state. 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% 70.0% 80.0% 90.0% 100.0% 1 2 3 4 Total PercentCategory Minimum Maximum Average

PAGE 132

115 Figure 5. 4. Average of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis total score results per strata, summarized per state (results are present ed in percentage of total categorical score). 0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0% Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Total Average total "Process and Plan Evaluation Instrument" scoreState Strata 1 Strata 2 Strata 3

PAGE 133

116 Table 5. 8. M inimum, maximum, and average categorical scores of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for strata 1 (results are presented in percentage of total categorical score). Category State 1 2 3 4 Total Minimum Arizona California 20.0 0.0 14.3 0.0 21.1 Colorado 0.0 0.0 0.0 0.0 0.0 Idaho 20.0 0.0 14.3 0.0 8.6 Montana 0.0 0.0 0.0 0.0 0.0 Nevada New Mexico 0.0 0.0 0.0 0.0 0.0 Oregon 0.0 0.0 0.0 0.0 0.0 Washington 0.0 0.0 0.0 0.0 0.0 Wyoming 0.0 0.0 0.0 0.0 0.0 Total 0.0 0.0 0.0 0.0 0.0 Maximum Arizona California 60.0 50.0 85.7 100 65.4 Colorado 20.0 0.0 28.6 66.7 21.7 Idaho 20.0 0.0 14.3 66.7 25.2 Montana 80.0 0.0 57.1 0.0 34.3 Nevada New Mexico 0.0 0.0 0.0 0.0 0.0 Oregon 60.0 50.0 42.9 0.0 38.2 Washington 20.0 0.0 14.3 0.0 8.6 Wyoming 20.0 50.0 42.9 33.3 28.2 Total 80.0 50.0 85.7 100 65.4 Average Arizona California 32.0 20.0 40.0 66.7 39.7 Colorado 10.0 0.0 7.1 16.7 8.5 Idaho 20.0 0.0 14.3 33.3 16.9 Montana 10.9 0.0 6.5 0.0 4.4 Nevada New Mexico 0.0 0.0 0.0 0.0 0.0 Oregon 23.3 8.3 14.3 0.0 11.5 Washington 15.0 0.0 10.7 0.0 6.4 Wyoming 5.0 12.5 10.7 8.3 9.1 Total 15.5 5.0 13.2 14.2 12.0

PAGE 134

117 Table 5. 9. Minimum, maximum, and average categorical scores of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for strata 2 (results are presented in percentage of total categorical score). Category State 1 2 3 4 Total Minimum Arizona 0.0 0.0 14.3 0.0 3.6 California 20.0 50.0 0.0 0.0 17.5 Colorado 20.0 0.0 14.3 0.0 8.6 Idaho 20.0 0.0 14.3 0.0 8.6 Montana 0.0 0.0 0.0 0.0 0.0 Nevada 20.0 0.0 14.3 0.0 8.6 New Mexico 0.0 0.0 0.0 0.0 0.0 Oregon Washington 0.0 0.0 14.3 0.0 3.6 Wyoming Total 0.0 0.0 0.0 0.0 0.0 Maximum Arizona 0.0 0.0 14.3 33.3 11.9 California 60.0 100 71.4 66.7 62.0 Colorado 80.0 50.0 57.1 33.3 46.8 Idaho 80.0 100 71.4 33.3 62.9 Montana 80.0 0.0 57.1 66.7 51.0 Nevada 20.0 0.0 14.3 0.0 8.6 New Mexico 20.0 0.0 14.3 33.3 16.9 Oregon Washington 40.0 50.0 42.9 100 58.2 Wyoming Total 80.0 100 71.4 100 62.9 Average Arizona 0.0 0.0 14.3 16.7 7.7 California 36.0 60.0 37.1 40.0 43.3 Colorado 34.3 7.1 28.6 14.3 21.1 Idaho 35.6 22.2 25.4 7.4 22.6 Montana 45.0 0.0 32.1 25.0 25.5 Nevada 20.0 0.0 14.3 0.0 8.6 New Mexico 10.0 0.0 7.1 8.3 6.4 Oregon Washington 26.7 8.3 21.4 38.9 23.8 Wyoming Total 29.5 15.0 24.3 19.2 22.0

PAGE 135

118 Table 5. 10. Minimum, maximum, and average categorical scores of the “ CWPP Implementation: Local Governance Evaluation Instrument ” document analysis score results for strata 3 (results are presented in percentage of total categorical scor e). Category State 1 2 3 4 Total Minimum Arizona 0.0 0.0 28.6 0.0 7.1 California 20.0 0.0 14.3 0.0 13.6 Colorado 0.0 0.0 0.0 0.0 0.0 Idaho 20.0 0.0 14.3 0.0 8.6 Montana 20.0 0.0 14.3 33.3 16.9 Nevada 40.0 50.0 28.6 0.0 33.2 New Mexico 0.0 0.0 14.3 0.0 17.1 Oregon 20.0 0.0 14.3 0.0 8.6 Washington 40.0 50.0 14.3 0.0 26.1 Wyoming 0.0 0.0 0.0 0.0 0.0 Total 0.0 0.0 0.0 0.0 0.0 Maximum Arizona 0.0 0.0 28.6 0.0 7.1 California 80.0 100 71.4 100 84.3 Colorado 80.0 100 85.7 100 78.9 Idaho 100 100 71.4 100 92.9 Montana 80.0 50.0 85.7 66.7 70.6 Nevada 40.0 50.0 42.9 100 54.6 New Mexico 40.0 0.0 28.6 66.7 20.2 Oregon 60.0 50.0 71.4 33.3 46.5 Washington 40.0 50.0 14.3 0.0 26.1 Wyoming 20.0 0.0 14.3 33.3 16.9 Total 100 100 85.7 100 92.9 Average Arizona 0.0 0.0 28.6 0.0 7.1 California 44.0 50.0 42.9 66.7 50.9 Colorado 40.0 35.7 32.7 23.8 33.0 Idaho 33.3 16.7 28.6 33.3 28.0 Montana 56.0 10.0 40.0 46.7 38.2 Nevada 40.0 50.0 35.7 50.0 43.9 New Mexico 20.0 0.0 21.4 33.3 18.7 Oregon 40.0 14.3 30.6 9.5 23.6 Washington 40.0 50.0 14.3 0.0 26.1 Wyoming 5.0 0.0 3.6 8.3 4.2 Total 36.0 22.5 30.0 30.0 29.6

PAGE 136

119 Implementation scores were translated using the same grading scheme classification as described in Table 5.6. This resulted in one county earning an A, one county earning a B, one county earning a C, four counties earning a D, 17 counties earning a high F, 63 counties earning a middle F, and 23 counties earning a low F. CWPP Composite Scores ( integration + implementation) Composite scores ranged from a minimum of 4.5% to a maximum of 65.2% with an average of 23.7%. Figure 5.5 documents the average total c omposite score, per strata, per state. Strata one scores had a minimum of 4.5%, a maximum of 54.9%, and an average of 18.2% (Table 5.11). Strata two scores had a minimum of 10.2%, a maximum of 53.0%, and an average of 25.2% (Table 5.11). Strata three score s had a minimum score of 4.5%, a maximum of 65.2%, and an average of 27.8%. Translating the total composite scores into letter grades (Table 5.6) resulted in the no counties earning an A, B, or C. One county earned a D, while 18 counties earned a high F, 52 counties earned a middle F, and 50 counties earned a low F. 0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0% 35.0% 40.0% 45.0% 50.0% Arizona California Colorado Idaho Montana Nevada New Mexico Oregon Washington Wyoming Total Average Average total composite scoreState Strata 1 Strata 2 Strata 3

PAGE 137

120 Figure 5. 5. Final composite score results per strata, summarized per state (results are presented in percentage of total categorical score). Table 5. 11. Minimum, maximum, and average composite score results for all strata. Strata Minimum Maximum Average Strata 1 4.5% 54.9% 18.2% Strata 2 10.2% 53.0% 25.2% Strata 3 4.5% 65.2% 27.8% Overall 4.5% 65.2% 23.7% Conditions of Effective CWPPs and Risk Reduction Implementation This section presents the results of inferential statistics for the “CWPP Process and Plan Evaluation Instrument,” CWPP Implementation Local Governance Instrument,” and the composite score of the two instruments in relationship to the independent variables: age, homeowner tenure, residency status, and income . For each of the two instruments and the composite score, the Spearman’s correlation coefficient results and the ordinal logistic regres sions results are presented. It is important to note two unexpected changes in analysis that arose during data processing and assumption testing. First, the Pearson’s product moment correlation was originally intended to be utilized instead of the Spear man’s correlation coefficient. The pilot dataset met the Pearson’s product moment correlation assumptions tests; however, the full research sample dataset failed Pearson’s test of normality, has a number of outliers that cannot be removed, and does not exhibit homoscedasticity. Therefore, the Spearman’s correlation coefficient was a more appropriate correlation test, and the assumptions were tested and met. Second, the intention was to run power analyses. However, because none of the ordinal regression results were statistically significant, the power analyses could not be processed. Please note that the lack of statistical significance is likely due to too few observations with high scores

PAGE 138

121 to get a signal. However, the results do provide ample anecdotal results to advance planning theory and practice, as presented in Chapter VI . CWPP Process and Plan Evaluation Inferential Statistics Spearman’s correlation coefficient results A Spearman's rank order correlation was run to assess the relationship between C WPP Process and Plan Evaluation scores (CWPP scores) and the independent variables in counties with CWPPs. One hundred and twenty counties were sampled. Preliminary analysis showed the relationship to be monotonic, as assessed by visual inspection of a scatter plot. There were no statistically significant correlation between CWPP scores and age, rs(118) = 0.110, p=0.233; homeowner tenure rs(118) = 0.128, p=0.163; or income rs(118) = 0.135, p=0.141. However, there was a statistically significant, small positive correlation between CWPP scores and the growth of WUI seasonal homes between 2000 and 2010, rs(118) = 0.180, p < 0.05. Ordinal logistic regressions results A cumulative odds ordinal logistic regression with proportional odds was run to determine the effect of age, homeowner tenure, residency status, and income on CWPP Process and Plan Evaluation scores. There were proportional odds, as assessed by a full likelihood ratio test comparing the fitted model to a model with varying location parameters, 2(8) = 2.618, p = 0.956. The deviance goodness of fit test indicated that the model was a good fit to the observed data, 2( 353) = 238.997, p = .10, but most cells were sparse w ith zero frequencies in 75.0% of cells. However, the final model did not statistically significantly predict the dependent variable over and above the intercept 2(4) = 2.726, p = 0.605. An increase in age (expressed in years) was associated wi th an increase in the odds of higher CWPP scores, with an odds ratio 2(1) = 0.620, p = 0.431.

PAGE 139

122 An increase in homeowner tenure (expressed in years) was associated with an increase in the odds of higher CWPP scores, with an odds ratio of 0.15, 95% CI [0.741, 1.130], which is not 2(1) = 0.680, p = 0.410. An increase in WUI seasonal housing units (expressed in dwelling units) was associated with an increase in the odds of higher CWPP scores, 2(1) = 0.260, p = 0.610. An increase in income (expressed in dollars) was associated with an increase in the odds of higher CWPP scores, with an odds ratio of 1.0, 95% CI [1.0, 1.0], which is not statistically 2(1) = 0.097, p = 0.756. CWPP Implementation Local Governance Evaluation Inferential Statistics Spearman’s correlation coefficient results A Spearman's rankorder correlation was run to assess the relationship between Implementation Local Governance Evaluation scores (LG scores) and the independent variables in counties with CWPPs. One hundred and twenty counties were sampled. Preliminary analysis showed the relationship to be monotonic, as assessed by visual inspection of a scatter plot. There were no statistically significant correlation between LG scores and age, rs(118) = 0.166, p=0.070 or income, rs(118) = 0.096, p=0.298. However, there was a statis tically significant, small negative correlation between LG scores and homeowner tenure, rs(118) = 0.208, p < 0.05. Additionally, there was a statistically significant, medium positive correlation between LG scores and the growth of WUI seasonal homes betw een 2000 and 2010, rs(118) = 0.480, p < 0.01. Ordinal logistic regressions results A cumulative odds ordinal logistic regression with proportional odds was run to determine the effect of age, homeowner tenure, residency status, and income on CWPP Implementation Local Governance Evaluation scores. The proportional odds test failed, as

PAGE 140

123 assessed by a full likelihood ratio test comparing the fitted model to a model with varying location parameters. A deeper investigation of the assumption of proportional odds was undertaken by running separate binomial logistic regressions on cumulative dichotomous dependent variables (six categories). The assumption is the proportional odds of the estimated parameters and by extension the odds ratio should be the same for each parameter in the equations. As you can see in Table 5.12, two of the four independe nt variables (age and homeowner tenure) are highly variable across the six categories. Power analyses were not considered because the regression results were not statistically significant. Table 5. 12. CWPP Implementation Local Governance Evaluation proportional odds test results. Independent variables Age Homeowner Tenure WUI Seasonal Housing Income Intercept B (Parameter Estimates) Category 1 .39 .175 0 0 .719 Category 2 .082 .277 .001 0 .022 Category 3 .059 .171 0 0 2.09 Category 4 .155 .582 0 0 1.093 Category 5 .339 .807 0 0 1.68 Category 6 .44 .527 .001 0 .16 Exp(B) (Odds Ratio) Category 1 1.04 1.192 1 1 .487 Category 2 1.086 1.319 .999 1 1.022 Category 3 1.06 1.186 1 1 1.022 Category 4 1.167 1.789 1 1 2.983 Category 5 1.404 2.242 1 1 2.983 Category 6 1.552 1.694 1.001 1 .852

PAGE 141

124 Composite Evaluation Inferential Statistics Spearman’s correlation coefficient results A Spearman's rank order correlation was run to assess the relationship between Composite Evaluation scores (C scores) and the independent variables in counties with CWPPs. One hundred and twenty counties were sampled. Preliminary analysis showed the relationship to be monotonic, as assessed by visual inspection of a scatter plot. There was no statistically significant correlation between C scores and income, rs(118) =0.013, p=0.888. However, there was a statistically significant, small negative correlation between C scores and age, rs(118) = 0.188, p < 0.05. Addi tionally, there was a statistically significant, small negative correlation between C scores and homeowner tenure, rs(118) = 0.238, p < 0.01. Finally, there was a statistically significant, medium positive correlation between C scores and the growth of W UI seasonal homes between 2000 and 2010, rs(118) = 0.432, p < 0.01. Ordinal logistic regressions results A cumulative odds ordinal logistic regression with proportional odds was run to determine the effect of age, homeowner tenure, residency status, and income on composite scores. There were proportional odds, as assessed by a full likelihood ratio test compa ring the fitted model to a model with varying location parameters, 2(8) = 8.542, p = 0.382. The deviance goodness of fit test indicated that the model was a good fit to the observed data, 2( 326 ) = 216.665, p = 0.665, but most cells were sparse with zero frequencies in 75.0% of cells . However, the final model statistically significantly predicte d the dependent variable over and above the intercept only model 2(4) = 18.372, p < 0.001. An increase in age (expressed in years) was associated with an increase in the odds of higher composite scores, with an odds ratio of 0.918, 95% CI [0.757, 1.114], 2(1) = 0.744, p = 0.388. An increase

PAGE 142

125 in homeowner tenure (expressed in years) was associated with an increase in the odds of higher composite scores, with an odds ratio of 0.824, 95% CI [0.654, 1.038], which is not statistically 2(1) = .2702, p = 0.10. An increase in WUI seasonal housing units (expressed in dwelling units) was associated with an increase in the odds of higher composite scores, with an odds ratio of 1.0, 95% CI [1.0, 1.001], which is no 2(1) = 8.968, p = 0.003. An increase in income (expressed in dollars) was associated with an increase in the odds of higher composite scores, with an odds ratio of 1.0, 95% CI [1.0, 1.0], which is not statistically significant 2(1) = 0.638 , p = 0.424.

PAGE 143

126 CHAPTER VI CONCLUSIONS ON CWPP AND LOCAL GOVERNANCE INTEGRATION IN REDUCING WILDFIRE RISK Discussion of the what the results mean in the application of the CWPP process and conclusions of this project on the CWPP and local governance integration of best practices in reducing wildfire risk are presented in this chapter. First, a discussion of the regression results is presented. Second, document analyses are explored for both the CWPP document analysis results and the local governance document analysis results. Third, theory, practice, and policy implications of this research project are discussed, including the updating of CWPPs and ordinances ; framing of the CWPP and comprehensive plans ; creating CWPP goals, objective s, and priorities; and improving CWPP mapping. Fourth, the limitations and avenues for future research are discussed. Finally, some ending conclusions are presented. Regression Results The intent of the regression analysis was to validate and identify t he generalizability of past, small case study research. Past research identified the following independent variables as barriers to engaging with wildfire risk reduction efforts: age, length of homeowner tenure, full time/parttime residency, and income. W hile, as reported in the previous chapter, none of the regression results were statistically significant, four emerging trends were revealed. The first of these trends is that the WUI is still growing, despite research demonstrating the need to restrict or limit growth in wildfire risk areas (Bhandary & Mulle r, 2009; Butsic et al., 2015; Headwaters Economics, 2016b; Mowery, 2008; Paterson, 2007; Syphard et al., 2013) . Between 2000 and 2010, 85 counties experienced greater than 0.25 km2 of W UI growth, with an average of 11.67 km2, which is concerning as this trend may continue, or even increase its rate. However, this is

PAGE 144

127 not a universal trend of growth, as the WUI shrunk in 24 counties ranging between 0.04 km2 and 5.10 km2 with an average of 0.88 km2. Furthermore, 11 counties experienced no to minimal WUI growth (0 0.25 Km2), with an average of 0.07 km2. The second emerging trend of increased population in the WUI further complicates the spatial growth of WUI. Population increased in 96 counties that were assessed in this study, adding an additional 1,917,868 people to the WUI (24 counties experienced a cumulative decrease of 9,516 people). According to wildfire theory and best practices, wildfire risk reduction needs to be addressed as a land use problem in order to stabilize or reduce the spatial extent of the WUI and stabilize or decrease the WUI population. However, according to this research, counties are largely not succeeding in this land use management. While two time period data points are insufficient to definitively identify ongoing trends, this eme rging trend suggests that current land use planning efforts have done little to alter development consumption within wildlands and the WUI, and this suggestion is also supported by Headwaters Economics’ (2016b) , Molly Mowery (2008) , Bustic et al. (2015) , Radeloff et al., (2005; 2017) and Rice and Davis (1991) . This WUI growth is concerning because communities and lives are continuing to be placed at risk to wildfire, and if significant land use control interventions are not enacted, this emerging trend is projected to continue unabate d because as market are unable to control development in highly desirable WUI locations (Gorte, 2013; Headwaters Economics, 2014, 2016a; Simon, 2016; Austin R Troy, 2001) . As only 16% of the WUI in the American West has been developed, there is significant potential for continued growth. In the local governance document analysis scores and results discussed below, reasons for the lack of local governance impact on slowing or eliminating grow th are illuminated.

PAGE 145

128 The third emerging trend is a small positive correlation between the increase of WUI seasonal home growth and the composite score of the CWPP and local governance instruments. After 2000, 73 counties updated their local governance poli cies and ordinances, despite not having updated them in at least a decade or more; a handful of counties had not updated their comprehensive plans, codes or ordinances since 1970 or 1980. It appears that seasonal home growth in the WUI may be serving as a motivation for counties to update CWPPs or local governance ordinances to potentially curtail vacation home growth and to be protective of the local sense of place. The policy and ordinance updates did not adequately address all wildfire best practices or nor did these counties update their CWPPs during the same time frame, and this suggests that the local counties were not necessarily spurred by a concern to wildfire. Local governance updates were more heavily influenced by local residents concerns for the pace, scale, and aesthetics of recent development trends. This is evident in the language of the comprehensive plans’ updated goals and objectives, types of zoning code, and subdivision ordinances. For example, Nevada County, California (TSS Consultants) comprehensive plans goals and objectives are: 1) fostering a rural quality of life; 2) sustaining a quality environment; 3) development of a strong diversified, sustainable local economy; and 4) planned land use patterns will d etermine the level of public services appropriate to the character, economy, and environment of the region. Digging further into Nevada County’s intentions, as documented in their general plan (TSS Consultants) , the rural character (the anti thesis to suburban sprawling development) and the preservation of scenic aesthetics, agriculture, logging, and mining drive the county decisions, not wildfire concerns. Similar intentions were expressed in all 73 county updates. However, whi le these policy and ordinance updates were spurred by concerns other than

PAGE 146

129 wildfire, these update cycles could potentially provide counties with an opportunity to concurrently update their CWPPs. The fourth emerging trend is a small negative correlation between longer homeowner tenure and the composite CWPP process and local governance composite scores. One would expect that the longer residents lived in the WUI, the more invested they would be in ensuring that their homes were protected by wildfire by engaging in all available avenues (e.g., CWPPs, local governance, and personal mitigation efforts). In fact, theory would suggest that the more fire seasons residents have experienced and the more wildfire mitigation knowledge with which they are presented, r esidents would actively pursue decreasing their wildfire risk. However, research has also shown that individuals experience catastrophic wildfires differently and engage in varying levels of wildfire education and mitigation efforts (T. W. Collins, 2008b; Crow et al., 2015; Kousky et al., 2011; Pavegli o et al., 2009) . Furthermore, Champ and Brenkert Smith (2016) have identified instances where residents of WUI communities in high risk areas have perceptions of risk dangerously lower than expected because these particular communities have not experienced reoccurring wildfires. Therefore, wildfire risk reduction ef forts have a very short window to capitalize on using wildfires as an action leverage point because homeowner memories of wildfire experiences can be short lived if they did not suffer devastating consequences of the fire (Champ & Brenkert Smith, 2016; Champ, Donovan, & Barth, 2013; Austin R Troy, 2001) . While local knowledge of longtime WUI residents is critically important to CWPP and local governance efforts, planners should not assume longtime WUI residency equates to high levels of wildfire risk reduction know ledge, action, or behavior.

PAGE 147

130 CWPP Process Document Analysis Scores and Results CWPP Document Analysis Scores and Results Context No county received a passing grade for the context section of the CWPP process . Only six counties documented how they handled past challenges to help the community understand their vulnerability and paths to success. A great example of doing this was Siskiyou County, California, which documenting past wildfire hazard challenges and efforts. Their CWPP executive summary documente d the return interval, acreage burned, and cost of all major fires in the county: There have been a number of major wildland fires in Siskiyou County over the years. During the current 2006 fire season, three major fires, i.e., Happy Camp complex (3,900 ac res/$9.5 million cost to date), Uncles Complex (16,400 acres / $10.1 million cost to date), and the Orleans Complex (16,000 acres / $16.8 million cost to date) have vividly demonstrated the devastating damage potential for wildland fire . Wildland fire has deadly implications . Recently, two helicopter personnel lost their lives during the Happy Camp Complex incident, and the Stanza Fire in 2002 cost three USFS firefighters their lives. Wildland fire is an extremely dangerous natural hazard. (FireSafe Council of Siskiyou County, 2008, p. 1) Additionally, Siskiyou County documents past mitigation efforts, including complete and incomplete projects. Their documentation serves as an example because it includes a robust documentation of project type, location, implementing party and cost, across public and private lands, and infrastructure investment an excerpt of which is provided below in Table 6.1.

PAGE 148

131 Table 6. 1. Siskiyou County, CA’s Fire Safe Council Projects in Siskiyou County Volunteer & Locally Funded Projects Implementing Party Cost Volunteer workday on 10/17/2003: 30 50 foot brushing and limbing on Old Beaver Creek Road. Cutting back weeds, vines, removing low limbs, slash and dead wood. 13 volunteers, 4 hours each = 52 hours total. Klamath River Fire Safe Council Private Lands Fire/Fuel Projects Funded by RAC May02/June03 Salmon River Fuel Reduction SRRC/Salmon River FSC $9,000.00 Forks Fire Hydrant Forks of Salmon Community Club $27,600.00 Sawyers Bar Fuel Plan and Reduction: For fire safe planning and fuels reduction in the Sawyers Bar area. SRRC/Salmon River FSC $27,600.00 Five counties documented their past collaborative efforts and lessons learned. Tuolumne County, California documented a long history of wildfire risk reduction collaboration. They also documented several issues and lessons learned that make CWPP, projects, grants, and contracts more difficult, specifically related to institutional issues: 1. The delay in announcing grants that were awarded through the National Fire Plan reduced the time frame to actually implement the grant projects. a. This has frustrated both the agency chiefs and cooperator sponsors of these projects. b. A more t imely process must be developed to streamline grant award notification. 2. USFS contract equipment, crews and CDF fire crews have been very involved in the implementation of fire plan projects. There are not enough crews and equipment to support the work load generated by the CWPP. CDF managers in Sacramento, BLM, National Park Service and USFS managers in Washington must continue to support the use of crews and equipment. Expansion of such programs must occur. 3. The National Fire Plan grants do not allow funding of maintenance projects that will treat fuels that have grown back in existing fuel breaks and treated areas.

PAGE 149

132 a. The only way that most of these maintenance measures will take place is through the use of grant dollars. b. Agencies, fire safe cou ncils, and strategic groups must communicate this fact to the grantors, otherwise maintenance of past treatment efforts will never occur. 4. Prescribed burns have become more difficult to execute for the following reasons: a. Implementation of more stringe nt air pollution rules Burn Days occur less frequently than before. b. More difficult to schedule equipment and personnel resources during fire season. Many burns were postponed or cancelled altogether because resources were committed to incidents or cover assignments. c. Early fall rains have caused burns to be cancelled. d. Due to lawsuits being filed against government officials following recent prescribed burn escapes that have caused property damage, many agency officials are not will ing to assume that liability. 5. Programmatic Environmental Impact Reports for performing VMP’s in coniferous forests need to be approved to avoid the current requirement of filing Negative Declarations for these projects. 6. Need to integrate both Nationa l Environmental Protection Agency (NEPA) and California Environmental Quality Assessment (CEQA) requirements into a single checklist to prevent the necessity of duplicating these efforts on projects where the Federal agencies and CDF are partners. 7. Delay in processing the new Five Party Agreement has delayed projects. 8. There is usually not enough Agency staff available to identify, plan and implement Pre Fire projects during the nonfire season. 9. Weather has affected project implementation in the foll owing ways: a. Snow at higher elevation projects has kept crews from working during the winter months. This has been an issue on two grant projects. b. As mentioned above in #4, rain has caused delays or cancellation of prescribed burns. (FireSafe Council of Siskiyou County, 2008, pp. vi vii) Tuolumne County raises valid concerns and highlights the difficulty of cross scale and jurisdictional collaboration and coordination; however, they have not explored ways in which they could contribute to solving these issues. Furthermore, Tuolumne County has not explored

PAGE 150

133 their own shortcomings in local governance, funding, mitigation programs, or participatory barriers. Trinity County, California on the other hand has acknowledged and documented their own shortcomings in past wildfire efforts: Historically, county or regional scale wildfire management planning effort s have often failed to involve or even acknowledge local residents' knowledge and expertise” (Trinity County Resource Conservation District & The Watershed Research and Training Center, 2010, p. 1) . Trinity went on to doc ument their process, how the community was engaged throughout the process, and how decisions were made. Forty one counties identified key people or organizations that were involved in earlier wildfire planning efforts who were also included in ongoing CWP P efforts. These stakeholders included community leaders, residents, search and rescue, planners, state and federal agencies, county government, church leaders, and industry leaders. Only t wo counties document their collaborative experience. Mendocino County, California documented the history of their wildfire efforts. This documentation provides a window into the historical capacity of Mendocino’s efforts, county residents tend to rely too heavily on fire suppression resources without taki ng responsibility for their own safety, thus putting both themselves and firefighters in harm’s way needlessly. (Mendocino County Board of Supervisors, Mendocino County Fire Chiefs' Association, & California Department of Forestry and Fire Protection, 2005, p. 4) Additionally, Mendocino County and Firewise councils acknowledged their lack of certain expertis e and strategically reached out to various people and organizations for specific support, such as Mendocino County Air Quality Management, Ranch Associations, and public meeting facilitators and trainers. Mendocino County specifically credited, “Ms. Kelly Elder, who

PAGE 151

134 professionally facilitates public meetings, trained all facilitators free of charge and contributed the ‘sticky wall’ concepts” (Mendocino County Board of Supervisors et al., 2005, p. 16) . No c ounties document past community conflicts and their potential barriers to the CWPP process, which is concerning because these past community and political issues can cause a lack of engagement and resentment, which derails the efficacy of the process and i mplementation. Perhaps this information is more difficult to collect, especially given the number of CWPPs that were written by nonlocal consultants. However, it is still a necessary step to facilitate engagement and to overcome participation barriers. G oals and Objectives Goals and objectives should structure the discussion and the wildfire mitigation efforts in a CWPP. Only 10 counties received passing scores for goals and objective: Del Norte County, California (B), Jefferson County, Colorado (B), Silver Bow County, Montana (B), Bannock County, Idaho (C), Clackamas County, Oregon (C), El Paso County, Colorado (C) Nevada County, California (C), Trinity County, California (C), Tulare County, California (C), and Valley County, Idaho (C). The more concerning figure is 15 CWPPs had no goals at all, despite HFRA requiring CWPPs to have clearly articulated goals. Additionally, HFRA also suggest CWPPs to have a WUI definition goal and fuel mitigation goals, yet only 50 counties had WUI definition goals. Seventythree counties had structural ignitibility goals. Seventy four counties had forest thinning goals. Only 31 counties had defensible space or home ignition zone goals. It is important to note that this study assessed the presence or absence of goals, not t he quality of them. So even when goals were present, many of them were simply repeating vague HFRA language and were not robust enough to lead to logical and time sensitive objectives and implementation strategies. As such, these counties had no way to evaluate CWPP

PAGE 152

135 implementation progress overtime in reducing wildfire risk. This issue is demonstrated by El Paso County, Colorado’s singular goal of creating a CWPP to comply with Colorado Legislature mandates and HFRA: Purpose, constraints, and goals – This Community Wildfire Protection Plan (CWPP) was created to comply with a mandate of the Colorado Legislature, while also meeting the requirements of CWPPs as defined in the Healthy Forest Restoration Act (HFRA). (Russell, Campbell , & Root, 2011, p. 1) The Owyhee County, Idaho goals and objectives were better, yet they are still inadequate because they do not include a WUI definition goal or a structure ignitibility reduction goal: The goals of this WildlandUrban Interface Fi re Mitigation Plan include: 1. Improve Fire Prevention and Suppression 2. Reduce Hazardous Fuels 3. Restore Fire Adapted Ecosystems 4. Promote Community Assistance Its three guiding principles are: 1. Priority setting that emphasizes the protection of comm unities and other highpriority watersheds at risk. 2. Collaboration among governments and broadly representative stakeholders. 3. Accountability through performance measures and monitoring for results. (Russell et al., 2011, p. 3) However, it is worth noting that while Owyhee County’s CWPP does not include a WUI definition goal, they do thoroughly define and map the WUI. This omission is more concerning for counties that do not have a WUI goal or thoroughly documented WUI locations. Many counties also simply copied HFRA and CW PP goals without including contextual, community values, or following best practices for constructing goals, objectives, and strategy. Indeed, Boulder County, Colorado had bulleted list of goals, as shown in Figure 6.1. While they did follow up their vague goals with purpose statements (Figure 6.2), these did not include tangible objectives or strategies with timeframe language. Further evidence of gaps in goals and

PAGE 153

136 objective include: 73 counties include a structural ignitibility goal or objective ; 74 count ies include a forest thinning goal or objective ; and 31 counties include a defensible space or home ignition zone goal and objective. Figure 6. 1. Boulder County, Colorado CWPP goals (Boulder County Board of County Commiss ioners, 2011, p. 1) Figure 6. 2. Boulder County, Colorado CWPP goals (Boulder County Board of County Commiss ioners, 2011, p. 1) Counties should model their goals and objectives after Crook County, Wyoming or Taos County, New Mexico because they exhibited all the requisite types of goals and objectives,

PAGE 154

137 including multiple frames and actions to overcome county limitations. However, missing from all CWPPs are timeconstrained and measurable objectives, and CWPPs could be dramatically improved with the inclusion of these. For example, Crook County, Wyoming’s goal “reduce dense undergrowth that fuels catastrophic fires through thinning and prescribed burns” (Crook County Commission, p. 1) should be improved adding a timeframe attached to assess progress towards the goal. There should also be a longer timeframe attached for assessment of risk because wildfire mitigation goals are dynamic, and need regular reassessment. Additionally, this goal could be increased by defining the areas and acreage needing to be thinned and burned. Multiple frames were included in 47 counties’ goals and objectives. The multiple frames that counties included were public health, ecosystem health, home and asse t protection, and education. While “frames” are not explicitly stated, they emerged in the executive summary, introduction, or goals and objectives. The importance on framing is evident in the kinds of goals and objectives the county pursues, methods of fuel thinning, and risk reduction. For example, Baker County, Oregon expressed the following goals and objectives: Goal: Creating Fire Adapted Communities • Identify areas at risk and existing hazards. • Promote cooperation, relationships, and partnerships among agencies, organizations, jurisdictions, and communities. • Improve pre suppression planning strategies among all agencies with protection responsibilities. • Encourage stakeholder participation in development of strategies that will reduce wildfire risk. • Promote fire prevention and education. • Provide education and prevention messages targeted at creating defensible space, fuels reduction and improved structure access. Goal: Enhancing Safe and Effective Response to Wildfires • Seek opportunities to maintain and improve interagency wildland fire presence and interagency emergency response systems. • Collaborate on opportunities to secure additional fire equipment and infrastructure to enhance fire response capability.

PAGE 155

138 • Provide intera gency wildland fire training among agencies, organizations, jurisdictions, among agencies, organizations, jurisdictions. Goal: Restoring and Maintaining Resilient Landscapes in Fire Adapted Ecosystems: • Identify and treat hazardous fuels. • Provide for rapid assessment and treatment of burned lands, including implementation of stabilization techniques. • Communities will encourage land management agencies to promote the control of invasive species and consider establishment of native seed and plant mate rial. • Provide for maintenance of fuels treatments. Baker County’s expanded frames include education, ecosystem health, suppression and safety, and governance. These multiple frames have not only created a CWPP that considers a wider range of implement ation strategies and impacts on risk reduction, it has also expanded their potential capacity. For example, reframing resilient fire adapted landscapes as an invasive species issue provides the opportunity to include nontraditional wildfire agencies and f unding mechanisms (e.g., USDA Invasive Species Program). Conversely, the framing of Prowers County Colorado is concerning because they have framed the WUI wildfire problem solely as a fire suppression capacity issue and public awareness / education issue. T herefore, this framing has limited their CWPP to only explore suppression and education strategies. For example, their discussion regarding structural ignitibility includes the following, “Public awareness of Firewise concepts and self implementation of applicable concepts may help landowners potentially reduce the risk of structural ignitability in the event of an encroaching wildfire” (Prowers C ounty Commissioners, 2013, p. 7) . Similarly, discussing longterm implementation plans, efforts focused on: Initiating public awareness in Firewise Concepts (i.e., Are You At Risk!, Access, Water Supply, Defensible Space, Trees and Shrubs, Construction Design and Materials, Interior Safety, and ‘What to do When’). Target audiences could include community and rural

PAGE 156

139 homeowners, rural businesses, and government entities, as appropriate. Providing ‘General Outdoor Fire Safety’ brochures/information to vari ous publics (smoking, outdoor camping fires, trash burning, agricultural burning and other controlled burns, spark arresters, etc .). Target audiences could include community and rural homeowners, rural businesses, recreational users, tourists, highway corr idor travelers, and railroad personnel. Provide overall information/awareness about Fire Bans, Red Flag Warnings . (Prowers County Commissioners, 2013, p. 8) While these frames of reference are not inherently bad, limiting to single frames or suppression/education frames are dangerous. R esearch has repeatedly shown that focusing only on suppression and awareness / education is not adequate in dr astically reducing wildfire risk (Champ & Brenkert Smith, 2016; Champ et al., 2013) . Counties used the goals and objectives in their CWPPs to address local values. Twenty one counties articulated local values in their goals and objectives. For example, Crook County, Wyoming framed wildfire mitigation goals and objectives as a potential su pplyside energy production frame and forest ecosystem health: Authorize the Healthy Forests Reserve Program to protect, restore and enhance degraded forest ecosystems on private lands to promote the recovery of threatened and endangered species; Encourage biomass energy production through grants and assistance to local communities creating market incentives for removal of otherwise valueless forest material; and Develop an accelerated program on certain Federal lands to combat insect infestations (Crook County Commission, p. 2) . Conversely, Boulder County, Colorado used a unique narrative to describe their values and how the CWPP goals and objectives responded to community concerns in promoting a sense of community, recreation, and ecological health through wildfire risk reduction efforts. They also included community values in the importance of pre planning for wildfires through proverbs.

PAGE 157

140 Proverbs from community members include, “a stitch in time saves nine,” “an ounce of prevention is worth a pound of cure,” and “procrastination is the thief of time,” (Boulder County Board of County Commissioners, 2011, p. 1) . CWPPs were framed at the correct scale for 101 counties, leaving 19 counties that were incorrectly framed. Incorrectly framed CWP Ps exhibited a similar trait : their primary focus was to motivate homeowners to reduce hazards on their properties. However, Williams et . al ., (2012) advise that county level plans should focus on reducing wildfire risk across the landscape. Community Capacity Communi ty capacity is paramount to adhering to wildfire risk management efforts. Six counties received passing grades (A) for community capacity: Bannock County, Idaho, Boulder County, Colorado, El Paso County, Colorado, Mendocino County, California, San Diego C ounty, California, and Valley County, Idaho. These counties thoroughly documented the CWPP process and resources available to support and maintain community participation. Most of the capacity documentation focused on suppression capacity (e.g., not enough equipment or water infrastructure) or economics. Economic concerns were focused on money to buy firefighting equipment and pay for fire fighters or community mitigation efforts. In fact, Mohave County, Arizona framed their CWPP process as a means to ident ify funding needs and opportunities Additional functions of a CWPP are to improve fire prevention and suppression activities, as well as to identify funding needs and opportunities to reduce the risk of wildland fire and enhance public and firefighter safe ty. (Mohave County Board of Supervisors, 2008, p. v) Additionally, a few counties also discussed community capacity issues regarding the political will to address the roo t causes of wildfire, community conflicts, and differing community values. Only a handful of counties such as Siskiyou, California acknowledged that

PAGE 158

141 past participation efforts were often compulsory, not fullyengaged, and in need of a change, as demonstrat ed from their preface: National and State Fire Plans mandate legitimate community based planning efforts with full stakeholder participation, coordination, project identification, prioritization, funding review, and multi agency cooperation. Past and exist ing fire planning efforts and documents often experience absence of meaningful community and stakeholder initiative and engagement, data and validation gaps, and multi jurisdictional disparities. Our intention is to acknowledge and catalyze every citizen’s responsibility for fire safety while creating local grassroots’ community buy in, and a sense of ownership driven by synergistic empowerment (FireSafe Council of Siskiyou County, 2008, p. ii). However, Siskiyou did not explicitly document their strategies for more meaningful engagement. All CWPPs should improve their documentation of strategies to overcome capacity concerns. Partnerships and Collaboration Nine counties received passing scores for partnerships and collaboration. Eightyone counties documented the core CWPP team , and 97 CWPPs documented participants beyond the core team. Many counties had a broad arra y of stakeholders, and Siskiyou County, California serves as a model due to its breadth and diversity of inclusion (Table 6.2). However , even Siskiyou County is missing several critical groups that are often instrumental in land development within the WUI: relators, architects, planners, surveyors, landscape architects, and civil engineers. These groups are also critical as they are the licensed professionals that are planning, designing, and certifying the health, safety, and welfare of new subdivisions and community development. Additional groups should also include large landowners, ranchers, energy extraction, health services, religious and nonprofit organizations. Counties should also consider nontraditional groups such as the Sierra Club, outdoor groups and clubs, and garden clubs because these groups have had recent successes in developing landscape codes (Lerch &

PAGE 159

142 Chambers, 2019) . No county documented who would coordinate grant writing, submissions, or future project funding and implementation decisions. A few counties did not include broader participation because it was deemed to be too difficult. El Paso County, CO stated: The standard guidance for developing a CWPP advises a community to first form a core team that includes all local stakeholders. That team then meets frequently to guide every step of the process, from research through draft writing, then public comment and eventual adoption. Some counties have successfully used this approach to develop all County CWPPs. However, El Paso County is large, highly populated, and politically diverse. Its jurisdictions (some of which overlap) inclu de eight municipalities, twenty one fire protection districts, two metropolitan districts with fire departments, five military installations, a state park and a national forest. Some parts of the County have many active citizen groups, while other areas ha ve very few. Its 2,158squaremile area consists of arid plains, dense forests, and high alpine environments, each with different wildfire concerns. A person from the southeastern grasslands would be challenged to represent or understand the interests of someone in the northwestern foothills. As we looked at this complex picture, we quickly saw that a core team that included all stakeholders would be too large and cumbersome to be effective. But to form one that was small enough to be manageable would require selecting representatives who still might not be able to represent all local concerns. (Russell et al., 2011, p. 3, ita lics added for emphasis.) Table 6. 2. Siskiyou County, CA’s list of critical stakeholders (FireSafe Council of Siskiyou County, 2008, pp. 1617) Federal Government U.S. Department of Agriculture (USDA) U.S. Forest Service: Shasta Trinity National Forest Klamath National Forest Modoc National Forest Natural Resources Conservation Service U.S. Fish and Wildlife Service Bureau of Land Management Bureau of Indian Affairs NOAA Fisheries Tribal Karuk Shasta Klamath State Government

PAGE 160

143 California Department of Forests and Fire California Department of Health California Department of Transportation California Department of Fish and Game California Highway Patrol California Department of Water Resources Office of Emergency Services Water Quality Control Board – State and Regional Local Government Siskiyou County Board of Supervisors Siskiyou County Department of Health Services Siskiyou County Chief Administrative Officer Siskiyou County Road Department Siskiyou County Sheriff’s Department Siskiyou County Planning Department Office of Emergency Services Siskiyou County Office of Education Siskiyou County Air Quality Management District Siskiyou County Schools Municipal/Emergency Services Volunteer fire departments City Councils throughout the County Ambulance services Red Cross Industry and Utilities Fruit Growers Supply Co AT&T Timbervest Siskiyou Telephone Roseburg Forest Products Co. Pacific Power and Electric Sierra Pacific Industries Verizon Wireless Timber Products Co. Edge Wireless Union Pacific Railroad Singular Grenada Irrigation District Water/Sewer districts and companies Montague Irrigation District Farmers Ditch Irrigation District Shasta Water Association Scott Valley Irrigation District Local Fire Safe Councils Butte Valley Fire Department Lower Scott River Road Fire Safe Council Copco/Bogus Fire Safe Council McCloud Fire Safe Council Dunsmuir Fire Safe Council Mt. Shasta Area Fire Safe Council Fire Safe Council of Siskiyou County Orleans/Somes Bar Fire Safe Council French Creek Fire Safe Council Quartz Valley Area Fire Safe Council Greater Weed Area Fire Safe Council Rattlesnake Creek Fire Safe Council Happy Camp Fire Safe Council Salmon River Fire Safe Council Juniper Flat Fire Safe Council Scott Bar Fire Safe Council Klamath River Fire Safe Council Scott Valley Fire Safe Council Lake Shastina Fire Safe Council Seiad Valley Fire Safe Council

PAGE 161

144 While garnering participation of a wide group of stakeholders is challenging, the purposeful exclusion of diverse stakeholders in participation to avoid conflicts is at the expense of creating a contextually responsive CWPP. The subversion of the participa tory process ensures that community values are not fully represented in the framing, goals, or objectives, and also potentially creates community barriers to successful implementation. Finally, participation is a potential learning opportunity that can dee pen citizens’ understanding and engagement in wildfire risk reduction. Furthermore, strategies exist to potentially overcome the participation barriers that El Paso, CO and other counties experience: 1) create a generalized county plan that fully engaged t he community, but then leave specific fire districts to create their own localized CWPP, or 2) provide an overarching framework through the county plan and embed several specific plans for the contextual community, district, or plant community. Participation recruitment and engagement processes were documented by 61 counties. Many counties used multiple recruitment strategies, including flyers, news releases, television and radio ads, website presence, blogs, key informant and stakeholder networking, YouTube, door to door soliciting, bus ads, and informational booths at community events. Asotin County, Washington included examples of their participation recruitment information so that future efforts can learn from successful and unsuccessful strategies. An example is included in Figure 6.3 below . Asotin County even included all the meeting minutes, presentation slides, and handout materials used at each event.

PAGE 162

145 Figure 6. 3. Asotin County, Washington public meeting announcement (Northwest Management Inc., 2008, p. 27) .

PAGE 163

146 T he roles and responsibilities of key stakeholders were documented by 30 counties. Documenting responsibilities of each key stakeholder is critical to ensuring action and accountability in creating and implementing CWPPs. Counties provided one of two forms of documentation: a narrative or organizational t able/chart. Bent County, Colorado provides an excellent example of a narrative description: Participants since the onset through direct meeting participation and/or email notifications and updates include the Bent County Sheriff, the Bent County Commission ers, the Bent County Office of Emergency Management, and representatives from Hasty/McClave Rural Fire Protection District, and Las Animas/Bent County Fire Protection District. These same individuals also represent their communities. Federal and state partners notified or that have participated over time include the U.S. Army Corps of Engineers John Martin Dam, Natural Resource Conservation Service, the Bureau of Land Management, John Martin Reservoir State Park, Colorado Division of Wildlife, and Colorado State Forest Service. Assisting with the mapping to jpg format was the Bent County Assessor’s Office. (Participants include or have included Gerry Oyen, David Encinias, Tom Wallace, Randy Freed, Clay Hasser, Julie Davis, Tandy Hasser, Karen Downey, Darrel Six, John Merson, Steve Keefer, Mike Smith, Ed Skerjanec, Fran Pannebaker, Donna Davis, NRCS lead Working Partners Group. Recent dates include – 1/21/10 (with AWOP meeting; 3/22/10 map review; 4/28 mapping w/county assessor office; 10/26/10 Maps & Firewis e ; 1/13/11 – CWPP draft review & 2011 Annual Plan). (Bent County Commissioners, 2011, pp. 45) El Paso County, Colorado provided details of a more lim ited steering committee, but they still completed a thorough job in communicating responsibilities and obligations with the CWPP team (Figure 6.4).

PAGE 164

147 Figure 6. 4. El Paso County, Colorado steering team responsibilities (Russell et al., 2011, p. 3) . More complicated organizational structures, such as Montrose County, Colorado, used tables to documented people, organization affiliations, and roles and responsibilities (Table 6.3). In cases like this one, it would be helpful to also have an organization flowchart to document how these participants interact with the decision making process.

PAGE 165

148 Table 6. 3. Excerpt (not a complete listing of participants) from Montrose County, CO CWPP development team’s roles and responsibilities (Anchor Point Group & AMEC Earth and Environmental, 2011b, pp. 911) . Name Organization Roles/Responsibilities Ike Holland, Emergency Manager Montrose County Primary point of contact and decision making, emergency response Rick Dunlap, Sheriff Greg Thorton Tad Rowan, Fire Chief Montrose Fire Protection District Community risk and value approval, development of community protection priorities, and prioritization of fuel treatment project areas and methods. Provided previous fuels treatment data. D ecision making process es were documented by 80 counties . Only 48 counties document ed implementation dates and timelines. The decision making processes documented were as follows: 1) driven by consultants and reviewed by core team member signatories, 2) prioritization by core team member, 3) prioritization by core team and supported by lim ited community engagement, 4) prioritization through a full participatory decisionmaking process, and 5) assignment of all projects within a certain risk zone the same priority (e.g., extreme wildfire risk areas with projects received the highest priority rating). Many counties used qualitative scoring ranging from high to low priorities. Most counties used the wildfire risk method to prioritize projects, which is alarming because this approach is shown to be seriously flawed because results have been coun ties with hundreds of projects that are all high priority and no readily available way to prioritize within this grouping. It is especially concerning because, as discussed above, many counties are inadequately funded and, therefore, have limited capacity to implement projects lists of this size. Thus, it is imperative to create a more nuanced mechanism to determine priorities. Counties that use the wildfire risk method should take cues from other

PAGE 166

149 counties that used a mix of core team member prioritization with limited public engagement. Prioritization tactics included Likert scale surveys, focus groups, community meeting charrettes and subsequent voting, per person scale of impact, and core team member discussion and rankings. Fremont County, WY utilized a hybrid approach that allowed space for discretionary decisions: The action items recommended in this chapter were prioritized through a group discussion and voting process. The action items in Tables 6.1 – 6.4 are ranked as “High”, “Moderate”, or “Low” pr iorities for Fremont County as a whole. The CWPP committee does not want to restrict funding to only those projects that are high priority because what may be a high priority for a specific community may not be a high priority at the county level. Regardle ss, the project may be just what the community needs to mitigate disaster. The flexibility to fund a variety of diverse projects based on varying criteria is a necessity for a functional mitigation program at the county and community level. (Northwest Management Inc., 2014, p. 140) Fremont County’s efforts are especially notable because their priorities span policy and planning effort s, fire prevention and education projects, infrastructure enhancements, resource and capability enhancements, proposed mitigation projects, and regional land management recommendations with estimated dates and timelines (Figure 6.5).

PAGE 167

150 Figure 6. 5. Fremont County, WY proposed project area priorities and timelines, excerpt (Northwest Management Inc., 2014, p. 145) . Forty four counties outline a CWPP update timeframe and process, and these processes fell into two broad categories: CWPP updates and CWPP implementation project updates. CWPP updates were described as either fiveor ten year cycles. However, despite 44 CWPPs suggesting update these timeframes, only 30 have been updated. Of these 30 updates, 7 were

PAGE 168

151 short time frame, quick error fixes (e.g., original created in 2004 and updated in 2006) and are currently overdue for another update. Over 75% of t he counties noted the need to update the core team, public, and CWPP to reflect yearly changes to project implementations. They also noted these updates need to be published in a public manner as either 1) dynamic appendices to the CWPP or 2) website host used for the CWPP. Further investigation revealed that only a handful of counties have maintained this yearly update cycle. Only six counties document social and economic vulnerabilities and how they plan to overcome them. The low number of counties frami ng vulnerabilities is not surprising because vulnerability is not yet widely accepted in the wildfire risk modeling world and nuances in vulnerability are difficult to ascertain in remote, rural census geographies (H. Brenkert smith et al., 2017; Chakraborty, Tobin Graham, & Montz Burrell, 2005; T. W. Collins, 2011; Cutter, Boruff, & Shirley, 2003; Cutter & Finch, 2008; Gaither et al., 2011; Lynn, 2003) . Curry County, Oregon provides a unique subsection within their CWPP entitled “3.5 Income, Poverty, and Special Needs” because they acknowledge, “In addition to an aging population, Curry County has a higher proportion of residents with special needs or experiencing poverty compared to the state” (Lynn, Ojerio, & Wolf, 2008, pp. 310) . Curry County’s economic status is categorized as distressed. Additionally, Curry County also documented citizens with unique needs that can limit evacuation mobility and engagement in personal wildfire mitigation efforts: The U.S. Census indicates that as of 2000, 28% of Curry County residents ages five and older had a disability. The same year statewide disability status was at 18.8%. According to the Census Bureau, citizens are considered to have a disability if they have one of the following conditions: a) a sensory disability such as deafness, blindness or significant impairment, or b) a physical disability that significantly limits their ability to perform basic physical activities, such as walking, lifting or carrying. As the median age in Curry County increases as the baby boomer

PAGE 169

152 generation ages, the number and percent of residents with a disability is likely to increase (Lynn et al., 2008, pp. 311) . However, Curry County did not suggest specific strategies to overcome these concerns or direct future development away from higher risk areas to accommodate citizens with special needs. Forty five CWPPs highlighted sustained engagement. Conversely, the remaining counties were a mix of one off engagement opportunities or simply did not engage with the community. Engagement efforts include both passive and active strategies. Passive strateg ies include advertising campaigns, press releases, and a web presence. Active strategies include public outreach events, field days (Figure 6.6), surveys, and ongoing community meetings regarding CWPP implementation. Four counties documented key community leaders that could serve as catalysts for sustained implementation engagement. A handful of counties acknowledged the need for future leader, and thus solicited future advocates (Figure 6.7), a strategy that should prove useful for counties that do not have a robust network of engaged community members. Ideally, as four counties documented, these stakeholders should have access to multiple social networks to leverage a more expansive organizational and grass roots engagement and implementation effort.

PAGE 170

153 Figure 6. 6. Benewah County, Idaho Forest Owners Field Day Announcement (Schlosser & Mierzwinski, 2012, p. 47)

PAGE 171

154 Figure 6. 7. Gunnison County, Colorado letter soliciting participation as a community wildfire mitigation advocate (WMA) (Anchor Point Group & AMEC Earth and Environmental, 2011a, p. B21) .

PAGE 172

155 Base Map Every county received a zero for their base mapping efforts. This is surprising because base maps should be the least controversial and debatable element of a CWPP. However, base maps proved to woefully inadequate. One county, Crook County, Wyoming provided no base maps at all, despite HFRA and CWPP strongly incentivizing their inclusion (Jakes et al., 2007; Jakes et al., 2012; One Hundred Eighth Congress of the United S tates of America, 2003; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004; The White House, 2003) . Another county’s maps, Harney County, Oregon, were illegible (Figure 6.8). The remaining counties provided little more than a single map that, at the least, provided community locations and major roads (Figure 6.9) and, at most, provided community locations, cursory WUI designations, major roads, fire district boundaries, and land ownership (Figure 6.10). Base mapping efforts do not need to be integrated into a single map or even a singlesection but could be comprised of many maps, logically spread throughout the CWPP. In order to adequately document risk and vulnerable populations within each county’s WUI, base mapping efforts should include inhabited areas; critical community infrastructure – e.g. hospital s, nursing homes, fire stations, emergency shelters, water availability and supplies, evacuation routes; and a preliminary WUI designation, and utility corridors (P. Berke et al., 2015; Samuel D Brody, 2003; Cox, 2006) . For example, a base ma p documenting emergency management should show the locations of critical fire suppression infrastructure (e.g., known water sources and capacities, fire station locations and serviceshed times), fuel breaks, hospitals, evacuation routes, dwellings, hospit als, and possible evacuation locations.

PAGE 173

156 Figure 6. 8. Harney County, Oregon CWPP base map (Barker & Glenn, 2013) .

PAGE 174

157 Figure 6. 9. Bingham County, Idaho CWPP base map (Covington, 2003) .

PAGE 175

158 Figure 6. 10 . Cochise County, Arizona CWPP base map (Cochise County Board of Supervisors, 2014) . Even more frustrating is that the need for expanded base mapping was acknowledged in their CWPPs, yet the counties did not complete it for the inclusion in the CWPP nor did they publish updates on these efforts . For example , Harney County, Oregon document ed how critical this information is: As stated throughout this plan, the process of developing a CWPP will help Harney County clarify and ref ine its priorities for the protection of life; property; critical infrastructure; significant recreation and scenic areas; and landscapes of historical, economic, or cultural value in the countywide WUI (Barker & Glenn, 2013, p. 1) Yet, they did not provide these elements in their base maps, even though they stated it was critical to do so:

PAGE 176

159 Community Base Map A community base map should be developed that may illustrate important features such as landownership, structures, roads, surface water, fire districts, or major utility corridors. The map's importance is that it illustrates community values from which recommendations concerning wildfire planning can occur (Barker & Glenn, 2013, p. 1) . As Harney County, Oregon acknowledged, knowing where these resources are is i mportant to pre fire planning efforts, but it is also important because it can direct citizens evacuation planning, support incident command activities, and orient nonlocal suppression support personal to county contexts. Additionally, it can provide transparency in why pre fire planning and incident command makes certain decisions. Risk Assessment Three counties received passing grades (C) for risk assessment. Most failures were due to three reasons: 1) assumed populations experienced hazard exposure and risk equally, 2) ignored continued development and changing community contexts contributions to evolving levels of risk, and 3) failed to acknowledge the evolving biophysical and climatic characteristics increasing risk. WUI boundary was document and defi ned by 93 counties. The 27 counties that did not, submitted a CWPP that failed to address a core tenant of HFRA CWPP s (Jakes et al., 2007; Jakes et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004) . Counties that received failing grades simply created risk maps, but they never identified where thei r WUI is located or what a WUI is. When counties provided definitions of the WUI in the CWPP, the definitions fell into three categories: Federal Registrar, simplistic, and complex. Counties that used the Federal Registrar simply adopted the Federal Regis trar’s WUI definition and applied it to their context, creating a WUI map. Simplistic approaches, such as Cibola County, New Mexico, defined the

PAGE 177

160 WUI as presence of housing within fire prone ecosystems or simply designated the whole county: Because of the rural nature of Cibola County along with its large contiguous blocks of wildland interlaced with low density subdivisions and scattered homes, the core group decided to designate the entire county as the Wildland Urban Interface (SEC Inc, 2006, p. 6) . This is inherently a poor tactic because it does not allow for land use planning and management, developing lesser risk areas over higher risk areas. C omplex WUI definitions used a multi factor approach to define the WUI, which were often a combination of wildfire risk modeling, presence of structures, presence of population, and a locally contextual buffer of development. For example, Boulder County, Colorado used fire risk modeling to define their WUI. Modeling eff orts used FlamMap to create a wildfire intensity map (crown fire potential and flame length), wildfire occurrence (conditional burn probability), community values at risk, wildfire hazard (likelihood of wildfire occurrence plus the anticipated fire intensi ty), major fire paths (computer simulation), and wildfire areas of concern. Boulder County used experts, community surveys, and focus groups to determine which community values were important and how important they were (Table 6. 4). It is worth noting that communities and structures were defined as areas with a structure density of greater than 64 structures per square mile. However, it is important to point out that missing from Boulder County’s community values at risk was recreational or other cultural s ites, which is particularly important given Boulder’s longterm investment in parks and openspace. The results are presented in Figure 6.11.

PAGE 178

161 Table 6. 4. Boulder County, CO community values at risk and weights of importance (Boulder County Board of County Commiss ioners, 2011, p. 66) . Values Weight of importance Communities 25 Homes 18 Water supply zones 17 Priority historical sites 15 Key ecological areas 14 Roads and railroads 11 Figure 6. 11 . Boulder County, CO wildfire risk assessment map results (Boulder County Board of County Commissioners, 2011, p. 67) The sophistication of the WUI definition presented in the CWPP was often related to the type consultant or experience of the planning department. Environmental scientists and geospatial consultants or planning departments with access to in house scientists used

PAGE 179

162 sophisticated modeling to define WUIs while planners and grassroots efforts typically relied on a community values based approach to defining the WUI or simply used whole county designations. Only one county, Campbell County, Wyoming, documented the demographic and economic context of risk over time. However, Campbell County did not necessarily document the traditional social changes (e.g., income, age, or disabilities), but instead focused on the difficulties in addressing explosive, unregulated growth in industrial infrastructure. Indeed, Campbell County stated, Several booms have taken place since the oil boom of the late 70's and early 80's. Along with oil, coal and most recently coal bed methane, Campbell County has, and is, experiencing dramati c population growth. Developments and subdivisions seem to be appearing over night with little regard to wildland fuels within and surrounding these new wildland urban interfaces. The wildland urban interface (WUI) is commonly described as the zone where structures and other human development meet and intermingles with undeveloped wildland or vegetative fuels. The WUI poses tremendous risks to life, property, and infrastructure and comprises one of the most dangerous and complicated situations firefighter s face. (Campbell County Fire Warden, p. 2) The county documented the link between increases in development and increases in ignitions and highlighted the danger posed by unplanned development: Ignitions related to development and population can be projected to increase correspondingly to growth as evidenced with fires ignited by railroads in the past. Initial Attack response will vary but on the whole will average between 30 60 minutes and are exasperated by a hodge podge system of roads accessing a multitude of developments. The complexity of wildfire incidents in the county are increasing proportionately to the increase in development and population. (Campbell County Fire Warden, p. 4) These issues are concerning due to the long fire suppression response times and the potential issues of evacuatin g volatile industrial operations during extreme wildfire events.

PAGE 180

163 Ninety five counties documented differing jurisdictional boundaries. However, 25 counties did not, which means they failed at providing a core HFRA CWPP component (Jake s et al., 2012; Resource Innovations Institute for a Sustainable Environment, 2008; Society of American Foresters, 2004) . Many counties simply included a land ownership map of the three key signatories. However, the best examples also documented the land ownerships t hat made their county wildfire issues unique, such as Campbell County, Wyoming. Campbell county included key private landowners and industrial land uses (Figure 6.12).

PAGE 181

164 Figure 6. 12 . Campbell County CWPP industrial development and land ownerships (Campbell County Fire Warden, p. 9).

PAGE 182

165 Seventytwo counties thoroughly documented the location of priority fuels reduction projects. Documentation took on two forms: narrative or narrative and maps. Albany County, Wyoming, provide a narrative similar to other narrative documentation that is often included with a thorough description and occasional photos; however, a map would have been helpful for nonlocals: Wold Tract Treat fuels along evacuation roads by thinning, pruning and slash disposal, dispose of slash within one mile of private property, thin reproduction in old cutting units, the to 40% canopy cover in the medium sized stands adjacent to the community and thin along road sides nearby. Cooperate with community to designate two evacuation routes through the national forest land back onto state highway 230 (Anchor Point Group, 2013, p. 6) . Okanogan County, Washington provides a great example of project description ranking and mapped project locations (Figure 6.13). The project list, type, acres, structures, miles, priority, and map location are beneficial to assess the impact and amount of efforts needed to implement. The only thing missing is a timeframe for the completion of efforts.

PAGE 183

166 Figure 6. 13 . Okanogan County, WA priority fuels reduction projects and map locations. Eightynine counties thoroughly docum ented the nonstatic nature of risk assessment and how seasonal, biophysical, community context changes risk over time. However, they did not identify the triggers that would cause the creation of an update to the risk map, and no counties documented their intended update cycle. However, 90 counties thoroughly documented the key factors, definitions, and process used to assign level of risk; the level of risk categorization utilized for these passing counties involved a minimum of three categories, and a few counties

PAGE 184

167 documented four categories of risk. Risk was assigned using one of two approaches: 1) state assessments augmented by field assessments, such as the Wildfire Hazard Information Extraction model or 2) contextually augmented GIS models . It is important to note that only a handful of counties used field assessments, and an example from Crook County, Wyoming can be found in Appendix G. GIS risk modeling efforts either used consultant proprietary modeling methods or accepted risk modeling best practice s, such as FARSITE and FlamMAP. These modeling efforts were often augmented by threats to community values. Boulder County, Colorado’s risk assessment, as described above, is a well documented example of integrating FlamMAP and county values (Figure 6.11). Counties failed risk assessment because 1) they either simply equated housing within unincorporated areas of the county as “ at risk ” , 2) had single risk categories, or 3) provided no documentation defining the i r three levels of risk. The biophysical nature of risk changes of time were documented by only seven counties documented. However, no county outlined plans or timetables to update risk mapping efforts despite their acknowledgements of change. Only four counties documented the relationship between r isk and vulnerable populations. Shoshone County, Idaho provided extensive documentation of a 1,100 acre heavy metal and mining superfund site that poses a threat to children and other vulnerable populations because, “the extent and nature of the cleanup that has occurred and is currently ongoing at the Bunker Hill Superfund Site present special considerations for Shoshone County” (Shoshone County Community Wildfire Protection Plan Committee, 2011, p. 21) . However, these counties did not discuss how to engage or assist vul nerable populations in overcoming challenges to riskreduction efforts. Furthermore, eight counties acknowledged climate changes effects on future

PAGE 185

168 risk, but they did not provide a course of action to integrate alternative climate futures and their effects on risk. Indeed, Nevada County, California stated, In addition to the vegetation and all of the anthropogenic impacts that have degraded natural fire regimes, climate change has also played an extensive role in altering fire occurrence and severity. Clima te change has influenced the vegetative cover and available burnable fuel across the Western landscape. In the past few years, fires have grown to record sizes, are burning earlier and longer, and are burning hotter and more intensely than they have in the past (Westerling et al. 2006). (TSS Consultants, pp. 1617) . Yet Nevada County did not include a future risk assessment or plans to do so. However, it is unclear why the remaining 112 counties did not include the changing nature of risk at all. Research would suggest the lack of climate change acknowledgement could be due to the politicized nature of climate change discussions and the barriers they create when implementing policy in predominatel y conservative, rural communities (Fisher, 2006; McCright & Dunlap, 2011) . Hazardous Fuels Reduction Thirty five counties received a passing grade (A) for hazardous fuels reduction. Mohave County, Arizona provided a thorough documentation of fuel modification and treatme nt plans, including the treatment categories of vegetation and slash across nine different treatment classifications and several treatment categories (Appendix H). Treatment category classifications included developed parcels less than two acres; undeveloped private parcels or single structure parcels greater than two acres; grassland fire breaks; oak/pinyon/juniper shrublands within the WUI; prescribed fire; escape and resource transportation corridors (federal and nonfederal lands); riparian areas (feder al, non federal, and private lands); conditional suppression areas (federal and nonfederal lands); and saltcedar removal for restoration purposes (federal and nonfederal lands). Treatment options for developed private parcels less than two acres were

PAGE 186

169 cat egorized into four home ignition zones of treatment, 010, 030, 30100, and 100600 feet respectively. Treatment options for undeveloped private parcels or single structure parcels greater than two acres as well as grassland fire breaks were categorized i nto two classifications: slopes less than 20% and slopes greater than or equal to 20%. Finally, oak/pinyon/juniper shrublands within the WUI were categorized into two treatment types: landscape treatment outside the firebreaks and within the firebreaks. C ounties earned failing grades this section of the CWPP process instrument scoring for at least one of the following four reasons. First, some counties only included single treatment types, such as defensible space. Second, counties did not identify speci fic locations or priorities for hazardous fuel reduction. Third, failing counties sol el y focused on either private or public lands and neglected the type of land that was not of focus. And finally, failing counties had no documentation or tracking of acr es needing treatment , which is surprising in a wildfire mitigation document . Reducing Structural Ignitability One county, Minidoka County, Idaho received a passing score (B) for reducing structural ignitability. The remainder of the counties received failing grades. Reasons for failing this section of the CWPP process instrument was due, in part, to not providing essential informat ion that should be readily available to the county. For example, the number of fires, their causes, and associated economic and home losses were documented by 73 counties. However, no county documented the number of homes saved from a fire. Documenting th e amount of structures saved and lost are integral to increasing landowner participation and highlight the benefits of hazardous fuel reduction within the home ignition zone and the utilization of Firewise building practices.

PAGE 187

170 Only 44 counties outlined cur rent applicable WUI wildfire codes and regulations. While these counties documented the best practices thoroughly in their CWPP, they often did not integrate them into enforceable code. The few counties that have created building codes ensure compliance by using subdivision design review and building inspections or permitting processes. While the counties often documented issues of noncompliant structures, they had no strategies or processes in place to reduce their ignitibility. Ten counties documented WUI growth trends and their methodology for assessing these trends. WUI growth was documented using the following: change in geographical WUI extent, number of additional dwellings within the WUI, and WUI population increase. But no county decided to preclude development from highrisk zones in order to reduce the structural ignitibility of future developments. Two counties outline WUI growth control strategies to direct WUI development away from highrisk areas. These counties recom mend in an abstract sense, the use of growth boundaries, open space, and zoning, but provide no regulations, strategies, or actions to integrate them into existing code. The counties receiving non passing grades earned those grades because they did not fol low structural ignitibility best practices, such as requiring fire safe roofs and walls, and allowed wood balcony and decks. It is also worth noting that no county has a plan to address existing, exempted structures that do not meet current code. Education and Outreach Nine counties received passing scores (C) for education and outreach: Boulder County, Colorado, Del Norte County, California, Madison County, Montana, and Siskiyou County, California. While all CWPPs provided or linked to educational materia l, such as brochures, websites, blogs, YouTube videos, or pamphlets documenting defensible space or ways to reduce structural ignitibility, no counties provided CWPPs translated into other languages. This is

PAGE 188

171 concerning given that many of these counties, pa rticular those in the southern American West, have high concentrations of Spanish speakers. Fortunately, educational material Californian counties is translated into Spanish because of CalFire multi lingual outreach efforts; however, this process of transl ation is ongoing and not available for all materials yet. Counties should be aware of potential language barriers and make efforts to reach nonEnglish speakers, in written material, community meetings, and emergency alerts. Deep public engagement, includ ing a wide range of county and state supported activities were provided by 87 counties. Efforts included public meetings, field trips, news casts, demonstration projects, school events, booths at farmers markets, household visits, youth workdays, Firewise workshops, and other community events. However, only 20 counties provided home planning preparedness checklist and guidelines (Appendix I) and personal evacuation route planning checklists (Appendix J) in their CWPP. This is particularly concerning given r epeated issues with evacuation preparedness (Lucas, 2018; McCaffrey, Rhodes, & Stidham, 2015; M. Moritz & Anderson, 2018) . Only 10 counties documented changing attitudes, awareness, and participation in humancaused wildfires and increased participation in fuel reduction efforts or woody debris disposal. Therefore, 110 counties are not fully aware of the effectiveness of past education and outreach efforts. All 10 of the counties that documented changing attitudes did so through ongoing surveys and anecdotal evidence fro m community events. To broaden outreach and education Otero County, New Mexico created a Facebook page and administered a survey, which showed that 65% of respondents felt their homes were at moderate to high risk from wildfire (SWCA Environmental Consulta nts, 2014). Additionally, only 16% felt their community was well prepared for a wildfire (SWCA Environmental Consultants, 2014). While 40% of homeowners

PAGE 189

172 thought they were well prepared, while the majority felt their neighbors were not (SWCA Environmental Consultants, 2014) . While 40% thought their own homes were prepared, the majority felt their neighbors were not (SWCA Environmental Consultants, 2014) . As Otero County suggests, additional outreach is needed to address the discrepancy between individual, neighborhood, and community wildfire preparedness. Emergency Management Capacity No county received a passi ng grade for emergency management capacity because : only 10 counties were able to document the percentage or number of homes and their locations in each fire district ; only 11 counties noted adequate incident command training; only 11 counties addressed animal agriculture evacuation, and only 4 counties addressed notification and evacuation planning and testing. Knowing the locations of homes is important because these locati ons are critical for incident command to develop points of control during a fire event and to support wildfire evacuation notifications. Counties that do not have the capacity to generate and maintain this information on their own would benefit from using the national building dataset, augmenting it with parcel contact information and rural addressing initiatives. It is worth noting that many counties with these issues did document rural addressing initiatives as critical to their CWPP efforts. Seventynine counties documented their fire suppression capacities; however , many counties noted their reliance on state and federal support due the lack of economic resources to expand their own capacities (Covington, 2003; Northwest Management Inc. & Adams County Wildfire Urban Interface Wildfire Mitigation Plan Committee, 2004; Western Washington University Huxley College, 2012) . In the absence of existing capacity documentation, many counties listed suppression needs, which included communication equipment and infrastructure,

PAGE 190

173 retention of vol unteer firefighters, mapping, training, update and creation of new facilities, and repairing, modernizing, and increasing the amount of suppression equipment. Every county had at least one incident command trained firefighter; however, all counties also expressed the need for more command trained firefighters and the need for funding ongoing training efforts. Boulder County, Colorado exemplified how to address animal and livestock preparedness and evacuation plans through the InterMountain Alliance (Boulder County Board of County Commissioners, 2011) . Boulder County planned to evacuate large animals to the County Fair Grounds and if this location is unsafe to surrounding county fairgrounds. Additionally, they had procedures in place to ensure animals have access to water and food. All counties could should also update CWPPs with external post fire agriculture recovery resources. Resource links should include federal, state, and county agriculture extension efforts to support short term recovery of agriculture economies and the long term recovery of grazing lands. M atthew Shapero (2018) provides a model of how counties and ranchers can better prepare for wildfire before, during, and after a fire event. Only four counties documented evacuati on plans, safety zones, vulnerable populations, community fire and personal communication systems, notifications, and evacuation tests. Lemhi County, Idaho provided model wildfire evacuation guidelines on a neighborhood basis (Appendix K). While every fire presents unique challenges, it is imperative that systems are in place and tested to ensure reliability and capacity. For example, preliminary reports from the Camp Fire in Paradise, California, suggested that deaths were the result of inadequate evacua tion route capacity and a failed phone based alert system (Hart, 2018) . Forty one county CWPPs documented integrat ing wildfire efforts into other hazard mitigation plans in order to coordinate and maximize planning capacity, potential grant

PAGE 191

174 resources, and various levels of state and federal support. Integration efforts included the national fire plan, multihazard mit igation plans, FEMA hazard management efforts, regional coordination efforts, and local guidelines. Judith Basin provided an extensive example list within the introduction of their CWPP (Northwest Management Inc., 2004) . Longterm Success Thirty counties received passing grades for long term success. Therefore, it is useful to discuss the individual best practices in this longterm success category to explore where CWPPs were on the right track to success and where the CWPPs were lacking. First, 58 counties documented CWPP projects that could be accomplished quickly to foster homeowner buyin and broaden support for long term efforts. Most of these counties broke their planning efforts into short (yearly), medium (5 years) and longterm (5+ years) plans. Yearly plans included education and outreach; rules, restrictions, ordinances, and enforcement; prevention planning; fuels mitigation; an d grants and administration. Counties that planned the minimum effort to fulfill this long term planning initiative generally focused on education and outreach activities. One such example is Prowers County, Colorado. This county detailed three education and outreach activities with specific timelines and implementation details, such as: Share the CWPP and Mitigation Assessment Maps with the community at large. Who Donna Davis & Team What Attend meetings to share Where Communities, fire departments, conservation districts When 2nd & 3rd Quarter (minimum three this year) Costs – TBD (Prowers County Commissioners, 2013, p. 28) Re source availability seems to be a concern in planning and implementing plans for longterm success. Prowers County (and many other counties) , acknowledged that they lack the resources to implement most of their yearly, medium, and longterm plans :

PAGE 192

175 There i s no funding for this plan at this time. Potential grant applications may be considerations (For example, an IMR National Fire Plan Community Assistance Grant from the Department of Interior, National Park Service may become available for application) (Prowers County Commissioners, 2013, p. 28) . One way that counties can overcome a lack of resources is coordinating their CWPP efforts with other related plans, planning activities, or infrastructure development (Samuel D Brody, 2003; R. Burby & Deyle, 2000; Fleeger, 2008; Jakes et al., 2012) . However, only 38% of the counties integrate their CWPPs into other plans. Typical county integration activities include, coordinating with local CWPPs, local fire plans, and annual county work plans. Additional plans that counties may consider are FEMA multi hazard mitigation plans, comprehensive planning, and federal and state wildfire efforts. An aspect of the long term planning that was particularly concerning is that only eight counties expressed a need to integrate CWPPs into county governance. This integration is paramount to enacting a number of mitigation activities and also regulating development in the WUI. Of these eight counties that did integrate their CWPPs, the governance plans focused on integrating county plans into local land use ordinances and state codes. For example, Benewah County, Idaho integrated wildfire planning activities into a broad array of regulatory tools: plans, policies, and programs, as outlined in Table 6.5. Benewah , used this same approach across all levels of governance, state and federal.

PAGE 193

176 Table 6. 5. Benewah County, ID’s legal and regulatory resources available for wildfire mitigation efforts (Schlosser & Mierzwinski, 2012, p. 34) Regulatory tool Name Description (effect on hazard mitigation) Hazards addressed Mitigation, prepar edness, response, or recovery Affects development in hazard areas? Plans Benewah county emergency operations plan Defines responsibilities All Preparedness and response No Flood response plan Identifies actions related to flood activities and safety of responders and the public Flood Preparedness and response No County comprehensiv e plan Defines level of importance All All Yes County wildfire mitigation plan Identifies threats and hazard mitigation activities for wildfire Wildfire All No Policies Zoning ordinance Identifies land use locations All Mitigation and preparedness Yes Subdivision ordinance Specifies densities All Mitigation and preparedness Yes Floodplain ordinance Identifies restricted or controlled areas Flood Mitigation, preparedness, and recovery Yes Site disturbance ordinance Controls construction disturbance All All Yes Programs County fire mitigation program Reduces threat Wildfire Mitigation and preparedness No

PAGE 194

177 The counties without governance integration plans provided utilizing reasoning for the omission similar to that of Calaveras County, California. Calaveras County expressly stated: This C.W.P.P. is not a legal document and is not intended to be an all encom passing document in regards to fire planning and management in Calaveras County. Additionally, this C.W.P.P. does not satisfy any regulatory permitting process, including CEQA analysis for any project proposed within. This plan recommends both general and specific projects, all of which are subject to the appropriate permitting and environmental review for the county in which they are proposed. Any public projects identified or proposed in this C.W.P.P. will be done only as funding allows. There is huge var iation in vegetation, weather conditions, geography and access throughout Calaveras County. There are also numerous government jurisdictions, with differing interests. Because of this, the discussions and recommendations in this C.W.P.P. have remained gene ral in nature, as to not conflict with stakeholder interests. Because this plan is a flexible planning tool, rather than a blueprint, general guidelines allow the project proponent to develop the most appropriate methods available for fuels treatment (Fullerton Management Group, 2011, p. 17) Th e mentality expressed by Calaveras County is concerning because they purposefully avoid conflict, whic h sacrificed deeper discussions to address the public safety issue of wildfire. Instead of taking this anti conflict route, counties should embrace what Innes and Booher (1999) describe as collaborative rationality to solve prob lems because it is more likely to result in feasible, meaningful, and legitimate decisions. The National Fire Protection Association’s Firewise program played a pivotal role in how counties encouraged local, non governmental engagement in adapting to living with wildfire. One hundred counties discussed the importance of Firewise certified communities or the creation of fire safe councils and the importance of these councils in the multi jurisdictional nature of managing wildfire in the WUI. For counties, Firewise served as a way to motivate and support voluntary citizen efforts in neighborhoods or small communities in order to reduce fuels and pre pare homes for wildfire.

PAGE 195

178 Local Governance Document Analysis Scores and Results As with the CWPP process score, the local governance document analysis scores were lower than what was hoped at the start of this project. Only eight counties out of the 120 total counties received passing grades. These counties were Bannock County, Idaho (A), San Diego County, California (B), Jefferson County, Colorado (C), Silver Bow County, MT (C); Trinity County, California (D), El Paso County, Colorado (D), Valley County, Idaho (D), and Nevada County, California (D). There is no discernable pattern of sociodemographics to suggest why some counties received higher test scores than others. While disappointing, these low grades were not a complete surprise due to the compl exities surrounding county land use planning efforts. California had a higher cumulative average score than any other states on average across all three sampled strata. This is not surprising given California’s emphasis on minimum state wildfire building standards, wildfire risk mitigation policies, CalFire’s active engagement with inspecting defensible space, California’s market wildfire incentives and disincentives, and strong statemandated local planning efforts. However, despite California’s longstan ding traditions of comprehensive and ecological planning and fire research, the state as a whole has failed to implement acceptable local governance best practices in order to mitigate wildfire risk. This is particularly concerning because California budge t and planning capacity is much higher than that of much of the American West. What is even more concerning is that California state law requires land use planning in very high fire hazard severity zones be reviewed and updated, utilizing the references of the most recent version of California’s “Fire Hazard Planning” (Gover nor's Office of Planning and Research, 2014) , provisions of SB 1241 ("SB1241. Kehoe Land use: general plan: safety element: fire hazard impacts.," 2012) , and other CalFire material

PAGE 196

179 (CAL FIRE, 2012) . If this land use planning has suff iciently utilized these materials, which fulfills local governance best practices, then the California county scores should have been passable scores across the board. Planning capacity is a grave concern in many counties in the American West, as over 75% of the counties that were studied have minimal planning departments (e.g. a planning department of less than two) to no planning departments or staff at all . Indeed, seven counties in Montana, Wyoming, Arizona, New Mexico, Washington, and Idaho had no z oning or building codes at all. Three counties went as far as to implement a “Code of the West,” in which they state that counties cannot and will not provide the same level of services in rural portions of the county due to their remote nature and the inhabitants’ culture of self reliance and minimal regulation (Lincoln County Board of Commissioners, 2008) . Minimal regulation means that these same counties continue to allow new development within these remote rural locations while refusing to provide necessary services or using land use and building controls to ensure Firewise development. This unregulated and unadvisable development is effectively subsidized by state and federal wildfire suppression efforts during wildfire events. Comprehensive Plans All but 28 counties implemented comprehensive or general plans. Ten counties scored 100% on their comprehensive plans: Bannock County, Idaho; Blaine County, Idaho; Glenn County, California; Jefferson County, Colorado; Kittitas County, Washington; Okanogan County, Washington; Riverside County, California; San Diego County, California; Trinity County, California; and Washoe County, Nevada. Thirteen other counties received passing grades (D): Adams County, Idaho; Costilla County, Colorado; Del Norte County, California; Inyo County, California; Monterey County, California; Nevada County, California; Park County,

PAGE 197

180 Montana; Pondera County, Montana; Power County, Idaho; Silver Bow County, Montana; Taos County, New Mexico; Tulare County, California; and Tuolumne County, California. However, approximately 70% of the comprehensive plans framed planning solely as economic development and land development growth, with no mention of existing wildfire hazards. Indeed, none of the comprehensive plans mentioned or coordinated with the county CWPPs. This result is not surpris ing given that counties with minimal planning infrastructure used land development consultants (e.g., planners, surveyors, civil engineers, land developers, and landscape architects) to draft their comprehensive plans, zoning, building codes, and ordinance s. In many cases, the same consultants were used by numerous counties. In the counties with their own planning departments, these same documents were often coordinated with community development boards and groups, whose only frame of emphasis is population and economic growth. Comprehensive plans with multiple frames – such as economic growth, environmental and cultural stewardship, enhancing quality of life, and health, safety and welfare – exhibited a higher degree of wildfire integration. This was evident in Butte Silver Bow County Growth P olicy , in that they defined several wildfire goals, objectives, and strategies. Specifically, they stated: to reduce the potential r isk to structure s located within the wildland u rban interface, Butte S ilver Bow will evaluate its subdivision ordinance and, as appropriate, adopt necessary regulations to assure that future subdivisions provide for: defensible space around structures; adequate ingress and egress to and from structures and developments to facilitate fire suppression activities; and adequate water supply for fire protection (Community Development Services of Montana, 2008, pp. 55) . However, as seen in the above quotation, Butte Silver Bow County was not restricting or directing development away from high risk areas, nor were they impl ementing a broader array of

PAGE 198

181 best practices. Instead, the county was simply implementing minimum suppression and evacuation standards. Del Norte, California’s comprehensive plan had a subsection specifically designated for fire hazards. It includes a sin gle goal, “To prevent and minimize the risk of loss of life, injury, and property damage result from unwanted fires” (Mintier & Associates, Jones & Stokes Associates, Lowens, & Del Norte County Community Development Department, 2003, pp. 26) with nine specific policies and five existing implementation programs. However, of concern is policy 2.E.3 The County should avoid development in areas identified as high or extreme fire hazard areas when possibl e. Where such development is permitted, structures located in extreme or high fire hazard areas should be constructed with fire resistant materials, utilizing fire resistant design standards, and the surroundings should be irrigated (Mintier & Associates et al., 2003, pp. 26) . It is co ncerning that the county qualifies the policy “when possible” with no further discussion about what this phrase means. Additionally, Del Norte says in policy 2.E.5, “the County should not approve major developments if fire fighting services are not availab le or are not adequate for the area,” (Mintier & Associates et al., 2003, pp. 26) , yet their CWPP documents numerous shortcomings in the county’s fire protection districts: The following table shows the extent of equipment resources currently available to CFPD. One structural engine, those used for structure fires such as homes, is 30 years old and needs to be replaced. CFPD currently houses the only ladder truck in the county and it is over 30 years old and needs to be replaced. They also have a 62 year old fire boat in need of replaceme nt and one rescue vehicle which is more than ten years old and in need of replacement. Fire hose, self contained breathing apparatus, and radio pagers have been identified as other priority needs for CFPD (Katelman, 2005, p. 59) . One wonders how these decisions were made and if the county has the political will to deny development in high risk areas because Del Norte’s WUI grew by 2.6 km2 between 2000 and

PAGE 199

182 2010 (Volker C. Radeloff et al., 2017) , despite policies restricting development in highrisk areas and under capacity fire protection districts. What is particularly concerning is that wildfire risk best practices are not evident in most comprehensive plans, codes or ordinances even though they were created with t he guidance of professionally licensed planners, surveyors, civil engineers, and landscape architects. Embedded within each of these profession’s professional licensure is a code of ethics regarding the health, safety, and welfare of the public. Yet, in wi ldfire prone states, there are no test requirements, liabilities, or responsibilities related to the professional ethics of wildfire risk reduction. This is a glaring omission in the licensure regarding public safety. The possibility is that licensed prof essionals engaged in land development could serve as a mechanism to integrate wildfire best practices within local governance and new land development. States with highrisk wildfire landscapes should require licensed professionals to certify plans meet mi nimum Firewise standards. Finally, comprehensive plans serve as a process and opportunity to document the coordination and integration of local governance activities in achieving a community’s goals and objectives. As such, comprehensive plans should provide strategies and integrate across systems, departments, codes, ordinances, and governance activities to address community concerns. Only 18 counties discussed how their comprehensive planning efforts used open space preservation, conservation, watershed management, or climate change planning to buffer existing and future development from wildfire risk or how to leverage these processes to further reduce wildfire risk . All 18 of these same counties suggested open space and park systems could provide community scale wildland fuel breaks; however, no counties provided specific requirements, strategies,

PAGE 200

183 funding mechanisms, developer incentives, or suggested geographic locations to best implement fuel breaks. Zoning Codes, Development Standards, Subdivision Guidelines, and HOA Requirements Zoning codes, development standards, subdivision guidelines, and HOA requirements did not integrate wildfire risk reduction best practices. Indeed, only ten counties codes received passing scores in this section of the local governance instrument: Silver Bow, Montana (B), Jefferson County, Colorado (B), Del Norte County, California (C), Valley County, Idaho (C), Tulare County, California (C), Trinity County, California (C), Nevada County, California (C), El Paso County, Colorado (C), Clackamas County, Oregon (C), and Bannock County, Idaho (C). The remaining 110 counties received an F. These scores were not surprising given the lack of zoning codes. Most county’s zoning efforts emphasized economic development and land use compatibilities but not hazards or wildfire. Research has shown that wildfire is a land use issue and risk can be reduced through zoning and land development ordinances (Headwaters Ec onomics, 2014) . If counties lack the capacity to update or rewrite their zoning codes, counties could use overlay zoning codes to reduce and management development in higher wildfire risk areas. However, only seven counties implemented a WUI overlay zoning code and development standards. No counties with zoning or development ordinances eliminated or restricted development in extreme or highrisk wildfire areas. Counties approached development in wildfire risk areas by suggesting development proposals un dergo review for wildfire risk and fire mitigation implementation, and these reviews were often under the purview of a planner that focused on wildfire efforts or a certified fire district representative. However, these counties provided no guidelines for development requirements, nor did they identify what the plan review should

PAGE 201

184 include or would entail. Additionally, they did not document how implementation would be monitored or identify the consequences for noncompliance. These glaring omissions cause serious questions about the real practice of the review and the ability to regulate development in these areas at all. Furthermore, counties also omitted the identification of land use regulatory authority processes over wildfire risk even though 50% of the counties mentioned the need for zoning to control land use development in wildfire risk areas. None of these identified land use planning activities that would update codes and ordinances to address risk. A handful of counties abdicated responsibility to a fire district manager to review design drawings; however, it remains unclear if fire district managers have the legal authority or the capacity to review and enforce compliance over large development projects. For example, Park County, Wyoming’s developme nt standards require “proposed development shall demonstrate compliance with recommendations of the local fire district for emergency vehicle access, firefighting water supply, wildfire mitigation, and requirements of the State Fire Marshal, if applicable” (input citation here for Park County Wyoming development standards, page 140). Yet, after searching all of Park County’s fire districts, no such published standards exist. While enforceable state codes exist for water supply, no such codes exist for veget ation mitigation. Zoning and other ordinances did not adequately address the home ignition zones. None of the 120 counties in this study had specific WUI wildfire risk reduction landscape ordinances, yet over a dozen counties had other landscape ordinances (e.g., street tre e requirements, parking lot planting requirements, and incompatible land use buffers). While all but a handful of counties mentioned defensible space or home ignition zone requirements, very few actually codified them or provided specific recommendations w ithin their guidelines. Over half of the recommendations

PAGE 202

185 were for a single or double zone system, despite current best practices recommending at least three zones that are adaptive to topographic conditions. Furthermore, all home ignition zone requirements are truncated at private property lines. This is a particularly worrisome result of private property rights overruling public safety regulation. If developers, planners, and other county/state officials prefer to truncate ignition zone requirements at p rivate property lines, then the zoning and subdivision design guidelines should at least ensure lots that are large enough to implement adequate home ignition zones on many different types of topography. Finally, none of the counties documented that land developments contribut e to elevated fire risk. In other words, more structures in the WUI equal higher fuel loads , and this fact should at least be acknowledged. Fourteen counties identified the use of trails, recreation, and open space areas as community l evel fuel break opportunities. However, only four counties provided specific details on fuel break requirements that adequately provide wildfire risk reduction protection. One of these counties documented incentives for implementing the fuel breaks. Howeve r, none of these counties discussed ongoing maintenance responsibilities or inspections to ensure acceptable fuel loads within the fuel breaks. All counties had subdivision design guidelines and development standards regarding fire suppression activities, specifically water supply standards, address signage, minimum road widths, and vehicle turnouts. However, not all county standards met the IBC 2018 recommendations. All counties had mechanisms for making exceptions to guidelines and standards. These except ions are often utilized in regards to dual access requirements in remote rural developments. No counties articulated processes or mechanisms to trigger redesigns and reconstruction of nonconforming roads. Anecdotal evidence suggested that very few countie s have the political will to implement robust zoning codes and ordinances to restrict

PAGE 203

186 development or minimize its impacts on wildfire or to implement robust inspection and maintenance programs, which is supported in the literature (Bhandary & Muller, 2009; Godschalk et al., 1998; Headwaters Economics, 2016b; Syphard et al., 2013) . Blaine County, Idaho instituted a “Mountain” overlay zoning district, to among other things, “prevent unsafe development into areas at risk to wildfires” (Blaine County Land Use Department, 2012, pp. 92121B) . The mountain overlay district documents do not allow development on slopes greater than 25%. Additionally, ingress and egress cannot navigate slopes greater than 25% or in fire chimneys or saddles. A fire chimney is a narrow side canyon usually tilted up toward a ridge line that draws hot air from a fire into it, speeding it uphill. A saddle is a low area between two high points on a ridgeline. In a wildfire, saddles act as channels for high winds as the herd from the fire flows uphill. The saddles exac erbate wildfire behavior, as such neither chimneys or saddles are locations where development should occur. It is worth noting, Blaine County’s mountain overlay zoning has these unique guidelines because it has multiple frames. Blaine County goes beyond the simple allocation of land use and development and also views zoning as a means to protect aesthetic value, ensure slope stability and public safety, and maintain water and wildlife habitat quality (Blaine County Land Use Department, 2012) . Building C odes Shockingly only six counties exhibited passing grades for building code s : Bannock County, Idaho (A), Boulder County, Colorado (A), El Paso County, Colorado (A), Mendocino County, California (A), San Diego County, California (A), and Valley County, Idaho (A). While advances in Firewise construction best practices continue to grow , researchers have known the basic tenants of Firewise building codes for a long time (J. Cohen, 2008; J D Cohen, 1991; Jack D. Cohen, 2000; Jack D. Cohen, 2000; Jack D Cohen, 2001; Jack D. Cohen & Finney, 2010;

PAGE 204

187 Menakis, Cohen, & Bradshaw, 2003; Reinhardt et al., 2008) . Indeed, research has shown that building to minimum Firewise building codes drastically increases the potential of structure survivability in all but the worst of wildfire conditions (Graham et al., 2012; Haines, Renner, & Reams, 2008; International Code Council Inc, 2003; Quarles, Natural, Advisor, & County, 2010) . Many counties did not write their own building codes, but rather they simply implemented the International Building Code (IBC) standards. While the current iteration of IBC 2018 contains robust wildfire specific building codes and WUI overlay zoning codes, over 60% of the counties have not updated their building codes since the early 2000s, meaning that they are not implementing IBC standards with wildfire s pecific codes. One extreme example of the updating cycle is Inyo County, California, which has not updated their building codes since 1982. This is woefully inadequate, especially for wildfire risk mitigation. While it is reasonable that the IBC 2018 code s have not been adopted because they are only one year old, the IBC has had wildfire related WUI and building codes since the early 2000s (International Code Council Inc, 2003) . However, previous adopted iterations of the IBC code have been ineffectively implemented because many counties did not adopt the full IBC code. In fact, in many cases, counties adopted only portions of t he code. For example, Cochise County, Arizona implemented the core IBC building codes of 2012, but specifically chose to not implement any of the wildfire specific building codes or WUI overlay codes. Rio Blanco County, Colorado implemented their building codes in 2016 and mentioned the need for current best practices regarding Firewise roofing materials in wildfire risk zones, but this county did not adopt Firewise building codes for s iding and decking; closed eaves and soffits; and protecting vents and wi ndows . This is highly

PAGE 205

188 concerning because emerging research has demonstrated that reducing wildfire risk is likely an all or nothing endeavor in regards to building practices (Lasky, 2018b) . Additionally, 20 counties had no building codes at all and simply refer red to state minimum building codes . Furthermore, many of the state standards are not wildfire specific or the standards are out of date. In fact, 12 counties referred to state building codes with no wildfire specificity. Five counties provided web links to state building code documents that no longer exist , and ten more counties referred t o state building codes hosted by a third party behind a paywall. Broken links and paywalls are problematic because the codes are effectively inaccessible and particularly serve as a socio economic barrier to many rural, lower income property owners. Plan Review and Inspection Procedures Eightyfive counties required plan reviews and inspection testing to ensure adequate water supply in order to assist in fire suppression activities. These inspections were to occur during one of two phases: pre development or during building occupancy permit inspections. The amount of water required of each private landowner varied by county and sometimes by location within a given county. Counties that required water tests met the minimum standards for rural residential ho me protection: 1,000 gpm for a minimum of 30 minutes . The storage volume and delivery mechanisms required to meet these minimums varies according to micro climatic conditions. In areas with many natural water sources (e.g., streams, lakes, and reservoirs), the amount of water required is lower than in arid environments. Counties varied wildly in how they assessed landowner’s implementation of home ignition zone best practices, ranging from no inspection to volunteer self assessed reporting to pre developme nt inspection to building occupancy permit inspections to neighbor reported non-

PAGE 206

189 compliance to random inspections. Inspecting authorities were also highly variable, ranging from self assessment reporting to fire district manager to planner to building inspe ctor to state forest agency representatives (e.g., CalFire or state forest service rangers) to consultants. Penalties for noncompliance were also variable: denial of building and occupancy permits, fines, leans, and partial reimbursement of fire suppressi on costs. California and Oregon both require inspections at a state level, but each are implemented differently. Both states require differing defensible space requirements in distinct wildfire risk zones. Both states also require ongoing monitoring and i nspections. California’s inspections are carried out by CalFire. CalFire’s inspection form can be found in Appendix L. CalFire has the right to inspect homes in the WUI and complete wildfire mitigation on private property. If mitigation work is inadequate , CalFire can fine the landowner twice. If the land owner is fined a third time within five years, CalFire can hire contractors to complete mitigation work, and the cost of that mitigation work will be billed to the homeowner ("CHAPTER 3. Mountainous, Forest , Brush and Grass Covered Lands," 2019) . If the homeowner does not pay, the cost of mitigation work will be assessed as a lean against the home, to be paid when the property is sold to a new owner ("CHAPTER 3. Mountainous, Forest , Brushand GrassCovered Lands," 2019) . While this seems like a robust policy, it is important to note that California has 4,426,803 homes in the WUI (V C Radeloff et al., 2005) . While not all of these WUI homes are in high risk areas, CalFire has inadequate inspection staffing to maintain inspections on all homes. As such, Californ ia recommends that homes in lower severity wildfire areas implement home ignition zone standards, but they are exempt from statemandated defensible space guidelines and inspections . This gap in requirements can prove catastrophic, as recent wildfires in C alifornia

PAGE 207

190 have shown. Figure 6.14, 6.15, and 6.16 shows three homes that were lost in the recent Paradise, California and were not compliant with defensible space standards. Note. This house is not compliant with home ignition zone best practices. Take note of the tree canopy overhanging the roof and the shrubbery touching the wood siding on the house. Figure 6. 14 . 6153 Laurel Dr. Paradise, CA, taken July, 2012 (Google) .

PAGE 208

191 Note. This house is not compliant with home ignition zone best practices. Take note of the tree canopy overhangi ng the roof, shrubbery touching the wood siding on the house, and needle and leaf debris on the roof. Figure 6. 15 . 565 Valley View Dr. Paradise, CA, taken July, 2012 (Google) .

PAGE 209

192 Note. This house is not compliant with home ignition zone best practices. Take note of the tree canopy overhanging the roof, density of trees within 30 100’ of the structure, shrubbery touching the wood siding on the house, and an unenclosed elevated wooden dec k. Figure 6. 16 . 5915 Pine View Dr. Paradise, CA, taken July, 2012 (Google) . Oregon requires defensible space, but the requirements vary by statedesignated risk zones: moderate, high, and extreme (Oregon Department of Forestry) . Oregon’s inspection forms are publicly accessible on their website (Oregon Department of Forestry) . Oregon’s inspections are completed via self assessment every five years, when the ownership changes, or when a new structure is added. While this is a unique approach to addressing county and state capacity for inspecting all WUI lots, research has shown that self assessment is often not a robust way of assessing the efficacy of compliance (J. M. Berry, West, & Dennehey, 1989; R. Gray, Kouhy, & Lavers, 1995; John, Loewenstein, & Prelec, 2012) . Oregon is empowered to collect up to $100,000 in suppression costs from each noncompliant homeowner and costs can exceed

PAGE 210

193 $100,000 if a landowner is found to be willful, negligent or malicious in the origin of a fire (Oregon Department of Forestry). While n o county require s fire insurance , Bannock County, Idaho strongly encouraged fire insurance for land development that had the potential for commercial liability. D evelopment in high risk portions of the WUI should strongly encourage or require fire insurance. Fire insu rance education is also important as demonstrated by Jack Thompson’s testimonial in Boulder County, Colorado’s CWPP: Jack Thompson has lost his home to wildfire, twice —in the Fourmile Canyon Fire and in the Black Tiger Fire. In 1989, he had a full replacem ent cost policy. When the final amount was tabulated for replacement of his home, the insurance company paid all of it. Jack’s situation, however, is much different following the Fourmile Canyon Fire. Full replacement cost policies have gone by the wayside following the enormous losses associated with the Oakland Hills fires in California and Hurricane Andrew in Florida. After the Fourmile Canyon Fire, Thompson was surprised to learn that his insurance would not cover the full cost of rebuilding his home. H e is not alone. Many people believed their insurance policies provided them adequate coverage and by paying their annual insurance premiums they have done their part in protecting their future. The lesson of the Fourmile Canyon Fire is, do not assume you a re fully insured —understand your policy and update it annually. Jack Thompson is rebuilding his home again, but this time he is not rebuilding the home he had. He can’t afford it (Boulder County Board of County Commissioners, 2011, p. 38) . Inspection procedures do not explicitly document ensuring roads were constructed to the proper width, turnouts or multiple ingress or egress installation. However, typically road engineering requires commissioning, inspections, and certification outside of wildfire inspections through either state highw ay departments or county agencies such as highway departments or public works. This process provides an opportunity to ensure that wildfire risk management is

PAGE 211

194 taken into consideration during road construction, which is essential in providing emergency rout es and emergency service access. Twenty counties require d predevelopment mitigation work to be completed before a new subdivision or home can be built . None of the twenty counties provided fuel loading guidelines or assessment process criteria documentatio n. Each county required a qualified professional to assess each site and its context for fuel load risk. Counties varied in who qualified to assess risk, ranging from planning department employees to forest service agents to a fire marshal to a sub consult ant. In many cases, assessors used development or home lot plans in concert with home ignition zone inspection check lists to assess fuel load risks (Appendix M). Counties then required the mitigation work to be completed prior to the issuance of entrance, grading, or building permits. Final Thoughts: Theory, Pr actice , and Policy Implications for HFRA Implications for Planning Theory The study of wildfire planning supports, challenges, and suggests several new avenues of inquiry within planning theory. Specifically, this study has two main contributions for planning theory. The first implication is methodological in relation to plan evaluations, and the second contribution for planning theory is the insight into wildfire plan making. Plan Evaluation The f irst contribution this study makes to planning literature is methodological, as it expands plan evaluation into wildfire plans. In order to discuss wildfire plan making theory, we

PAGE 212

195 need to evaluate the process, plans, and outcomes of wildfire planning efforts.6 It is important to note here that there are few, if any, large N studies that have evaluated wildfire plans across diverse contexts. While several planning studies were conducted for broadly defined hazards and flooding, few studies wer e found that addressed wildfire plan evaluation. The seminal work of William Baer (1997) formalized the concepts of plan evaluation. His work identified five types of plan evaluation: plan assessment, plan testing and evaluation, critique, research and professional evaluations, and post hoc evaluation of plan outcomes (Baer, 1997) . While plan evaluation research has addressed various aspects of hazard planning, these research efforts mainly focused on comprehensive planning or flooding coas tal hazard issues (Baer, 1997; P. Berke & Godschalk, 2009; Frazier et al., 2013; Guyadeen & Seasons, 2016a, 2016b; Kim & Tran, 2018; Laurian et al., 2010; Lyles & Stevens, 2014; Oliveira & Pi nho, 2010) . The few studies that focused on wildfire plan evaluation were limited to what Baer (1997) calls research and professional evaluations and post hoc plan outcomes. However, research and professional evaluations had two primary limitations. First, the studies were often conducted within the fire science literature, not planning literature, and they also focused on small N plan comparisons within the same state or across a small subset of states (e.g., Jakes et al., 2007; Jakes et al., 2011; Reams et al., 2005; Srivastava & Laurian, 2006) . Expanding the N allows for a 6 For a comprehensive literature review of the plan evaluation literature, please read the works of Baer (1997); P. Berke and Godschalk (2009); Samuel D. Brody and Highfield (2005); Connell and Daoust Filiatrault (2017); Guyadeen and Seasons (2016a) ; Kenitzer (2016); Kim and Tran (2018); Laurian et al. (2010); Loh (2011); Lyles and Stevens (2014); Norton (2008); Oliveira and Pinho (2010); Seasons (2003); Shahab, Clinch, and O’Neill (2017); Spurlock (2019); Stevens, Lyles, and Berke (2014); and Tang a nd Brody (2009) . For a dee per history of plan making literature please read Abrams, Nielsen Pincus, Paveglio, and Moseley (2016); Samuel D. Brody, Kang, and Bernhardt (2010); R. J. Burby (2003); R. J. Burby et al. (1997); Dalton and Burby (1994); De Leo and Forester (2017); Frazier et al. (2013); Loh (2011); Loh and Norton (2013); March (2010); May et al. (1997); and Vidyarthi et al. (2012) .

PAGE 213

196 deeper contextual analysis of social, physical, and governance scale influences on county wildfire planning efforts to reduce landscape scale wildfire risk. Second, many of the studies focused on local (i.e., subcounty) scale wildfire planning efforts (e.g., Ager, Kline, & Fischer, 2015; Muller & Schulte, 2011) . It is widely documented that subcounty scale wildfire planning efforts are effective at engaging homeowners and neighborhoods in wildfire mitigation; however, as noted previously, it is ineffective and reducing landscape scale wildfire risk factors. Hence, i t is also necessary to study county scale wildfire planning efforts. Additionally, post hoc plan evaluations exhibited two limitations. First, most post hoc plan evaluations primarily focused on social outcomes of the planning process (e.g., capacity and knowledge) (e.g., H. J. J. o. F. Brenkert Smith, 2011; Guyadeen & Seasons, 2016a; K. C. Nelson et al., 2010) . The focus on social outcomes c an be attributed to post modernist planners influences on plan evaluations, which conceives the plan as a symbolic outcome of the large goal of achieving community dialogue. I would argue that the primary goal of wildfire planning is a matter of health, sa fety, and welfare, and thus, the primary concern of wildfire planning is wildfire risk reduction. However, I also feel this does not negate post modernist plan evaluation concerns, or the benefit of social outcomes. As repeatedly articulated throughout thi s document, public participation and the community capital that it creates is critical to wildfire risk reduction planning. The second limitation of post hoc wildfire plan evaluations is that no study has coupled the evaluation of 1) the process, plan content, policy implementation, and 2) the built environment and biophysical reduction of risk. The plan evaluation instruments and methods discussed in Chapter IV overcome these limitations by bringing plan evaluation methodologies into wildfire planning whi le creating a novel methodological approach to understand plan

PAGE 214

197 making process, content, and policy outcomes, and biophysical outcomes. The results of these plan evaluations provide unique insights into county scale wildfire plan making. Plan Making While this research study spanned multiple states, each state enabled county or local level land use plan making. The results of this study facilitate a theoretical discussion on three plan making principles: scale, capacity, and commitment. The fire science literature advocates for county scale CWPPs and wildfire planning to reduce landscape wildfire risk. Indeed, it has become a documented best practice (Society of American Foresters, 2004; Williams et al., 2012) . While the county scale makes theoretical sense given its focus on wideranging, heterogeneous land use issues and need for multi jurisdictional coordination of resources, the results of this study show that counties are unable to manage wildfire risk adequately. This study therefore highlights a disconnect between fire science best practices and plan making theory. This outcome is supported by Norton et al.’s (2018) findings, where county governance also demonstrated a lack of progress in land use planning efforts for coastal management. Modernist plan making theory suggests counties inability to reduce wildfire risk could be explained by a lack of robust federal and state mandates (Abrams et al., 2016; R. J. Burby et al., 1997; Dalton & Burby, 1994; May et al., 1997) , which is partially true. Indeed, s tates with a longstanding state influence on planning enablement, such as California, on average , scored higher than other states. California has made significant strides in mandating stringent building codes on new construction, wildfire planning, and requiring and supporting county comprehensive planning. This is supported by emerging research which suggests that homes built to the 2009 California State Building codes have a much higher chance of survival, as evident in the recent Camp and Tubb fires (Guerin, 2018; The Sacramento Bee, 2019) .

PAGE 215

198 Conversely, states with weak state mandates or home rule states (e.g. , New Mexico) on average scored worse. This would support, as the modernist plan evaluation and plan to make literature suggest, the need for coordinat ing stricter federal and state wildfire planning mandates . However, state enabled mandates are only a partial explanation as to why California, on average, outperformed counties in other states. California’s higher average scores were also due to counties access to state financial and technical capacity. California’s comprehensive planning and wildfire mandates are facilita ted through funding, incentives, and disincentives. Additionally, CalFire provides extensive technical assistance. The outcomes of this research would support that lower scores can be attributed to limited financial capacity, but financial capacity alone does not explain higher or lower scores. Indeed, many counties acknowledged limited capacity as a constraint to wildfire suppression, planning activities, and project implementation. States without strong mandates, but adequate funding, scored significantly lower than California on average, suggesting that mandates need to be coupled with adequate financial resources. Despite California’s advancements, no California county received a passing composite scores due to commitment issues. Commitment broadly refe rs to the planner’s willingness to act (May et al., 1997; Norton et al., 2018) . This concept is demonstrated when counties use planning as a governmental function to address substantive objectives and the appropriate activities to advance those objectives, such as using land use regul ations to reduce wildfire risk (Dalton & Burby, 1994; May et al., 1997) . The results of this study confirm the previous research findings: county governments show little commitment to regulate or enforce land use regulations (Philip R. Berke & Campanella, 2006; Samuel D. Brody et al., 2010; R. J. Burby, 2003; R . J. Burby et al., 1997;

PAGE 216

199 Dalton & Burby, 1994; Dalton, Conover, Rudholm, Tsuda, & Baer, 1989; Frazier et al., 2013; Norton et al., 2018) . This could be because most of the wildfire plan making literature focuses on post modernist app roaches to the planning process, specifically public participation. These efforts have spanned both county and subcounty scales. However, as this study’s results show, the over emphasis of post modernist approaches in wildfire plan making has created inef fective county plans. A s Baer (1997) predicted, incorporating post modern criteria in federal and state wildfire planning mandates have created vague and unimplementable plans because of the overemphasis of social process outcomes. As such, the vagueness of HFRA is not genius, as Jakes et al. (2011) suggest, but in reality, a barrier to meaningful risk reduction. Indeed, the wildfire plans evaluated in this study lacked clear visions, goals, objectives, and pathways to implementation. Thus, these plans do not reduce landscape risk. Instead, the WUI and catastrophic wildfire lo sses continue to grow. These results were not surprising because similar outcomes have been described across the comprehensive and hazard planning literature (P. Berke & Godschalk, 2009; Philip R. Berke, Yan, & Stevens, 2009; Samuel D. Brody & Highfield, 2005; Frazier et al., 2013; Kim & Tran, 2018; Loh, 2011; Prater Carla & Lindell Mi chael, 2000; Tang & Brody, 2009) , which has served to fuel the ongoing debate between modernist and post modernist approaches in hazard planning, suggesting there is still a strong need for ongoing research in areas of public participation in wildfire planning. For example, despite the wildfire plan making literature emphasis on participation processes and social outcomes, wildfire planning has yet to demonstrate it has moved beyond nonparticipation and token participation. Indeed, as discussed in Chapter VI, virtually all counties pursed, what Arnstein (1969) defines as therapy (e.g., community defensible space days), informing (e.g., the one way flow of information in how to reduce private

PAGE 217

200 citizens property of wildfire risk), con sultation (e.g., attitude surveys, community and open houses), and placation (e.g., appointed key stakeholders with no decisionmaking authority). A post modernist could explain this outcome as the counties lacked the capacity, knowledge, or commitment to engage in what Arnstein (1969) refers to as degrees of citizen power. This indeed may be true because many counties had little to no planning staff and lacked the financial means to maintain consultants or ongoing participation efforts. However, this argument overlooks the complexities of scale and its impacts on putting theory into practice. I would argue that it is more likely the result of incomplete post modernist theories. Post modernist planners that advocate for partnerships, delegation, and citizen control levels of participati on have documented successes at small municipal and neighborhood scales and suggest that these concepts are scalable across multi jurisdictional levels of governance (P. Berke & Godschalk, 2009; P. R. Berke, Dixon, & Ericksen, 1997; Goodchild, 1990; Newig & Fritsch, 2009; Norton, 2008; Tang & Brody, 2009) . This also suggests that wildfire planning should happen at subcounty scales. However, it is worth noting that fire science and hazard plan evaluation literature has continually confirmed, l ocal hazard plans share the same flaws as county wildfire planning because they also have done little to address risk, even while engaging in various degrees of citizen power (Philip R. Berke et al., 2009; Samuel D. Brody & Highfield, 2005; Kenitzer, 2016; Kim & Tran, 2018; Laurian et al., 2010) . Additionally, sub county scales prove to have similar, if not greater planning and financial capacity issues, requiring higher levels of governance support (Philip R. Berke & Campanella, 2006; Dalton et al., 1989; Norton et al., 2018) . As federal and state entities become involved, there is a need for coordination and generalization, which are often counter to post modernist theories. Without mandates and coordination, planning efforts have become piecemeal and fragmented, and therefore not able to

PAGE 218

201 reduce landscape scale risk (Dalton et al., 1989) . Thus, county scale wildfire planning is critical to risk reduction; however, it has proven difficult to integrate public participation. This study’s results suggest public engagement is a worthy activity because, even at token levels of participation, counties often achieved higher scores i n categories other than public engagement if they engaged their constituents. However, all counties in this study acknowledged limitations to their public engagement, often due to the complexity and large differences in county, ecological, and social compositions (see the public participation discussions in Chapter VI). Admittedly some counties lack the commitment to engage the public; however, many counties wish to, but struggle to, move beyond token participation. These results show that post modernist community engagement theories and methodologies are not as scalable as suggested. Thus, new avenues of research are needed to engage the public at larger and more heterogeneous scales, which integrate both the modernist mandates of improving risk reduction w hile also more deeply engaging the public. Finally, in order to increase theoretical understandings of the complexities regarding commitment to wildfire risk reduction, key concepts from modernist and post modernist planners should be integrated. Both the fire science and planning literature has long advocated the best solution to reducing wildfire risk are building codes and land use regulations. However, no level of governance precluded development from wildfire risk areas, and this, suggests that new ave nues of research are needed to enhance the political will in committing to land use regulations needed to address WUI wildfire risk. It is worth noting that as I write this section, significant advancements in commitment are being made regularly. However, before recent catastrophic wildfires in California, the commitment had stopped short of strict land use regulations. Emerging efforts by the American

PAGE 219

202 Planning Association (APA) and California are normalizing technical knowledge best practices for public participation and using planning, specifically land use regulations, and codes to address wildfire risk. These recent actions reinforce the notion that catastrophic fire seasons can influence increasing commitment to wildfire risk reduction, at least in the relatively shortterm. In April 2019 the American Planning Association (APA) published PAS report 594: Planning the wildland urban interface, which documents new national policies and planning frameworks that influence the WUI, latest science, best pract ices, and data that is driving better understanding of the WUI and appropriate land use solutions (Mowery, Read, Johnston, & Wafaie, 2019) . The report contains an extensive array of the best practices referred to throughout this document. It also suggests similar future research agendas: development morphologies on wildfire risk and structure survivability, metanalysis, and evaluation of state and local planning best practices, and co benefits of WUI regulations for sustainability. The report undercuts its power, like the CWPP discussion in Chapter IV. The report is rife with language such as: Planning the WildlandUrban Interface is the latest report in that series. I can think of no more important topic for planners to address to fulfill their fundamental mission of promoting the public health, safety, and welfare (Mowery et al., 2019, p. 3) and intentional actions by planners to mitigate negative consequences in th e built environment, including the effects of destructive wildfires on our communities, are an essential responsibility to ensure a safe and resilient future (Mowery et al., 2019, p. 8) . These statements set up the need for strong WUI planning. However, the authors conclude the executive summary with: In many ways, the existence of the WUI may be inevitable. Humans will always seek a relationship with nature and will want to live near forests, grasslands, and other wildlands (Mowery et al., 2019, p. 8) .

PAGE 220

203 While it is true that human behavior cannot be completely regulated, this quote does send a message of planning futility. Likewise, the report argues for rethinking the WUI by familiarizing planners with WUI issues and concepts and educating planners through tangible planning solution case studies, which would prime planners to engage in land use wildfire risk reduction solutions. However, the report does not acknowledge planni ng’s failure to address WUI growth in extremely high wildfire hazard risk areas throughout the United States. Research, as discussed above, has repeatedly shown that this failure to address is a problem that planning practice faces across a range of hazard s, not just wildfire. While it is too early to determine how these inadequacies in the report’s wording will impact results, it is likely that this report will exhibit little to no effect on planners in reducing landscape wildfire risk, similar to the prev ious APA report by Schwab et al. (2005) . As a professional organization, the APA should have taken a stronger stance and mandated wildfire risk design and planning best practices through the maintaining of licensure (see the professional licensure discussion in Chapter VI). Some would a rgue that licensure changes are not needed because large fire events are pushing varying levels of government to rethink its legislation. For example, California has shown a re emergence to land use regulation to reduce wildfire risk. In September 2018, G overnor Brown signed 29 measures into law, ranging from forest management, mutual aid, emergency alerts, post disaster recovery, power grid fire caused liability, and to electric garage doors. It is important to note SB 182 Local government: planning and z oning: wildfires , which received California Senate approval and is currently at the Assembly pending approval ("Local government: planning and zoning: wildfires.," 2019) . SB 182, for the first time specifically mandates the regular update of hazard (i.e., safety) plans within comprehensive plans every eight years. Additionally, it requires several regulatory updates to land use codes and ordinances. First,

PAGE 221

204 counties must designate and adopt a very highrisk area overlay zone to reasonably direct new development away from veryhigh risk areas; when adopted, the county is prohibited from developing the said area. Second, it requires counties to adopt a plan to address the existing development of veryhigh risk areas. Third, it requires the coordination and inspection of wildfire risk reduction standards. Fourth, SB 182 requires collaboratively posting local ordinances, policies, and best practices relating to land use planning in very high fire risk areas. Finally, it prioritize wildfires risk reduction project grants to counties removing and directing development away from veryhigh fire risk areas and retrofitting noncompliant existing development. While these mandates are unprecedented and a huge step in the right direction, it is worth noting there are still loopholes in place because this is mandated unless “the city or county makes specified findings, findings based on substantial evidence in the record” ("Local government: planning and zoning: wildfires.," 2019, p. 96) . Likewise, it is still unclear how very highrisk zones are collaboratively negotiated. Thus, it remains to be seen how many exemptions are made to this mandate. Furthermore, recent fires have shown to have catastrophic consequences even in areas with designations below veryhigh risk, as evidenced by the 2017 Napa Valley fire. Indeed, emerging mega fires have produced disastrous effects kilometers from the fire front due to embers igniting homes – suggesting the need for a stricter risk mapping and a t iered mandate structure addressing homes and development patterns in lower wildfire risk zones (M. Moritz & Anderson, 2018; M. A. Moritz et al., 2014) . Implications for Practice The purpose of this research was to evaluate whether and under what conditions county governments are implementing wildfire risk reduction best practices across the American West .

PAGE 222

205 The question was how well do counties integrate wildfire risk reduction best practices within the CWPP process and subseq uently CWPPs into the county land planning codes and ordinances. Furthermore, this study attempted to explore what social, economic, demographic and geographic factors predict the level of best practice integration of CWPP inputs and outputs (Figure 1.1 and 5.14) ? This research provides a baseline that identifies gaps between best practice research and implementation. Understanding the levels at which best practices are integrated and the conditions under which they are integrated is beneficial to planners and consultants because it provides contexts that can be leveraged for deeper integration of best practices or external barriers that need to be moderated before integration can be successful. This will allow future research to identify why some tasks – such as framing – are so difficult while others are much easier. Practitioners can then adjust their processes and methods accordingly, focusing on known gaps or challenges in order to integrate best practices. Ensu ring the integration of best practices provides confidence in the veracity of the content to broader community participants, and also provides a sense of legitimacy in the process and county governance efforts in protecting public health, safety, and welfa re. Veracity and legitimacy are important to maintaining community engagement in wildfire planning efforts. They are also important in order to mobilize landowners to engage in mitigation efforts. In this project, I first created two document analysis inst ruments to evaluate the CWPP planning process and the land development codes. Then I explored changes in WUI geographic extent and population growth. Finally, I evaluated the socioeconomic and demographic contexts under which CWPPs and land development codes and ordinances are developed. The results of which provide suggestions to improve CWPPs and county governance regulatory approaches to reducing wildfire risk.

PAGE 223

206 Updating the CWPPs and ordinances The CWPP process scores were very low, shockingly so whe n results were first tallied. However, given the dates of initial CWPP development, these scores underscore the need for comprehensive updates to those CWPPs. Over 57 of 120 CWPPs selected for this study were created before researchers had an opportunity t o evaluate and publish CWPP process and best practices. This conforms with the slow process by which research is often disseminated and integrated into practice. Therefore, while best practices research has documented robust CWPP processes and content best practices, they have not been integrated into CWPPs because the literature is more current than most CWPPs. Best practice literature has merged within three time periods: 20032004 (One Hundred Eighth Congress of the United States of America, 2003; Societ y of American Foresters, 2004), 20072008 (Fleeger, 2008; Jakes et al., 2007; Resource Innovations Institute for a Sustainable Environment, 2008; Rodman & Stram, 2008), and 20112012 (Jakes et al., 2012; Williams et al., 2012). Yet, 57 of the CWPPs in this study were created before 2008, and 84 other CWPPs were created before 2011. Furthermore, in reviewing the citations and CWPP process documentation, many of the post 2011 CWPPs were still quoting literature from the early 2000s. However, if the community is to believe the efficacy and immediacy of the WUI wildfire problem, then CWPPs need to undergo more regular and substantial updates as well as commit to integrating current best practices in those updates. Updates should respond to both the evolving clim atic and biophysical changes to risk and community context. In many cases, this will require updating and expanding the CWPPs frame to include additional and broader community perspectives. Counties need to make every effort to update yearly project lists and review major updates no more than 10 years apart. Additionally, comprehensive plans are often on a similar

PAGE 224

207 update cycle. Integrating CWPP updates into the comprehensive planning process could assist some counties in overcoming planning capacity issues. Framing CWPPs and Comprehensive Plans Th e findings of this study demonstrates that. framing is critical to understanding how communities define issues and supports Williams et al. (2012) ’s results. The framing of the CWPPs and comprehensive plans was often unclear. Both documents need to more clearly articulate their frames of reference to ensure plans are based on facts, has continuity, and provides transparency. For CWPPs to reduce the effects of catastrophic wildfire consequences, they need to integrate a diversity of frames. Similarly, comprehensive plans are addressing a multitude of complex system dynamics. Frame diversity explains why different planning efforts approach differing goals, objectives, strategies, and participation processes. Frames even define th e core teams are, ‘who’ the broader community is, and how involved they are. Finally, they also define management expectations, options, and outcomes. Counties with single frames often limit themselves to predefined solutions. For example, a CWPP frame tha t includes emergency response will focus on suppression capacity, training, equipment, and home and infrastructure protection. Yet, a CWPP that focuses on homeowner responsibility will focus on education, outreach, homeowner mitigation efforts and home ignition zones. Both of these singular frames often preclude efforts utilized through the other frame and also preclude using prescribed fire as a means to reduce risk and maintain ecosystem health. These singular frames, in isolation, consider all fire as a threat and not as a tool. Integrating multiple frames is challenging. It is often the most difficult step in any planning process because, as with any planning problem, there is not a correct, single solution. It

PAGE 225

208 requires deep, sustained, and broad community participation. Each person and subcommunity has its own perspectives, solutions, morals, values, and strategies, which are often conflicting. However, counties should not avoid these conflicts to generate a CWPP or comprehensive plan. They should embr ace conflicts in a structured and productive process. By trying to eliminate conflict, voices and options are also eliminated. Counties should engage in effective participatory, collaborative, and mediated processes, with open and well structured dialogue . Participatory and community engagements’ best practices have been documented at length within the literature (Bihari, Hamin, & Ryan, 2012; Brummel et al., 2012; Cheng & Sturtevant, 2012; Cilliers & Timmermans, 2014; Fleeger & Becker, 2008, 2010; B. Gray, 2004; Innes & Booher, 2014; Sanoff, 2000) . Admittedly, deeply engaged planning processes are time and resource intensive, often outside the expertise and financial means of many of the 120 reviewed counties. For example, there were several instances where CWPPs were guided by elected coun ty sheriffs, who are not trained in participatory planning. Many counties overcame these challenges by acquiring external grants in order to hire consultants. However, a consultant’s frame is also crucial to how CWPPs are framed. For example, CWPPs created by forestry consults over emphasized fire as a threat to lumber commodities and often integrated logging strategies that research has shown to be less effective than other wildfire mitigation strategies (Ayres et al., 2016; Busenberg, 2004) . If consultants are hired, it is helpful if they are multi disciplinary in order to overcome their own inherently biased frames. A nother option is to hire collaborative planning consultants, which would allow the core team and community to provide the material to create the plans. Undoubtedly, federal and state funds are inadequate to support multiple, repeated planning efforts. Federal and state governments could and should do more to ensure wildfire

PAGE 226

209 funding and other land management funds can support year round planning and mitigation efforts, rather than simply supporting seasonal fire suppression efforts. Additionally, if countie s integrate comprehensive planning and CWPPs into a single process, they open up additional funding opportunities, as many states (e.g., California, Washington, and Oregon) financially support comprehensive planning efforts. If counties have exhausted all traditional means of supporting CWPP development, such as grants, consultants, state and federal agencies, and nonprofits; there are several non traditional means to enhance county capacity. Two of the higher scoring CWPPs were directed by academic instit utions. Curry County, Oregon’s CWPP was created through the University of Oregon’s Resource Innovations Institute for a Sustainable Environment (Lynn et al., 2008) . Another, CWPP, Mason County Community Wildfire Protection Plan, was created as a project based learning experience by an under graduate and graduate ecosystem management course, under the guidance of their professor and a steering committee (Western Washington University Huxley College, 2012) . Finally, it is important to note that m any CWPPs do a great job at public participation. However, they still have a significant wildfire risk problem. Some countie s have done a great job at developing robust core teams and reaching out to a broadly defined public. However, one frame is chronically and critically missing – wildfire risk is a land development problem. This frame is noticeably unrepresented in the goal s and objectives of most CWPPs because it reflects the participatory composition of CWPP teams. Other than a few CWPP consultants, there was a noticeable absence of professionally licensed land developers and consultants (i.e. architects, site planners, en gineers, landscape architects, land developers, surveyors, landscape contractors, contractors, and realtors). This absence of licensed professionals tasked with ensuring public

PAGE 227

210 safety, omits the responsibility inherent in the licensing process and therefore omits the hard development restricting mitigation efforts necessary to keep the public safe. CWPP Goals , Objectives , and Priorities CWPPs can be strengthened by using a framework of goals, objectives, and strategies. As stated, many CWPPs failed becau se they either simply reiterated HFRA goals or did not articulate clear, measurable goals. Furthermore, many county CWPPs only stated goals and very few had objectives. Additionally, no county had clearly stated, time limited objectives or strategies paramount in creating accountability to these objectives. As a result, there is no clear way to evaluate the success of implementing a particular CWPP due to these issues with goals and objectives. To improve CWPPs, goals cannot be a reiteration of HFRA’s goa ls; th ey need to respond to how the county frames a CWPP and its core values . While goals should lso be general statements, they should be specific enough to assess whether progress has been made in achieving them. Objectives need, first, to be included and, second, to be more specific and a subset of the goals in order to provide measurable strategies and a time frame for those strategies to be completed. To ensure effective objectives, counties should evaluate them to ensure they are S.M.A.R.T.: Specific – do they tell how much of what is achieved, and by when? Measurable – can information about the objective be collected, detected, or obtained? Achievable – is it feasible to complete? Relevant – does it contribute to the frame, scale, and reducing wildfi re risk? Timed – does it have a time frame?

PAGE 228

211 CWPPs should clearly articulate the strategies, programs, or policies that they intend to pursue to implement their goals and objectives. Finally, CWPPs should prioritize their projects and, therefore, the imp lementation plan. If counties acknowledge significant economic challenges and capacity issues, not every project can be of the highest priority. To help refine priorities, counties should consider ranking high, medium, and low priorities across the short , medium , and longterm. Short term projects should be easily feasible with minimal capacity and money. Longer term projects, with significant lead times, are then free to pursue grants and other partnerships to ensure implementation. While this prioritiza tion does not remove the limitations of funding, it does provide strategies to work within such constraints. CWPP Mapping – Base Maps, Vulnerability, and Projects CWPP mapping efforts are overwhelmingly inadequate and need to improve. First of all, many ma ps were illegible. This is unacceptable because, even if needed information were provided, there would be no way to access that information from these maps. Second, many maps lacked critical information. In regards to mapping, in particular, HFRA is too vague. HFRA should provide stricter requirements on the content that is included in CWPPs, specifically in regards to base maps, risk, vulnerability, and projects. Stricter requirements does not preclude HFRA from remaining communityvalue driven and contex tual. In fact, requirements ensure that CWPPs better support mitigation and suppression efforts local contexts across multiple agencies and jurisdictions. Mapping efforts should begin with a clearly articulated WUI definition and defined boundari es on a map. Base maps should include all critical values and infrastructure. Table 6. 6 outlines minimum base map components and their associated categories. To ensure the legibility

PAGE 229

212 of these maps, it will often take more than one map to communicate all of the nec essary values and infrastructure. Finally, hazard and risk need to clearly defined and mapped to ensure continuity in updates and transparency in the process. Risk mapping efforts should also expand to include concepts of vulnerability, many other hazard p lanning efforts (e.g., flood and earthquake) already have and provide a framework to do so. Integrating vulnerability allows CWPPs to acknowledge that not every community member has the same capacity to mitigate, evacuate, or recover from a fire event. In order to ensure all barriers are considered and fair support is provided to all community members, communities need to know what the barriers are and where they are concentrated. Minimally counties should understand and map to the extent possible wildfire vulnerabilities. Counties should consider an adaptation of the CDC’s social vulnerability index (SVI), but at a finer resolution when possible (Flanagan Barry, Gregory Edward, Hallisey Elaine, Heitgerd Janet, & Lewis, 2011) . Table 6. 6. Base map requirements Suppression infrastructure Water sources Access points Locked gates Pre planned points and lines of attack Fire stations, with access times Fire district boundaries Air support zones Fire roads and existing trails Evacuation and emergency services infrastructure Evacuation routes Shelters (e.g., churches, community centers, or schools) Hospitals and other emergent care facilities Police stations Vulnerable communities Age Disabilities Poverty Hospitals

PAGE 230

213 Nursing homes Homeless shelters Community values WUI boundary Communities Cultural and natural sensitive areas Parks and openspaces Trails Zoning Codes, Development Standards, Subdivision Guidelines, and HOA Requirements As demonstrated by this research, zoning codes, development standards, subdivision guidelines, HOA requirements are an underutilized resource in directing development away from high risk wildfire areas and minimizing risk. While research has begun to explore land use regulatory approaches to wildfire risk reduction, there is not a comprehensive detailed accounting of each regulatory approach and its effects on wildfire risk. As such local governance has implemented broader rules of thumbs, such as no development on slopes over 2030% or in fire chimneys. Future research should provide a comprehensive meta analysis of land use regulatory efforts and their impacts on wildfire risk. To assess future development, as allowed under their land use regulations, coun ties should use an anticipatory geodesign process as outlined in Figure 6.17. This process will facilitate a deeper understanding of where development risk can be mitigated and where stricter land use controls need to be implemented to direct development a way from higher risk areas. Additionally, simulations can also enhance policy learning, by linking risk and mitigation strategies to specific properties and communities.

PAGE 231

214 Figure 6. 17 . Anticipatory modeling process. Building Codes Building codes should be current and updated on a more frequent schedule . Procedures need to be in place to ensure compliance and limited to no exception can be allowed. Research has repeatedly shown that Firewise construction is an all or nothing endeavor (Jack D. Cohen, 2000; Jack D. Cohen & Finney, 2010; Lasky, 2018b; Menakis et al., 2003) . Yet, millions of homes are of non compliant, WUI homes. Neither, the literature or CWPPs provided insight into how to address existing structures a nd unsafe developments. Future research should develop strategies, processes, and best practices for addressing these homes. Review Process and Inspections Many of the nuances of the health safety and welfare aspects of land development are ensured through planning and design review processes and construction administration inspections. However, prior to these efforts, design plans are created by licensed professionals. Land Use Regulations and Community Values Urban Growth Modeling and Morphologies Wildfire Vulnerability Modeling Spatial Fire Planning

PAGE 232

215 As discussed earlier, these professionals should be included in the CWPP collaboration process. Counties and states with wildfire risk should engage each profession’s licensing authority to ensure professionals are adequately trained and tested on applicable minimum Firewise development, landscaping, and construction best practices. Just as landscape architects certify to the health, safety, and welfare of subdivision grading and drainage plans, they should also certify their design is Firewise. These efforts can include ongoing HOA and homeowner management plans that the owner or HOA are bound to maintain. Counties should also minimally define an implementation and ongoing maintenance inspection review process. These processes can be multi tie red, due to the large number of homes in the WUI. Future research should expand on inspection best practices, specifically the efficacy of self reporting in personal wildfire risk reduction implementation and other current inspection practices. It is worth noting that advances in machine learning and groundand aerial based highresolution remote sensing offer promising possibilities to better inspecting expansive geographies and in flagging potential issues for more detailed in person field assessments. Implications for HFRA: Flaws and Improvements Flaws The low CWPP scores exhibited by the 120 counties reviewed in this study are the result of HFRA’s vagueness and simplistic incentive structure and requirements. As stated earlier, the only HFRA requireme nts of CWPPs are 1) collaboration, 2) prioritize fuel reduction, 3) measures to reduce structural ignitability, and 4) have three entities mutually agree to the final CWPP content (applicable city or county government, local fire department(s), and state e ntity responsible for forest management). Benefits and incentives of having a CWPP include: a locally defined WUI and boundary for maximum fuel treatment mitigation funding, priority to projects

PAGE 233

216 and treatments identified in a CWPP, and access to a expedited environmental assessment process (One Hundred Eighth Congress of the United States of America, 2003; Society of American Foresters, 2004) . Past research has debated the potential results of HFRA’s vagueness and the benefits that it provides to county and local governance, specificall y their ability to customize wildfire mitigation efforts to local community values (Jakes et al., 2011) ; however, this study demonstrates that the vagueness often defaulted to the minimum status quo rather than utilizing the opportunity to the fullest extent. CWPPs consistently failed to integrate content and process best practices. Indeed, many counties, as detailed in Chapter VI, failed 1) to integrate robust participation, community values, meaningful goals , and objectives, 2) to adequately prioritize mitigation projects, 3) to reduce structure ignitibility best practices, and 4) to use adequate risk and WUI definitions and mapping. Indeed, counties often implemented the minimal status quo of content and participatory processes , resulting in CWPPs with little effect on wildfire risk reduction, as evidenced by the increasing severity and devastation of recent wildfire activity in the American West. The status quo is the result of HFRA’s incentive structure, in which counties must simply demonstrate that they have a CWPP in order to gain access to the HFRA resources. There is no further review to validate content or process before gaining access to the resources. Improvements HFRA needs to improve to ensure county CWPPs are integ rating best practices in reducing wildfire risk. Simply having a CWPP is ineffective. To improve CWPPs, HFRA should be amended to include minimum requirements to gain access to a sliding scale of incentives, as outlined in Table 6.7. The improvements listed below are necessary because research has repeatedly shown that minimal requirements with no standard review produces ineffective plans

PAGE 234

217 (Baer, 1997; P. R. Berke et al., 1997; Philip R. Berke, Roenick, Kaiser, & Burby, 1996; Dalton & Burby, 1994) . Table 6. 7 HFRA CWPP sliding scale incentives. Level 1 Requirements: CWPPs must be no more than ten years old. Goals, Objectives, Strategies, and Actions should be S.M.A.R.T. Implement current IBC WUI development and building codes. Create, define, and map Wildland, WUI interface and intermix, and nonWUI development with three levels of associated risk (low, medium, and high). Base maps must be legible and contain the criteria outlined in Table 6.6. Signatories participation1 Priori tized projects with a timeline of implementation. Incentives: Expedited fuel treatment and environmental review process. Level 2 Requirements: Level 1 requirements. Goals, Objectives, Strategies, and Actions should be S.M.A.R.T. and demonstrate the integration of community values. Informing and consultation participation1 Enhanced prioritized projects with a timeline for implementation organized into low, medium, and high priorities distributed across the short , medium , and long term. Incentives: Expedited fuel treatment and environmental review process. Access to part ially, federally subsidized CWPP planning support and expertise. Gain access to small CWPP planning and wildfire risk reduction project grants. Level 3 Requirements: Level 2 requirements. Create, define, and map Wildland, WUI interface and intermix, and nonWUI development with four levels of associated risk (low, medium, high, and extreme). Integrate climate change into modeling future risk. Prohibit new development in areas identified as an extreme risk. Partnership and delegated participation 1 Incentives: Expedited fuel treatment and environmental review process. Gain access to full federally subsidized CWPP planning support and expertise. Gain access to large CWPP planning and wildfire risk reduction project grants.

PAGE 235

218 Note: 1 Level one partici pation follows the existing HFRA CWPP participation structure, which suggests three signatories: applicable local government, local fire departments, and state entity responsible for forest management. Level two participation requires a demonstratable achi evement of what Arnstein (2019) calls “degrees of token participation.” Level three participation requires a demonstratable achievement of what Arnstein (2019) calls “citizen power.” Limitations and Future Research Limitations As with all research, there are limitations that future research should address. The first limitation is that the s tudy focused on county CWPPs; however, county CWPPs are likely to be influenced by state mandates, policies, and wildfire planning culture, as demonstrated by California’s higher average scores across both instruments. Likewise, county CWPP analysis preclu ded evaluating the interactions between county CWPPs and local level wildfire planning efforts and land use ordinances and codes, which could provide further explanation of the strength or weakness of county CWPP efforts or local Firewise best practices in tegration. It is worth noting that while the research design did not specifically address the measurement of state influences on county wildfire planning, the fact that many county CWPPs referenced state efforts required these referenced state materials to be reviewed for the document analysis. This provided significant anecdotal evidence of the interplay between the state and the county in the discussion of county results as well as provided guidance for future research efforts. The second limitation is t hat the evaluation instruments were created from fire science and hazard planning literature by a single researcher and simply evaluated documents for the presence or absence of each indicator. While this is a normal practice in dissertation research, it does potentially introduce the biases of a single researcher. Relying on peer reviewed best practices from the fire science and hazard planning literature does mitigate this limitation, though

PAGE 236

219 it would be wise in future research to utilize an expert panel for instrument review. Additionally, evaluating the presence or absence of best practices in an instrument does not uncover the nuances of plan quality within a given indicator. Ideally, in future research, the instrument would be accompanied by a qualit ative deep dive into the quality of instrument categories. The third limitation is that this study relied on secondary data, specifically CWPP and local governance documents. Relying on secondary data allowed the analysis of a large n sample; however, it precluded delving into the nuances of CWPP process participation, politics, power, and influences on decisionmaking. Finally, the fourth limitation is that the statistical results were not statistically significant, likely due to too few observations with high scores to get a signal. Future research could increase the n of this study; however, the possibility remains that there will be too few high score observations to result in statistical significance. Future Research Plans To better c apture the influences of jurisdictions, geographic scale, and the nuances of CWPP decision making processes, a smaller n case study research effort should be conducted. A smaller n case study should be conducted using a subset of high and low scoring count ies from this study’s 120 counties. A state and local government CWPP, wildfire mandates, codes, and ordinances, and land use codes and ordinances will help researchers understand the interactive effects of jurisdictions and geographic scale on county wildfire efforts. Each CWPP’s core team should be interviewed, using semi structured interviews, to better understand the levels of participation, politics, power, and influences on decisionmaking. Additionally, each county’s planning staff should also be int erviewed using semi structured interviews to better understand the politics, power, and influences on the county’s use of Firewise building and land use

PAGE 237

220 practices to mitigate wildfire risk. These interviews will help researchers understand why decisionmak ers made certain decisions and why certain best practices were included or left out. A community survey should be administered to understand community perceptions on wildfire risk and the participation, politics, power, and influences on CWPP and county governance decisionmaking, to triangulate the interview results with the values of a broader public. As stated in Chapter IV the CWPP Process and Plan Content and CWPP Implementation: Local Governance Evaluation Instrument s were created from a comprehensi ve literature review. Future research should identify a focus group panel of fire science and hazard planning experts to ensure the rigorousness and validity of the indicators and themes within each instrument. Finally, future research should seek to expand on the statistics used to address questions of concern. Conducting a larger sample and expanded document analysis will produce a larger n, which would allow cross group comparisons. Depending on the additional sampling methodologies used, comparisons could be made between states, income, and planning scales using a factorial ANOVA. Additionally, a larger n facilitates the ability to search for a statistical signal in the ordinal regressions, if no signal is still present the variables could be converted to a binary ordinal regression (e.g., pass or fail scores). Conclusion The intermix of fire prone landscapes, firedependent wildland vegetation, development, people, complex land ownership mosaics, and climate change has created a dangerous and untenable wildfire problem in the WUI. The results show that current CWPP and local governance efforts have done little to reduce or alter WUI expansion and they have not robustly integrated wildfire risk reduction best practices. Indeed, the severity of the proble m continues to grow, which is not surprising given the low CWPP and local governance integration scores.

PAGE 238

221 The similarity of CWPP scores can be explained by the author meta data, which shows that 41 of the 120 CWPPs in this study were created by 13 unique c onsultants. In fact, Northwest Management, Inc, Anchor Point Group, and Wildland Fire Associates accounted for 15, 5, and 4 CWPPs, respectively. It appears that the contextual nature of CWPPs has been subverted by a homogeneity in consultant knowledge, met hodologies, and CWPP process and content structure. Indeed, consultants often reused whole sections across multiple CWPPs, a potential by product of lowest bidder consultant contracts. These results beg the question: are CWPPs really contextual? The result s suggest they are not. CWPPs exhibit a lack of diversity in scores and content. HFRA can be strengthened using two mechanisms: 1) providing access to additional funding for ongoing updates only if CWPP minimum standards are meet, and 2) requiring certain CWPP content by providing more detailed content guidelines. While there were no single exemplary CWPP plans, higher scoring plans did share similar characteristics. First, they engaged a wide array of community members through multiple means of engagement (e.g., town hall meetings, surveys, and focus groups meetings). Despite not reaching citizen control, these means of engagement still produced more robust CWPPs and governance outcomes. For specific examples and details see the "CWPP Document Analysis Sco res and Results" in Chapter VI. Second, the CWPP framed wildfire across multiple frames, such as a hazard and beneficial to ecosystem health and fuel load mitigation service. Finally, CWPPs were considered part of the comprehensive planning process, although it is worth noting this was only due to state mandates, not the choice of the county engaging in CWPP activities. Despite their shortcomings, I still believe CWPPs and local governance do afford the opportunity to significantly reduce wildfire risk, but only if counties are willing to embrace difficult conversations and the conflict that arises during them. Indeed, land development

PAGE 239

222 discussions are often framed in terms of supporting and sustaining economic growth. In fact, the only counties that demonstr ated negative WUI growth were counties that also discussed experiencing negative economic growth. The findings reinforce the need for ensuring multiple frames are included in CWPPs and comprehensive planning efforts. These two documents set the tone by whi ch all other efforts, policies, and strategies are pursued and implemented. This will require designers and planners to be more proactive in ensuring Firewise best practices are a foundational principle in all codes, ordinances, and regulations within fire prone landscapes. The arguments I introduce in this document demonstrate how planners can improve CWPPs and local governance policies in relation to wildfire risk reduction planning outcomes. To ensure tighter integration of best practices between CWPPs and local governance, I suggest using the process outlined in Figure 6.18. Figure 6. 18 . Embedded comprehensive planning and local regulatory CWPP process. Connecting this research study’s findings to theory despite many theories suggesting that community conflicts are bad and need remediated (Few, Brown, & Tompkins, 2007; Glucker, Driessen, Kol hoff, & Runhaar, 2013; Senecah, 2004) , this study supports the emerging

PAGE 240

223 beliefs conflicts are good and need to be openly discussed to ensure all voices and strategies are heard and considered (Lundberg, Richardson, & Hongslo, 2018) . Indeed, Innes and Booher (2014) suggests that if not all key stakeholders are engaged, then the process is not truly collaborative and it is to be considered a failure. This study reinforces three of the tenants of policy learning theory, as discussed by Brody (2003) . In particular, if C WPPs are to improve, they must first establish a precedence of excellence. This precedent has been shown to lead to an increase in learning and improved plans. As such, counties must do a better job of ensuring ongoing, engaged efforts to update CWPPS. Brody (2003) suggests policy issues should not be vague or abstract, but rather be tangible and real, connected to specific sites and properties. CWPPs and local governance efforts that did this typic ally scored higher than those that did not. As Brody (2003) predicted, these same counties exhibited a deeper and more committed engagement with communities. Finally, this research identifies a lar ge gap in the literature. The largest barriers to the utilization of land use regulatory controls of WUI development and wildfire risk reduction practices are economic and political. Future theoretical research should explore the structural, legal, socio e cological, and political ecology nature of land use regulations and landscape scale risk reduction efforts. This projects’ findings show that CWPPs need to better integrate process and content best practices. Additionally, the findings show planners need to improve the integration of wildfire risk reduction frames into strong county visions, while improving all regulatory tools and mechanisms to ensure Firewise land use development.

PAGE 241

224 REFERENCES Abrams, J., Nielsen Pincus, M., Paveglio, T., & Moseley, C. (2016). Community wildfire protection planning in the American West: homogeneity within diversity? Journal of Environmental Planning and Management, 59(3), 557572. doi:10.1080/09640568.2015.1030498 Abukhater, A. (2011). GIS based community risk assessment. Paper presented at the ArcUser, Redlands, CA. Agee, J. K., Bahro, B., Finney, M. A., Omi, P. N., Sapsis, D. B., Skinner, C. N., . . . Phillip Weatherspoon, C. (2000). The use of shaded fuelbreaks in landscape fire management. Forest Ecology and Management, 127, 5566. doi: http://dx.doi.org/10.1016/S03781127(99)001164 Agee, J. K., & Skinner, C. N. (2005). Basic principles of forest fuel reduction treatments. Forest ecology management, 211(1 2), 83 96. Ager, A. A., Day, M. A., McHugh, C. W., Short, K., GilbertsonDay, J., Finney, M. A., & Calkin, D. E. (2014). Wildfire exposure and fuel management on western US national forests. Jour nal of Environmental Management, 145, 5470. doi:10.1016/j.jenvman.2014.05.035 Ager, A. A., Kline, J. D., & Fischer, A. P. (2015). Coupling the Biophysical and Social Dimensions of Wildfire Risk to Improve Wildfire Mitigation Planning. Risk Analysis, 35(8) , 13931406. doi:10.1111/risa.12373 Ager, A. A., Vaillant, N. M., & Finney, M. A. (2011). Integrating fire behavior models and geospatial analysis for wildland fire risk assessment and fuel management planning. Journal of Combustion, 2011. doi:10.1155/2011/572452 Ager, A. A., Vaillant, N. M., Finney, M. A., & Preisler, H. K. (2012). Analyzing wildfire exposure and source sink relationships on a fire prone forest landscape. Forest Ecology and Management, 267, 271283. doi:10.1016/j.foreco.2011.11.021 Albini, F. A. (1976). Estimating wildfire behavior and effects. Paper presented at the General Technical Report INT 30. Alexander, M. E., & Cruz, M. G. (2013). Are the applications of wildland fire behaviour models getting ahead of their evaluation again? Environ mental Modelling and Software, 41, 6571. doi:10.1016/j.envsoft.2012.11.001 Anchor Point Group. (2013). Albany County Community Wildfire Protection Plan. Retrieved from Albany County, WY Anchor Point Group, & AMEC Earth and Environmental. (2011a). Gunnison County, Colorado Community Wildfire Protection Plan Retrieved from Gunnison County, CO

PAGE 242

225 Anchor Point Group, & AMEC Earth and Environmental. (2011b). Montrose County, Colorado Community Wildfire Protection Plan . Retrieved from Montrose County, CO Anderss onsko, Y. (2016). Effective and sustainable flood and landslide risk reduction measures: an investigation of two assessment frameworks. International Journal of Disaster Risk Reduction, 7 , 374392. doi:10.1007/s1375301601065 Arnstein, S. R. (1969). A L adder Of Citizen Participation. Journal of the American Institute of Planners, 35 (4), 216224. doi:10.1080/01944366908977225 Arnstein, S. R. (2019). A Ladder of Citizen Participation. Journal of the American Planning Association, 85(1), 2434. doi:10.1080/ 01944363.2018.1559388 Arroyo, L. A., Pascual, C., & Manzanera, J. A. (2008). Fire models and methods to map fuel types: The role of remote sensing. Forest Ecology and Management, 256(6), 12391252. doi: https://doi.org/10.1016/j.foreco.2008.06.048 Ayres, A., Degolia, A., Fienup, M., Kim, Y., Sainz, J., Urbisci, L., . . . Tague, C. (2016). Social science/natural science perspectives on wildfire and climate change. Geography Compass, 10(2), 6786. Baer, W. C. (1997). General Plan Evaluation Criteria: An Approach to Making Better Plans. Journal of the American Planning Association, 63(3), 329344. doi:10.1080/01944369708975926 Barker, J., & Glenn, A. (2013). Harney County Community Wildfire Protectio n Plan. Retrieved from Harney County, OR Bent County Commissioners. (2011). Community Wildfire Protection Plan: Bent County Fire . Retrieved from Bent County, CO Berke, P., & Godschalk, D. (2009). Searching for the Good Plan: A Meta Analysis of Plan Quali ty Studies. Journal of Planning Literature, 23(3), 227240. doi:10.1177/0885412208327014 Berke, P., Newman, G., Lee, J., Combs, T., Kolosna, C., Berke, P., . . . Salvesen, D. (2015). Evaluation of Networks of Plans and Vulnerability to Hazards and Climate Change: A Resilience Scorecard. Journal of American Planning Association, 81, 287302. doi:10.1080/01944363.2015.1093954 Berke, P. R., & Campanella, T. J. (2006). Planning for Postdisaster Resiliency. The Annals of the American Academy of Political and Soc ial Science, 604, 192207. Berke, P. R., Dixon, J., & Ericksen, N. (1997). Coercive and Cooperative Intergovernmental Mandates: A Comparative Analysis of Florida and New Zealand Environmental Plans. Environment and planning B: Planning and design, 24(3), 451 468. doi:10.1068/b240451

PAGE 243

226 Berke, P. R., Roenick, D. J., Kaiser, E. J., & Burby, R. (1996). Enhancing Plan Quality: Evaluating the Role of State Planning Mandates for Natural Hazard Mitigation. Journal of Environmental Planning and Management, 39(1), 7996. doi:10.1080/09640569612688 Berke, P. R., Yan, S., & Stevens, M. (2009). Integrating Hazard Mitigation into New Urban and Conventional Developments. Journal of Planning Education and Research, 28(4), 441455. doi:10.1177/0739456X09331550 Berry, A. (2007). Forest Policy Up in Smoke: F ire Suppression in the United States , Bozeman, MT. Berry, J. M., West, R. L., & Dennehey, D. M. (1989). Reliability and validity of the Memory Self Efficacy Questionnaire. Developmental Psychology, 25(5), 701. Bhandary, U., & Muller, B. (2009). Land use planning and wildfire risk mitigation: an analysis of wildfire burned subdivisions using highresolution remote sensing imagery and GIS data. Journal of Environmental Planning and Management, 52, 939955. doi:10.1080/09640560903181147 Bihari, M., Hamin, E. M., & Ryan, R. L. (2012). Understanding the Role of Planners in Wildfire Preparedness and Mitigation. ISRN Forestry, 2012, 112. doi:10.5402/2012/253028 Blaine County Land Use Department. (2012). Blaine County, Idaho County Code. Retrieved from https://www.sterlingcodifiers.com/codebook/index.php?book_id=450 Boulder County Board of County Commissioners. (2011). Boulder County Community Wildfire Protection Plan: Wildfire + Unpre pared = Disaster, Management + Community = Protection . Retrieved from Boulder, CO Bradley, G. A. (1984). The urban/forest interface. In G. A. Bradley (Ed.), Land use and forest resources in a changing environment (pp. 316). Seattle, Washington: Universit y of Washington Press. Brenkert smith, H., Meldrum, J. R., Champ, P. A., & Barth, C. M. (2017). Where you stand depends on where you sit: qualitative inquiry into notions of fire adaptation. 22. Brenkert Smith, H. J. J. o. F. (2011). Homeowners' perspecti ves on the parcel approach to wildland fire mitigation: The role of community context in two Colorado communities. 109(4), 193200. Brody, S. D. (2003). Are we learning to make better plans? A longitudinal analysis of plan quality associated with natural hazards. Journal of Planning Education and Research, 23, 191201. doi:10.1177/0739456X03258635 Brody, S. D., & Highfield, W. E. (2005). Does Planning Work?: Testing the Implementation of Local Environmental Planning in Florida. Journal of the American Planning Association, 71(2), 159175. doi:10.1080/01944360508976690

PAGE 244

227 Brody, S. D., Kang, J. E., & Be rnhardt, S. (2010). Identifying factors influencing flood mitigation at the local level in Texas and Florida: the role of organizational capacity. Natural Hazards, 52(1), 167184. doi:10.1007/s1106900993645 Brown, R. T., Agee, J. K., & Franklin, J. F. ( 2004). Forest restoration and fire: principles in the context of place. Conservation Biology, 18, 903912. doi:10.1111/j.1523 1739.2004.521_1.x Brummel, R. F., Nelson, K. C., & Jakes, P. J. (2012). Burning through organizational boundaries? Examining inter organizational communication networks in policy mandated collaborative bushfire planning groups. Global Environmental Change, 22, 516528. doi:10.1016/j.gloenvcha.2011.12.004 Bryant, B. P., & Westerling, A. L. (2014). Scenarios for future wildfire risk in California: links between changing demography, land use, climate, and wildfire. Environmetrics, 25 , 454471. doi:10.1002/env.2280 Brzuszek, R. F., & Walker, J. B. (2008). Trends in community fire ordinances and their effects on landscape architecture prac tice. Landscape Journal, 27, 142153. doi:10.3368/lj.27.1.142 Burby, R., & Deyle, R. (2000). Creating hazard resilient communities through landuse planning. Natural Hazards Review, 1, 99106. doi:10.1061/(ASCE)15276988(2000)1:2(99) Burby, R. J. (2003). M aking plans that matter: Citizen involvement and government action. American Planning Association. Journal of the American Planning Association, 69 (1), 3349. doi: http://dx.doi.org/10.1080/01944360308976292 Burby, R. J., May, P. J., Berke, P. R., Kaiser, E. J., Dalton, L. C., & French, S. P. (1997). Making governments plan: State experiments in managing land use : JHU Press. Burns, M., & Cheng, A. S. (2007). Framing the Need for Active Management for Wildfire Mitigation and Forest Restoration. Society & Natural Resources, 20 (3), 245259. doi:10.1080/08941920601117348 Busenberg, G. (2004). Wildfire management in the United States: the evolution of a policy failure. Review of Policy Research, 21 (2), 145156. Butsic, V., Kelly, M., & Moritz, M. (2015). Land use and wildfire: a review of local interactions and teleconnections. Land, 4, 140156. doi:10.3390/land4010140 CAL FIRE. (2012). Fire Prevention. Retrieved from http://calfire.ca.gov/fire_prevention/fire_prevention CAL FIRE. (2018a). Wildfire is coming. Are you ready toGO. In C. FIRE (Ed.), (pp. 8). Sacramento, CA: CAL FIRE.

PAGE 245

228 CAL FIRE. (2018b). Wildfire is coming. Are youSET. In C . FIRE (Ed.), (pp. 8). Sacramento, CA: CAL FIRE. Campbell County Fire Warden. Campbell County Community Wildfire Protection Plan . Retrieved from Campbell County, WY Carmel, Y., Paz, S., Jahashan, F., & Shoshany, M. (2009). Assessing fire risk using Monte Carlo simulations of fire spread. Forest Ecology and Management, 257, 370377. doi:10.1016/j.foreco.2008.09.039 Carmo, M., Moreira, F., Casimiro, P., & Vaz, P. (2011). Land use and topography influences on wildfire occurrence in northern Portugal. Landscape and Urban Planning, 100, 169176. doi:10.1016/j.landurbplan.2010.11.017 Chakraborty, J., Tobin Graham, A., & M ontz Burrell, E. (2005). Population Evacuation: Assessing Spatial Variability in Geophysical Risk and Social Vulnerability to Natural Hazards. Natural Hazards Review, 6 (1), 2333. doi:10.1061/(ASCE)1527 6988(2005)6:1(23) Champ, P. A., & Brenkert Smith, H. (2016). Is Seeing Believing? Perceptions of Wildfire Risk Over Time. Risk Analysis, 36(4), 816 830. doi:10.1111/risa.12465 Champ, P. A., Donovan, G. H., & Barth, C. M. (2013). Living in a tinderbox: wildfire risk perceptions and mitigating behaviours. International Journal of Wildland Fire, 22(6), 832840. CHAPTER 3. Mountainous, Forest , Brush and GrassCovered Lands, State of California, Division 4. Cong. Rec. Section 4291. (2019). Cheng, A. S., & Sturtevant, V. E. (2012). A framework for assessi ng collaborative capacity in communitybased public forest management. Environmental Management, 49, 675689. doi:10.1007/s0026701198016 Chuvieco, E., Martnez, S., Romn, M. V., Hantson, S., & Pettinari, M. L. (2014). Integration of ecological and soci oeconomic factors to assess global vulnerability to wildfire. Global Ecology and Biogeography, 23, 245258. doi:10.1111/geb.12095 Cilliers, E., & Timmermans, W. (2014). The importance of creative participatory planning in the public place making process ( Vol. 41). Clark, R. N., & Stankey, G. H. (2007). Integrated research in natural resources: the key role of problem framing: DIANE Publishing. Cochise County Board of Supervisors. (2014). Cochise County Community Wildfire Protection Plan . Retrieved from Coc hise County, AZ

PAGE 246

229 Cochrane, M. A., Moran, C. J., Wimberly, M. C., Baer, A. D., Finney, M. A., Beckendorf, K. L., . . . Zhu, Z. (2012). Estimation of wildfire size and risk changes due to fuels treatments. International Journal of Wildland Fire, 21, 357367. doi:10.1071/WF11079 Cohen, J. (2008). The WildlandUrban Interface Fire Problem. Forest History Today, Fall , 2026. doi:10.1071/WF07131 Cohen, J. D. (1991). A Site Specific Approach for Assessing the Fire Risk to Structures at the Wildland Urban Interface . Fire and the Environment : Ecological and Cultural Perspectives international symposium, 69 , 252256. Cohen, J. D. (2000). Examination of the home destruction in Los Alamos associated with the Cerro Grande Fire July 10, 2000. Retrieved from Missoula, Montana: https://www.fs.usda.gov/treesearch/pubs/4686 Cohen, J. D. (2000). Preventing disasters: Home ignitability in the wildlandurban interface. Journal of Forestry, 98, 1521. Cohen, J. D. ( 2001, November). Wildland–urban fire — a different approach. Paper presented at the Proceedings of the Firefigther Safety Summit, International Association of Wildland Fire, Missoula, MT. Cohen, J. D., & Finney, M. A. (2010, November, 2010). An examination of fuel particle heating during fire spread. Paper presented at the Proceedings of the VI International Conference on Forest Fire Research, Coimbra, Portugal. Colburn, J. (2007). The Fire Next Time: Planning Land Uses in the WildlandUrban Interface. 81, 146. Collins, B. M., Kramer, H. A., Menning, K., Dillingham, C., Saah, D., Stine, P. A., & Stephens, S. L. (2013). Modeling hazardous fire potential within a completed fuel treatment network in the northern Sierra Nevada. Forest Ecology and Management, 310, 156166. doi:10.1016/j.foreco.2013.08.015 Collins, T. W. (2008a). The political ecology of hazard vulnerability: marginalization, facilitation and the production of differential risk to urban wildfires in Arizona's White Mountains. Journal of Political E cology, 15, 2155. Collins, T. W. (2008b). What influences hazard mitigation? Household decision making about wildfire risks in Arizona's White Mountains. The Professional Geographer, 60, 508526. doi:10.1080/00330120802211737 Collins, T. W. (2011). House holds, forests, and fire hazard vulnerability in the American West: a case study of a California community. Environmental Hazards, 6, 2337. doi:10.1016/j.hazards.2004.12.003 Community Development Services of Montana. (2008). Butte Silver Bow County Growth Policy . Retrieved from Butte Silver Bow County, MT

PAGE 247

230 Connell, D. J., & Daoust Filiatrault, L. A. (2017). Better Than Good: Three Dimensions of Plan Quality. Journal of Planning Education and Research, 38(3), 265272. doi:10.1177/0739456X17709501 Cooperative Forestry Assistance Act of 1978, Pub. L. No. 95 313 (1978 July 1, 1978). County of Lemhi Wildland Urban Interface: Wildland Fire Hazard, Risk, & Mitigation Plan. (2006). Retrieved from Lemhi County, ID Covington, E . (2003). Bingham County Wild Land Mitigation Fire Plan. Retrieved from Bingham County, ID Cox, W. (2006). Emergency evacuation report Card: 25 urban areas could face greater challenges than New Orleans experienced after Hurrican Katrina . Retrieved from Crook County Commission. Crook County Community Wildfire Protection Plan . Retrieved from Cook County, WY Crow, D. A., Lawhon, L. A., Koebele, E., Kroepsch, A., Schild, R., & Huda, J. (2015). Information, resources, and management priorities: agency outrea ch and mitigation of wildfire risk in the western United States. Risk, Hazards & Crisis in Public Policy, 6 , 6990. doi:10.1002/rhc3.12073 Cruz, M. G., & Alexander, M. E. (2013). Uncertainty associated with model predictions of surface and crown fire rates of spread. Environmental Modelling and Software, 47, 1628. doi:10.1016/j.envsoft.2013.04.004 Cruz, M. G., Alexander, M. E., & Wakimoto, R. H. (2003). Assessing canopy fuel stratum characteristics in crown fire prone fuel types of western North America. I nternational Journal of Wildland Fire, 12(1), 39 50. Cutter, S. L., Boruff, B. J., & Shirley, W. L. (2003). Social Vulnerability to Environmental Hazards*. Social Science Quarterly, 84(2), 242261. doi:10.1111/15406237.8402002 Cutter, S. L., & Finch, C. (2008). Temporal and spatial changes in social vulnerability to natural hazards. Proceedings of the National Academy of Sciences, 105(7), 2301. doi:10.1073/pnas.0710375105 Dalton, L. C., & Burby, R. J. (1994). Mandates, plans, and planners: Building local commitment to development management. American Planning Association. Journal of the American Planning Association, 60(4), 444. Dalton, L. C., Conover, M., Rudholm, G., Tsuda, R., & Baer, W. C. (1989). The Limits of Regulation Evidence from Local Plan Impl ementation in California. Journal of the American Planning Association, 55(2), 151168. doi:10.1080/01944368908976015

PAGE 248

231 De Leo, D., & Forester, J. (2017). Reimagining planning: moving from reflective practice to deliberative practice a first exploration in the Italian context. Planning Theory & Practice, 18 (2), 202216. doi:10.1080/14649357.2017.1284254 Deyle, R. E., & Smith, R. A. (2000). Riskbased taxation of hazardous land development. Journal of the American Planning Association, 66, 421434. doi:10.1080/01944360008976125 Dondo Bhler, M., Curth, M. d. T., & Garibaldi, L. A. (2013). Demography and socioeconomic vulnerability influence fire occurrence in Bariloche (Argentina). Landscape and Urban Planning, 110, 6473. doi:10.1016/j.landurbplan.2012.10.006 Ebert, U., & Welsch, H. (2004). Meaningful environmental indicators: a social choice approach. Journal of Environmental Economics and Management, 47, 270283. Elia, M., Lafortezza, R., Colangelo, G., & Sanesi, G. (2014). A streamlined approach for the s patial allocation of fuel removals in wildland –urban interfaces. Landscape Ecology, 29, 17711784. doi:10.1007/s1098001400707 Few, R., Brown, K., & Tompkins, E. L. (2007). Public participation and climate change adaptation: avoiding the illusion of incl usion. Climate Policy, 7 (1), 4659. doi:10.1080/14693062.2007.9685637 Finney, M. A. (2001). Design of regular landscape fuel treatment patterns for modifying fire growth and behavior. Forest Science, 47 , 219228. Finney, M. A. (2002). Fire growth using mi nimum travel time methods. Canadian Journal of Forest Research, 32 , 14201424. doi:10.1139/x02068 Finney, M. A. (2006). An overview of FlamMap fire modeling capabilities. Fuel Management How to Measure Success: Conference Proceedings. Fort Collins, CO. Fi nney, M. A., McHugh, C. W., Grenfell, I. C., Riley, K. L., & Short, K. C. (2011). A simulation of probabilistic wildfire risk components for the continental United States. Stochastic Environmental Research and Risk Assessment, 25 , 9731000. doi:10.1007/s004770110462z Fire History Society. U.S. Forest Service history fire suppression. U.S. Forest Service History, Policy & Law, Fire Suppression. FireSafe Council of Siskiyou County. (2008). Siskiyou County Wildfire Protection Plan. Retrieved from Siskiyou County, CA Fisher, D. R. (2006). Bringing the Material Back In: Understanding the U.S. Position on Climate Change. Sociological Forum, 21(3), 467494. doi:10.1007/s1120600690262

PAGE 249

232 Flanagan Barry, E., Gregory Edward, W., Hallisey Elaine, J., Heitgerd Janet, L., & Lewis, B. (2011). A Social Vulnerability Index for Disaster Management. In Journal of Homeland Security and Emergency Management (Vol. 8). Fleeger, W. E. (2008). Collaborating for Success: Community Wildfire Protection Pl anning in the Arizona White Mountains. Journal of Forestry, 106(2), 78 82. Fleeger, W. E., & Becker, M. L. (2008). Creating and sustaining community capacity for ecosystem based management: Is local government the key? Journal of Environmental Management, 88, 13961405. doi:10.1016/j.jenvman.2007.07.018 Fleeger, W. E., & Becker, M. L. (2010). Decision processes for multijurisdictional planning and management: community wildfire protection planning in Oregon. Society & Natural Resources, 23 , 351365. doi:10.1080/08941920802120067 Forthofer, J. M., Butler, B. W., & Wagenbrenner, N. S. (2014). A comparison of three approaches for simulating fine scale surface winds in support of wildland fire management. Part I. Model formulation and comparison against measure ments. International Journal of Wildland Fire, 23, 969981. doi:10.1071/WF12089 Frazier, T. G., Walker, M. H., Kumari, A., & Thompson, C. M. (2013). Opportunities and constraints to hazard mitigation planning. Applied Geography, 40, 5260. doi: https://doi.org/10.1016/j.apgeog.2013.01.008 Fullerton Management Group. (2011). Calaveras County Community Wildfire Protection Plan. Retrieved from Calaveras County, CA Gaither, C. J., Poudyal, N. C., Goodrick, S., Bowker, J. M., Malone, S., & Gan, J. (2011). Wildland fire risk and social vulnerability in the Southeastern United States: An exploratory spatial data analysis approach. Forest Policy and Economics, 13(1), 2436. doi: https://doi.org/10.1016/j.forpol.2010.07.009 Gardner, P. D., Cortner, H. J., & Widaman, K. (1987). The risk perceptions and policy response toward wildland fire hazards by urban home owne rs. Landscape and Urban Planning, 14, 163172. doi:10.1016/01692046(87)900211 Gebru, T., Krause, J., Wang, Y., Chen, D., Deng, J., Aiden, E. L., & Fei Fei, L. (2017). Using deep learning and Google Street View to estimate the demographic makeup of neighborhoods across the United States. Proceedings of the National Academy of Sciences, 114(50), 1310813113. doi:10.1073/pnas.1700035114 %J Proceedings of the National Academy of Sciences Geographic Area Coordination Centers. Common terms NFDRS. National GACO Website Portal. Retrieved from https://gacc.nifc.gov/rmcc/predictive/fuels_fire danger/drgloss.htm

PAGE 250

233 Gill, A. M., & Stephens, S. L. (2009). Scientific and social challenges for the management of fire prone wildland–urban interfaces. Environmental Research Letters, 4 , 034014. doi:10.1088/17489326/4/3/034014 Glucker, A. N., Driessen, P. P. J., Kolhoff, A., & Runhaar, H. A. C. (2013). Public participation in environmental impac t assessment: why, who and how? Environmental Impact Assessment Review, 43 , 104111. doi: https://doi.org/10.1016/j.eiar.2013.06.003 Godschalk, D., Kaiser, E. J., & Berke, P. (1998). Integrating haz ard mitigation and local landuse planning. In R. J. Burby (Ed.), Cooperating with nature: confronting natural hazards with landuse planning (pp. 85118). Washington, D.C.: Joseph Henry Press. Collaborating for transformative resilience: shared identity in the U.S. Fire Learning Network 339358 (MIT Press 2012). GonzlezCabn, A., Haynes, R. W., McCaffrey, S., Mercer, E., & Watson, A. (2007). Fire social science research –selected highlights . Retrieved from Portland, Oregon Goodchild, B. (1990). Planning and the Modern/Postmodern Debate. The Town Planning Review, 61 (2), 119137. Google (Cartographer). Streetview, digital images Gorte, R. (2013). The Rising cost of wildfire protection. Paper presented at the Headwaters Economics, Bozeman, MT. Gov ernor's Office of Planning and Research. (2014). Fire Hazard Planning: General Plan Technical Advice Series. Retrieved from Sacramento, CA Graham, R., Finney, M., McHugh, C. W., Cohen, J., Calkin, D. E., Stratton, R. D., . . . Nikolov, N. (2012). Fourmile Canyon Fire Findings . Gray, B. (2004). Strong opposition: frame based resistance to collaboration. Journal of Community & Applied Social Psychology, 14(3), 166176. doi:10.1002/casp.773 Gray, R., Kouhy, R., & Lavers, S. (1995). Corporate social and envir onmental reporting: a review of the literature and a longitudinal study of UK disclosure. Accounting, Auditing Accountability Journal, 8 (2), 4777. Grayzeck Souter, S. A., Nelson, K. C., Brummel, R. F., Jakes, P., & Williams, D. R. (2009). Interpreting fe deral policy at the local level: the wildland –urban interface concept in wildfire protection planning in the eastern United States. International Journal of Wildland Fire, 18(3), 278289. Green, K., Finney, M., Campbell, J., Weinstein, D., & Landrum, V. ( 1995). Fire. Using gis to predict fire behavior. Journal of Forestry, 93, 21 25.

PAGE 251

234 Greiving, S., Fleischhauer, M., & Lckenktter, J. (2006). A Methodology for an integrated risk assessment of spatially relevant hazards. Journal of Environmental Planning and Management, 49, 119. doi:10.1080/09640560500372800 Guerin, E. (2018). New houses, built to fire code, burned down anyway in California's 2017 Thomas Fire. California building codes are not keeping up with wildfires. Retrieved from https://cronkitenews.azpbs.org/2018/12/28/california new homes built to fire code burneddownanyway/ Guyadeen, D., & Seasons, M. (2016a). Evaluation Theory and Practice: Comparing Program Evaluation and Evaluation in Planning. Journal of Planning Education and Research, 38(1), 98110. doi:10.1177/0739456X16675930 Guyadeen, D., & Seasons, M. (2016b). Plan Evaluation: Challenges and Directions for Future Research. Planning Practice & Research, 31(2), 215228. doi:10.1080/02697459.2015.1081335 Haas, J. R., Calkin, D. E., & Thompson, M. P. (2013). A national approach for integrating wildfire simulation modeling into wildland urban interface risk assessments within th e United States. Landscape and Urban Planning, 119, 4453. doi:10.1016/j.landurbplan.2013.06.011 Haines, T. K., Renner, C. R., & Reams, M. A. (2008). A Review of state and local regulation for wildfire mitigation. In T. P. Holmes, J. P. Presetemon, & K. L. Abt (Eds.), The Economics of Forest Disturbances: Wildfires, Storms, and Invasive Species (Vol. 79, pp. 273293). Dordrecht: Springer. Harris, L. M., McGee, T. K., & McFarlane, B. L. (2011). Implementation of wildfire risk management by local governments in Alberta, Canada. Journal of Environmental Planning and Management, 54, 457475. doi:10.1080/09640568.2010.515881 Hart, A. (2018). California wildfires reveal alert system shortcomings. Just and public safety. Retrieved from https://www.govtech.com/public safety/California Wildfires Reveal AlertSystem Shortcomings.html Hayes, J. L., Ager, A. A., & Barbour, R. J. (2004). Methods for integrated modeling of landscape change: Interior Northwest Landscape Analysis System. Paper presented at the General technical report PNW, Portland, OR. Headley, R. (1916). Fire suppression , Washington, DC. Headwaters Economics. (2014). Reducing wildfire risks to communities: solutions for controlling the pace, scale, and pattern of future development in the wildlandurban interface , Bozeman, MT. Headwaters Economics. (2016a). Do insurance policies and rates influence home development on fire prone lands?, Bozeman, MT.

PAGE 252

235 Headwaters Economics. (2016b). Land use planning to reduce wildfire risk: lessons from five western cities. Retrieved from Bozeman, MT The Healthy Forest Restoration Act of 2003, Pub. L. No. 10 8148, H.R. 1904 Stat. (2003 12/03/2003). Heyerdahl, E. K., Brubaker, L. B., & Agee, J. K. (2001). Spatial controls of historical fire regimes: a multiscale example from the interior west, USA. Ecology, 82, 660678. doi:10.1890/00129658(2001)082[0660:SCOH FR]2.0.CO;2 Holland, M., March, A., Yu, J., & Jenkins, A. (2013). Land use planning and bushfire risk: CFA referrals and the February 2009 Victorian fire area. Urban Policy and Research, 31, 4154. doi:10.1080/08111146.2012.736255 Horney, J., Nguyen, M., S alvesen, D., Dwyer, C., Cooper, J., & Berke, P. (2017). Assessing the quality of rural hazard mitigation plans in the southeastern United States. Journal of Planning Education and Research, 37, 5665. doi:10.1177/0739456X16628605 Horney, J., Nguyen, M., Sa lvesen, D., Tomasco, O., & Berke, P. (2016). Engaging the public in planning for disaster recovery. International Journal of Disaster Risk Reduction, 17, 3337. doi: http://dx.doi.org/10.1016/j.i jdrr.2016.03.011 Hughes, R., & Mercer, D. (2009). Planning to Reduce Risk: The Wildfire Management Overlay in Victoria, Australia. Geographical Research, 47(2), 124141. doi:10.1111/j.17455871.2008.00556.x Husari, S., Nichols, T. H., Sugihara, N. G., & St ephens, S. L. (2006). Fire and fuel management. In N. G. Sugihara, J. W. Van Wagtendonk, J. Fites Kaufman, K. E. Shaffer, & A. E. Thode (Eds.), Fire in California's ecosystems (pp. 444465): University of California Press. Innes, J. E., & Booher, D. E. (1999). Consensus building and complex adaptive systems: A framework for evaluating collaborative planning. Journal of the American Planning Association, 65, 412423. doi:10.1080/01944369908976071 Innes, J. E., & Booher, D. E. (2014). A turning point for plan ning theory? Overcoming dividing discourses. Planning Theory, 14(2), 195213. doi:10.1177/1473095213519356 International Code Council Inc. (2003). 2003 International UrbanWildland Interface Code . Retrieved from Country Club Hills, IL: Jakes, P. J., Burns , S., Cheng, A., Saeli, E., Nelson, K., Brummel, R., & Grayzeck, S. (2007). Critical elements in the development and implementation of community wildfire protection plans (CWPPs). The fire environment — innovations, management, and policy; conference proceedings RMRSP 46CD . Destin, FL: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station.

PAGE 253

236 Jakes, P. J., Esposito, C., Burns, S., Nelson, K. C., Cheng, A. S., Sturtevant, V. E., & Williams, D. R. (2012). Best management practices for c reating a community wildfire protection plan, Newton Square, PA. Jakes, P. J., Nelson, K. C., Enzler, S. a., Burns, S., Cheng, A. S., Sturtevant, V., . . . Staychock, E. (2011). Community wildfire protection planning: is the Healthy Forests Restoration Act 's vagueness genius? International Journal of Wildland Fire, 20, 350363. doi:10.1071/WF10038 Jakes, P. J., & Sturtevant, V. (2013). Trial by fire: Community Wildfire Protection Plans put to the test. International Journal of Wildland Fire, 22(8), 11341143. Johansen, C., Horney, J., & Tien, I. (2017). Metrics for evaluating and improving community resilience. Journal of Infrastructure Systems, 23, 1 11. doi:10.1061/(ASCE)IS.1943555X.0000329. John, L. K., Loewenstein, G., & Prelec, D. (2012). Measuring the prevalence of questionable research practices with incentives for truth telling. Psychological science, 23(5), 524 532. Kalies, E. L., & Yocom Kent, L. L. (2016). Tamm Review: Are fuel treatments effective at achieving ecological and social objectives? A systematic review. Forest Ecology and Management, 375, 8495. doi: https://doi.org/10.1016/j.foreco.2016.05.021 Kane, J. M., Varner, J. M., Knapp, E. E., & Powers, R. F. (2010). Understory vegetation response to mechanical mastication and other fuels treatments in a ponderosa pine forest. Applied Vegetation Science, 13(2), 207220. doi:10.1111/j.1654109X.2009.01062.x Katelman, T. (2005). Del Norte Fire Safe Plan: Community Wildfire Protection Plan . Retrieved from Del Norte County, CA Keane, R. E., Drury, S. a., Karau, E. C., Hessburg, P. F., & Reynolds, K. M. (2010). A method for mapping fire hazard and risk across multiple scales and its application in fire management. Ecological Modelling, 221, 218. doi:10.1016/j.ecolmodel.2008.10.022 Keane, R. E., Garner, J. L., Schmidt, K. M., Long, D. G., Menakis, J. P., & Finney, M. A. (1998). Development of input data layers for the FARSITE fire growth model for the Selway Bitterroot wilderness complex, USA. Paper presented at the General Technical Report RMRS GTR3, Ogden UT. Keane, R. E., Mincemoyer, S. A., Schmidt, K. M., & Garner, J. L. (2000). Mapping vegetation and fuels for fire management on the Gila National Forest Complex, New Mexico. Paper presented at the USDA Forest Service RMRS GTR 46CD, Missoula, MT. Kenitzer, Z. E. (2016). How Plan Implementation Fails: Examining the role of Experience, Expectations, and Externalities. The Ohio State University, Retrieved from http://rave.ohiolink.edu/etdc/view?acc_num=osu1465815464

PAGE 254

237 Kim, W. H., & Tran, T. (2018). An Evaluation of Local Comprehensive Plans Toward Sustainable Green Infrastructure in US. Sustainability, 10(11). doi:10.3390/su10114143 Kocher, S., & Butsic, V. (2017). Governance of Land Use Planning to Reduce Fire Risk to Homes Mediterranean Fr ance and California. Land, 6, 24. doi:10.3390/land6020024 In harm’s way: homeowner behavior and wildland fire policy 178199 (Taylor & Francis for RFF Press 2011). Kramer, H. a., Collins, B. M., Kelly, M., & Stephens, S. L. (2014). Quantifying ladder fuels: a new approach using LiDAR. Forests, 5 , 14321453. doi:10.3390/f5061432 Kreye, J. K., Kobziar, L. N., & Zipperer, W. C. (2013). Effects of fuel load and moisture content on fire behaviour and heating in masticated litter dominated fuels. International Journal of Wildland Fire, 22(4), 440445. LANDFIRE. (2010). The LANDFIRE Project: supporting fire and land management across the nation. Lasky, D. (2018a). Fantastic failure: false hope and the Four Mile Canyon Fire. Retrieved from https://fireadaptednetwork.org/fantastic failure false hope and the four mile cany onfire/ Lasky, D. (2018b). Fantastic Failure: False Hope and the Four Mile Canyon Fire. Fantastic Failure. Retrieved from https://fireadaptednetwork.org/resource/notesfield falsehope four mile canyon/ Laurian, L., Crawford, J., Day, M., Kouwenhoven, P., Mason, G., Ericksen, N., & Beattie, L. (2010). Evaluating the Outcomes of Plans: Theory, Practice, and Methodology. Environment and planning B: Planning and design, 37(4), 740757. doi:10.1068/b35051 Lerch, A., & Chambers, C. (2019). How Ashland passed a wildfire ordinance (hint: we removed our ordinance blinders). Fire adapted communities learning network. Retrieved from https://fireadaptednetwork.org/how ashlandpassed a wildfire ordinance/?utm_source=Fire+Adapted+Communities+Learning+Network+Blog&utm_ca mpaign=46cd04af4f FAC+Blog+Updates+Campaign&utm_medium=email&utm_term=0_95a7a7b25246cd04af4f 202251969 Lincoln County B oard of Commissioners. (2008). The code of the west . Retrieved from Lincoln County, NM: https://www.lincolncountynm.gov/wpcontent/uploads/2017/12/Code_of _the_West_1c_bw.pdf Local government: planning and zoning: wildfires., SB 182, California Senate (2019). Loh, C. G. (2011). Four Potential Disconnects in the Community Planning Process. Journal of Planning Education and Research, 32(1), 3347. doi:10.1177/ 0739456X11424161

PAGE 255

238 Loh, C. G., & Norton, R. K. (2013). Planning Consultants and Local Planning. Journal of the American Planning Association, 79(2), 138147. doi:10.1080/01944363.2013.883251 Lucas, J. (2018). California’s disaster communication plan failed during deadly wildfires. Retrieved from https://www.ravemobilesafety.com/blog/californias disaster communicationplan Lundberg, A. K., Richardson, T., & Hongslo, E . (2018). The consequences of avoiding conflict: lessons from conservation planning for Europe's last wild reindeer. Journal of Environmental Planning and Management , 120. doi:10.1080/09640568.2017.1409197 Lyles, W., Berke, P., & Smith, G. (2014). A compa rison of local hazard mitigation plan quality in six states , USA. Landscape and Urban Planning, 122, 8999. doi:10.1016/j.landurbplan.2013.11.010 Lyles, W., & Stevens, M. (2014). Plan Quality Evaluation 1994–2012: Growth and Contributions, Limitations, an d New Directions. Journal of Planning Education and Research, 34 (4), 433450. doi:10.1177/0739456X14549752 Lynn, K. (2003). Wildfire and rural poverty: disastrous connections. Natural Hazards Observer. Lynn, K., Ojerio, R., & Wolf, J. (2008). Curry County Community Wildfire Protection Plan . Retrieved from Curry County, OR Mandelker, D. R. (1976). The Role of the Local Comprehensive Plan in Land Use Regulation. Michigan Law Review, 74(5), 899973. doi:10.2307/1287830 Manning, R. (2012). In a major policy re versal, the Forest Service is fighting every fire this year -but at what cost? In OnEarth Magazine (pp. 4 7). New York, NY. Manson, S., Schroeder, J., Riper, D. V., & Ruggles, S. (2018). IPUMS National Historical Geographic Information System [Database]. March, A. (2010). Practising theory: When theory affects urban planning. Planning Theory, 9(2), 108125. doi:10.1177/1473095209357864 May, P. J., Burby, R. J., Ericksen, N. J., Handmer, J. W., Dixon, J. E., Michaels, S., & Smith, D. I. (1997). Environmental Management and Governance: Intergovernmental Approaches to Hazards and Sustainability. Environmental International, 23(2), 266266. McCaffrey, S., Rhodes, A., & Stidham, M. (2015). Wildfire evacuation and its alternatives: perspectives from four Unite d States’ communities. International Journal of Wildland Fire, 24, 170178. doi: http://dx.doi.org/10.1071/WF13050 McCright, A. M., & Dunlap, R. E. (2011). The Politicization of Climate Change and Polarizati on in the American Public's Views of Global Warming, 2001–2010. The Sociological Quarterly, 52 (2), 155194. doi:10.1111/j.15338525.2011.01198.x

PAGE 256

239 McHugh, M. L. (2012). Interrater reliability: the kappa statistic. Biochemia medica: Biochemia medica, 22(3), 276282. Meldrum, J. R., Barth, C. M., Falk, L. C., Brenkert Smith, H., Warziniack, T., & Champ, P. A. (2015). Living with wildfire in Delta County, Colorado: cross community comparisons. Paper presented at the Research Note RMRS RN 67, Fort Collins, CO. Menakis, J. P., Cohen, J., & Bradshaw, L. (2003). Mapping wildland fire risk to flammable structures for the conterminous United States. Fire Conference 2000: The First National Congress o n Fire Ecology, Prevention, and Management. Tallahassee, FL: Tall Timbers Research Station. Mendocino County Board of Supervisors, Mendocino County Fire Chiefs' Association, & California Department of Forestry and Fire Protection. (2005). Mendocino County Community Wildfire Protection Plan . Retrieved from Mendocino County, CA Metzenbaum, J. (195 7). The History of Zoning--A Thumbnail Sketch. Western Reserve Law Review, 9 , 36 42. Miller, C., & Ager, A. A. (2013). A review of recent advances in risk analysis for wildfire management. International Journal of Wildland Fire, 22, 114. doi:10.1071/WF11114 Mintier & Associates, Jones & Stokes Associates, Lowens, S., & Del Norte County Community Development Department. (2003). Del Norte County General Plan. Retrieved from Mohave County Board of Supervisors. (2008). Mohave County Community Wildfire Protec tion Plan . Retrieved from Mohave County, AZ Moritz, M., & Anderson, S. (2018). Wildfires are inevitable fatalities and homes losses are not. The Conversation Environment. Retrieved from https://www.scientificamerican.com/article/wildfires are inevitable mdash fatalities andhomes lossesare not/ Moritz, M. A., Batllori, E., Bradstock, R. A., Gill, A. M., Handmer, J., Hessburg, P. F., . . . Syphard, A. D. (2014). Learning to coexist with wildfire. In Nature (Vol. 515, pp. 5866). Mowery, M. (2008). Wildfire and development: why stronger links to landuse planning are needed to save lives, protect property, and minimize economic risk. (Masters in City Planning), Massachusetts Institute of Technology, Boston, MA. Mowery, M., Read, A., Johnston, K., & Wafaie, T. (2019). Planning the wildlandurban interface (PAS Report 594). Muir, W. D. (1941). Forest fire control in the United Sta tes of America , New Haven, CT. Muller, B., & Schulte, S. (2011). Governing wildfire risks: what shapes county hazard mitigation programs? Journal of Planning Education and Research, 31, 6073. doi:10.1177/0739456X10395895

PAGE 257

240 National Association of State Fore sters. (2017). Communities at Risk Report Fiscal Year 2016. Retrieved from Washington, D.C.: National Fire Protection Association. (2016a). The Basics of defensible space and the “home ignition zone”. Firewise Communities . National Fire Protection Asso ciation. (2016b). Firewise Toolkit. Wildfire Preparedness. National Forest Foundation. (2019). Grant Programs & Resources. Retrieved from https://www.nationalforests.org/grant programs Nationa l Interagency Fire Center. (2018). Federal fire fighting costs (suppression only). Fire Information Statistics. Retrieved from https://www.nifc.gov/fireInfo/fireInfo_statistics.html Nat ional Volunteer Fire Council. (2019). Wildland Fire Assessment Program Home Assessment Checklist. In N. V. F. Council (Ed.): National Volunteer Fire Council. Nelson, K. C., Souter, S. G., Jakes, P. J., & Williams, D. R. (2010). Social learning in a policy mandated collaboration: community wildfire protection planning in the eastern United States AU Brummel, Rachel F. Journal of Environmental Planning and Management, 53(6), 681699. doi:10.1080/09640568.2010.488090 Nelson, R. M. (2000). Prediction of diurnal change in 10h fuel stick moisture content. Canadian Journal of Forest Research, 30, 10711087. Newig, J., & Fritsch, O. (2009). Environmental governance: participatory, multilevel – and effective? Environmental Policy and Governance, 19(3), 197214. doi:10.1002/eet.509 Nielsen pincus, M., Ribe, R. G., & Johnson, B. R. (2015). Spatially and socially segmenting private landowner motivations, properties, and management: a typology for the wildland urban interface. Landscape and Urban Planning, 137, 112. doi:10.1016/j.landurbplan.2014.11.020 North, M., Collins, B. M., & Stephens, S. (2012). Using Fire to Increase the Scale, Benefits, and Future Maintenance of Fuels Treatments. Journal of Forestry, 110(7), 392401. doi:10.5849/jof.12021 Northwest Management Inc. (2004). Minidoka County, Idaho, WildlandUrban Interface Wildfire Mitigation Plan . Retrieved from Minidoka County, ID Northwest Management Inc. (2008 ). Asotin County, Washington Community Wildfire Protection Plan . Retrieved from Asotin County, WA Northwest Management Inc. (2014). Fremont County, Wyoming Community Wildfire Protection Plan . Retrieved from Fremont County, WY

PAGE 258

241 Northwest Management Inc., & Adams County Wildfire Urban Interface Wildfire Mitigation Plan Committee. (2004). Adams County, Idaho, WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Adams County, ID Norton, R. K. (2008). Using content analysis to evaluate local maste r plans and zoning codes. Land Use Policy, 25 (3), 432454. doi: https://doi.org/10.1016/j.landusepol.2007.10.006 Norton, R. K., David, N. P., Buckman, S., & Koman, P. D. (2018). Overlooking the coast: Limited local planning for coastal area management along Michigan’s Great Lakes. Land Use Policy, 71 , 183203. doi: https://doi.org/10.1016/j.landusepol.2017.11.049 Oldham, C. H. (2016). Wildfire liability and the federal government: a double edged sword. Arizona State Law Journal, 48, 15. Oliveira, V., & Pinho, P. (2010). Evaluation in Urban Planning: Advances and Prospects. Journal of Planning Literature, 24(4), 343361. doi:10.1177/0885412210364589 One Hundred Eighth Congress of the United States of America. (2003). Healthy Forest Restoration Act of 2003. Oregon Department of Forestry. Fire Protection in the Wildland Urban Interface. Fire. Retrieved from https://www.oregon.gov/ODF/Fire/Pages/UrbanInterface.aspx Ottmar, R. D., Blake, J. I., & Crolly, W. T. (2012). Using fine scale fuel measurements to assess wildland fuels, potential fire beha vior and hazard mitigation treatments in the southeastern USA. Forest Ecology and Management, 273, 13. doi:10.1016/j.foreco.2011.11.003 Papadopoulos, G. D., & Pavlidou, F. N. (2011). A comparative review on wildfire simulators. IEEE Systems Journal, 5, 233243. doi:10.1109/JSYST.2011.2125230 Parks, S. A., Miller, C., Parisien, M. A., Holsinger, L. M., Dobrowski, S. Z., & Abatzoglou, J. (2015). Wildland fire deficit and surplus in the western United States, 1984–2012. Ecosphere, 6 (12), 113. doi:10.1890/ES1500294.1 Paterson, R. G. (2007). Wildfire hazard mitigation as "safe" smart growth. In A. R. Troy & R. G. Kennedy (Eds.), Living on the edge: economic, institutiona and management perspectives on wildfire hazard in the urban interface (Vol. 6, pp. 4371). Oxford, UK: Elsevier Ltd. Paveglio, T. B., Jakes, P. J., Carroll, M. S., & Williams, D. R. J. E. M. (2009). Understanding Social Complexity Within the Wildland–Urban Interface: A New Species of Human Habitation? , 43(6), 10851095. doi:10.1007/s002670099282z Paveglio, T. B., Prato, T., & Hardy, M. (2013). Simulating effects of land use policies on extent of the wildland urban interface and wildfire risk in Flathead County, Montana. Journal of Environmental Management, 130, 2031. doi:10.1016/j.jenvman.2013.08.036

PAGE 259

242 Poudyal, N. C., JohnsonGaither, C., Goodrick, S., Bowker, J. M., & Gan, J. (2012). Locating spatial variation in the association between wildland fire risk and social vulnerability across six southern states. Environmental Management, 49, 623635. doi:10.1007/s002670119796z Prater Carla, S., & Lindell Michael, K. (2000). Politics of Hazard Mitigation. Natural Hazards Review, 1 (2), 7382. doi:10.1061/(ASCE)15276988(2000)1:2(73) Prowers County Commissioners. (2013). Commun ity Wildfire Protection Plan Prowers County . Retrieved from Prowers County, CO Quarles, S. L., Natural, U., Advisor, R., & County, C. C. (2010). Home Survival in Wildfire Prone Areas : Building Materials and Design Considerations. 1 22. Radeloff, V. C., Hammer, R. B., Stewart, S. I., Fried, J. S., Holcomb, S. S., & McKeefry, J. F. (2005). The wildland urban interface in the United States. Ecological Applications, 15, 799805. doi:10.1890/04 1413 Radeloff, V. C., Helmers, D. P., Kramer, H. A., Mockrin, M. H., Alexandre, P. M., Bar Massada, A., . . . Stewart, S. I. (2017). The 19902010 wildlandurban interface of the conterminous United States . Geospatial data. Reams, M. A., Haines, T. K. T., Renner, C. R., Wascom, M. W., & Kingre, H. (2005). Goals, obstac les and effective strategies of wildfire mitigation programs in the wildland –urban interface. Forest Policy and Economics, 7, 818826. doi:10.1016/j.forpol.2005.03.006 Reinhardt, E. D., Keane, R. E., Calkin, D. E., & Cohen, J. D. (2008). Objectives and considerations for wildland fuel treatment in forested ecosystems of the interior western United States. Forest Ecology and Management, 256(12), 19972006. doi: https://doi.org/10.1016/j.foreco.2008.09.016 Resource Innovations Institute for a Sustainable Environment. (2008). Community wildfire protection plan evaluation guide , Eugene, OR. Riccardi, C. L., Ottmar, R. D., Sandberg, D. V., Andreu, A., Elman, E., Kopper, K., & Long, J. (2007). The fuelbed: a key element of the Fuel Characteristic Classification SystemThis article is one of a selection of papers published in the Special Forum on the Fuel Characteristic Classification System. Canadian Journal of Forest Research, 37(12), 23942412. doi:10.1139/X07143 Rice, C. L., & Davis, J. (1991). Landuse planning may reduce fire damage in the urbanwildland intermix . Retrieved from Berkeley, CA Riebsame, W. E., Gosnell, H., & Theobald, D. M. (1996). Land use and landscape change in the Colorado mountains I: theory, scale, and pattern. Mountain Research and Development, 16, 395405. doi:10.2307/3673989

PAGE 260

243 Rodman, S., & Stram, J. (2008). Community guide to preparing and implementing a community wildfire protection plan. Paper presented at the A supplemental re source guide to preparing a community wildfire protection plan: a handbook for wildland –urban interface communities, March 2004, Eugene, OR. Rothermel, R. C. (1972). A mathematical model for predicting fire spread in wildland fuels. Paper presented at the Research Paper INT 115, Ogden, UT. Russell, K., Campbell, M., & Root, D. (2011). Community Wildfire Protection Plan for Unincorporated El Paso County "A Continuing Process". Retrieved from El Paso County, CO Sabatier, P., & Mazmanian, D. (1980). The Implementation of Public Policy: A Framework of Analysis. The Policy Studies Journal, 8 (4), 538560. Sang, L., Zhang, C., Yang, J., Zhu, D., & Yun, W. (2011). Simulation of land use spatial pattern of towns and villages based on CA Markov model. Mathematical and Computer Modelling, 54, 938943. doi:10.1016/j.mcm.2010.11.019 Sanoff, H. (2000). Community participation methods in design and planning: John Wiley & Sons. SB1241. Kehoe Land use: general plan: safety element: fire hazard impacts., 1241, California State Senate (2012). Schlosser, W. E., & Mierzwinski, J. F. (2012). Benewah County WildlandUrban Interface Wildfire Mitigation Plan Update . Retrieved from Benewah County, ID Schoennagel, T., Balch, J. K., Brenkert Smith, H., Dennison, P. E., H arvey, B. J., Krawchuk, M. A., . . . Whitlock, C. (2017). Adapt to more wildfire in western North American forests as climate changes. Proceedings of the National Academy of Sciences, 114(18), 4582. doi:10.1073/pnas.1617464114 Schwab, J., Meck, S., & Simone, J. (2005). Planning for wildfires : American Planning Association Chicago. Seasons, M. (2003). Monitoring and Evaluation in Municipal Planning: Considering the Realities. Journal of the American Planning Association, 69(4), 430440. doi:10.1080/01944360308976329 SEC Inc. (2006). Community Wildfire Protection Plan Ciboloa County, NM . Retrieved from Cibola County, NM Senecah, S. L. (2004). The trinity of voice: The role of practical theory in planning and evaluating the effectiveness of environmental parti cipatory processes. In S. Depoe, J. W. Delicath, & M. F. A. Elsenbeer (Eds.), Communication and public participation in environmental decision making (pp. 1333). Albany, NY: SUNY Press.

PAGE 261

244 Shahab, S., Clinch, J. P., & O’Neill, E. (2017). Impact based planning evaluation: Advancing normative criteria for policy analysis. Environment and Planning B: Urban Analytics and City Science, 46(3), 534550. doi:10.1177/2399808317720446 Shapero, M. (2018). When wildfire hits the ranch: lessons learned from the Thomas Fir e. Retrieved from https://fireadaptednetwork.org/wildfire hits ranch lessons learned thomas fire/ Shoshone County Community Wildfire Protection Plan Committee. (2011). Shoshone County, Idaho Community Wildfire Protection Plan. Retrieved from Shoshone County, ID Silcox, F. A. (1910). Fire prevention and control on the national forests. Paper presented at the Yearbook of Department of Agriculture, Washington, DC. Simon, G. L. (2016). Flame and fortune in the American West: urban development, environmental change and the Great Oakland Hills Fire. Oakland, CA: University of California Press. Smith, A. M. S., Kolden, C. A., Paveglio, T. B., Cochrane, M. A., Bowman, D. M. J. S., Moritz, M. A., . . . Abatzoglou, J. T. (2016). The Science of Firescapes: Achieving Fire Resilient Communities. BioScience, 66(2), 130 146. doi:10.1093/biosci/biv182 Smith, J. W., Leahy, J. E., Anderson, D. H., & Davenport, M. A. (2013). Communi ty/Agency Trust and Public Involvement in Resource Planning. Society & Natural Resources, 26 , 452471. doi:10.1080/08941920.2012.678465 Society of American Foresters. (2004). Preparing a community wildfire protection plan: a handbook for wildland urban int erface communities, Bethesda, MD. Southard, L. F. (2011). The history of cooperative forest fire control and the Weeks Act. Forest History Today, Spring/Fal , 1720. Spurlock, D. (2019). Using plan and ordinance quality to evaluate the implementation of ri parian buffer policies. Landscape and Urban Planning, 183, 111. doi: https://doi.org/10.1016/j.landurbplan.2018.10.007 Srivastava, R., & Laurian, L. (2006). Natural hazard mitigation in loca l comprehensive plans: the case of flood, wildfire and drought planning in Arizona. Disaster Prevention and Management, 15, 461483. doi:10.1108/09653560610669936 Communities and wildfire policy 109 126 (Resources for the Future Press 2008). Steelman, T. A ., & Burke, C. A. (2007). Is wildfire policy in the United States sustainable? What is the wildfire problem? Journal of Forestry , 67 72. Stephens, S. L., & Collins, B. M. (2007). Fire policy in the urbanwildland interface in the United States: what are the issues and possible solutions? In A. R. Troy & R. G. Kennedy

PAGE 262

245 (Eds.), Living on the edge: economic, institutiona and management perspect ives on wildfire hazard in the urban interface (pp. 3342). Oxford, UK: Elsevier Ltd. Stephens, S. L., McIver, J. D., Boerner, R. E. J., Fettig, C. J., Fontaine, J. B., Hartsough, B. R., . . . Schwilk, D. W. (2012). The Effects of Forest Fuel Reduction Tre atments in the United States. BioScience, 62, 549560. doi:10.1525/bio.2012.62.6.6 Stephens, S. L., & Moghaddas, J. J. (2005). Experimental fuel treatment impacts on forest structure, potential fire behavior, and predicted tree mortality in a California mixed conifer forest. Forest Ecology and Management, 215(1), 2136. doi: https://doi.org/10.1016/j.foreco.2005.03.070 Stevens, M. R., Lyles, W., & Berke, P. R. (2014). Measuring and reporting interc oder realiability in plan quality evaluation research. Journal of Planning Education and Research, 34, 7793. doi:10.1177/0739456X13513614 Stevens, M. R., & Shoubridge, J. (2015). Municipal hazard mitigation planning: a comparison of plans in British Colum bia and the United States. Journal of Environmental Planning and Management, 58, 19882014. doi:10.1080/09640568.2014.973479 Stewart, S. I., Radeloff, V. C., Hammer, R. B., & Hawbaker, T. J. (2007). Defining the wildland– urban interface. Journal of Forestr y, 105, 201207. Stratton, R. D. (2004). Assessing the effectiveness of landscape fuel treatments on fire growth and behavior. Journal of Forestry, 102, 3240. Stratton, R. D. (2006). Guidance on spatial wildland fire analysis: models, tools, and techniques. Paper presented at the General Technical Report RMRS GTR183, Fort Collins, CO. Stratton, R. D. (2009). Guidebook on LANDFIRE fuels data acquisition, critique, modification, maintenance, and model calibration. Retrieved from Fort Collins, CO SWCA Environmental Consultants. (2014). Otero County Community Wildfire Protection Plan. Retrieved from Otero County, NM Syphard, A. D., Bar Massada, A., Butsic, V., & Keeley, J. E. (2013). Land Use Planning and Wildfire : Development Policies Influence Future Probability of Housing Loss. PLoS ONE, 8 , e71708. doi:10.1371/journal.pone.0071708 Tang, Z., & Brody, S. D. (2009). Linking Planning Theories with Factors Influencing Local Environmental Plan Quality. Environment and planning B: Planning and design, 36(3), 522537. doi:10.1068/b34076 Teie, W., & Weatherford, B. (2000). A report to the Council of Western State Foresters: fire in the West, the wildland/urban interface fire problem. In: Deer Valley Press, Rescue, California, USA.

PAGE 263

246 The Sacramento Bee. (2019). Build to survive: homes in California's burn zones must adopt fire safe code. Retrieved from https://www.govtech.com/em/preparedness/EDITORIAL Build to Survive Homes in Californias Burn Zones Must Adopt Fire Safe Code.html The White House. (2003). Healthy Forests an initiative for wildfire prevention and stronger communities. Policies in Focus. Retrieved from https://georgewbushwhitehouse.archives.gov/infocus/healthyforests/ Theobald, D. M., & Romme, W. H. (2007). Expansion of the US wildland–urban interface. Landscape and Urban Planning, 83, 340354. doi:10.1016/j.landurbplan.2007.06.002 Thompson, C. N. (2001). Urban Wildland Interface Communities Within the Vicinity of Federal Lands That Are at High Risk From Wildfire . (66 FR 751). Federal Reg istrar: United States Department of Agricutlure Forest Service Thompson, M. P., Ager, A. A., Calkin, D. E., Finney, M. A., & Vaillant, N. M. (2012). The science and opportunity of wildfire risk assessment: novel approaches and their applications in risk as sessment . Thompson, M. P., Haas, J. R., GilbertsonDay, J. W., Scott, J. H., Langowski, P., Bowne, E., . . . Grenfell, I. C. (2015). Development and application of a geospatial wildfire exposure and risk calculation tool. Environmental Modelling & Software , 63, 6172. doi:10.1016/j.envsoft.2014.09.018 Tian, G., Ouyang, Y., Quan, Q., & Wu, J. (2011). Simulating spatiotemporal dynamics of urbanization with multiagent systems A case study of the Phoenix metropolitan region, USA. Ecological Modelling, 222, 11291138. doi:10.1016/j.ecolmodel.2010.12.018 Tra vis, W. R. (2007). New geographies of the American West: land use and the changing patterns of place (2nd ed.). Washington, D.C.: Island Press. Trego, G. D. (2014). We didn't start the fire...and we won't pay to stop it: financing wildfire management in Am erica's wildland urban interface. William and Mary Environmental Law and Policy Review, 36, 595634. doi:10.1525/sp.2007.54.1.23. Trinity County Resource Conservation District, & The Watershed Research and Training Center. (2010). Trinity County Community Wildfire Protection Plan . Retrieved from Trinity County, CA Troy, A. R. (2001). Natural hazard policy and the land market: an assessment of the effects of the California Natural Hazard Disclosure Law. In Arbor Ciencia Pensamiento Y Cultura (pp. 184). Troy , A. R. (2001). Natural hazard policy and the land market: an assessment of the effects of the California Natural Hazard Disclosure Law. (Doctor of Philosophy), University of California, Berkeley

PAGE 264

247 TSS Consultants. Community Wildfire Protection Plan: Nevada County, California. Retrieved from Nevada County, CA United States Forest Service Rocky Mountain Research Station. (2016). Fire danger rating. Wild Fire Assessment System . Vaillant, N. M., & Reinhardt, E. D. (2017). An Evaluation of the Forest Service H azardous Fuels Treatment Program —Are We Treating Enough to Promote Resiliency or Reduce Hazard? Journal of Forestry, 115 (4), 300308. doi:10.5849/jof.16067 Van Wagner, C. E. (1977). Conditions for the start and spread of crown fire. Canadian Journal of Fo rest Research, 7 , 23 24. Van Wagner, C. E. (1993). Prediction of crown fire behavior in two stands of jack pine. Canadian Journal of Forest Research, 23, 442449. van Wagtendonk, J. W. (1996). Use of a deterministic fire growth model to test fuel treatme nts. Paper presented at the Sierra Nevada Ecosystem Project: Final report to Congress, vol. II, University of California, Davis. Vaux, H. J. (1982). Forestry’s hotseat: The urban/forest interface. Veblen, T. T., Kitzberger, T., & Donnegan, J. (2000). Clim atic and human influences on fire regimes in ponderosa pine forests in the Colorado front range. Ecological Applications, 10, 11781195. doi:10.1890/10510761(2000)010[1178:CAHIOF]2.0.CO;2 Vidyarthi, S., Winkle, C., Smith, J., Zhang, T., Kawamura, K., & Hoch, C. (2012). Holistic, inclusive and practical: teaching plan making at the core. The Town Planning Review, 83(6), 625645. Volker, R. C., Helmers, D. P., Kramer, H. A., Mockrin, M. H., Alexandre, P. M., Bar Massada, A., . . . Stewart, S. I. (2017). The 19902010 wildlandurban interface of the conterminous United States geospatial data [Vector Digital Data]. Wales, B. C., Suring, L. H., & Hemstrom, M. a. (2007). Modeling potential outcomes of fire and fuel management scenarios on the structure of for ested habitats in northeast Oregon, USA. Landscape and Urban Planning, 80, 223236. doi:10.1016/j.landurbplan.2006.10.006 Westerling Anthony, L. (2016). Increasing western US forest wildfire activity: sensitivity to changes in the timing of spring. Philosophical Transactions of the Royal Society B: Biological Sciences, 371 (1696), 20150178. doi:10.1098/rstb.2015.0178 Western W ashington University Huxley College. (2012). Mason County Community Wildfire Protection Plan. Retrieved from Mason County, WA Whitnall, G. (1931). History of zoning. The Annals of the American Academy of Political Social Science, 155 (2), 114.

PAGE 265

248 Williams , D. R., Jakes, P. J., Burns, S., Cheng, a. S., Nelson, K. C., Sturtevant, V., . . . Souter, S. G. (2012). Community wildfire protection planning: the importance of framing, scale, and building sustainable capacity. Journal of Forestry, 110, 415420. doi:10.5849/jof.12001 Winter, G., McCaffrey, S., & Vogt, C. A. (2009). The role of community policies in defensible space compliance. Forest Policy and Economics, 11, 570578. doi:10.1016/j.forpol.2009.07.004 Wolters, E. A., Steel, B. S., Weston, D., & Brunson, M. (2017). Determinants of residential firewise behaviors in central Oregon. The Social Science Journal, 54, 168178. doi: http://dx.doi.org/ 10.1016/j.soscij.2016.12.004 Yun, S., Chen, C., Li, J., & Tang, L. (2011). Wildfire spread simulation and visualization in virtual environments. ICSDM 2011 Proceedings 2011 IEEE International Conference on Spatial Data Mining and Geographical Knowledge Services, 315319. doi:10.1109/ICSDM.2011.5969054

PAGE 266

249 APPENDIX A CWPP PROCESS AND PLAN CONTENT EVALUATION INSTRUMENT This evaluation instrument is intended to provide an overall quality score of the CWPP process and plan. Note, the evaluation targets wha t is deemed most important, as defined by the literature, to complete a successful CWPP process, and plan. CWPP Title: CWPP Location: Coder Name: Date: List of Pertinent Documents: Yes (Yes = 1) No (No = 0) CWPP Process and Plan 1. Context 1.1. Does the CWPP plan or meeting minutes remind community members of how they handled past challenges, such as a wildfire or environmental disaster; to help the community understand how it is vulnerable and create a sense of urgency for developing a CWPP? 1.2. Does the CWPP plan or meeting minutes show that previous collaborative efforts in the community, whether wildfire planning or other projects were studied, to identify how they were successful and use lessons from those experiences to lay the groundwork fo r doing a CWPP? 1.3. Does the CWPP plan or meeting minutes identify people who were involved in earlier collaborative or wildfire planning efforts and bring their experience to developing a CWPP? 1.4. Does the CWPP plan or meeting minutes acknowledge the comm unity has little or no experience with collaboration or wildfire planning, and document how they overcame this inexperience? 1.5. Does the CWPP plan or meeting minutes identify previous disagreements within a community, related to wildlife or not, that threaten the CWPP process, and document how they were addressed early in the process to prevent them from becoming barriers? 2. Goals and Objectives 2.1. Does the CWPP plan clearly articulate and define the CWPPs goals and objectives. 2.2. Does the CWPP’s goals and objectives include a WUI definition goal and objective? 2.3. Does the CWPP’s goals and objectives include a structure ignitability goal and objective? 2.4. Does the CWPP’s goals and objectives include a forest thinning goal and objective? 2.5. Does the CWPP’s goals and objectives include a defensible space goal and objective? 2.6. Are the CWPP’s goals and objectives framed to contain multiple frames – e.g. forest health, saving lives, water protection?

PAGE 267

250 2.7. Does the CWPP’s goals and objectives also contain local values and interests? 2.8. Is the CWPP’s geographic scale compatible with its goals and objectives – e.g. a smaller scale CWPP, at the neighborhood or community level, if the stated goals are to motiva te homeowners to reduce hazards on their properties while a larger scale plan, such as at the county level, should be used if the goal is to reduce wildfire risk across the landscape? 3. Community Capacity 3.1. Does the CWPP plan or meeting minutes document resources that can help CWPP participants work together? 3.2. Does the CWPP plan or meeting minutes document capacity issues – e.g. funding, economic, political agendas – and suggest ways to overcome them? 4. Partnerships and Collaboration 4.1. Does the CWPP plan identify the core team? 4.2. Does the CWPP plan identify other local, state, and federal partners (may also be part of the core team)? 4.3. Does the CWPP plan or meeting minutes articulate the process by which additional stakeholders – e.g. forest management groups, city council members, resource advisory committees, HOAs, Division of Wildfire/Fish and Game, Department of Transportation, local and state emergency management agencies, water districts, utilities, recreation organizations, environmental organizations, forest products interests, local chambers of commerce, and watershed councils – and the community at large was solicited for inclusion in the process and use multiple methods to do so – e.g. radio ads, bus ads, emails? 4.4. Does the CWPP plan articulate the role each of the aforementioned groups play, including their responsibilities? 4.5. Does the CWPP plan identify the process for setting priorities? 4.6. Does the CWPP plan identify dates and timelines for implementation? 4.7. Does the CWPP plan identify how often the CWPP will be evaluated and fall within a five to ten year time frame? 4.8. Does the CWPP plan identify social vulnerabilities and how to overcome them – e.g. assistance to low income and underserved residence, social service organizational support, and diversity of populations served? 4.9. Does the CWPP plan and meeting minutes reflect continued and sustained engagement throughout the CWPP process? 4.10. Does the CWPP plan or meeting minutes identify leaders, from within the community or drawn from outside it, who can mobilize others and serve as catalysts for action, and recruit them for your CWPP process? 4.11. Does the CWPP plan or meeting minutes articulate the involvement of people in the CWPP process who have access to multiple social networks and can serve as intermediaries between the networks? 5. Base Map 5.1. Does the CWPP plan contain a base map that identifies the following: inhabited areas; critical community infrastructure – e.g. hospitals, nursing homes, fire stations, emergency shelters, water availability and

PAGE 268

251 supplies, and evacuation routes; and a prelim inary WUI designation zone? 6. Risk Assessment 6.1. Does the CWPP plan articulate how the WUI boundary is defined? 6.2. Does the CWPP plan document how the community has changed over time, this should include all of the following: population; age; percentage of youth; percentage of elderly; number of housing units; percent of owner and renter occupied housing units; percent age of people in the labor force; percentage of families below the federal poverty line; unemployment rate; length of homeowner tenure; full time/part time residency status; and income? 6.3. Does the CWPP plan document and describe the differing jurisdiction al boundaries – e.g. USFS, Parks, and other private and public land owners and managers? 6.4. Does the CWPP plan identify where the priority fuels projects are located in each community? 6.5. Does the CWPP plan identify data sources used in the risk assessment process, including new or updated data that may change the risk assessment and influence fuel priorities – e.g. hazards, risks, protection and response capabilities, structural vulnerabil ities, community values and resources, low income and vulnerable populations? 6.6. Does the CWPP plan outline the key factors, definitions and process used to assign risk that minimally includes three categories of risk, which were developed on the followin g minimum criteria: slope, aspect, and fuels? 6.7. Does the CWPP plan outline evaluation criteria and processes to measure changing risk overtime? 6.8. Does the CWPP plan identify the percent of vulnerability populations at risk and how they are going to be engaged in risk reduction? 6.9. Does the CWPP plan acknowledge and provide a course of action for monitoring and addressing long term climate change dr iven changes in wildfire risk? 7. Hazardous Fuels Reduction 7.1. Does the CWPP plan identify hazardous fuels reduction on public and private land? This should include prioritization and approximate acres needing treatment and the strategies to do so – e.g. defensible space, forest thinning techniques that include: presc ribed fire, mechanical thinning, mastication, plowing with a bulldozer, and applying herbicide. This information should also include the number and percentage of homes needing defensible space treatments and number and percentage of homes needing treatment in vulnerable communities. 8. Reducing Structural Ignitability 8.1. Does the CWPP plan describe past resource losses – e.g. number of human caused fires, number of lightning caused fires, and number homes lost to fires, suppression costs, and economic losses? 8.2. Does the CWPP plan describe past resources saved – e.g. number of homes and economic savings? 8.3. Does the CWPP plan outline wildfire codes and regulations – e.g. building materials, roof types, windows – and where they are applicable, including recommend updates to local governance?

PAGE 269

252 8.4. Does the CWPP plan document the past trends of WUI expansion and outline a process for continuing to monitor and report WUI expansion? 8.5. Does the CWPP plan document and outline local governance measures for best control WUI growth and directing it away from moderate to high risk areas? 9. Education and Outreach 9.1. Does the CWPP plan discuss how the public was and will be engaged, this should include multiple avenues – e.g. public meetings field trips, demonstration projects, household visits, youth engagement, and community events? 9.2. Does the CWPP plan discuss the role of fire insurance and evacuation plans, including templates/examples of evacuation plans? 9.3. Does the CWPP plan document changing attitudes and awareness of wildfire – e.g. change in humancaused wildfires, increased participation in local fuel reduction programs and woody debris disposal? 9.4. Does the CWPP plan and meeting minutes reflect attempts to provide translations of all relevant materials to non English speakers living in the WUI? 10. Emergency Management Capacity 10.1. Does the CWPP plan document the number and percentage of homes in each fire district? 10.2. Does the CWPP plan document the capacity of emergency management – e.g. number and percent of trained and/or certified fire fighters and crews; fire suppression equipment; and response times? 10.3. Does the CWPP plan address incident command training? 10.4. Doe s the CWPP plan address animal and livestock preparedness and evacuation plans? 10.5. Does the CWPP plan include or link to a community evacuation plan and scheduled times to test it, this should include: local neighborhood evacuation plans; safety zones; residential and vulnerable population evacuation plans; community and fire personal communication systems; and resource lists? 10.6. Does the CWPP plan or meeting minutes reflect the coordination of CWPPs with other hazard mitigation plans – e.g. meet FEMA requirements for natural hazard mitigation plans? 11. Long term Success 11.1. Does the CWPP plan outline the incorporation of projects into the CWPP that can be accomplished quickly to foster homeowner buy in and broaden support for the longer term effort? 11.2. Does the CWPP plan discuss other related plans or link to other types of plans and process to augment resources, broaden support, and enhance implementation? 11.3. Does the CWPP plan discuss how the CWPP will be implemented into formal governance structure? 11.4. Does the CWPP plan document the existence of or encourage the participation in the Firewise community program or have a Fire Safe Council?

PAGE 270

253 APPENDIX B LOCAL GOVERNANCE INSTRUMENT This evaluation instrument is intended to provide an overall quality score of the implementation of CWPPs. Note, the evaluation targets what is deemed most important, as defined by the literature, to complete a successful implementation CWPPs. CWPP Title: CWPP Location: Coder Name: Date: List of Pertinent Documents: Yes (Yes = 1) No (No = 0) CWPP Implementation: Local Governance 1. Comprehensive Plan 1.1 Does the comprehensive plan acknowledge and document the integration of the CWPP, WUI, and wildfire risk into the planning process? 1.2 Is there clear evidence that the comprehensive plan responds to wildfire risk – e.g. directing or limiting development and critical infrastructure away from moderate to high risk areas; and protection of resources? 1.3 Does the comprehensive plan document and strategize ways to use open space preservation, conservation, watershed management, or climate change planning to buffer existing and future development from wildfire risk or how to leverage these processes to further reduce wildfire risk? 2. Zoning Codes, Development Standards, Subdivision Design Guidelines, and HOA Ordinances 2.1 Are there WUI and wildfire risk reduction specific zoning codes – e.g. WUI or risk zone overlay districts or WUI codes? 2.2 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances include or link to forest thinning and defensible space location priorities and requirements? 2.3 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances set or link to water supply standards defensible space requirements, resource protection requirements, and responsibilities of ongoing maintenance? 2.4 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances outline penalties for lack of compliance? 2.5 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances restrict development within high wildfire risk areas? 2.6 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances contain documented incentives for developers to plan and implement open space and trails to act as fuel breaks? 2.7 Does the zoning code, development standards, subdivision design guidelines or HOA ordinances set adequate minimum road widths?

PAGE 271

254 3. Building Codes 3.1 Are there local building codes that address fire resistant building codes that cover the following topics: fire rated roofing, siding, and decking; closed eaves and soffits; and how to protect windows and vents? 3.2 Does the local building codes meet or exceed the current International WUI Building Code requirements for ignition resistant construction? 4. Plan Review and Inspection Procedures 4.1 Do the plan review or inspection procedures include water tests and inspections to ensure adequate water supply requirements? 4.2 Are there inspection procedures to monitor mitigation implementation and maintenance and proper wildfire building code implementation for new construction? 4.3 Do the plan review or inspection procedures require developers or occupants to have fire insurance? 4.4 Does the plan review process or inspection procedures require pre development mitigation work? 4.5 Does the plan review or inspection procedures ensure multiple ingress and egress options for developments?

PAGE 272

255 APPENDIX C PILOT STUDY WRITEU P Appendix C summarizes the initial pilot study results. Please note the tables and figures that appear within this appendix are not listed in the front matter. Pilot Study Results This subsection summarizes the results of the pilot s tudy in three additional subsections. The first sub section presents WUI and independent variable descriptive statistics and the document coding results for “CWPP Process and Plan Evaluation Instrument” and “CWPP Implementation: Local Governance Evaluation Instrument." The second subsection, entitled “Regression Analyses,” documents the results of the correlation and ordinal regressions. The third sub section, entitled “Discussion,” presents and discusses qualitative document coding findings and issues an d solutions regarding errors, uncertainties, and sensitivities within the pilot data and analyses. Descriptive Statistics and Document Coding Results Descriptive Statistics The WUI in Boulder County, Colorado grew by 28.92 kilometers2 (11.17 miles2) betwe en 2000 and 2010 (Radeloff et al., 2017) . Overall the WUI in Boulder County is increasing; however, two CWPP locations experienced no growth, and three CWPP locations experienced negative growth. Table Appendix C .1 presents a summary of individual WUI changes. However, as expected the WUI as defined by Radeloff et al. (2017) is not entirely covered by a CWPP. This difference was not unexpected as each CWPP has a unique WUI definition, as allowed and encouraged by HFRA and CWPP best practices. Boulder County has 12 local CWPPs that cover 278.53 kilometers2 (107.54 miles2) of WUI. Despite this difference, Radeloff et al.’s (2017) WUI

PAGE 273

256 dataset does provide a uniform measure that allows us to compare WUI changes over geographic space and time for each CWPP. Figure Appendix C .1 shows the geographic locations of each CWPP within Boulder County. Table Appendix C. 1. Boulder County and CWPP WUI change from 19902000 in square kilometers (square miles in parentheses). ID 2000 WUI 2010 WUI Delta Change Boulder_001 21.18 (8.18) 22.37 (8.64) 1.19 (.46) Boulder_002 5.71 (2.2) 5.70 (2.2) .01 ( .004) Boulder_003 8.89 (3.43) 8.95 (3.46) .06 (.02) Boulder_004 3.39 (1.31) 3.50 (1.35) .11 (.04) Boulder_005 7.56 (2.92) 7.39 (2.85) .17 ( .04) Boulder_006 10.68 (4.12) 10.68 (4.12) .0 (0) Boulder_007 7.47 (2.88) 7.41 (2.86) .06 ( .07) Boulder_008 5.19 (2) 5.50 (2.12) .31 (.12) Boulder_009 8.27 (3.19) 9.33 (3.6) 1.06 (.41) Boulder_010 18.74 (7.24) 22.77(8.79) 4.03 (1.56) Boulder_011 5.86 (2.26) 5.86 (2.26) .0 (0) Boulder_012 27.45 (10.6) 28.16 (10.87) .71 (.27) Boulder County 338.13 (130.55) 367.05 (141.72) 28.92 (11.17) Note. WUI calculations are based on Radeloff et al. ’s (2017) WUI classification and Boulder County’s CWPP GIS boundary file.

PAGE 274

257 Figure Appendix C. 1. Geographic location and extent of Boulder County, Colorado local CWPPs. Eleven of the twelve CWPPs were prepared for and organized around local fire districts, while one CWPP was developed and arranged around a municipal government, the City of Boulde r. A single consultant completed over 64 percent of the CWPPs within Boulder County. Many of the CWPPs were created in 2006, 2007, or 2008 while two CWPPs were created in 2011, Boulder_007 and Boulder_009. Boulder_005 updated their CWPP in 2010. Only four of

PAGE 275

258 the CWPPs documents were written collaboratively, engaging multiple, public agency authors. Table Appendix C .2 provides a complete list of Boulder County’s local CWPPs with unique identifiers, titles, clients, and authors. Table Appendix C. 2. CWPP identifiers and relevant CWPP documents. CWPP Unique Identifier CWPP Title CWPP Location/Prepared For Author Boulder_001 Boulder Mountain Fire Protection District Wildland Urban Interface Community Wildfire Protection Plan Boulder Mountain Fire Protection District Anchor Point Boulder_002 Four Mile Fire Protection District Wildland Urban Interface Community Wildfire Protection Plan Four Mile Fire Protection District, Anchor Point Boulder_003 Boulder Rural Fire Protection District Wildland Urban Interface Community Wildfire Protection Plan Boulder Rural Fire Protection District Anchor Point Boulder_004 City of Boulder Wildland Urban Interface Community Wildfire Protection Plan City of Boulder Anchor Point Boulder_005 Rocky Mountain Fire Wildland Urban Interface Community Wildfire Protection Plan Rocky Mountain Fire Anchor Point Boulder_006 Sugar Loaf Fire District Bou lder, Colorado Wildland Urban Interface: Community Wildfire Protection Plan Sugar Loaf Fire Protection District, Anchor Point Boulder_007 Nederland Fire Protection District Community Wildfire Protection Plan Nederland Fire Protection District, Anchor Point Boulder_008 Community wildfire protection plan Gold Hill Fire Protection District Finn et al. Boulder_009 Lefthand Fire Protection District Community Wildfire Protection Plan Lefthand Fire Protection Dist rict Greenwood Sustainability LLC Boulder_010 Lyons Fire Protection District Community Wildfire Protection Plan Lyons Fire Protection District Owen et al. Boulder_011 Community wildfire protection plan Sunshine Fire Protection District Stratton et al. Boulder_012 Community Wildfire Protection Plan Allenspark, Colorado Allenspark Fire District Walter et al.

PAGE 276

259 Implementation documents included 13 documents of various types. F our countywide documents cover every CWPP location, including a comprehensive plan, landuse code, building codes, and a hazard mitigation plan. Four CWPPs had additional local m unicipal policy and code implementation documents, including comprehensive community plans, building regulations and zoning and subdivision guidelines . Table Appendix C .3 outlines the specific documents coded for each CWPP ID. Table Appendix C. 3. Applicable implementation documents per CWPP unique ID. CWPP ID Applicable implementation documents Boulder_001, 002, 003, 004, 005, 006, 007, 008, 009, 010, 011, and 012 Boulder County Comprehensive Plan Boulder County Land Use Code Boulder County Building Code Amendments Hazard Mitigation Plan: Boulder County Boulder_004 Boulder Valley Comprehensive Plan Title 9 Land Use Code and Title 10 Structures Boulder_007 Envision Nederland 2020 Chapter 16 Zoning and Chapter 18 Building Regulations Boulder_005 Town of Superior Comprehensive Plan Chapter 16 Land Use Chapter 18 Building Regulations Boulder_010 The Town of Lyons Comprehensive Plan Zoning, Subdivision, and Building Regulations Building Guides Table Appendix C .4 summarizes the mean and delta change of the independent variables for each CWPP. The general trend of ages within each CWPP is middle age, with an average of 41 in 2000 and 46 in 2010 and trending older with an average change of five years. The length of homeowner tenure in 2000 was seven years and almost ten years in 2010, with an average change of three years. In 2000, 11 percent of dwe lling units were classified as secondary or vacation homes, while in 2010, 15 percent of home s classified are secondary homes. The general trend increased by four percent, with the largest growth in Boulder_004, Boulder_003, and Boulder_011, and Boulder_005. However, three CWPPs exhibited decreases in secondary home

PAGE 277

260 classifications. The average income in each census decade exhibited a wide range, $48,000 to $91,000 in 2000 and $69,000 to $112,000 in 2010. The average in 2000 was $70,400, while the average i n 2010 was $91,000, with a median change of approximately $21,000. However, not all CWPPs inc reased; equally, the range of increase was particularly diverse from as little as $7,000 to as much as $45,000. Please note, that each CWPP did not align with cens us block groups, which is discussed in the “Discussion” sub section below. Table Appendix C. 4. Boulder CWPP mean and delta descriptive statistics for the independent variables. ID Age Length of Homeowner Tenure Residency Status Income 2000 2010 2000 2010 2000 2010 2000 2010 M M

PAGE 278

261 A summary of “Process and Plan Evaluation Instrument” categorical scores and total scores are presented in Table Appendix C .5. These scores exhibited a compressed tot al score range of .23 to .43, with an average of .345. Many of the categories exhibited similar scores and compressed ranges of scores, with few outliers, except for category 2.0. A summary of “CWPP Implementation: Local Governance Evaluation Instrument” c ategorical scores and total scores are presented in Table Appendix C .6. The full “CWPP Implementation: Local Governance Evaluation Instrument” document analysis score results are presented in Appendix B . The implementation totals scores exhibit a wider ran ge than the previous scores, ranging from .06 to .65. However, this wider range was mainly exhibited in the implementation of category two and several outliers in each category. The average score was .42. Total composite scores, combining “Process and Plan Evaluation Instrument” and “CWPP Implementation: Local Governance Evaluation Instrument” are presented in Table Appendix C.7. Composite scores ranged from .21 to .46 with an average of .38 Table Appendix C. 5. Summary of categorical scores for Boulder County CWPP “Process and Plan Evaluation Instrument” document analysis score results (results are presented in percentage of total score). ID 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 Total Boulder_001 .4 .5 0 .27 0 .33 1 .4 .75 .33 .5 .39 Boulder_002 0 .5 .5 .45 0 .56 1 .2 .25 .33 .5 .38 Boulder_003 0 .5 0 .27 0 .44 1 .2 0 .17 .25 .27 Boulder_004 0 .5 0 .27 0 .44 1 .2 0 .17 .25 .27 Boulder_005 .2 .5 0 .45 0 .56 1 .2 .25 .17 .25 .36 Boulder_006 .2 .5 0 .45 0 .56 1 .2 .25 .17 .25 .34 Boulder_007 .2 .75 0 .45 0 .56 1 .2 .25 .33 .25 .38 Boulder_008 0 1 .5 .55 1 .56 1 .2 .25 0 0 .41 Boulder_009 0 .75 0 .45 1 .67 1 .4 .25 .33 .25 .43 Boulder_010 0 .38 0 .36 0 .44 1 .4 .25 .33 0 .30 Boulder_011 0 0 0 .36 0 .44 1 .4 .25 0 0 .23 Boulder_012 .8 .5 0 .64 0 .22 1 .2 0 .17 .25 .38

PAGE 279

262 Table Appendix C. 6. Boulder County CWPP “CWPP Implementation: Local Governance Evaluation Instrument” document analysis score results (results are presented in percentage of total score). ID 1.0 2.0 3.0 4.0 Total Boulder_001 0 .43 1 .6 .47 Boulder_002 0 .43 1 .6 .47 Boulder_003 0 .43 1 .6 .47 Boulder_004 1 .57 1 .4 .65 Boulder_005 0 .14 0 0 .06 Boulder_006 0 .43 1 .6 .47 Boulder_007 .33 .29 1 .4 .41 Boulder_008 0 .43 1 .6 .47 Boulder_009 0 .43 1 .6 .47 Boulder_010 0 .14 0 .4 .18 Boulder_011 0 .43 1 .6 .47 Boulder_012 0 .43 1 .6 .47 Table Appendix C. 7. Composite document analysis scores (“Process and Plan Evaluation Instrument” total score + “CWPP Implementation: Local Governance Evaluation Instrument” total score). ID Composite Total Score Boulder_001 .43 Boulder_002 .42 Boulder_003 .37 Boulder_004 .46 Boulder_005 .21 Boulder_006 .40 Boulder_007 .39 Boulder_008 .44 Boulder_009 .45 Boulder_010 .24 Boulder_011 .35 Boulder_012 .42 Cohen’s Kappa was computed to check the interrater reliability of each rater’s question scoring. The researchers targeted a moderate to perfect level of agreement, see Table Appendix C .8 for asso ciated Cohen’s Kappa values. It took two rounds of coding to achieve acceptable Kappa values for all coding questions. Table Appendix C. 8. Cohen’s Kappa levels of reliability interpretation (Mchugh, 2012) .

PAGE 280

263 Value of Kappa Level of Agreement % of Data that are Reliable 0 .20 None 0 – 3% .21 .39 Minimal 4 14% .40 .59 Weak 15 34% .60 .79 Moderate 35 63% .80 .90 Strong 64 81% Above .90 Almost Perfect 82 100% The resulting kappa for each question is reported in Table Appendix C .9. After the first round of scoring, eight questions received an unacceptably low kappa score, specifically questions: 1.3, 2.5, 2.6, 4.3, 4.9, 6.1, 8.2, and 10.1. The two researchers me t and reviewed the causes for each low score. Several causes were identified, which are presented in Table Appendix C .10. After fixing each coding error, each researcher recoded the eight low questions and the second round of Cohen’s Kappa was completed . T he second round results provided acceptable Kappa ranges for the eight low scoring questions (Table Appendix C .9). The final results are: 42 questions (75%) have almost perfect interrater reliability, two questions (4%) have strong interrater reliability, and 12 questions (21%) have moderate reliability. Table Appendix C. 9. CWPP Process and Plan E valuation Instrument Cohen’s Kappa.

PAGE 281

264 Question Round 1 Kappa Round 2 Kappa 1.1 1.00 1.2 1.00 1.3 0.44 1.00 1.4 0.63 1.5 1.00 2.1 1.00 2.2 1.00 2.3 0.63 2.4 0.63 2.5 0.43 0.75 2.6 0.43 0.75 2.7 1.00 2.8 1.00 3.1 1.00 3.2 1.00 4.1 1.00 4.2 1.00 4.3 0.57 0.75 4.4 0.80 4.5 1.00 4.6 1.00 4.7 1.00 4.8 1.00 4.9 0.50 1.00 4.10 0.63 4.11 1.00 5.1 1.00 6.1 0.58 1.00 6.2 1.00 6.3 1.00 6.4 1.00 6.5 0.83 6.6 1.00 6.7 1.00 6.8 1.00 6.9 1.00 7.1 1.00 8.1 1.00 8.2 0.13 0.75 8.3 1.00 8.4 1.00 8.5 1.00 9.1 1.00 9.2 0.63 9.3 1.00 9.4 1.00 10.1 0.47 1.00 10.2 0.75 10.3 0.63 10.4 1.00 10.5 1.00 10.6 1.00

PAGE 282

265 11.1 1.00 11.2 1.00 11.3 1.00 11.4 0.75 Table Appendix C. 10. Intercoder reliability discussion results. Question Issue 1.3 Coder one inadvertently coded Boulder_006 and Boulder_007 as a “1” when they meant “0.” 2.5 Coder one inadvertently coded Boulder_007 as a “0” when they meant a “1.” 2.6 Coder one initially coded Boulder_009 as a “1”; however, after discussion, the goals and objectives were so vague they do not contain multiple frames of reference. 4.3 Coder two initially coded Boulder_005 as a “0”; however, after review, the information was contained in a separate appendix . Thus it was recoded to a .” 4.9 Coder two initially coded Boulder_007 as a “0,” but upon re review Appendix, E showed sustained involvement. 6.1 Coder two initially coded Boulder_004 as a “1” when they meant a “0.” 8.2 Coder two initially coded Boulder_001 as a “0”, upon further discussion of the question, it was recoded a “1.” Coder one initially miscoded Boulder_005 a “0” when they meant a “1.” 10.1 Coder one initially miscoded Boulder_002 as a “1” when they meant “0. ” Coder one initially miscoded Boulder_011 and Boulder_012 as a “0” when they meant “1.” A single coder, coder one, coded all relevant implementation documents for the pilot study using the “ CWPP Implementation: Local Governance Evaluation Instrument ” ( Appendix B ). Since coder one will also code all the research study implementation documents as well, there is no need for an interrater reliability measure. The complete list of documents are listed in Table Appendix C .3. Score results are presented above. If multiple researchers assist in coding future implementation documents , an intercoder reliability test will be performed. Regression Analyses A Pearson product moment correlation coefficient was computed to assess the relationship betwe en each dependent variable and the independent variables. There were no statistically significant results, likely due to a pilot n = 12. Table Appendix C .11 summarizes the results. Table Appendix C. 11. Variable Pearson Correlations for Median Delta Change of Independent Variables.

PAGE 283

266 Dependent Variables Independent Variables Document Analysis Score Implementation Analysis Score Composite Score Age .436 .055 .221 Homeowner Tenure .445 .534 .338 Residency Status .431 .065 .104 Income .213 .173 .248 Note. Put a note here about it being an n = 12 and a correlation calculated to the .01 level (2tailed). T o calculate ordinal regressions, the document analysis, implementation analysis, and composite scores were transformed into ordinal categories using a standard grading scale (Table Appendix C.12) . A review of the literature did not provide a standard scale, so a standard academic grading scale was used because 1) a majority of people are familiar with the conversion of percentage scores t o letter grades, 2) it provides the necessary dependent variable type for ordinal regressions, and 3) i t is often used in educational research statistics to evaluate effective learning . The initial ordinal regression resulted in an error because all of the dependent variables received an F letter grade, so additional grade categories were added , low F, middle F and high F (Table Appendix C.12). Table Appendix C. 12. Scoring to letter grade conversion chart. Percentage Letter Grade .9 – 1 A .8 – .89 B .7 – .79 C .6 – .69 D .39 – .59 F High .19 – .38 F Middle 0 – .18 F Low The model case processing summaries are presented in Table Appendix C .13. None of the model fitting information reported a statistically significant score (<. 05) (Table Appendix C .14). All models reported an acceptable goodnessof fit or failed to reject the null hypothesis

PAGE 284

267 for both Pearson and Deviance tests; however, the composite score only marginally passed ( Table Appendix C .15). The Nagelkerke pseudo r squared test explains 100 percent of the document analysis score, 46.5 percent of the document implem entation analysis score, and 14.2 percent of the composite score (Table Appendix C .16). However , the test for parallel lines, a test for proportional odds, suggests exercising caution in interpreting the outcomes of the regression analysis because all of the scores are lower than .05 (Table Appendix C .17). No independent variable parameter estimate is statistically significant (Table Appendix C .18). Table Appendix C. 13. Case P rocessing Summary . Dependent Variable Grade N Marginal Percentage Document Analysis Score F – Middle 9 75% F – High 3 25% Document Implementation Score F Low 2 16.7% F Middle 9 75% F High 1 8.3% Composite Score F – Middle 3 25% F – High 9 75% Table Appendix C. 14. Model Fitting Information. Model Significance Document Analysis Score .009 Document Implementation Score .261 Composite Score .876 Table Appendix C. 15. Goodness of fit Model Test Significance Document Analysis Score Pearson 1 Deviance 1 Document Implementation Score Pearson .865 Deviance .845 Composite Score Pearson .075 Deviance .092

PAGE 285

268 Table Appendix C. 16. Pseudo R square. Model Nagelkerke Document Analysis Score 1 Document Implementation Score .465 Composite Score .142 Table Appendix C. 17. Test of Parallel Lines. Model Significance Document Analysis Score 0 Document Implementation Score 0 Composite Score 0 Table Appendix C. 18. Parameter estimates. Model Variable Significance Estimate Document Analysis Score Age .999 7.877 Homeowner Tenure .998 26.652 Residency Status 0 483.777 Income .997 .006 Document Implementation Score Age .501 .515 Homeowner Tenure .194 2.279 Residency Status .637 6.590 Income .254 0 Composite Score Age .413 .615 Homeowner Tenure .697 .246 Residency Status .772 3.476 Income .500 .0000066 Discussion Coding and S coring Discussion Overall, Boulder County CWPPs scored lower than anticipated on the document , implementation, and composite scores. However, it is worth noting that many of these documents were prepared before the publication of CWPP best practices. Additionally, as previously mentioned, over 64 percent of the CWPPs within Boulder County w as completed by a single consultant. This consultant used similar processes and documentation across all their CWPPs.

PAGE 286

269 CWPP Process and Plan Context CWPP best practices suggest that the CWPP document acknowledges and articulates past challenges, vulnerabilities, experience, and disagreements that are potential barriers to a positive process or implementation. Additionally, positive past experiences and processes can be used as leverage points during the CWPP process. However, few to none of the Boulder CWPPs reported on past collaborative efforts. Indeed, none of the CWPPs documented the core team’s or community’s collaborative process or wildfire experience level. Finally, none of the CWPPs documented potential barriers to collaboration or wildfire risk reduction efforts, leaving the process open to failure if underlying conflicts are not addressed. Goals and Objectives While all 12 CWPPs have goals and objectives, they are often vague and do not provide a clarity of purpose or explicitness of procedural actions that are understandable to both stakeholders and public officials. Indeed, the CWPPs goals and objectives are too vague to evaluate and measure adequately and of ten focus on education and not federally mandated requirements such as prioritized fuel reduction locations, priorities, and methods. For example, Boulder Mountain Fire Protection District’s CWPP states: Promote community awareness: Quantification of the community's hazards and risk from wildfire will facilitate public awareness and assist in creating public action to mitigate the defined hazards) (Anchor Point, 2006 p. 2) . While a noble goal and objective, numerous studies have shown that behavior change s are not solely encouraged by passive, information sharing, they require active engagement and additional incentives or punishments .

PAGE 287

270 Community Capacity CWPPs focused on economic capacity, with emphasis on federal and state grants. However, the CWPPs did not document local or nonprofit economic capacity. Additionally, these CWPPs did not articulate political capacity or issues of community and political will. Likewise, they did not document support groups, organizations, or entities that could help facilitate a CWPP process or assist in implementation ( e.g., volunteer groups, consultants). Partnership and Collaboration Most CWPPs documented the core team; however, they did not articulate the ir roles . Wh ile most core teams consisted of local, state and federal government officials, fire district representatives, and report and process consultants, the often also engaged the broader public through public open house meetings. However, several groups were noticeably missing: real estate agents, developers, and land development consultants (architects, planners, landscape architects, and engineers ). This is particularly concerning because of the well document WUI expansion into high wildfire risk areas, lack of forethought in designed risk mitigation strategies, and lack of integration in local governance. Compounding these issues are the lack of scheduling, or prioritization of implementation; thus, causing the potential for lack of accountability. Indeed, none of the CWPPs included a means of plan evaluation or updates. Likewise, the CWPPs do not document social vulnerabilities that can cause underlying social and economic issues that impede risk reduction efforts. Indeed, many of the reports simply cited the same county level population and economic growth trends. Base Map In general base, maps were often titled and illegible due to a lack of labels , poor resolution, and refined cartographic best practices. Many of the base maps were maps of wildfire

PAGE 288

271 risk, for the entire fire district; however, they never articulated if this was all considered WUI or not a critical requirement of CWPPs. Instead , they often highlighted project focus areas but did not define how they decided on these locations. Additionally, they may or may not have mapped project focus areas. Additionally, many base maps did not provide documentation for critical community infrast ructure (e.g., hospitals, nursing homes, fire stations, emergency shelters, water availability and supplies, and evacuation route s) . Figure Appendix C .2 is a typical base map. Figure Appendix C. 2. Typical base m ap.

PAGE 289

272 Risk Assessment Each CWPP measured and mapped risk; however, not all the CWPPs reported their methodology, making CWPPs difficult to evaluate and update. Anchor Point authored CWPPs used a combination of Fire Regime Condition Class and fire behavior potential modeling to assess risk. The remaining documented CWPPs used LANDFIRE derived datasets. However, none of the CWPPs expand their risk assessment to evaluate social or economic vulnerabilities. Nor do they evaluate or integrate response capacity a nd times or accessibility to the fire source or water. Likewise, the CWPPs do not document changing trends or risk or climate change impacts. It is also interesting to note, that none of the CWPPs acknowledge the role increased development plays in increas ing risk. Hazardous Fuels Reduction Hazardous fuels reduction is mentioned in each CWPP, with areas of priority. However, few provide dates for implementation or document the current state of defensible space treatments. Nor do the CWPPs document the fuel loads or thinning necessary on large swaths of surrounding forest land. Indeed, the strategies for hazardous fuels reduction only provides limited direction of treatment options, e.g., thinning and defensible space, but do not describe all the possibilities and applications of thinning, e.g. , prescribed fire, mechanical thinning , mastication, plowing with a bulldozer, and applying herbicide. The documents often reference other documents that are dated and do not contain current best practices in defensible space or thinning standards, such as variable thinning and defensible spac e distances dependent on topography. Indeed, many of the defensible space guidelines recommend two zones and not the current practice of three zones.

PAGE 290

273 Reducing Structural Ignitibility CWPP plans documented the number of homes and other resources saved and destroyed in past fires; however, the CWPPs do not document the economic loss or savings. The CWPPs do mention and address structural ignitibility, but often in vague terms or acceptance of noncompliant existing structures. For example, Boulder_002 states, The most important element for the improvement of life safety and property preservation is for every home in the study area to have compliant, effective defensible space. This is es pecially important for homes with wood roofs and homes located on steep slopes, in chimneys, saddles, or near any other topographic feature that contributes to fire intensity. An aggressive program of evaluating and implementing defensible space for homes will do more to limit fire related property damage than any other single recommendation in this report ( Anchor Point, 2006, p. 34) . However, to date, these strategies have not been implemented, nor have current building codes been documented or referred to for new construction or for updating existing noncompliant structures. More importantly , development and landuse codes and planni ng have ignored the locations of structures as development continues to expand into the WUI. The CWPPs that do mention current building codes often contain broken links or outdated material and do not address existing noncompliant dwellings. Indeed, many dwellings are allowed to be replaced in kind. Many of the CWPPs document the population growth and building growth within the broader context of Boulder County but do not address the local growth of the WUI or the geographic change of the WUI footprint. Nor do they document landuse best practices for directing development away from highrisk areas. More importantly, CWPPs recommendations for fuel reduction and reducing structural ignitibility projects are often limited to a few municipal projects, with lit tle to no discussion regarding assessing private property or the larger contiguous wildland fuel reduction needs.

PAGE 291

274 Education and Outreach The pilot CWPPs contain a great deal of information regarding how the public was engaged and how they will continue engagement, outreach, and education. These efforts often cover a wide spectrum of activities to private property consultation efforts to community events and demonstrations. One missing engagement opportunity is youth engagement efforts. Only a few CWPPs ment ion and articulate the need and benefits of fire insurance. However, CWPPs could be improved by providing examples and templates for household evacuation plans. None of the CWPPs document the communities changing attitudes and awareness of wildfire risk or document efforts and materials to reach out to nonEnglish speakers living within the WUI. Emergency Management Capacity Roughly 50% of the CWPPs document the number of homes within the WUI or fire district. Much of the emergency management documentation is focused on fire suppression response capacity. However, the se CWPPs do not mention ongoing incident command training efforts, animal and livestock preparedness and evacuations plans, or community evacuation plans. These plans should include local neigh borhood evacuation plans; safety zones; residential and vulnerable population evacuation plans; community and fire personal communication systems; and resource lists, in addition to scheduled tests . None of the CWPPs mentioned how hazard mitigation plans a re coordinated with CWPPs . Longterm Success CWPPs focus longterm success efforts by encouraging Firewise community participation and the creation of a Fire Safe Council. However, not many of the communities have joined Firewise or created Fire Safe Coun cils. Additionally, the CWPPs have no documentation of longterm success strategies, e.g. , no quick short term success projects,

PAGE 292

275 broader resources and planning initiatives to facilitate implementation, and no formal governance implementation. Without these items the long term success of these CWPPs is questionable. CWPP Implementation: Local Governance Comprehensive Plan Only one CWPP, Boulder_004, integrates the CWPP into their comprehensive plan. Indeed, there is clear evidence this comprehensive plan an d CWPP attempt to direct new development and critical infrastructure away from moderate to highrisk areas; however, they do not address changing risk or risk to existing development. The remaining CWPPs do not leverage the possible ways broader comprehens ive planning efforts regarding open space preservation, conservation, watershed management, or climate change planning to reduce wildfire risk. Zoning Codes, Development Standards, Subdivision Guidelines, and HOA Ordinances All the CWPPs development stand ards set adequate road widths for new construction; however, they do not document or address existing roads that are sub standard and need to be updated. Nor do they have a plan in place for updating said roads. None of the local governance implementations limit development in highrisk wildfire areas. Additionally, they do not incentivize the implementation of fuel breaks. None of the local governance codes and ordinances outline penalties for noncompliance. However, lack of incentives and disincentives is often because there are no clear Firewise design zoning or development guidelines or requirements. Building Codes All but two of the locations mention current international WUI building codes. However, it is worth n oting that plans are only reviewed if a building permit is required and on site

PAGE 293

276 inspections are not always performed to ensure WUI Firewise compliance. Additionally, many existing structures can be updated and rebuilt per outdated codes. Boulder_005 and Bouler_010 need to update their building codes. Plan Review and Inspection Procedures Local governments do have inspection and test procedures in place to test new development for adequate water supply. Additionally, many have ongoing efforts to ensure the maintenance and future testing of water supplies to ensure adequate water volume and pressure for fire suppression efforts. Additionally, local governance has procedures in place to ensure new subdivisions has multiple ingres ses and egress routes; however, they do not address existing development or solitary remote enclaves. No local gover nment requires developers or occupants have fire insurance. Additionally, no local government has inspection procedures in place for implementation or maintenance of proper WUI building code implementation or require pre development inspection of wildfire mitigation work. Error, Uncertainty, and Sensitivity The pilot study highlighted two modifiable areal unit problems (MAUP). The first is created by using CWPPs as the summary unit of analysis. The second is created by using census block groups for the ind ependent variables. A MAUP occurs whenever administrative boundaries are used to delineate units of analysis and to summarize data. CWPP WUI and project boundaries are created during the CWPP process. It is a localized deliberative process, emphasizing loc ally defined and identified priorities. As such, each community could define their boundaries and priorities differently, making comparisons difficult. Likewise, the census block groups are set by the Census Bureau. Due to the nature of census data , only t he Census Bureau has access to the data provided by each respondent . T hus the data cannot be summarized

PAGE 294

277 using CWPP boundaries. Combining these datasets introduces additional possibilities of error and uncertainty; however, without the ability to use differ ing delineations of boundaries for both CWPPs and census data it is not possible to test the sensitivity of the pilot study’s results. Additional error and uncertainty arise due to many small CWPP boundaries overlapping and sharing similar census block group units ( Figure Appendix C .3).

PAGE 295

278 Figure Appendix C. 3. The modifiable areal unit problem between CWPP boundary and census block groups.

PAGE 296

279 APPENDIX D CWPP BIBLIOGRAPHY Alpenfire LLC, & SE Group. (2012). Rio Blanco County Community Wildfire Protection Plan Update. Retrieved from Rio Blanco County, CO . Alpine Enterprises Inc. (2004). Blaine County Community at Risk Fire Mitigation Plan. Retrieved from Blaine County, ID . Alpine Sa fety Committee. (2010). Alpine Community Wil dfire Protection Plan. Retrieved from San Diego County, CA . Anchor Point Group. (2009). Inyo County, California: Community Wildfire Protection Plan. Retrieved from Inyo County, CA . Anchor Point Group. (2013). A lbany County Community Wildfire Protection Plan. Retrieved from Albany County, WY . Anchor Point Group, & AMEC Earth and Environmental. (2011a). Gunnison County, Colorado Community Wildfire Protection Plan. Retrieved from Gunnison County, CO . Anchor Point Group, & AMEC Earth and Environmental. (2011b). Montrose County, Colorado Community Wildfire Protection Plan. Retrieved from Montrose County, CO . Anchor Point Group LLC, & The Placitas Group Inc. (2008). San Miguel County, New Mexico Wildland Urban Inter face Community Wildfire Protection Plan. Retrieved from San Miguel County, NM . Arctos Research. (2005). Community Wildfire Protection Plan for Lake County, Montana. Retrieved from Lake County, MT . Baker County Commissioners. (2015). Baker County Communit y Wildfire Protection Plan. Retrieved from Baker County, OR . Barker, J., & Glenn, A. (2013). Harney County Community Wildfire Protection Plan. Retrieved from Harney County, OR . Barraclough, C., Buscombe, K., & Bowne, E. (2007). Community Wildfire Protect ion Plan. Retrieved from Fairplay, CO . Beck Consulting. (2005). Fallon County Community Wildfire Protection and Pre Disaster Mitigation Plan. Retrieved from Fallon County, MT . Bent County Commissioners. (2011). Community Wildfire Protection Plan: Bent County Fire. Retrieved from Bent County, CO . Big Sky Hazard Management LLC. (2007). Musselshell County, Montana Community Wildfire Protection Plan. Retrieved from Musselshell County, MT . Big Sky Hazard Management LLC. (2014). Madison County, Montana Community Wildfire Protection Plan. Retrieved from Madison County, MT . Board of Clackamas County Commissioners. (2018). Clackamas Community Wildfire Protection Plan. Retrieved from Clackamas County, OR . Bonneville County Commissioner. (2015). Bonneville County Community Wildfire Protection Plan (CWPP). Retrieved from Bonneville County, ID . Boulder County Board of County Commissioners. (2011). Boulder County Community Wildfire Protection Plan: Wildfire + Unprepared = Disaster, Management + Community = Pro tection. Retrieved from Boulder County, CO. Campbell County Fire Warden. Campbell County Community Wildfire Protection Plan. Retrieved from Campbell County, WY .

PAGE 297

280 Clearwater County Community Wildfire Protection Plan Committee. (2011). Clearwater County, Ida ho Community Wildfire Protection Plan. Retrieved from Clearwater County, ID. Cochise County Board of Supervisors. (2014). Cochise County Community Wildfire Protection Plan. Retrieved from Cochise County, AZ . Colvin, S., & Community Wildfire Protection Pla n Steering Group. (2009). Sherman County Community Wildfire Protection Plan. Retrieved from Sherman County, OR . Cossitt Consulting, Beck Consulting, & Herzberg, R. (2005). Wibaux County Community Wildfire Protection and Pre Disaster Mitigation Plan. Retrieved from Wibaux County, MT . Cossitt Consulting, & Herzberg, R. (2005a). McCone County Community Wildfire Protection and Pre Disaster Mitigation Plan. Retrieved from McCone County, MT . Cossitt Consulting, & Herzberg, R. (2005b). Prairie County Community Wildfire Protection and Pre Disaster Mitigation Plan. Retrieved from Prairie County, MT . County of Lemhi Wildland Urban Interface: Wildland Fire Hazard, Risk, & Mitigation Plan. (2006). Retrieved from Lemhi County, ID . Covington, E. (2003). Bingham County Wild Land Mitigation Fire Plan. Retrieved from Bingham County, ID . Crook County Commission. Crook County Community Wildfire Protection Plan. Retrieved from Cook County, WY . Diablo Fire Safe Council. Coummunity Wildfire Protection Plan: Con tra Costa County, California. Retrieved from Contra Costa County, CA . Dolores County Community Fire Plan. Retrieved from Dolores County, CO . Douglas County Community Wildfire Protection Plan Steering Committee. (2013). Douglas County, Washington Communit y Wildfire Protection Plan. Retrieved from Douglas County, WA . Dynamic Corporation. (2004). Wildland Fire Mitigation Plan Power County, Idaho. Retrieved from Power County, ID . Ellis, J., Wallace, G., & Reeves, S. (2005). Missoula County Community Wildfir e Protection Plan. Retrieved from Missoula County, MT . Emergency & Environmental Response Solutions. (2004). Jefferson County Idaho Wildland/Urban Interface Fire Mitigation Plan. Retrieved from Jefferson County, ID . firelogistics Inc. (2008). Golden Va lley County Community Wildfire Protection Plan. Retrieved from Golden Valley, MT . FireSafe Council of Siskiyou County. (2008). Siskiyou County Wildfire Protection Plan. Retrieved from Siskiyou County, CA . Fox Logic LLC. (2005). Community Wildfire Protect ion Plan (CWPP) Butte Silver Bow County, MT. Retrieved from Butte Silver Bow County, MT . Franklin County Community Wildfire Protection Plan Steering Committee. (2014). Franklin County, Washington Community Wildfire Protection Plan. Retrieved from Frankl in County, WA . Fremont County Sheriff's Office Wildland Fire. Fremont County Community Wildfire Protection Plan. Retrieved from Fremont County, CO. Fullerton Management Group. (2011). Calaveras County Community Wildfire Protection Plan. Retrieved from Cal averas County, CA . Gallamore, A., Griffin, T., Snart, R., Keith, R., & Fitzgerald, F. S. Jefferson County Community Wildfire Protection Plan. Retrieved from Jefferson County, CO .

PAGE 298

281 Gardiner, R. (2016). Taos County CWPP Update: Connecting Communities and Wa tersheds. Retrieved from Taos County, NM . Glenn County Resource Conservation District. (2011). Glenn County Community Widlfire Protection Plan. Retrieved from Glenn County, CA . Gullickson, K. (2005). Converse County Mountain Community Wildfire Protection Plan. Retrieved from Converse County, WY . Hammond, D. (2009). Custer County Idaho Wildland/Urban Interface Fire Mitigation Plan. Retrieved from Custer County, ID . Hawkins, J. R. (2009). Riverside Unit Fire Management Plan. Retrieved from Riverside County, CA. Hawley, L. (2013). Douglas County Community Wildfire Protection Plan. Retrieved from Douglas County, OR . Hollis, F., & Boykin, D. (2006). Socorro County Community Wildfire Protection Plan. Retrieved from Socorro County, NM. Hulbert, J. H. (2009). Polk County, Oregon Community Wildfire Protection Plan. Retrieved from Polk County, OR . Johnson County Fuel Mitigation Committee. (2017). Johnson County Community Wildfire Protection Plan. Retrieved from Johnson County, WY . Katelman, T. (2005). Del N orte Fire Safe Plan: Community Wildfire Protection Plan. Retrieved from Del Norte County, CA . Kittitas County Board of County Commissioners. (2009). Kittitas County Wildfire Protection Plan. Retrieved from Kittitas County, WA . La Plata County Community Wildfire Protection Plan (CWPP). Retrieved from La Plata County, CO. Land Stewardship Associates, L. (2008). Costilla County Community Wildfire Protection Plan. Retrieved from Costilla County, CO . Lincoln County, Montana Community Wildfire Prot ection Plan. (2005). Retrieved from Lincoln County, MT . Lynn, K., Ojerio, R., & Wolf, J. (2008). Curry County Community Wildfire Protection Plan. Retrieved from Curry County, OR . Marion County Board of Commissioners. (2008). Marion County Community Wildf ire Protection Plan. Retrieved from Marion County, OR . Maxim Technologies: Engineering & Environmental Consultants, & Bear Paw Development Corporation. (2005a). Blaine County Montana Community Wildfire Protection Plan. Retrieved from Blaine County, MT . Maxim Technologies: Engineering & Environmental Consultants, & Bear Paw Development Corporation. (2005b). Phillips County Community Wildfire Preparedness Plan. Retrieved from Phillips County, MT . Mendocino County Board of Supervisors, Mendocino COunty Fir e Chiefs' Association, & California Department of Forestry and Fire Protection. (2005). Mendocino County Community Wildfire Protection Plan. Retrieved from Mendocino County, CA . Modoc County Fire Safe Council. (2005). Community Wildfire Protection Plan: M odoc County, California. Retrieved from Modoc County, CA . Mohave County Board of Supervisors. (2008). Mohave County Community Wildfire Protection Plan. Retrieved from Mohave County, AZ .

PAGE 299

282 Monterey Fire Safe Council. (2010). Monterey County Community Wildfi re Protection Plan. Retrieved from Monterey County, CA . Montoya, N. (2008). Union County Community Wildfire Protection Plan. Retrieved from Union County, NM . North Wind Inc. (2004a). Camas County, Idaho Wildland Fire Hazard Mitigation Plan. Retrieved fro m Camas County, ID . North Wind Inc. (2004b). Gooding County, Idaho Wildland Fire Hazard Mitigation Plan. Retrieved from Gooding County, ID . Northwest Management Inc. (2004a). Judith Basin County, Montana Wildland Urban Interface Wildfire Mitigation Plan. Retrieved from Judith Basin County, MT . Northwest Management Inc. (2004b). Minidoka County, Idaho, WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Minidoka County, ID . Northwest Management Inc. (2004c). Petroleum County, Montana Wildla ndUrban Interface Wildfire Mitigation Plan. Retrieved from Petroleum County, MT . Northwest Management Inc. (2004d). Twin Falls County, Idaho WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Twin Falls, ID . Northwest Management Inc. (2004e). Valley County, Idaho WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Valley County, ID . Northwest Management Inc. (2005). Owyhee County, Idaho WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Owyhee County, ID . Nor thwest Management Inc. (2007a). Liberty County, Montana Community Wildfire Protection Plan. Retrieved from Liberty County, MT . Northwest Management Inc. (2007b). Pondera County, Montana, Community Wildfire Protection Plan. Retrieved from Pondera County, M T . Northwest Management Inc. (2007c). Washington County, Oregon Community Wildfire Protection Plan. Retrieved from Washington County, OR . Northwest Management Inc. (2008a). Asotin County, Washington Community Wildfire Protection Plan. Retrieved from Asot in County, WA . Northwest Management Inc. (2008b). Garfield County, Washington Community Wildfire Protection Plan. Retrieved from Garfield County, WA . Northwest Management Inc. (2009). Lincoln County, Washington Community Wildfire Protection Plan. Retriev ed from Lincoln County, WA . Northwest Management Inc. (2013). Okanogan County, Washington Community Wildfire Protection Plan. Retrieved from Okanogan County, WA . Northwest Management Inc. (2014). Fremont County, Wyoming Community Wildfire Protection Plan. Retrieved from Fremont County, WY . Northwest Management Inc., & Adams County Wildfire Urban Interface Wildfire Mitigation Plan Committee. (2004). Adams County, Idaho, WildlandUrban Interface Wildfire Mitigation Plan. Retrieved from Adams County, ID . Oregon Natural Hazards Workgroup. (2008). Lane County Community Wildfire Protection Plan. Retrieved from Lane County, OR . Otero County Sheriff and Emergency Manager’s Office. (2013). Community Wildfire Protection Plan Otero County Fire. Retrieved from Otero County, CO . Pend Oreilled County Coordinator of Emergency Services. (2011). Pend Oreille County CWPP 2011 Update. Retrieved from Pend Oreille County, WA .

PAGE 300

283 Prowers County Commissioners. (2013). Community Wildfire Protection Plan Prowers County. Retrieved from Prowers County, CO . Rath, D. (2008). Sweet Grass County Community Wildfire Protection Plan. Retrieved from Sweet Grass County, MT . Resource Concepts Inc. (2009). Carson City Community Wildfire Protection Plan. Retrieved from Carson C ity, NV . resourceLogic LLC. (2004). Community Fire Plan Moffat County, Colorado. Retrieved from Moffat County, CO . Rogers, G. (2016). Klamath County Community Wildfire Protection Plan. Retrieved from Klamath County, OR . Routt County Office of Emergency M anagement (OEM). (2010). Routt County Community Wildfire Protection Plan. Retrieved from Routt County, CO . Russell, K., Campbell, M., & Root, D. (2011). Community Wildfire Protection Plan for Unincorporated El Paso County "A Continuing Process". Retrieved from El Paso County, CO . San Juan County Community Wildfire Protection Plan Steering Committee. (2012). San Juan County, Washington Community Wildfire Protection Plan/Wildfire Risk Assessment. Retrieved from San Juan County, WA . Schlosser, W. E., & Mi erzwinski, J. F. (2012). Benewah County WildlandUrban Interface Wildfire Mitigation Plan Update. Retrieved from Benewah County, ID . SEC Inc. (2006). Community Wildfire Protection Plan Ciboloa County, NM. Retrieved from Cibola County, NM . Secrest Fire So lutions LLC. (2014). Park County Community Wildfire Protection Plan. Retrieved from Part County, MT . Shoshone County Community Wildfire Protection Plan Committee. (2011). Shoshone County, Idaho Community Wildfire Protection Plan. Retrieved from Shoshone C ounty, ID . Simons, T. (2006). Larimer County Fire Plan. Retrieved from Larimer County, CO . Steve Holl Consulting, & Wildland Rx. (2010). Final Tulare County Community Wildfire Protection Plan. Retrieved from Tulare County, CA . Stillwater County Communit y Wildfire Protection Plan and Pre Disaster Mitigation Steering Committee. Stillwater County Community Wildfire Protection Plan. Retrieved from Stillwater County, MT . SWCA Environmental Consultants. (2008a). De Baca County Community Wildfire Protection Pl an. Retrieved from De Baca County. SWCA Environmental Consultants. (2008b). De Baca County Community Wildf ire Protection Plan. Retrieved from De Baca County, NM: SWCA Environmental Consultants. (2014). Otero County Community Wildfire Protection Plan. Ret rieved from Otero County, NM . Technical Forestry Services LLC. (2008). Park County Community Wildfire Protection Plan. Retrieved from Park County, WY . Technical Forestry Services LLC. (2018). Big Horn County Community Wildfire Protection Plan 2017 Update . Retrieved from Big Horn County, WY . Teton All Hazard Mitigation Planning Committee. (2016). Teton County, Idaho Multi Jurisdiction All Hazard Mitigation Plan. Retrieved from Teton County, ID . The Bannock County LEPC, & Bannock County Emergency Services. (2003). Bannock County Wildland Urban Interface Fire Mitigaton Plan. Retrieved from Bannock County, ID .

PAGE 301

284 The South Central Mountain Resource Conservation & Development Council Inc. (2014). Lincoln County New Mexico Community Wildfire Protection Plan. Retrieved from Lincoln County, NM . Thomas, M., Wingert, J., Taylor, D., Lange, D., Bowne, E., Janzen, P., . . . Maxwell, K. (2009). Chaffee County Community Widlfire Protection Plan. Retrieved from Chaffee County, Colorado. Tillamook County Local Coordinating Group. (2010). Tillamook County Community Wildfire Protection Plan. Retrieved from Tillamook County, OR . Trinity County Resource Conservation District, & The Watershed Research and Training Center. (2010). Trinity County Community Wildfire Protection Plan. Retrieved from Trinity County, CA . TSS Consultants. Community Wildfire Protection Plan: Nevada County, California. Ret rieved from Nevada County, CA . Western Washington University Huxley College. (2012). Mason County Community Wildfire Protection Plan. Retrieved from Mason County, WA . Wheeler County Court. (2006). Wheeler County Community Wildfire Protection Plan. Retri eved from Wheeler County, OR . Wildland Fire Associates. (2007). Curchill County, Nevada Landscape Scale Wildland Fire Risk/Hazard/Value Assessment. Retrieved from Chuchill County, NV . Wildland Fire Associates. (2008a). Elko County, Nevada Landscape Scale Wildland Fire Risk/Hazard/Value Assessment. Retrieved from Elko County, NV . Wildland Fire Associates. (2008b). Lincoln County, Nevada Landscape Scale Wildland Fire Risk/Hazard/Value Assessment. Retrieved from Lincoln County, NV . Wildland Fire Associates . (2009). Washoe County, Nevada Landscape Scale Wildland Fire Risk/Hazard/Value Assessment. Retrieved from Washoe County, NV . Yakima County CWPP Steering Committee. (2015). Yakima County, Washington Community Wildfire Protection Plan. Retrieved from Yakim a County, WA . Yuma County Office of Emergency Management. (2010). Yuma County Community Wildfire Protection Plan. Retrieved from Yuma County, AZ .

PAGE 302

285 APPENDIX E LOCAL GOVERNANCE BI BLIOGRAPHY 2030 Update Tulare County general plan. (2012). Retrieved from Tulare County, CA . Adams County Land Use Department. (2006a). Adams County Subdivision Ordinance. Retrieved from Adams County, ID . Adams County Land Use Department. (2006b). Adams County Zoning Ordinance. Retrieved from Adams County, ID . Ayres, A., Degolia, A., Fienup, M., Kim, Y., Sainz, J., Urbisci, L., . . . Tague, C. (2016). Social science/natural science perspectives on wildfire and climate change. Geography Compass, 10(2), 6786. Bannock County Planning and Development Council. Wildland Interface Code. Retrieved from Bannock County, ID . Bannock County Planning and Development Council. The Zoning Ordinance of Bannock County, Idaho. Retrieved from Bannock County, ID . Bannock County Planning and Development Council. (1997). Subdivision Ordinance of Bannock County, Idaho. Retrieved from Bannock County, ID . Bannock County Planning and Development Council. (1998). Building Code Ordinance of Bannock County, Idaho. Retrieved from Bannock County, ID . Benewah County Planning and Zoning Commission. (2018). Benewah County, Idaho County Code. Retrieved from https .//www.sterlingcodifiers.com/codebook/index.php?book_id=877 Bent County Administrator. (2016). Planning and zoning manual. Retrieved from http .//www.bent county.net/document_center/index.php#revize_document_center_rz59 Berry, J. M., West, R. L., & Dennehey, D. M. (1989). Reliability and validity of the Memory Self Efficacy Questionnaire. Developmental Psychology, 25(5), 701. Bingham County Planning & Zoning. (2005). Bingham County Comprehensive Plan. Retrieved from Bingham County, ID . Bingham County Planning & Zoning. (2012). Bingham County Zoning Ordinance. Retrieved from Bingham County, ID . Birnie, M. (2017). Gunnison County Strategic Plan. Retrieved fr om Gunnison County, CO . Blaine County Land Use Department. (2012). Blaine County, Idaho County Code. Retrieved from https .//www.sterlingcodifiers.com/codebook/index.php?book_id=450 Blaine County Land Use Department. (2018). Blaine County Comprehensive Pla n Update. Retrieved from Blaine County, ID . Bonneville County Planning Commission. (2011). Bonneville County Zoning Ordinance. Retrieved from Bonneville County, ID . Bonneville County Planning Commission. (2013). Bonneville County Comprehensive Plan. Retr ieved from Bonneville County, ID . Boulder County Land Use Department. (2017). Boulder County Building Code. Retrieved from Boulder County, CO . Boulder County Land Use Department. (2018a). Boulder County comprehensive plan . goals, policies, and maps eleme nt. Retrieved from Boulder County, CO . Boulder County Land Use Department. (2018b). Boulder County Land Use Code. Retrieved from https .//www.bouldercounty.org/propertyand land/land use/planning/land use code/

PAGE 303

286 Busenberg, G. (2004). Wildfire management in the United States . the evolution of a policy failure. Review of Policy Research, 21(2), 145156. Butte Silver Bow Planning Board. The Butte Silver Bow, state of Montana, zoning ordinance. Retrieved from Butte Silver Bow County, MT . Butte Silver Bow Planning Board. (2009). The subdivision regulations of the consolidated city and county of Butte Silver Bow State of Montana. Retrieved from Butte Silver Bow County, MT . Butte Silver Bow Planning Board. (2012). International Building Code (2012 edition). Retrieved from Butte Silver Bow County, MT . http .//www.co.silverbow.mt.us/620/Building Code CAL FIRE. (2018a). Wildfire is coming. Are you ready toGO. In C. FIRE (Ed.), (pp. 8). Sacramento, CA . CAL FIRE. CAL FIRE. (2018b). Wildfire is coming. Ar e youSET. In C. FIRE (Ed.), (pp. 8). Sacramento, CA. CAL FIRE. Calaveras County Board of Supervisors. (2018). Calaveras County California code of ordinances. Retrieved from https .//library.municode.com/ca/calaveras_county/codes/code_of_ordinances?nodeId=C ALAVERAS_CO_CALIFORNIA_MUNICIPAL_CODE_PR Calaveras County Planning Commission. (2014). The Calaveras County General Plan. Retrieved from Calaveras County, CA . Camas County Planning and Zoning Commission. (2007). Camas County, Idaho Subdivision Ordinance. Retrieved from Camas County, ID . Camas County Planning and Zoning Commission. (2016a). Camas County, Idaho Zoning Ordinance. Retrieved from Camas County, ID . Camas County Planning and Zoning Commission. (2016b). Camas County, Idaho Zoning Ordinance. Retr ieved from Camas County, ID . Chaffee County Building Department. (2015). Chaffee County Building Codes and Design Criteria. Retrieved from http .//www.chaffeecounty.org/BuildingDept AdoptedCodes andDesign Criteria Chaffee County Planning and Zoning Depa rtment. (2014). Chaffee County Land Use Code. Retrieved from http .//www.chaffeecounty.org/Planningand ZoningLand Use Code CHAPTER 3. Mountainous, Forest , Brush and GrassCovered Lands, State of California, Division 4. Cong. Rec. Section 4291. (2019). Clarion Associates, & Ernst, S. (2008). Bannock County, Idaho Comprehensive Plan. Retrieved from Bannock County, ID . Clerk of the Board. (2018). Riverside County Ordinances. Retrieved from https .//countyofriverside.us/aboutthecounty/countyordinanc es.aspx Cochise County. (2015). Cochise County Comprehensive Plan. Retrieved from Cochise County, AZ . Cochise County. (2016). Cochise County Subdivision Regulations. Retrieved from Cochise County, AZ . Cochise County. (2017). Cochise County Zoning Ordinance. Retrieved from Cochise County, AZ . Community Development Services of Montana. (2008). Butte Silver Bow County Growth Policy. Retrieved from Butte Silver Bow County, MT . Consensus Planning Inc. (2000). Chaffee County Comprehensive plan. Retrieved from Chaffee County, CO .

PAGE 304

287 Contra Costa County Department of Conservation and Development. Contra Costa County Subdivision Ordinance. Retrieved from Contra Costa County, CA . Contra Costa County Department of Conservation and Development. (2005). Contra Costa C ounty General Plan. Retrieved from Contra Costa County, CA . Contra Costa County Department of Conservation and Development. (2016). Building codes, ordinances, and technical guidelines for building design. Retrieved from Contra Costa County, CA . Contra C osta County Department of Conservation and Development. (2017). Contra Costa County Zoning. Retrieved from Contra Costa County, CA . Costilla County Mitigation Advisory Committee. (2015). Draft Costilla County, Colorado Multi Jurisdictional Multi Hazard Plan. Retrieved from Costilla County, CO . Costilla County Planning Commission. Costilla County Land Use Code. Retrieved from Costilla County, CO . Costilla County Planning Commission. (1999). Costilla County Comprehensive Plan. Retrieved from Costilla Cou nty, CO . County of Glenn California. (1975). Glenn County Code. County Code Directory. Retrieved from https .//www.countyofglenn.net/govt/county code County of San Diego Planning and Development Services. (2017a). County Building Codes. Building Services. Retrieved from https .//www.sandiegocounty.gov/content/sdc/pds/bldg.html County of San Diego Planning and Development Services. (2017b). Zoning Ordinance. Retrieved from https .//www.sandiegocounty.gov/content/sdc/pds/zoning.html County of Tuolumne Community Resources Agency. (2018). 2018 Tuolumne County General Plan. Retrieved from Tuolumne County, CA . Custer County Executive Planning and Zoning Department. (2018). Custer County Planning, Zoning, Codes, and Ordinances. Retrieved from http .//www.co.custer. id.us/departments/executive/planning and zoning/ Del Norte County. (2016). County of Del Norte County Code. Retrieved from Del Norte County, CA. Department, C. C. B. P. Clearwater County Ordinances. Retrieved from https .//www.clearwatercounty.org/departme nts/building_plan/PlanningSub/zoningordina nce/index.php Department, V. C. P. Z. (2010). Valley County Comprehensive Plan. Retrieved from Valley County, ID . Dolores County Planning Commission. (2012). Dolores County Development and Land Use Regulations. Re trieved from Dolores County, CO . El Paso County Planning and Community Development. (2018). El Paso County, Colorado Land Development Code. Retrieved from https .//planningdevelopment.elpasoco.com/landdevelopment code/ Fallon County Planning Board. (2017) . Fallon County, City of Baker and Town of Plevna Subdivision Regulations. Retrieved from Fallon County, MT . Few, R., Brown, K., & Tompkins, E. L. (2007). Public participation and climate change adaptation. avoiding the illusion of inclusion. Climate Poli cy, 7(1), 4659. doi .10.1080/14693062.2007.9685637

PAGE 305

288 Flanagan Barry, E., Gregory Edward, W., Hallisey Elaine, J., Heitgerd Janet, L., & Lewis, B. (2011). A Social Vulnerability Index for Disaster Management. In Journal of Homeland Security and Emergency Management (Vol. 8). Fremont County Department of Planning and Zoning. (2014). Subdivision regulations of Fremont County, Colorado. Retrieved from Fremont County, CO . Fremont County Department of Planning and Zoning. (2015). Fremont County Masterplan. Retrieved from Fremont County, CO . Fremont County Department of Planning and Zoning. (2018). Fremont County, Colorado Zoning R esolution. Retrieved from Fremont County, CO . Glucker, A. N., Driessen, P. P. J., Kolhoff, A., & Runhaar, H. A. C. (2013). Public participation in environmental impact assessment . why, who and how? Environmental Impact Assessment Review, 43, 104111. doi .https .//doi.org/10.1016/j.eiar.2013.06.003 Golden Valley County Commissioners. (2012). Golden Valley Comprehensive Plan. Retrieved from Golden County, MT . Gooding County Planning and Zoning Commission. (2010). Gooding County Comprehensive Plan. Retrieved from Gooding County, ID . Gooding County Planning and Zoning Commission. (2011). Gooding County Subdivision Ordinance. Retrieved from Gooding County, ID . Gooding County Planning and Zoning Commission. (2013). The Zoning Ordinance of Gooding County, Idaho. Retrieved from Gooding County, ID . Gooding County Planning and Zoning Commission. (2014). Gooding County Building Code. Retrieved from Gooding County, ID . Gunnison County Board of County Commissioners. (2013). Gunnison County Land Use Resolution. Retrieved from Gunnison County, CO . Harland Bartholomew & Associates Inc. (2014). Nevada County General Plan. Retrieved from Nevada County, CA . HNTB. (2003). Prowers County Master Plan. Retrieved from Prowers County, CO . Inyo County Clerk of the Board. (2019). Inyo County Code. Retrieved from http .//www.qcode.us/codes/inyocounty/ Jefferson County Board of County Commissioners. (2019). Planning and Zoning. Document Center Development and Transportation. Retrieved from https .//www.jeffco.us/D ocumentCenter Jefferson County Comprehensive Plan Committee. (2005). Jefferson County Comprehensive Plan. Retrieved from Jefferson County, ID . Jefferson County Planning and Zoning Commission. (2015). Jefferson County Zoning Ordinance. Retrieved from Jefferson County, ID . Jefferson County Planning and Zoning Division. (2017). Jefferson County Comprehensive Master Plan. Retrieved from Jefferson County, CO . Jones & Stokes Associates, BRW, Mintier & Associates, & Applied Development Economics. (2001). Goals and Policies Report for the Inyo County General Plan. Retrieved from Inyo County, CA . Judith Basin County Planning Board, & Stahly Engineering & Assoc. Inc. (2016). Judith Basin County Growth Policy. Retrieved from Judith Basin County, MT . La Plata Count y Community Development Services. (1997). Code of La Plata County, Colorado. Retrieved from http .//online.encodeplus.com/regs/la plata co/doc viewer.aspx#secid 1

PAGE 306

289 La Plata County Community Development Services. (2017). La Plata County, Colorado Comprehensive Plan. Retrieved from La Plata County, CO . Lake County Planning Department. Subdivision regulations within Lake County. Retrieved from https .//www.lakemt.gov/planning/SubdivisionMain.html Lake County Planning Department. Zoning districts and regulations within Lake County. Retrieved from https .//www.lakemt.gov/planning/zoning.html Land Solutions LLC. (2018). Lake County, Montana 2018 Growth Policy. Building a Community of Communities. Retrieved from Lake County, MT . Larson, G., & Lincoln County Planning Department. (2009). Lincoln County Growth Policy. Retrieved from Lincoln County, MT . Lemhi County Planning and Zoning Commissioners. (2007). Lemhi County Comprehensive Plan. Retrieved from Lemhi County, ID . Lemhi County Planning and Zoning Commissioners. (2017). Lemhi County Development Code. Retrieved from Lemhi County, ID . Lincoln County Board of Commissioners. (2008). The code of the west. Retrieved from Lincoln County, NM . https .//www.lincolncountynm.gov/wpcontent/uploads/2017/12/Code_of_the_West_1c_bw.pdf Lincoln County Planning Department. (2015). Lincoln County Subdivision Regulations. Retrieved from Lincoln County, MT . LSC Transportation Consultants Inc, & Trinity County Planning Commission. (2012). Trinity County general plan. Ret rieved from Trinity County, CA . Lundberg, A. K., Richardson, T., & Hongslo, E. (2018). The consequences of avoiding conflict . lessons from conservation planning for Europe's last wild reindeer. Journal of Environmental Planning and Management, 120. doi .10.1080/09640568.2017.1409197 Mapp, M. J., Gardner, R., & Don Acheson Riedesel & Assoc. (2008). Twin Falls County Comprehensive Plan. Retrieved from Twin Falls, ID . Matthew Bender & Company Inc. (2018). Trinity County, CA code of ordinances. Retrieved from https .//library.municode.com/ca/trinity_county/codes/code_of_ordinances Minidoka County and Rupert City Planning and Zoning Commission. (2013). Minidoka County/City Comprehensive Plan. Retrieved from Minidoka County, ID . Minidoka County Planning and Zoni ng Commission. (2018). Minidoka Community Development Building and Zoning Codes. Retrieved from http .//www.minidoka.id.us/179/CommunityDevelopment Building andZonin Mintier & Associates, Jones & Stokes Associates, Lowens, S., & Del Norte County Community Development Department. (2003). Del Norte County General Plan. Retrieved from Mintier Harnish & Associates. (1988). Modoc County General Plan. Goals, Policies, and Action Program. Retrieved from Modoc County, CA . Modoc County Planning Department. Modoc County building code. Retrieved from Modoc County, CA . Modoc County Planning Department. (1991). Modoc County zoning ordinance. Retrieved from Modoc County, CA . Moffat County Planning Department. Moffat County Zoning Resolution. Retrieved from Moffat Cou nty, CO . Moffat County Planning Department. (2003). Moffat County Master Plan. Retrieved from Moffat County, CO .

PAGE 307

290 Moffat County Planning Department. (2004a). Minor subdivision regulations for Moffat County. Retrieved from Moffat County, CO . Moffat County Planning Department. (2004b). Subdivision regulations for Moffat County. Retrieved from Moffat County, CO . Mohave County Board of Supervisors. (2015). Mohave County Building Safety Code. Retrieved from Mohave County, AZ . Mohave County Department of Deve lopment Services. (2015). Mohave County Zoning Regulations. Retrieved from Mohave County, AZ . Mohave County Planning Department. (2008). Mohave County General Plan. Retrieved from Mohave County, AZ . Mohave County Planning Department. (2010). Mohave Count y Land Division Regulations. Retrieved from Mohave County, AZ . Mohave County Planning Department. (2011). Mohave County Minor Land Division Regulations. Retrieved from Mohave County, AZ . Monterey County board of supervisors. (1992). Monterey County Calif ornia code of ordinances. Retrieved from https .//library.municode.com/ca/monterey_county/codes/code_of_ordinances?nodeId=TI T1GEPR Monterey County planning commission. (2010). County of Monterey general plan. Retrieved from Monterey County, CA . Montrose County Board of County Commissioners. (2015). Montrose County Subdivision Regulations. Retrieved from Montrose County, CO . Montrose County Board of County Commissioners. (2016a). Montrose County Building Code. Retrieved from Montrose County, CO . Montrose County Board of County Commissioners. (2016b). Montrose County Zoning Resolution. Retrieved from Montrose County, CO . Montrose County Planning Department. (2010). 2010 Montrose County Master Plan. Retrieved from Montrose County, CO . National Volunteer Fire Council. (2019). Wildland Fire Assessment Program Home Assessment Checklist. In N. V. F. Council (Ed.) . National Volunteer Fire Council. Otero County Land Use Department. (2013). Otero County Land Use Code. Retrieved from Otero County, CO . Owyhee County Community Development Department. (2009a). Owyhee County Subdivision Regulations. Retrieved from Owyhee County, ID . Owyhee County Community Development Department. (2009b). Owyhee County Zoning Regulations. Retrieved from Owyhee County, ID . Owyhee County Community Development Department. (2010). Owyhee County Comprehensive Plan. Retrieved from Owyhee County, ID . Park County Planning Department. (2011). Park County Land Use Regulations. Retrieved from http .//parkco.us/189/LandUse Regulations PMC. (2009). The County of Mendocino. General plan. Retrieved from Mendocino County, CA . Power County Planning and Zoning Board. (2018a). Power County Comprehensive Plan. Retrieved from Power County, ID . Power County Planning and Zoning Board. (2018b). Power County, Idaho County Code. Retrieved from https .//www.sterlingcodifiers.com/codebook/index.php?book_id=838

PAGE 308

291 Prowers County Planning Commission. (2006). Prowers County Subdivision Regulations. Retrieved from Prowers County, CO . Prowers County Plannin g Commission. (2012). Prowers County Zoning Regulations. Retrieved from Prowers County, CO . Quad Consultants, BrownBuntin Associates Inc., & Dowling Associates. (1993). Policy Plan. Glenn County General Plan. Retrieved from Glenn County, CA . Riverside C ounty Planning Department. (2015). County of Riverside general plan. Retrieved from Riverside County, CA . Roser, J., Crafton, B., Cobble, J., Davis, B., Priest, D., King, T., & Wilson, L. (2014). Camas County, Idaho Comprehensive Land Use Plan. Retrieved from Camas County, ID . RPI Consulting LLC. (2016). Park County Strategic Master PLan. Retrieved from Park County, CO. Sage Community Resources. (2006). Adams County Comprehensive Plan. Retrieved from Adams County, ID . Shoshone County Planning and Zoning Department. Shoshone County Ordinance and Code Retrieved from https .//shoshonecounty.id.gov/planningzoning/ Siskiyou County Board of Supervisors. (2018). Siskiyou County, California code of ordinances. Retrieved from https .//library.municode.com/ca/siski you_county/codes/code_of_ordinances Siskiyou County Planning Department. (1973). General plan Siskiyou County, California. Retrieved from Siskiyou County, CA . sites southwest. (2007). Lincoln County Comprehensive Plan. Retrieved from Lincoln County, NM . https .//www.lincolncountynm.gov/wpcontent/uploads/2017/12/Final_Comp2.pdf Supervisors, M. C. B. o. (2018). Code of Ordinances. Retrieved from https .//library.municode.com/ca/mendocino_county/codes/code_of_ordinances Teton County Comprehensive Plan Committ e. (2012). Comprhensive Plan A vision and framework 20122030. Retrieved from Teton County, ID . Teton county Planning and Zoning Department, & Teton County Building Department. (2017). Teton County Code and Policies. Retrieved from https .//www.tetoncoun tyidaho.gov/codePolicy.php The Board of Supervisors of the County of Nevada California. (1972). The Administrative Code of the County of Nevada, California. Retrieved from http .//qcode.us/codes/nevadacounty/view.php?topic=0&frames=on The Board of Supervisors of the County of San Diego. (2017). County of San Diego 2017 consolidated fire code. Retrieved from San Diego County, CA . Tulare County Resource Management Agency. (2018). Tulare County Planning and Building. Retrieved from https .//tularecounty.ca.gov/ rma/index.cfm/planningbuilding/ Tuolumne County California County Counsel. (2018). Tuolumne County Ordinance Code. Retrieved from https .//www.tuolumnecounty.ca.gov/165/Tuolumne CountyOrdinance Code#17 Twin Falls Planning and Zoning Department. (2018). Twin Falls County, Idaho County Code. Retrieved from https .//www.sterlingcodifiers.com/codebook/index.php?book_id=525 University of California, & Interagency Engineering Working Group. (2000). Property Inspection Guide, 2000 version. Retrieved from California .

PAGE 309

292 Valley County Planning & Zoning Department, & Valley County Planning & Zoning Department. (2018). Valley County, Idaho County Codes. Retrieved from https .//www.sterlingcodifiers.com/codebook/index.php?book_id=922 Wardlaw, M. (2018). County of San Diego General Plan. Retrieved from San Diego County, CA . Western Slope Consulting LLC, & Central Mountain Planning LLC. (2011). Rio Blanco County Master Plan. Retrieved from Ri Blanco County, CO . Wildfire Prevention Associates LLC. (2018). Valley County Wildland Urban Interface Fire Protection Plan Manual. Retrieved from Valley County, ID . Yuma County Building Official. (2003). Dry Hydrants a nd Construction Standards. Retrieved from Yuma County, AZ . Yuma County Department of Development Services. (2015). Yuma County Subdivision Regulations. Retrieved from Yuma County, AZ . Yuma County Department of Development Services. (2017). Yuma County 2020 Comprehensive Plan. Retrieved from Yuma County, AZ . Yuma County Department of Development Services. (2018). Yuma County Zoning Ordinance. Retrieved from Yuma County, AZ .

PAGE 310

293 APPENDIX F GIS DATA BIBLIOGRAP HY Manson, S., Schroeder, J., Riper, D. V., & R uggles, S. (2018). IPUMS National Historical Geographic Information System [Database]. Volker, R. C., Helmers, D. P., Kramer, H. A., Mockrin, M. H., Alexandre, P. M., Bar Massada, A., . . . Stewart, S. I. (2017). The 19902010 wildlandurban interface of the conterminous United States geospatial data [Vector Digital Data].

PAGE 311

294 APPENDIX G CROOK COUNTY, WYOMI NG ’ S RISK FIELD ASSESSM ENT The following is an excerpt of Crook County, Wyoming’s CWPP field assessment narratives (Crook County Commission, pp. 35) .

PAGE 312

295

PAGE 313

296 A PPENDIX H MOHAVE COUNTY, ARIZ ONA FUEL TREATMENT AND MODIFI CATION PLAN Source: (Mohave County Board of Supervisors, 2008, pp. 5253) .

PAGE 314

297

PAGE 315

298 APPENDIX I CAL FIRE’S WILDFIRE EVACUATION GUIDE. Source: (CAL FIRE, 2018a) .

PAGE 316

299

PAGE 317

300

PAGE 318

301

PAGE 319

302

PAGE 320

303

PAGE 321

304

PAGE 322

305

PAGE 323

306 APPENDIX J CAL FIRE’S EVACUATIO N PLANNING GUIDE Source: (CAL FIRE, 2018b) .

PAGE 324

307

PAGE 325

308

PAGE 326

309

PAGE 327

310

PAGE 328

311

PAGE 329

312 APPENDIX K MODEL COMMUNITY WIDLFIRE EVACUATION GUI DELINES Source: ( County of Lemhi Wild land Urban Interface: Wildland Fire Hazard, Risk, & Mitigation Plan , 2006) .

PAGE 330

313

PAGE 331

314

PAGE 332

315

PAGE 333

316

PAGE 334

317

PAGE 335

318

PAGE 336

319

PAGE 337

320

PAGE 338

321 APPENDIX L CALFIRE’S HOME IGNI TION ZONE INSPECTION FORM . Source: (Headwaters Economics, 2014, pp. A 2 A 3)

PAGE 339

322

PAGE 340

323 APPENDIX M WFAP HOME ASSESSMENT CHECKLIST. Source: (National Volunteer Fire Council, 2019) .

PAGE 341

324

PAGE 342

325

PAGE 343

326

PAGE 344

327

PAGE 345

328

PAGE 346

329

PAGE 347

330