University of Southern California Dissertations and Theses

Urban green space accessibility and environmental justice: a GIS-based analysis in the city of Phoenix, Arizona

(USC Thesis Other)

Urban green space accessibility and environmental justice: a GIS-based analysis in the city of Phoenix, Arizona

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Urban Green Space Accessibility and Environmental Justice:
A GIS-Based Analysis in the City of Phoenix, Arizona

by

Shuk Wai So

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)

August 2016

Copyright ® 2016 by Shuk Wai So
iii

Table of Contents
List of Figures ................................................................................................................................ vi
List of Tables ............................................................................................................................... viii
Acknowledgements ........................................................................................................................ ix
List of Abbreviations ...................................................................................................................... x
Abstract .......................................................................................................................................... xi
Chapter 1 Introduction .................................................................................................................... 1
1.1 Why Urban Green Space Access Matters ........................................................................... 1
1.2 Existing Research Gaps ...................................................................................................... 2
1.3 Objective ............................................................................................................................. 3
1.4 Study Area .......................................................................................................................... 4
1.5 Organizational Framework ................................................................................................. 6
Chapter 2 Background and Literature Review................................................................................ 8
2.1 Environmental Justice ......................................................................................................... 8
2.1.1 Defining Environmental Justice .................................................................................... 9
2.1.2 Environmental Justice in Phoenix ............................................................................... 10
2.2 Urban Green Space ............................................................................................................ 10
2.3 Access to Green Spaces ..................................................................................................... 12
2.4 Environmental Justice, Urban Green Space, and GIS ....................................................... 13
2.4.1 Methodologies for Studying Urban Green Space Accessibility ................................. 13
2.4.2 Buffer Approach to Measure Urban Green Space ...................................................... 14
2.4.3 Network Analysis Approach to Measure Urban Green Space .................................... 16
2.4.4 Comparing Two Approaches to Measure Urban Green Space ................................... 17
iv

Chapter 3 Methodology ................................................................................................................ 18
3.1 Study Area ........................................................................................................................ 19
3.2 Data Sources And Description Of Spatial Datasets .......................................................... 20
3.3 Data Preparation................................................................................................................ 23
3.3.1 Digitizing Spatial Data ................................................................................................ 23
3.3.2 Demographic and Population Data ............................................................................. 25
3.4 Network Analysis.............................................................................................................. 27
3.4.1 Park Service Areas Determination .............................................................................. 27
3.4.2 Green Space Provision Enhancement ......................................................................... 32
Chapter 4 Results .......................................................................................................................... 35
4.1 Overall Park Service Areas .............................................................................................. 35
4.2 Green Space Access Results ............................................................................................ 36
4.2.1 Green Space Access: White Population ..................................................................... 36
4.2.2 Green Space Access: Black Population ..................................................................... 39
4.2.3 Green Space Access: Asian Population ..................................................................... 42
4.2.4 Green Space Access: Hispanic Population ................................................................ 45
4.2.5 Green Space Access: American Indian Population ................................................... 48
4.2.6 Green Space Access: Overall ..................................................................................... 51
4.3 Green Space Provision Enhanced Results ........................................................................ 51
4.3.1 Green Space Provision Enhanced: White ................................................................. 52
4.3.2 Green Space Provision Enhanced: Black .................................................................. 53
4.3.3 Green Space Provision Enhanced: Asian .................................................................. 54
4.3.4 Green Space Provision Enhanced: Hispanic ............................................................. 56
v

4.3.5 Green Space Provision Enhanced: American Indian ................................................... 57
4.3.6 Green Space Provision Enhanced: Overall .................................................................. 58
Chapter 5 Discussion And Conclusion ......................................................................................... 67
5.1 Summary of Results ......................................................................................................... 67
5.1.1 First Analysis .......................................................................................................... 67
5.1.2 Second Analysis ...................................................................................................... 67
5.2 Significance of Findings .................................................................................................. 68
5.3 Limitations ....................................................................................................................... 70
5.4 Future Research ............................................................................................................... 70
References ..................................................................................................................................... 74

vi

List of Figures
Figure 1 Location of the City of Phoenix. ...................................................................................... 5
Figure 2 Input features and buffers (Flater 2011) ......................................................................... 15
Figure 3 Summary of workflow .................................................................................................... 19
Figure 4 Manually digitized park access points ............................................................................ 24
Figure 5 Park access points within the City of Phoenix ............................................................... 25
Figure 6 Park service area polygons based on park access points ................................................ 28
Figure 7 An example of clipped parcels ....................................................................................... 29
Figure 8 An example of dissolved parcel boundaries ................................................................... 31
Figure 9 An example of future green space location study area using erase and dissolve tool .... 33
Figure 10 Percentage of White population that can access green space within 0.25 and 0.5 mile
service areas .................................................................................................................................. 37
Figure 11 Number of White people that can access green space within 0.5 mile service areas ... 38
Figure 12 Percentage of Black population that can access green space within 0.25 and 0.5 mile
service area.................................................................................................................................... 40
Figure 13 Number of Black people that can access green space within 0.5miles ........................ 41
Figure 14 Percentage of Asian population that can access green space within 0.25 and 0.5 mile
service areas .................................................................................................................................. 43
Figure 15 Number of Asian people that can access green space within 0.5 miles ....................... 44
Figure 16 Percentage of Hispanic population that can access green space within 0.25 and 0.5
mile service areas .......................................................................................................................... 46
Figure 17 Number of Hispanic people that can access green space within 0.5 miles .................. 47
vii

Figure 18 Percentage of American Indian population that can access green space within 0.25 and
0.5 mile service areas .................................................................................................................... 49
Figure 19 Number of American Indian people that can access green space within 0.5 miles ...... 50
Figure 20 Comparing percentage of each racial group able to access public green space at 0.25
and 0.5 miles ................................................................................................................................. 51
Figure 21 Future possible green space locations: White .............................................................. 53
Figure 22 Future possible green space locations: Black ............................................................... 54
Figure 23 Future possible green space locations: Asian ............................................................... 55
Figure 24 Future possible green space locations: Hispanic .......................................................... 57
Figure 25 Future possible green space locations: American Indian ............................................. 58
Figure 26 Park pressure result (White) ......................................................................................... 60
Figure 27 Park pressure result (Black).......................................................................................... 61
Figure 28 Park pressure result (Asian).......................................................................................... 62
Figure 29 Park pressure result (American Indian) ........................................................................ 63
Figure 30 Park pressure result (Hispanic) ..................................................................................... 64
Figure 31 Overall park pressure result .......................................................................................... 66

viii

List of Tables
Table 1 Data used in this study ..................................................................................................... 20
Table 2 Attributes of Park Boundary Layer .................................................................................. 21
Table 3 Number of Parks of different park types ......................................................................... 21
Table 4 Demographic attribute definition ..................................................................................... 26
Table 5 Scoring system ................................................................................................................. 32
Table 6 Scoring system ................................................................................................................. 34
Table 7 Summary of Park Service Areas ...................................................................................... 36
Table 8 Summary of the White Population................................................................................... 37
Table 9 Summary of the Black Population ................................................................................... 39
Table 10 Summary of the Asian Population ................................................................................. 42
Table 11 Summary of the Hispanic Population ............................................................................ 45
Table 12 Summary of the American Indian Population ............................................................... 48

ix

Acknowledgements
I thank my committee chair, Dr. Daniel Warshawsky for the help, guidance, and encouragement
during the many stages of this project. I would also like to thank my committee members, Dr.
Elisabeth Sedano and Dr. Laura Loyola for all of their support.

x

List of Abbreviations
ANGSt Accessible Natural Greenspace Standards
GIS Geographic Information System
NRPA National Recreation and Parks Association

xi

Abstract
Research studies show that urban green spaces promote physical activity, health of urban
residents, and psychological well-being. However, most public urban green space is not
distributed equally and fairly. In addition, access to public green space is often stratified based
on race and income level. The objective for this research is to assess the level of environmental
justice in the city of Phoenix, Arizona, and to answer the following questions: 1) how accessible
are public parks or green spaces within a walking distance of 0.5 miles for White, Black, Asian,
Hispanic, and American Indian populations; and 2) which areas need more public green spaces
or parks? The accessibility of public green space refers to the distance travelled from a
residential area to the nearest public green space. This study utilizes network analysis to
investigate how accessible public parks or green spaces are to residents of the City of Phoenix,
categorize by race, and which areas need more public green space in the City of Phoenix. A
geodatabase from the US Census Bureau with pre-defined shapefiles and demographic data, as
well as city parcel shapefiles from the City of Phoenix Open Data Portal are combined using the
intersect tool in ArcMap. Results show that the White population does not have a higher
percentage that live nearby public green space. The Asian population has the lowest public green
space accessibility and the Hispanic population has the highest public green space accessibility,
but also the highest park pressure. According to the future possible green space locations
analysis and park pressure analysis, the demand of public green spaces for Whites and Hispanic
people are the highest as compared to other groups. Given these research findings, this study
suggests that geospatial analysis should be utilized in future environmental justice scholarship.

1

Chapter 1 Introduction
Urban green spaces, by definition, are open spaces in urban areas that are primarily
covered by vegetation which can be public or private (Baycan-Leven et al. 2002). Using this
definition, urban green space can include parks, community gardens, natural reserves, golf
courses, and forests. In this study, only public urban green spaces are being studied because these
public green spaces are free of charge and most people are not able to access private green
spaces such as golf courses. Urban green space access, in this research, refers to the distance
travelled from a residential area to the nearest public green space. There is a growing amount of
research on park access using Geographic Information Systems (GIS) to study environmental
justice (Comber et al. 2008; Coutts et al. 2010; Sister et al. 2007). The research herein uses GIS
network analysis to analyze how accessible public parks or green spaces are for different racial
populations and to find out which areas need to have more public green spaces in the City of
Phoenix.
1.1 Why Urban Green Space Access Matters
Access to urban green space can have a great influence on human health. Research shows
that urban green spaces promote physical activity, improve the general public health of urban
residents, and enhance psychological well-being (Wolch et al., 2014). In addition to these
important components, Mitchell and Popham(2007) argue that people who live near green spaces
are healthier than people who live farther from green spaces.
Parks are one form of public green space in an urban setting. People can relax or exercise
at parks, and parks help improve the environmental and air quality in a dense urban development
(Parsons 2015) because parks usually have vegetation. However, loss of natural landscape and
2

green space due to rapid urbanization is occurring and might be detrimental to human health
(Coutts et al. 2010). To ensure green spaces are distributed equitably within a city, a measuring
system needs to be developed and implemented.
1.2 Existing Research Gaps
Previous studies have shown that in the United States people of color, typically live in the
urban core where public green space is scarce and poorly maintained (Heynen et al. 2006; Wolch
et al. 2014). Therefore, race and ethnicity have become major factors in planning urban land use.
To improve this environmental inequality, it is necessary to identify how serious the problem is
in different communities.
Network analysis within a GIS can calculate how much time is needed to travel from one
location to another. Previous studies use GIS and network analysis to determine how different
socio-economic groups, ethnic groups, and religious groups access urban green space in the
United Kingdom; their work shows that access to green space is uneven amongst different
groups (Comber et al. 2008; Kuta et al. 2014). Chapter 2 will discuss the details of the above
examples and their methodologies.
The reason that many research studies use GIS to perform environmental justice analysis
is because it can effectively solve different social issues after identifying the possible issues.
Future planners should pay attention to the importance of equal access to green space, because
all people living within a city deserve equal access to public green spaces. In addition, research
studies can increase the awareness of using geographic information sciences so that more
scholars can use it to address different types of social or environmental issues. This research will
contribute to scientific knowledge because it is the first to use network analysis to study the
differential accessibility of public urban green space, based on race, in the city of Phoenix, AZ.
3

Network analysis allows urban planners, landscape architects, and the government to understand
how environmental justice affects cities and to help communities have more equitable access to
healthier living environments (Sister et al. 2010). In addition, network analysis can be a new
methodology for urban planners to analyze the existing neighborhoods that are in need of
renewal. In spite of the fact that the chance of urban renewal or reconstruction of existing cities
tends to be limited, this study can produce a model for new urban developments based on
socially equitable access to public green spaces.
There are two common measurement techniques that are used to study accessibility:
Euclidean and network analysis. Most planning authorities use Euclidean distance, also called
straight line distance, to measure accessibility (Coutts et al. 2010; Coutts et al. 2013; Moseleya et
al. 2013), but this technique is over-simplifies the real world because it does not account for
barriers to movement across city space. In contrast, network analysis is based on the actual roads
and their associated speeds and is much more accurate in a study of accessibility. Much current
scholarship debates which approach is more accurate and dependable in measuring accessibility
(Ghanbari and Ghanbari 2013; Steadman 2004).
1.3 Objective
The objective for this research is to assess a key aspect of environmental justice in the
city of Phoenix, AZ: access to public green space. Recent scholarship shows that, in many cities,
public urban green space is not distributed equally, and access to public green space is often
stratified based on race and income. Boone et al. (2009) find that more African Americans have
access to parks within 400 meters walking distance in Baltimore, Maryland while White people
have access to more acreage of parks within walking distance. Many research studies have
revealed that the distribution of parks often disproportionately benefits mostly White and more
4

affluent groups (Abercrombie et al., 2008; Wolch et al., 2005; Wolch et al., 2014). The uneven
distribution of green spaces has become a serious environmental justice concern. In spite of the
fact that it is hard to alter an existing neighborhood, it is important to examine where inequalities
exist and ways that these inequalities can be overcome.
The purpose of this study is to answer the following research questions:
1. How accessible are public parks or green spaces for White, Black, Asian, Hispanic,
and American Indian populations within a walking distance of 0.5 miles?
2. Which communities need increased access to public green spaces?
1.4 Study Area
The study area of this research is the City of Phoenix, AZ. Figure 1 shows the boundary
of the City of Phoenix and its location. Phoenix is a good case study, because it is a large and
sprawling city that has great racial diversity but segregated neighborhoods. The City of Phoenix
encompasses an area around 516.70 square miles. According to the 2013 U.S. Census data, the
City of Phoenix has more than 1,513,000 people and the population has increased 4.5% since
April 2010. White is the majority racial group in Phoenix; Hispanic or Latino is the second
highest racial group; and finally, Black, Asian, and American Indian are the third, fourth, and
fifth racial groups respectively.
5

Figure 1 Location of the City of Phoenix.
6

Within the city of Phoenix, according to the 2013 U.S. Census, the White population is
mostly located in the northern and eastern Phoenix. The Hispanic population is mainly located in
the southern and western Phoenix. The Black population is mostly located in southern Phoenix.
The Asian population is mainly located in the center of Phoenix. The American Indian
population is mostly located in the north and south.
After the Mexican-American war in 1848, Mexico lost its northern part (that is the
southern part of Phoenix) to the United States and the Mexican residents became the largest
minority group in Phoenix, joined by smaller populations of African Americans, Asians, and
American Indians. Bolin et al. (2005) argue that racial categories and attendant social relations
were constructed by the Whites in the late 19th and early 20th centuries to produce a stigmatized
zone of racial exclusion and economic marginality in South Phoenix. This shows that racial
inequality has been present in Phoenix since 19th century and continues to influence the racial
distribution in Phoenix. Therefore, for this research, I hypothesized that the White population has
a better access to public urban green space than other races, such as Asian, Hispanic or Latino,
Black or African American, and American Indian.
1.5 Organizational Framework
This thesis is organized into four additional chapters. Chapter 2 reviews the literature on
environmental justice of urban green spaces and discusses the geospatial techniques and
methodologies for environmental justice analysis. Chapter 3 describes the data sources and
processing necessary to carry out the analysis. Additionally, it details methodology of the future
possible green space locations, network analysis, and park pressure analysis. Chapter 4 reviews
and interprets the results of the network analysis and discusses the study outcomes while
7

identifying possible areas of the improvement. Chapter 5 concludes the thesis by discussing the
future research directions of the environmental justice analysis of public urban green space.

8

Chapter 2 Background and Literature Review
The purpose of this chapter is to examine the history and existing research on
environmental justice, previous research studies on urban green spaces, and compare different
geospatial approaches to study green spaces access. This chapter is organized into four major
subsections: environmental justice, urban green spaces, access to green spaces, and geographic
information systems.
2.1 Environmental Justice
Since the US Civil Rights Movement emerged in the 1960s, racial inequity has become
an ongoing issue and influenced the development of the environmental justice movement in the
late 1970s and early 1980s. For example, in the late 1970s, there was a protest over a proposed
polychlorinated biphenyls (PCB) landfill location in Warren County, North Carolina (Frumkin
2005; Sister et al. 2010), a predominantly African American community. Although the landfill
was eventually approved, they challenged the proposed landfill as an act of "environmental
racism" (Frumkin, 2005; Sister et al., 2010). The term "environmental racism" refers to any low-
income group or minority community that is exposed to chemical waste, pollution, degraded
environments, or toxic waste that affects their health (Massey 2004). There were over 30,000
gallons of waste oil contaminated with PCBs that were illegally discharged in Warren County.
As a result, 60,000 tons of PCB contaminated soil were collected and later disposed of in the
landfill specifically created for this particular purpose in a predominantly African American and
low-income community in Warren County (Sister et al. 2010). This instance had a profound
effect on the environmental justice movement.
9

2.1.1 Defining Environmental Justice
As mentioned above, minorities might face different types of environmental justice issues
around their neighborhoods. It is an environmental injustice as the privileged or the authorities
often choose to build landfills, power plants, or other hazardous buildings that affect human
health near low-income and minority neighborhoods (Sister et al. 2010) and limit access to green
space.
These low-income and ethnic minorities often suffer from severe environmental pollution
and degradation (Massey 2004). Chemical or toxic wastes lead to human health problems, such
as asthma or cancer. People who live nearby these areas often have health issues due to the poor
level of the living environment. Access to urban green space can help reduce these health issues,
however, it is limited. The high quality of the living environment should not be limited to the
privileged and the affluent groups. Massey (2004) argues that the levels of income,
environmental quality, and access to health care can affect human health. Therefore, the minority
groups should also have a similar living environment and facilities as the privilege groups have.
The demand for urban green spaces is increasing because people can get fresh air,
socialize with friends, or play with children. However, most of the research has shown that
minority groups have less access to these green spaces. Wolch et al. (2005) argue that some
minority groups lack access to parks and green spaces in Los Angeles as the city has grown and
become increasingly dense. Minority groups usually live in the inner city, areas usually without
good-planning behind their built environment. Therefore, people who live in those areas often
lack recreation facilities such as green spaces.
The privilege groups have the ability to alter their living environments while the poor and
minority group might not have the money to modify their living environments and they have to
10

rely on the public urban green space. A healthy neighborhood should have plenty of green spaces
for people to rest, social, and exercise. However, where are these urban green spaces usually
located? Are they equally distribute in the city? These two questions need to be further studied
and analyzed when researchers are interested in studying accessibility.
2.1.2 Environmental Justice in Phoenix
As mentioned, minority and low-income communities often suffer from environmental
harm and risk (Environmental Protection Agency 2009). Environmental injustices are not evenly
distributed in the city of Phoenix. Scholars have examined environmental justice in the Phoenix
metropolitan area and the socio-spatial distribution of different kinds of facilities in the Phoenix
area in relation to the demographics of nearby neighborhoods. In their research, Grineski et al.
(2007) found that social-class and ethnicity are direct related to the distribution of air pollution.
Latinos, immigrants, and low-income residents have the higher exposure to pollutants than the
White (Grineski et al. 2007).
Similarly, Bolin et al. (2005) studied how racial categories and attendant social relations
were constructed by the Whites to produce a stigmatized zone of racial exclusion and economic
marginality in South Phoenix during late 19th to early 20th centuries. Bolin et al. (2005) argued
that the historical development of socio-spatial effect produced unequal and unsafe
environmental burdens in low-income and minority communities in Southern Phoenix. Therefore,
understanding the current and historical distribution of different racial groups and environmental
hazards are important to study environmental inequality.
2.2 Urban Green Space
Increasingly, researchers have started to focus on the distribution of green space access in
urban settings. Comber et al. (2008) studied green space access for different religious and ethnic
11

groups in Leicester, UK, and they found that Indian, Hindu, and Sikh groups, which are the
ethnic minorities in Leicester, have limited access to green space. Kuta et al. (2014) studied
urban green space accessibility for different socio-economic groups in the UK as well and they
found that socio-economically deprived group lack access to green space within 300m from the
residence. Sotoudehnia and Comber (2011) studied physical and perceived accessibility to urban
green space in the UK, and they found that only 15% of the population in Leicester meet the
physical access up to 300m. However, Nicholls (2001) studied accessibility and distributional
equity within a system of public parks in Bryan County, Texas using GIS and the Mann-Whitney
U test procedure in SPSS and the results show that no inequality was present. The above
examples indicate that the awareness of environmental justice and urban green space are
increasing and more people care about environmental inequality. More importantly, these studies
found that environmental inequality does exist in many places.
Section 2.1.1 has discussed the issues of environmental justice and how it affects to
human life; therefore, it is necessary to understand how to solve this issue. Urban green space is
one of the environmental factors that can benefit human health. With plenty of green spaces in
urban areas, human health can be improved, as increased vegetation improves air quality and
reduces the temperature of high heat concrete spaces. In their research study, Harlan and Ruddell
(2011) found that people who are physiologically susceptible, socioeconomically disadvantaged,
and live in the most degraded environments have higher risk of health issues.
Boone et al. (2009) agree that people who live nearby urban green spaces benefit from
access to public space and opportunities for social interactions. In addition, Giles-Corti et al.
(2005) found that people who live close to green spaces are three times more likely to get the
12

recommended amount of exercise than other people. Maller et al. (2006) also found that urban
green spaces can help improve mental health. It is clear that urban green spaces benefit residents.
2.3 Access to Green Spaces
Accessibility can have a broad meaning in general. However, in green space literature
and in this study, accessibility refers to the walking distance between the access points of the
green spaces and the residential areas.
The United Kingdom provides a set of guidelines called Accessible Natural Greenspace
Standards (ANGSt) for evaluating the provision of and access to green spaces (Comber et al.
2008). The standards are listed below:
 No one should live more than 300m from their nearest area of green space of at least 2
hectare in size.
 There should be at least one accessible 20 hectare site within 2 km from residential area.
 There should be one accessible 100 hectare site within 5 km.
 There should be one accessible 500 hectare site within 10 km.

In spite of the fact that the ANGSt model provides a detailed set of guidelines, it is not
suitable for every city or country. Some countries might not have as many green spaces as in the
UK, and not all green spaces are accessible. Therefore, particularly in the US, almost every city
has their own set of standards for green spaces. A quarter mile has become the standard distance
threshold that people are willing to walk to reach a park or recreation area (Boone et al. 2009).
However, the city of Phoenix, AZ aims to have parks or green spaces for the entire population
within 0.5 miles (800 meters). As a result, this study uses 0.5 miles to define the walking
13

distance. In order to measure the walking distance from residential areas to the nearest green
space, GIS is a helpful tool to calculate the distance, as well as the time required.
2.4 Environmental Justice, Urban Green Space, and GIS
While much research has argued that there is a connection between urban green space
and environmental justice, much of the existing research has not adequately leveraged GIS.
Traditional studies of accessibility have used the geometric perspective approach to maximize
the efficiency of distribution networks and minimize the system costs (Nicholls 2001). However,
that analysis does not take into account the distribution of outcomes or benefits among users
(Nicholls 2001). Today, many studies are examining the most suitable methodology for
analyzing green space or park access using GIS (Ghanbari and Ghanbari 2013; Hass 2009).
Section 2.1 showed that environmental inequality in urban green spaces exists in some
places and not everyone can access green spaces within a suitable walking distance. In order to
explore this environmental inequality problem in Phoenix, GIS can help identify the issues and
provide guidance for future urban planning. For example, Nicholls (2001) used GIS to help
leisure service providers to enhance the planning and management of their facilities, so that they
can provide a better service to the public. There are a few common methods that researchers
usually use to study accessibility and these methods are compared in the following sub-sections.
2.4.1 Methodologies for Studying Urban Green Space Accessibility
GIS plays an important role in analyzing green spaces access and environmental justice.
There are two common methodologies to study green space access using GIS: buffering and
network analysis. One of the simplest methods to measure accessibility is called the buffer
approach. It defines park accessibility according to a specific distance and is represented by a
buffer zone. The reason why it is called the buffer approach is because people who are within or
14

covered by the buffer zone have access to the park. The buffer zone is created around a point, a
line, or a polygon by entering a specific distance. Studies use the buffer approach because the
concept of a buffer is easy to understand and it plays an important role in many geoprocessing
workflows involving proximity or distance analysis (Flater 2011). Another method is called the
network analysis approach. It is based on the actual distance of the roads. In addition, the
network analysis approach is based on the actual speed and types of road. It is, in this research, a
desirable method to analyze the accessibility of urban green spaces because residents are more
likely to follow the actual road to the public park or green space rather than using a straight-line
buffer radius to the nearby public green space, thus it is more accurate.
2.4.2 Buffer Approach to Measure Urban Green Space
The buffer approach, also called the Euclidean buffer or covering approach, is based on a
straight line distance and does not take any blockages, barriers, or walking patterns into account.
It can be created based on a point, a line, or a polygon feature in ArcGIS (Figure 2). Ghanbari
and Ghanbari (2013) argue that the buffer approach can be efficient for simple and general
analyses. However, for some complicated analyses, the buffer approach is not a good choice
because of its simplicity.
15

Figure 2 Input features and buffers (Flater 2011)

For example, Coutts et al. (2010) and Coutts et al. (2013) use the Euclidean buffer
analysis in their research study. Both studies are quite similar and about how human health is
related to green space access. Their study areas are in Florida and they use the census tract level
and county-level as their scales. The Euclidean distance was adapted in this research which made
this research less accurate as Euclidean distance is based on the straight-line, not the road
network. In addition, Euclidean distance is a distance around a given location with a fixed
distance or time and is shorter than the actual routes. As a result, errors might occur when the
user is following the actual routes to the parks or green space. It is less accurate to calculate the
distance and time from one place to the parks and green spaces as it does not take barriers into
account.
16

2.4.3 Network Analysis Approach to Measure Urban Green Space
Network analysis, as mentioned above, calculates the access based on the road network
and the types of road (i.e., local street and freeway). In a network dataset, a junction is
represented by a node and a street and other route is represented by a straight or curved line. A
service area in network analysis approach is a region that encompasses all accessible streets that
are within a specific parameter (ESRI ArcGIS Resources). For example, a 0.25 miles park
service area includes all the streets that can be reached within 0.25 miles from the park access
point. The service area can be used to identify how many people can access the park.
Furthermore, it can be used to calculate the shortest route from point A to point B and to provide
the best routes between points. The network analysis approach provides a much more accurate
result than the buffer approach.
Instead of using the covering or buffer approach, many studies prefer using the network
analysis approach. For instance, Comber et al. (2008) used network analysis to determine the
parks and green space access for different ethnic and religious groups in the UK. Bennet et al.
(2012) used network analysis approach to measure the walking distance to the nearest
playground and to estimate the number of users of a playground using the playground's service
area. Kuta et al. (2014) applied network analysis approach to determine the accessibility to green
space for socio-economically deprived groups. Pearce et al. (2006) utilized network analysis
approach to measure community resource accessibility in New Zealand. The reason why a
number of studies prefer the network analysis approach is because it has an advantage over the
covering approach as it reflects the actual travel and avoid all the barriers that make routes
inaccessible by pedestrians (Moseleya et al. 2013).
17

2.4.4 Comparing Two Approaches to Measure Urban Green Space
By comparing the above two approaches, the network analysis approach is more accurate
and suitable for environmental justice analysis such as urban green space access. Higgs et al.
(2012) found that the results of the distance to green spaces will be affected when the research
used the undesirable methodology. Using the wrong method may lead to inaccurate results and
conclusion.
The buffer approach could be used in some general and simple analysis. However, urban
planning, for example, should consider using network analysis approach. Ghanbari and Ghanbari
(2013) compared the buffer approach and the network analysis approach and found that there are
differences between these two approaches. The authors argue that barriers and times are the
important factors that cause errors in the buffer approach, and that urban planners should pay
attention to these for planning and spatial management (Ghanbari and Ghanbari 2013). The
buffer approach would be suitable when the researcher is interested in studying the geographic
distance between points while network analysis approach is suitable when the researcher is
studying transportation and travel times (Morganstern 2015). In addition, the network analysis
approach is suitable when the scale is large so that the readers can clearly see the routes or areas
in more detail. In this research, walking distance is calculated according to the street network and
therefore, the network analysis approach is the most suitable and appropriate to use.

18

Chapter 3 Methodology
This study of the City of Phoenix, AZ analyzes environmental justice with respect to public
green space. It explores the relationship between public green space access and demographic
diversity. The idea of this research study originates from Comber et al (2008) and the
methodology is derived from that developed by Sister et al. (2007) and Hass (2009). GIS is used
and Network Analyst in ArcMap 10.3 is the major tool in this research. This study uses the
above methodology to answer two questions: 1) How accessible are public parks or green spaces
for White, Black, Asian, Hispanic, and American Indian populations within a walking distance
of 0.5 miles?; and 2) which areas need to have increased access to public green spaces?
The data processing methodology follows two main steps: 1) identify park entrance
points using the ArcGIS imagery basemap and digitize all the park access points manually; and 2)
integrate the city parcel layer and demographic data layer to obtain populations by racial groups.
The date of the ArcGIS basemap imagery is from 6/4/2013 and 6/5/2013, and the resolution of
the imagery is 0.3 meters. Figure 3 provides a summary workflow of this study.
19

Figure 3 Summary of workflow

3.1 Study Area
The study area of this project is the City of Phoenix, Arizona shown in Figure 1 (refer to
Chapter 1). The City of Phoenix is the capital of Arizona and it is the largest city in Arizona. The
City of Phoenix encompasses an area of 516.70 square miles and has more than 1,513,000 people.
According to the 2013 U.S. Census data, the White population is mostly located in the northern
and eastern Phoenix and make up a large proportion of the city's population. The Hispanic
Data
Acquisition
• Municipal Park Boundary (shapefile)
• Street and Road Network (sdc format)
• Demographic data by race (shapefile)
• City Boundary (shapefile)
• City Parcel (shapefile)
Data
Processing
• Digitize all the park access points manually
• Integrate the city parcel and demographic layer using the
"Intersect" tool
• Create park service area polygons using Network Analyst
tool
Analysis and
Results
• Measure the accessible of public urban green spaces for
different races
• Locate the areas that do not have enough green spaces
• Measure the park pressure
• Calculate the population
20

population is mainly located in the southern and western Phoenix and is the second largest racial
group. The Black population is mostly located in southern Phoenix. The Asian population is
mainly located in the center of Phoenix. The American Indian population is mostly located in the
north and south.
3.2 Data Sources and Description of Spatial Datasets
The primary datasets for this research study are listed in Table 1 below. Data for this
study were mostly obtained through two sources: 1) City of Phoenix Open Data Mapping Portal
Website and 2) U.S. Census Bureau TIGER/ Line with selected demographic data and census
block group shapefiles.
Table 1 Data used in this study
Data Purpose Source
Municipal Park Boundary To create park access points City of Phoenix Open Data
Portal
Park Access point To create park service area
polygons
Through digitizing
Street Network To calculate time and distance UCLA Geography Department
Demographic/ Population To calculate race population US Census Bureau TIGER
City Boundary To display the study area City of Phoenix Open Data
Portal
City Parcel Combine with population data City of Phoenix Open Data
Portal

The municipal park boundary layer consists of 191 parks with an overall total area of
4,443 acres and the park types include 26 mini parks, 83 neighborhood parks, 44 community
parks, 8 district parks, and 30 undeveloped parks. The size of a mini park is around 0 to 1 acres;
the size of a neighborhood park ranges from 2 to 29 aces; the size of a community park ranges
from 11 to 72 acres; the size of a district park ranges from 62 to 325 acres; and the size of an
undeveloped park ranges from 6 to 209 acres. Because undeveloped parks are not well
21

maintained and people are not allowed access; in this case, undeveloped parks will not be
analyzed. In other words, this study only focuses on the mini, neighborhood, community, and
district parks: 161 parks in total. Table 2 shows the attributes of the park dataset. Table 3
summarizes the number of parks of the layer.
Table 2 Attributes of Park Boundary Layer
Attribute Data Type
Object ID Integer
Park Name Text
Park Size (Acres) Integer
Park Type Text
Park Address Text

Table 3 Different Park Types and Number
Park Type Numbers
Mini 26
Neighborhood 83
Community 44
District 8
TOTAL 161

One of the challenges is that the park boundary layer does not include the park entrances.
It is necessary to identify the park entrance points because people may use the nearest entrance
from their home, and therefore, the network analysis would be more precise if the actual access
points are used instead of the center point of parks. Some small parks might only have one
entrance or access point, while the large parks might have several entrances. Boone et al. (2009)
suggested that a centroid could be used as the destination point for small parks. However, in
order to make the analysis more accurate, digitizing the access points is necessary for both small
and large parks. The details of digitizing park access points will be discussed in section 3.3 Data
Preparation.
22

Park access points are digitized manually. The newly created park access point layer
includes the attributes such as ID and park name so that it can easily be joined to the park
boundary layer for further analysis. The park access points serve as the facility location and are
used in the network analysis to determine the service areas so that number of people that within
the service areas can be known.
The street network layer was obtained from the University of California, Los Angeles
(UCLA) Geography Department. This layer was originally a TIGER 2000-based streets dataset,
enhanced by ESRI and Tele Atlas, and prepared for routing within the StreetMap Find Route
dialog. In this dataset, there are three types of roads with different speeds: 1) Major highways, 2)
Roads, and 3) Streets. Since this analysis calculates accessibility using walking distance,
highways are eliminated as no one is allowed to walk on highways and therefore, only streets
and roads that are under 35 miles per hour (MPH) are used.
The demographic and population layer were obtained from the 2009-2013 American
Community Survey (ACS) 5-year estimates data and contains demographic information in
Arizona. The tables are joined to the TIGER/ Line block group level shapefile. A block group is
the smallest geographic unit for which the U.S. Census Bureau reports a full range of
demographic statistics (ESRI GIS Dictionary). According to the Census website, the population
range of each block group is from 600 to 3,000 people with an optimum size of 1,500 people
(U.S. Census Bureau). Because demographic information is only available at the block group
level, this analysis utilizes the block group data to show how different races access their nearest
urban green spaces. The limitation of these data is that the pre-defined block group polygon does
not accurately represent the population in residential areas as it also includes areas such as water
23

bodies and mountain areas. Therefore, it is necessary to modify and re-shape these block group
polygons by the city parcel layer.
The city parcel layer was obtained from the City of Phoenix Open Data Portal. This
dataset shows address, zip code, and the area of each parcel. In spite of the fact that this dataset
does not indicate the population, it is combined with the Census block group data, as described in
section 3.3.2.
3.3 Data Preparation
3.3.1 Digitizing Spatial Data
Including the mini, neighborhood, community, and district parks, there are a total of 161
municipal parks that need to be analyzed. This study does not use what Boone et al. (2009)
suggested, which is to use a centroid as the access point for small parks, and determines park
access points for all parks. Using an ArcMap World Imagery basemap, it was easy to identify the
parkaccess points. The basemap provides one-meter satellite and aerial imagery and the year of
the satellite imagery was from 2013. There are a total of 879 park access points after digitizing.
The intersections of the main road and pavement inside the park are counted as the park access
point (see Figure 4 below). Digitizing manually has the potential to introduce errors, but this
error is presumably smaller than using a buffer zone created from the perimeter of a park as a
measure of accessibility.
24

Figure 4 shows the digitized park access points as an example. The red polygon is the
boundary of the park. The above example has eight access points, making it more accessible to
people who live in all directions of the park. The procedure for digitizing the access point is to
zoom in to the park layer and look for the road intersections on the aerial imagery. All park
access points were digitized using the same method. Figure 5 shows all the park access points
within the City of Phoenix.
Figure 4 Manually digitized park access points
25

Figure 5 Park access points within the City of Phoenix
3.3.2 Demographic and Population Data
The pre-defined polygon unit of the US Census block group shapefile does not accurately
represent the population in residential areas because it includes large areas of vegetation, water,
26

and some other locations where the population is zero. The city parcel clarifies where residential
areas are. As a result, it is necessary to combine and re-shape the city parcel layer and the
demographic block group shapefile to obtain a better estimation of population.
The "Intersect (Analysis)" Tool from ArcToolbox is used to combine the above two
datasets. A new created shapefile contains a more precise population than the US Census block
group shapefile as the polygons became smaller and contain only the residential areas. The
attribute table still contains racial classifications and other necessarily fields. The most important
fields of the attribute table are the population totals by race. Table 4 shows the field names and
their definitions that are necessary for the study. The newly combined shapefile and the
population data are used in the later analysis (see Section 3.4.1).
Table 4 Demographic attribute definition
Field Definition
B02001e2 White (non-Hispanic) alone total population
B02001e3

Black or African American alone total
population
B02001e4 American Indian and Alaska Native alone total
population
B02001e5 Asian alonetotal population
B03002e12 Hispanic or Latino total population

27

3.4 Network Analysis
After digitizing the access points of the parks, network analysis is performed. New park
service area polygons are created using Network Analyst. All the park access points are loaded to
the facilities in this category as this layer stores the network locations that are used as facilities in
service area analysis. There are two rings of park service areas, 0.25 and 0.5 miles, which are the
walking distances that people are willing to walk to public green spaces or parks from their
home.
3.4.1 Park Service Areas Determination
Using the function Service Area from Network Analyst, 0.25 miles and 0.5 miles service
area polygons are created (See Figure 6). No U-turns are allowed as this study assumes people
walk to the nearest public park from their home without back tracking. Parcels that are within the
polygons show the number of people that can access each of the municipal parks.

28

Figure 6 Park service area polygons based on park access points
29

The city parcel-block group intersect layer that is created in the data preparation section
is clipped using the service area polygons (Figure 7).

Figure 7 An example of clipped parcels

30

Parcels are still shown in Figure 7, but the boundaries of these parcels are dissolved after
using the "dissolve" function (Figure 8). The purpose of the "dissolve" function was used to
eliminate and dissolve the boundary. Using the dissolve function can lower the chance of
miscalculating, because it can combine small multiple features based on the same attribute and
create a larger feature in the output feature class (Figure 8). The purpose of the "clip" and
"dissolve" functions that were used in this data preparation is to exclude the population which
are outside the service area polygons, so as to not easily introduce errors. Therefore, this method
can calculate the population and demographics which are based on the percentages that within
the service area polygons in the later analysis. The areas of the parcels are used to calculate the
percentage of how many people are within the park service areas and can therefore access the
municipal parks based on the following formula:
P
0
P
1
×100%
where:
P
0
= Area of the clipped polygons
P
1
= Area of the unclipped polygons
31

Figure 8 An example of dissolved parcel boundaries
32

3.4.2 Future Possible Green Space Locations
Table 5 shows the scoring system of the future green space locations analysis. 1 being the
most suitable location for future parks with the least acres per person available in that area. This
guideline may not be suitable for every city as some cities have fewer or more parks than the
others. The acres per person for the entire city of Phoenix is 0.0021 and it does not meet the
recommendation of 0.006-0.01 acres per person in average; therefore, this research derives and
adds one more suitable level above the NRPA that recommended. As a result, this extra level can
show in a greater detail and to see how each racial groups that within service area polygon need
extra public green spaces.
Using the service area polygons that were created from the last analysis, the park service
area polygons are removed and the boundary is dissolved base on the population (Figure 9). As a
result, I can exclude those people who can access to green spaces. Using the equation of total
park area divided by the number of people, it is possible to calculate how many people outside of
the park service areas do not have enough green spaces and to identify the suitable locations for
the future green spaces. According to the park boundary layer, there are total 3,344 acres of
parks, excluding the undeveloped parks, in the City of Phoenix. A score, based on the acres per
person, is calculated and displayed as a thematic map. A high score (1) in an area means it does
not have enough public parks and needs more. This analysis is performed for each of the
different races. The results are discussed in Chapter 4.
Table 5 Scoring system
Suitability (acres per persons) Score Category
> 0.011 3 Low
0.007-0.01 2 Medium
< 0.006 1 High
33

Figure 9 An example of future green space location study area using erase and dissolve tool
34

3.4.3 Park Pressure Analysis
Park pressure analysis is to measure the potential demands on parks within the park
service area polygons that were created in section 3.4.1. Park pressure is defined as acres per
person or resident if each person were to utilize the nearest park (Sister et al. 2007). The method
of this analysis is similar to section 3.4.2. The major difference is that this analysis utilizes park
service area polygons and the demographic data that are within the polygons. The equation of
this analysis is the area of each park divided by the number of people inside the park service area
polygons. Table 6 shows the scoring system of this analysis, 1 being the greatest park pressure
with the least acres per person available in that service area. The calculated score is displayed as
a thematic map. A high score (1) means the park pressure is high in that service area and there is
a need to put more parks.
Table 6 Scoring system
Park Pressure (acres per persons) Score
> 0.011 3
0.007-0.01 2
< 0.006 1

35

Chapter 4 Results
The ultimate goal of this study is to measure the physical accessibility of public urban green
space for the five racial groups, and to explore the level of environmental inequality in the city of
Phoenix, AZ based on the real world situation and road network conditions. This method
classifies the population by race and utilizes the green space access points or entrances that have
been digitized using the World Images basemap from the ArcGIS. The distance between a
residence and the green space access points is executed using the Network Analyst Tool in
ArcGIS. This proposed method mostly depends on the Network Analyst to calculate the walking
distance in order to find out which group has the lowest public green space accessibility in
Phoenix. The results for the green space access measurements for each race are reported in this
chapter.
4.1 Overall Park Service Areas
This section summarizes the total public green spaces that are accessible for Phoenix
residents. As mentioned in Chapter 3, the Network Analysis is conducted using both 0.25 miles
and 0.5 miles as the acceptable walking distances. Table 7 shows the total park area that is
covered by the 0.25 miles and 0.5 miles park service area polygons and the number of people
that can access these green spaces. Using the equation of the total area divided by the total
population, I can then calculate the acres per person.

36

Table 7 Summary of Park Service Areas
0.25 miles 0.5 miles
Total Park Area (acres) 3,344 3,344
Total Population (people) 674,005 990,119
Acres per person 0.00496 0.00337

The above table shows that more than six hundred thousand people have access to the
nearest public green space within a walking distance of 0.25 miles from their residence and near
one million people can access public green space within a distance of 0.5 miles. However, within
the total area of public green space is constant at slightly more than three thousand acres and
hence there is high park pressure (<0.006 acres per person) in general in the city of Phoenix.
4.2 Green Space Access Results
This section includes and explains the detailed results of public green space access for
each racial group. Each group, White, Black, Asian, Hispanic, and American Indian, is discussed
separately in the following sub-sections. As mentioned in the previous chapter, the percentage of
green space access by population is calculated using the number of people in the service area
polygons divided by the number of people outside the polygons. The below sub-sections used
Equation 1 to calculate the percentage of green space access for each racial groups.
4.2.1 Green Space Access: White Population
Table 8 shows the 0.25 and 0.5 miles public green space access range for the White
population. Among the White population, a bit more than 40% and 60% can access to the nearby
public green space within 0.25 miles and 0.5 miles walking distance respectively as shown in
Figure 10. Figure 11 shows the number of White people, divided into four classes that can access
37

public green space within 0.5 miles. The classification of the population value is based on the
percentage of the population within each service areas in the White group: 0-25% (dark green),
26-50% (light green), 51-75% (orange), and 76-100% (red). Northeastern and Southwestern
Phoenix, as shown as orange and red colors, have more White people that use and access the
nearby public green space.
Table 8 Summary of the White Population
0.25 mile service areas 0.5 mile service areas
Population (Inside) 519,319 762,948
Population (Outside) 1,235,471

Figure 10 Percentage of White population that can access green space within 0.25 and 0.5 mile
service areas
38

Figure 11 Number of White people that can access green space within 0.5 mile service areas
39

4.2.2 Green Space Access: Black Population
There are around ten thousand Black people in Phoenix, which is ten times less than the
White population. Table 9 shows the summary of the Black population with access. Both
percentages of the White and Black people that can access to the public green space are quite
similar, but the percentage of Black is a bit more than the White. Approximately 45% and 65%
of the Black population can access nearby public green areas of 0.25 miles and 0.5 miles walking
respectively (Figure 12). The classification of the population value is based on the percentage of
the population within each service areas in that racial group: 0-25% (dark green), 26-50% (light
green), 51-75% (orange), and 76-100% (red). Most of the Black population, more than 75%, has
access to the public green space in Southwestern Phoenix which is indicated with red color in
Figure 13. Most public green spaces that within 0.5 miles walking distance have less than 25% of
the Black people which is indicated with dark green color.
Table 9 Summary of the Black Population
0.25 mile service area 0.5 mile service area
Population (Inside) 47,666 67,624
Population (Outside) 104,671

40

Figure 12 Percentage of Black population that can access green space within 0.25 and 0.5 mile
service area
41

Figure 13 Number of Black people that can access green space within 0.5miles
42

4.2.3 Green Space Access: Asian Population
The city of Phoenix does not have a large Asian population, compared to other races and
there are roughly only fifty thousand people that identified as Asian according to the census
(Table 10). Less than 40% of the Asian can access the nearby public green space within 0.25
mile walking distance and 52% of them can walk to the public green space in 0.5 miles (Figure
14) which is both a lower percentage of the total population compared to both the White and
Black populations, but also reflects a smaller number of people. The classification of the
population value is based on the percentage of the population within each service areas: 0-25%
(dark green), 26-50% (light green), 51-75% (orange), and 76-100% (red). Figure 15 indicates
that the Asian mostly go to public green space in the northern Phoenix, which is indicated with
red color and is located between W Peoria Ave. and W Dunlap Ave., and N 19th Ave. and the
Black Canyon Freeway. Less than 25% (as shown as dark green color) of them can access to
public green space within 0.5 miles walking distance.
Table 10 Summary of the Asian Population
0.25 mile service area 0.5 mile service area
Population (Inside) 18,538 26,483
Population (Outside) 50,261

43

Figure 14 Percentage of Asian population that can access green space within 0.25 and 0.5 mile
service areas
44

Figure 15 Number of Asian people that can access green space within 0.5 miles
45

4.2.4 Green Space Access: Hispanic Population
The Hispanic, or Latino population is the second largest, after the White population, in
the city of Phoenix (Table 11). However, the total population of the Hispanic population is still
only about half of the White population. Figure 16 shows the percentage of Hispanic population
that can access to the nearby public green space. More than 46% and 67% of the Hispanic can
access public green space within 0.25 and 0.5 miles respectively. Southern and Southwestern
Phoenix have the highest concentration of Hispanic population that can access to public green
space within 0.5 miles which is indicated in light green, orange, and red colors in Figure 17.
Table 11 Summary of the Hispanic Population
0.25 mile service areas 0.5 mile service areas
Population (Inside) 285,797 417,453
Population (Outside) 623,459

46

Figure 16 Percentage of Hispanic population that can access green space within 0.25 and 0.5
mile service areas
47

Figure 17 Number of Hispanic people that can access green space within 0.5 miles
48

4.2.5 Green Space Access: American Indian Population
The American Indian population is the smallest demographic group in Phoenix with only
thirty thousand people (Table 12). However, the green space access percentage is quite similar to
the White population. Around 40%and 60% of the American Indian population has access to
public green space within 0.25 and 0.5 miles of walking distance respectively (Figure 18).
Although the percentages are quite similar, the difference between the White and the American
Indian is most of the American Indian (as shown as red color) live near Sunsets Garden (between
W Dunlap and W Butler Dr., and N 35th Ave. and N 39th Ave.) in Western Phoenix and most of
them use public green space in that area (Figure 19).
Table 12 Summary of the American Indian Population
0.25 mile service areas 0.5 mile service areas
Population (Inside) 14,055 20,879
Population (Outside) 33,295

49

Figure 18 Percentage of American Indian population that can access green space within 0.25 and
0.5 mile service areas
50

Figure 19 Number of American Indian people that can access green space within 0.5 miles
51

4.2.6 Green Space Access: Overall
The public green space access percentages of the five different racial groups in Phoenix
are comparable, except the Asian population, which range from 40% to 45% for 0.25 miles walk
and 60% to 65% for 0.5 miles walk (Figure 20). The White population does not have a higher
percentage that live nearby the public green space or better green space accessibility.
Nevertheless, the Asian population has the lowest percentage able to access public green space
within both 0.25 miles and 0.5 miles. While the Hispanic and the Black populations have the
highest percentages able to access public green space among the racial groups, their percentages
are almost the same.

Figure 20 Comparing percentage of each racial group able to access public green space at 0.25
and 0.5 miles

4.3 Future Possible Green Space Locations Results
As mentioned in Chapter 3, future possible green space analysis calculates park acre per
the number of people in the city of Phoenix and uses the score to identify which neighborhoods
need to create new public green space for residents. There are three categories: low, medium, and
52

high. High score indicates that it is suitable to put public green spaces in that area (refer to
Section 3.4.2 for the scoring system).
4.3.1 Future Possible Green Space Locations: White
The White population is the majority race in the city of Phoenix, but many of them do not
have enough public green spaces around their residence. Figure 21 shows the suitable locations
to site public green spaces. The red color represents the areas that scored high (1) and are
therefore suitable to put more public parks or green spaces in the future while blue (3) and light
green (2) colors represent the areas that are less suitable of putting more public green spaces.
Almost the entire city of Phoenix is red, except a few places are blue color.
53

Figure 21 Future possible green space locations: White
4.3.2 Future Possible Green Space Locations: Black
The Black population, however, has a lower demand of public green space than the White
population. Figure 22 below shows that the percentage of blue and red colors are quite average.
54

Nevertheless, there are some areas still suitable to put extra public green spaces for the Black
population. The red color indicates that the suitable locations of the possible public green spaces
in the future.

Figure 22 Future possible green space locations: Black
4.3.3 Future Possible Green Space Locations: Asian
The Asian population is one of the smallest populations in the city of Phoenix. Figure 23
suggests that there is quite enough green space for the small amount of Asian people as many
55

areas have shown as blue color. However, Northern Phoenix, especially, needs extra public green
spaces as shown in Figure 23 below. There are more than 50% of blue color and light green color
in Figure 23 and this represents public green spaces are less suitable for the Asian population in
the future.

Figure 23 Future possible green space locations: Asian
56

4.3.4 Future Possible Green Space Locations: Hispanic
Hispanic is another racial category that does not have adequate public green spaces and
Figure 24 suggests that many areas are suitable to place public green spaces for the Hispanic
population. Although the Hispanic population in Phoenix is high, not all Hispanic people live
close by a public green space or can access a nearby public green space easily. Some of the His
panic population may live far away from the public green space and may require to walk a long
distance to there. As a result, Figure 24 shows many areas (shown as red color) that are suitable
to put public green spaces in the future.
57

Figure 24 Future possible green space locations: Hispanic
4.3.5 Future Possible Green Space Locations: American Indian
The American Indian population has the least demand of public green spaces in the city
of Phoenix among the five racial groups. Figure 25 suggests that Phoenix has suitable amount of
public green spaces for the small amount of the American Indian population. Most of the areas
that are shown in Figure 25 are blue and there is only a few suitable places to put green space,
for instance, in the southwestern part which shows a concentration of red color.
58

Figure 25 Future possible green space locations: American Indian
4.3.6 Future Possible Green Space Locations: Overall
As shown in Figure 21 through 25, public urban green space is insufficient for high
population races, such as White and Hispanic. There are 3,344 acres of green space in the city of
Phoenix; however, many people are not able to access a public green space within 0.5 miles from
their residence. Low population races, such as American Indian and Asian, have the lowest
demand of public green spaces in the future. In contrast, high population races, such as White
59

and Hispanic, have the highest demand of public green spaces in the future. This analysis
suggests that population and green space locations have direct relationship. High population
racial groups require more public green spaces while low population racial groups require fewer
public green spaces. Although this study cannot determine where the exact location is to place
extra public green space, it indicates the necessary and the importance of green space and shows
how many people have the difficulty to access public green space in an acceptable walking
distance.
4.4 Park Pressure Analysis Results
Park pressure measures park acres per the number of people using the park service areas
that were created in the first analysis. The service area that has high park pressure indicates a
dearth of park resource relative to the potential demand in that specific area.
4.4.1 Park Pressure Analysis: White
The park pressure is defined as acres per person, therefore the number of people
accessing an area and the size of a park are the factors that can affect the park pressure level. For
the White population, as shown in Figure 26 below, about 30% are facing a high park pressure.
The red color represents that a park has a high park pressure. The blue color represents the park
pressure is low and the green color represents the park pressure is medium.
60

Figure 26 Park pressure result (White)
4.4.2 Park Pressure Analysis: Black
The park pressure result shows that the Black population has a low level of park pressure
in the city of Phoenix. Figure 27 indicates that the majority of the park service area polygons are
blue in color and only a few of the park service area polygons are red. Most of the Black
population can access more than 0.011 acres of public green space from their residence.
61

Figure 27 Park pressure result (Black)
4.4.3 Park Pressure Analysis: Asian
The Asian population, as shown in Figure 28 below, exerts a medium level of park
pressure on parks in the city of Phoenix. Central Phoenix has the highest park pressure level as
the majority of the park service area polygons are red in color; otherwise, the rest of them are
62

mostly blue and green in color. There are many areas in Central Phoenix where the Asian
population has less than 0.006 acres per person of public green space, but most parks have a low
park pressure with 0.011 to 23 acres per person accessible.

Figure 28 Park pressure result (Asian)
63

4.4.4 Park Pressure Analysis: American Indian
The park pressure level for the American Indian population is high in southwestern and
northeastern Phoenix (Figure 29). The park pressure level based on the American Indian
population is relatively low in the central Phoenix as it is shown in blue and green colors. This
could be due to the fact that not many American Indians live in Central Phoenix.

Figure 29 Park pressure result (American Indian)
64

4.4.5 Park Pressure Analysis: Hispanic
The park pressure level for the Hispanic population is also relatively low (Figure 30).
Many parks have a low park pressure of 0.011 to 36 acres per person, as shown in blue. However,
the park pressure level in central Phoenix based on the Hispanic population is especially high if
compared to other parts of Phoenix, with many parks allowing for less than 0.006 acres per
person.

Figure 30 Park pressure result (Hispanic)
65

4.4.6 Park Pressure Analysis: Overall
Figure 31 shows the park pressure level for entire population in the city of Phoenix. It is
clear that central Phoenix has a high level of park pressure and it is suggested the demand of
public green spaces and parks in central Phoenix is very high. Many Phoenix residents have less
than 0.006 acres of green spaces in Central Phoenix. Otherwise, the blue indicates that Phoenix
residents have around 0.011 to 1.102 acres per person of public green spaces within 0.5 miles
walking distance from their residences.
66

Figure 31 Overall park pressure result

67

Chapter 5 Discussion and Conclusion
This study examines the accessibility of existing public urban green spaces in the city of
Phoenix, Arizona, using ArcGIS network analysis. This chapter discusses the major findings and
observations of the study, as well as the contribution to existing research on environmental
inequality and public urban green space access in the United States. The chapter concludes and
provides recommendations for future research on urban green space access.
5.1 Summary of Results
5.1.1 First Analysis
There are several conclusions that can be drawn from the first analysis, which calculates
the number of people that can access public green spaces within 0.25 and 0.5 miles of their
residences. Secondly, results show that the White population does not have greater access to
public green space than other peoples. However, the analysis shows a distinct result that the
Asian population has lower access to public green spaces than other peoples. In addition, among
of the five racial populations, the Hispanic population has the highest percentage with access to
public green space.
5.1.2 Second Analysis
The results of the future possible green space locations analysis showed a great contrast
among the racial groups. As the percentage of Whites and Hispanics are the highest in the city,
the proportion and demand of the public green space needed is also the highest. Figures 21 to 25
show the suitable locations of public green spaces for different racial groups. Minority
neighborhoods, such as Asian and American Indian neighborhoods, exhibited similar results and
these neighborhoods do not need much public green space in the future. This demonstrates that
68

overall population size and green spaces have a direct relationship. Further planning can make
use of this analysis for reference.
5.1.3 Third Analysis
The results of the park pressure analysis showed the level of park pressure for the
different racial groups. This analysis included the park pressure for the White, the Black, the
Asian, the American Indian, the Hispanic populations, and as well as the overall population in
the city of Phoenix. Results show that the White population exerts the highest level of park
pressure among the racial groups while the Black population exerts the lowest level of park
pressure. In general, the highest level of park pressure in the city of Phoenix is in the Central
Phoenix area.
5.2 Significance of Findings
The major conclusion of this study is that the White population in the city of Phoenix
does not have better access to public green space while the Hispanic and the Black population
have better access to public green space. The results of this study are quite similar to the primary
conclusion of Boone et al. (2009), that a higher percentage of the Black population has access to
parks within walking distance than the White population. If the city of Phoenix is mostly made
up of White people, while they are not the highest percentage of living nearby public green space,
this may suggest that public green space is more favorable to the Hispanic and the Black
populations and less favorable to the Asian population. In other words, environmental injustice
does exist in the city of Phoenix, but in this study, the White population is not the racial group
that can access public green spaces easily. The public green space accessibility is more favorable
to the Hispanic and the Black population.
69

Chapter 2 has mentioned that environmental inequality does exist in many parts of the
world. Some facilities in neighborhoods might be more unfavorable to some specific racial
groups. However, this study shows that some racial groups, such as the Black population, have
better access to nearby public green space than the White population. In spite of the fact that this
study shows that some racial groups of color have a higher public green space access than the
White population, based on the principle of environmental justice, there is a need for further
policy interventions to promote equitable green space access.
While the future possible green space analysis does provide a greater understanding of
the need of public green space, it only shows general areas that are in need of extra public green
spaces for specific racial groups. Further analysis and research are needed to investigate the ideal
locations for these public green spaces. Although this study focuses on how different racial
groups access nearby public green spaces, park and green space are not only for one specific
racial group in the real world. For example, children and the elderly probably walk to the nearby
playground or green space. As a result, public parks and green spaces may be in high demand in
areas where there are many children and elderly.
The park pressure analysis provides a general index for each racial group to show where
the highest and lowest levels of park pressure are based on the amount of public green space and
the number of people. This analysis shows that the White population, compared to other racial
groups, exerts the highest park pressure. A large population of White people is probably the best
explanation for this situation. Overall, this analysis suggests that the central part of Phoenix
needs to increase the acres of public green space because the population is relatively high and
there are not enough acres of public green space for the people within 0.5 miles walking distance
from their residences.
70

5.3 Limitations
Based on the methods of this study, there are limitations that might affect the results.
First of all, besides the methodology itself, the data such as population, street networks, and park
entrance locations, these datasets can cause errors during the analysis. This study utilizes
secondary data that are obtained from other people or agencies, thus this study is not able to tell
if these data contained errors during the data collection process.
On the other hand, the methodology is mainly based on the street networks and the
ArcMap software. All the calculations, such as the distance and time, are calculated
automatically in ArcMap. Therefore, this study might contain errors that came from the datasets.
The second analysis uses the US census demographic data to calculate park acres per person in
Phoenix to determine the suitable location of public green spaces. However, there are many
factors that can influence the location of the possible green spaces in reality. The weakness of
this analysis is that it is only focused on the scale of how different racial populations lack access
to public green spaces and does not consider other factors that might affect the results. Future
research on green space accessibility can pay attention on other factors to find out suitable
locations to put green spaces. Site suitability analysis might be a good method to locate green
spaces for different racial populations.
5.4 Future Research
Although this study reveals that the White population does not always have better access
to nearby public green space, the results of this study highlight some limitations of investigating
environmental inequality and public urban green space access. This section provides suggestions
for future research.
71

Public and private green spaces can have great influence on the environmental inequality
research. This research only focuses on the public green space and does not include private green
space such as golf course. This research showed that the White people have less access to public
green space in Phoenix than people of other races. However, the result might be totally different
if private green space is included in the research. Phoenix has a lot of private golf courses and
these golf courses will have a much lower park pressure than the public green spaces. The White
population might have a limited access to public green space, but they could have a better access
to private green spaces such as golf courses. Future research can analyze both public and private
green spaces to examine whether environmental inequality exists and which racial group has
better access to urban green space in the city of Phoenix.
Accessibility of public urban green space may involve many factors, and these factors
can affect the results of the analysis. Various types of green spaces, including parks, sport
grounds, or vegetation covered spaces, have a significant influence on the public urban green
space access measurements. It is necessary to classify the types of green space in the future green
space study or accessibility study.
Furthermore, the time and month can affect the number of people using the public green
space. This is especially important to the park pressure analysis. More parents bring their
children or pets to public green space during holidays or weekends or after-school. For instance,
a public green space does not have much people during off-peak hours but it does not mean that
there are no people going to that public green space during peak hours. This will affect the
results and more importantly, these result will be biased toward the number of visits.
In addition, the study limited the boundary within the city of Phoenix only. There are
possible public green spaces that are outside the Phoenix boundary but can be reached within 0.5
72

miles walking distance. Future research could also include those accessible green spaces and
does not need to set a fixed boundary for the study area. Ideally, this research would enrich the
existing environmental justice research.
Moreover, future accessibility analysis can include other variables beyond race, including
social status and religion. These variables can be used to compare green space access in different
neighborhoods. The ultimate goal of accessibility analysis is to stimulate the awareness of
environmental inequality and to let more people understand how environmental inequality
influences the health and quality of life of everyday life in cities.
Finally, the network analysis from ArcMap is an automated process and does not require
advanced level programming language or scripting. This might affect the accuracy of the results
because of the limited data input. Future studies could, if possible, develop an advanced
measurement model that would allow researchers to input the necessary data such as population,
access points, road types, and execute multiple solutions for comparison and analysis. While
developing the model would be a benefit for accessibility analysis, it is also important to identify
the park access points accurately. Some parks or green spaces either contain more than one
entrance or do not have a clear entrance. For example, there are no park access points data in
Phoenix, and this data is needed to digitize the access points manually. However, digitizing
process may introduce errors. Therefore, it is recommended to acquire a list of park entrance
coordinates or addresses for geocoding in the future research. An accurate access point dataset
can help decrease the data preparation time while increasing the accuracy during the network
analysis.
The benefits of the network analysis approach have been discussed in this study and
many studies utilize this approach to explore environmental justice. The results of this study
73

highlight some limitations of investigating urban green space access. The primary results of this
study do not necessarily reflect all realities of green space access in the city of Phoenix; however,
this network analysis approach has utilized an important research method that can be used in
future research studies.

74

References
Abercrombie, Lauren, James Sallis, Terry Conway, Lawrence Frank, Brian Saelens, and James
Chapman. 2008. "Income and Racial Disparities in Access to Public Parks and Private
Recreation Facilities." American Journal of Preventive Medicine 34(1): 9-15.
Baycan-Levent,Tuzin, Eveline van Leeuwen, Caroline Rodenburg, and Peter Nijkamp. 2002.
"Development and Management of Green Spaces in European Cities: A Comparative
Analysis." Paper presented at the 38th International Planning Congress on “The Pulsar
Effect” Planning with Peaks, Glifada, Athens, Greece, September 21-26, 2002., 237-247.
Bennet, Scott, Nikolaos Yiannakoulias, Allison Williams, and Peter Kitchen. 2012. "Playground
Accessibility and Neighbourhood Social Interaction Among Parents." Social Indicators
Research 108: 199-213.
Bolin, Bob, Sara Grineski, and Timothy Collins. 2005. "The Geography of Despair:
Environmental Racism and the Making of South Phoenix, Arizona, USA." Human
Ecology Review 12(2): 156-168.
Boone, Christopher, Geoffrey Buckley, Morgan Grove, and Chona Sister. 2009. "Parks and
People: An Environmental Justice Inquiry in Baltimore, Maryland." Annals of the
Association of American Geographers 99(4): 767-787.
Comber, Alexis, Chris Brunsdon, and Edmund Green. 2008. "Using a GIS-based network
analysis to determine urban greenspace accessibility for different ethnic and religious
groups." Landscape and Urban Planning 86: 103-114.
Coutts, Christopher, Mark Horner, and Timothy Chapin. 2010. "Using GIS to model the effects
of green space accessibility on mortality in Florida." Geocarto International 25(6): 471
-484.
75

Coutts, Christopher, Timothy Chapin, Mark Horner, and Crystal Taylor. 2013. "County-Level
Effects of Green Space Access on Physical Activity." Journal of Physical Activity and
Health10(2): 232-240.
Environmental Protection Agency. Environmental Justice Resource Guide: A Handbook for
Communities and Decision-Makers. Washington, DC: Environmental Protection Agency,
2009.
ESRI ArcGIS Resources. ArcGIS Help 10.1 Service area analysis. July 2, 2014.
http://resources.arcgis.com/EN/HELP/MAIN/10.1/index.html#//004700000048000000
(accessed September 18, 2015).
ESRI Support. ESRI GIS Dictionary: Block group.
http://support.esri.com/en/knowledgebase/GISDictionary/term/block%20group (accessed
September 16, 2015).
Flater, Drew. "Understanding Geodesic Buffering: Correctly use the Buffer tool in ArcGIS."
ArcUser, 2011: 33-37.
Frumkin, Howard. 2005. "Guest Editorial: Health, Equity, and the Built Environment." Environ
Health Perspect 113(5): 290-291.
Ghanbari, Atefeh,and Mahdi Ghanbari. 2013. "Assessing Spatial Distribution of Tabriz Parks by
GIS (Compared Network Analysis and Buffering)." Geography and Environmental
Planning Journal 50(2): 57-60.
Giles-Corti, Billie, Melissa Johnson, Matthew Knuiman, Catherine Collins, Kate Douglas, and
Kevin Ng. 2005. " Increasing walking: How important is distance to, attractiveness, and
size of public open space? ." American Journal of Preventive Medicine 28(2S2): 169-176.
76

Grineski, Sara, Bob Bolin and Christopher Boone. 2007. "Criteria Air Pollution and
Marginalized Populations: Environmental Inequity in Metropolitan Phoenix, Arizona."
Social Science Quarterly 88: 535-554.
Harlan, Sharon, and Darren Ruddell. 2011. "Climate change and health in cities: Impacts of heat
and air pollution and potential co-benefits from mitigation and adaptation."
Environmental Sustainability 3: 126-134.
Hass, Kara. "Measuring Accessibility of Regional Parks: A Comparison of Three GIS
Techniques." Master‟s thesis, San Jose State University, 2009.
Heynen, Nik, Harold Perkins, and Parama Roy. 2006. "The political ecology of uneven Green
Space: The Impact of Political Economy on Race and Ethnicity in Producing
Environmental Inequality in Milwaukee." Urban Affairs Review 42: 3-25.
Higgs, Gary, Richard Fry, Mitchel Langford. 2012. "Investigating the implications of using
alternative GIS-based techniques to measure accessibility to green space." Environment
and Planning B: Planning and Design 39: 326-343.
Kuta, Abdullahi, Joseph Odumosu, Oluibukun Ajayi, Nanpon Zitta, Hassan Samail-Ija, and
Ekundayo Adesina. 2014. "Using a GIS-Based Network Analysis to Determine Urban
Greenspace Accessibility for Different Socio-Economic Groups, Specifically Related to
Deprivation in Leicester, UK." Civil and Environmental Research 6(9): 12-20.
Maller, Cecily, Mardie Townsend, Anita Pryor, Peter Brown, and Lawrence St Leger. 2006.
"Healthy nature healthy people: „contact with nature‟ as an upstream health promotion
intervention for populations." Health Promotion International 21(1): 45-54.
Massey, Rachel. Environmental Justice: Income, Race, and Health. Medford: Global
Development And Environment Institute, Tufts University, 2004.
77

Mitchell, Richard, and Frank Popham. 2007. "Greenspace, urbanity and health: relationships in
England." J Epidemiol Community Health 61: 681-683.
Morganstern, Seth. "Disparities In Food Access: An Empirical Analysis Of Neighborhoods In
The Atlanta Metropolitan Statistical Area."Master's Thesis, University of Southern
California, 2015.
Moseley, Darren, Mariella Marzano, Jordan Chetcuti, and Kevin Watts. 2013. "Green networks
for people: Application of a functional approach to support the planning and management
of greenspace." Landscape and Urban Planning 116: 1-12.
Nicholls, Sarah. 2001. "Measuring the accessibility and equity of public parks: a case study using
GIS." Managing Leisure 6: 201-219.
Parsons, Jonathan. "Mapping Uniformity Of Park Access Using Cadastral Data Within Network
Analyst In Wake County, Nc."Master's Thesis, University of Southern California, 2015.
Pearce, Jamie, Karen Witten, and Phil Bartie. 2006. "Neighbourhoods and health: a GIS
approach to measuring community resource accessibility." Journal of Epidemiology and
Community Health60(5): 389-395.
Sister, Chona, Jennifer Wolch, and John Wilson. 2010. "Got green? addressing environmental
justice in park provision." GeoJournal 75(3): 229-248.
Sister, Chona, John Wilson, and Jennifer Wolch. Park Congestion and Strategies to Increase
Park Equity. Los Angeles: University of Southern California GIS Research Laboratory
and Center for Sustainable, 2007.
Sotoudehnia, Fariba , and Lex Comber. "Measuring Perceived Accessibility to Urban Green
Space: An Integration of GIS and Participatory Map."Paper presented at the AGILE,
April 18-22, 2011.
78

Steadman, Philip. 2004. "Guest editorial: Developments in space syntax." Environment and
Planning B: Planning and Design 31: 483-486.
US Census Bureau. Geographic Terms and Concepts - Block Groups. December 6, 2012.
https://www.census.gov/geo/reference/gtc/gtc_bg.html (accessed September 27, 2015).
Wolch, Jennifer, Jason Byrne, and Joshua Newell. 2014. "Urban green space, public health, and
environmental justice: The challenge of making cities „just green enough‟." Landscape
and Urban Planning 125: 234-244.
Wolch, Jennifer, John Wilson, and Jed Fehrenbach. Parks and Park Funding in Los Angeles: An
Equity Mapping Analysis. Los Angeles: GIS Research Laboratory University of Southern
California, 2005.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

Exploring the relationship between urban forests and ethnicity in West Covina using GIS and object-based image analysis

PDF

Using pedestrian accessibility indicators to locate schools: a site suitability analysis in Greenville County, South Carolina

PDF

Spatial narrative of the invasive lionfish in the Western Atlantic and Caribbean oceans: a GIS story map

PDF

Urban areas and avian diversity: using citizen collected data to explore green spaces

PDF

Finding food deserts: a study of food access measures in the Phoenix-Mesa urban area

PDF

Does the Bay Area have a social center? Delimiting the postmodern urban center of the San Francisco Bay Area

PDF

A spatial analysis of violent crime cold spots: testing the capable guardian component of the routine activity theory

PDF

A comparison of pixel-based, object-based, deep learning, and data fusion image classification approaches in delineating urban tree canopy in City Terrace

PDF

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

PDF

A spatial analysis of veteran healthcare accessibility

PDF

Implementing spatial thinking with Web GIS in the non-profit sector: a case study of ArcGIS Online in the Pacific Symphony

PDF

Utilizing 311 service requests as a signature of urban location in the City of Los Angeles

PDF

Spatial distribution of the endangered Pacific pocket mouse (Perognathus ssp. pacificus) within coastal sage scrub habitat at Dana Point Headlands Conservation Area

PDF

A critical assessment of the green sea turtle central west Pacific distinct population segment utilizing maxent modeling on nesting site locations

PDF

The role of amenities in measuring park accessibility: a case study of Downey, California

PDF

A comparison of weighted and fuzzy overlays in mapping landslide susceptibility, south-central front range, Colorado

PDF

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

PDF

Developing, maintaining, and employing crowd-sourced geospatial data in support of helicopter landing zone surveys for disaster response operations

Asset Metadata

Creator So, Shuk Wai (author)

Core Title Urban green space accessibility and environmental justice: a GIS-based analysis in the city of Phoenix, Arizona

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 06/17/2016

Defense Date 01/08/2016

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

Tag environmental justice,GIS,Inequality,network analysis,OAI-PMH Harvest,Phoenix,urban green space

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Warshawsky, Daniel (committee chair), Loyola, Laura (committee member), Sedano, Elisabeth (committee member)

Creator Email shukwais@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-254295

Unique identifier UC11281154

Identifier etd-SoShukWai-4446.pdf (filename),usctheses-c40-254295 (legacy record id)

Legacy Identifier etd-SoShukWai-4446-1.pdf

Dmrecord 254295

Document Type Thesis

Format application/pdf (imt)

Rights So, Shuk Wai

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