University of Southern California Dissertations and Theses

The wildland-urban interface in Lassen County, California: a change analysis 2000–2015

(USC Thesis Other)

The wildland-urban interface in Lassen County, California: a change analysis 2000–2015

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE WILDLAND-URBAN INTERFACE IN LASSEN COUNTY, CALIFORNIA:

A CHANGE ANALYSIS 2000 – 2015

by

Joshua D. Greetan

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
(GEOGRAPHIC INFORMATION SCIENCE AND TECHNOLOGY)

May 2016

Copyright 2016 Joshua D. Greetan

ii

ACKOWLEDGMENTS
I would like to thank my thesis advisor, Dr. Darren Ruddell, and my two thesis committee
members, Dr. Katsuhiko “Kirk” Oda and Dr. Travis Longcore, for their feedback, guidance, and
suggestions throughout the whole process. I would also like to thank Dr. Robert Vos for helping
me to clarify my ideas into a thesis proposal. Finally, I would like to thank my family and friends
for their love and support throughout the entire degree program.

iii

TABLE OF CONTENTS
ACKOWLEDGMENTS ................................................................................................................. ii
LIST OF TABLES ......................................................................................................................... iv
LIST OF FIGURES ....................................................................................................................... vi
LIST OF ABBREVIATIONS ....................................................................................................... vii
ABSTRACT ................................................................................................................................. viii
CHAPTER 1: INTRODUCTION ................................................................................................... 1
1.1 Wildland-Urban Interface ............................................................................................. 1
1.2 Motivation ..................................................................................................................... 3
1.3 Case Study Area: Lassen County California ................................................................ 3
1.4 Organizational Framework .......................................................................................... 7
CHAPTER 2: BACKGROUND AND RELATED WORK ........................................................... 8
2.1 Geographic Modeling of the WUI ................................................................................ 8
2.1.2 Dasymetric Mapping Approaches................................................................ 10
2.2 Definition Effect ......................................................................................................... 17
2.3 Summary ..................................................................................................................... 20
CHAPTER 3: METHODS AND DATA ...................................................................................... 21
3.1 Method Framework ..................................................................................................... 21
3.1.1 Output Products ........................................................................................... 21
3.1.2 : Data Sources .............................................................................................. 21
3.2 Data Selection ............................................................................................................. 23
3.2.1 Residential Parcels ....................................................................................... 24
3.2.2 Housing Structures as Points ....................................................................... 26
3.2.3 Calculating the WUI .................................................................................... 30
3.2.4 Removal of Non-Wildland Fuel Types ........................................................ 34
3.3 Wildfire Hazard Assessment....................................................................................... 34
3.4 Hardware Used............................................................................................................ 35
CHAPTER 4: RESULTS .............................................................................................................. 36
4.1 Residential Parcel Identification ................................................................................. 36
4.2 Analyzing the WUI ..................................................................................................... 37
4.3 Land Ownership .......................................................................................................... 39
4.4 Residential Structures ................................................................................................. 42
4.4.1 Fire Hazard Severity Zones (FHSZ) ............................................................ 42
4.4.2 Fire Responsibility Areas ............................................................................. 47
CHAPTER 5: DISCUSSION AND CONCLUSIONS ................................................................. 50
5.1 Benefits of the Cadastral-Based Approach ................................................................. 50
5.2 Disadvantages of the Cadastral-Based Approach ....................................................... 51
5.2.1 Limitations of Housing Data ........................................................................ 51
5.2.2 Limitations of Other Data and GIS Tools .................................................... 53
5.3 Future Research .......................................................................................................... 55
5.3.1 Existing Vegetation ...................................................................................... 55
5.3.2 Home Ignition Zone .................................................................................... 55
5.3.3 Social Variables ........................................................................................... 56
5.4 Conclusion .................................................................................................................. 56
REFERENCES ............................................................................................................................. 58

iv

Appendix A ................................................................................................................................... 61

v

LIST OF TABLES
Table 1: Past WUI studies & their methods.................................................................................. 16
Table 2: Datasets and Sources ...................................................................................................... 24
Table 3: Lassen County Assessor Use Code Index ...................................................................... 25
Table 4: Changes in WUI land ownership acreage 2000- 2015 ................................................... 40
Table 5: Number of homes within each FHSZ from 2000-2015 .................................................. 43

vi

LIST OF FIGURES
Figure 1: Lassen County public versus private lands ..................................................................... 5
Figure 2: Example of dasymetric layers, to be removed, overlaid upon census blocks (Tully
2013) .....................................................................................................................................….. 11
Figure 3: Example of census block after the removal of dasymetric layers (Tully 2013) ……... 11
Figure 4: Comparison of result of Wilmer & Aplet (2005) and Stewart et al. (2005) (Source:
Stewart et al. 2009) ............................................................................................................... 19
Figure 5: Parcels in Lassen County .............................................................................................. 27
Figure 6: Parcel features converted to centroids ........................................................................... 28
Figure 7: Hand digitization of houses as points. Centroids moved to location of residence. ....... 29
Figure 8: 2000 housing stock buffer before and after wildland fuel removal .............................. 32
Figure 9: 2015 housing stock buffer before and after wildland fuel removal .............................. 33
Figure 10: Map showing WUI growth from 2000 to 2015 ........................................................... 38
Figure 11: Map showing 2015 WUI and land ownership ............................................................. 41
Figure 12: Number of homes in 2000 & 2015 WUIs and their relation to FHSZ ........................ 44
Figure 13: Total 2015 WUI acreage as it relates to FHSZ ........................................................... 45
Figure 14: 2015 WUI overlaid on CAL FIRE FHSZ data ............................................................ 46
Figure 15: 2015 WUI acreage by responsibility area ................................................................... 48
Figure 16: 2015 WUI overlaid over responsibility areas .............................................................. 49

vii

LIST OF ABBREVIATIONS
APN Assessor Parcel Number
BLM Bureau of Land Management
CAL FIRE California Department of Forest and Fire Protection
CFPZ Community Fire Protection Zone
CWPP Community Wildfire Protection Plan
FHSZ Fire Hazard Severity Zone
FRA Federal Responsibility Area
GIS Geographic Information System
HFRA Healthy Forest Restoration Act of 2003
LRA Local Responsibility Area
NLCD National Land Cover Database
SRA State Responsibility Area
USFS United States Forest Service
USGS United States Geological Survey
WUI Wildland-Urban Interface

viii

ABSTRACT

The expansion of low-density residential development into wildlands places a variety of stresses
on Earth’s natural systems. Urban sprawl is one of the many factors contributing to water
shortages, higher resource costs, and increasingly destructive storms, all of which fit under the
umbrella of global climate change (Gencer 2013). In California, it is wildfires that capture the
headlines during the summer and fall months. With wildfires becoming a common occurrence
within the California landscape, it is crucial that local government and agencies work to create
proactive approaches to mitigate the severity of these events. The areas of chief concern occur
where structures and infrastructure are intermixed or adjacent to wildland fuels; this area is
known as the wildland-urban interface (WUI). Mapping the WUI is a useful resource for
assessing which communities and developed areas are most vulnerable to wildfire. County
cadastral data was used to assess the spatial extent of the WUI within Lassen County, California.
This method is a large improvement to past studies because it eliminates the need to estimate the
areas where housing is located. California Department of Forest and Fire Protection (CAL
FIRE) Fire Hazard Severity Zone Maps were then overlaid to identify the areas in most need of
emergency planning and preventative action. Results show that in 2015, the spatial extent of
Lassen County’s WUI was 1,016.67 sq. km. (392.54 sq. mi) and accounted for 101,666.68
hectares (251, 222.83 acres) or 8.32% of the study area. The WUI area included 5,456
residential structures (49.51% of the total housing), which was a 13.22% increase and a 1.89
percentage point increase from 2000. 81% of the WUI occurs on private lands.

1

CHAPTER 1: INTRODUCTION
The expansion of low-density residential development into wildland areas places a variety of
hardships on Earth’s natural systems. As communities sprawl, they replace agricultural lands
and natural habitat, while simultaneously pillaging resources (Benfield 2001). Urban sprawl is
one of the many factors contributing to severe drought, higher resource costs, and increasingly
destructive storms, all of which fit under the umbrella of global climate change (Gencer 2013).
In California, it is wildfires that capture the headlines during the summer and fall months.
Sprawling urban development and human alteration of natural fire regimes have increased
California’s susceptibility to wildfires, which increases the risk to lives, natural resources, and
the built environment (Theobald 2007). When residents place their homes in the wildland-urban
interface they increase their vulnerability to wildfire.
1.1 Wildland-Urban Interface
The wildland-urban interface (WUI) refers to the area where human development meets natural
areas such as forest, shrub lands, or grasslands (Radeloff et al. 2005). The US federal
government recognizes three types of WUI: interface, intermix, and occluded. Definitions for
each of the three types were published in the 2001 Federal Register (66 FR 751 2001). Interface
communities are located immediately adjacent to wildland fuels. This means that there is a clear
line between development and the wildland fuels. Interface communities typically have a higher
density than intermix areas, but have been operationalized in past studies using a housing density
threshold of one housing unit per 40 acres (Stewart et al. 2003; Radeloff et al. 2005; Wilmer &
Aplet 2005; Tully 2013).
Intermix communities are comprised of structures that are dispersed throughout wildland
fuels. Intermix communities do not have a clear line of separation between development and

2

fuels. The housing density threshold of one housing unit per 40 acres is also used to
operationalize the definition (Stewart et al. 2003; Radeloff et al. 2005; Tully 2013). The
occluded community is generally located within an urban area, with structures that immediately
abut an isolated area of wildland fuels. The two major elements that distinguish between the
types of WUI are the location and density of the community in relation to wildland fuels.
Previous studies have relied upon a variety of geospatial techniques to implement the Federal
Register’s definitions (Stewart et al. 2003; Radeloff et al. 2005; Wilmer and Aplet 2005,
Theobald and Romme 2007; Tully 2013).
Nationally, humans cause 90% of wildfires (National Park Service n.d). The California
Department of Forestry and Fire Protection (CAL FIRE) estimates that 95% of fires, within the
State are caused by humans. Therefore, the probability for wildfires increases as the WUI
expands. The importance of the WUI increases as development encroaches upon the
surrounding natural environment, making these types of communities more prone to wildfires.
California has the most homes in the U.S. located within the WUI at 5.1 million units (Radeloff
et al. 2005). Furthermore, 7% of the state’s landmass, 24% of the state’s population, and 26% of
its housing units are located within the WUI (Tully 2013). This places a large financial strain
upon local, state, and government agencies. These agencies not only provide funding for
proactive wildfire mitigation measures, but must also bear the costs of fighting wildfires.
The purpose of this study is to conduct a change analysis of the wildland urban interface
(WUI) in Lassen County, California between 2000 and 2015. The study examines if the spatial
extent of the WUI has increased or decreased and where any changes have occurred. The WUI
maps are overlaid on CAL FIRE’s Fire Hazard Severity Zone Map for Lassen County. Together,
these maps will help to identify the areas in most need of fire and natural resource management.

3

Specifically, this study will help local and state fire districts identify the areas in most need of
fuel reductions, fire mitigation, and resource allocations.
1.2 Motivation
In the conterminous United States, 38.5% of the nation’s housing units are located within a WUI
and about 99 million people live within a WUI (Radeloff et al. 2005; Stewart et al. 2010). The
Healthy Forest Restoration Act of 2003 (HFRA) helps to protect communities against
catastrophic wildfires by allocating federal funding to help reduce fire risk to homes located
within a WUI. Specifically, HFRA provides funding for hazardous fuel reduction on federal
lands. This is primarily based upon the dangers associated with a wildfire on federal lands
spreading to a nearby community. The HFRA prioritizes project funding to communities with a
community wildfire protection plan (CWPP). This ensures that local agencies have the
opportunity to influence where hazardous fuel reduction projects take place within their
jurisdiction.
California Senate Bill 1241 mandates that city and county jurisdictions revise the Safety
Element chapter of their General Plan to address wildfires. Specifically, these jurisdictions must
delineate local, state, and federal responsibility areas and address preventative wildfire measures
for all CAL FIRE “very high” fire hazard severity zones areas. It also requires that the Safety
Element be reviewed and updated as necessary to address the risk of fire in these areas in
accordance with the Governor’s Office of Planning and Research’s “Fire Hazard Planning”
document. The proposed study would help to fulfill the County of Lassen’s State-required duty.
1.3 Study Area: Lassen County, California
Lassen County is located in the northeast section of California; it is bordered by Nevada to east.
The county’s area is 12,224.74 km² (4,720 mi²). In 2012, it was also the site of the Rush Fire,

4

the second largest wildfire recorded in California history. The fire consumed 127,701.38
hectares (315,557 acres). Lassen County is particularly interesting as all twenty of its
communities were listed in the Federal Register’s List of Communities at Risk to Wildfire in
2001. The majority of the county is comprised of public lands. The county, state, and federal
government own or manage 63% of the land (Figure 1).

5

Figure 1: 2015 Land Ownership in Lassen County. Map shows the distribution of public versus private lands

6

Previous academic studies have mostly addressed the mapping of the WUI at a national
scale (Stewart et al. 2003; Radeloff et al. 2005; Theobald and Romme 2007; Tully 2013);
however, a handful of studies address the issue within smaller, more focused case study areas
(Wilmer and Aplet 2005; Rozmajzl 2012). This project assesses the WUI at a scale in between
those of the past endeavors, which will allow for a more precise and accurate measurement of the
WUI.
Since the study doubles as a professional project, I had access to County data not
available to the public. With access to the County Assessor’s GIS files, the potential to model
housing density at a more precise level than the 2010 Census may exist. This study expands
upon dasymetric techniques used in Wilmer and Aplet’s (2005) model for mapping the
Community Fire Protection Zone (CFPZ). One improvement to their model will be the
utilization of cadastral data to eliminate the need of US Census Blocks for estimating home
locations.
The Assessor’s parcel database contains a wealth of property and owner information and
has the potential to map housing density in rural settings not recorded by the decennial Census.
The Lassen County Public Works Department’s transportation infrastructure shapefiles are more
current than the 2010 Census roads shapefile. Finally, the County’s rapport with other local,
state, and federal agencies aided in the completion of this project; CAL FIRE was especially
integral to this study, as they provided a wealth of data and informational resources.
CAL FIRE’s most recent Fire Hazards Severity Zone Map for Lassen County was
published in 2007. The updated WUI map builds upon CAL FIRE’s hazard research and also
provided the opportunity to utilize the most current datasets. These datasets include: 2015
Lassen County parcel data and National Land Cover Database (NLCD).

7

1.4 Organizational Framework
This thesis contains four additional chapters. Chapter 2 explores previous WUI mapping studies
in order to analyze the variety of approaches available and the differences between them. This
chapter serves as the cornerstone for operationalizing the WUI in the methods chapter. Chapter
3 presents the proposed cadastral-based approach for this study. Additionally, it provides the
framework for measuring and analyzing the WUI. Chapter 4 presents the results of the proposed
modeling technique. It is comprised of the outputs mentioned in Chapter 3. Chapter 5 discusses
the implications of the study results, the successes and shortfalls associated with the proposed
approach, and suggests opportunities for future research.

8

CHAPTER 2: BACKGROUND AND RELATED WORK
Although many studies use the same data and implement the federal definitions in a similar
manner, the techniques employed yield conflicting outputs (Stewart et al. 2009; Tully 2013).
The following chapter provides a discussion of the discrepancies between data, models, and the
analytical techniques of previous WUI studies. The first section discusses the history of WUI
mapping and the variety of methods used over time. The second section explores the definition
effect (Platt 2010; Tully 2013).
2.1 Geographic Modeling of the WUI
The first study to operationalize the Federal Register’s definitions for WUI used 2000 US Census
Block Group data with a minimum housing density of 1 housing unit per 40 acres, and 30-meter
pixel National Land Cover Data (NLCD 1992) to map the WUI in the conterminous US (Stewart
et al. 2003). The study included the following land cover classes as wildlands in its assessment:
forests (coniferous, deciduous, and mixed), native grasslands, shrubs, wetlands, and transitional
lands (e.g. clear-cuts). It excluded orchards, arable lands (agricultural lands and row crops), and
pasture (Stewart et al. 2003). The analysis addresses two types of WUI: interface and intermix.
The study’s geographic modeling is primarily based upon the definitions provided within
the Federal Register (66 FR 751 2001); however, Stewart et al. further define each of the two
types to visualize them within a geographic information system (GIS). According to the Federal
Register, the interface WUI occurs “within the vicinity” of wildlands vegetation, however the
term “vicinity” is not defined. Stewart et al. use a 2.4 km (1.5 mi) buffer that extends from
Census blocks meeting the minimum housing density criteria. This is based on the estimated
distance a firebrand can travel from a wildfire, through the air, and potentially ignite a home or
fuel near it (California Fire Alliance 2001; Stewart et al. 2003). Interface areas have 1 or more

9

homes per 40 acres, have less than 50% vegetation cover, and are within 1.5 mi of an area
comprising 500 ha (1236 acres) with vegetation cover of more than 75%. Intermix areas have
the same criteria for housing density threshold, but must have more than 50% vegetation cover
(Stewart et al. 2003).
While Stewart et al. calculate the land mass and locations for both interface and intermix
separately, they use the term WUI interchangeably for both types and aggregate the two types for
the final map deliverable. They concluded that the WUI covered 9.3% (175 million acres) of the
conterminous US and 36.7% (42.2 million) of the total housing units. A major limitation of the
study was that it did not address fire risk as national data was not yet available.
Radeloff et al. (2005) further refined Stewart et al’s (2003) efforts to map the WUI within
the conterminous US by modifying vegetation classifications. The vegetation and land cover
types used were nearly identical to those of Stewart et al. (2003), with the addition of
grasslands/herbaceous and woody and emergent herbaceous wetlands. The study expanded upon
the classes omitted with the following exclusions: low- and high-intensity residential,
commercial/industrial, orchards/vineyards, pasture/hay, row crops, small grains, fallow, urban
recreational grasses, bare rock/sand/clay, quarries open water, and perennial ice/snow (Radeloff
et al. 2005). Similar to Stewart et al. (2003) housing density and vegetation coverage were then
calculated for each Census block to operationalize intermix and interface areas.
Definitions for interface remained similar to those used by Stewart et al. (2003) and had
only a few additional criteria. Radeloff et al. used a minimum threshold area of 3.1 sq. mi. for
heavily vegetated areas within the 1.5 mi buffer in order to exclude urban parks from the
assessment. Another key alteration to Stewart et al.’s (2003) model was the splitting of Census
blocks.

10

Radeloff et al. split and then removed portions of Census blocks falling outside of the 1.5
mile buffer area from wildlands. The results of the analysis were similar to those of Stewart et
al.’s (2003) analysis. Radeloff et al. concluded that the WUI comprised 9.4% (177.7 million
acres) of the conterminous US and included 38.5% (44.3 million) of the total housing units.
2.1.2 Dasymetric Mapping Approaches
Dasymetric mapping is a type of areal interpolation that uses ancillary data to more accurately
distribute an attribute over a large areal unit. When mapping the WUI, dasymetric techniques
are used to remove areas such as hydrographic features and public lands, unlikely to contain
housing units from the US Census blocks. Housing unit density is then recalculated using the
original housing unit counts divided by the area of the refined Census block. The removal of
areas unlikely to contain housing allows for a more accurate spatial distribution of housing
across the Census blocks. Wilmer and Aplet (2005) was one of the first studies to use these
methods for mapping the WUI (see Figures 2 and 3).

11

Figure 2: Example of dasymetric layers, to be removed, overlaid upon census blocks (Tully 2013)

Figure 3: Example of census block after the removal of dasymetric layers (Tully 2013)

12

In 2003 Congress adopted the Healthy Forest Restoration Act (HFRA). The HFRA helps
to protect communities against catastrophic wildfires by allocating federal funding to help reduce
fire risk to homes located within a WUI. Specifically, HFRA provides funding for hazardous
fuel reduction on federal lands. This is primarily based upon the dangers associated with a
wildfire on federal lands spreading to a nearby community. The HFRA prioritizes project
funding to communities with a community wildfire protection plan (CWPP). This ensures that
local agencies have the opportunity to influence where hazardous fuel reduction projects take
place within their jurisdiction.
Since the HFRA provides incentive for communities to establish a CWPP, Wilmer and
Aplet (2005) aimed to develop a consistent method for using a GIS to map communities prone to
wildfires. Wilmer and Aplet’s study sought to define and map an area they refer to as the
“community fire planning zone” (CFPZ) in three separate case study areas: The Colorado Front
Range, the central Idaho Ecosystem, and the Greater Yosemite area in California. The study
built upon earlier models by using a dasymetric mapping approach to remove public lands from
the analysis (Schmidt et al. 2002; Aplet and Wilmer 2003; Stewart et al. 2003).
Wilmer and Aplet removed public lands from 2000 Census blocks, recalculated housing
density for each block, and then selected those which met the minimum housing density
threshold of 1 housing unit per 40 acres. Instead of discriminating between the two WUI
categories (intermix and interface), Wilmer and Aplet isotropically buffered the modified census
blocks by 0.5 mi in order to encapsulate a community’s physical footprint. They then removed
non-wildland fuel classes (water, barren, rock, agriculture and urban areas) from census blocks
meeting density criteria to approximate the CFPZ. Since the WUI is defined as the developed
area intermixed or adjacent to wildland fuels, this decreases the spatial extent of the WUI but

13

retains the CFPZ area. While the study did not map the WUI at a national scale it created a
simple method for calculating the CFPZ and WUI by utilizing dasymetric approaches to modify
census blocks.
Theobald and Romme (2007) sought to refine previous methods for modelling the WUI
(Stewart et al. 2003; Aplet and Wilmer 2003; Wilmer and Aplet 2005; Radeloff et al. 2005) in
order to create a more spatially precise map of the WUI within the conterminous Unites States.
Expanding upon Wilmer and Aplet’s dasymetric mapping techniques (Wilmer and Aplet 2005),
Theobald and Romme utilized the USGS’s Protected Areas Database (PADUS) to remove public
lands and protected areas from 2000 Census blocks. The study also utilized Census hydrography
data to remove water features from Census blocks.
Unlike previous works (Stewart et al. 2003; Wilmer and Aplet 2005; Radeloff et al.
2005), the Theobald and Romme buffered blocks that met the minimum housing density
threshold of 1 housing unit per 40 acres with a variable-width buffering technique based upon
cost-weighted distances of surrounding vegetation types and their probability to spread wildfire.
Instead of solely using a uniform buffer, based upon the estimated distance firebrands can travel
from the front of a wildfire, this approach creates a more liberal estimation of risk zones
(California Fire Alliance 2001). This produced a drastically reduced estimate of the WUI. When
compared to the estimates of past studies (Stewart et al. 2003; Radeloff et al. 2005; Wilmer and
Aplet 2005). The study used three buffer intervals: 0.5 mi, 0.5-1.0 mi, and 1.0-2.0 mi. These
figures are based upon treatment zones for fuel reductions and proactive fire mitigation (Nowicki
2002; Hann and Strohm 2003; and Theobald and Romme 2007).
Theobald and Romme combined 2001 NLCD (30-m resolution) and US Department of
Agriculture’s (USDA) FUELMAN datasets (1km resolution (Schmidt et al. 2002)) to classify

14

vegetation types within the vicinity of the WUI. In addition, Theobald and Romme (2007) also
used 2004 Census TIGER files and road density calculations to weight housing density.
Theobald and Romme classified road density into four different types, and then removed areas
with the lowest road density classification, on the premise that these areas would not have
enough residential development to meet the housing density threshold. The study concluded the
WUI comprised 115.1 million acres and contained 12.5 million homes (Theobald and Romme
2007).
Similar to previous approaches Tully used a binary dasymetric approach to remove areas
of census blocks assumed to not contain housing units (Wilmer and Aplet 2005; Theobald and
Romme 2007); specifically, water features, public and protected lands, and low-road density
areas were removed from census blocks (Tully 2013). Housing density was then recalculated
based upon the original housing unit count and the modified block area (Tully 2013). The most
noteworthy improvement to those of previous efforts was the integration of the USGS
LANDFIRE Existing Vegetation Height (EVH) into analysis.
LANDFIRE EVH data allows for a more accurate and precise variable width buffering
that is based upon local vegetation types and their height. The zone for firefighter safety can be
calculated by creating a linear buffer four times the maximum sustained flame height (Butler and
Cohen 1998). Maximum flame height is estimated to be twice the height of existing vegetation
(Nowicki 2002; Tully 2013). Tully buffered EVH pixels by a maximum distance of eight times
the height of existing vegetation, and then rounded up to the nearest 30-meter pixel. The largest
safety zone buffer was 510 meters.
Tully then used a corridor function to create a directional buffer around census blocks
meeting the minimum housing density threshold. This allows for an estimated defensible space

15

to be created through the buffering process. This estimation can be used to help clear and reduce
wildland fire fuels within or near the WUI. Non-wildland classes were then removed from
census blocks; however, community buffers were retained in these areas (Tully 2013). Tully
concluded that the 2010 WUI covered 11.79% (227,376,491 acres) of the conterminous United
States and encapsulated 48.45% of the total housing units (Table 1).

16

Table 1: Past WUI studies & their methods
Author Stewart et al. Wilmer & Aplet Radeloff et al. Theobald & Romme Tully
Year Published 2003 2005 2005 2007 2013
Housing Density Data 2000 Census
2000 Census modified by
removing lands
2000 Census
2000 Census
modified by
removing public
lands, and distributed
using road density
from 2004 TIGER
line files
2000 Census & 2010 Census
modified by removing public lands,
low road density areas from 2004 &
2010 TIGER line files
Density Threshold 1 housing unit/ 40 acres 1 housing unit/40 acres
1 housing unit/40
acres
1 housing unit/ 40
acres
1 housing unit/ 40 acres
Vegetation NLCD 1992 NLCD 1992 NLCD 1992
NLCD 1992/
FUELMAN 2002
LANDFIRE 1.0.5 (2011)
Resolution 30 meters 30 meters 30 meters
30 meters augmented
with 1sq.km.
FUELMAN
vegetation
subcategories
30 meters
Buffer Distance 2.4 km 0.8km 2.4km 0.8-3.22km 4m-510m
Buffer Method Isotropic Isotropic Isotropic Variable Variable
Dasymetric Overlays None Public Lands None
Public Lands, Low
Road Density Areas
Public Lands, Low Road Density
Areas, National hydrography
Vegetation Classes
Included
Forests, native,
grasslands, shrubs,
wetlands, transitional
lands
Forests, shrubland,
grasslands, herbaceous
wetlands
Forests, native
grasslands, shrubs,
wetlands, transitional
lands
Forests, shrubland,
grassland, wetlands,
transitional lands
LANDFIRE Existing Vegetation
(Herb, Shrub, Forest)
Categories of WUI
Interface: more than 1
housing unit per 40
acres, have less than
50% vegetation, and are
within 2.41 km of an
area (made up of one or
more contiguous Census
blocks) over 500 ha that
is more than 75%
vegetated

One category defined as
the Community Fire
Protection zone (CFPZ)
>1 housing unit per 40
acres
Interface: more than 1
housing unit per 40
acres, having less
than 50% vegetation
and are within 2.41
km of an area( made
up of one or more
contiguous Census
blocks) over 500 ha
that is more than 75%
vegetated

Interface:>1 unit per
2.4 acre
(based upon 250
people/sq. mi)and
>10 ha patch

> 1 housing unit per 40 acres
Intermix: 1 unit per
2.4-40 acres
Intermix: 1 housing
unit per 40 acres, and
have more than 50%
vegetation
Intermix: 1 housing unit
per 40 acres, and have
more than 50%
vegetation
Acres Identified 175,124,915 n/a 177,707,043 115,173,663
215,116,195 (2000)
227,376,491(2010)
Percent of
conterminous US
9.3 n/a 9.4% 6%
11.15% (2000)
11.79% (2010)
Housing Units in
WUI
42,297,763 n/a 44,348,628 12,500,000
53,269,202 (2000)
63,408,552 (2010)
Percent of housing in
conterminous US
36.7% n/a 38.5% 13%
46.25% (2000)
48.45% (2010)

17

2.2 The Definition Effect
Although each of the studies operationalize the Federal Register’s definitions, subtle differences
in purpose, implementation, and approaches result in divergent WUI mapping outputs (Stewart et
al. 2009; Platt 2010; Tully 2013). This is more commonly referred to as “the definition effect”
(Platt 2010; Tully 2013). While studies may utilize the same datasets and housing density
threshold of one housing unit per 40 acres to operationalize WUI definitions differences in
purpose can drastically alter the final output (Stewart et al. 2009). For example, Wilmer and
Aplet (2005) and Stewart et al. (2005) both use the same US Census data to measure housing and
the same National Land Cover Data to characterize vegetation, but their variations in purpose
result in different outputs.
While Wilmer and Aplet (2005) and Stewart et al. (2005) use the same datasets and
housing density thresholds to map the WUI, their methods are geared towards different
objectives. The purpose of the Wilmer and Aplet (2005) study is to identify areas where
wildland fuels can be treated; Stewart et al.’s. (2005) focus is to determine where and in which
type of WUI housing structures are located. Both methods start by identifying areas that meet
the target housing density threshold and then characterize vegetation within or adjacent to the
blocks; however, the Wilmer and Aplet (2005) method is primarily concerned with identifying
vegetative fuels with the goal of reducing them through wildland fuel treatments (Stewart et al.
2009). Stewart et al.’s (2005) methods seek to quantify the number of housing units within the
area in order to aid policymakers in addressing wildfire management policy, in hopes of
providing protection to human settlements in hazardous areas. While these small changes in
focus may seem insignificant, Stewart et al. (2009) conclude that the Wilmer and Aplet (2005)
method results in the classification of double the spatial extent of Stewart et al.’s (2005) method.

18

Stewart et al. (2009) use the state of California as an example to illustrate the differences
in WUI mapping outputs between the methods of Wilmer and Aplet (2005) and Stewart et al.
(2005). The large discrepancy between these two models stems primarily from the buffering
techniques used to address each respective study’s purpose. Wilmer and Aplet (2005) use a
0.5mi buffer based upon the HFRA definition for treating wildland fuels outside of the CFPZ,
while Stewart et al. utilize the National Fire Plan’s (2001) 1.5 mi buffer, which is based upon the
estimated distance a firebrand can be carried by wind. Wilmer and Aplet (2005) first buffer the
CPZ and then utilize a pixel-by-pixel retention approach, which only removes individual 30-
meter pixels in nonvegetated areas after the initial buffer has already been completed.
Conversely, Stewart et al. (2005) excludes Census blocks that do not meet their operational
definition of 50% wildland vegetation. These two seemingly subtle differences result in Wilmer
and Aplet’s (2005) method having twice as much WUI for the state of California (Figure 4).

19

Figure 4: Comparison of result of Wilmer & Aplet (2005) and Stewart et al. (2005) (Source: Stewart et al. 2009)

20

Density thresholds are another critical component that often results in large discrepancies
between models. Densities also revolve around interpretation of various WUI definitions in
accordance to the focus of the study. Density can be measured for several WUI attributes such
as population, vegetation, and housing; slight changes in the distribution of any of these can
drastically alter the extent of the WUI. For example, previous studies have excluded entire
Census blocks not meeting 50% vegetation density criteria (Stewart et al. 2003, Stewart et al.
2005, Radeloff et al. 2005); this results in an underestimation of the spatial extent of the WUI.
In a similar fashion, Theobald and Romme (2007) used the Federal Register definition of 250
people per square mile instead in lieu of the housing density threshold of 1 housing unit per 40
acres for calculating the interface WUI. Furthermore, the study added the parameter that the
interface area has to comprise a minimum of 10ha (24.7 acres) to promote clustering and
eliminate isolated islands of WUI. This is problematic to many rural towns adjacent to wildland
fuels that do not meet the population density; when small communities are excluded, based upon
density parameters, it exponentially decreases the likelihood that these communities will receive
federal assistance for fuel treatments.
2.3 Summary
Discrepancies between WUI mapping outputs are largely based upon the wide variety of
definitions of the WUI. The variety in methods exists because maps vary with purpose and the
intended audience (Stewart et al. 2009). It also demonstrates the elusiveness of the WUI.
Because methods differ with purpose, no one single method can be all-encompassing when
measuring the WUI. Therefore, each mapping approach should explicitly state the problem
being addressed, the methods used, and the limitations of the datasets; this is especially true
when the results have the power to shape policy.

21

CHAPTER 3: METHODS AND DATA
This study proposes a method for measuring the spatial extent of the WUI as part of a larger
effort to assess how the WUI in Lassen County has changed from 2000 – 2015. Additionally,
the study utilizes county cadastral data to digitize housing unit location, to measure the number
of homes located in and out of the WUI. The digitized points representing housing structures
were overlaid upon the Lassen County CAL FIRE Fire Hazard Severity Zone Map in order to get
an accurate assessment of the WUI from a fire hazard standpoint.
3.1 Method Framework
The proposed method uses county cadastral data to locate residential parcels to map housing
units in their actual geospatial location. This approach allows for a more accurate and precise
estimate of the WUI. Use dwelling locations also negates the need to disaggregate Census
blocks. The following workflow will be used to assess the WUI: 1) Locate residential parcels by
running a query of assessor land use codes; 2) Create residential parcel layer from selected
features in parcel layer; 3) Convert residential parcel polygons to centroids and then locate points
to home location; 4) Buffer homes by 0.5 miles (HFRA 2003; Wilmer and Aplet 2005); 5)
Remove non-wildland fuel types (water, rock, barren, agriculture, urban land cover classes)
(Wilmer and Aplet 2005); 6) Compare extent of WUI and locations of homes to CAL FIRE
hazard maps; 7) Compare extent of WUI with land ownership types; 8) Compare extent of WUI
to fire responsibility areas.
3.1.1 Output Products
The proposed workflow will yield both feature datasets and tabular data. Together, these output
types will be used to frame the scope of the relation between human settlements and areas prone
to wildfire. Feature datasets will be used to create a series of maps that illustrate the spatial

22

extent of the WUI in 2000 and in 2015, the location of housing structures in relation to the WUI,
and the relation between WUI, residences, and fire hazard severity zones. These maps will then
be used to make a variety of comparisons, the most important of which will be to assess the
differences in using conventional WUI measurements versus the measurements of a structure-
based approach to fire hazard. The study will also quantify multiple features.
Tabular data will provide a wealth of information to local government agencies including
planning and emergency service departments. Data tables will be used to quantify the number of
acres and houses within and outside of the WUI, detail how those numbers have changed
temporally since 2000, and provide an assessment of fire risk across the county. The tabular data
will also assess housing structures hazard to wildfires, by providing numerical counts for housing
in moderate, high, and very high fire hazard areas. The generation of both graphics and tabular
data for analysis will rely heavily upon cadastral data.
3.1.2 Data Sources
Parcel data, acquired from the Lassen County Assessor Office, are integral to identifying the
location of residential parcels and housing structures within the county. The cadastral data along
with parcel ownership data was used to improve upon conventional modeling approaches for
mapping the WUI (Stewart et al. 2003; Radeloff et al. 2005; Wilmer and Aplet 2005). Together,
these two sets of data will be used to identify the location of all residential parcels and then
digitize the location of housing structures on a parcel.
The ArcGIS tools critical to achieving this workflow (see Appendix A) are “select-by-
attribute”, “select-by-location”, “join”, “buffer”, “summarize”, “reclassify”, “combine”, “zonal
geometry to table”, “extract mutli values to points”, and “feature to raster.” The join tool joins
features with features and/or tables by a shared attribute value. The assessor parcel data will be

23

joined with parcel ownership data by the assessor parcel number (APN); this unique number is
associated with only one parcel and is assigned by the tax assessor for record-keeping and
identification purposes. The “select-by-attribute” tool selects objects and/or features based upon
specific attribute values. The “select-by-location” tool features from a target layer based upon
their spatial location to features within a source layer. The “summarize” tool generates an output
summary table that provides various statistical calculations on attribute information such as
counts, averages, and minimum and maximum values. Once converted to raster format using the
feature to raster tool, raster will be reclassified. The WUI rasters will then be combined with
other rasters using the “combine” tool in order to assess the spatial relationship of WUI to other
layers. “Zonal geometry to table” will then be used to calculate the area of each raster value.
“Extract multi values to points” can then be used to quantify the number of homes in each of the
various raster values.
3.2 Data Selection
The data necessary for completing the study were acquired from multiple governmental agencies
(Table 2). The data used for this study include: NLCD land cover data (USGS 2001, 2011),
parcel data (Lassen County Assessor Office 2015), parcel ownership information table (Lassen
County Assessor office & CREST database 2015), BLM land ownership types and perennial
water bodies. With exception of the data sourced from Lassen County agencies, all data for this
study are free and easily accessible through digital downloads. All Lassen County datasets were
sourced from the Planning and Building Services Division’s drives.

24

Table 2: Datasets and Sources
Dataset Source Target Variables Spatial
Resolution
Temporal
Resolution
Lassen County
Roads
Lassen County
Public Works
Road networks
County of
Lassen
2004
NLCD
US Geological
Survey
Land cover for
vegetation
identification
County of
Lassen
2001
2011
Lassen County
Assessor Parcels
Lassen County
Assessor Office
Residential Parcels
County of
Lassen
June 2015
Parcel Ownership
Information
Lassen County
Assessor Office &
Crest Database
Assessor land use
codes (table will be
joined with parcel
layer)
N/A June 2015
World Imagery esri
Will be used to hand
digitize parcels
Global Compilation
Fire Hazard Severity
Zones

CAL FIRE Severity Zones
County of
Lassen
2007
Fire Responsibility
Areas

CAL FIRE
Fire Responsibility
Areas
County of
Lassen
2008
Land Ownership
Types

BLM
Will be used to
identify the various
agencies and
governments that
comprise the
county
California 2012

3.2.1 Residential Parcels
This study proposes an improvement to past WUI mapping approaches by utilizing cadastral
parcel data to temporally assess the change in the spatial extent of the WUI from 2000-2015.
This study uses Wilmer & Aplet’s (2005) methods for identifying the CFPZ as a means for
mapping the WUI. A dasymetric approach will be used to isolate all residential parcels in order
to eliminate the areas where housing does not exist (Wilmer & Aplet 2005; Theobald and
Romme 2007; Tully 2013). This improves upon conventional WUI mapping methods by
identifying the exact location of residential parcels, instead of redistributing housing units across
refined Census blocks by removing land where housing is assumed not to exist.

25

Once the CREST parcel ownership table is joined with the assessor parcel layer, using
the APN as the join field, a query can be conducted utilizing the “select-by-attribute” tool. A
query expression was built which identifies all assessor use codes to select all residential parcels.
Use codes help the Assessor to describe the different land uses occurring on a parcel for tax
purposes and land value assessments (Table 3). The selected parcels were used to create a new
layer, “Residential Parcels”.

Table 3: Lassen County Assessor Use Code Index

26

After creating the residential parcel layer, housing was digitized for each parcel using
aerial imagery and the assessor database. Housing density was not factored into this study for
two reasons: 1) the distribution of housing was known 2) this study uses the definition that WUI
occurs where development is intermingled or adjacent to wildland fuels. The need to utilize
Census block layers is negated entirely by creating residential parcel and housing layers. The
majority of past studies were conducted at the national level, a complete dataset of the
conterminous US were necessary; this is why the US Census blocks were chosen. Census blocks
are comprised of arbitrary boundaries and increase the modified aerial unit problem (MAUP).
MAUP makes it appear as if housing is evenly distributed across the Census blocks, while a
cadastral-based approach gives the exact parcel location.
3.2.2 Housing Structures as Points
The US Census does not account for all housing structures within Lassen County, this study
generated its own count by digitizing housing structures as points. The residential parcels will be
converted to centroids using the “feature-to-point” tool. Centroids were then digitized, by
moving the centroids over the residence location using aerial imagery (see Figure 5, 6, and 7).

27

Figure 5: Lassen County Assessor Parcels (2015).

28

Figure 6: Residential parcels converted to centroids.

29

Figure 7: Digitization of houses as points. Centroids moved to location of residence.

The assessor use codes provide the number of residences per parcel. Parcels with more than one
housing structure have multiple points, representative of each residence. There are a few
limitations associated with this approach.
Aside from the amount of time it takes to hand digitize all of the housing, there are a few
issues and limitations that must be addressed. This study only digitizes housing units, because
the assessor does not keep a numerical count of structures on the property; this will also
significantly reduce the time digitizing. When the residence cannot be delineated between other
structures, the point was placed in between the structures in question. When tree cover is too
thick to identify the residence location the point remained a centroid. The final key issue is that

30

the Lassen County Assessor’s Office does not keep annual legacy files for their parcel database
or GIS system.
The mapping of residential parcels in the year 2000 requires slightly more effort since the
Assessor’s Office does not keep legacy files for each year. The Lassen County Planning and
Building Services Department keeps a record of permits for all new residences constructed or
placed within the county. This information is organized by a parcel’s APN. The record also
shows whether a previous residence was replaced by a new one. A report of all homes built or
placed from January 2000 to June 2015 was generated in order to remove residential parcels
created after 2000.
All APNs, listed in the generated permit report, were then selected using the “select-by-
attribute” tool. The selected features were then copied to create a new layer that shows the
growth of residential parcels in the county. The 2015 residential layer was then copied and
duplicated; the same selected features were removed from one of the 2015 layers, which yielded
all residential parcels as of January 1, 2000. This is beneficial to the study as it allows the same
parcel-based approach to calculate a historic change in WUI from 2000-2015. It also shows the
number of homes constructed within the WUI in the same timeframe.
3.2.3 Calculating the WUI
All homes were buffered by 0.5 mi (HFRA 2003; Wilmer & Aplet 2005) (see Figures 7 & 8).
Many other studies use a 1.5 mi buffer that is based upon the estimated distance a firebrand can
travel, which can result in ignition; however, this tends to overestimate the WUI. The 1.5 mi
buffer should be used in areas of steep topography and thick vegetation, both of which have the
potential to create particularly hazardous conditions (HFRA 2003). These buffers do not endorse
the clearing of a 0.5 mi area around all residential parcels and communities; rather, they denote

31

the areas where the community and governmental agencies should focus their attention in order
to create a defensible space and proactively reduce wildfire risk (Wilmer & Aplet 2005). After
the 0.5 mi buffers have been rendered for each residential parcel layer, they were converted to
30-meter raster layers and non-wildland vegetation types were removed.

32

Figure 8: 2000 housing stock buffer (0.5mi) in yellow. 2000 WUI in red (after removal of non-wildland fuels)

33

Figure 9: 2015 housing stock buffer (0.5 mi) in yellow. 2015 WUI in red (after removal of non-wildland fuels)

34

3.2.4 Removal of Non-Wildland Fuel Types
The 2001 and 2011 USGS’s NLCD land cover datasets were used to examine vegetation cover
within the study area. The classes of vegetation to be categorized as non-wildland fuel types are
as follows:
Class 11 (open water), 12 (perennial ice/snow), 22 (developed, low intensity), 23
(developed, medium intensity), 24 (developed, high intensity), 31 (barren land:
rock/sand/clay), 81 (pasture/hay), and 82 (cultivated crops) (Radeloff et al. 2005; Wilmer
& Aplet 2005).
Vegetation types categorized as wildland fuels will be weighted a “1” and non-wildland fuel
types will be weighted a “0”. Non-wildland classes were then removed (Wilmer & Aplet 2005;
Tully 2013).
3.3 Wildfire Hazard Assessment
Once the WUI was defined and non-wildland fuel classes had been removed, CAL FIRE’s
Lassen County Fire Hazard Severity Zone map was overlaid. The map has three zone types:
moderate, high, and very high hazard severity. These were weighted 1, 2, and 3 respectively.
The number of homes within these zones was then quantified. Percentage of homes, private
versus public lands, and the number of homes in and out of WUI were all calculated for each
zone using the “summarize” and the “extract multi values to points” tools. This allowed for a
number of comparisons to be made between a conventional WUI measurement and hazard based
measurement.

35

3.4 Software and Hardware Used
-esri ArcGIS 10.3.1
-VMWare Horizon Client
-Apple iMac
-2.0 GHz Intel Core i5
-16 GB RAM 1600 MHz DDR3
-OS X Yosemite, Version 10.10.5
-Apple MacBook Pro
-2.9 GHz Intel Core i7
-8GB RAM 1600 MHz DDR3
-OS X Yosemite, Version 10.10.5
-hp desktop
-Intel Core i3-3220 CPU at 3.30 GHz
-4GB RAM
-Windows 7 Enterprise 32-bit Operating System with Service Pack 1 installed

36

CHAPTER 4: RESULTS
The goal of this study was to develop a cadastral-based measurement technique for locating and
analyzing spatial change in the WUI. The identification and location of residential parcels
throughout the county is integral to gathering the precise location of the WUI and is a major
improvement over previous studies as it eliminates the need to estimate where housing exists.
Digitization of housing structures is another logical improvement as it not only provides the
location of the structure(s), but also enables multiple opportunities for analysis through location
and quantification.
This chapter reviews the lessons learned from the utilizing a cadastral-based approach for
measuring the WUI. The results include the spatial extent of the WUI in the years 2000 and
2015. The 2000 and 2015 WUIs and housing stocks were then combined with other layers (land
ownership types, fire responsibility areas, and CAL FIRE fire hazard severity zones) in order to
analyze their relationships and measure change. Maps were then created to illustrate the spatial
location of the WUI in comparison to these layers.
4.1 Residential Parcel Identification
As mentioned in Chapter 3, Lassen County Assessor land use codes were used to identify
residential parcels within the study area. A total of 10,195 residential parcels were identified.
Together, with the assessor database and aerial imagery, homes were then digitized as points.
There were 11,019 homes digitized; these homes comprise the 2015 housing stock for the study.
A list of all building permits issued for the construction or placement of new homes, by the
Lassen County Planning & Building Service Department, from January 1, 2001 to June 2015 was
generated. These homes and/or parcels were removed from the 2015 housing stock using the
assessor parcel number (APN). The new selection was then used to create a 2000 residential

37

parcel layer and 2000 housing stock layer. There were 9,360 residential parcels and 10,120
homes. The two housing stock layers were the foundation for determining the WUI.
4.2 Analyzing the WUI
Once all of the homes were located and digitized, they were buffered by .5 mi, based upon the
HFRA recommendation (HFRA 2003). The buffers were then dissolved to merge overlapping
geometry from the buffers created for each home. The buffer layers were then converted to 30-
meter rasters and combined with the reclassified land cover rasters to remove the buffered areas
not containing wildland fuels, using the “Combine” tool (Wilmer & Aplet 2005). The “Zonal
Geometry to Table” tool was then used to calculate the spatial extent of the 2000 and 2015
WUIs.
The spatial extent of the 2000 WUI was 961.08 sq. km. (371.08 sq. mi.) and accounted
for 96,108.01 hectares (237,488.06 acres) or 7.86% of the study area. Unsurprisingly, the WUI
grew proportionately with the number of new homes placed in the county over 15 years. The
spatial extent of the 2015 WUI was 1,016.67 sq. km. (392.54 sq. mi) and accounted for
101,666.68 hectares (251,223.83 acres) or 8.32% of the study area. The spatial extent of the
WUI grew by 0.057 %, 55.59 sq. km. (21.46 sq. mi.), or 5,558.68 hectares (13,735.77 acres)
from 2000-2015. Most of this growth can be attributed to the 899 homes placed in low density
regions of the county. Most of the growth was seen in the northern region and in the satellite
communities in close proximity to Susanville (Figure 10).

38

Figure 10: Change in WUI 2000 -- 2015

39

4.3 Land Ownership
It is important to differentiate between the types of land ownership where the WUI occurs in
order to identify who is responsible for proactively treating fuels within the area. Although the
majority of WUI is typically located on private property, it can also extend into public lands.
WUI on public lands is usually attributed to the buffers extending into these property types, but
can also result from private housing on or surrounded by public lands. Lassen County fits this
typical WUI profile.
In 2000, 78,503.43 hectares (193,985.94 acres) or 81.70% of the WUI was located on
private lands, 10,833.79 hectares (26.770.85 acres) or 11.27% on lands managed by the Bureau
of Land Management (BLM), 3,789.43 hectares (9,363.88 acres) or 3.94% on U.S. Forest
Service lands, and the remaining 2961.24 hectares (7317.38 acres) or 3.09% spread over county,
native, National Park Service (NPS), and federal military lands. These numbers changed
minimally over the 15 year time frame for the study. In 2015, 82,504.39 hectares (203,872.51
acres) 81.17% of the WUI was located on private lands, 12,096.76 hectares (29,891.69 acres)
11.90% on BLM lands, 3,974.38 hectares (9,820.90 acres) 3.91% on Forest Service lands, and
the remaining 3,069.84 hectares (7,585.74 acres) or 3.02% distributed through county, native,
NPS, and federal military lands (see Table 4 and Figure 11).

40

Table 4: Changes in WUI land ownership acreage 2000- 2015
Land Ownership
WUI
Acreage
2000
% of total
acreage
WUI
Acreage
2015
% of
total
acreage
Change
from
2000-2015
(acres)
Percent
Change
Private 193,985.94 81.70% 203,872.51 81.27% 9,886.57 5.10%
County 231.27 0.10% 243.88 0.10% 12.61 5.45%
BLM 26,770.85 11.28% 29,891.69 11.90% 3120.84 11.66%
U.S. Forest Service 9,363.88 3.94% 9,820.90 3.91% 457.02 4.88%
National Park
Service
775.93 0.33% 770.37 0.31% -5.56 -0.72%
State 4,197.46 1.76% 4,389.38 1.75% 191.92 4.57%
Native 797.93 0.34% 788.81 0.31% -9.12 -1.14%
Military 1,314.79 0.55% 1,393.30 0.55% 78.51 5.97%

41

Figure 11: 2015 WUI and land ownership

42

4.4 Residential Structures
Although digitization of the housing as points is labor intensive, the housing stock datasets allow
for multiple analyses to be conducted. Specifically, the study was able to quantify the number of
housing structures in relationship to the 2000 & 2015 WUIs, designated fire agency
responsibility areas, and CAL FIRE fire hazard severity zones (CAL FIRE 2007). The “Extract
Multi Values to Points” tool was used to quantify the number of houses in each of the various
layers. In 2000, 4,819 residential structures (47.62%) were located within the WUI. Out of the
899 new homes placed within the county from 2000-2015, 637 homes (70.86%) were placed in
existing WUI or near wildland fuels, which resulted in the expansion of the WUI. The total
number of residential structures located within the WUI, in 2015, was 5,456 (49.51%), which
resulted in an increase of 13.22% from 2000.
4.4.1 Fire Hazard Severity Zones (FHSZ)
The study used a method which focuses on identifying wildland fuels in close proximity to
housing in order to prioritize treatment in these areas (Wilmer & Aplet 2005; Stewart et al.
2009). The model does not calculate fire hazard however, which makes it difficult to determine
which communities are in most need of fuel treatment and resource allocations. In 2007, CAL
FIRE calculated fire hazard severity zones for the entire state. Their models factor in numerous
variables such as fire history, existing and potential fuels, flame length, firebrands, topography,
and average weather patterns by region, in order to provide an independent estimate of fire
hazard for each county (CAL FIRE 2007). The integration of this data supplements the study by
quantifying the number of homes in each zone in 2000 and 2015. This data also identifies which
hazard zones the communities and the WUI are located in.

43

In 2000, 2,024 homes were located in a “Very High” hazard zone, and 1,569 of these
structures were also found in the WUI (77.52%). 1,837 residential structures were located in
“High” zones, 726 (39.52%) of which were in the WUI. 3,327 were located in zones classified
as “Moderate”, 1,950 (58.61%) of which were in the WUI. The number of homes in each of
these zone classifications increased through 2015 (Table 5). In 2015, there were 2,299 homes in
“Very High” zones, 1,984 homes in “High” zones, and 3,714 in “Moderate” zones. The number
located in a WUI also increased in each of these three zones (Figures 12).

Table 5: Number of homes within each FHSZ from 2000-2015
Fire Hazard
Severity Zones
# of Homes
in WUI
(2000)/ % of
total homes
per zone
Total # of
Homes
(2000)
# of Homes
in WUI
(2015)/ %
of total
homes per
zone
Total # of
Homes in
WUI (2015)
%
increase
in WUI
%
increase
total
Very High
1,569
(77.52%)
2,024
1,786
(77.69%)
2,299 13.83% 13.59%
High
726
(39.52%)
1,837
829
(41.78%)
1,984 14.19% 8.00%
Moderate
1,950
(58.61%)
3,327
2,245
(60.45%)
3,714 15.13% 11.63%
Non-wildland
/Non-urban
435
(48.49%)
897
460
(46.61%)
987 5.75% 10.03%
Urban/Unzoned
139
(6.83%)
2,035
136
(6.68%)
2,035 -0.02% 0.00%
TOTAL
4,819
(47.62%)
10,120
5,456
(49.51%)
11,019 13.21% 8.88%

44

Figure 12: Comparison of the number of homes in the 2000 and 2015 WUIs and their relation to FHSZ

The overwhelming majority of the lands in the county are classified as “moderate” (see Figures
13 and 14). This category also has the most homes in 2000 and 2015. Lands and homes
classified as “very high” are the second largest FHSZ type. The “high” class is the third largest
group for both lands and homes in both case study years. While the WUI only accounts for a
small portion of the county, 36% of it lies within areas classified at either “very high” or “high”.
These areas should be prioritized for proactive fire prevention activities.

0
500
1000
1500
2000
2500
FHSZ Housing 2000
FHSZ Housing 2015

45

Figure 13: Total 2015 WUI acreage as it relates to FHSZ
60,944.77
28,416.01
143,793.39
17,731.51
355.39
-
20,000.00
40,000.00
60,000.00
80,000.00
100,000.00
120,000.00
140,000.00
160,000.00

46

Figure 14: 2015 WUI overlaid on CAL FIRE FHSZ data

47

4.4.2 Fire Responsibility Areas
Fire responsibility areas indicate which government agency is responsible for providing fire
protection, for maintaining and treating fuels in the area, and also who is financially responsible
for a fire. In California, if a home is located within a state responsibility area (SRA) then that
home owner must pay an annual “SRA Fire Prevention Fee” for each habitable structure on their
property. These costs help pay for fuel reduction activities, FHSZ mapping, fire prevention
education, and a variety of other fire prevention activities (CAL FIRE 2012). Some local fire
districts also impose one time fire fees based upon the square footage of a new home in addition
to annual fire prevention fee.
In 2000, 4,235 homes were located in SRAs, 5,823 in local responsibility areas (LRAs),
and 62 in federal responsibility areas (FRAs). The number of homes in all three of these
responsibility areas grew through 2015. In 2015, 4,955 homes were located in SRAs, 5,997 in
LRAs, and 67 in FRAs. While FRAs comprise the largest responsibility type spatially, 70% of
the WUI is in SRAs (Figure 15). Because few homes are located on federal lands, it is not
surprising that the bulk of the WUI is in state responsibility areas. The 0.5 mile isotropic buffer
used for mapping the WUI is rather conservative in comparison to those of past endeavors. If the
buffer was changed to either one or one and a half miles the WUI would extend further into
federal lands, changing the SRA and land ownership percentages (Figure 16).

48

Figure 15: 2015 WUI acreage by responsibility area
42,376.83
33,477.27
175,354.73
30.91
-
20,000.00
40,000.00
60,000.00
80,000.00
100,000.00
120,000.00
140,000.00
160,000.00
180,000.00
200,000.00
FRA LRA SRA Undefined

49

Figure 16: 2015 WUI overlaid over fire responsibility areas

50

CHAPTER 5: DISCUSSION AND CONCLUSIONS
The main objective of this study was to create a cadastral-based approach to map the 2000 &
2015 WUIs in Lassen County, California. The additions of digitized housing stock, FHSZ,
responsibility areas, and land owner information allowed a variety of comparisons and change
analyses to be conducted. Government agencies, property owners, and emergency response
crews can all benefit from understanding more about the relationships the WUI has with these
variables and how they have evolved over the past 15 years. This final chapter provides a
summary of the study, a discussion of its results, their limitations, and suggestions and
implications for future research.
5.1 Benefits of the Cadastral-Based Approach
The main advantage to using a cadastral-based approach is that it increases the levels of
precision and accuracy for home location and the WUI. The utilization of local cadastral data
negates the need to estimate, through dasymetric techniques, where housing is located. Previous
studies, most of which were at the national scale, used Census blocks in combination with other
datasets to remove the areas where housing was not likely to exist (i.e. public lands, water
features) (Radeloff et al. 2005; Wilmer & Aplet 2005; Theobald & Romme 2007); however, this
study used Lassen County Assessor parcel data to isolate all residential parcels within the study
area. Esri World Imagery was then used to digitize housing structures. The use of assessor
parcel information and aerial imagery yield a more accurate rendering of the WUI. Another
major advantage is the ability to quantify the number of homes in the WUI, in FHSZs, and in
responsibility areas.
Although the digitization of homes as points was extremely time consuming, it allowed
for the number of homes in the WUI and other data layers to be quantified. The digitization only

51

provides government agencies with statistical figures on the number of homes in areas of “very
high” and “high” severity zones, but also provides emergency response crews the exact locations
of homes in the area. Understanding the relationship between residential development, the WUI,
and wildfire hazard zones is integral to proactive wildfire management. It not only helps to
pinpoint where development occurs. and therefore where fire has the potential to cause human
fatalities and destroy infrastructure and resources, but also identifies the areas where wildland
fires could be permissible and actually help to restore the environment and natural fire regimes
(Tully 2013). Finally, identifying the location of homes in the WUI and highest severity zones
can help to prioritize state and federal assistance for preventive fire activities and fuel treatments.
5.2 Disadvantages and Limitations of the Cadastral-Based Approach
Although the utilization of Assessor parcel data offers multiple improvements to the endeavors
of past studies, it also comes with a few disadvantages and many limitations. The biggest
disadvantage was the time it took to create the two housing stock layers. While it only took an
hour to create the 2000 and 2015 residential parcel layers, it took nearly 75 hours of digitization
to create the housing stocks. Another issue is accessibility to data. While the general public
might be able to request a copy of the parcel layer for the county, the Crest database that contains
the use code is for official county use only and would not be made accessible to the public;
however, this information measures could be taken to keep the sensitive data separate from the
land use code.
5.2.1 Limitations of Housing Data
The assessor data used was from June 2015, as they only update their GIS database once a year,
so this would not represent all housing built through the end of 2015. Another problem that
arose was the discrepancies between the Assessor’s office and the County Planning and Building

52

Department. When Planning and Building conducted a query of all new manufactured and
constructed homes, from January 2001- June 2015, 63 of the residential parcels within the report,
were not being assessed with a residential use code and therefore were not factored into the
study. This number is insignificant as it equates to 0.01% of the 2015 housing stock. The City
of Susanville was unable to generate a similar report with APNs, so new residential parcels built
from 2000-2015 could not be determined for the City of Susanville; however, the Municipal
Planning Office was able to state that 250 homes had been constructed or placed since 2000.
Again, this number is insignificant as it equates to 0.02% of the 2015 housing stock.
Another limiting factor was that some parcels had more than one home. Since the
method used to create the 2000 housing stock layer was to subtract homes and parcels listed in
generated permit report by the county, the study was unable to determine which home came first
from the aerial imagery used. This would not have a significant impact on the study as it would
typically only affect the 2000 housing layer and then would only shift the 2000 WUI a matter of
feet. The decision to represent housing as points likely provides a much smaller WUI estimate.
The representation of housing structures as points results in a conservative estimate of the
WUI. The choice to represent housing as points saved an immense amount of time, as digitizing
housing footprints as polylines would take a few months; however, buffering a small point
results in much smaller WUI than buffering a 1,000 sq. ft. home. Another major limitation is
that only housing structures were buffered, therefore accessory structures, commercial, and
industrial buildings were not included. Although these buildings would be considered
development, they were not included because of the time it would take to digitize all buildings
within the county. Additionally, the assessor’s office does not keep a record of the number of
structures on a parcel within their GIS; they only include housing counts within their database.

53

As previously mentioned, housing structures were represented, and the attribute
information does not contain the number of units within each structure. The 2014 Census
QuickFacts for Lassen County had a count of 12,741 housing units for the county, 13.4% of
these units were found in multi-unit structures. 1,707 housing units were located in multi-unit
structures; when this number is subtracted from the total count it leaves 11,034 units, which is
just 15 more units than the 2015 housing stock created for this study. These calculations
validated that the accuracy and validity of the housing stock datasets. The number of seasonal
homes in the county was unavailable.
A decision to not use community density calculations was made as the study used
cadastral data instead of Census blocks. Many of the prior WUI studies were at the national
scale and relied upon Census blocks. In order to more accurately estimate housing location they
used dasymetric techniques to disaggregate the blocks. Target housing densities were then used
in order to prevent isolated pockets of WUI across the nation. However, since this studies focus
was more oriented towards wildfire hazards in relation to housing it did not use a target housing
density. Therefore, any housing located in or next to wildland fuels was considered a part of the
WUI. The inclusion of the isolated areas provides a more accurate estimate of the actual WUI
than taking a community based approach that excludes isolated rural housing. It also allowed
for an analysis of the WUI to multiple other layers to be conducted. Housing counts within each
of the other layers were then quantified.
5.2.2 Limitations of Other Data and GIS Tools
The “Zonal Geometry as Table” tool resulted in slight discrepancies in figures between various
datasets, despite using the raster snap function and all raster resolutions being 30-meter. For
example, the tool yielded a calculation of 1016.67 sq. km. for 2015 WUI layer, but an area of

54

1016.45 sq. km. for land ownership with WUI. The tool was reran and yielded slightly different
calculations each time. This is attributed to resampling error, as all the rasters have a 30-meter
resolution (esri, 2011).
The CAL FIRE FHSZ and responsibility area data were published in 2007. Further the
CAL FIRE FHSZ data is an assessment of hazard, not risk. “Hazard” is based upon physical
conditions and the likelihood an area will burn over a given time period, and does not account for
fuel treatment activities. Fire “risk” is defined as the susceptibility of the development to fire. It
accounts for all fuel treatments that have been conducted in the area, fire infrastructure systems,
defensible space, and fire resistant building materials (CAL FIRE 2007). Therefore, fire risk is
not factored into this study. All other limitations within the study revolve around the age of the
data used.
The structural based approach used in this study allowed for quantification of the number
of homes located within and out of the WUI. It also allowed for a quantification of the number
of homes located within each of the fire hazard severity zones. It was interesting to see the
differences between the number of homes in both the WUI and FHSZ layers, and the number for
each FHSZ but not within the WUI. In 2015, 77.69% of the homes located in a “very high”
FHSZ were also in the WUI. Only 41.78% of the homes located in a “high” FHSZ were also in
the WUI. This shows that the WUI is not an accurate assessment of wildfire hazard to homes; in
California, a combination of CAL FIRE’s expertise (FHSZ data) and a WUI assessment should
be utilized to give a more meaningful approach. In locations outside of California, it is critical to
factor in other agencies’, with an expertise in regards to fire, data and resources.
The NLCD data used for the two WUIs were from 2001 and 2011. It is unlikely that land
cover changed drastically between 2011 and 2015 within the study area. The layer for land

55

ownership type was from 2012, and did not include county lands. County lands had to be
queried from the assessor parcel data, converted to raster, combined, and then new figures were
calculated.
5.3 Future Research
The study could be easily replicated by other agencies, as long as they had access to parcel data
with the number of homes on any given parcel within the attribute table. While the study offered
logical improvements using a cadastral-based approach for mapping the WUI, it could still be
further improved. Automated structure digitization could help reduce the time digitizing
housing. Automated digitization would be especially useful to counties with a much larger
housing stock. Automated digitization also has the potential to yield structural footprint
geometry, which would give a more accurate WUI assessment; most importantly, it could
digitize all structures within a study area, and could therefore create a more accurate estimate of
existing development and change.
5.3.1 Existing Vegetation
Another possible improvement would be to incorporate existing vegetation data into the
assessment. For this study NLCD land cover data was used in order to keep consistency;
however, the integration of LANDFIRE data could also improve the study. LANDFIRE existing
vegetation type (2012) is more accurate as it offers actual vegetation on the ground, while NLCD
land cover data averages a percent coverage. Datasets within the LANDFIRE database, that
would be useful, include: existing vegetation types, existing vegetation cover, and existing
vegetation height. These datasets could be used to give estimates of potential flame height, burn
rates, and help fire agencies calculate the distance of vegetation to treat. The use of LANDFIRE
datasets would inevitably yield a more accurate assessment of fire hazard. The integration of

56

social variables such as population, demographic data, and land values could also supplement the
study.
5.3.2 Home Ignition Zone
Cohen (1998; 2001) defines the home ignition zone as the area 100-150 feet around the
residence. Fuel treatments in this zone will reduce a home’s probability of ignition from radiant
heat (Cohen & Butler 1998; Cohen 2001; Platt 2010). This zone could be calculated using a
Euclidean distance buffer approach. All homes located within the WUI could then be buffered
by 150 feet. The rendering of the home ignition zone would delineate the area for defensible
space immediately around a residence. The area beyond the home ignition zone is referred to as
the “Community Protection Zone” and is captured by using the HFRA recommended 0.5 mile
buffer (Wilmer & Aplet 2005; Theobald and Romme 2007; Platt 2010).
5.3.3 Social Variables
The integration of population, demographic data, and land value figures all have the potential to
substantially supplement this type of study. Population and demographic data could help analyze
the distribution of people living in and out of the WUI. It could also help to assess which houses
are occupied year round and which are seasonal. Land value figures could reveal if disparities
exist that keep certain income groups from living in urban areas. They could also help to
estimate the potential costs of damage if a wildfire were to hit a community.
5.4 Conclusion
The main motivation for the study was to measure the change in WUI from 2000 – 2015,
so that local governmental agencies and communities could be made aware of the growing
hazard posed by wildfire in the WUI. By making the hazards known, communities can begin to

57

address these issues; the best way to do this is through a community wildfire protection plan
(CWPP). In order to be eligible to receive federal assistance for fuel treatment activities,
communities must have an adopted CWPP. The Lassen County Fire Safe Council is a non-profit
corporation made up of various local, state, and federal agencies, as well as other institutions.
They have helped facilitate the CWPP process for the majority of the county. Together, with the
partnering agencies, they prioritize high priority fuel reduction projects throughout the county.
Upon completion, this study will be presented on behalf of the County of Lassen Planning
Division at a Fire Safe Council meeting. Additionally, other components of this study will also
live outside of academia.
The 2015 housing stock was distributed to emergency service agencies, so that they can
quickly identify home location; this is especially useful when trying to determine home location
on larger parcels. The County Planning and Building Department will also be able to maintain
and use the housing and residential parcel layers for the Housing Element of the County General
Plan and for other long range analyses.
The WUI measurement provided by this study is a rather conservative estimate as it uses
the HFRA 0.5 mi buffer. Many other studies use a 1.5 mi buffer that is based upon the estimated
distance a firebrand can travel, which can result in ignition; however, this tends to overestimate
the WUI. The 1.5 mi buffer should be used in areas of steep topography and thick vegetation,
both of which have the potential to create particularly hazardous conditions (HFRA 2003).
Again, it is important to emphasize that the 0.5 mi buffer is not a recommendation to clear a
distance of a 0.5 mi around all housing in the WUI; rather, it should serve as the focal area for
communities to assess wildfire hazard and risks, infrastructural needs for combatting wildfires,
and wildfire fuel reduction activities, all in order to improve public safety.

58

References
Benfield, F.K. 2001. “Once There Were Greenfields: How Urban Sprawl is Undermining
America’s Environment, Economy, and Social Fabric.” New York: Natural Resources
Defense Council.
Bhandary, Uddab. 2007. “Vulnerability to Natural Hazards: A Study of Wildfire-Burned
Subdivisions in the Wildland-Urban Interface Using IKONOS Imager and GIS Data.”
University of Colorado at Denver, 327-495.
Butler, B.W. and J.D Cohen. 1998. “Firefighter Safety Zones: How Big is Big Enough?” Fire
Management Notes 58 (1): 13–16.
California Department of Forestry and Fire Protection (CAL FIRE). 2012. “Fire Hazard Severity
Zone Development.”, accessed October 12, 2015,
http://www.fire.ca.gov/fire_prevention/fhsz_maps_lassen.
California Fire Alliance. 2001. “Characterizing the Fire Threat to Wildland-Urban Interface
Areas in California.” accessed September 29, 2015,
http://frap.fire.ca.gov/projects/wui/525_CA_wui_analysis.pdf.
California Senate. 2012. “Senate Bill No. 1241.”
Cohen, J.D. 2001. “Wildland-Urban Fire – A Different Approach.”, accessed October 21, 2015,
http://extension.oregonstate.edu/sorec/sites/default/files/urban_wildfire_diff_approach.pd
f.
County of Lassen. 2012. “Lassen County General Plan: Land Use Element.”
Esri. 2011. “How Zonal Geometry Works.” ArcGIS Desktop 9.3 Help, accessed December 29,
2015,http://webhelp.esri.com/arcgisdesktop/9.3/index.cfm?TopicName=How%20Zonal%
20Geometry%20works.
Executive Department. 2015. “Executive Order B-29-15.” State of California.
Gencer, Ebru A. 2013. “Natural Disasters, Urban Vulnerability, and Risk Management: A
Theoretical Overview.” SpringerBriefs in Environment, Security, Development and
Peace 7: 7-43.
Hammer, Roger, Susan S. Stewart, and Volker Radeloff. 2009. “Demographic Trends, the
Wildland-Urban Interface, and Wildfire Management.” Society & Natural Resources 22
(8): 777-782.

59

Hann, W.J. and D.J. Strohm. 2003. “Fire Regime Condition Class and Associated Data for Fire
and Fuels Planning: Methods and Applications.” Proceedings of the Conference on Fire,
Fuel Treatments, and Ecological Restoration: Proper Place, Appropriate Time, Colorado
State University, April 2002. USDA Forest Service Proceedings RMRS-P-29, 397-433.,
accessed October 1, 2015,
http://www.fs.fed.us/rm/pubs/rmrs_p029/rmrs_p029_397_434.pdf.
Haight, R.G., T. David Cleland, B. Roger Hammer, B. Volker Radeloff, T. Scott Rupp.
“Assessing Fire Risk in the Wildland-Urban Interface.” Journal of Forestry 102 (7):41-
48.
Liu, Z., C. Michael Wimberly, Aashi Lamsal, L. Terry Sohl, J. Todd Hawbaker. 2015. “Climate
Change and Wildfire Risk in an Expanding Wildland Urban Interface: A Case Study
from the Colorado Range Corridor.” Landscape Ecology, 30 (10): 1-15.

Massada et al. 2009. “Wildfire Risk in the Wildland–Urban Interface: A Simulation Study in
Northwestern Wisconsin.” Forest Ecology and Management 258 (9): 1990-1999.

Miller et al. 2003. “Re-Introducing Fire at the Urban/Wild-land Interface: Planning for Success.”
Forestry 76 (2):253-260.

Martinuzzi, Sebastian., I. Susan Stewart, P. David Helmers, H. Miranda Mockrin, B. Roger
Hammer, C. Volker, Radeloff. 2015. “The 2010 Wildland-Urban Interface of the
Conterminous United States.”, accessed October 15, 2015,
http://www.nrs.fs.fed.us/pubs/48642.

National Park Service. “Wildfire Causes.”, accessed October 15, 2015,
http://www.nps.gov/fire/wildland-fire/learning-center/fire-in-depth/wildfire-causes.cfm.

Nowicki, B. 2002. “The Community Protection Zone: Defending Houses and Communities from
the Threat of Forest Fire.” Center for Biological Diversity.

Platt, J.V. 2010. “The Wildland-Urban Interface: Evaluating the Definition Effect.” Journal of
Forestry 103 (8): 383-389.

Radeloff, V. C. 2005. “The Wildland-Urban Interface in the United States.” Ecological
Applications 15 (3): 799-805.
Rozmajzl, Mary, “Delineating the Wildland Urban Interface Using Publicly Available Geospatial
Data.” Master’s thesis, University of Nebraska at Omaha, 2012.
Schmidt, K.M., J.P. Menakis, C.C. Hardy, W.J. Hann, and D.L. Bunnell. 2002. “Development
of Coarse-Scale Spatial Data for Wildland Fire and Fuel Management.” General
Technical Report RMRS-GTR-87. US Department of Agriculture, Forest Service, Rocky
Mountain Research Station, Fort Collins, CO.

60

Stewart S., V.C. Radeloff , and R.B. Hammer. 2003. “Characteristics and Location of the
Wildland- Urban Interface in the United States.”, accessed October 13, 2015,
http://www.nrs.fs.fed.us/pubs/gtr/gtr_nrs1/stewart_1_197.pdf.
Theobald, David, and H. William Romme. 2007. “Expansion of the US Wildland–Urban
Interface.” Landscape and Urban Planning 83 (4): 340-354.

Thomas, Douglas S., and T. David Butry. 2014. “Areas of the U.S. Wildland–Urban Interface
Threatened by Wildfire during the 2001–2010 Decade.” Natural Hazards 71 (3): 1561-
1585.
Tully, Jacob P, "The Wildland-Urban Interface in the Conterminous United States 2000-2010."
Master’s thesis, Western Washington University, 2013.
US Congress. 2003. “H.R. 1904: Healthy Forest Restoration Act of 2003.”
US Federal Register. 2001. “Urban Wildland Interface Communities within the Vicinity of
Federal Lands that are at High Risk from Wildfire.” National Archives and Records
Administration.
US Forest Service. 2013. “Wildfire, Wildlands, and People: Understanding and Preparing for
Wildfire in the Wildland-Urban Interface.” US Department of Agriculture.
US Forest Service. 2014. “The Rising Cost of Wildfire Operations: Effects on the Forest
Service’s Non-Fire Work.” US Department of Agriculture.
Whitman, Ellen, Eric Rapaport, and Kate Sherren. 2013. “Modeling Fire Susceptibility to
Delineate Wildland–Urban Interface for Municipal-Scale Fire Risk Management.”
Environmental Management 52 (6): 1427-1439.
Wilmer B., G. Aplet. 2005. “Targeting the Community Fire Planning Zone: Mapping Matters.”
The Wilderness Society, Washington, D.C.

61

Appendix A: Flow Charts for GIS Work

62

63

Abstract (if available)

Abstract The expansion of low-density residential development into wildlands places a variety of stresses on Earth’s natural systems. Urban sprawl is one of the many factors contributing to water shortages, higher resource costs, and increasingly destructive storms, all of which fit under the umbrella of global climate change (Gencer 2013). In California, it is wildfires that capture the headlines during the summer and fall months. With wildfires becoming a common occurrence within the California landscape, it is crucial that local government and agencies work to create proactive approaches to mitigate the severity of these events. The areas of chief concern occur where structures and infrastructure are intermixed or adjacent to wildland fuels

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The impact of definition criteria on mapped wildland-urban interface: a case study for ten counties along the Oregon-California border

PDF

Pre-incident plan mapping in Kern County's wildland urban interface

PDF

Geospatial analysis of the Round Fire: a replication of burn severity analyses in the Sierra Nevada

PDF

Modeling burn probability: a Maxent approach to estimating California's wildfire potential

PDF

Impacts of vegetation management on wildfire severity: a study of the 2021 Caldor fire

PDF

San Diego Wildfire Hazards Information Center mashup

PDF

Using Maxent modeling to predict habitat of mountain pine beetle in response to climate change

PDF

Spatial analysis of human activities and wildfires in the Willamette National Forest

PDF

Installing public electric vehicle charging stations: a site suitability analysis in Los Angeles County, California

PDF

Visualizing historic space through the integration of geographic information science in secondary school curriculums: a comparison of static versus dynamic methods

PDF

A geospatial analysis of income level, food deserts and urban agriculture hot spots

PDF

Historical ecology of the Split Oak Forest in east central Florida

PDF

Community gardens for social capital: a site suitability analysis in Akron, Ohio

PDF

Distribution and correlates of feral cat trapping permits in Los Angeles, California

PDF

Site suitability analysis: small-scale fixed axis ground mounted photovoltaic power plants in Fresno, CA

PDF

Risk analysis and assessment of non‐ductile concrete buildings in Los Angeles County using HAZUS‐MH

PDF

An analysis of the North Rainier Elk Herd area, Washington: change detection and habitat modeling with remote sensing and GIS

PDF

Investigating bus route walkability: comparative case study in Orange County, California

PDF

Developing a replicable approach for the creation of urban climatic maps for urban heat island analysis: a case study for the city of Los Angeles, California

PDF

Preparing for the next major southern California earthquake: utilizing HAZUS with soils maps and ShakeMaps to predict regional bridge damage and closures

Asset Metadata

Creator Greetan, Joshua D. (author)

Core Title The wildland-urban interface in Lassen County, California: a change analysis 2000–2015

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 02/19/2016

Defense Date 02/05/2016

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag fire hazard severity,Lassen County,OAI-PMH Harvest,wildfire,wildland urban interface,wildland-urban interface,WUI

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ruddell, Darren (committee chair), Longcore, Travis (committee member), Oda, Katsuhiko (committee member)

Creator Email greetan@usc.edu,jgreetan@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-209674

Unique identifier UC11277166

Identifier etd-GreetanJos-4119.pdf (filename),usctheses-c40-209674 (legacy record id)

Legacy Identifier etd-GreetanJos-4119-0.pdf

Dmrecord 209674

Document Type Thesis

Format application/pdf (imt)

Rights Greetan, Joshua D.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA