Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Redefining urban food systems to identify optimal rooftop community garden locations: a site suitability analysis in Seattle, Washington
(USC Thesis Other)
Redefining urban food systems to identify optimal rooftop community garden locations: a site suitability analysis in Seattle, Washington
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
REDEFINING URBAN FOOD SYSTEMS TO IDENTIFY OPTIMAL ROOFTOP
COMMUNITY GARDEN LOCATIONS:
A SITE SUITABILITY ANALYSIS IN SEATTLE, WASHINGTON
by
Ana E. Stoudt
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 2015
Copyright 2015 Ana E. Stoudt
ii
DEDICATION
I dedicate this paper to my parents, friends, and coworkers for their constant support throughout
this entire process. Without them, I would not have made it here today.
iii
ACKNOWLEDGMENTS
I am forever grateful to my advisor, Professor Warshawsky, for his encouragement and guidance
throughout this experience.
iv
TABLE OF CONTENTS
DEDICATION ................................................................................................................................ ii
ACKNOWLEDGMENTS ............................................................................................................. iii
LIST OF TABLES ........................................................................................................................ vii
LIST OF FIGURES ..................................................................................................................... viii
LIST OF ABBREVIATIONS ......................................................................................................... x
ABSTRACT ................................................................................................................................... xi
CHAPTER 1: INTRODUCTION ................................................................................................... 1
1.1 Urban Agriculture and Food Systems ................................................................................... 1
1.2 Existing Research Gaps ......................................................................................................... 2
1.3 Using GIS to Rank a Building’s Rooftop Community Garden Potential ............................. 5
CHAPTER 2: RELATED WORK .................................................................................................. 7
2.1 Urban Food Systems ............................................................................................................. 7
2.1.1 Food Utilization .............................................................................................................. 8
2.1.2 Food Availability ............................................................................................................ 8
2.1.3 Food Access .................................................................................................................... 9
2.2 Food Deserts ........................................................................................................................ 10
2.3 Urban Agriculture ............................................................................................................... 12
2.3.1 Intra- and Peri- Urban Agriculture ............................................................................... 12
2.3.2 Seattle’s Urban Agriculture .......................................................................................... 13
v
2.3.3 Defining Urban Agriculture in GIS .............................................................................. 13
2.4 Rooftop Gardens and GIS ................................................................................................... 16
CHAPTER 3: METHODS ............................................................................................................ 19
3.1 Variables.............................................................................................................................. 25
3.1.1 Roof Size ...................................................................................................................... 25
3.1.2 Water Access ................................................................................................................ 26
3.1.3 USDA Food Deserts ..................................................................................................... 26
3.1.4 Food Banks ................................................................................................................... 27
3.1.5 Community Gardens (P-Patches) ................................................................................. 28
3.1.6 Bus Stops ...................................................................................................................... 28
3.1.7 Schools and Education Centers (SEC) ......................................................................... 29
3.2 On-Site Assessments ........................................................................................................... 30
3.3 Limitations .......................................................................................................................... 31
CHAPTER 4: RESULTS .............................................................................................................. 32
4.1 Overall Ranking Results...................................................................................................... 32
4.2 Variable Ranking Results .................................................................................................... 35
4.2.1 Roof Size ...................................................................................................................... 35
4.2.2 Water Access ................................................................................................................ 37
4.2.3 Proximity to USDA Food Desert.................................................................................. 39
4.2.4 Proximity to Food Banks .............................................................................................. 41
vi
4.2.5 Proximity to Community Gardens ................................................................................ 43
4.2.6 Proximity to Bus Stop................................................................................................... 45
4.2.7 Proximity to Schools and Education Centers ............................................................... 47
4.3 Seattle’s “Landscape of Opportunity”................................................................................. 49
4.4 Ground-truthing/On-site Analysis ....................................................................................... 51
4.4.1 305 Harrison Street ....................................................................................................... 51
4.4.2 300 Mercer Street ......................................................................................................... 53
4.4.3 301 Mercer Street ......................................................................................................... 56
4.4.4 801 South Dearborn Street ........................................................................................... 58
4.4.5 2203 Airport Way South ............................................................................................... 61
4.4.6 4201 West Marginal Way Southwest ........................................................................... 63
4.4.7 1250 Denny Way .......................................................................................................... 65
CHAPTER 5: DISCUSSION AND CONCLUSION ................................................................... 69
5.1 Key Observations ................................................................................................................ 69
5.2 Contrasts with Prior Studies ................................................................................................ 70
5.3 Recommendation for Future Research ................................................................................ 72
REFERENCES ............................................................................................................................. 75
APPENDIX A ............................................................................................................................... 81
APPENDIX B ............................................................................................................................... 82
vii
LIST OF TABLES
Table 1 Building characteristic ranking system ............................................................................ 26
Table 2 Food availability and food accessibility characteristic ranking system ........................... 29
Table 3 Food utilization characteristic ranking system ................................................................ 30
Table 4 Final food system variable ranking .................................................................................. 33
viii
LIST OF FIGURES
Figure 1 Elements associated with the three components of food systems (Source: Ingram et al.
2005) ............................................................................................................................................... 9
Figure 2 USDA food deserts, in red, which are located in south Seattle (Source: USDA Food
Access Research Atlas, 2014) ....................................................................................................... 11
Figure 3 High-level methodology framework .............................................................................. 20
Figure 4 Data acquisition, preparation, and integration workflow ............................................... 24
Figure 5 Overall food system variable ranking system (source for base map provided by Esri) . 34
Figure 6 Proportional map of roof size variable ranking system .................................................. 36
Figure 7 Frequency of proximity to current community garden variable ranking ....................... 37
Figure 8 Proportional map of water access variable ranking system ............................................ 38
Figure 9 Frequency of water access variable ranking ................................................................... 39
Figure 10 Proportional map of proximal USDA Food Desert variable ranking system ............... 40
Figure 11 Proportional map of proximal food bank variable ranking system .............................. 42
Figure 12 Frequency of proximity to food banks variable ranking .............................................. 43
Figure 13 Proportional map of proximity to current garden variable ranking system .................. 44
Figure 14 Frequency of proximity to current community garden variable ranking ..................... 45
Figure 15 Proportional map of proximity to bus stop variable ranking system ........................... 46
Figure 16 Frequency of proximity to bus stop variable ranking ................................................... 47
ix
Figure 17 Proportional map of proximity to SEC variable ranking system ................................. 48
Figure 18 “Landscape of Opportunity” for rooftop community gardens in Seattle depicts the
possible locations for a rooftop garden, excluding the roof size variable (see Longcore et al.
2011) ............................................................................................................................................. 50
Figure 19 Aerial view of 305 Harrison Street ............................................................................... 52
Figure 20 Ground view of 305 Harrison Street ............................................................................ 53
Figure 21 Aerial view of 300 Mercer Street ................................................................................. 55
Figure 22 ground view of 300 Mercer Street ................................................................................ 56
Figure 23 Aerial view of 301 Mercer Street ................................................................................. 57
Figure 24 Aerial view of 301 Mercer Street displaying the roof’s rolling geometry ................... 58
Figure 25 Aerial view of 801 South Dearborn Street ................................................................... 59
Figure 26 Ground view of 801 South Dearborn Street ................................................................. 60
Figure 27 View of 801 South Dearborn Street from Interstate 5 .................................................. 61
Figure 28 Aerial view of 2203 Airport Way South ...................................................................... 62
Figure 29 Aerial view of 4201 West Marginal Way Southwest ................................................... 64
Figure 30 Ground view of 4201 West Marginal Way Southwest ................................................. 65
Figure 31 Aerial view of 1250 Denny Way .................................................................................. 67
Figure 32 Ground view of 1250 Denny Way................................................................................ 68
x
LIST OF ABBREVIATIONS
APA American Planning Association
FAO Food and Agriculture Organization
FDA Food and Drug Administration
GIS Geographic Information Systems
PAH Polynuclear Aromatic Hydrocarbons
PCB Polychlorinated Biphenyls
RUAF Resource Centres on Urban Agriculture & Food Security
SEC School and Education Centers
UA Urban Agriculture
USDA United States Department of Agriculture
WAGDA Washington State Geospatial Data Archive
xi
ABSTRACT
As urbanization has increased in recent decades, urban food systems have become stressed,
reducing food security (Gregory, Ingram, and Brklacich 2005). Community gardens that occupy
a city’s vacant lots have been known to combat food insecurity (Oulton 2012, Colasanti 2009),
but many compact cities lack space to garden. One solution has been the development of rooftop
gardens (Tian and Jim 2012). In recent decades, Seattle, Washington has increased the number
of community gardens, but like many urbanizing centers, the city lacks vacant lots for gardening.
With limited ground availability in Seattle and an ever increasing demand to expand upon the
city’s community garden program, otherwise known as P-Patches, to combat this rapid
expansion and improve food security, the city has started to become more creative with its urban
spaces through activities such as rooftop gardens (Forbes 2013, Cronin 2013, Greene 2013,
Seattle.gov 2014). The goals of this thesis are the following: (1) determine criteria to represent
Seattle’s food system in a site-suitability analysis to improve food security; (2) rank 33 potential
buildings using this spatial index; (3) and perform a ground-truthing exercise to complete and
on-site assessment of the seven highest ranked buildings. By taking a more holistic approach
when selecting variables, buildings were identified that not only provided the structural needs of
a rooftop community garden, but are optimally located within the city’s food system based on
availability, accessibility, utilization, the three main components that comprise an urban food
system (Gregory, Ingram, and Brklacich 2005). Future studies should examine further
modification of selecting for these food system variables, which could then provide a more
accurate and realistic representation of urban food systems as a means to improve food security.
1
CHAPTER 1: INTRODUCTION
Over the past decade, rates of urbanization have climbed as the world‘s population has grown
from approximately 6 billion to over 7 billion with almost half the planet’s population occupying
urban areas. Rapid growth of urban areas has been linked with the degradation of land,
deforestation, pollution, climate change, increased unemployment, gentrification, the
overburdening of current infrastructure, and lack of urban amenities (UNCHS 2001b, Kisner
2008, UNFPA 2014). The expansion of the city center and development of new cities has caused
policy makers and governments alike to examine new and creative ways to reduce these adverse
effects such as urban agriculture (UA) to improve the city’s food system.
1.1 Urban Agriculture and Food Systems
Urban agriculture, in one form or another, has been present in the United States since the
1890’s as a way to “provide land and technical assistance to unemployed workers in large cities
and to teach civics and good work habits to youth” (University of Missouri, Extension 2009, 1).
During World War I and the Great Depression, the government promoted community gardens as
a source for domestic crops and as a way for the unemployed to grow their own food. Continuing
into the Second World War, community gardens were rebranded as Victory Gardens in hopes of
improving the country’s morale and influence people to grow food for their own personal
consumption (Parks 2012). This trend then continued into the 1970’s where community
gardening took on the form of what we most commonly see today in response to a poor
economic environment, environmental concerns, and a rise in urban abandonment. Urban
residents utilized these gardens as a means to improve their neighborhood’s social capital
through expanding green spaces into vacant lots, improving economic support, and stabilizing
food security.
2
When a city has experienced an economic downturn, it is common for the unoccupied,
derelict land to be utilized for community gardens. Most of the gardens in these once neglected
and vacant parcels of land can provide a city with the opportunity for revitalization (Oulton
2012, Colasanti 2009). This type of urban agriculture is prevalent in American cities such as
Detroit, Chicago, and New York City, where thousands of acres of vacant plots have been given
over to the community to grow food while also improving social capital and minimize
environmental strains (Deelstra and Girardet 2000, UNEP 2011, Oulton 2012).
Urban food systems have been prone to a reduction in food security; however, as a rise in
community gardens have been known to combat the negative effects of rapid urbanization and
population growth unique to cities. Cities that include community gardens in urban development
plans create sustainable food systems, alas, only a handful of cities include community gardens
in urban development planning, the most prominent being Seattle (Hou, Johnson, Lawson 2009).
Seattle has been known as one of the leaders in the urban agricultural movement in the past five
years by promoting their P-Patch program for community development and reduction in food
insecurities (Schukoske 1999, Erikson 2009, Dunn 2010). Ground-level community gardens
work well for cities that have vacant lots such as Akron, Ohio and Detroit, Michigan, but are
seen as occupying valuable ground space in growing cities such as Hong Kong, New York, and
Seattle (Oulton 2012, Tian and Jim 2012). Incorporating green roofs or rooftop gardens could be
part of the solution to improving food security within such cities.
1.2 Existing Research Gaps
Current community garden studies show correlation between lower socio-economic
status and proximity to gardens on the ground possibly to help mitigate the presence of food
deserts (Wilkinson 2012). As these studies have been completed in regards to community
3
gardens on the ground, little if no research has been on the spatial distribution of rooftop gardens
relative to urban food systems. Based off of the association between ground-dwelling gardens
and food deserts, would it be ideal to consider food desert boundaries as a selective trait in
determining potential rooftop garden locations?
Food systems can be comprised of three components: (1) food availability; (2) food
accessibility; (3) food utilization (Gregory, Ingram, and Brklacich 2005). Availability
encompasses the production, distribution and exchange of food, while food accessibility may be
better defined as the allocation, affordability, and preference (Gregory, Ingram, and Brklacich
2005). Once food is available and accessible, food utilization make take place, otherwise defined
as nutritional values, food safety, and social values (Gregory, Ingram, and Brklacich 2005).
Food security is achieved when a food system functions so that “all people, at all times,
have physical and economic access to sufficient, safe and nutritious food to meet their dietary
needs and food preferences for an active and healthy life” (FAO 1996), 1. When food systems
are stressed, food security dwindles, therefore could urban food systems be accurately measured
through a spatial index? What variables should be measured?
Rooftop gardens or green roofs have become popular for a multitude of reasons. In
Syumi Rafida Abdul Rahman and Hamidah Ahmad’s Green roofs as urban antidote: A review
on aesthetic, environmental, economic and social benefits, the authors state that some green roof
benefits include a reduction in interior temperature, increased building/roof lifespan and fire
resistance, improve sound insulation, provide small-scale open green space, filter storm water
runoff, and raise aesthetic values of a buildings appearance. They have become increasingly
popular in “compact cities” such as Hong Kong which are strapped for ground space (Tian and
Jim 2012).
4
In Yuhong Tian and C.Y. Jim’s Development potential of sky gardens in the compact city
of Hong Kong, the authors utilize GIS techniques aligning a green space layer with roof and/or
podium outline layers to identify potential or actual roof gardens. The sky gardens were defined
by 5 subcategories, not including vegetation for human consumption, but rather trees and
shrubbery.
A similar analysis is completed in Danielle Berger’s A GIS Suitability Analysis of The
Potential for Rooftop Agriculture in New York City a set of GIS criteria is developed in order to
identify rooftop for large scale urban agriculture. Through trial and error, the author decides on a
ranked system for its potential to house a rooftop farm by examining size, zoning, floor height,
and year built. These variables were selected based off of the building construction and
restrictions unique to New York City. Although her results might provide useful for the city and
large scale farming, a more generalized approach is needed if we seek to increase the prevalence
of green roofs. Existing research like Tian and Jim, or Berger analyze green roofs spatially in
using a variety of different methodologies and variables based on the aspect of green roof
benefits they hope to examine, such as locating vegetative roof locations or roof tops for
commercial agriculture. Yet none have been completed on the community garden aspect of
rooftop gardens which focuses on providing the collective benefits of improving food security
and possibly slowing the negative effects of rapid expansion.
Studies on green roofs tend to consider many of the same variables needed for analysis
such as space, zoning, and stability. However, with so many different types of green roofs (from
grasses to produce) and varying catalysts for implementing such roofs, these variables are
defined uniquely for the gardens purpose. Yet, how can rooftop community gardens be defined
for a spatial index in accordance with Seattle’s food system? And which regions within the city
5
would benefit the most from such garden?
Similar research has dealt with large rooftop urban agriculture operations providing a
generalized geospatial analysis when locating potential gardens in New York City. This research
has provided a great building block for identifying what components of a roof top garden are
necessary for geospatial analysis for America’s East Coast; however the building codes, climate,
and city structure greatly vary from that of Seattle’s. Due to Seattle’s green initiatives, food
systems, and permit requirements, how will this affect how spatial variables will be defined for
potential rooftop gardens?
1.3 Using GIS to Rank a Building’s Rooftop Community Garden Potential
Food systems can be comprised in a variety of ways (Wrigley et al., 2002, Morton and
Blanchard 2007, USDA Economic Research Service 2009) but struggle to accurately depict a
city’s food system to identify optimal locations to improve food security. Also we know that a
spatial index can provide a more in depth analysis in a site suitability analysis (Oulton 2012,
Opfer 2010), but can a holistic approach accurately identify potential rooftops?
This paper seeks to rank buildings through a site suitability analysis within Seattle
identifying buildings that could house a rooftop community garden by adequately defining
Seattle’s urban food system. Selecting for the variables that best represent Seattle’s food system
(availability, accessibility, and utilization), a spatial index allows for a more in depth
examination of not only a buildings structural potential but how well a building is spatially
situated within the urban food system. Identifying the locations of rooftop gardens will reduce
development of redundant community gardens in the city.
The remainder of this thesis includes four chapters. Chapter Two will first examine
related works on urban agriculture defining the various types of urban agriculture and food
6
systems. Similar urban agriculture studies involving the use of GIS provide a better
understanding of how GIS can play a supportive role in analysis, ending with a look at variables
needed for geospatial green roof analyses. Chapter Three reviews the study area, Seattle, why
variables were selected for and how they are defined. Also identifying data sources and
preparation, the methodology of the spatial index analysis, and concluding with the ground-
truthing exercise. Chapter Four will provide the visual documentation and results of the study.
While Chapter Five will discuss this thesis in its entirety and propose future research.
7
CHAPTER 2: RELATED WORK
Most published literature on community gardens are focused on their optimum location. Many
scholars have discussed locating ground-level community gardens through GIS (Elwood 2002,
Randall, Churchill, and Baetz 2003, Greene and Pick 2011), however, there are very few GIS
studies have been completed on locating roof top gardens (Tian and Jim 2012). This literature
review is divided into three sections that are necessary for establishing a thorough understanding
of urban agriculture. The first section discusses and defines how urban food systems have been
categorized in prior studies in association with food security. The second section is a review of
urban agriculture and the classification of various types of urban agriculture in accordance with
the city of Seattle and the American Planning Association (APA). Finally, the literature review
ends with the most recent study of rooftop gardens and discusses the methods used for locating
potential rooftop gardens with GIS. From this framework it is possible to illustrate how GIS can
further expand the analysis of rooftop gardens in addition to contributing to food system
research.
2.1 Urban Food Systems
A sustainable urban food system is important for a variety of reasons. Urban food
systems provide access to healthy nutritious food to city residents. Creating sustainable urban
agriculture can help stabilize food insecurity within the city (Francis, C. et al. 2003, Gregory,
Ingram and Brklacich 2005, FAO 2006, Boko, et al. 2007, Freedman and Bell 2009, RUAF
2010). Activities involving food systems have been known to create economic value and build
habitable communities that provide safe, affordable, and fresh food to residents (RUAF 2010).
For all intents and purposes, this study focuses on reduction of food insecurity within urban food
systems by food asset mapping.
8
In regards to food security, an urban food system can be comprised of three things: food
utilization, food access, and food affordability (Francis, C. et al. 2003, Gregory, Ingram, and
Brklacich 2005, FAO 2006, Boko, et al. 2007). When a food system is strong in all three aspects
of utilization, access, and affordability, there is a reduction in food insecurities. Accurately
examining and defining these criteria in food asset mapping is critical to correctly identifying
food insecurities within a city. The following sections provide definitions of these criteria and
how they can be measured.
2.1.1 Food Utilization
Food utilization can be represented in the food system by numerous measurements.
Utilization has been known to encompass food safety, social value, nutritional value (Gregory,
Ingram, and Brklacich 2005, Boko, et al. 2007).
Food safety is important for public health, especially in regards to nutrition and the
prevention of spreading foodborne illnesses. Food safety can encompass handling, labeling,
additives, and inspection of foods (FSMA 2014). Food safety in the United States is regulated on
a state and federal level; most commonly know is the Food and Drug Administration (FDA
2014). FDA inspection acts have helped improve food safety (Nielsen 2010), ultimately
providing safe food to citizens.
2.1.2 Food Availability
The third variable to defining urban food systems is food availability. Food Availability
deals with factors like production, distribution, and exchange of food in a community (Gregory,
Ingram, and Brklacich 2005; FAO 2006, Boko, et al. 2007).
Availability is important in an urban food system as this is the basis for not only
determining how much food is available but how food is distributed spatially throughout the city,
9
and if foods are available for exchange. Figure 1 shows all three components as displayed in
GECAFS science plan and implementation strategy (Ingram et al. 2005).
Figure 1 Elements associated with the three components of food systems (Source: Ingram
et al. 2005)
2.1.3 Food Access
The affordability, allocation, and preference of food are related to food access (Gregory,
Ingram, and Brklacich 2005, FAO 2006). Food should be accessible to residents in allocation
and economic standings should not restrict access to safe and healthy foods.
Accessibility to fresh produce is important for public health, as well as socio-economic
aspects and development processes (Gregory, Ingram, and Brklacich 2005; Boko et al. 2007).
Human health is dependent upon obtaining foods to reduce susceptibility to diseases. This in turn
can undermine livelihoods and alter socio-economic structures, reducing social protections if not
properly managed. Limited access to healthy and affordable foods is commonly defined as a
food desert (USDA 2009).
10
2.2 Food Deserts
Food security is achieved when a food system functions so that “all people, at all times,
have physical and economic access to sufficient, safe and nutritious food to meet their dietary
needs and food preferences for an active and healthy life” (FAO 1996). When food systems are
stressed, food security dwindles. The United States Department of Agriculture first defined food
deserts as an, “area in the United States with limited access to affordable and nutritious food,
particularly such an area composed of predominantly lower income neighborhoods and
communities” (USDA Economic Research Service 2009, a: 1) in 2009 with a yearlong study
assessing areas within the nation with limited access to nutritious and affordable. Figure 2
depicts Seattle’s food deserts by USDA’s definition, which are located in the southern region of
the city.
11
Figure 2 USDA food deserts, in red, which are located in south Seattle (Source: USDA
Food Access Research Atlas, 2014)
12
2.3 Urban Agriculture
Urban agriculture (UA) is the practice of “growing, processing, and distributing of food
through intensive plant cultivation and animal husbandry in and around cities” (Bailkey and Nasr
2000:6). Studies have discussed the costs, benefits, and planning of urban agriculture within a
city (Mougeot 2006). Tip 244 from Seattle’s Department of Planning and Development, Seattle’s
Land Use Code, classifies UA into five distinct urban agriculture uses: animal husbandry,
aquaculture, community garden, horticulture, and urban farm (see Appendix A) (DPD 2014).
This in combination with the American Planning Association (APA) provides a basic
understanding of the types of urban agriculture.
2.3.1 Intra- and Peri- Urban Agriculture
Urban agriculture can be categorized into two classifications, intra- or peri-urban
agriculture (Mougeot 2000, FAO 2007). Intra-urban agriculture refers to UA located within a
city or urban space, while peri-urban agriculture refers to UA along the fringes of urban areas,
more along the border of a city and rural setting. Although these are the subdivisions of UA tend
to dictate the scale at which UA is found, various types of UA can be found throughout intra-
and peri-urban agriculture (FAO 2007).
Although UA can be classified as the raising of animals or growing of plants for food and
various other uses within or around the city, UA is commonly seen throughout cities as
community, institutional, or rooftop gardens (Smit et al 1996b, Mouget 2000, FAO 2007). These
gardens can provide produce to feed the surrounding communities and also provide, social
capitalism, education to students on farming and sustainability, as well as local economic
development through selling of produced at farmers markets/grocery stores (FAO 2007, Oulton
2012, P-Patch Community Gardens 2014).
13
2.3.2 Seattle’s Urban Agriculture
Seattle’s best known in urban agriculture for its community gardens called P-Patches.
The P-Patch Community Gardening Program was first introduced in 1973, started by the Picardo
(family) Farm, thus commemorating the family with the letter “P” in P-Patch (P-Patch
Community Gardens 2014). Since the first garden, the program has grown to 88 gardens
throughout the city, totaling 32 acres. Some of the main goals of the program are to grow the
community, feed the hungry, education, and improve community food security (P-Patch
Community Gardens 2014). Community Environment and Planning (CEP) outlined Seattle’s UA
advocacy and outreach from the 2009 urban agriculture visioning meetings. It is important to
note that these community gardens are maintained by members of the community, and without
constant membership renewals and new members these community gardens would not prosper.
2.3.3 Defining Urban Agriculture in GIS
The APA released a report in 2011 known as the Urban Agriculture: Growing Healthy,
Sustainable Communities, which discusses the use of urban land for community gardens and its
implications for urban land use planning (Hodgson, K., Campbell, M.C., Bailkey, M. 2011). The
report provides a great commentary of the topic through case studies of urban agriculture within
North America. Starting with the history of urban agriculture, the APA discusses the
industrialization of North American cities and the effect this had on food production. The APA
equates this resurgence in community gardens to “deindustrialization, depopulation, increase in
acreage of vacant land, and the failures of urban renewal but also to immigration” (Hodgson, K.,
Campbell, M.C., Bailkey, M. 2011, 12). Many of these community gardens are managed by
community members, nonprofit organizations, and local governments.
Similar to the five types of urban agriculture defined by Seattle’s Land Use Code, the
14
APA report states that there are various determinants that will shape what type of urban
agriculture can be grown. It has been determined that land is the primary requirement for urban
agriculture as a lack of growing space can greatly reduce vegetation productivity. Most studies
discuss the use of vacant lots as a solution to determining where new community gardens could
be placed (Oulton 2012); however this does not work well for municipalities that have been
“built-out.” The report discusses how although there might be a lack of usable vacant lots in
these built-out cities, there is such great diversity in the types of urban agriculture that many
gardens or farms can be adapted to a variety of size, shapes, and locations (Hodgson, K.,
Campbell, M.C., Bailkey, M. 2011, 23).
Deterrents of urban agriculture can not only be a lack of available land, but soil
contamination (Lovell 2010), thus affecting the ability if the community to grow safe and healthy
produce. Seattle’s Duwamish neighborhoods are an excellent example of the negative impact of
contaminated soils. According to a report released by the EPA in April, 2010, “sediments in the
Lower Duwamish Waterway are contaminated with polychlorinated biphenyls (PCBs),
polynuclear aromatic hydrocarbons (PAHs), phthalates, and inorganics,” thus advising residents
to refrain from growing their own food in the ground soil.
Geographic Information Systems (GIS) has been a technical tool for urban planning over
the past couple of decades. There have been multiple GIS studies involved with urban food
production and activities such as Dongus, S. and Drescher, A.W. (2000) Resource Centres on
Urban Agriculture & Food Security’s, Use of Geographic Information Systems (GIS) and Global
Positioning Systems (GPS) for Mapping Urban Agricultural Activities and Open Space in Cities,
which first examined inner city food production with GIS.
The authors of the study clearly state the benefits of using Geographic Information
15
Systems (GIS) for determining potential open lands as well as taking an inventory of current
vegetable production:
Visualization of spatial data, particularly the distribution of agricultural open spaces in a
city
Simple analytical functions such as calculation of the sizes of agricultural areas
Possibility for data overlay in order to investigate relations with various relevant factors,
e.g. designated land use, irrigation water quality, socioeconomic variables etc.
Potential for updating digital maps in the future, and extension to a greater range of topics
and layers
Possibility to print hardcopies of maps showing any desired selection of topics and areas
in any scale, for discussions with stakeholders
Linkage of vector data in maps with attribute data such as type of crops grown or number
of farmers
High flexibility: According to the respective local contexts and available data sources, a
wide variety of spatial data can be integrated and combined for optimal outcome:
Satellite imagery, aerial photography (digital or analogue), topographic or thematic maps
of all scales, cadastral maps, GPS measurements etc. (Dongus & Drescher 2000, 1).
It has become present that GIS is used as a valuable tool when identifying, analyzing, and
portraying urban agriculture for the above reasons.
Numerous GIS studies have been completed with a particular type of urban agriculture
known as community gardens. Allison Oulton’s Community Gardens for Social Capital: A Site
Suitability Analysis in Akron, Ohio, and other site suitability analyses of community gardens, this
thesis takes this kind of analysis to new heights. Many dissertations discuss the benefits of on-
16
the-ground community gardens and how they can benefit a community, however very few
discuss as such for rooftop gardens. Rooftop gardens require a different set of criteria in order to
be built compared those of the traditional on-the-ground gardens. The criteria primarily used for
a site suitability analysis for the on-the-ground gardens deals with: vacant lots, zoning, soil type,
and accessibility to water and sunlight (Wilkinson 2012).
Although there has been much discussion of the methodology for site suitability analyses
of community gardens on the ground, there has yet to be such a discussion for rooftop
community gardens. When completing a site suitability analysis for a community garden, certain
factors must be taken into consideration such as: soil type, slope, sunlight, parcel size, and water
accessibility (Oulton 2012). However, when placing a garden on a rooftop terrace there are
similar and different criteria that must be taken into consideration such as: roof square footage,
zoning, building age, roof pitch, and possibly height (Berger 2013). Other criteria that must be
taken into consideration with these are possibly accessibility to light and water, while soil would
be a nonexistent factor for rooftop garden analysis.
2.4 Rooftop Gardens and GIS
Over the past 5 years there has been an increase in GIS studies completed on green roofs.
Yuhong Tian and C.Y.’s 2012 paper provided a great introduction into sky gardens and how GIS
can be applied to identifying potential green spaces. The most recent study of rooftop gardens, in
particular, and GIS was completed by Danielle Berger, which is an excellent springboard for this
study’s methodology section. The author follows the process of first identifying potential
rooftops from the buildings shapefile, then selecting by the zoning attributes, number of floors,
and then the area and year built (Berger 2013).
Tian and Jim 2012 study examined “sky gardens” in Hong Kong, particularly focusing on
17
regional requirements for vegetation to be present on a roof or podium. The authors developed a
methodology involving geospatial data such as building layers, land use layers and green space
layers in tandem to identify roof gardens and podium gardens by land uses (see Appendix A).
The variables prove to be translational to Seattle with some alterations primarily due to
geographic location. Hong Kong is prone to typhoon damages which ultimately could affect the
types of vegetation grown on the roofs. Seattle has a much more temperate climate, ideal for
growing crops year round. In the results sections, the authors shed summed up potential areas by
a variety of different categories such as land use, building story, and district. Although this
information is educational, it might not be necessary for this study.
Berger’s study is completed in the context of New York City. New York City is home to
8 million people in 305 square miles, and is known for its extreme temperatures, high
inequalities in income and availability of fresh foods, and large levels of stormwater runoff
(Berger 2013). The author was driven by these factors to expand upon NYC’s already large
levels of urban agriculture through the use of spatial data and GIS modeling. Her resulting site
suitability methodology was defined by three factors (1) square footage (2) roof live load
capacity (3) and building ranking system. The result then presented her with the buildings that
had green roof potential. Although the case study was completed in the context of New York
City, where the average building height towers above that here in Seattle, it does provide a nice
basis for what criteria might be the most important, as well as why it was selected. Although
Berger’s study worked well for a more industrialized version of rooftop urban agriculture in New
York City, not every city has the proper datasets or similar zoning ordinances to identify rooftop
gardens using her GIS methodology.
The City of Seattle will require a similar, yet different methodology than the one
18
presented in Berger’s analysis for an assortment of reasons. Berger created a building ranking
system to identify least suitable (value of 1) to most suitable (with a value of 6) based on a
combination of factors. A rooftop garden site suitability analysis in Seattle can differ slightly
from this. It is agreed that the analysis should fall under specific city boundaries, such as
Seattle’s city limits. However, Berger limits her study by commercially or manufacturing zoned
buildings, greatly reducing the number of potential rooftop gardens. The proposed methodology
in the Seattle study examined a larger variety of zoned buildings based off of the green roof
zoning requirements for the city. Number of floors will not be examined in this study either due
to the more agreeable climate of Seattle and a lack of a complete dataset containing number of
floors. Many of the “noxious or city utility” defined buildings have great potential to house green
roofs in Seattle, therefore this criteria was removed from the study.
Continuing differences between the proposed study and that of Berger’s are the roof
square footage and roof live load capacity. Again, the author divided the roof square footage
into multiple classifications of small (<5,000 square feet), medium (5,000-40,000 square feet),
and large (>40,000 square feet) (Berger 2013). These numbers appeared far too large for the
study at hand, primarily due because the study seeks to examine rooftop gardens versus the
industrial sized gardens Berger examines. Finally is the measure of roof live load capacity,
necessary for supporting rooftop urban agriculture. Berger measured roof live load capacity by
the building’s year of construction. Her reasoning was that buildings built before 1968 had a
lower live load capacity while buildings built after 1968 had higher capacities based off of
NYC’s building code changes (Berger 2012). Since building codes differ from city to city, this
same criteria would not work well for the Seattle analysis. Stability is therefore best measures by
an on-site assessment of a building engineer.
19
CHAPTER 3: METHODS
To identify ideal buildings for rooftop community gardens, a combination of seven variables
were considered. Physical and proximal variables were analyzed through a spatial index of food
security constructed for this analysis. The methodology follows three main steps: (1) identify city
owned buildings and prescreen for suitable size; (2) buffer, analyze, and rank vacant parcels by
basic garden requirements and food system criteria; (3) ground-truth through on-site assessment
(see Figure 3).
Previously, green roofs have been measured through remote sensing using a combination
of building, land use and green space layers (Tian and Jim 2012). This method only examined
green roofs rather than locating roof top gardens as a means to resolve food insecurities.
Although Oulton (2012) ranked her buildings under two separate criteria, physical and social
capital, due to time restrictions and available data, a unified scoring system was used in this
study to rank the available buildings.
The spatial index of food security drew from the basic building requirements from other
green roof literature (Tian and Jim 2012) and the definition of food systems and security
(Gregory, Ingram, & Brklacich 2005). The term food system encompasses (1) food availability –
production, distribution, and exchange; (2) food accessibility – allocation, affordability, and
preference; (3) food utilization – societal and nutritional values, and safety (Gregory, Ingram, &
Brklacich 2005).
20
Figure 3 High-level methodology framework
21
These food system variables were utilized in this spatial index. Food availability,
accessibility, and affordability can be mapped as geographic reference points to identify the
spatial framework of Seattle’s food system by selecting certain variables to represent
components that make up a food system. The use of a spatial index involving these variables will
better represents the nuances and complexities of urban food systems, providing a more holistic
approach.
To map these variables, the index used in Seattle attempted to derive these geographic
reference points from these categories including United States Department of Agriculture
(USDA) Food Deserts, food banks, current community gardens (P-Patches), transit stations (bus
stops), and schools/education centers. Water appurtenances were also mapped as representative
geographic reference points of civic engagement. The spatial distribution of the food system was
reflected through these geographic reference points, which were weighted equally. These
categories were selected for based off of prior studies’ classifications, garden requirements,
interviews with government and planning officials, and available data.
Variables were weighted for individual addresses rather than individual polygon entities.
The term address refers to any building that resides at the postal address provided within the
linked city’s building-assessors shapefile. This study was restricted to buildings that were city
owned to promote lower turnover rates, similar to that of the P-Patch gardens, and any
commercial, or manufacturing and industrial centers, to adhere to Seattle’s urban agriculture
zoning codes (see Appendix A). The Seattle Public Utilities maintained the data for the building
outline shapefile which was downloaded through the Washington State Geospatial Data Archive
(WAGDA). This data was joined with the King County Department of Assessments tabular data
to provide address and other building information. This thesis chose to rank buildings under a
22
unified scoring system, producing a final ranking of the buildings as an average rank (see Figure
4).
In this model, buildings analyzed were selected through a preliminary screening
involving size, land use, and property owner. The minimum size was greater than or equal to
10,000 square feet, to provide an adequate amount of produce to the surrounding community. In
this analysis, greater roof size was considered an asset and a building’s rank was higher with
greater roof space. Any buildings where the land use was not commercial, or manufacturing and
industrial centers were excluded in the preliminary screening to adhere to Seattle zoning code’s
(see Appendix A) and speaking with Branin Burdette an employee as a Planner for Seattle Land
Use. Current community gardens within Seattle occupy parcels owned by the City of Seattle to
reduce garden turnover compared to privately owned lands (Macdonald 2014). Subsequently,
only city owned buildings were selected for in this analysis to reduce rooftop community garden
turnover. Following this preliminary screening, buildings were ranked by the food system
characteristics.
24
Figure 4 Data acquisition, preparation, and integration workflow
25
3.1 Variables
The following subsections examine which variables were selected for in this analysis,
where the data was downloaded, and how the variable was measured. Variables were identified
and selected based off of previous studies, literature, and interviews with Seattle zoning officials,
Seattle Urban Farm Company, and Seattle Department of Planning representatives.
3.1.1 Roof Size
In order to provide produce yields to improve food security, it is assumed that more
growing space provides more growing potential. Seattle’s minimum ideal sized P-Patch is 4,000
square feet for on ground gardens (Seattle Department of Neighborhoods 2014). Similar to on
the ground gardens, rooftop gardens are limited by available roof space (or size), therefore larger
sized gardens could accommodate more productions increasing the number of people served by
the garden (NRCS 2014). This is assuming that there are enough community gardeners to
maintain the rooftop gardens. This variable was included in the analysis as roof size which
represents growing capacity.
Square feet of roof area will be treated as an asset in this model; (1) a larger number of
community members may partake in gardening with more available space, and (2) a larger roof
area will provide the community with a greater amount of produce thus improving the food
system and food security. This variable was measured through The City of Seattle’s building
roof outline polygon layer from 2012. Size ranking categories were determined by the Jenks
natural break within the dataset. The most desirable buildings score a five, as this represents the
buildings with the largest roof area available, while scoring a one means the roof had little square
footage.
26
3.1.2 Water Access
Water access is a necessity when it comes to growing plants (Seattle Department of
Neighborhoods 2014). Therefore, proximity to water availability was included a variable in this
analysis, where a building scored highest when closest to a water appurtenance and lowest when
beyond 500 feet, meaning non-available as a water source, as adapted from Oulton’s study.
Water sources that are located on the building are defined as “along boarder” in Table 1.
Although a new water meter can be installed by Seattle Public Utilities, installation of an
appurtenance can end up costing the gardeners more money than has been budgeted (Seattle
Department of Neighborhoods 2014, Seattle Public Utilities 2014) make it less ideal to have a
garden being located too far from a water source.
Water access for irrigation is a variable adapted from Community Gardens for Social
Capitol (Oulton 2012). This variable was assessed through the City of Seattle’s water
appurtenances point shapefile. Each building was ranked according to its proximity to the nearest
appurtenance (see Table 1).
Table 1 Building characteristic ranking system
Roof Size Water Access
Building Address
5 = >200,000 sq. ft.
4 = >100,000 sq. ft.
3 = >40,000 sq. ft.
2 = >20,000 sq. ft.
1 = <20,000 sq. ft.
5 = Along boarder
4 = within 50 ft.
3 = within 100 ft.
2 = within 500 ft.
1 = greater than 500 ft.
3.1.3 USDA Food Deserts
Food deserts, as defined by the United States Department of Agriculture, are defined as
“census tract with a substantial share of residents who live in low-income areas that have low
levels of access to a grocery store or healthy, affordable food retail outlet,” primarily using
27
census tracts as the unit for analysis. Low-income and low-access communities are defined as
follows:
1. They qualify as "low-income communities", based on having: a) a poverty rate of 20
percent or greater, OR b) a median family income at or below 80 percent of the area
median family income; AND
2. They qualify as "low-access communities", based on the determination that at least 500
persons and/or at least 33% of the census tract's population live more than one mile from
a supermarket or large grocery store (10 miles, in the case of non-metropolitan census
tracts) (USDA 2014, 1).
The USDA Food Access Research Atlas was downloaded in Excel spreadsheet format from
http://www.ers.usda.gov/ which was then joined to the Census Bureau's Cartographic Boundary
File to provide spatial context.
3.1.4 Food Banks
Related to food availability, distribution and exchange of produce is most effectively
measured through food banks. To improve community food security, Seattle P-Patchers donate
more than ten tons of produce each year to neighborhood food banks (Seattle Department of
Neighborhoods 2014). In order to continue this trend, it would seem ideal to have to have food
banks being located as close as possible to their sources to reduce cost of transport and time
between ‘farm to table’. This variable was measured for the study in Seattle by the proximity to
food banks, and was assessed through food banks. Staying consistent with the previous variables,
buildings located closest to food banks scored higher than those located farther away. This
variable was assessed though a combination of manually digitization and the King County GIS
Center’s Food Facilities shapefile totaling 49 food banks.
28
3.1.5 Community Gardens (P-Patches)
Although the City of Seattle does not restrict the establishing of a new P-Patch
community garden based on its proximity to current community garden locations, it may prove
useful to eliminate the development of redundant gardens. By selecting this as a variable in a
spatial index rather than a singular binary variable, allows for a more flexible analysis where
proximity to current community gardens does not skew a buildings ranking too much.
The variable was measured as proximity to current P-Patches, where a score of five
means a gardens is farther away from a building and a score of one indicates extremely close
proximity to an established P-Patch. This variable was measured proximally to remain consistent
with the previous variables and definition of the USDA food desert. A point shapefile of existing
Seattle community gardens listed on WAGDA (maintained by The City of Seattle) was utilized
in this analysis. These points were then used in the proximity analysis. A total of 82 points were
identified.
3.1.6 Bus Stops
Limited accessibility to fresh fruits and vegetables is one of the key concerns for food
deserts as defined by the USDA (USDA 2014). A community is considered low-access if a third
of the population lives more than 1 mile from a supermarket or grocery store, so it is important
for people to be located within that 1 mile radius of fresh foods (USDA 2014). In order to ensure
the public can access the produce with ease, proximity to bus stops was included as a variable.
Again, the closer a bus stop was located to a building, the higher that building scored in the
analysis. Specifically bus stops were selected as Seattle has a robust metro system that is easily
accessible by the public.
Food accessibility was represented in this analysis by Metro bus stop locations. The City
29
of Seattle maintains the point shapefile containing transit stops. Preference was given to
buildings within closer proximity to bus stops in accordance of USDA Food Desert
classifications (USDA 2014, 1). Rankings are detailed in Table 2.
Table 2 Food availability and food accessibility characteristic ranking system
Proximity to
USDA Food
Desert
Proximity to Food
Banks
Proximity to Current
Community Gardens
Proximity to Bus
line
Building
Address
5 = within food
desert
4 = within 1/4 mi. of
food desert
3 = within 1/2 mi. of
food desert
2 = within 1 mi. of
food desert
1 = greater than 1 mi.
from food desert
5 = within 1/4 mi. of
food bank
4 = within 1/2 mi. of
food bank
3 = within 3/4 mi. of
food bank
2 = within 1 mi. of
food bank
1 = Greater than 1 mi.
of food bank
5 = greater than 1 mi. of p-
patch
4 = within 1 mi. of p-patch
3 = within 3/4 mi. of p-patch
2 = within 1/2 mi. of p-patch
1 = within 1/4 mi. of p-patch
5 = within 1/4 mi. of bus
stop
4 = within 1/2 mi. of bus
stop
3 = within 3/4 mi. of bus
stop
2 = within 1 mi. of bus
stop
1 = greater than 1 mi.
from bus stop
3.1.7 Schools and Education Centers (SEC)
To ensure that the future rooftop garden will be utilized by the community, proximity to
schools and education centers will help rank a building by its potential to add societal value. As
food utilization can be measured by societal value (Gregory, Ingram, and Brklacich 2005), and
schools or education centers as a location for community outreach in educating the public on
UA.
To measure food utilization, public schools and community education centers were
combined into one single layer. Point shapefiles were downloaded from The City of Seattle and
merged into 1 comprehensive layer. A proximity analysis was performed where the closer a
building was to a SEC the higher the building scored (see Table 3). This variable was selected
for to allow for students to easily access these gardens for hands-on learning.
30
Table 3 Food utilization characteristic ranking system
Proximity to School/Education Center
Building
Address
5 = within 1/4 mi. of SCC
4 = within 1/2 mi. of SCC
3 = within 3/4 mi. of SCC
2 = within 1 mi. of SCC
1 = greater than 1 mi. of SCC
3.2 On-Site Assessments
A ground-truthing exercise will be performed at the end of the analysis for the seven
highest ranked buildings. These visits will help verify the data, and measure the realities of each
building site to determine if the building would actually be suitable for a rooftop community
garden. The visits will entail a more qualitative approach to the analysis rather than the
quantitative approach taken earlier for the spatial index. Ground-truthing is important to assess
the accuracy of remotely sensed data by actually measuring in the field (ESRI 2014).
On-site assessments of the top seven buildings were performed to ground-truth results.
Physical indicators were assessed to verify not only that the building was located at the correct
address, but that the buildings were easily accessible (via transit and the public could access
them), and city owned. Aerial images were then used to examine the building’s rooftops to
determine the roofs likelihood of housing a rooftop garden. Some factors examined were usable
space, noticeable damage, and verification of the buildings presence. Buildings either were found
as acceptable or unacceptable by meeting this criterion. The on-site assessment is expected to
align accordingly with the geospatial site-suitability analysis.
31
3.3 Limitations
There were a few limitations to this methodology. Availability of data restricted which
variables were selected for in this analysis. Stability is a great example of a dataset that would be
time consuming and expensive to acquire and create. Before rooftop gardens can be built, a
rooftop assessment must be completed by the city (see Appendix A). Garden layout is greatly
affected by stability and vice versa (Seattle Urban Farm Company 2014). This was a highly
generalized analysis, were food system variables could be measured a variety of ways. Although
the scoring methodology provides a stable framework to first rank buildings, the methodology
works best in tandem with the ground-truthing exercise.
Variables that were excluded, other than stability, were alternative rooftop characteristics
and distance to freeways. Distance to freeways was not included due to a lack of applicable data
such as building height. Since the rooftop height was unavailable, it would be difficult to
determine if a potential rooftop gardens proximity from a ground-level freeways as well as
determining a roof’s distance from an overpasses. The other rooftop variables not included were
unavailable in accessible data and could not easily and accurately be obtained.
32
CHAPTER 4: RESULTS
Of the available 417,283 building entries within the Seattle municipality, only 6,948 were zoned
accordingly. Only 177 of those buildings were owned by the government. Forty-one of those
building entries passed the preliminary size screening. The minimum size of analysis was 10,000
square feet in roof size, consolidating entries located at the same address. A total of thirty-three
buildings qualified for the adjusted ranking.
4.1 Overall Ranking Results
The results of the overall ranking system returned results that largely matched the
ground-truthing exercises in the field (the on-site assessment of observable variables and feel of
suitability of a rooftop community garden for each location). It was slightly unexpected that the
best buildings were located in the center of the city rather than south Seattle where the only food
deserts are present, and would appear to offer large roof sizes of warehouses. Table 4 shows the
buildings ranked in order of highest-to-lowest score according to the contrived methodology,
while Figure 5 Overall food system variable ranking system.
The total score possible for each building was 35. The results returned an average total
score (P) of 21.09 with a standard deviation of 2.6. The median score was 22. No building was
between ¼ mile and the boundary of a food desert. No building was beyond ¾ mile from a
transit station, or beyond ¾ mile from a SEC.
33
Table 4 Final food system variable ranking
ADDRESS
Roof
Size
Water
Access
Proximity
to USDA
Food
Desert
Proximity
to Food
Banks
Proximity
to Current
Community
Gardens
Proximity
to Bus
Stop
Proximity
to School/
Education
Center
Total
(out of
35)
305 HARRISON ST 98109
5 4 1 5 2 5 5 27
300 MERCER ST 98109
4 4 1 4 3 5 5 26
301 MERCER ST 98109
4 3 1 4 3 5 5 25
801 S DEARBORN ST 98134
3 4 1 5 2 5 5 25
2203 AIRPORT WAY S 98134
5 3 1 3 3 5 4 24
4201 WEST MARGINAL WAY
SW
4 3 2 1 5 4 5 24
1250 DENNY WAY 98109
3 4 1 5 1 5 5 24
400 S SPOKANE ST 98134
4 2 2 1 4 5 5 23
225 WARREN AVE N 98109
2 3 1 5 2 5 5 23
918 S LANDER ST 98134
3 3 1 4 2 5 4 22
8100 2ND AVE S 98108
2 1 3 4 3 5 4 22
1133 N 100TH ST 98133
2 3 1 2 5 5 4 22
860 TERRY AVE N 98109
2 5 1 4 2 5 3 22
800 ALOHA ST 98109
2 4 1 4 2 5 4 22
302 THOMAS ST 98109
2 2 1 5 2 5 5 22
232 1ST AVE N 98109
1 2 1 5 2 5 5 21
130 S KENYON ST
1 2 3 4 4 5 2 21
1126 N 98TH ST 98103
2 4 1 2 2 5 4 20
6605 13TH AVE S
2 1 5 2 2 5 3 20
907 NW BALLARD WAY 98107
2 4 1 3 2 5 3 20
4200 AIRPORT WAY S
2 3 2 1 2 5 4 19
1500 N 34TH ST 98103
2 3 1 3 2 5 3 19
614 NW 46TH ST 98107
2 3 1 3 2 5 3 19
12600 STONE AVE N 98133
2 3 1 1 3 5 4 19
1519 12TH AVE 98122
2 3 1 4 2 5 2 19
1300 N 130TH ST 98133
1 3 1 1 3 5 5 19
10528 5TH AVE NE
1 4 1 1 2 5 5 19
615 DEXTER AVE N 98109
1 2 1 3 3 5 4 19
1350 N 34TH ST 98103
2 2 1 3 2 5 3 18
10735 STONE AVE N 98133
2 3 1 1 3 5 3 18
1727 ALASKAN WAY S 98134
1 4 1 2 5 3 2 18
1318 N 128TH ST 98133
1 3 1 1 3 5 4 18
810 MARTIN LUTHER KING JR
WAY S
1 3 1 4 1 5 2 17
34
Figure 5 Overall food system variable ranking system (source for base map provided by
Esri)
35
4.2 Variable Ranking Results
The following section examines each individual variable used in the score system and
identifies trends in the data.
4.2.1 Roof Size
The majority of buildings (48%) scored a two for the roof size variable. The average roof
size score was 2.27, with a median score of 2, and a standard deviation of 1.14, frequency of
ranking is displayed in Figure 7. Buildings that scored a four or higher were clustered around
Seattle’s city center and downtown core. There were only two buildings that scored a 5: (1) 305
Harrison St. and (2) 2203 Airport Way S. Figure 6 shows the roof size scoring for each building.
Unexpectedly, buildings towards the center of the city had larger roof sizes. Since south
Seattle is known as a manufacturing and industrial center that is occupied with large roofed
warehouse, it was unforeseen that the larger roofs would not occupy that area of the city.
36
Figure 6 Proportional map of roof size variable ranking system
37
Figure 7 Frequency of proximity to current community garden variable ranking
4.2.2 Water Access
Figure 8 shows each buildings’ water access score. Only 860 Terry Ave N scored a 5
which is located in the Queen Anne neighborhood. The average water access score was 3.03,
with a median of 3, and a stand deviation of 0.90. The only two buildings that scored a two for
water accessibility were located in south Seattle. Figure 9 depicts the frequesncy of water access
rankings.
The results for proximity of water access were as expected due to the wealth of water
appurturanances that are located throughout the city. It is unclear why the buildings in south
Seattle were located farther away from a water appurtance, and could possibly be associated with
land use.
0
2
4
6
8
10
12
14
16
18
1 2 3 4 5
Roof Size Ranking
Frequency
38
Figure 8 Proportional map of water access variable ranking system
39
Figure 9 Frequency of water access variable ranking
4.2.3 Proximity to USDA Food Desert
Figure 10 shows the scoring of each building’s proximity to a USDA Food Desert. Only
1 building scored a five, 6605 13TH Ave S, which is located in south Seattle. The average score
was 1.33, with a median score of 1, and a standard deviation of 0.84. No buildings scored a four
for proximity to USDA Food Deserts, while an overwhelming number of buildings scored a one.
These results were exactly as expected in this analysis as the USDA food deserts only
occupy south Seattle. Buildings in south Seattle have scored relatively low in all other variables,
so the USDA food desert variable made fairly little impact on the buildings overall ranking.
0
2
4
6
8
10
12
14
16
1 2 3 4 5
Water Access
Frequency
40
Figure 10 Proportional map of proximal USDA Food Desert variable ranking system
41
4.2.4 Proximity to Food Banks
The majority of buildings were located within ¼ mile to ½ mile of a food bank. Figure 11
shows the score of each building’s proximity to food banks. The average score for proximity to a
food bank was 3.03, while the median was 3 and the standard deviation was 1.45, Figure 12
depicts such frequency of scores. Five buildings in northern Seattle scored a one from the
devised methodology. This could be due to the fact that only food banks that were located with
Seattle’s municipal boundary were examined. These buildings possibly could have ranked higher
in the scoring system if food banks beyond the municipal boundary were considered.
This was a fairly interesting outcome, and the index identified two major regions within
the city that a lacking food banks to dispense produce grown from gardens. Oddly enough, these
two areas are well known as residential areas within the city. Possibly placing a food bank closer
to the three lowest ranking buildings in south Seattle may assist in diminishing Seattle’s food
deserts.
42
Figure 11 Proportional map of proximal food bank variable ranking system
43
Figure 12 Frequency of proximity to food banks variable ranking
4.2.5 Proximity to Community Gardens
The majority (52%) of buildings scored a two, or between ¼ mile and ½ mile, for their
proximity to community gardens. Figure 13 depicts each buildings score for proximity to current
community gardens. The average score for proximity to community gardens was 2.60, with a
median score of 2, and a standard deviation of 1.01, Figure 14 depicts the frequency of each
ranking. It is important to note that this variable only ranks buildings on their proximity to a
community garden, but does not weight each garden by their food production nor if a garden is
categorized as a “Giving Garden.”
0
1
2
3
4
5
6
7
8
9
10
1 2 3 4 5
Proximity to Food Banks
Frequency
44
Figure 13 Proportional map of proximity to current garden variable ranking system
45
Figure 14 Frequency of proximity to current community garden variable ranking
4.2.6 Proximity to Bus Stop
Figure 15 shows how each building scored on its proximity to a bus stop using the
devised methodology. An overwhelming majority (93%) of buildings scored five, or were within
¼ mile of a bus stop. The average score was 4.90, with a median score of 5, and a standard
deviation of 0.38. 4201 West Marginal Way SW was the only building to score a four, while
1727 Alaskan Way S was the only building to score a three. No buildings scored a two or one
(see Figure 16).
0
2
4
6
8
10
12
14
16
18
1 2 3 4 5
Proximity to Community Garden
Frequency
46
Figure 15 Proportional map of proximity to bus stop variable ranking system
47
Figure 16 Frequency of proximity to bus stop variable ranking
4.2.7 Proximity to Schools and Education Centers
Figure 17 displays each building’s score for their proximity to a school or education
center. The average score was 3.90, with a median of 4, and standard deviation of 1.03. No
buildings were located beyond one mile from a SEC, while a majority of buildings were within
¼ of a mile to a SEC. The highest scoring buildings were centralized in north and central Seattle.
0
5
10
15
20
25
30
35
1 2 3 4 5
Proximity to Bus Stop
Frequency
48
Figure 17 Proportional map of proximity to SEC variable ranking system
49
4.3 Seattle’s “Landscape of Opportunity”
The multi-criteria analysis of the above variables, excluding roof size, identified potential
opportunities for rooftop community gardens as shown in Figure 18. The constructed map
identifies areas where the installation of a rooftop community garden might be advantageous to
the local community. South Seattle’s manufacturing and industrial center was identified as the
region that could potentially benefit the most from a community garden.
50
Figure 18 “Landscape of Opportunity” for rooftop community gardens in Seattle depicts
the possible locations for a rooftop garden, excluding the roof size variable (see Longcore et
al. 2011)
51
4.4 Ground-truthing/On-site Analysis
The top seven buildings were selected for the ground-truthing exercise. These buildings
scored a 24 or higher in the overall ranking. Overall, the results of the ground-truthing exercise
supported the results of the model in scoring a building’s potential improve food security through
installation of a community rooftop garden.
Ground truthing proved to be highly valuable in this analysis as one building was absent
in the on-site assessment versus the geospatial analysis. Geospatial data becomes out-of-date as
soon as it is gathered. This proved to be true in this analysis, as there was no building standing at
site 1250 Denny Way which will be discussed in the subsequent sections.
4.4.1 305 Harrison Street
305 Harrison St. scored the highest on the spatial index with a score of 27 out of 35. This
building is known as the Chihuly Garden and Glass Exhibition Hall in Seattle Center. The
building scored a five in roof size, proximity to food banks, bus stops, and SECs variables, while
only scoring one in proximity to USDA Food Deserts. The building was easily accessible by foot
and was recently constructed in 2012. Incidentally, the roof already houses a partial green roof;
with shrubs growing on a section of the roof (see Figure 19 and Figure 20).The building was
once used as a warehouse for Seattle Center per the metadata, offering a relatively level surface
for growing.
Although the property is centrally located within the city, specifically the Seattle Center
House, the area is more known to support tourism rather than a residential area. Community
gardens do not need to be located directly within a residential area, it may be best to recruit
gardeners to maintain the plots. Oddly enough, the only other rooftop community garden in the
city (the Up Garden) is located just two minutes walking time from 305 Harrison Street. This
52
could reveal that the placement of the Up Garden may have gone through a similar qualitative
site suitability analysis.
Figure 19 Aerial view of 305 Harrison Street
53
Figure 20 Ground view of 305 Harrison Street
4.4.2 300 Mercer Street
This building is located fairly close to 305 Harrison St. as it also occupies the Seattle
Center region. The building is a parking garage directly across the street from Seattle’s opera
house, McCaw Hall. 305 Mercer St. scored 26 out of 35, receiving primarily fours in roof size,
water access, and proximity to food banks. Again, the building’s lowest score was for its
proximity to USDA Food Deserts. The building shows great potential structure-wise for a
rooftop community garden, as the Up Garden was constructed on top of a similar parking
structure (see Figure 21 and Figure 22).
54
This parking garage could prove to be one of the better locations for the next rooftop
community gardens because of its structure type, geographic location, and accessibility by
pedestrians. This building came in a close second in the site-suitability analysis, revealing that
the spatial index could be better altered to weight variables differently than every variable
weighted equally. Located close to Seattle Center but not within in it (located on the boundary
between residential and commercial) and on multiple bus lines, the roof of the garage can be
easily accessed by the community. The produce grown from the garden could easily be
transported to multiple food banks within the area, and could provide societal benefits to the
surrounding schools.
Like the 305 Harrison Street, the only major downfall of this building is its proximity to
food deserts. Being over 1 mile away from a food desert is undesirable when trying to eradicate
food insecurities within the city. Although the building is some distance away from the USDA
food deserts, if there was an effective means of transport to get food from the northern city
gardens to the south, then 300 Mercer Street may be a viable building for a new rooftop
community garden.
55
Figure 21 Aerial view of 300 Mercer Street
56
Figure 22 ground view of 300 Mercer Street
4.4.3 301 Mercer Street
The Phelps Center (301 Mercer St.) is located diagonally from 300 Mercer St. The center
is houses the Pacific Northwest Ballet. The building is easily accessible by foot however exhibits
a unique rolling rooftop geometry (see Figure 23 and Figure 24). Overall, the building has great
potential for housing a rooftop community garden that could easily be produce a large amount of
vegetation which could be conveniently accessed and utilized by the public.
Being so close in proximity to 300 Mercer Street, the property shared similar scores using
the spatial index; however ground truthing revealed that this building may not be as suitable for a
rooftop community garden as 300 Mercer Street. Although rooftop garden installations can be
57
customized to each roof, the uneven surface of 301 Mercer Street may prove to be more difficult
than the others, while possible costing more. This increase in cost could be a deterrent for
selecting this site for a new garden.
Figure 23 Aerial view of 301 Mercer Street
58
Figure 24 Aerial view of 301 Mercer Street displaying the roof’s rolling geometry
Photography by Public47 Architects
4.4.4 801 South Dearborn Street
This building is located in south Seattle and is known as an emissions testing center.
Although the building scored high for the proximity to SECs and bus stops, the property did not
seem open to the general public due to a chain link fence and barbwire surrounding the perimeter
(see Figure 25 and Figure 26).
59
Figure 25 Aerial view of 801 South Dearborn Street
60
Figure 26 Ground view of 801 South Dearborn Street
Despite the hostile exterior, the building houses a fairly favorable rooftop for a garden;
with a relatively flat surface, railing along the perimeter, and plenty of sun exposure (see Figure
27). The slope of the roof and already present railings would allow for easy setup of the garden.
The building was easily accessible by bus, and located in very close proximity to a food bank,
making it a prime location in regards to accessibility and utilization. Alas, if the public is unable
to access the rooftop, not just the building, then a building is considered unfavorable for a
community rooftop garden.
61
Figure 27 View of 801 South Dearborn Street from Interstate 5
4.4.5 2203 Airport Way South
Located in the industrial section of Seattle, 2203 Airport Way S. scored a 24 out of 35
total points. This property scored the most points with roof size and proximity to bus stops, while
scoring the lowest with a one for the proximity to USDA Food Deserts. The building was easily
accessible by foot, however showed a fairly ‘cluttered’ rooftop, potentially reducing usable
rooftop space for growing. Figure 28 is an aerial image of 2203 Airport Way South.
Examining aerial images, the roof shows a fair inhospitable rooftop for a garden due to
the varying levels of rooftop debris. The debris can affect the surface area used, ultimately
discrepancies in available roof size and usable roof size, such that usable is the substrate on
which the plants would grow. The rooftop of 2203 Airport Way South would therefore have
minimal usable roof area.
62
Figure 28 Aerial view of 2203 Airport Way South
63
4.4.6 4201 West Marginal Way Southwest
The Seattle Parks & Recreation warehouse occupies 4201 W Marginal Way SW (see
Figure 29 and Figure 30). This building scored 24 out of 35 total points, scoring the highest in
proximity to community gardens and SECs, while scoring a one in proximity to a food bank. The
aerial imagery reveals a continuous flat rooftop that could allow for easy garden installation.
This was the only property out of the top seven buildings from the on-site assessment that scored
above a one for the proximity to a USDA Food Desert. This is the most western roof in South
Seattle, being one of the only rooftops that could supply fresh produce to the prone to USDA
food desert south Seattle.
Although 4201 West Marginal Way Southwest was ranked sixth in the index, the
building is a Seattle Parks and Recreation office, which could make the property more ideal for a
rooftop community garden as they passed the Parks and Green Space Levy in 2008 that
encouraged the development of new P-Patches (Seattle Department of Neighborhoods 2014).
Building a rooftop community garden at this location could not only assist with the reduction in
food insecurities, but also be a means to promote these rooftop community gardens for social
capitalism by being a demo garden.
64
Figure 29 Aerial view of 4201 West Marginal Way Southwest
65
Figure 30 Ground view of 4201 West Marginal Way Southwest
4.4.7 1250 Denny Way
This was the outlier in the ground-truthing exercise, as there was a vacant lot where the
building should have been standing (see Figure 31 and Figure 32). The parcel is currently owned
by Seattle City Light (SCL) and is undergoing construction for a new electrical substation. As a
former Greyhound Bus Maintenance Facility, the parcel has undergone decontamination over the
past 2 years, however, because of its prior use, the on-the-ground gardens should not be utilized.
The Denny Substation Project has recently received approval from the Seattle Design
Commission and is currently developing the Final Environment Impact Statement.
66
Although there is not currently a building occupying the parcel, the property scored fairly
well on other variables such as water access, proximity to food banks, bus stops, and SECs. Due
to the other high scoring variables, other than roof space, the future building could prove to be an
ideal location for a rooftop community garden. The Denny Substation Project promotes a
“substation that will fit with the neighborhood, advance community goals, and meet the City's
goals for sustainability” (Seattle City Light 2014), which could mean that since the project is still
in its design phase, that Seattle City Light would be accommodating to incorporating rooftop
community gardens into their design.
67
Figure 31 Aerial view of 1250 Denny Way
68
Figure 32 Ground view of 1250 Denny Way
69
CHAPTER 5: DISCUSSION AND CONCLUSION
This site-suitability analysis appeared to be effective utilizing food systems index as an indicator
for rooftop community garden planning in conjunction with multiple variables such as physical,
social, and economical variables adapted for urban agriculture. Satisfactory buildings fitting the
set of devised criteria were identified and ranked accordingly therefore considering this project a
success. The spatial index provided a well-rounded profile for each building, allowing for
selection with preference of varying criteria priority.
5.1 Key Observations
Results in this study suggest that ideal buildings for rooftop community gardens are
localized in the center of the city even though Seattle’s food deserts are primarily located in
south Seattle. By ranking buildings using a multivariable spatial index, buildings were better
identified when Seattle’s food system was measured holistically. The benefits of using a spatial
index allows for community members, planners, and government officials to weight for variables
that are most desirable when determining where to place a rooftop community garden. This
analysis weighted all variables equally as to examine if the food system was lacking in any areas
of accessibility, utilization, or availability.
Higher ranking buildings from the index notably occupied the higher valued regions from
the “Landscape of Opportunity” map. The manufacturing and industrial center appeared to have
a disproportionate number of buildings located in that region compared to other sections of the
city. This could be due to the larger rooftops associated with warehouses that are currently being
underutilized. Although, it may appear that not many people would be living in these regions, it
could house a large homeless population whom are in most need of an accessible food source. It
is important to note that Seattle’s only USDA food deserts are located in this region. Building a
70
rooftop community garden could be helpful if a cost benefit analysis was completed.
As there have been minimal GIS studies on defining rooftop community gardens to
improve food insecurities, this paper did well by drawing upon previous food systems, urban
agriculture and green roof papers. Defining a food system by availability, accessibility, and
utilization worked well for this study because it allowed for the customization of geospatial
variables unique to Seattle. This ultimately provides a good basis for other studies to select their
own variables unique to their city with different weighted values.
Variables in this study appeared to provide an accurate reading of the urban food system
therefore able to identify areas lacking in food security. Although the methodology was applied
to the City of Seattle, the spatial index may be applied to other cities by encouraging a new
selection of variables. A common variable that is used for site suitability analyses of green roofs,
stability, was purposely omitted from this study as this is a variable that can truly be measured by
an engineer per Seattle’s city planning requirements (see Appendix A). The overall food system
variable ranking methodology was sound in assessing variables and can readily be used to
effectively select suitable buildings to improve food security.
The ground-truthing and on-site assessment provided confirmation of the suitability
ranking of each building. This provided a qualitative observation of the variables to back up the
spatial data that was used for remote sensing. In the case of 1250 Denny Way, this exercise
proved to be an invaluable asset, as there was no building standing.
5.2 Contrasts with Prior Studies
The clustering of higher ranked buildings located in the city’s downtown core suggest
that regardless of structural potential to support a rooftop garden, Seattle’s food system may
better support community rooftop gardens in unforeseen regions within the city. The “Landscape
71
of Opportunity” map provides insight to this consideration. This analysis improves upon
previous analyses by Tian and Jim (2012) and Berger (2013) that take more of a one dimensional
approach at locating ideal buildings for rooftop gardens where the main determinants are
building area and stability. Stability, being the hardest variable to measure remotely, was
purposely excluded from this analysis for that very reason, as well as an onsite assessment by an
engineer is required prior to installing a garden.
This study ranked buildings for ideal rooftop community garden locations specific to
Seattle’s food system, discussing the intricacies associated with how urban food systems are
measured Similar to Oulton’s (2012) analysis, the spatial index allowed for a quantitative
approach to a fairly fluid selection of variables. If the variables can be justified to best suit the
city’s specific food system, then the ranking system can accurately inform urban planners of
insecurities within their system.
This analysis only examined on type of urban agriculture, rooftop community gardens.
The APA, City of Seattle, and many other organizations define other categories of UA, which
require different conditions to thrive within a city which may not be best suited by this analysis.
Similar to Dongus & Drescher (2000), this study really only examined the intra-urban areas of
the city rather than the peri-urban. This did inhibit the study from examining food banks,
community gardens, and USDA food deserts beyond Seattle’s municipal boundary, which may
ultimately influence a buildings score. A possible resolution to this analysis would be to
complete a network analysis of sorts on the flow of production to consumption. By first
identifying the movement of food, a more customized boundary and scale of analysis can be
completed.
Overall, this site-suitability analysis did a satisfactory job at representing Seattle’s urban
72
food system by utilizing aspects of food availability, accessibility, utilization as described in
Gregory, Ingram, and Brklacich (2005). By first understanding the components of an urban food
system, researchers will then be able to selectively define a particular city’s geospatial food-
scape based off of available data and important factors desired by community members and
officials. This could then be applied to not only rooftop gardens, but on the ground gardens, solar
panels to provide optimal energy outputs and green roofs to reduce the heat-island effect, all by
first identifying the geospatial ecosystem to identify areas of need, effective regions, and
ultimately optimal placement.
5.3 Recommendation for Future Research
Similar to Oulton’s (2012) spatial index, based on physical and social variables, this basic
methodological framework can provide the means to easily rank parcels or buildings according
to the research goals. Weighting the variables by importance to city planners and community
members could provide a more tailored approach to scoring a buildings potential by a
community’s ‘wants’ verses a community’s ‘needs.’
As this study was completed in a community garden-friendly city with a well-developed
transit system, food system variables could easily be selected for. It would be interesting to first
apply the same spatial index to another dense urban city, but could this index be used to assess
food insecurities in a variety of population regions? It is also important to note that although a
garden may not be located in a food desert, that a community could not be experiencing food
insecurities. Future research could possible examine what threshold of food security must a
community reach before it redirects its produce to the next community in need.
Even though the 1250 Denny Way site currently did not have a building occupying that
address, this parcel shows the greatest potential for incorporating a rooftop community garden
73
into the future building design. Referring back to Table 4, this property scored highest in food
accessibility and utilization variables rather than roof size. These variables may be more difficult
to change, and therefore could be valued or weighted more than roof size in future analyses.
Stability has been proven to be one of the more difficult variables to define in a
geospatial analysis. If there were a better way to define roof stability in such an analysis, this
could greatly assist urban planners not only for rooftop community gardens, but solar panel,
turbine, and other alternative food and energy sources in crowded urban dwellings. Other
variables not included in this analysis such as rooftop characteristics like stability, material,
slope, wind, sunlight, temperature, and number of building stories, could in future analysis,
greatly strengthen this study.
In addition to this site-suitability analysis, a cost-benefit analysis would also be another
area of future research. By estimating and examining the strengths and weaknesses of each
rooftop’s ability to positively affect the community’s food system. Determining the actual cost of
installing one of these rooftop gardens, fiscal and otherwise, should ultimately be outweighed by
the benefit to the community and the community’s food system.
This study was able to identify buildings to house a rooftop community garden by
defining a spatial index representing Seattle’s urban food system which may be altered and
weighted depending upon the city. To expand upon this analysis, the introduction of new
variables could help assist city planners by calculating how much these rooftop community
gardens could impact a regions food system. Defining an urban food system is much like
defining an ecosystem, where the alteration of one variable could greatly affect the interpretation
of the geospatial food-scape. With this analysis, ground-truthing proved to be an invaluable tool
74
to assessing the efficacy of the spatial index. If possible, it would be impressive if this step could
be eliminated for such an analysis.
75
REFERENCES
Bailkey, Martin, and Joe Nasr. 1999. From brownfields to greenfields: Producing food in North
American cities. Community Food Security News. Fall 2000: 6.
Berger, Danielle. 2013. A GIS suitability analysis of the potential for rooftop agriculture in New
York City.
Boko, Michel, Isabelle Niang, Anthony Nyong, Coleen Vogel, Andrew Githeko, Mahmoud
Medany, Balgis Osman-Elasha, Ramadjita Tabo, and Pius Yanda. 2007. Climate change
2007: Impacts, adaptation and vulnerability. Contribution of Working Group II to the Fourth
Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge
University Press, Cambridge UK: 433-67.
Colasanti, Kathryn J. A. 2009. Growing food in the city: Two approaches to exploring scaling up
urban agriculture in Detroit.
Cronin, Brenda. 2013. Seattle a leader in job growth. Dow Jones & Company, Inc., 25 August
2013.
http://online.wsj.com/news/articles/SB10001424127887323423804579024702316622012
Deelstra, Tjeerd, and Herbert Girardet. 2000. Urban agriculture and sustainable cities. Bakker N.,
Dubbeling M., Gündel S., Sabel-Koshella U., De Zeeuw H.Growing Cities, Growing
Food. Urban Agriculture on the Policy Agenda.Feldafing, Germany: Zentralstelle Für
Ernährung Und Landwirtschaft (ZEL): 43-66.
Dongus, Stefan, and Axel W. Drescher. 2006. The use of GIS, GPS and aerial imagery for
mapping urban agricultural activities on open space in cities. GIS Development Weekly 20.
Dunn, Richard A. 2010. The effect of fast-food availability on obesity: An analysis by gender,
race, and residential location. American Journal of Agricultural Economics: aaq041.
76
Elwood, Sarah A. 2002. GIS use in community planning: A multidimensional analysis of
empowerment. Environment and Planning A 34 (5): 905-22.
EPA. 2010. Final Remedial Investigation Report, Environmental Protection Agency. Last
accessed 28 September 2014.
http://www.epa.gov/region10/pdf/sites/ldw/remedial_investigation_7-9-2010.pdf.
ESRI. 2014. GIS Dictionary: Ground Truth. Last Accessed 16 December 2014.
http://support.esri.com/en/knowledgebase/GISDictionary/term/ground%20truth.
FAO. 1996. Report of the World Food Summit. FAO. November 13-17.
http://www.fao.org/docrep/003/w3548e/w3548e00.htm.
Forbes. 2013. America’s 20 fastest growing cities. Forbes.com. Accessed 30 November 2013
http://www.forbes.com/pictures/mhj45mhlf/6-seattle-wa/.
Francis, Charles, G. Lieblein, S. Gliessman, TA Breland, N. Creamer, R. Harwood, L.
Salomonsson, J. Helenius, D. Rickerl, and R. Salvador. 2003. Agroecology: The ecology
of food systems. Journal of Sustainable Agriculture 22 (3): 99-118.
Freedman, Darcy A., and Bethany A. Bell. 2009. Access to healthful foods among an urban food
insecure population: Perceptions versus reality. Journal of Urban Health 86 (6): 825-38.
Greene, Richard P., and James B. Pick. 2011. Exploring the urban community: A GIS approach.
Pearson Prentice Hall.
Greene, Jay. 2013. Warehouse growth reflects Amazon’s evolving tax strategy. The Seattle
Times. 29 October 2013.
Gregory, Peter J., John Ingram, and Michael Brklacich. 2005. Climate change and food security.
Philosophical Transactions of the Royal Society of London.Series B, Biological Sciences
360 (1463) (Nov 29): 2139-48.
77
Hodgson, Kimberley, Marcia Caton Campbell, and Martin Bailkey. 2011. Urban agriculture:
Growing healthy, sustainable places. American Planning Association.
Hou, Jeffrey, Julie Johnson, and Laura J. Lawson. 2009. Greening cities, growing communities:
Learning from Se att le’ s urban community gardens. Landscape Architecture Foundation.
Kisner, C. 2008. Green roofs for urban food security and environmental stability. Climate
Institute. Accessed 16 December 2014. http://www.climate.org/topics/international-
action/urban-agriculture.htm.
Longcore, Travis, Christine Lam, Mona Seymour, Alina Bokde. 2011. “LA Gardens: Mapping
to Support a Municipal Strategy for Community Gardens.” University of Southern
California GIS Research Laboratory Research Report. Accessed 1 December 2014.
http://spatial.usc.edu/wp-content/uploads/2014/03/LAGardens-Research-Report.pdf.
Lovell, Sarah Taylor. 2010. Multifunctional urban agriculture for sustainable land use planning
in the United States. Sustainability 2 (8): 2499-522.
Macdonald, Rich. 2014. P-Patch Community Gardens. Seattle Department of Neighborhoods.
Accessed 28 September 2014.
http://www.seattle.gov/neighborhoods/ppatch/aboutPpatch.htm.
Morton, Lois Wright, and Troy C. Blanchard. 2007. Starved for access: Life in rural America’s
food deserts. Rural Realities 1 (4): 1-10.
Mougeot, Luc JA. 2006. Growing better cities: Urban agriculture for sustainable development.
IDRC.
78
National Resources Conservation Service. 2014. Community Garden Guide Vegetable Garden
Planning and Development. Accessed 28 September 2014.
http://www.nrcs.usda.gov/Internet/FSE_PLANTMATERIALS/publications/mipmcot940
7.pdf.
Opfer, Pamela R. 2010. Using GIS technology to identify and analyze ‘Food deserts’ on the
southern Oregon coast.
Oulton, Allison. 2012. Community Gardens for Social Capital: A Site Suitability Analysis in
Akron, Ohio.
Parks, Kevin. 2012. Community gardening has long history in the U.S. This Week Community
News. 20 October 2012. http://extension.missouri.edu/p/MP906-4.
Rahman, Syumi Rafida Abdul, and Hamidah Ahmad. 2012. Green roofs as urban antidote: A
review on aesthetic, environmental, economic and social benefits. Sixth South East Asian
Technical Consortium in King Mongkut University of Technology Thonbur, Bangkok,
Thailand.
Randall, Todd A., Cameron J. Churchill, and Brian W. Baetz. 2003. A GIS-based decision
support system for neighbourhood greening. Environment and Planning B 30 (4): 541-64.
RUAF. 2010. The growth of cities in East-Africa: consequences for urban food supply. RUAF
Foundation for the World Bank. Accessed 10 August 2014.
http://www.ruaf.org/sites/default/files/The%20growth%20of%20cities%20in%20East%2
0Africa--FINAL-%2020-12-2010%20_2__1.pdf.
Schukoske, Jane E. 1999. Community development through gardening: State and local policies
transforming urban open space. NYUJ Legis & Pub. Policy 3 : 351.
79
Seattle. gov. 2014. Seattle Department of Neighborhoods. 2014. P-Patch community gardens.
Accessed 26 November 2013.
http://www.seattle.gov/neighborhoods/ppatch/aboutPpatch.htm.
Seattle City Light. 2014. Denny Substation Project. Seattle City Light. Accessed 27 November
2014. http://www.seattle.gov/light/dennysub/substation.asp.
Seattle Green Factor. 2014. The Seattle Green Factor. Seattle Department of Planning and
Development. Accessed 13 March 2014.
https://www.seattle.gov/DPD/cityplanning/completeprojectslist/greenfactor/whatwhy/def
ault.htm.
Seattle Public Utilities. 2014. Obtain or Upgrade Water Service. Accessed 13 October 2014
http://www.seattle.gov/util/MyServices/Water/WaterService/index.htm.
Seattle Urban Farm Company. 2013. The Seattle Urban Farm Co. Accessed 17 November 2013.
http://www.seattleurbanfarmco.com/.
Tian, Yuhong, and CY Jim. 2012. Development potential of sky gardens in the compact city of
Hong Kong. Urban Forestry & Urban Greening 11 (3): 223-33.
UNCHS. 2001. Urbanization. United Nations Environment Programme. Accessed 18 November
2014 http://www.unep.org/geo/geo3/english/403.htm.
UNEP. 2011. The Urban Environment. United Nations Environment Programme Regional Office
for North America. Accessed 18 November 2014.
http://www.rona.unep.org/cities/index.html.
UNFPA. 2014. UN Llaunches ICPD beyond 2014 global review report. United Nations
Population Fund. Accessed 18 November 2014.
http://icpdbeyond2014.org/uploads/browser/files/icpd_global_review_report.pdf.
80
University of Missouri, Extension. 2009. Community gardening toolkit, the history of
community gardening.” Accessed 14 November 2013.
http://extension.missouri.edu/p/MP906-4.
USDA. 2014. Food Deserts. USDA. http://www.ers.usda.gov/data-products/food-access-
research-atlas/download-the-data.aspx#.U0LYmPldWSo.
Wilkinson, Renee. 2012. Equal access: Providing urban agricultural benefits to under-served
communities.
Wrigley, Neil. 2002. 'Food deserts' in British cities: Policy context and research priorities. Urban
Studies 39 (11): 2029-40.
81
APPENDIX A
82
APPENDIX B
Abstract (if available)
Abstract
As urbanization has increased in recent decades, urban food systems have become stressed, reducing food security (Gregory, Ingram, and Brklacich 2005). Community gardens that occupy a city’s vacant lots have been known to combat food insecurity (Oulton 2012, Colasanti 2009), but many compact cities lack space to garden. One solution has been the development of rooftop gardens (Tian and Jim 2012). In recent decades, Seattle, Washington has increased the number of community gardens, but like many urbanizing centers, the city lacks vacant lots for gardening. With limited ground availability in Seattle and an ever increasing demand to expand upon the city’s community garden program, otherwise known as P-Patches, to combat this rapid expansion and improve food security, the city has started to become more creative with its urban spaces through activities such as rooftop gardens (Forbes 2013, Cronin 2013, Greene 2013, Seattle.gov 2014). The goals of this thesis are the following: (1) determine criteria to represent Seattle’s food system in a site-suitability analysis to improve food security
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Community gardens for social capital: a site suitability analysis in Akron, Ohio
PDF
Developing and implementing a GIS-based framework to identify optimal locations for clean water wells in sub-Saharan Africa
PDF
Disparities in food access: an empirical analysis of neighborhoods in the Atlanta metropolitan statistical area
PDF
Does the Bay Area have a social center? Delimiting the postmodern urban center of the San Francisco Bay Area
PDF
A geospatial analysis of income level, food deserts and urban agriculture hot spots
PDF
Measuring seasonal variation in food access: a case study of Everett, Washington
PDF
Monitoring parks with inexpensive UAVs: cost benefits analysis for monitoring and maintaining parks facilities
PDF
A comparison of urban land cover change: a study of Pasadena and Inglewood, California, 1992‐2011
PDF
Finding food deserts: a study of food access measures in the Phoenix-Mesa urban area
PDF
Calculating solar photovoltaic potential on residential rooftops in Kailua Kona, Hawaii
PDF
Utilizing 311 service requests as a signature of urban location in the City of Los Angeles
PDF
Use of GIS for analysis of community health worker patient registries from Chongwe district, a rural low-resource setting, in Lusaka Province, Zambia
PDF
Using Landsat and a Bayesian hard classifier to study forest change in the Salmon Creek Watershed area from 1972–2013
PDF
Effect of spatial patterns on sampling design performance in a vegetation map accuracy assessment
PDF
Estimating populations at risk in data-poor environments: a geographically disaggregated analysis of Boko Haram terrorism 2009-2014
PDF
A Maxent-based model for identifying local-scale tree species richness patch boundaries in the Lake Tahoe Basin of California and Nevada
PDF
Selection of bridge location over the Merrimack River in southern New Hampshire: a comparison of site suitability assessments
PDF
Urban green space accessibility and environmental justice: a GIS-based analysis in the city of Phoenix, Arizona
PDF
Site suitability analysis: small-scale fixed axis ground mounted photovoltaic power plants in Fresno, CA
PDF
Obesity and healthy food accessibility: case study of Minnesota, USA
Asset Metadata
Creator
Stoudt, Ana E.
(author)
Core Title
Redefining urban food systems to identify optimal rooftop community garden locations: a site suitability analysis in Seattle, Washington
School
College of Letters, Arts and Sciences
Degree
Master of Science
Degree Program
Geographic Information Science and Technology
Publication Date
01/20/2015
Defense Date
11/18/2014
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
GIS,OAI-PMH Harvest,urban food systems
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Warshawsky, Daniel N. (
committee chair
), Lee, Su Jin (
committee member
), Longcore, Travis R. (
committee member
)
Creator Email
ana.e.stoudt@gmail.com,stoudt@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-523594
Unique identifier
UC11297714
Identifier
etd-StoudtAnaE-3121.pdf (filename),usctheses-c3-523594 (legacy record id)
Legacy Identifier
etd-StoudtAnaE-3121.pdf
Dmrecord
523594
Document Type
Thesis
Format
application/pdf (imt)
Rights
Stoudt, Ana E.
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
Tags
GIS
urban food systems