Assessment of Ecosystem Services of Urban Green Space As A Tool For Sustainable Urban Development - The Case of Xi'an, China

Download as pdf or txt
Download as pdf or txt
You are on page 1of 227

Wissenschaftszentrum Weihenstephan für Ernährung,

Landnutzung und Umwelt


Lehrstuhl für Strategie und Management der Landschaftentwicklung

Assessment of Ecosystem Services of


Urban Green Space as a Tool for
Sustainable Urban Development – The
Case of Xi’an, China

Tong Cai
Vollständiger Abdruck der von der Fakultät Wissenschaftszentrum
Weihenstephan für Ernährung, Landnutzung und Umwelt der
Technischen Universität München zur Erlangung des akademischen
Grades eines
Doktors der Agrarwissenschaften (Dr. agr.)
genehmigten Dissertation.

Vorsitzender: Prof. Dr. Anton Fischer


Prüfer: 1. Prof. Dr.-Ing. Stephan Pauleit
2. Prof. Dr. Jürgen Breuste

Die Dissertation wurde am 14.03.2018 bei der Technischen Universität


München eingereicht und durch die Fakultät Wissenschaftszentrum
Weihenstephan für Ernährung, Landnutzung und Umwelt am 01.06.2018
angenommen.
Contents
Acknowledgements........................................................................................................................... IX
Abstract ............................................................................................................................................... X

1 Introduction.........................................................................................................1
Urbanization and Urban Challenges...................................................................................1
Importance of Urban Green Spaces ...................................................................................2
Aim and Objectives of the Research ..................................................................................4
Organization of the Research ..............................................................................................4
2 Literature Review...............................................................................................7
Urban Green Spaces and Ecosystem Services ..................................................................7
2.1.1 Concept and Classification of Urban Green Spaces ................................................7
2.1.2 Concept and Classification of Ecosystem Services in Urban Green Spaces.........9
2.1.3 Contributions of Urban Green Spaces to the Provision of Ecosystem Services 11
2.1.4 Relationship between Ecosystem Services and Features of Urban Green Spaces13
Assessment and Evaluation of Ecosystem Services Provided by Urban Green Spaces15
2.2.1 Monetary and Non-monetary Evaluation of Ecosystem Services Provided by
Urban Green Spaces .................................................................................................................. 15
2.2.2 Indicators for the Evaluation of Ecosystem Services Provided by Urban Green
Spaces ..................................................................................................................................... 18
2.2.3 Processing the Indicators for the Assessment of Ecosystem Services Provided
by Urban Green Spaces ............................................................................................................. 20
Land Cover Mapping as an Assessment Tool for Regulating Services of Urban
Green Spaces................................................................................................................................... 22
2.3.1 Land Cover Types within Urban Green Spaces and Their Relation to Ecosystem
Services ..................................................................................................................................... 22
2.3.1.1 Definition and Classification of Land Cover ................................................ 22
2.3.1.2 Relationship between Land Cover and Regulating Ecosystem Services... 26
2.3.2 Vegetation Structure Types within Urban Green Spaces and their Relation to
Ecosystem Services.................................................................................................................... 27
2.3.2.1 Definition and Classification of Vegetation Structure.................................. 27
2.3.2.2 Relationship between Vegetation Structure and Regulating Ecosystem
Services ............................................................................................................................ 30
Green Space Features and their Relation to Human Use and Perception ................... 34
2.4.1 Human Use of Urban Green Spaces....................................................................... 34
2.4.2 Recreation Activities in Urban Green Spaces........................................................ 35
2.4.3 Respondents’ Perceptions of Urban Green Spaces ............................................... 37
I
2.4.4 Respondents’ Preferences for Urban Green Spaces.............................................. 38
Ecosystem Services as an Approach to Support Green Space Planning..................... 39
Research Questions ........................................................................................................... 42
3 Materials and Methods .................................................................................. 44
Study Area.......................................................................................................................... 44
3.1.1 Definition ................................................................................................................... 44
3.1.2 Geographic Conditions ............................................................................................ 44
3.1.3 Social, Economic and Environmental Conditions ................................................ 47
Methods .............................................................................................................................. 49
3.2.1 Outline of Methods ................................................................................................... 49
3.2.2 A Rapid Ecosystem Service Assessment Approach............................................. 50
3.2.2.1 Summary of the Approach.............................................................................. 50
3.2.2.2 Indicators used for Assessment ...................................................................... 51
3.2.2.3 Indicator Rating ................................................................................................ 53
3.2.2.4 Aggregation ...................................................................................................... 53
3.2.3 Land Cover Mapping ............................................................................................... 55
3.2.3.1 Data Collection................................................................................................. 55
3.2.3.2 Data Processing and Analysis......................................................................... 59
3.2.4 Questionnaires........................................................................................................... 60
3.2.4.1 Sample Size of the Questionnaire Survey ..................................................... 60
3.2.4.2 Question Design ............................................................................................... 62
3.2.4.3 Data Processing and Analysis......................................................................... 64
3.2.5 Document analysis.................................................................................................... 64
4 Planning and Development of Green Spaces in Xi’an............................. 66
Overview of the planning documents ............................................................................. 66
4.1.1 Emergence of a Green Space Structure Plan (1953-1972)................................... 66
4.1.2 The Green Space Structure Plan from the 1980s (1980-2000)............................ 67
4.1.3 The Green Space Structure Plan from the 1990s (1995-2010)............................ 68
4.1.4 The Green Space Structure Plan from the 2000s (2008-2020)............................ 69
4.1.5 Summary of the Green Space Plan ......................................................................... 70
The Development of Urban Green Spaces in Practice .................................................. 72
Challenges in the Planning of Urban Green Spaces ...................................................... 75
5 Types of Land Cover and Vegetation Structures within Green Spaces
of Xi’an...................................................................................................................... 78
Land Cover Types within Urban Parks of Xi’an ........................................................... 78
Vegetation Structure Types in Urban Parks of Xi’an .................................................... 80
Discussion .......................................................................................................................... 84
II
5.3.1 Comparison of Land Cover / Vegetation Structure Types in Xi’an city with
those Reported in Similar Studies............................................................................................. 84
5.3.2 The Relationships between Land Cover / Vegetation Structure and Park
Variables ..................................................................................................................................... 85
6 Citizens’ Use, Perceptions, Assessments and Demands for Green
Spaces in Xi’an ........................................................................................................ 88
Response Rate and Respondents’ Characteristics.......................................................... 88
Citizens’ Use Behaviors of Green Spaces ...................................................................... 89
6.2.1 Use Types and Variations among Social Groups.................................................. 89
6.2.2 Citizens’ Use Frequencies, Times and Durations of Visits, and Variations
among Social Groups................................................................................................................. 90
6.2.3 Citizens’ Use Frequency and Duration of Visits, and Variations among Park
Variables ..................................................................................................................................... 93
Citizens’ Recreational Activities in Green Spaces......................................................... 96
6.3.1 Citizens’ Recreational Activities in Green Spaces and Socioeconomic
Differences .................................................................................................................................. 96
6.3.2 Citizens’ Recreational Activities in Green Spaces and Differences of Park
Variables ..................................................................................................................................... 99
Citizens’ Perceptions of Green Spaces.......................................................................... 101
6.4.1 Citizen’ Perceptions of Green Space Types and Socioeconomic Differences. 101
6.4.2 Citizen’ Attitudes Towards the Importance of Ecosystem Services of Green
Spaces and Socioeconomic Differences ................................................................................ 103
6.4.3 Citizen’ Perceptions of the Importance of Ecosystem Services of Green Spaces
and Differences of Park Variables.......................................................................................... 106
Citizens’ Assessments of Green Spaces........................................................................ 108
6.5.1 Citizens’ Assessments of Green Spaces and Socioeconomic Differences....... 108
6.5.2 Citizens’ Assessments of Green Spaces and Differences of Park Variables.... 113
Citizens’ Demands for Green Spaces............................................................................ 115
6.6.1 Citizens’ Demands for Green Spaces and Socioeconomic Differences ........... 115
6.6.2 Citizens’ Demands for Green Spaces and Differences of Park Variables........ 118
Discussion ........................................................................................................................ 120
6.7.1 Discussion of the Respondents’ Green Space Use Behaviors ........................... 120
6.7.1.1 Comparison of Green Space Use Behaviors with Other Studies.............. 120
6.7.1.2 The Influence of Socioeconomic Variables on Respondents’ Use
Behaviors .......................................................................................................................... 121
6.7.1.3 The Influence of Park Variables on Respondents’ Use Behavior ............ 122
6.7.2 Discussion of Respondents’ Recreational Activities in Green Spaces ............. 123
III
6.7.2.1 A Comparison of Respondents’ Recreational Activities with Similar
Studies .......................................................................................................................... 123
6.7.2.2 The Influence of Socioeconomic Variables on Respondents’ Recreational
Activities .......................................................................................................................... 124
6.7.2.3 The Influence of Park Variables on Respondents’ Recreational Activities124
6.7.3 Discussion of Respondents’ Perception of Green Spaces .................................. 125
6.7.3.1 Comparison of Respondents’ Perceptions with Similar Studies............... 125
6.7.3.2 The Influence of Socioeconomic Variables on Respondents’ Perceptions of
Ecosystem Services ............................................................................................................. 127
6.7.3.3 The Influence of Park Variables on Respondents’ Perceptions of
Ecosystem Services ............................................................................................................. 127
6.7.4 Discussion of the Respondents’ Assessments of Green Spaces ........................ 128
6.7.4.1 Comparison of Respondents’ Assessments with Similar Studies ............ 128
6.7.4.2 The Influence of Socioeconomic Variables on Respondents’ Assessments129
6.7.4.3 The Influence of Park Variables on Respondents’ Assessments .............. 130
6.7.5 Discussion of Respondents’ Demands for Green Spaces .................................. 131
6.7.5.1 Comparison of Respondents’ Demands with Similar Studies .................. 131
6.7.5.2 The Influence of Socioeconomic Variables on Respondents’ Demands. 132
6.7.5.3 The Influence of Park Variables on Respondents’ Demands.................... 132
7 Assessment of Ecosystem Services Generated by Urban Parks in Xi’an134
Regulating Services of Urban Parks in Xi’an............................................................... 134
7.1.1 Indicator Ratings of Regulating Services of Urban Parks in Xi’an................... 134
7.1.2 Assessment of Integrated Indicators of Regulating Services of Urban Parks in
Xi’an ................................................................................................................................... 138
7.1.3 Citizens’ Assessments of Regulating Services of Urban Parks in Xi’an.......... 140
Recreational Services of Urban Parks in Xi’an ............................................................ 141
7.2.1 Citizens’ Assessments of Recreational Services of Urban Parks in Xi’an ....... 141
7.2.2 Integrated Assessment of Recreational Services of Urban parks in Xi’an ....... 142
Discussion ........................................................................................................................ 146
7.3.1 The Influence of Park Variables on the Assessment of Regulating Services... 146
7.3.2 Discussion of the Assessment of Recreational Services..................................... 150
7.3.2.1 Comparison of the Assessment of Recreational Services with Similar
Studies .......................................................................................................................... 150
7.3.2.2 The Influence of Park Variables on the Assessment of Recreational
Services .......................................................................................................................... 151
8 Final Discussion and Conclusions .............................................................. 153
Discussion of the Results ................................................................................................ 153
IV
Suggestions for Improving the Green Spaces of Xi’an ............................................... 156
Limits of This Study and Suggestions for Future Research........................................ 162
8.3.1 Limits and Suggestions for the Non-monetary Assessment Method................ 162
8.3.2 The Limits and Suggestions for the Questionnaires............................................ 163
8.3.3 Suggestions for Future Studies .............................................................................. 164

Bibliography .................................................................................................................................... 166


Appendix and Annex ..................................................................................................................... 184
Appendix 1 .................................................................................................................................. 184
Appendix 2 .................................................................................................................................. 185
Appendix 3 .................................................................................................................................. 186
Appendix 4 .................................................................................................................................. 187
Appendix 5 .................................................................................................................................. 189
Annex 1 ........................................................................................................................................ 194
Annex 2 ........................................................................................................................................ 199
Curriculum vitae ............................................................................................................................ 211

V
List of Figures
Figure 1.1 Research flow chart .............................................................................................................6
Figure 3.1 Site map of Xi’an and the 22 surveyed sites (own drawing) ....................................... 46
Figure 3.2 Procedure for applying the methods of this research .................................................... 49
Figure 3.3 Height variations of woody plants (author’s own drawing) ........................................ 56
Figure 4.1 Master plan of Xi’an city (1953-1972) (source: Xi'an Urban Planning Bureau) ..... 67
Figure 4.2 Master plan of Xi’an city (1980-2000) (source: Xi'an Urban Planning Bureau) ...... 68
Figure 4.3 Master plan of Xi’an city (1995-2010) (source: Xi'an Urban Planning Bureau) ...... 69
Figure 4.4 Master plan of Xi’an city (2008-2020) (source: Xi'an Urban Planning Bureau) ...... 70
Figure 4.5 The regional development of the green space system of Xi’an city based on the plans
from the 1980s, 1990s and 2000s (Source: Yuan, 2014)................................................................ 74
Figure 4.6 Locations of urban parks within the areas of Xi’an city inside the third ring road in
the year 2009 (source: Wang, 2010) ................................................................................................. 74
Figure 5.1 Comparison of land cover and vegetation structure between Children’s Park (left)
and Xing Qing Park (right) ................................................................................................................ 84
Figure 6.1 Mean rankings of main recreational activities in green spaces.................................... 98
Figure 6.2 Mean rankings of importance of ecosystem services ................................................. 105
Figure 6.3 Mean rankings of the assessments................................................................................ 112
Figure 6.4 Mean rankings of the survey respondents’ demands for specific features in urban
green spaces ....................................................................................................................................... 117
Figure 7.1 Cover ratio of vegetation and water in the 22 investigated parks in Xi’an............... 134
Figure 7.2 Relative proportions of vegetation layers in the 22 investigated parks in Xi’an...... 135
Figure 7.3 Relative proportions of tree cover in the 22 investigated parks in Xi’an.................. 136
Figure 7.4 Relative proportions of mature trees in the 22 investigated parks in Xi’an.............. 137
Figure 7.5 Relative proportions of evergreen species in the 22 investigated parks in Xi’an. ... 138

List of Tables
Table 2.1 Classification of urban green spaces in China (source: CMC, 2007).............................8
Table 2.2 Classification of urban parks in China (source: CMC, 2007) ....................................... 10
Table 2.3 Ecosystem services generated by green and water areas in urban regions (source:
Niemeläet al., 2010)........................................................................................................................... 12
Table 2.4 Valuation methods used for ecosystem services generated by urban green spaces
(source: Baycan-Levent and Nijkamp, 2004 and 2005a) ............................................................... 16
Table 2.5 Examples of land cover units within urban parks (own drawing) ................................ 25
Table 2.6 Examples of vegetation structure types within urban green spaces (own drawing) ... 29
Table 3.1 List of the 22 investigated parks (source: Xi'an Ancient Architecture and Landscape
Architecture Design Institute, 2009) ................................................................................................. 45
VI
Table 3.2 List of indicators used for the assessment of ecosystem services ................................. 52
Table 3.3 Approach for the rapid assessment of ecosystem services ............................................ 54
Table 3.4 Land cover and vegetation structure types within urban parks of Xi’an city .............. 58
Table 3.5 Questionnaire samples at the 22 selected urban parks ................................................... 61
Table 3.6 Government documents on urban green spaces in Xi’an city ...................................... 65
Table 4.1 History of the structure of green spaces in Xi’an............................................................ 71
Table 4.2 Annual budgets for urban green spaces in Xi’an over five years (Source: statistical
year book of investment in Chinese cities 2010-2014)................................................................... 73
Table 5.1 Percentages of different land cover types within the 22 investigated urban parks of
Xi’an city.............................................................................................................................................. 79
Table 5.2 Vegetation structures within urban parks of Xi’an city ................................................. 81
Table 5.3 Percentages of different vegetation structures within the 22 studied urban parks of
Xi’an city.............................................................................................................................................. 83
Table 6.1 Respondents’ personal information.................................................................................. 88
Table 6.2 Types of often-used green spaces (percentage)............................................................... 89
Table 6.3 Use frequencies, times and durations of visits to green spaces ..................................... 91
Table 6.4 Use frequency by socioeconomic variables.................................................................... 92
Table 6.5 Duration per visit by socioeconomic variables ............................................................... 92
Table 6.6 Use frequency by variables of park characteristics ....................................................... 94
Table 6.7 Duration per visit by variables of park characteristics ................................................... 95
Table 6.8 Main reasons to use green spaces for recreation (percentage) ...................................... 97
Table 6.9 Activities in green spaces by respondents’ socioeconomic variables .......................... 98
Table 6.10 Respondents’ recreational activities by variables of park characteristics ................ 100
Table 6.11 Importance of different types of green spaces (percentage)...................................... 102
Table 6.12 Importance of different types of green spaces by socioeconomic variables ........... 102
Table 6.13 Perceptions of the importance of ecosystem services generated by green spaces
(percentage) ....................................................................................................................................... 104
Table 6.14 Perceptions of the importance of ecosystem services by respondents’ socioeconomic
variables ............................................................................................................................................. 105
Table 6.15 Respondents’ perceptions of ecosystem services by variables of park characteristics107
Table 6.16 Assessments of the condition of the entirety of green spaces in Xi’an city
(percentage) ....................................................................................................................................... 109
Table 6.17 Assessment of the condition of the entirety of green spaces in Xi‘an by respondents’
socioeconomic variables (percentage) ............................................................................................ 109
Table 6.18 Assessments of green spaces used daily by the citizens (percentage)...................... 111
Table 6.19 Assessments of green spaces used daily by the respondents based on socioeconomic
variables ............................................................................................................................................. 112
VII
Table 6.20 Respondents’ assessments by variables of park characteristics ................................ 114
Table 6.21 Respondents’ demands for specific features in urban green spaces (percentage)... 116
Table 6.22 Respondents’ demands by socioeconomic variables ................................................. 118
Table 6.23 Respondents’ demands by variables of the parks’ characteristics............................. 119
Table 7.1 Assessment of integrated indicators of regulating services ......................................... 139
Table 7.2 Integrated assessment of recreational services by park area........................................ 145

VIII
Acknowledgements

I would like to take this opportunity to convey my sincere acknowledgements to all


those who offered me their help during my PhD study.

The first people I thank is my supervisor, Professor Stephan Pauleit, Chair for Strategic
Landscape Planning and Management, Technical University of Munich. Without his valuable
supervision and endless patience, it would have been impossible to complete this research. His
critical directions, inspiring instructions and stimulating discussions have led to my full
exposure to scientific research and enabled me to grasp the core skills of research. All of these
learning opportunities have played a vital role in the completion of this dissertation, and will
continuously and positively contribute to my future work of academic research.

Second, I shall convey my special thanks to my second supervisor, Professor Jürgen Breuste,
Chair for the Urban and Landscape Ecology, in the department of Geography and Geology,
University Salzburg. His kind encouragements increased my confidence for the research. His
sharp and insightful comments and constructive suggestions have greatly helped me to improve
and finish my thesis.

Third, I am also grateful to the staff at the Chair for Strategic Landscape Planning and
Management for their warmhearted help during my PhD study.

Last but not the least, I shall also thank my family for their full understanding, support and love.

Tong Cai
December 2017

IX
Abstract
This study contributes to using ecosystem services as a concept for the design and planning of
green spaces, taking the City of Xi’an, Central China, as a case study. The aims of the study
were to (a) develop and apply an approach for the rapid appraisal of ecosystem services
provided by urban green space; (b) identify the relationships between park features and their
ecosystem services; (c) assess park visitors’ use habits, perception and appreciation of
ecosystem services; (d) draw conclusions for the future design and management of urban green
space in Xi’an and Chinese cities.

The methodology consisted of a combination of methods from the natural and social sciences.
Documentary analyses revealed the goals of green space plans and policies related to ecosystem
services supply. Visual inspection of satellite imagery provided data of land cover which was
complemented by a field survey to further analyse the types and compositions of land cover and
vegetation structures of green spaces. Data on the use, recreational activities, perceptions,
assessments and demands for green spaces by urban residents were collected by a questionnaire
survey. Descriptive statistics were employed to show general patterns, whereas statistical tests
such as t-test and one-way ANOVA were used to explore whether variations of socioeconomic
backgrounds and park characteristics were related to green space use, perception, assessment
and demands. Finally, based on the assessment indicators which were derived from the land
cover, vegetation structure and the user questionnaire, an integrated approach was developed
using simple rating rules to assess the ecosystem services generated by green spaces of Xi’an.

Document analyses show that the government of Xi’an has increasingly realized the
importance of ecosystem services. Green spaces’ plans and managements have also stressed the
need to develop multifunctional green spaces in order to provide ecological and recreational
green spaces.

The results of the field survey reveal that six types of land cover and eleven types of vegetation
structure can be distinguished in the 22 urban parks of Xi’an city. Vegetation accounts for most
of the area. Multi-layered trees covering shrubs and lawns and trees covering shrubs and lawns
are the dominant types of vegetation structure. Older parks usually show more variations in
compositions of land cover and vegetation structure. Larger parks located within the third ring
road have a higher percentage cover of lawn and the older parks located within the second ring
road of Xi’an have more playgrounds.

Most of the respondents in Xi’an frequently use the nearby urban parks for short stays (1-2
hours). Elderly people are the main user groups of green spaces. Respondents show preferences
X
for passive recreational activities when they visit green spaces, such as to enjoy the scenery, to
have large greenery and to contact with nature. The elders enjoy a broader range of activities in
green spaces compared with the other age groups. Also, respondents prefer to do activities in
the larger and newer parks and in the parks with 26-50% of multi-layered vegetation and
evergreen species.

Respondents have the view that shading, O2 release, recreational places, contact with nature and
aesthetics are the most important services provided by green spaces. They perceive these
ecosystem services more important in the small to medium and medium sized parks (5-10 ha
and 10-30 ha) with relatively higher percentage of multi-layered vegetation, tree and mature
trees and medium percentage of evergreen species.

Overall, less than half of the respondents positively assessed green spaces of Xi’an.
Comparatively, the more positive assessments are for vegetation coverage, microclimate,
cleanliness and scenic beauty while the provision of water area, water quality and air quality
were rated less favourably. Not surprisingly, for respondents which have children air quality
and availability of playgrounds were important criteria for assessment of parks. Older parks,
which are located within the second ring road and have a size of 10-30 ha and 0-5 ha, received
relatively higher positive assessments.

Moreover, the respondents clearly expressed the need for more shade, opportunities to watch
nature, flowers and better management of green spaces. However, demands differed between
age groups and levels of education. Older respondents wanted more quiet places and play areas
for children. Students wanted more sporting facilities and employed wanted more cafes and
restaurants. In addition, respondents interviewed in the older parks located within the second
ring road expressed wishes to increase provision of a larger number of ecosystem services such
as more safe places. Their needs for the ecosystem services are also influenced by the different
coverage of vegetation structures. For example, respondents in the parks with 26-50% of multi-
layered vegetation and 0-25% of evergreens demanded more shade.

Generally, in Xi’an city, urban parks provided similar levels of regulation for microclimate and
air quality and these regulation functions were predominantly considered to be high or medium
high. The older parks with a small to medium size and located within the second ring road
received higher assessments on regulation functions. However, all parks were considered to
have a low value in recreational services regardless of park character.

All the findings contribute new knowledge on regulating ecosystem services, landscape
preferences, leisure studies and citizen assessments and demands of ecosystem services for the
planning, design and management of urban green spaces. Six general strategies using
XI
ecosystem services as approaches can be proposed to advance the quality of green spaces: to
maintain sufficient green spaces in the limited available lands; to maintain the number of
vegetation structure types and plant species; to balance the provisions of regulating and
recreational services; to incorporate citizens’ considerations into the process of green space
planning; and to develop a system for better management of green spaces.

XII
Kurzfassung

Weltweit hat die Urbanisierung zu vielfältigen Problemen geführt wie etwa der Degradierung
von Natur und Umweltverschmutzung in Städten, einem hohen Verbrauch von Energie und
natürlichen Ressourcen wie auch hohen Treibhausgasemissionen.

Städtische Grünflächen können einige dieser Probleme durch die Bereitstellung von
Ökosystemleistungen vermindern. Wie städtische Grünflächen zu planen und zu pflegen sind,
um durch ihre Ökosystemleistungen die städtische Umwelt- und Lebensqualität zu erhöhen,
und eine nachhaltige sowie klimaresiliente Stadtentwicklung zu fördern, ist daher zu einem
wichtigen Forschungsgebiet geworden.

Diese Untersuchung soll dazu beitragen, den Ansatz der Ökosystemleistungen für die
Gestaltung und Planung verstärkt zu verwenden, mit der Stadt Xi’an in Zentralchina als
Fallstudie. Ziele der Untersuchung waren (a) einen Ansatz für die einfache und rasche
Erfassung und Bewertung von Ökosystemleistungen städtischer Grünflächen zu entwickeln
und zu erproben, (b) die Beziehungen zwischen der Beschaffenheit der Parkanlagen und den
Ökosystemleistungen zu analysieren, (c) die Nutzung der Parks, sowie die Wahrnehmung und
Wertschätzung der Ökosystemleistungen durch die Parkbesucher zu ermitteln, und (d)
Schlussfolgerungen für die zukünftige Gestaltung und Pflege von städtischen Grünflächen in
Xi’an und anderen chinesischen Städten zu ziehen.

Die Vorgehensweise bestand aus einer Kombination von natur- und sozialwissenschaftlichen
Methoden. Eine Dokumentenanalyse zeigte, wie Ökosystemleistungen in der städtischen
Grünflächenplanung gegenwärtig Berücksichtigung finden. Durch die visuelle Auswertung von
Satellitenbildern wurden Landbedeckungsdaten erhoben, ergänzt um ein Felderhebung für die
nähere Analyse von Landbedeckung und Vegetationsstrukturen in den Grünflächen. Daten zur
Grünflächennutzung, zu den Erholungsaktivitäten, der Wahrnehmung von Grünflächen, ihrer
Wertschätzung sowie der Nachfrage nach Grünflächen wurden durch fragebogengestützte
Interviews durchgeführt. Diese Daten wurden einerseits mit deskriptiven Statistiken
ausgewertet, während die Beziehungen zwischen Nutzungs- und Wahrnehmungsmustern und
sozioökonomischen Merkmalen der Parkbesucher mit der Parknutzung, ihrer Wahrnehmung,
Bewertung und Nachfrage nach Ökosystemleistungen durch t-Tests und einfache
Varianzanalysen (ANOVA) überprüft wurden. Schließlich wurde aus den Daten zu den
Landbedeckungsarten, der Vegetationsstruktur und den Interviewergebnissen ein integrierter
Ansatz zur Bewertung der Ökosystemleistungen städtischer Grünflächen entwickelt und für
Xi’an angewendet.

XIII
Die Auswertung der Planungsdokumente zeigt, dass die Regierung von Xi’an den
Ökosystemleistungen der Grünflächen zunehmende Bedeutung beimisst. Die
Grünflächenpläne enthalten auch Ziele zur Entwicklung von multifunktionalen Grünflächen,
die sowohl ökologische als auch Erholungsfunktionen erfüllen sollen.

In den Feldbegehungen wurden in den 22 städtischen Parks von Xi’an sechs


Landbedeckungsarten und elf Vegetationsstrukturtypen unterschieden. Die Vegetation nimmt
die größten Flächenanteile an den Grünflächen ein. Mehrschichtige Gehölzstrukturen und
Baum-Strauch-Rasen-Strukturen sind die vorherrschenden Vegetationsstrukturtypen. Ältere
Parkanlagen zeigen eine größere Variation der Landbedeckungsarten und
Vegetationsstrukturen. Größere Parkanlagen innerhalb der dritten Ringstraße haben einen
höheren Anteil an Rasenflächen und ältere Parks innerhalb der zweiten Ringstraße von Xi’an
zeichnen sich durch mehr Spielflächen aus.

Die Mehrzahl der befragten Parkbesucher Xi’ans besuchen regelmäßig die nahegelegenen
Parks für einen kurzen Aufenthalt von 1-2 Stunden Dauer. Ältere Menschen sind die
Hauptnutzergruppe. Die Befragten bevorzugen passive Erholungsformen wie etwa das
Betrachten des Parks sowie Natur zu erleben. Die älteren Parkbesucher üben eine größere
Anzahl von Aktivitäten in den Parks aus als die jüngeren Besucher. Die älteren Parkbesucher
bevorzugen dabei größere und jüngere Parkanlagen, sowie Parks, in denen mehrschichtige und
immgrüne Gehölze einen Flächenanteil zwischen 26-50% einnehmen.

Die Parkbesucher sind der Auffassung, dass die Verschattung, Sauerstofferzeugung, das
Vorhandensein von Erholungsflächen, der Naturkontakt und die Ästhetik die wichtigsten
Ökosystemleistungen sind. Sie halten die Bereitstellung von Ökosystemleistungen für
besonders hoch in den kleineren und mittelgroßen Parkanlagen (5-10 ha und 10-30 ha) mit
einer mehrschichtigen Vegetationsstruktur, großen Bäumen und einem mittleren Anteil von
immergrünen Arten.

Insgesamt bewerteten weniger als die Hälfte der Parkbesucher den Zustand der Grünflächen
von Xi’an positiv ein. Vergleichsweise besser bewertet wurden die Vegetationsbedeckung, die
klimatischen Leistungen, die Sauberkeit und die Schönheit der Grünflächen während das
Vorhandensein von Wasserflächen, die Wasserqualität und die Luftqualität weniger gut
bewertet wurden. Für Befragte, die Kinder haben, waren erwartungsgemäß die Luftqualität und
das Vorhandensein von Spielplätzen besonders wichtige Bewertungskriterien. Ältere
Parkanlagen innerhalb der zweiten Ringstraße und mit einer Größe von 0-5 ha bzw. 10-30 ha
wurden besser als die andern Grünflächen bewertet.

XIV
Die Befragten wünschten sich mehr Schatten, Gelegenheiten für die Naturbeobachtung,
Blumen und eine bessere Grünflächenpflege. Die Bedürfnisse unterschieden sich aber
zwischen den Altersgruppen und Bildungsniveaus. Ältere Parkbesucher wünschten sich mehr
ruhige Plätze und Spielplätze für Kinder. Studenten waren dagegen an mehr Sporteinrichtungen
interessiert und Angestellte wollten zusätzliche Cafés und Restaurants. Die Besucher der älteren
Parks innerhalb der zweiten Ringstraße äußerten den Wunsch nach mehr Ökosystemleistungen
und auch nach erhöhter Sicherheit. Ihre Nachfrage nach Ökosystemleistungen stand in
Beziehung zur Vegetationsstruktur. So wünschten sich beispielsweise Besucher von
Parkanlagen mit einem Anteil mehrschichtiger Vegetation von 26-50% sowie einem Anteil von
0-25% an immergrünen Gehölzen mehr Schatten.

Insgesamt stellten die Parks in Xi’an in vergleichbarem Umfang klimaregulierende und die
Luftqualität verbessernde Ökosystemleistungen bereit und diese Leistungen wurden als
überwiegend hoch oder mittelhoch eingeschätzt. Die älteren, kleinen bis mittelgroßen
Parkanlagen innerhalb der zweiten Ringstraße wurden dabei besser bewertet. Auf der anderen
Seite erhielten alle Parkanlagen eine niedrige Bewertung in Bezug auf ihre Erholungsqualität.

Diese Ergebnisse haben eine hohe Bedeutung für die Planung, Gestaltung und die Pflege der
Grünflächen, um ihre Ökosystemleistungen zu stärken. Sechs Strategien werden dazu
vorgeschlagen: Sicherung der Grünflächen in dicht bebauten Stadtgebieten; Sicherung
vielfältiger Vegetationsstrukturen und der Biodiversität in den Parks, Berücksichtigung und
sorgfältige Abwägung von Maßnahmen zur Bereitstellung von regulierenden und Erholungs-
Funktionen, und die Verbesserung der Grünflächenpflege.

XV
1 Introduction
Urbanization and Urban Challenges
Urbanization is a global phenomenon. In today’s world, the urban population accounts for 54%
of the total global population (WHO, 2015). By the year of 2050, it is expected that more than
66% of the world population of more than six billion people will live in cities (UN 2014).
Nowadays urbanization in Asia is proceeding at a rate much higher than the world average (Ooi
2009). Particularly, the growth of megacities with a population of over 10 million is
characteristic of urbanization in Asia (Ooi 2009). China as the biggest country of Asia has
experienced a particularly fast urbanization, industrialization, and a dramatic population
explosion in cities (Shuqing Zhao 2006). The percentage share of the urban population in China
more than doubled from 19% in 1980 to approximately 47% in 2010 and is excepted to
reached 59% in 2025 (UN 2010). In 1980, there were only 51 cities with half a million
inhabitants in China. Between 1980 and 1995, another 51 were added to the group and,
between 1995 and 2010, 134 additional cities in China crossed the half a million threshold. By
2025, it is expected that there will be another 107 cities with more than half a million
inhabitants in China (UN 2010).

Urbanization is considered as a basis for rapid social development and economic growth, but at
the same time it can have great negative impacts on the quality of human life and the
environment (NRC 2001), such as global warming and waste generation (Georgi and Dimitriou
2010; Grimm NB 2008; M. Feliciano 2008). In addition to these negative impacts both on
regional and global scales, urbanization also leads to environmental problems within urban
areas, for instance, poor housing conditions in crowded cities, traffic congestion, a shortage in
natural resources such as clean drinking water (Li et al. 2005; Min et al. 2011). Furthermore,
urbanization causes the destruction, fragmentation and the disturbance of natural ecosystems
(Blair 1999; McKinney 2002). The replacement of vegetation areas by artificial land cover,
such as buildings and roads leads to increased mean temperatures compared with surrounding
rural areas and the various human activities in urban areas lead to higher concentrations of
green-house gases and atmospheric pollutants (Grimm NB 2008; WHO 2002). Moreover, the
impervious surfaces decrease infiltration of rainwater, as a result, surface runoff is greatly
increased, which may lead to the heightened risk of flooding (Pauleit 2000; Tyrväinen 2005).
The deteriorating environment also results in physical and mental health problems for urban
dwellers (Shuqing Zhao 2006).

Now these urban problems are severe in most Asian countries where economic development
has been prioritized over environmental concerns. The most obvious example is the air in Asia’s
cities which is among the most polluted in the world (Ooi 2009). China has already exposed to
1
nearly all these problems of the environment since the 30 years’ rapid urbanization. In extreme
cases, the growing urban areas are becoming increasingly vulnerable to natural disasters, such
as, the frequent dust storms in the northern areas of China, e.g. Beijing and Shijiazhuang (Qian
et al. 2002; Shuqing Zhao 2006).

Importance of Urban Green Spaces


Since the awareness of environmental degradation, a national strategy for sustainable
development - China’s Agenda 21 (Department of Planning Committee of China 1994) - has
been established, aiming to reduce the negative environmental impacts of economic
development while at the same time aiming to maintain economic and social benefits, and
finally attain a durable provision for human needs (Kyrkou and Karthaus 2011). This strategy
emphasizes land use management and the development of greener cities (Shuqing Zhao 2006).

In response, the concept of the “ecological city” was proposed by city planners to meet these
challenges. The eco-city is a broad concept which encompasses economic, social and
environmental objectives (Yijun 2011). Within this concept, urban green space has been raised
as a topic of particular importance in recent years. Amount of public green spaces, public parks
and recreation areas are often mentioned as important factors to make cities more livable,
pleasant and attractive for citizens (Kyrkou and Karthaus 2011).

In reality, the role of urban green spaces has been recognized at least since the 19th century
when it was considered as a response to the problems created by industrialization in Europe and
elsewhere (Nicol and Blake 2000). Since then, urban green space is thought to be helpful of
better city development (Hansen et al. 2015; Jim and Chen 2008a). At present, green spaces
planning increasingly adopts the concept of urban green infrastructure, i.e. a coherent and
multifunctional network of green spaces that can, for instance, encompass public parks,
woodlands and farmlands near urban areas as well as private green spaces (e.g. gardens) to
address the challenges of urbanization. The multiple values of green spaces are conceptualized
as ecosystem services (Bolund and Hunhammar 1999; Hansen et al. 2015; Tratalos et al. 2007;
Tyrväinen 2001). In order to achieve the basic goal of sustainable development the following
ecosystem services are essential, for instance, moderating the urban heat island effect (Leo et al.
2016; Yuan 2007), removing pollutants from air and water (Cohen et al. 2014), providing
recreational opportunities and aesthetic enjoyment thereby raising the quality of urban life (Rall
et al. 2017; Tzoulas et al. 2007), protecting soil quality, abating noise and reducing energy
consumption (Cohen et al. 2014; Yeh 2009). Furthermore, green spaces can provide a basis for
biodiversity, such as provide habitats for plants and animals (Hiroaki T. Ishii 2010), also can
characterize and enhance the image of the cities, such as Central Park in New York, Royal
Parks in London, and Red Square in Moscow (Nasar 1997). Well-designed urban green spaces
can achieve many functions simultaneously, for instance, a reduction in energy consumption by
2
shading the buildings and a contribution to abating noise and air pollution around the buildings
(Bolund and Hunhammar 1999). A network of high quality green spaces can also attract
business, retail and leisure facilities and thus strengthen the local economy (Cianga and Popescu
2013).

China has a long history of garden planning, tree planting along roads and water edges (Yu et al.
2006). The core of traditional Chinese gardening is to imitate the beauty of nature (Lou 2003).
Therefore aesthetic qualities are rather emphasized than ecological values. This tradition
influences the greening policy and style of contemporary Chinese landscape architecture and
leads to researches about urban green spaces mostly focus on the design and planning of
recreational green spaces. However, there is still a severe lack of the recognition of ecosystem
services generated by urban green spaces and their benefits to the health of urban ecosystems.
Whilst an understanding of the multiple functions of urban green spaces, such as the above-
mentioned ecological and cultural ecosystem services’ provisions is reasonably well-developed
in science, it also needs to be well integrated into the planning, design and management
processes (Yli-Pelkonen V 2005). However, reliable and robust approaches to effectively
support this integration are often absent (Tyrväinen 2001). Therefore, it is necessary to develop
approaches that comprehend the ecosystem services supply and demand of green spaces, and
enhance the integration of the ecosystem services concept into green space planning and
management.

Xi’an as a very important inland city in China is very worthy of such research. In recent
decades, the city has experienced a high speed urbanization with tremendous land use changes
and hence a formed compact city fabric (Lu and Peng 2006). There are 11,166 persons per km2
in central city area (XUPB 2010). The high urban density reduces the accessibility of unsealed
surface for residents. Urban green lands are particularly limited compared with other land uses.
The government of Xi’an city has realized the severity of these problems and undertaken a
series of studies and policies, for instance, the 1995-2010 Master Plan of Xi’an for urban
environmental protection, in particular for air quality and water quality (XAEPB 2000) and the
“Urban Green System Plan (1995-2010)” for keeping ecological functions started in 1995.
Since then, the green cover ratio had risen to approximately 32% in 2010, which is slightly
higher than the “national garden city” norm of 30% (Xi’an Statistic Yearbook 2011). While
comparing the green coverage with the other national garden cities, such as 45% in Beijing and
38% in Shanghai, it is relatively low, not to mention such cities with higher greening quality
such as Paris, Stockholm (see the website http://www.stats-
sh.gov.cn/tjnj/nj11.htm?d1=2011tjnj/C1016.htm and http://www. bjstats.gov.cn/bjsq/csjs/). The
contributions of green spaces thus need more attention and conservation. The deterioration of
natural urban ecosystems and the request for high quality of urban life provide a challenge for
the city to maintain and increase the natural values and functions of existing green spaces, as
3
well as to introduce more green spaces into urban areas. In practice, in Xi’an city, little is known
about whether green space quality and the provision of ecosystem services have actually
improved, and whether the ecosystem services provided by urban green spaces meets the
citizens demands. Such unknowns provide development potentials and give us an opportunity
for choosing Xi’an as our case study, where the urban green spaces urgently need to be
improved by comprehensive planning, policies and management relying on ecosystem
services.

Aim and Objectives of the Research


Since the increasing realization of the importance of green spaces and their ecosystem services
to alleviate the problem of urbanization and hence to benefit sustainable urban development, the
overall aim of this study is to contribute to using the concept of ecosystem services for design
and planning of green spaces in Chinese cities. The objectives of the study are :

Objective 1: Develop and apply approach for the assessment of ecosystem services provided by
urban green space;

Objective 2: Identify the relationship between park features and ecosystem services;

Objective 3: Assess urban park use, perception and appreciation of ecosystem services by park
users;

Objective 4: Explore how green space planning and management can better contribute to
enhancing ecosystem services in Chinese cities.

Organization of the Research


The research will be presented in four stages (see figure 1.1) and assigned to seven chapters.

This chapter has outlined some of the issues underpinning current urban green spaces debates
and described the research objectives. Chapter 2 reviews the relevant concepts, typologies and
research development regarding urban green spaces and their ecosystem services, also
introduces the evaluation methods of ecosystem services mainly regarding non-monetary
assessments and questionnaires. It also summarizes a series of ecosystem services indicators
generated by urban green spaces using land cover mapping method for the assessment in the
next stage. Chapter 3 describes the case study – Xi’an City including the geographical location
and the socio-economic and environmental conditions, the research area and the urbanization
process. It also introduces the research methods used in this study, including a simple non-
monetary assessment, land cover mapping, questionnaires and document analysis. The simple
assessment method embraces an indicators system, assessing rules and aggregation rules. In
4
addition, the procedures of data collection during field survey were described in a few
paragraphs. Chapter 4 shows the urban green spaces plans and development of Xi’an city from
the establishment of first plan. Chapter 5 shows the types of land cover and vegetation structure
of green spaces in Xi’an city and discusses the research findings with other studies. Chapter 6
reveals respondents’ use habits, perceptions, assessments and demands for the green spaces in
urban parks of Xi’an city. It also tries to explain the reasons of respondents’ choices and
compares the questionnaire findings with other studies. Chapter 7 shows the assessment results
of regulating services using simple non-monetary assessments and the assessments results of
recreational services using questionnaires. It tries to explain the possible reasons for these
findings as well. Finally, chapter 8 will draw the final discussion and conclusions and make
suggestions for improvements of green spaces in Xi’an and in other Chinese cities.

5
Ecosystem Services of Urban Green Urban Green Spaces Planning
Spaces

Regulating Services of Urban Green Cultural Services of Urban Green


Spaces Spaces

Regulation of Micro-climate and Air Recreation of Urban Green Spaces


Quality of Urban Green Spaces

- Respondents’ assessments for


` Types of land cover and vegetation recreational services
structure
- Respondents’ use habits, preferences,
perceptions, anticipations of green spaces

Questionnaires
Land cover mapping
Document analysis

Figure 1.1 Research flow chart

6
2 Literature Review
Urban Green Spaces and Ecosystem Services
2.1.1 Concept and Classification of Urban Green Spaces
Definitions of green spaces are manifold and depend on the purpose of the study in question
(Swanwick et al. 2003). In some cases, the term urban green space refers to those public and
private open spaces in urban areas, which are predominantly covered by vegetation, either
natural or artificial, and are directly or indirectly available to the public (Baycan-Levent et al.
2009; Haq 2011; Jim 2003). By this definition, green spaces may include water features,
buildings and impervious surfaces such as paved trails and amenities, in addition to grasses,
trees, shrubs and other vegetation (Dunnett 2002; Forsyth 2003; Jim 2003; Lo and Jim 2011).

Due to the breadth of the definition, a vast variety of urban places can be described as urban
green spaces. Considering their characteristics, such as naturalness, type, size and location
within the cities, as well as their multiple functions ranging from nature conservation to
recreation, a distinction of urban green spaces from other green space types can be made. For
example, from an urban ecology perspective, urban green spaces can be broadly grouped into
four types of urban nature: remnants of pristine landscapes such as natural woodlands; remnants
of cultural landscapes such as intensively managed forests and agricultural lands; designed
landscapes such as artificial gardens and parks, and other urban greenery, e.g. street trees; and
urban wilderness, i.e. areas of former urban use recolonized by vegetation (Kowarik 2005). By
contrast, a multipurpose classification has been suggested by Swanwick et al. (2003). A
distinction is made between functions of green spaces (e.g. recreation), economic value and
habitats. Therefore, urban green spaces include parks, gardens, farmland, school grounds,
wetlands, woodlands, et cetera. Similarly, in Australia, Ambrey and Fleming (2014) considered
the multi-functionality and naturalness, and hence defined green spaces with examples
including public parks, community gardens, cemeteries, sports fields, national parks and
wilderness areas. In the context of China, according to the “Standards for the Classification of
Urban Green Spaces”, urban green spaces are divided into nine categories: public park, square,
nursery, green buffer, attached green space, residential green space, roadside/street green space,
riparian green space, and scenic forest (see Table 2.1) (CMC 2007). This classification
considers both multiple human functions and naturalness.

7
Table 2.1 Classification of urban green spaces in China (source: CMC, 2007)

Original urban Reclassified patch Abbreviation Description


green space type type
Public park Public park PU Open to the public (includes community
parks); provides education, pleasure and
recreation; has natural and planted
vegetation
Plaza-green space PL Open to the public; provides open space,
recreational opportunities; has planted
vegetation, seldom trees, most is shorter
shrubs and grassland; low diversity
Nursery NU Propagating and cultivating vegetation,
breeding and supplying saplings for urban
greening
Green buffer GR Linear corridors protecting high-voltage
transmission lines, shielding wind and
filtering pollutants; with planted vegetation
Attached green Attached green space AT Attached to industrial, commercial, utility
space land, etc.; contains planted vegetation; low
diversity
Residential green RE Attached to residential areas, including
space those planted and maintained by the
individual residents; includes communal
green space serviced by the local
community (excludes PU and PL).
Provides aesthetic, amenity-recreation
venues; limited plant diversity
Roadside green space RO Linear corridors between sidewalks, curbs
or island patches in crossroads; serves to
buffer people from traffic, reduces noise,
solar radiation, etc.; contains planted
vegetation; limited plant diversity
Other green Riparian green space RI Linear corridors along watersheds;
space primarily natural habitat type; often high
plant diversity
Scenic forest SC Open to the public; serves to protect and
preserve flora, fauna and provides scenic
amenities; a mosaic of remnant- or
naturalized habitat types

8
No matter which classification is used for urban green spaces, urban parks are always included
as a type of green space. According to the terms stipulated in “basic term standards” of Chinese
landscape architecture (CMC 2007), urban parks are defined as green spaces available for the
public to visit, appreciate, rest in, participate in activities and do sports. Lin included fire
protection, disaster prevention and refuge as further functions of such parks (Lin 1995).
Therefore, urban parks are used not only for public recreation (CMC 2007) but also for
assembly in case of fire, earthquake, and other emergencies (Lin 1995).

Parks have been classified in different ways, but size, facilities, functions and accessibility are
often the main criteria. For example, the US classification paid more attention to the distinctions
between park functions, such as playing areas, sports fields, zoos and memorial parks
(Bonsignore 2003). In the city of Los Angeles, the park definitions include recreation center,
pocket/mini park, open space, dog park and playground (Luo and Li 2015). In China,
researchers have attempted to divide the urban parks into historical sites, recreational parks,
comprehensive parks, community parks and ecological parks (Jin and Hong 2015). An official
classification - the “Classification of Chinese Cities” - has been published by the Construction
Ministry of P.R. China in 2007. Urban parks were classified into four categories and 31 sub-
types according to the classification (CMC 2007) (see Table 2.2). This classification focuses not
only on area and service radius, but also on the function and shape of urban parks. A park’s area
and service radius are the basic data for assessing the provision of ecosystem services to the
surroundings.

2.1.2 Concept and Classification of Ecosystem Services in Urban


Green Spaces
Ecosystem services are defined as the benefits humans derive, directly or indirectly, from
ecosystem functions (Costanza et al. 1997; De Groot et al. 2002; MA 2003). It is also used to
refer to the aspects of ecosystems utilized, actively or passively, to produce human well-being
(Davidson 2013; Fisher et al. 2009; TEEB 2010). Bolund and Hunhammar (1999) introduced
the term “urban ecosystem services” for the first time. The authors outlined the value and
benefits that urban residents can obtain from internal ecosystems located within a city.

Compared to undisturbed natural ecosystems, urban ecosystems have been influenced by


humans to varying degrees during the process of urbanization (McIntyre et al. 2000).
Considering entire urban areas as urban ecosystems, urban ecosystems are described as
complex, dynamic bio-physic-social entities, in which people live at high densities, or where the
built infrastructure covers a large proportion of the land surface (Maes et al. 2016; Pickett et al.
2008; Pickett et al. 2001). Human activities modify the natural components of urban
ecosystems by changing the vegetation cover, removing plants and top soil, introducing exotic
9
species, as well as polluting the air, water and soil (Redman 1999).

Table 2.2 Classification of urban parks in China (source: CMC, 2007)

Park type Subdivision Area Service


(ha) radius
(km)
Suburban Suburban national
scenery park scenery park
Suburban provincial
scenery park
Suburban municipal
scenery park
Country park Country comprehensive 50 5-10
park
Country wetland park
Country forestry park
Farm recreation park
Cemetery
Other country park
Municipal park Comprehensive park Municipal comprehensive park 10-50 3-5
District comprehensive park 10 1-3
Belt park Manmade belt park
Natural belt park
Local park Square
Residential park 5-10 0.5-1
Roadside park 0.001 0.3-0.5
Neighborhood park 0.001 0.3-0.5
Other local park
Specialized Educational park Zoo
park Botanical garden
Defense park
Other educational park
Theme park Sculpture park
Musical park
Amusement park
Other theme park
Children’s park
Sports park
Historical park
Memorial park
Other specialized park

10
Accordingly, the land use types in urban areas are changed. Natural areas have been
transforming into built-up areas including buildings, rooftops, sidewalks, roads and parking lots
(Arnold and Gibbons 1996). Ecological consequences of the altered urban land uses have
resulted in environmental problems, such as air pollution, heat island effect, habitat destruction
and degraded water quality, with impacts on human well-being and quality of urban life
(Bolund and Hunhammar 1999; Gardiner et al. 2013; McKinney 2002). Urban ecosystem
services can locally mitigate these problems, since they are primarily, but not exclusively,
related to the environmental functions provided by urban green spaces (De Groot et al. 2002;
Gardiner et al. 2013; Niemelä et al. 2010; Tratalos et al. 2007; Whitford et al. 1998).

Niemelä et al. (2010) discussed the ecosystem services of urban green spaces and adapted a
classification of these ecosystem services into three categories: provisioning services, regulating
services and cultural services (see Table 2.3). Provisioning services are material benefits that
ecosystem services generate, such as food, fresh water and timber. Regulating services, such as
the regulation of microclimate and air quality, gas cycles and water infiltration, are essential
preconditions for other ecosystem services. Cultural services are immaterial benefits that
humans derive from ecosystems, e.g. through recreation, health benefits and education (MA
2005). Similar to the TEEB (2010) and CICES (2016) classifications, this classification also
does not include the MA (2005) category “supporting services”, and habitat provision was
merged into the category of “regulating services”.

2.1.3 Contributions of Urban Green Spaces to the Provision of


Ecosystem Services
There is an increasing amount of research on the contributions of urban green spaces to
temperature regulation, air purification, biodiversity conservation, carbon sequestration and
noise reduction, as well as human health and recreation (e.g. Cao et al., 2010; Zhang et al.,
2014; Yin et al., 2011 and McPhearson et al., 2013). For instance, green spaces, and especially
large areas of city parks, can become islands which are cooler and more humid than the
surrounding urban environment on hot summer days (Barradas 1991; Martina Petralli 2009).
The possible maximum air temperature reduction attributed to green spaces was found to be 1-
2 °C, and sometimes even 5-7 °C (Katayama et al. 1993; Kikegawa Y 2006; Rosenfeld et al.
1998). Parks with a wide variety of vegetation also play a particularly important role in
amending the air quality in urban areas. For example, in Shanghai, China, Yin et al. (2010)
found that parks remove 2-35% of total suspended particulate (TSP), 2-27% of SO2, and 1-21%
of NO2 in different seasons (Yin et al. 2011).

11
Table 2.3 Ecosystem services generated by green and water areas in urban regions
(source: Niemeläet al., 2010)

Group Service Service-generating unit


Provisioning services Timber Different tree species
Food Different species in land-, freshwater-
and marine ecosystems
Fresh water, soil Groundwater infiltration, suspension
and storage
Regulating services Microclimate regulation Vegetation, water
Gas cycles (O2 production, CO2 Vegetation, especially forests
consumption)
Carbon sequestration and storage Vegetation, especially trees
Habitat provision Biodiversity
Reduction of air pollution Vegetation, soil microorganisms
Noise reduction Protective green areas, thick/wide
forests, soft surfaces
Rainwater absorption Vegetation, unsealed surfaces, soil,
water
Water infiltration Wetlands, vegetation,
microorganisms, water
Pollination Insects, birds, mammals
Production of topsoil and Litter, invertebrates, microorganisms
maintenance of its nutrient content
Cultural services Recreation Biodiversity, especially in parks,
forests and water ecosystems
Psycho-physical and social health Forests
benefits
Science education, research and Biodiversity
teaching

Urban green spaces provide recreational opportunities and amenities, which are key factors of
life satisfaction and psychological well-being (Jim and Chen 2006a; Leitner and Leitner 1996).
Parks provide a peaceful and relaxing setting for stressed urban inhabitants. They are “second
living-rooms” for people living in confined quarters. Moreover, they offer sports fields and
activity spaces for an increasingly overweight city population. Green spaces are also
environments for children to play and to discover the world; they are active and social places
for youth (Aarts et al. 2012; Granzin and Williams 2012; Woolley 2006).

Moreover, urban green spaces provide valuable habitats for animals and plants, and some
species respond strongly to environmental changes and dispersal corridors (Bolger et al. 2001).
12
Therefore, urban green spaces possess a considerable potential to contribute to the biodiversity
of cities (Kühn 2004; Millard 2010). For example, birds are attracted by the rich supply of
foods and the wide array of potential nesting sites in green spaces, and this sometimes includes
even rare and endangered species (Volker Heidt 2008). Parks often have higher biodiversity
than other types of urban green space such as woodlands, gardens and green roofs, with more
woody plant species, due to their often high levels of habitat diversity and microhabitat
heterogeneity (Jim and Chen 2008b, 2009). Therefore, parks can constitute particularly
important hotspots of biodiversity in the cityscape (Cornelis and Hermy 2004).

2.1.4 Relationship between Ecosystem Services and Features of


Urban Green Spaces
All the natural components of urban ecosystems, such as vegetation, forests, water, soil, animals
and biodiversity, can be generators of ecosystem services (see Table 2.3). Nevertheless,
biodiversity, water, and especially vegetation, can be considered the most important generators
of ecosystem services in urban green spaces.

Biodiversity
Biodiversity is related to the generation of urban ecosystem services (Cardinale et al. 2012;
Costanza et al. 2007; Haines-Young and Potschin 2010) (see Table 2.3). Relevant studies have
demonstrated that urban biodiversity contributes to multiple ecosystem services which are very
important for human well-being (Schneiders et al., 2012 and Thompson et al., 2014). Diversity
of plant species brings about more diverse animal communities, since different animals rely on
different plants. Furthermore, biodiversity is positively influenced by diversity of land uses.
Additionally, higher biodiversity offers greater opportunities for interactions of people with
nature (Miller 2002; Young 2010), helps foster an active lifestyle, and therefore benefits human
health (Chiesura 2004).

Water
Water surfaces have also been demonstrated as generators of ecosystem services in urban areas.
For example, water bodies provide recreational opportunities, such as boating, fishing, and
swimming (Qiu 2009). Water surfaces affect not only the surface temperature, but also the air
temperature, since water evaporates into the air absorbing excess heat in the process (Givoni
2000; Robitu et al. 2006; Sun and Chen 2012). Additionally, water possesses considerable
storage capacity for the heat from solar radiation (Spronken-Smith et al. 2000). A study from
the Petaling district, Malaysia, found that the lowest land-surface temperatures during the
daytime appeared in areas with high-density forests and large water bodies (Buyadi et al.
2014a). Moreover, water regulates the temperature at the micro- and local- scale, such as within
gardens and city parks (e.g. Givoni et al., 2000; Robitu et al., 2006; Huang et al., 2008 and Sun
13
et al., 2012). In Beijing, 197 water bodies were found to reduce the air temperature by amounts
ranging from 0.03 to 2.2 °C/hm, with an average of 0.54 °C/hm (Sun and Chen 2012). The
thermal capacity of water is more obvious in the less-humid and arid locations (Kassem 1994).
For example, in the arid area of Tabriz, Iran, the surface temperature of water was 5-9 °C cooler
than on land, including areas covered with vegetation (Amiri et al. 2009).

Vegetation
Vegetation, and especially trees, plays the most important role in generating ecosystem services
in urban areas, such as microclimate regulation, reduction of air pollution, and recreation
(Bolund and Hunhammar 1999; Niemelä et al. 2010).

Vegetation can moderate the excess heat that accumulates in urban areas through evaporative
cooling, shading, wind speed control and thermal capacity (Akbari et al. 2001; Andrade and
Vieira 2007; Volker Heidt 2008). Evapotranspiration by plants consumes energy from solar
radiation so that leaves and the surrounding air are less warmed up (C.S.B. Grimmond and
T.R.Oke 1991; Taha HG 1988). Trees absorb and reflect the biggest part of solar radiation, and
thus prevent the warming of the land surface underneath (Akbari et al. 2001; Katayama et al.
1993; McPherson 1994; Picot 2004). Vegetation affects air movements and heat exchange by
increasing surface roughness and by providing corridors for ventilation in built areas (Bonan
1997). The phenomenon that air temperatures in winter are affected by wind speed reduction
has been described by e.g. Akbari et al. (2001) and Volker Heidt (2008). However, this effect
critically depends on the type of vegetation. At night, tree cover may retain warm air under the
canopy, while open grassland may promote cooling of the air by convection (Armson et al.
2012; Potchter et al. 2006).

Urban vegetation, and especially trees, can directly or indirectly improve the local air quality
(Nowak 2002). Vegetation reduces the amount of air pollution in cities by trapping, absorbing
and degrading pollutants (Beckett et al. 1998; Paoletti et al. 2011). Dry deposition, including
stomatal uptake and non-stomatal deposition on the plant surface, is a major mechanism by
which plants absorb pollutants from the air (Fowler et al. 1989; McPherson 1998). Gaseous
pollutants such as NO2, O3 and SO2 may be directly absorbed through leaf stomata (C.L 2002).
Water-soluble pollutants can be dissolved in the water layer on the moist leaf surface
(McPherson 1994). After the absorption, gases will diffuse into the intercellular space and may
be absorbed by water films or react with inner leaf surfaces (Smith 1990).

Particulate pollutants can be captured on the leaf surface, which results in either trapping them
there or dispersing them into the soil during rain (C.L 2002; Givoni 1991b). However, the
intercepted particles are often re-suspended into the atmosphere, washed off by rain or dropped
to the ground with leaf and twig fall (Nowak et al. 2006). Consequently, vegetation is only a
14
temporary retention site for many atmospheric particles.

Furthermore, it should also be noted that vegetation can negatively affect air quality through the
emission of volatile organic compounds (VOC), thus promoting the formation of smog and O3,
particularly at ground level (Chameides et al. 1988). Consequently, it has been suggested to use
low-VOC-emitting tree species (Haider 1996; Nowak 2000). Moreover, the production of
pollen in spring by trees contributes to particulate pollution and impacts those who are allergic
to pollen (Beckett et al. 1998). Despite these potential negative effects, increasing vegetation is
considered a helpful way to reduce air pollution.

Vegetated areas are also important for recreation in cities. Vegetation generates different colors,
textures, sounds and feelings depending on the season, time of day and weather conditions
(Miller 1988), and it can be used as a screen and buffer, which plays an important role in
blocking objectionable views and offering beautiful views to the residents (Millard 2010; Miller
1988). Furthermore, for many people, planting, cultivating and managing vegetation is in itself
a form of recreation. Also, the fruits, nuts, leaves, wood and wood chips harvested from trees
and other urban vegetation add to the recreational activities processes. In many cases, green
spaces are supplemented with playgrounds and sports fields so that they make up a
comprehensive recreational system. At the city scale, green spaces can be created both as
landscape scenery and for recreation by using a multitude of available engineering and
landscape skills (Millard 2000).

Assessment and Evaluation of Ecosystem Services


Provided by Urban Green Spaces
2.2.1 Monetary and Non-monetary Evaluation of Ecosystem
Services Provided by Urban Green Spaces
Given the importance of ecosystem services generated by urban green spaces, their different
values should be taken into account so that suitable decisions on the planning and management
of urban green spaces can be made. The various values with regard to ecosystem services
generated by urban green spaces have been highlighted both theoretically and empirically.
Among the reviewed literature, Baycan-Levent and Nijkamp (2004 and 2005) valued the
ecosystem services of urban green spaces by different approaches, including ecological,
economic, social, planning and multi-dimensional valuation (see Table 2.4). This classification
emphasizes the complex and multi-dimensional structure of urban green spaces, and reflects a
comprehensive evaluation from several perspectives at a conceptual level.

15
Table 2.4 Valuation methods used for ecosystem services generated by urban green
spaces (source: Baycan-Levent and Nijkamp, 2004 and 2005a)

Values of urban green Functions and Valuation methods


spaces contributions
Ecological values e.g. biological Monetary valuation Avoided costs, cost-benefit
diversity, analysis, willingness to pay
genetic diversity, Non-monetary Species and ecosystem
climate regulation, valuation richness indices, biodiversity
absorption of air index, keystone processes,
pollution health index, ecosystem
resilience and stability
analysis, hierarchical
structure, population
viability analysis, eco-
regions or eco-zones
Economic values e.g. provision of Monetary valuation Travel cost, hedonic pricing,
food, wood, fruits, avoided costs, replacements
and energy cost, stated preference
methods
Social and cultural e.g. sports, Monetary valuation Travel cost, tourism
values sightseeing, revenues, willingness to pay
culture, Non-monetary Questionnaires, contingent
aesthetics, valuation valuation
social interaction
Planning values e.g. city image, Monetary valuation Cost-benefit analysis,
energy reduction, willingness to pay, hedonic
tourism, shelter pricing
Non-monetary Geographical information
valuation system method, multi-
criteria decision method
Multi-dimensional e.g. scientific Monetary valuation Financial analysis, cost-
values research, education, benefit analysis, cost-
policy effectiveness analysis,
tourism revenues, taxes
revenues
Non-monetary Performance analysis, multi-
valuation criteria decision method,
meta-analysis, value transfer,
rough-set analysis, fuzzy-set
analysis, content analysis

16
At a more general level, valuation methods of ecosystem services provided by urban green
spaces can be divided into monetary and non-monetary ones (see Table 2.4). Monetary
valuations usually include the stated preference method, hedonic pricing, contingent valuation,
avoided costs, and willingness to pay. Methods used to determine economic values have the
common characteristic of using monetary units as an indicator, which can be derived by
different methods. For instance, the provisioning ecosystem services, consisting of directly
marketable goods, such as drinking-water, food and raw materials are valued through market
observation of reference prices (Tong et al. 2007). By contrast, studies of regulating services
traditionally use stated preference methods to derive the values based on secondary markets.
Among the monetary methods, hedonic pricing, stated preference methods and contingent
valuation, were most frequently used (Costanza et al. 2006; Tyrväinen 2005) to evaluate
services such as recreational (Tyrväinen and Miettinen 2000) and aesthetic benefits (Sander et
al. 2010), as well as air quality (Bayer et al. 2009).

Although there has been a thrust to apply monetary methods to value ecosystem services, these
methods are inappropriate for the totality and plurality of values which are characteristic of non-
monetary indicators (TEEB 2010). Moreover, monetary values are generally highly context-
dependent with regard to the socio-ecology, politics, and economics at any given time. Most
economic, social and ecological values attached to urban green spaces are non-priced
environmental benefits, which include pleasant scenery or peace and quiet - i.e. potential
recreational opportunities (De Groot et al. 2002). Since such benefits are difficult to integrate
into the monetary assessment procedure, it is not always necessary to express the values of
ecosystem services in monetary terms.

Non-monetary valuations can be achieved using a large number of methods, such as contingent
valuation, questionnaires, geographical information system (GIS) methods, multi-criteria
decision methods, and performance analysis. The non-monetary assessments focus on
regulating and supporting services. Among regulating services, air purification (e.g. Jim and
Chen 2009 and Escobedo et al. 2011), the cooling effect of trees and parks (e.g. Shashua-Bar
and Hoffman 2000), as well as carbon storage and sequestration (e.g. Lal 2004), are of primary
interest.

Social, planning, and multi-dimensional values are most directly associated with cultural
ecosystem services, such as psycho-physical health, place value, sense of community and
identity, recreational and educational opportunities (Chiesura 2004). These values reflect
emotional, affective and symbolic views attached to natural ecosystems in urban settings, which
in most cases cannot be adequately expressed in terms of commodities or monetary units.
Therefore, methods for assessing the value of cultural ecosystem services are usually aiming to
establish their importance to human well-being (e.g. Ambrey and Fleming 2011 and Calvet-Mir
17
et al. 2012). For example, using the data from questionnaire surveys, multivariate exploratory
techniques, analysis of variance (ANOVA) and chi-squared test are the most common methods
adopted to assess the cultural ecosystem services. Some studies used both questionnaires and
interviews to document citizens’ preferences for recreation (Kaplowitz, 2001 and Qiu et al.,
2015). While these methods do not provide monetary valuations of recreation, they can provide
useful information on the importance of recreation to people in ways that monetary assessments
cannot. Also, some studies used a combined assessment of social and ecological values to
compare the provision of ecosystem services with the perception of well-being (Fuller et al.
2007). In recent years, an increasing number of studies have concentrated on applying more
integrated evaluation methods. For example, questionnaires integrating collaborative
geographic information system (GIS) techniques were used to determine people’s attitudes
towards urban green spaces (Balram and Dragicevic 2005).

However, most of the studies included in a review on urban ecosystem services (Haase et al.
2014) were conducted in developed countries, and nearly 60% of the studies focused on a
single service, while multiple service valuations are still rare. Many studies presented the urban
ecosystem services at a regional or city scale, whereas only a few were conducted at a
neighborhood or local-site scale. For the whole city, the values of ecosystem services are rarely
addressed at the operational level, and little work has been done on how the ecosystem services
approach may be used to better incorporate non-monetary values in urban planning (Gómez-
Baggethun and Barton 2013).

2.2.2 Indicators for the Evaluation of Ecosystem Services


Provided by Urban Green Spaces
Ecosystem service indicators are used to evaluate the supply of, demand for, and trends of
provision of ecosystem services, making the conditions, trends and rates of change of
ecosystem services easily understandable (Layke 2009; Sparks et al. 2011). Indicators should be
easily understood by policy makers and other non-scientific audiences so that the importance of
the ecosystem services generated by urban green spaces for the economy, environment and
human well-being will be well-appreciated. With the help of such indicators, ecosystem
services of urban green spaces can be represented and evaluated easily, and policy-makers can
make decisions based on evidence.

A broad range of indicators have been used to assess the ecosystem services provided by urban
green spaces (Sparks et al. 2011). Different indicators have been used to assess provisioning,
regulating and cultural services, such as those pertaining to recreation, biodiversity, ecology,
climatology and forestry (Sparks et al. 2011). For example, the recreation services can be
measured using indicators such as size and structurce of green spaces (e.g. Ulf G. Sandström,
18
2009 and Onder et al., 2011), the number of visits and the frequcy of use (e.g. Carlos Smaniotto
Costa et al., 2004), distance and accessibility of green spaces (e.g. Onder et al., 2011 and Gupta
et al., 2012), willingness to pay (e.g. Lorenzo et al., 2000 and Casado-Arzuaga et al., 2013), the
motivation of users (e.g. Carlos Smaniotto Costa et al., 2004 and Lo and Jim, 2010a), the
number of features with specific value, and increases in real-estate value (e.g. Nowak and
Dwyer, 2007). Regulating service indicators, such as assessments of air temperature regulation,
removal of air pollutants and noise reduction, were also referred in some of the reviewed
studies. For example, changes of temperature (Jim and Chen, 2008 and Tallis et al., 2011), air
pollutants such as, NOx, O3, benzene, particulate matters and VOC (Zhang et al., 2014), and the
amount of dB decreased by green elements (De Ridder et al., 2004), were used as indicators to
assess the corresponding regulating services of urban green spaces. Biodiversity maintenance
can be revealed by indicators of plant species and habitat continuity (e.g. Tzoulas and James,
2011 and Hermy and Cornelis, 2000).

Furthermore, attempts have been made to assess in a single study several or all factors including
ecological regulation, biodiversity conservation, recreational services,quantity and quality, and
planning of urban green spaces using composite indicators. For example, in order to assess the
quantity, quality, use and planning, as well as the development and management of urban green
spaces, the URGE (Development of Urban Green Spaces to improve the quality of life in cities
and urban regions) project (2004) developed a detailed worksheet of criteria and indicators at
both the city- and site level, including a set of unfolded indicators - shade area at noon, air
quality, species diversity, frequency of use, number of entrance and number of parking places
(Carlos Smaniotto Costa et al. 2004).

Baycan-Levent et al. (2009) developed a multi-dimensional evaluation method which considers


the quantity, accessibility, changes and planning of green spaces. The indicators can be
represented by e.g. the proportion of green space area per 1000 inhabitants, ten-year change in
green space, and existing planning goals (Baycan-Levent et al. 2009). In another study, Stern
(2010) developed a set of indicators, including both objective (e.g. stratification, vegetation
density and proportion of evergreen vegetation for the assessment of air purification; vegetation
coverage and tree proportion for the assessment of temperature regulation, and habitat size and
shape for assessment of biodiversity) and subjective ones (e.g. local residents’ assessments of
the use and facilities of green spaces).

In summary, the indicators varied among different assessment methods. There is an increasing
trend to develop and test ecosystem service indicators from a wide scale to a local/site scale
(Layke 2009; Sparks et al. 2011). Researchers should analyze, monitor, and efficiently measure
the conditions, characteristics, trends, and rates of change of ecosystem services in order to infer
suitable indicators (Layke 2009; Sparks et al. 2011). However, the research is still faced with
19
some difficulties and shortages. (1) The use of particular indicators depends on the availability
of data and policy objectives. The indicators should be measurable, able to convey information,
and be sensitive to change (Layke et al. 2012); (2) Some researchers use a single indicator or a
number of connected indicators to evaluate a specific ecosystem service, wheras others use
several indicators. While a single indicator is in most cases not sufficient for the assessment
aims, several indicators sometimes provide overlapping information; (3) Since the indicators
are used to reduce the complexity of assessments, they do not necessarily provide a
comprehensive understanding of all services; (4) Numerous indicators and metrics of urban
ecosystem services with different use quality and applicability are still conceptual in nature, and
it has not been clearly demonstrated how exactly these indicators are linked to the underlying
ecosystem services (Haase et al. 2014).

2.2.3 Processing the Indicators for the Assessment of Ecosystem


Services Provided by Urban Green Spaces
Various models are used to evaluate the provision of ecosystem services, including empirical,
GIS-based, statistical and survey-based models. A large number of studies have used empirical
methods or models to quantify ecosystem services using different indicators. This is especially
true for regulating services such as air pollution reduction and local climate regulation.
However, models used for urban valuation need to be adjusted to the complex and multi-
functional urban environment (Pataki et al. 2011). GIS-based models are useful for providing
analyses because indicators based on spatial data, such as land cover and land use derived from
maps are increasingly available. Consequently, aerial photographs and satellite imagery can
serve as a basis for estimating particular ecosystem services associated with vegetation types
and other landscape features. This method provides intuitive results, albeit suitable models are
still needed (e.g. BUGS project, 2004; Onder et al., 2011 and Gupta et al., 2012).

Surveys are often used to analyze the recreational services provided by urban green spaces. In
particular, questionnaire surveys can evaluate the use, activities, perceptions and preferences of
the users of urban parks (Chiesura 2004; Tyrväinen et al. 2007). For example, in a visual quality
assessment method, visitors were asked to rate a selected photo with conceptual parameters and
physical features on a desirability scale as 1, 2, 3, 4, 5 (1 the lowest, 2 low, 3 medium, 4 high
and 5 the highest). The final assessment was inferred by simply adding the total scores. This
method takes users’ appreciation and preferences as the basis, emphasizing both functional and
visual satisfaction.

A number of studies assessed the ecosystem services by rating the ecosystem indicators
according to their contribution to specific services, such as poly-functional assessment methods
and multi-dimensional methods. In these studies, indicators were graded with scores or levels
20
based on the amount, percentage, performance or ability (e.g. URGE project, 2004; Sandström,
2009; Stern, 2010). For example, in the poly-functional assessment method (PFAM) used in the
URGE project, each indicator was given a score. Zero points indicated that the criterion cannot
be fulfilled (e.g. water quality cannot be assessed if there is no water) (Carlos Smaniotto Costa
et al. 2004). In practice, the ranging scales can be changed locally. Researchers can decide
whether or not to apply weights to individual criteria according to the situation at hand. If a
weight is applied, each indicator would be multiplied by its impact factor and the weight.
Finally, the total score of all indicators under a criterion were aggregated to obtain the absolute
functional values. The results were expressed per criterion in scores between 1 and 5. One point
indicated that the criterion was very poorly fulfilled, and 5 points indicated that the criterion was
fulfilled very well. Generally, the PFAM assessment method can be used for individual green
spaces, and is adequate to compare several green spaces with each other. Also, the results of this
assessment can be visualized per criterion via GIS, using the software ArcView (Carlos
Smaniotto Costa et al. 2004). However, such methods are very time-consuming due to the large
number of indicators involved, and are not suitable for indicators which cannot be ranked.

Stern (2010) designed similar procedures for assessing both objective and subjective indicators
of ecosystem services. All the indictors were endowed with the same weight, and were ranked
based on percentage, e.g. vegetation cover was rated as: low (0-35%), medium (36-70%) and
high (71-100%) (Stern 2010). The remaining indicators were dealt with in a similar way, but
with different percentage scales. For example, the tree proportion ranged between 0-25% (low),
26-50% (medium) and 51-100% (high) due to the varying impacts of each indicator’s
percentage on ecosystem services. The final assessment results were deduced via a pairwise
comparison, and represented by the ranking of low, medium and high values.

In the multidimensional evaluation methods used by Baycan-Levent et al. (2009), quantitative


indicators (e.g. ratio scales) were obtained from government departments and qualitative
indicators (e.g. yes/no ordinal scales) were obtained from questionnaires. All criteria were
assessed using a series of simple rules, e.g. “higher is better”, “existence is better”, “experience
is better” and “increase is better”. This collection of diverse indicators limits the applicability of
quantitative analysis techniques and therefore calls for a regime analysis, which is based on a
generalization of pairwise comparisons. If there was no prioritization of the criteria in the
evaluation process, all criteria were assigned the same numerical weight value. Finally, the
performances of green spaces were transferred into scores via the regime vector module, after
which they could be ranked and compared. However, this multi-dimensional method requires a
set of characteristic attributes and judgment criteria from a methodological perspective
(Baycan-Levent et al. 2009).

The review of previous studies shows that a large number of indicators, criteria and models
21
have been used for the assessment of ecosystem services generated by urban green spaces.
Indicators were usually quantified (e.g. by percentage and amount, yes/no answer, or
mentioned/not mentioned), and then endowed with scores to indicate their impact/value on
ecosystem services of urban green spaces. The results of assessment were aggregated by
summing up all the indicators, and then shown on maps in GIS software or calculated using
modules. All these methods require the assistance of GIS software, since most of them used the
data from remote sensing, and some rely on GIS software to visualize the results. Additionally,
some assessment results were summarized by adding all the indicators and then averaged them
using simple arithmetic means, or by pairwise comparison and transformation into categories
such as qualified/unqualified, bad/moderate/good or low/medium/high.

The reviewed studies also suggest that the evaluation methods of urban green spaces can be
comprehensive, integrated and multi-disciplinary, by using a combination of data from remote
sensing, government departments, field surveys and questionnaires with the help of GIS, SPSS
and module analysis. However, integrated valuation methods such as multi-criteria analysis
need to be introduced to both local and regional planners.

Land Cover Mapping as an Assessment Tool for


Regulating Services of Urban Green Spaces
2.3.1 Land Cover Types within Urban Green Spaces and Their
Relation to Ecosystem Services
2.3.1.1 Definition and Classification of Land Cover
Since ecosystem services in urban areas are clearly directly related to land cover (e.g. MEA,
2005), a good understanding of land cover is required in order to plan and maintain ecosystem
services. Land cover maps, which represent the dominant landscape cover in a particular area,
offer a helpful tool to discern land cover types and hence the predominant ecosystem services.

Land cover can be defined as what can be observed on the earth’s surface (Di Gregorio 2000),
which results from past and present human activities and thereby leads to relatively stable
features lasting for a long time (Pauleit and Duhme 2000). Land cover is the result of land use
by changing the earth’s surface. Land cover, and specifically vegetation change, is one of most
sensitive indicators of environmental change such as deforestation, overgrazing and
urbanization, and hence has a profound effect on the characteristics of ecosystems. Surface
cover is a synonym for land cover, and is defined as the characteristics and elements of
vegetation and/or built-up structures on earth’s surface (Pauleit and Breuste 2011). A limitation
of surface/land cover is that it only provides a two-dimensional representation of earth’s surface.
22
Therefore, in this study, vegetation composition and layers will also be assessed to represent the
vertical dimension.

New technologies for data collection and processing are currently available for land cover maps
which offer considerable improvements over traditional methods. These new technologies
include integrated remote sensing and Geographical Information System software. Remote
sensing plays an important role in generating land cover information, not only due to its explicit
spatial representation of earth’s surface, but also due to its frequent temporal coverage and
relatively low observation costs (Batista and Haertel 2010). More recently, the Google Earth
(GE) tool has quickly been developed for applications in many fields. The high-spatial-
resolution images released by GE have been used as a free and open data source for land cover
classification and as a visualization tool for land cover maps (Mering et al. 2010). It has been
proved that land cover classification based on GE images is not statistically significantly
different from the classification based on QuickBird images (Hu et al. 2013). Moreover, GE
imagery performs well for mapping land cover types with good spatial characteristics in terms
of shape and context (e.g. road and river), and somewhat less so for grassland and woodland,
which requires a high spectral signature (Hu et al. 2013). Considering this limitation, other
ancillary data are suggested to be included to improve the classification accuracy of GE
imagery, such as observation data from field surveys (e.g. Cornelis and Hermy 2004).

In addition, as the research aims vary, the use of data is either detailed or rough, and the
sampling scale may be different, which leads to differences of classification of urban land cover.
Almost all the research on land cover at the city level includes roads/transport systems,
cemeteries, residential, industrial and commercial areas, car parks, agricultural land, water-
covered areas, parks, woodland, and abandoned land (e.g. Di Gregorio, 2000; Pauleit and
Duhme, 2000; Young and Jarvis, 2001; Akbari et al., 2003 and Livingston et al., 2003). At the
site level, the present literature reveals that urban parks usually include buildings, pavements,
open soil, vegetation and water (see Table 2.5).

The green-cover ratio, which is defined as the proportion of cover by vegetated surfaces in a
given piece of land, are normally used as an indicator in urban planning in Chinese cities. In
Pauleit and Duhme’s (2000) classification, vegetation is subdivided into trees, shrubs, rough
grass/ tall herbs, lawns and flower beds (see Table 2.5). Vegetation accounts for 80% of all land
covers, and lawn is the predominant type, covering 80% of the vegetated surface area. Some
researchers attempted to explore the vegetation in more detail. Sekliziotis (1980) categorized
vegetation according to the delineation of grass type, e.g. turf grass, rough grass and infested
grass (see Table 2.5). Young and Jarvies (2001) classified vegetation by taking into
consideration its type (e.g. trees and lawns), distribution (e.g. trees and tree rows), as well as
their height (e.g. tall herb vegetation and turf grass) (see Table 2.5).
23
Hermy and Cornelis (2000) saw the classification from a different perspective. They divided the
habitat units of urban parks into planar, linear and point-like, according to their surface shapes.
This classification has more detailed subdivisions, especially in terms of vegetation. For
example, the planar elements were defined as including forests, planted trees and shrubs, crops
and vegetables, fallow land, ponds, buildings, car parks, etc. (see Table 2.5). Furthermore, the
forest category was subdivided into deciduous wood, coniferous wood and mixed wood. Later,
in 2004, the authors simplified the classification based on the version from 2000, though the
differences were not very large (see Table 2.5). It was observed in field surveys that woodland
and individual trees occupied a large proportion among all the habitat types. Gao et al. (2010)
took a biotope mapping approach to habitat classification (see Table 2.5). This classification is
based on a maximum six-level hierarchal variable. The first level is the division of vegetation,
open soil, water areas and sealed impermeable “hard surfaces”, which include concrete, asphalt
and construction such as buildings and pavements (see Table 2.5). The remaining levels all
pertain to the variations of vegetation structure, which will be described in the next section.

Generally, the comparison of various approaches shows a broad array of land cover
classifications, which provides a useful basis for later research (see Table 2.5). The land cover
types of urban green spaces can be simply classified into three categories: concrete
construction, vegetation and water. Concrete constructions usually include buildings, icehouses,
roads, walls and pavements. Vegetation mostly includes trees, shrubs, grass, and lawns. No
matter which classification was used, the planar elements, especially forests, trees and
grasslands, accounted for the overwhelming majority of land surface area within urban green
spaces.

24
Table 2.5 Examples of land cover units within urban parks (own drawing)

Land cover Sub-divisions of Sekliziotis Pauleit andHermy and Young and Cornelis Gao et al.
land cover (1980) Duhme Cornelis Jarvis and Hermy (2012)
(2000) (2000) (2001) (2004)
Built / √ √ √ √ √
Buildings
Icehouse √ √
Roads √ √ √
Walls √ √ √
Pavements / √ √ √ √
Concrete
Car parks √
Open soils √ √ √ √ √
Vegetation Forest √ √ √
Trees / planted √ √ √ √ √
trees
Park-wood and √ √ √
orchard
Tree rows √ √ √
Shrubs √ √ √
Hedges √ √ √
Lawn √ √
Turf grass √ √
Rough grass √ √
Infested grass √
Pasture √ √
Grassland √ √ √ √
Hay meadow √ √
Tall herb √ √ √ √
vegetation
Flower beds √ √ √
Vegetables √ √ √
Arable crops √ √
Reed √
communities
Water Water / √ √ √
freshwater
Bank √ √
Pool √ √ √
Brook √ √

25
2.3.1.2Relationship between Land Cover and Regulating Ecosystem

Services
There are important differences between land cover types, such as impervious surfaces, water,
groups of trees, shrubs, rough grassland, arable land and flower beds, in terms of providing
ecological services such as removal of air pollutants, regulation of the microclimate, rainwater
infiltration and surface run-off. Some studies have explicitly referred to this topic. For example,
Carlson and Arthur (2000) used land cover changes to reveal the related changes of surface
microclimate and hydrology. Janssen et al. (2008) estimated the air pollution based on land
cover data, while Lehmann et al. (2014) used vegetation structure types as an approach for
identifying micro-climate effects. All of these studies demonstrated that large areas of woodland
and forest have obvious ameliorating effects on local microclimate, water run-off, and even air
quality in urban areas.

Within a green space, temperature is affected by the proportion of vegetation, pavements and
water bodies. Large vegetated areas bring about markedly lower temperatures than the
surrounding buildings and pavements (e.g. Nishimura et al., 1998 and Ayman Hassaan Ahmed,
2011). For example, in Tainan, Taiwan, a study showed that at a scale of 100 meters, under the
same environmental conditions, a 10% increase of vegetated surface reduced the temperature
by 0.14-0.32 °C (Kuo 2000). Another study in Merseyside, UK, showed that the surface
temperature at noon in an area with 50% of vegetation cover as around 7 °C lower than that of
one with 15% vegetation cover among four sites (Whitford et al. 2001).

Conversely, impervious surfaces generate comparatively higher temperatures (Chi-Ru Chang et


al. 2007; Weng and Yang 2004). Thus, the higher the proportion of impervious surface area, the
higher the land surface temperature (Chen et al. 2006; Weng 2001). For example, in Taipei,
Taiwan, Chang et al. (2007) found that, at summer noon, parks smaller than 3 ha with more
than or equal to 50% of paved surfaces were on average warmer than their surroundings.
Moreover, this effect increased with park size (Chi-Ru Chang et al. 2007). Temperature
measurements in greater Manchester, UK, showed that concrete and asphalt surfaces in parks
were some 18 °C warmer than the local air temperature (Armson et al. 2012). Therefore, it is
possible to decrease the temperature of urban green spaces by reducing the paved area.

Water also plays a fundamental role in evaporative cooling in hot climates (Akbari et al. 2001;
Givoni 1991b). Water features in parks come in different forms such as fountains, cascades,
water channels, ponds and novel water facilities including waterfalls and spray fountains
(Ayman Hassaan Ahmed 2011). A study of Tennoji Park in Osaka, Japan, showed that during
the daytime, the maximum surface temperature was around 30 °C in the pond and 39 °C on the
26
paved stone with a spray fountain, compared to a searing 50 °C on the paved area (Nishimura et
al. 1998). The temperature measurements in eight sites within a park in Cairo, Egypt, showed
that the air temperature increased by up to 4.7 °C going from tree canopies (32.5 °C), cascades
(33.2 °C), fountains (35.3 °C), lakes (35.5 °C), hilltops (35.6 °C), resting areas (35.7 °C), and
pavements (35.8 °C), to linear pathway (37.2 °C) (Ayman Hassaan Ahmed 2011).

Similarly, the regulation of air quality by urban green spaces is mainly mediated by vegetation.
Accordingly, air quality improvements increase with increased percentage of tree cover and
decreased mixing-layer heights (Nowak et al. 2006). Pollution in big cities is caused mainly by
traffic emissions, which are attenuated within a 100-m-wide patch of vegetation (Gilbert et al.
2003). Accordingly, the pollutant concentrations showed a notable decrease from the outside to
the interior of parks. A study in a park in Essen, Germany, measured the concentrations of NO,
O3 and NO2 in different land cover types at varying wind speeds. The results demonstrated
increased emissions of NO and NO2 along green areas, residential areas, secondary roads and
motorways (Kuttler and Strassburger 1999). A study conducted in 55 US cities found that in
urban areas with 100% tree cover (i.e., contiguous forest stands), the average air quality
improvements during the daytime of the in-leaf season were around 2% for particulate matter,
O3, and SO2 (Nowak et al. 2006). In some cities, short-term air quality improvements (one hour)
in areas with 100% tree cover were estimated to be as high as 16% for O3 and SO2, 9% for NO2,
8% for particulate matter, and 0.03% for CO (Nowak et al. 2006).

2.3.2 Vegetation Structure Types within Urban Green Spaces


and their Relation to Ecosystem Services
2.3.2.1 Definition and Classification of Vegetation Structure
A comprehensive classification of land cover is the basis for land cover mapping. However,
most classifications have primarily been divided into land-use types and habitat types (see Table
2.5). Land cover types only offer information on the horizontal dimension, while vegetation
structure concerns both horizontal and vertical information. Vegetation structure thus provides
more detailed information on vegetated land cover. Characterizing the structure of vegetation is
particularly important for assessing regulating ecosystem services such as microclimate
moderation and filtering of air pollutants, as well as habitat and recreational quality of green
spaces at the site level (Sandström et al. 2006).

For these reasons, the land cover classifications shown in Table 2.5 should be complemented
with more detailed information on vegetation structures. Vegetation structure can be defined as
“the organization in space of the individuals composing a vegetation type or association”
(Dansereau 2011). In other words, vegetation structure refers to the composition and height

27
variability of trees, shrubs and grasses in an area (Tzoulas and James 2010a).

Based on previous studies concerning vegetation structure within urban green spaces (e.g.
Wittig, 1983; Hercock, 1997, Tzoulas and James, 2010 and Gao et al., 2012), the vegetation
structure can be classified into trees, shrubs, and herbs-and-grasses (see Table 2.6). Without
considering height, Witting (1983) divided up the vegetation mainly by spatial layout of trees,
and classified it into rows of trees, groups of trees and single trees (see Table 2.6). Considering
the vertical variations of vegetation, Hercock (1997) proposed a division based on the height
ranges of trees, shrubs and herbs. Trees were thus classified into < 10 m, 10-30 m and > 30 m
categories. Shrubs were divided into < 1 m, 1-2 m, and > 2 m, and herbs into < 5 cm, 5-20 cm,
21-50 cm and 51-100 cm categories (see Table 2.6). Later, Tzoulas and James (2010) adapted
the height of trees, shrubs and herbs according to their local growth (see Table 2.6).

It is worth noting that Gao et al. (2012) identified the vegetation structure with more
comprehensive factors including cover of trees and shrubs, age of the trees, and plant types and
configurations (see Table 2.6). For example, areas with over 80% coverage by trees and shrubs
were classified as forests, which were further described based on their layers as one-layered,
two-layered and multi-layered (see Table 2.6). Within a forest, trees were divided by age into
those of less than 30 years, 30-80 years, more than 80 years, and clear cut (Gao et al. 2012) (see
Table 2.6). Even in the same age group, trees were subdivided into deciduous, deciduous and
coniferous, coniferous and swamp categories.

In general, all the above-mentioned classifications considered the height variations of


vegetation. An obvious feature in the study by Gao et al. (2012) is that plant types and
configurations were considered in the course of classification. Such a way of classification
provides much more information and details on the trees in the study area, and therefore helps
to explore the relationship between biodiversity/recreation services and habitats. In this regard, a
detailed characterization of vegetation is required to describe its influences on ecosystem
services such as biodiversity, recreation, air quality and microclimate.

28
Table 2.6 Examples of vegetation structure types within urban green spaces (own drawing)

Type of Subtype of vegetation structure


vegetation Wittig (1983) Hercock (1997) Tzoulas and James Gao et al. (2012)
structure (2010)
Wood Wood Forest < 30 years Deciduous/ One-layered/
30-80 years deciduous and two-layered/
> 80 years coniferous/ multi-layered
Clear cut coniferous/
swamp
Trees Row of trees Trees (> 30 m) Grove, clump of < 30 years Deciduous/ One-layered/
Group of trees Trees (10 m - 30 High trees (≥ 10 m) trees, thicket, tree 30 - 80 years deciduous and two-layered/
m) belt or avenue coniferous/ multi-layered
Single trees Trees (< 10 m) Low trees > 80 years coniferous/swamp
(5 - 9.9 m)
Shrubs Hedges Shrubs (> 2 m) Bushes
Shrubs (1-2 m) (1 - 4.9 m)
Shrubs (< 1 m)
Herbs and Herb plants Grasses High grasses and Succession area
grasses (51 cm - 1 m) forbs
(20 - 99 cm)
Hay meadow Herbs Meadow
(21 - 50 cm)
Pasture Sedges Low grasses and Grazed land area
(5 - 20 cm) forbs
(5 - 19 cm)
Lawn Ground covers Ground flora (≤ 4 Lawn
(< 5 cm) cm)
Reed communities
29
2.3.2.2 Relationship between Vegetation Structure and
Regulating Ecosystem Services
Vegetation composition
Vegetation composition and layers were found to have a significant influence on the
regulation of microclimates and air quality. Multi-layered vegetation, in which tree-,
shrub- and herb layers overlap, would promote the decrease of temperature and air
pollution within green spaces, mainly by increased shading, evapotranspiration, and
increased carbon uptake and storage per unit area (Armson et al. 2012; Souch 1993;
Spronken-Smith and Oke 1998).

There are many studies on temperature regulation by urban green spaces. Generally,
air temperature effects are smaller than surface temperature effects. In different urban
parks in the USA, a temperature study of ten different parks in the summer indicated
that the largest air temperature decrease at 1.5 m height was 5 °C in a park with a
mixture of grass, trees, shrubs and flowerbeds, followed by around 4 °C above open
grass and grass with large tree borders (Spronken-Smith and Oke 1998). Another
study conducted in Bloomington, Indiana, USA, measured and compared the air
temperature at 1.5 m height in three green environments: individual trees over grass,
clumps of trees over grass, and individual trees over concrete. The results suggested
that the cooling effect of groups of trees or individual trees over lawn was stronger
than that of the individual trees over concrete. By evening, the latter were
approximately 0.5 °C warmer than the surrounding areas (Souch 1993). Another study
in Utah, USA, recorded the surface and air temperature data for three days in
locations with vegetation over turf and over asphalt. The average air temperature and
surface temperature of asphalt was 31.5 °C and 56.6 °C, which was 0.4 °C and
25.9 °C higher than over turf, respectively. Finally, the results demonstrated that trees
planted into turf reduced the air temperature more strongly, since evapotranspiration
rates were higher than those for trees over asphalt due to longer stomatal closure
induced by higher leaf surface temperatures (Kjelgren and Montague 1998). Similar
conclusions can be inferred from the research of D. Armson et al. (2012). They
suggested that having both trees and grass would be even better than either alone, as
the surface temperature of grass in shade can be 4-7 °C cooler than the surrounding
air (Armson et al. 2012).

In China, a study of urban parks in Xi’an city showed that the air temperature at 1.5 m
height at sites with trees over shrubs and lawns was on average 3.76 °C lower than at
pavements, compared to a 1.54 °C reduction over lawns (Qin Yaomin 2006). A similar
study in Zizhu Park, Beijing, showed that the three-day-average temperature
decreased by approximately 2 °C with an increased quantity of combinations of trees
and shrubs (Wu et al. 2006). Similarly, a study measuring the air temperature at 2
meters of height in- and outside of 61 parks in Taipei revealed that a higher combined
cover of trees and shrubs correlated with cooler temperatures within the parks
compared to their surroundings, especially at summer noon (Chi-Ru Chang et al.

30
2007).

In addition to temperature regulation, multi-layered vegetation, e.g. trees over shrubs


and grass, have been suggested to possess a higher capacity for the reduction of air
pollutants. Moreover, this effect is much more pronounced during summer under all
conditions (e.g. Gromke, 2011; Nowak, 2006 and Wania et al., 2012). Air quality
models which quantify the effects of plant-related functional attributes on pollution
reduction are often based on relatively accurate estimates of vegetation structure (e.g.
Gromke, 2011 and Wania et al., 2012). Vegetation structure influences the canopy
area, which is a main determinant of the capacity for air pollutant removal.

The combination of trees and shrubs is important because planting bushes in or


between trees is more effective in terms of pollution reduction than a forest of the
same size (Heidt and Neef 2008). This may be due to the increased deposition of air
pollutants (Cavanagh et al. 2009; Yin Shan 2007 ). A study monitoring the particulate
matter in Riccarton Bush, Christchurch, New Zealand, explored the variations of
particulate matter by vegetation structures. The results showed that the particulate
matter at the monitoring sites with similar coverage of one/two-layered trees over
shrubs ranged from 21.3 µg/m3 to 23.1 µg/m3 (Cavanagh et al. 2009). However, in
areas with the same vegetation coverage but only a single tree layer, the particulate
matter increased up to 31.9 µg/m3 (Cavanagh et al. 2009). Similar results were
obtained in a study of particulate air pollution concentrations in six green spaces in
Shanghai, China by Yin Shan et al. (2007).

Tree Cover
In China, trees are usually defined as woody plants with a main stem higher than 5 m
(Chen 1990). In different studies, the size of the crown area (C.L 2002) or the tree’s
canopy was considered as the most important factor for mitigating excess urban heat
(Chi-Ru Chang et al. 2007; Shashua-Bar and Hoffman 2000; Weng and Yang 2004;
Zhou et al. 2011), because trees absorb and reflect the biggest part, probably up to
80% of the solar radiation, by their dense crown (Gillespie 1995; Picot 2004). Solar
radiation was thought to be the determining factor of temperature change, especially
in the summer. Canopy shading is determined by tree height and size, foliage density,
canopy volume, spacing of the trees, and growth factors such as cultivation and
irrigation regime (Potchter et al. 2006; Shashua-Bar and Hoffman 2000). Moreover,
trees can affect the temperature through transpiration, shading, altering the wind
speed, and modifying the storage and exchanges of heat among urban surfaces
(McPherson 1998; Nowak and Dwyer 2007; Spronken-Smith and Oke 1998; Weng
and Yang 2004). Usually, one or more of these microclimatic factors of trees produces
a high cooling efficiency. Therefore, it was suggested that more trees as opposed to
grassy lawns and shrubs should be planted to moderate the air temperature.

Trees influence the microclimate at a range of scales, from an individual tree to a


forest. The reduction of solar radiation induced by groups of trees is much greater
31
than that induced by isolated trees (Andrade and Vieira 2007). For instance, based on
modeling, increasing tree cover by 25% in Sacramento and Phoenix would decrease
the summer surface temperatures by 3.3-5.6 °C (Akbari H. 1992). A land surface
temperature measurement in Shah Alam City indicated that the temperature in groups
of trees was on average 3 °C lower than in other vegetation (Buyadi et al. 2014b). By
contrast, under the canopy of individual trees or small groups of trees, the reduction of
air temperature amounted to only 0.7-1.3 °C compared with near-open space, at
summer noon in Bloomington, Indiana, USA (Souch 1993). Therefore, increasing the
tree canopy coverage can be considered as an effective method to reduce excess urban
heat via vegetation management (Zhou et al. 2011).

Among trees, mature ones have the most stable cooling effects due to the large shaded
area and large total leaf surface (Huang et al. 2008). According to empirical
observations, such as those done by Hercock, 1997 and Tzoulas and James, 2010,
mature trees or high trees can be simply defined as trees with a height of more than 10
m. A fact that mature trees with large crowns contribute 80% of the cooling effect in
summer was corroborated by 1500 hours of temperature data from six partially shaded
areas in an urban complex in Tel-Aviv, Israel (Shashua-Bar and Hoffman 2000). In
fact, 80-85% of solar radiation in summer was neutralized by big trees with full leaf
coverage (Andrade and Vieira 2007). Even an individual large tree can have a
valuable effect on temperature reduction. For example, measurements in the suburbs
of Sacramento showed that the air temperature under the crowns of mature trees was
1.7-3.3 °C lower than in adjoining areas without trees (Taha HG 1988). Even a 3.6 °C
decrease of temperature during the summer daytime has been documented in the
shade of a large single tree in Miami, Florida (JR 1989).

On very hot days, temperature measurements in Gulbenkian Park and the surrounding
built-up areas in Lisbon, Portugal, showed that the lowest temperatures (30-33 °C)
were found in those parts of the park which were under dense tree shade (Andrade
and Vieira 2007). A study in the subtropical city of Gaborone, Botswana, showed that
at noon, densely vegetated areas were up to 2 and even 5 °C cooler than surrounding
sites (Andrade and Vieira 2007; Jonsson 2004). Moreover, the temperature changes
among the measurement sites were strongly associated with their shade degree.
Simply put: the higher the shade, the lower the temperature (Andrade and Vieira
2007; Jonsson 2004).

Similarly, a temperature comparison was conducted at three urban parks with different
vegetation compositions in the city of Tel Aviv, Israel. The results showed that an
urban park containing tall trees with a wide canopy had the maximum cooling effect
during daytime, reducing air temperatures by up to 3.5 °C. By contrast, dense,
medium-sized trees reduced the temperature by up to 2.5 °C. Remarkably, an urban
park covered with grass and a few low trees can be warmer than the built-up area
during the day (Potchter et al. 2006).

32
Trees have a stronger ability to abate air pollution than other vegetation types because
the filtration capacity of vegetation increases with an increase of the leaf coverage per
unit area of land (Givoni 1991b). In addition to their large canopy leaf surface area,
twigs and branches create a rough surface that effectively mediates the removal of
coarse particulate air pollution via a dry deposition process (Beckett et al. 1998).
Furthermore, the air turbulence created by trees increases the uptake of air pollutants
more than what was observed with shorter vegetation (Givoni 1991b). Usually, a large
leaf area for shading and high effective transpiration together increase the mitigation
of pollutant concentrations (C.L 2002). Therefore, the order of pollution removal
efficiency of vegetation is suggested to be trees > shrubs > grass. It has been
demonstrated that increased tree cover can lead to greater total pollution removal
(Nowak, 1994; McPherson et al., 1997; Scott et al., 1998; Nowak et al. 2006; C.Y.Jim
et al., 2007; Yin et al., 2011 and Paoletti et al. 2011). For example, in West Midlands
County, a model called “fine resolution atmospheric multi-pollutant exchange
transport”, predicted that an increase of total tree cover from 3.7% to 16.5% would
reduce average PM10 concentrations by 10%, from 2.3 to 2.1 µg/m3. Similarly, in
Glasgow, increasing the tree cover from 3.6% to 21% was predicted to reduce the
PM10 concentration by 7%, from 1.26 to 1.17 µg/m3 (McDonald et al. 2007).
Moreover, tall vegetation can capture particles more efficiently from the air by dry
deposition than low vegetation types (Gallagher et al. 1997; McDonald et al. 2007). In
fact, the tree crown diameter and height are the two most important factors affecting
the elimination of air pollution (C.L 2002; McPherson 1994). Therefore, dense and
continuous cover with large trees will contribute more to the improvement of air
quality than a cover of low trees (Jim and Chen 2008a).

For instance, large, healthy trees greater than 76 cm in diameter can remove 60 to 70
times more pollutants per year than smaller ones with diameters of about 8 cm
(McPherson 1994). Mature woodlands are obviously better than low vegetation at
removing gaseous pollutants, because their canopies have a much greater surface
roughness than other vegetation types (Manning 1980), and their large leaf area can
effectively control the dispersal of particulates for miles (Fowler et al. 1989). A
measurement of seasonal pollutant concentrations in an urban park in Shanghai,
which included SO2, NO2, and total suspended particulate (TSP), showed that in
summer the percentages of SO2 and NO2 decreased with increasing tree crown volume
and coverage. The tree crown volumes were calculated via the tree crown diameter
and height in relation to the surface area of the plots. Finally, SO2 removal efficiencies
were reduced from nearly 30% to 10% due to a four-fold decrease of tree crown
volume and coverage (Yin et al. 2011).

Evergreen Species
Air pollution removal by vegetation relies not only on tree cover, but also on the in-
leaf season. Trees remove gaseous air pollution primarily by uptake via leaf stomata.
The condition of the plants’ stomata can therefore alter the plants ability to absorb air
pollutants (Nowak 2002). Evergreen species that grow throughout the year can
33
contribute more to pollutant removal than deciduous ones, since air pollutant removal
occurs principally during the in-leaf season, and the increased length of the growing
season with longer foliage retention can result in greater total pollution removal
(Paoletti et al. 2011). This is especially important since in many Chinese cities the
emissions of air pollutants are increased during the winter. The efficiency of air
pollution removal by evergreen plants is therefore decidedly superior to that of
deciduous species (Jim and Chen 2008a).

Based on data from 55 cities in the USA, Nowak et al. (2006) demonstrated that an
increased in-leaf season of urban trees leads to greater total pollution removal because
the greatest effect of urban trees on O3, SO2, and NO2 occurs during the daytime of
the in-leaf season, when trees are actively transpiring water (Nowak et al. 2006).
Cavanagh et al. (2009) measured the concentrations of PM10 in a broadleaf evergreen
urban forest plot in New Zealand. The results showed that the mean PM10
concentration of 31.5 µg/m3 outside the forest was lowered to 22.4 µg/m3 within it
(Cavanagh et al. 2009). Manes et al. (2012) estimated the total removal of O3 by the
tree groups in Rome, Italy, for two years. They found that the great majority of O3 was
removed by evergreen broadleaves during the whole year, ranging from 0.002 to
0.049 Mg/ha, while in summer conifers cleaned the highest proportion of O3. As a
comparison, the O3 removal by deciduous broadleaves ranged from 0 to 0.042 Mg/ha,
which was significantly lower than that of evergreen broadleaves. Most notably, the
O3 removal in winter by evergreen broadleaves amounted to around 0.005 Mg/ha,
compared with practically 0 Mg/ha by deciduous broadleaves (Manes et al. 2012).
Green Space Features and their Relation to
Human Use and Perception
2.4.1 Human Use of Urban Green Spaces
There have been a number of studies on human use of green spaces in the past
decades. Urban residents typically appreciate and actively use urban green spaces,
albeit the time and frequency of use vary from site to site. Mornings and afternoons
are the preferred visiting times during a day. Stay time per visit is relatively short e.g.
no more than one hour. For example, in Singapore, 85% of the residents declared that
they used a park at least once per week and mainly on weekends (Yuen et al. 1999).
The time choice was similar to the findings in Bari, Italy, but only 58% of the
respondents in the study were frequent park users (Sanesi and Chiarello 2006). In
Guangzhou, China 70% of the citizens reported to be frequent park visitors, and
morning and evening were the preferred times to visit (Jim and Chen 2006a). In
Copenhagen, Afternoon and midday were the most preferred times of the day to visit.
Most citizens (82.2%) visited green spaces several times a month, and 74.3% stayed
for 15 - 60 min. (Peschardt et al. 2012). Similarly, in Germany, Chile and Spain, most
citizens stayed no more than one hour per visit (Priego et al. 2008).

The distance to green spaces has been shown to be one of the most important factors

34
that affect the frequency of use. A distance of 300 - 400 m, which corresponds to a 5
minutes’ walk, is often regarded as the threshold for daily visits to green spaces.
Beyond this distance, the number of visits declines rapidly (Coles and Bussey 2000;
Derkzen 2012; Grahn and Stigsdotter 2003; Nielsen and Hansen 2007). Those green
spaces closer to home and familiar to people are usually most welcomed by women
and children (Bell et al. 2004; Coles and Bussey 2000; Grahn and Stigsdotter 2003;
Jim and Chen 2006b). Nevertheless, other studies, such as Barbosa et al. (2007),
Comber et al. (2008) and Kessel et al. (2009), showed that at least 90% of the
respondents do visit green spaces within 900 - 1000 m, rather than 300 - 400 m, if
convenient public transportation is available. Therefore, a workable distance or
convenient access is generally considered an important factor for the use of green
spaces.

Coles and Bussey (2000) reported that the characteristics of a green space, such as the
size and conditions of facilities, the area of play grounds and the number of available
benches, have effects on its use (Coles and Bussey 2000). Larger parks are more
frequently visited than smaller ones because of the abundant provision of various
kinds of facilities and features (Coles and Bussey 2000; Jim and Chen 2006a;
Rydberg and Falck 2000). However, well-designed small green spaces were perceived
as more popular than the larger ones (Forsyth 2003). Some studies deemed that
gender, income and education also affect the usage of urban green spaces, though the
influences were complex and very location-dependent. For example, young and
middle-aged citizens were found to be the main park users in studies by Yuen et al.
(1999), Sanesi and Chiarello (2006), and Dunnett et al. (2002). However, similar
studies by Jim and Chen (2006 b) and Toftager et al. (2011) showed entirely opposite
results.
2.4.2 Recreation Activities in Urban Green Spaces
Citizens’ recreation activities in urban green spaces are the basis for understanding the
recreation services of green spaces, and thus provide information for the management,
design and maintenance of urban green spaces. Recreation can be divided into two
broad categories: active and passive recreation. Active recreation encompasses
activities in which people are physically active and mobile, that can be done
individually or in groups (Burgess 1988; Ho 2005). Passive recreation encompasses
activities that provide visual, emotional, socializing or relaxing enjoyment, and
include mostly stationary activities that are investigative and acquisitive in nature
(Burgess 1988; Ho 2005).

People therefore make various use of green spaces, including a number of active
activities such as exercising, jogging, bicycling, walking, waking the dog, playing
games and doing sports, as well as passive activities such as dating, resting, getting
fresh air, sightseeing, relaxing, enjoying and watching nature, or social contact and
interaction (Burgess 1988; Chiesura 2004; Dunnett 2002; Yuen et al. 1999). These
activities in parks are also a reason why people visit parks. Consequently, the more

35
activities a park offers people, the more reasons and opportunities people will have to
go and participate in these activities, meaning that park use will increase. Interestingly,
people in developed and developing countries engage in similar activities in parks.
For example, in the UK visitors went to green spaces mainly for walking, dog-
walking, playing with children, sitting to relax and enjoy nature, and exercising
(Dunnett 2002; Tzoulas and James 2010b). Similarly, visitors in Guangzhou, China
inclined to go to green spaces for relaxation, staying in a peaceful place, physical
exercise, nature appreciation and aesthetic pleasure (Jim and Chen 2006a). By
contrast, in Beijing, China, the young respondents preferred to go to green spaces for
walking, sightseeing and jogging (Zhang et al. 2015). There were also other preferred
activities in green spaces, such as to have fresh air in Bari, Italy, (Sanesi and Chiarello
2006), or to encounter wildlife and to enjoy natural landscapes in the UK and Sweden
(Grahn and Stigsdotter 2010; Özgüner and Kendle 2006). In reality, the combination
of the above activities is what makes well-designed urban green spaces particularly
attractive. Therefore, urban green spaces with a variety of facilities and features will
be more popular, better known and more frequently visited (Grahn and Stigsdotter
2010).

Despite the large number of common uses, people from different social groups use
green spaces differently. In fact, age, gender, race, education, employment and
residence all appear to influence visitors’ activities in urban green spaces. For
example, researchers who have compared participation by age found in general that as
the age of visitors increased, their participation in outdoor recreation and physical
activity decreased. Especially, with visitors’ increasing age there is a decline in the
preference for participation in physical activity (Burgess 1988; Chiesura 2004;
Dunnett 2002; Yuen et al. 1999). Sports and meeting others were pursued more by the
young, while relaxing, being with children and enjoying nature appeared to be the
preference of adults and the elderly (Chiesura 2004). In addition, males typically
chose urban green spaces for sports, which was not the case for females (Sanesi and
Chiarello 2006). Females usually engaged in stationary activities that are associated
with childcare and family groups (Sanesi and Chiarello 2006).

The amount and quality of green spaces can affect the patterns of residents’ activities,
frequency of everyday recreation, and the way in which knowledge about the
environment is acquired (Herzele and Wiedemann 2003). Larger green spaces
attracted more physical activities (Qiu and Nielsen 2015). Access is also particularly
critical for the recreational use of green spaces. Proximity will increase the
recreational activities in green spaces due to less time being available in today’s
society (Coles and Bussey 2000). Indeed, having green spaces nearby in the
neighborhood has been shown to increase outdoor recreational activities (Grahn and
Stigsdotter 2003; Toftager et al. 2011).

Moreover, vegetation has been shown to be one of the characteristics that influence
recreation. Trees and grass might foster daily activities and experiences important for
36
children’s development (Taylor et al., 1998). In urban parks of US cities, grass is
positively associated with relaxation and rest, indulging, eating, play and reading
(More 1985). By contrast, shrubs are negatively associated with most activities. Their
very presence reduces the space available for activity, but the effect is not strong as
the correlations are neither large nor significant (More 1985). Trees have interesting
effects in two dimensions - number and size. The number of trees is positively
correlated with many activities, but especially with sleeping and reading, likely since
these two activities may occur in the shade. Large trees fostered conversing, play,
reading and sleeping (More 1985). Preferences for such “green exercise” may result
from being physically active and directly exposed to nature (Pretty et al. 2005). After
all, doing exercise along with others (even strangers) while enjoying beautiful scenery
and perhaps fresh air, is a desirable thing in itself for most people (King et al. 2000).
For example, women in New York felt more fulfilled when undertaking exercise in
the prospect Park than in other places, e.g. streets and indoor gyms (Krenichyn 2004).
Furthermore, parks were identified by runners as a more restorative environment than
city streets or sidewalks (Bodin and Hartig, 2003). Notwithstanding, how respondents’
preferences for recreational activities can be influenced by different vegetation
structures, such as vegetation types, layers and compositions, still needs further
consideration.
2.4.3 Respondents’ Perceptions of Urban Green Spaces
Perceptions are an important factor which determines the use of urban green spaces,
and they are also the basis for assessing them. Perception is the gathering of
information through the senses including seeing, hearing, touching, tasting, and
smelling (Dawes 1972). Characterizing individual park users’ perceptions can better
support the integration of all interested groups, optimize local benefits, and increase
success in community and green space planning by using cooperative management
strategies (Gerd 2002).

Measuring respondents’ perceptions towards urban green spaces has been mostly
done through structured questionnaire surveys (Balram and Dragicevic 2005). In
some studies, respondents ranked visual aesthetics and shading as the most important
values of green spaces (Lorenzo 2000; Wang et al. 2016). A survey of urban forests in
Finland showed that residents commonly thought the social values such as outdoor
activities and exercise to be important, while the benefits associated with
environmental quality, such as wind protection, pollution mitigation and climate
regulation were perceived as of relatively low importance. Economic values, such as
timber production, were considered rather unimportant (Tyrväinen 2001). Similar
research in Guangzhou, China also supported such findings. People appreciated more
the recreational function of green spaces and their contribution to the improvement of
the environment. In comparison, their functions as wildlife habitats and their
economic value attracted little attention (Jim and Chen 2006a). However, the
respondents in Bari, Italy and Phnom Penh, Cambodia considered the function of
climate improvement more important than leisure (Sanesi and Chiarello 2006; Yen et

37
al. 2016). Thus, outdoor recreation and aesthetic qualities are generally widely
highlighted, and economic values of urban green spaces are ranked relatively low.

Common perceptions of urban green spaces can be linked to different societal and
personal attributes. While nature-related benefits (e.g. wildlife habitat) were
highlighted in the USA (Lorenzo et al., 2000) and northwestern European cities
(Tyrväinen, 2001 and Tyrväinen et al., 2007). Climatic (e.g. shading) and
environmental contributions (e.g. air pollution reduction) received the most attention
in China (Jim and Chen, 2006a) and Italy (Sanesi and Chiarello, 2006). Some studies
revealed that personal backgrounds particularly influence the perception of urban
green spaces. A study in Guangzhou, China showed that younger (<30 years old) and
richer residents (> RMB 3000/month) were more likely to emphasize the importance
of ecosystem services generated by green spaces (Jim and Chen 2006a). In Bari, Italy,
the climatic function of urban green spaces was particularly identified by males over
65 years of age (Sanesi and Chiarello 2006). Regarding urban green spaces as places
for socialization and leisure was more prevalent among the younger of both genders,
while seeing them as places for children to play in was favored by women (Sanesi and
Chiarello 2006). By contrast, in Helsinki, Finland, women paid more attention to
recreational and health benefits than other groups (Tyrväinen et al. 2007).

While residents have common perceptions of green spaces, differences were found
across different study areas and social groups in the reviewed studies, because
perceptions are usually affected by a joint effect of social, cultural, environment and
personal factors in particular societies. Such differences still need further research in
order to provide suitable green spaces for each group of users. For this purpose, it is
important to carry out detailed surveys, interviews and observational studies, so that
more in-depth insights can be acquired on people’s behavior, preferences and
perceptions. In practical policy makings, common perceptions are relatively easy to
meet. However, the individuality of social groups remains a challenge for green space
management.
2.4.4 Respondents’ Preferences for Urban Green Spaces
Urban green spaces with their natural features such as trees, water, flowers and
grasses are predominately welcomed (Yuen et al. 1999). Residents in East Midlands,
UK declared they wanted a more natural appearance of urban green spaces (Bell et al.
2004). In the USA, half of the residents appreciated flowers and vegetation at parks
(Pincetl and Gearin 2005). By contrast, water features such as streams and rivers were
found to be highly preferred in the studies by Dwyer et al. (1989) and Rydberg and
Falck (2000).

Apart from natural elements, people share an appreciation for other features of urban
green spaces. They usually prefer peace, quiet and cleanliness in urban green spaces
(Jim and Chen 2006b; Tyrväinen et al. 2007; Yuen et al. 1999). Some wanted more
children’s playgrounds, athletic fields and shelters (Özgüner and Kendle 2006). Some

38
asked for more recreational facilities, such as tables, benches and drinking water
(Herzele and Wiedemann 2003; Oguz 2000). Furthermore, some respondents wanted
more species of vegetation, playgrounds and recreational activities provided by green
spaces (Yen et al. 2016). Moreover, visitors from different gender and age groups may
have different demands for urban green spaces. For example, green spaces closer to
home were most welcomed by women and children (Bell et al. 2004). Last but not
least, convenient public transportation was highlighted by the elderly in some cases
(Jim and Chen 2006b; Sanesi and Chiarello 2006).

In summary, the literature review in this section reveals that a majority of the studies
addressing the respondents’ use habits, recreational activities, perceptions,
assessments and anticipations of urban green spaces were conducted in developed
countries or cities, particularly in Europe and North America, and few similar efforts
were made in developing countries. Citizens’ use behaviors, perceptions and
expectations are rarely taken into consideration in the planning and management of
green spaces in developing countries. During the city-construction history of China,
policymakers seldom enquired the public’s preferences and opinions. Consequently,
urban green spaces may fail to meet consumers’ demands, prevent certain groups of
people from using them, attract undesirable elements or activities, and in extreme
cases, be abandoned by users (Burgess 1988). In order to fill this research gap, more
studies need to be conducted in developing countries that face high pressures in the
conservation, provision and development of urban green spaces. Conversely, engaging
citizens in the planning and management of urban green spaces can bring more
benefits to more people (Dunnett 2002; Marcus CC 1998). In China, economic growth
and increased public awareness of social affairs has created a need to consider public
perceptions during the planning and development of urban green spaces. As new
green spaces will be built and old parks with outmoded design need renovation, the
government should incorporate citizens’ preferences and expectations into decision
making (Lorenzo 2000).
Ecosystem Services as an Approach to Support
Green Space Planning
A high speed of urbanization leads to dense cities and land use competition, and under
land use pressure, the planning of green spaces is always secondary to commercial
planning. However, with increasing awareness of the importance of green spaces for
sustainable development, the status of urban green spaces in urban planning has been
raised, especially from ecological, cultural and economic perspectives (Baycan-
Levent et al. 2009; James et al. 2009). The concept of ecosystem services is new and
unfamiliar to many actors in green space planning, although the issues encompassed
by the concept have been included in urban planning principles based on sustainable
development. Introducing the concept could therefore help promote the development
of a more comprehensive understanding of ecosystem services in the minds of civil
servants, decision makers and citizens alike (Niemeläet al. 2010).

39
Generally, five principles that can be considered in green space planning were
repeatedly described in the literature: (1) Accessibility for the public will increase the
green spaces’ attraction and use (Grahn and Stigsdotter 2003; Nielsen and Hansen
2007); (2) Coherence and sufficient size of green spaces can increase the ecosystem
services, even if some ecological benefits can be provided by well-managed green
spaces of limited size (Tratalos et al. 2007); (3) Variation in character and type of
green spaces provides better functions - e.g. natural, recreational and well-kept parks
with vegetation and wetlands (Gill et al. 2007; McPherson et al. 1997; Nowak et al.
2006); (4) Distribution of green spaces influences the quality of their ecosystem
services, e.g. public parks, quiet places and fresh-air-generating zones which are
located in high-density residential areas may enhance the provision of recreation and
clean air (Givoni 1991b; Heidt and Neef 2008); (5) Public participation is encouraged
in the process of green space planning. Involving the public in decision making is a
driving force for the acceptance and implementation of green space planning (Li et al.
2005). Practitioners, researchers and politicians should deal with the issues of the
entire planning process of urban green structure through a communicative process.

In order to enhance their ecological functions, urban green spaces should be viewed in
an interconnected way with consideration of their size, diversity and distribution
within the city, their history, as well as the design and management of individual
green spaces (Gilbert 1989). The interconnection of all types of green spaces within a
city was termed green space structure (Kaliszuk and Szulczewska 2005). Therefore,
green spaces should be considered as an organized structure rather than many
individual amenities. Every city has its own distinctive type of green space structure
that results from the interaction of certain natural and human processes (see Table 2.1).
Based on the origin and functions of green spaces, three layers of green structure can
be identified: (1) The layer of natural landscapes that were there before the city came
into existence. This includes forests, wetlands and arable land; (2) The urban layer or
the amenity green structure. This layer includes public parks, residential, industrial
and commercial green spaces, and city squares; (3) The linear layer, such as roads and
green buffers (e.g. Kendle and Forbes 1997).

The ecological perspective is increasingly being applied to the study of urban green
space planning. The corresponding studies provide valuable indicators of the
ecosystem services generated by urban green spaces (as described in previous
sections), as well as approaches that may optimize these services, such as the planning
concepts of green belts (e.g. Bo et al., 2011), green corridors (e.g. Hellmund and
Smith, 2006; Jim and Chen, 2003), green wedges (e.g. Li et al., 2005) and green
infrastructure (e.g. Pauleit et al., 2011). For example, in some European cities, such as
Helsinki and Oslo, green space corridors were planned for multiple social and
ecological functions, including recreation, biodiversity conservation and ventilation
(e.g. Bednarek, 1990; Blazejczyk and Kuchcik, 2001, Werquin et al., 2005). The
contribution of woodlands to the reduction of air pollution from roads was explored,
and detailed guidelines for the design of tree belts along the roads were developed
40
(e.g. Beckett et al., 1998). Additionally, some studies discussed the relationships
between green space structure planning and climate regulation, such as the ones
conducted in Munich (Pauleit and Duhme, 2000) and Warsaw (e.g. Szulczewska and
Kaliszuk, 2003). A green-finger design was considered to concentrate the climatic
function of green spaces in the case of Warsaw (Szulczewska and Kaliszuk, 2003). In
developing countries such as China, such studies mainly focused on a small number
of large metropolitan regions. For instance, Jim and Chen (2003) applied landscape
ecological principles to the planning of green spaces in Nanjing, China. Li and Wang
(2003) proposed a method for the evaluation, planning and prediction of ecosystem
services of urban green spaces in Yangzhou City in China, using landscape principles.
In Beijing, the riverside green corridors can be used to create recreational open spaces
for people with easy access, and encompass a variety of plant species, textures and
colors. Roadside trees can separate the traffic line and reduce the noise. Moreover,
their combined network can act as a green corridor for wildlife. For green space
structure planning, the “star” settlement structure managed to connect green wedges
(urban parks, urban forests and agricultural land) with green corridors (road trees and
riverside trees) to reduce the heat island effect and enhance air ventilation (Li et al.
2005).

The above planning approaches consider urban green space as an integrated structure
and attempt to optimize many of the functions of urban green spaces. Indeed, many of
these functions are often interconnected. For example, a well-connected green
structure can help protect, improve and create ecological corridors that enable plants
and animals to move between core habitat areas. This may provide a better chance of
survival for vulnerable populations (Tjallingii 2005). Combining green and blue (i.e.
water) structures will improve both sustainable water management and the quality of
green spaces for recreation (Tjallingii 2005).

This literature review indicates that planning practices of green spaces in dense cities
need to permit the inclusion of urban green spaces that are close to people, coherent
and of sufficient size, varied, well-maintained, and are places where people can
engage in development. These rather general and non-quantified quality guidelines
may serve to provide various ecosystem services, benefits and values, but will need to
be adapted to site-specific conditions as a prerequisite for functional densification
(Berg et al. 2012). Moreover, detailed information on ecosystem services of green
spaces is still lacking, and its application in green space structure planning has not yet
been discussed in detail, especially in developing countries. Therefore, more
integrated green space planning and management practices are required to meet the
ecological and current social demands in environments under strain. More detailed
targets and goals of green space structure planning should be established using
ecosystem service approaches and considering the interactions between different
ecosystem services and green space structure planning. More research is needed to
further understand which properties and potentials of urban green space structure can
provide urban ecosystem services in dense cities.
41
Research Questions
The overview presented above shows a number of themes of ecosystem services
provided by green spaces, including their definitions, classifications and various
assessment methods. Beyond these research achievements, several open questions and
research gaps, such as multidimensional assessment methods and incorporation of
peoples’ assessments into green space planning, need to be highlighted and explored
further. The focus of this work therefore is the development of integrated assessment
methods for ecosystem services using land cover mapping and visitors’ questionnaire
surveys, and the connection of ecosystem services to green space planning and
development.

The research questions themselves are split into three related sections: the assessment
of regulating services, assessment of recreational services, and green space planning
and development. Each of these is first discussed as an independent issue, and then
collectively to show the relationships between them.

Although the thesis reviews a number of assessment methods for ecosystem services,
rapid and simple assessment methods using land cover types as assessment indicators
are still needed in the context of limited data and time. The review presented in this
chapter leads to the following research questions which further detail the aim and the
four objectives of this research as outlined in chapter 1.3.

Objective 1: Develop and apply approach for the assessment of ecosystem services
provided by urban green space.

Based on this objective, the corresponding research question is proposed - How can
land cover and vegetation structure mapping be applied with a rapid and simple
approach to assess the regulating services of urban green spaces in Xi’an city?

Objective 2: Identify the relationship between park features and ecosystem services.

Two questions were proposed in order to achieve objective 2.

Question 1. How does land cover influence the regulating services of green spaces in
Xi’an city? For example, what are the main land cover types which influence the
regulating services of green spaces, and how do they influence the regulating services?

Question 2. Which other factors (e.g. park area, age and location) also influence
regulating services, and how?

Objective 3: Assess urban park use, perception and appreciation of ecosystem services
by park users.

42
In this study, these basic questions were asked first to determine the people’s use,
assessments, demands for, and perceptions of green spaces. The most important
research questions which needed to be answered here are the relationship between
land cover types and peoples’ use and assessment of green spaces. This new
perspective links land cover types with recreational services, and hence fills the
research gap:

1. What are the people’s use habits, preferred recreational activities, demands for, and
perceptions of green spaces in Xi’an? What are the predominant factors which
influence peoples’ use habits, preferred recreational activities, demand for and
perceptions of green spaces?

2. How do users assess green spaces in Xi’an, especially the regulating and
recreational services? What are the predominant factors (e.g. land cover types, green
space characteristics and socioeconomic variables) that influence these assessments?

Objective 4: Explore how green space planning and management can better contribute
to enhancing ecosystem services in Chinese cities.

Ecosystem services of urban green spaces and urban green space planning and
development have been separately paid attention for many years. However, their
combinations, such as considering ecosystem services in green space planning and
development, still remain to be investigated in many developing cities such as Xi’an.
The following questions will therefore be used to attempt to answer how to use the
assessment of ecosystem services to achieve better green space planning and
development:

1. How are the urban green spaces planned and developed in Xi’an? Are ecosystem
services taken into account in the planning and management of parks in Xi’an city?

2. What are the main problems and critical issues for urban green space planning and
development in Xi’an, both at the city and site level? How can the problems in urban
green space planning and development be mitigated using the ecosystem services
approach?

43
3 Materials and Methods
Study Area
3.1.1 Definition
As one of the strongly developing large cities in China, Xi’an has experienced
massive land use changes from non-urban to urban, accompanied by the degradation
of environmental conditions. At the same time, the awareness of important benefits of
urban green spaces has increased. Some improvement approaches have been used to
conserve the existing and develop new green spaces in the urban area. However,
urban green space design, planning and development often lack a scientific basis and
rationale. Moreover, the idea of solving local environmental and ecological problems
using urban green spaces is seldom promoted. These circumstances make Xi’an a
suitable area to conduct research aiming at better planning and conservation of green
spaces.

The study area is confined to the central parts of Xi’an city. Given the high population
density, rapidly developed economy, crowded urban fabric, deteriorating environment,
and urgent expectations for a higher quality of life, it is warranted to conduct a
scientific study of the urban green spaces of the city to inform better planning,
management and conservation. The studied central area is located within the Third
Ring Road, covering an area of 403 km2 with a population of 4.5 million (XUPB
2010). The 22 urban public parks within this area were chosen as the research sites of
this study (see Table 3.1 and Figure 3.1). Among these parks, one is a children’s park,
one is a sports park, and three are municipal parks. The remaining seventeen are all
district parks.
3.1.2 Geographic Conditions
Geographically, Xi’an is situated in the middle of the Guanzhong Plain, which is a
flood plain created by eight rivers and streams in Central China, located around
33°39’-34°45’N latitude and 107°40’-109°49’E longitude (XUPB 2010). The
elevation of the study area is on average 412 meters above sea level. The Li Mountain
lies east of the city and the Qinling Mountain to the south. The Feng River runs
through the west of the city while the Wei River traverses the north. The distance from
the east to the west of Xi’an is 204 km, and 116 km from south to north (XUPB
2010).

Influenced by the East Asian monsoon, Xi’an has a semi-moist temperate continental
climate, which is characterized by brief and dry springs and autumns, hot and humid
summers, and cold and dry winters. The average air temperature is 15.5°C annually,
and reaches its lowest in January at -0.9°C and its highest in July at 26.4°C
(XAALADI 2009). The annual average rainfall is about 600 mm, most of which falls
from August to late October (XAALADI 2009). Dust storms often occur during
March and April as the city rapidly warms up. In the summer months, there are
frequent but short thunderstorms, and in winter, snow occurs occasionally but rarely
44
settles for long.

The area hosts a large number of vegetation types. Before urbnization, the natural
vegetation was bushlands, meadows, coniferous forests, mixed forests of conifers and
borad-leaves, and deciduous broad-leaf forests (XAALADI 2009). Some species from
natural habitats have been preserved in the urban environment. For example, relic
species of the ice age, such as Ginkgo biloba, Cercidiphyllum japonicum, and
Tetracentron sinense, have survived in this area. Since the 1850s, a number of exotic
species have been introduced and widely planted, such as Platanus spp. and Quercus
variabilis. At present, the city region contains more than 2200 plant species, among
which approximately 1550 are spermatophytes (XAALADI 2009).

Table 3.1 List of the 22 investigated parks (source: Xi'an Ancient Architecture
and Landscape Architecture Design Institute, 2009)

Park Area (ha) Recorded Year Green space type


area (ha) built
Children's Park 2.11 3.0 1928 Children’s park
Tu Men Road Park 2.79 2.8 1997 Roadside park
Fang Zhi Park 3.44 3.0 1978 Neighborhood park
Ci En Si Yi Zhi Park 4.43 4.5 1997 Neighborhood park
Xi Qu Da Guan Yuan 4.44 4.2 2001 Neighborhood park
Lian Hu Park 5.25 6.4 1916 Residential park
Min Su Park 6.01 5.2 2001 Residential park
Xin Ji Yuan Park 6.05 6.0 1997 Residential park
Wen Jing Park 6.25 7.6 2004 Residential park
Mu Ta Si Park 6.31 6.8 2009 Residential park
Mu Dan Yuan 6.95 7.0 2007 Residential park
Lao Dong Park 7.33 7.7 1965 Residential park
Ge Ming Park 7.75 10.4 1927 Residential park
Huan Cheng Xi Yuan 10.96 10.5 2004 District comprehensive park
Yong Yang Park 11.24 11.7 2006 District comprehensive park
Chang Le Park 17.16 19.6 1977 District comprehensive park
Feng Qing Park 22.52 21.1 2004 Municipal comprehensive park
Qu Jiang Tang City Wall 31.99 31.0 2008 District comprehensive park
Park
City Sports Park 33.84 35.4 2006 Sports park
Tang Yan Road Tang City 40.57 44.4 2006 District comprehensive park
Wall Park
Xing Qing Park 48.99 50.2 1958 Municipal comprehensive park
Qu Jiang Yi Zhi Park 50.67 53.0 2008 Municipal comprehensive park
(Note: data of recorded park areas are taken from the field survey)

45
Figure 3.1 Site map of Xi’an and the 22 surveyed sites (own drawing)

1. Children's Park; 2. Tu Men Road Park; 3. Fang Zhi Park; 4. Ci En Si Yi Zhi Park; 5. Xi Qu Da Guan Yuan; 6. Lian Hu Park; 7. Min Su Park; 8. Xin Ji
Yuan Park; 9. Wen Jing Park; 10. Mu Ta Si Park; 11. Mu Dan Yuan; 12. Lao Dong Park; 13. Ge Ming Park; 14. Huan Cheng Xi Yuan; 15. Yong Yang
Park; 16. Chang Le Park; 17. Feng Qing Park; 18. Qu Jiang Tang City Wall Park; 19. City Sports Park; 20. Tang Yan Road Tang City Wall Park; 21.
Xing Qing Park; 22. Qu Jiang Yi Zhi Park

46
3.1.3 Social, Economic and Environmental Conditions
Xi’an, the provincial capital of Shaanxi province, is well known for its rich historical
and cultural resources. It was the capital of thirteen dynasties for over 3,000 years of
Chinese history and was once the largest city in the world, during the Tang dynasty.
Xi’an is also the political, economic, cultural, scientific and technological center of
northwest China. Lying at the crossroads of several main inland arteries, Xi’an holds a
powerful position among the central and western Chinese cities, and has an
abundance of high-tech and light industries, universities and research institutions
(XUPB 2010; Yongming et al. 2006).

The social development of Xi’an is supported by its economic development, and both
jointly enhance the quality of life. After the implementation of reforms and the
opening-up policy of the country in 1978, the economy of Xi’an grew consistently
with an increasing speed over the last two decades. The yearly gross domestic product
(GDP) growth rate has constantly surpassed 13% since 1995. As of 2011, the GDP of
the city was 386,421 billion yuan (Xi’an Statistical Yearbook 2012,
http://tjj.xa.gov.cn/ptl/index.html).

Accompanied by the rapid population growth and economic development, the urban
area gradually expanded from the inner city-wall area, namely the built-up area
enclosed by the ancient city wall, to the outside area, which can be divided by the
three ring roads. In 2010, Xi’an had a population of 9.2 million and a total area of
approximately 10108 km2 (XUPB 2010). The inner area spans approximately 12 km2
with a high population density of 36,885 persons/km2. This is where the main
commercial and political center of the city is now located, housing 37% of retail and
25% of commercial offices (Xi’an Statistical Yearbook 2010). Because of the
numerous businesses, houses and institutes, this area contributes a large portion of the
total GDP of the city. However, such achievements in economic development and
urbanization have created serious congestion problems. As a result, the government
had to relocate the commercial center to northern Xi’an, starting in 2004 (XAALADI
2004). Urban land use expanded continually from the city core to the periphery,
changing the suburbs into urban areas. Especially after the 1990s, the city has
enlarged rapidly, and the urban scale and structure have changed dramatically.
Worryingly, a significant amount of arable land has been irreversibly taken over by
businesses and industry (Lu and Peng 2006).

In addition, fast social development, economic growth and urbanization have brought
a number of environmental problems to Xi’an. The climate of Xi’an has changed in
the last 60 years (1951-2013), whereby the mean temperature increased by 1.3°C in
the urban area - twice as much as the average global temperature increase (Che
Huizheng 2005; Mengtao and Zhou 2016). Annually, the average temperature at noon
is above 30, 35 and 40°C for 75, 20 and 7 days, respectively (Li 2002). Moreover,
between 1993 and 2013, the average difference of air temperature between the

47
suburban and urban areas was 0.35°C and 0.4°C in the hottest month - July (Liu 2008;
Mengtao and Zhou 2016; Yufeng Liu et al. 2015). The intensification of the urban
heat island phenomenon is thus obvious. Especially, between 2007 and 2012, the
increase of air temperature due to the urban heat island effect was 1.06°C on average
(Yufeng Liu et al. 2015).

Furthermore, one of the factors leading to the urban heat island effect is the decreased
air humidity due to the large areas of hard/impervious surfaces in cities (Memon et al.
2011; Oke 2011). Vegetation, and especially tree species with high transpiration rates
and water use efficiency can increase air humidity. Considering the prolonged periods
of low temperatures in the winter months, large amounts of deciduous broad-leaf tree
species are superior to other species for planting in Xi’an city because of their better
winter survival.

During the same period, the monthly rainfall, the relative air humidity and sunshine
duration decreased. The number of sunny days has been reduced because of the
increased particles in the air (Che Huizheng 2005). In 2011, the average daily air
particulate matter (0.118 mg/Nm3) of urban areas exceeded the standard by 18%
(ADB 2003 ). The average air pollution data of SO2 (0.042 mg/Nm3) and NO2 (0.041
mg/Nm3) stayed at the upper limits of the national air quality standards. Moreover,
water pollution has become severe in many rivers. Aquatic habitats in many rivers
have been disturbed by the discharge of sewage, solid waste and other pollutants
(ADB 2003 ). For example, Ba River and Chan River have been heavily polluted by
drainage of chemical materials from domestic industry. Although the acute and
chronic exposure to pollutants can kill off vegetation, death of vegetation induced by
environment pollution (e.g. air pollution and water pollution) is rare in Xi’an city
(Ouyang 2016; Xu et al. 2013; Yang 2007). However, the potential injury of trees by
pollutants can be found in many cases, including for example reduced photosynthesis
due to blocked stomata of leaves caused by air pollutants such as dust in the air
(Ouyang 2016; Yang 2007). This kind of injury will obviously decrease plant growth.

As described in Chapter 2, vegetation, and especially trees, can be significant sources


of cool air in summer, as well as active absorbers of air pollutants. Due to their large
areas of vegetation and trees, urban green spaces, and especially parks, can be used to
mitigate the environmental deterioration caused by urbanization. Furthermore, urban
green spaces provide recreation opportunities, and thus enhance the environmental
quality and human well-being. In recent years, the government has realized the severe
urban problems and tried to implement a number of national and municipal measures
to contain and control the urban heat island effect and rampant air pollution.
Considering the continued fast urbanization and population growth in the study area,
long-term approaches should focus on introducing well-developed multi-functionality
to green spaces, providing especially regulating and recreational functions.

48
Methods
3.2.1 Outline of Methods
The main aim of this chapter is to outline the methodology used in this research.
Taking Xi’an city as a case study, the methodology was applied to 22 public urban
parks.

In general, the procedure of the approach is as outlined in Figure 3.2.

Rapid Assessment of Ecosystem Services of Urban Green Spaces


Assessment Indicators

Indicators of regulating services Indicators of recreation services

Land cover mapping Questionnaires

Indicator Rating

Aggregation

Assessment of Regulating Services Assessment of Recreation Services

Ecosystem services as an approach

Enhancement of the planning and development of urban green spaces

Questionnaires Document analysis

Peoples’ participation in urban green Current urban green space planning


space planning and development and development

Figure 3.2 Procedure for applying the methods of this research

As Figure 3.2 shows, the methods of this research aimed to enhance urban green
space planning and development according to assessments of indicators and citizens’
attitudes towards green space ecosystem services, as well as the analysis of
government documents concerning green space planning and development.

The rapid assessment of ecosystem services attempts to assess the regulating and
recreational services of green spaces in Xi’an. Theoretically, this approach consists of
indicators, indicator ratings and aggregations. All the information in the indicators
was derived from a field survey, Google Earth (GE) images and a questionnaire
49
survey. The indicators of the provision of regulating services were determined using
land cover mapping, based especially on the vegetation structure types. The indicators
of recreational services were determined by users’ direct assessments collected via a
questionnaire. In addition to respondents’ direct assessments of recreational services,
their use habits, preferred recreation activities, perceptions and demands for green
spaces were also used to invoke citizens’ participation in green space planning and
development.

Moreover, the documents used for the planning and development of green spaces in
Xi’an city were analyzed in order to identify insufficiencies of current green spaces.
Subsequently, ecosystem services, questionnaires and document analysis were used as
approaches to assess and enhance the urban green spaces of Xi’an.
3.2.2 A Rapid Ecosystem Service Assessment Approach
3.2.2.1 Summary of the Approach
In general, the rapid ecosystem service assessment method encompasses composite
indicators, indicator ratings and aggregations (see Tables 3.2 and 3.3). The catalogue
of composite indicators and indicator rating rules is based on a review of the relevant
scientific literature (see Sections 2.2, 2.3 and 2.4). For ease of operation and
comparison, the indicators should be kept simple and in a format that can be
quantified.

In some studies, the indicators were weighted according to their importance. For
instance, importance may be decided based on preferences derived from
questionnaires (e.g. Curtis, 2004 and Gu et al., 2010). Other studies have attempted to
evaluate ecosystem services without weighting indicators, so that all indicators carry
the same weight (e.g. Liu et al., 2002; Chen et al., 2009 and Stern, 2010). The
approach applied here also considers all indicators to have the same weight. On the
one hand, this approach relies not only on the assessments of citizens but also on
assessments of land cover and vegetation structure. As such, it is unsuitable to allocate
weights by considering the preferences of citizens only. On the other hand, many
studies have demonstrated the fact that regulating- and cultural services are all very
important services provided by green spaces (e.g. Bolund and Hunhammar, 1999 and
Jim and Chen, 2009). To our best knowledge, no study has been conducted on the
basis of some such services being of greater importance than others.

Owing to specific conditions such as local temperatures, varying measurements and


complex vegetation structures, it would be very difficult to ascertain a precise
relationship between indicators and the provision of corresponding ecosystem
services (temperature regulation, air pollution abatement etc.). Hence, it is difficult to
reveal the relationships between decreased air temperature / air pollutants and certain
types of vegetation, such as street trees or groups of trees (e.g. Mao et al., 1993 and
Jim et al., 2003), and to use this relationship elsewhere directly. However, in order to
produce an evaluation, a linear variation trend between an indicator and its impact on
50
ecosystem services can be approximated (see Table 3.3). The process by which to
identify an indicator’s impact can be seen as the indicator rating. Aggregation rules
are then created in order to combine all indicator impacts into a final assessment
result. These rules should be kept as simple as possible and should obey logic (see
Table 3.3).

In practice, for a selected green space, the assessment will be achieved through three
stages (see Tables 3.2 and 3.3): (1) apply a catalogue of composite indicators to the
green space, with collected data being used as the basis for the assessment; (2) assign
to each indicator its corresponding class range and impact score, with the help of
ArcGis 10.0 software (ESRI inc., USA); and (3) combine the indicators’ impact
scores into the final assessment results of specific ecosystem services via aggregation
rules.
3.2.2.2 Indicators used for Assessment
Indicators of Regulating Services
Land cover and vegetation structure are the basis of regulating service assessment (see
Section 2.3). In many cases, the indicators related to the assessment of air temperature
regulation and air pollution reduction generated by urban green spaces are connected
with the land cover, especially vegetation. In reality, the regulating functions of
microclimate and air quality regulation are connected with each other. Many air
pollutants are temperature-sensitive and lower temperatures decrease air pollution,
especially in summer (Cardelino and Chameides 1990) (see details in Section 2.3.2).
As discussed in section 2.3.2, vegetation and water cover (1.1), multi-layered
vegetation cover (1.2), tree cover (1.3) and mature tree cover (1.4) were used as
indicators to assess microclimate regulation in this study. Based on these indicators, in
conjunction with the cover by evergreen species (1.5), air pollution removal was
assessed (see Table 3.2).

Indicators of Recreation Services


Questionnaires are proposed as a method by which to evaluate the ecosystem services
of green spaces (see Section 2.4). In this study, the recreational services of urban
parks were assessed directly by citizens (see section 2.4.2). Since the main recreation
services of urban green spaces are focused on aesthetic enjoyment, exercise, rest and
children’s play (Chiesura 2004; Tzoulas and James 2010b; Zhang et al. 2015),
questions related to these items were posed as indicators (2.1-2.10) to assess
recreational services. These indicators were: (1) scenic beauty (2.1); (2) sporting
facilities: the number of sporting facilities (2.2), the location of sporting facilities
(2.3) and the maintenance of sporting facilities (2.4); (3) rest: the number of resting
facilities (2.5), the location of resting facilities (2.6) and the maintenance of resting
facilities (2.7); (4) children’s play: the area of playgrounds (2.8), the location of
playgrounds (2.9) and the maintenance of playgrounds (2.10) (see Table 3.2).

51
Table 3.2 List of indicators used for the assessment of ecosystem services

Catalogue of Indicators of the Provision of Ecosystem Services


- Regulating services of urban green spaces (indicators: 1.1-1.5)
- Recreation services of urban green spaces (indicators: 2.1-2.10)
Nr. Indicators Indicator description
1.1 Microclimate Land cover Cover of vegetation Proportion of vegetation and water areas
and air quality and water in relation to the total green space area of
the site

1.2 Vegetation Multi-layered Proportion of vegetation with more than


structure vegetation cover two layers in relation to the total vegetated
area of the site
1.3 Tree cover Proportion of area covered by tree crowns
in relation to the total vegetated area of
the site
1.4 Mature tree cover Proportion of area covered by mature tree
crowns (tree height ≥10 m) in relation to
the total vegetated area of the site
1.5 Air quality Evergreen species Proportion of area covered by the crowns
cover of evergreen species in relation to the total
vegetated area of the site
2.1 Recreation Questionnaire Scenery Percentage of respondents satisfied with
scenery
2.2 Number of sporting Percentage of respondents satisfied with
facilities the number of sporting facilities
2.3 Location of sporting Percentage of respondents satisfied with
facilities the location of sporting facilities
2.4 Maintenance of Percentage of respondents satisfied with
sporting facilities the maintenance of sporting facilities
2.5 Number of resting Percentage of respondents satisfied with
facilities the number of resting facilities
2.6 Location of resting Percentage of respondents satisfied with
facilities the location of resting facilities
2.7 Maintenance of Percentage of respondents satisfied with
resting facilities (e.g. the maintenance of resting facilities
table or bench)
2.8 Area of playgrounds Percentage of respondents satisfied with
the area of playgrounds
2.9 Location of Percentage of respondents satisfied with
playgrounds the location of playgrounds
2.10 Maintenance of Percentage of respondents satisfied with
playgrounds the maintenance of playgrounds

52
3.2.2.3 Indicator Rating
As can be seen from the theoretical sections (see Section 2.3), regulating services
include positive relationships between indicators of air temperature / air pollution and
the decrease of temperature and air pollutants. Although the relationship cannot be
shown accurately without local measurements, it can be approximated in a rough and
simple manner (see Table 3.3). The increased cover ratio of indicators (e.g. vegetation
and water, multi-layered vegetation, trees, mature trees and evergreen species) raises
the possibility of temperature decrease and air pollutant reduction in green spaces, and
this effect becomes increasingly evident when the cover ratio increases (see Section
2.3). Based on these facts, the indicators of air temperature and air pollution were
simply classified into four scales, by logical, gradual and equal intervals of their cover
ratio (see Table 3.3): 0-25% (very small), 26-50% (small), 51-75% (medium) and 76-
100% (large).

The indicators from the questionnaires (2.1-2.10) were judged by respondents. In


order to assess the provision of recreation services, the respondents were asked to
select their answers from “excellent’’, “good”, “fair”, “poor” and “unclear”. Among
all answers, the total proportion of positive answers (“excellent’’ and “good”) were
used as satisfaction assessments, which were then divided by the gradual and average
intervals: 0-25% (very dissatisfied), 26-50% (dissatisfied), 51-75% (satisfied) and 76-
100% (very satisfied) (see Table 3.3). Therefore, the higher the number of positive
assessments, the higher the degree of satisfaction.

Upon ascertaining the indicators and indicator ratings, scores were allocated to denote
indicator impacts in order to simplify the aggregation in the next step. For each
indicator, one point was allocated for the class range of 0-25%, while two points and
three points were allocated for 26-50% and 51-75%, respectively. Four points were
allocated for 76-100% (see Table 3.3). In a word, indicators with higher proportions
attained a higher score and represented a greater impact on regulating services and
recreation services.
3.2.2.4 Aggregation
Aggregation represents the deduction of results from indicators and rating rules. In
this study, it can be seen from the theoretical part (see Sections 2.3 and 2.4) that all
the indicators have positive interactions with each other. Therefore, the full impact
scores of specific ecosystem services can be inferred by summing up the impact
scores of all referred indicators (1.1 - 1.4 / 1.1 - 1.5 / 2.1 - 2.10). Then, the final
assessments can be divided into very low (1 - 4 points / 1 - 5 points / 1 - 10 points),
low (5 - 8 points / 6 - 10 points / 11 - 20 points), medium (9 - 12 points / 11 - 15
points / 21 - 30 points) and high values (13 - 16 points / 16 - 20 points / 31 - 40
points) by averaging the total impact scores (see Table 3.3).

53
Table 3.3 Approach for the rapid assessment of ecosystem services
Catalogue of Indicators used to Assess the Provision of Ecosystem Rating Aggregations
Services
Nr. Indicators Class range Class Impact Total impact score Assessment
description score
1.1 Microclimate Cover of vegetation and water 0-25% Very small 1 1-4 1-5 Very low
1.2 and air quality Multi-layered vegetation 26-50% Small 2 5-8 6 - 10 Low
51-75% Medium 3 9 - 12 11 - 15 Medium
1.3 Tree cover 76-100% Large 4 13 - 16 16 - 20 High
1.4 Mature tree cover
1.5 Air quality Evergreen species cover
2.1 Recreation Scenery satisfaction 0-25% Very dissatisfied 1 1 - 10 Very low
2.2 Satisfaction with the number of sporting 26-50% Dissatisfied 2 11 - 20 Low
facilities 51-75% Satisfied 3 21 - 30 Medium
2.3 Satisfaction with the location of sporting 76-100% Very satisfied 4 31 - 40 High
facilities
2.4 Satisfaction with the maintenance of sporting
facilities
2.5 Satisfaction with the number of resting
facilities
2.6 Satisfaction with the location of resting
facilities
2.7 Satisfaction with the maintenance of resting
facilities (e.g. tables or benches)
2.8 Satisfaction with playground area
2.9 Satisfaction with the location of playgrounds
2.10 Satisfaction with the maintenance of
playgrounds

54
In practice, the assessment of ecosystem services can be inferred by (1) implementing
the indicators, their rating scales and scores; (2) averaging the total indicators’ scores
into four ranges, to which four values are assigned; and (3) summing the indicators’
scores and then assigning them to their corresponding ranges and values (see Table
3.3). For example, if there is a park with 72% of water and vegetation cover, 65.5% of
multi-layered vegetation, 30.3% of tree cover, 41.2% of mature tree cover and 32.2%
of evergreen species cover, the park would get a score of twelve. The total scales of
regulating services resulted in 20 points. By averaging, very low values ranged from 1
to 5 points, low values ranged from 6 to 10 points, medium ranged from 11 to 15
points, and high values ranged from 16 to 20 points. Thus, this park would be
considered as having medium value in terms of regulating services.
3.2.3 Land Cover Mapping
In this study, the method of land cover mapping was used to collect and compute the
information of the indicators of regulating services as aforementioned, including
cover of vegetation and water, multi-layered vegetation, tree cover, mature tree cover
and evergreen species cover.
3.2.3.1 Data Collection
The indicators of regulating services are related to land cover and vegetation structure
of the study areas (22 urban parks), which were classified via observations in field
surveys and interpretation of Google Earth (GE) images. Similar approaches were
employed by Young et al., 2001; the URGE project, 2004, and Nicole Stern, 2010.

Generally, the main tasks of field surveys were to distinguish the types of land cover
and vegetation structure, rather than to determine the boundaries of different land
cover types. It is notable that there is no standard approach to image interpretation,
although visual interpretation was preferred in this case. During the on-site-
confirmation process, boundary illegibility (always induced by tree crowns) was
unavoidable. However, small disparities do not appear to significantly impact the land
cover proportions or the provision of ecosystem services by adjacent patches. For
instance, Tratalos et al. (2007) ignored illegible patch areas of less than 10 m2 because
they considered that small areas of sealed surfaces (e.g. paths through parks or narrow
roads) did not represent a barrier for the dispersal of most plant and animal species
(Tratalos et al. 2007). Similarly, in this study, boundary illegibility of land cover areas
smaller than 10 m2 was not indicated, and any unclear boundaries connected with
vegetation were counted as vegetation for the sake of consistency.

Prior to the field survey, additional survey sheets of land cover and vegetation
structure were generated based on the literature review presented in section 2.3 (see
Appendices 2 and 3). Site maps and printed GE images of the urban parks (image
source: Google Earth 6.1) were prepared to aid the classification and boundary
identification. Moreover, in cases where a type of land cover or vegetation structure
was not listed on the survey sheet, the type was named and recorded in the
supplementary column in Appendices 3 and 4.
55
Especially, possible classifications of vegetation height were defined for local
contexts in this study. According to the consensus of landscape architecture in China
and definitions taken from the book Dendrology for Landscape Architecture (Chen
1990), trees are defined as woody plants with a main stem taller than 5 m. Shrubs are
defined as woody plants with no obvious main stem, normally below 5 m in height
(Chen 1990). Considering the height variability of vegetation (e.g. Hercock, 1997 and
Tzoulas and James, 2010), the height variations of vegetation in urban green spaces of
Xi’an city can be classified into: (1) herbs; the herb layer is mainly composed of
mown lawns (< 10 cm), flowers (< 20 cm, 20 - 50 cm or 50 - 100 cm) and bamboos
(< 2 m); (2) shrubs, defined as all woody vegetation with a height below 5 m, except
trees; (3) low trees: trees with a height of 5 - 10 m; and (4) mature trees or high trees:
trees with a height greater than 10 m (see Figure 3.3).

Figure 3.3 Height variations of woody plants (author’s own drawing)

First Field Survey


In this study, field surveys were conducted both at the city level and the park level.
The first field survey as a pre-test of land cover and vegetation structure types was
conducted during June 2010. This pre-test aimed to modify the survey sheets for
implementation of the second field survey. Based on government documents (Table
3.6), the 22 urban parks’ area, location, history and characteristics were reviewed.
Then, after the observation of said 22 parks, which was carried out by visits on foot
and by car, six public parks including Lian Hu Park, Yong Yang Park, Feng Qing
Park, City Sports Park, Xing Qing Park and Qu Jiang Yi Zhi Park were selected for
testing (see Figure 3.1 and Appendix 1). Finally, some 172.51 ha of park area were
investigated, which accounted for over fifty percent of the total area.

The steps undertaken during the first field survey at each of the six parks were:
(1) Preparation - proposing the possible land cover types

Beforehand, the locations of the park boundaries were drawn on the corresponding
printed GE images. After on-site-confirmation, these were modified to improve
accuracy. All patches within each park were numbered and identified in terms of their
land cover types using survey sheet 1 (see Appendix 2), with the aid of GE images
56
and park site maps. Provisional patch boundaries were drawn onto the printed GE
images.

(2) Direct visual observation - determining the land cover and vegetation structure
types. During the field survey at each park, land cover and vegetation structures were
adjusted to local situations and recorded.

Firstly, the land cover types were determined. The land cover types and boundaries
were confirmed on-site and modified at each patch. In this step, the point was to
discern the vegetation and water areas from the other land cover types such as
pavements, buildings and playgrounds (see Appendix 2). All the existing land cover
types have been marked in the survey sheet 1 (see Appendix 2). If there were land
cover types which did not conform to any of the given types, they were recorded as
supplementary types (see Appendix 2).

Then, the types of vegetation structure were determined. Within vegetated patches, the
information regarding vegetation structures including vegetation layers and vegetation
compositions was visually estimated and recorded on survey sheet 2 (see Appendix
3). Vegetation layers were counted as one, two, three and multilayers (more than three
layers). The different types of vegetation and their compositions were based on the
presence of trees, shrubs and herbs, such as multi-layered trees over shrubs and lawns,
groups of trees and groups of shrubs over lawns. Generally, vegetation types were
classified by their appearance, physiological features and height range, i.e. the herbs
were usually 10-20 cm, shrubs were always less than 5 m and trees were 5-10 m high
(see Appendix 3 and the Section 3.2.3.1).

(3) Information supplementation

During the process of investigation, photographs were taken to show the patch
appearance, land cover and vegetation structure.

According to the demonstration of the first field survey, the direct visual obseration
method worked well in the local context of Xi’an city. Types of land cover and
vegetation structure were categorized as shown in Table 3.4, which can be seen as
providing classifications of the land cover and vegetation structure of green spaces in
Xi’an city. Based on this table, the revised survey sheet 3 (see Appendix 3) was used
in the formal field survey, i.e. the second field survey.

57
Table 3.4 Land cover and vegetation structure types within urban parks of Xi’an
city

Land cover types Sub-types


Vegetation Trees Evergreen coniferous woodland
Evergreen broadleaf woodland
Deciduous broadleaf woodland
Shrubs Evergreen coniferous shrubs
Evergreen broadleaf shrubs
Deciduous broadleaf shrubs

Lawns
Flower beds
Water plants
Bamboo
Trees covering shrubs
Trees covering lawns
Shrubs covering lawns
Trees covering shrubs and lawns
Multi-layered trees covering
shrubs and lawns
Pavements with
scattered vegetation
Pavements
Playgrounds
Buildings Business buildings
Garden architecture
Water bodies Lakes
Fountains

Second Field Survey


The second field survey was conducted in all of the 22 chosen public urban parks in
Xi’an city (including the six parks included in the first field survey) during July and
August of 2010.

The survey procedures were:


(1) Preparation - proposing the possible land cover types

Within each park, the preparations were repeated as in the first step of the first field
survey, while all provisional patches were identified using the adapted types of land
cover (see Table 3.4).

(2) Direct visual observation - determining all the information concerning land cover

58
and vegetation structure. The information on each patch was recorded using survey
sheet 3 (see Appendix 4).

(i) Determining the cover of vegetation and water


In this step, the point was to differentiate with certainty the ground coverage of
vegetation and water areas from the other five types such as pavements, buildings and
playgrounds (see Table 3.4). All the numbered provisional patches were confirmed
on-site, modified, and finally recorded in the second column of survey sheet 3 (see
Appendix 4). The patch boundaries were also confirmed on-site and drawn on the GE
images.

Subsequently, the steps focusing on the vegetation patches included:


(ii) Determining multi-layered vegetation
Vegetation layers were counted as one, two, three and multilayer (more than three
layers); the different types of vegetation and their compositions based on trees, shrubs
and herbs were recorded in survey sheet 3, using table 3.4. All the data were recorded
in the third column of survey sheet 3. Then, the percentage of multilayered vegetation
was estimated and recorded in the corresponding columns (see Appendix 4).

(iii) Determining the coverage with trees and mature trees


The percentage of trees and mature trees in each vegetation patch was estimated by
walking around the patch ground, and recorded using gross intervals of 0-25%, 26-
50%, 51-75% and 76-100% (see Appendix 4).

(iv) Determining the coverage with evergreens


Similarly, at each vegetation patch, the proportion of evergreen trees and shrubs was
estimated and recorded using the same gross intervals of 0-25%, 26-50%, 51-75% and
76-100% (see Appendix 4).

As a supplement, the percentages of deciduous trees and shrubs, lawns and other
types of herbs at each patch were estimated at the same time.
3.2.3.2 Data Processing and Analysis
Subsequent to the data collection, land cover maps of the 22 urban parks were drawn
up and digitized using ArcGis10.0 (ESRI Inc., USA) using the following procedures:

(1) Using the software GetScreen1.1.1.0, GE images of the 22 parks were downloaded
from Google Earth 6.1 (Google Inc., USA). The GE images were acquired on
cloudless summer days before 2010, and were the same ones that were used in the
field survey in order to determine the land cover and vegetation structure types;

(2) Geo-referencing the GE images of the 22 parks individually in ArcMap10.0 (ESRI


Inc., USA) using the coordinates WGS_1984_UTM_Zone_49S;

59
(3) Establishing a new shape file (.shp) for each park in ArcCatalog 10.0 (ESRI Inc.,
USA) using the same coordinates as in the geo-referenced images;

(4) Tracing the figure of land cover and vegetation structure on the shape layer of
each park based on the geo-referenced images, with the help of prepared drafts on the
printed GE images;

(5) Input of the textual information related to land cover and vegetation structure into
the attribute tables in each park’s shape file, in order to allow surface classifications,
map generation and relevant data output;

(6) Filling in the common types of land cover and vegetation structure with the same
colors and inserting a north arrow, scale and legend on the shape layer, and finally
outputting the digitized maps.

Using the shape files of the parks, the area of each patch was automatically generated
by ArcMap10.0 (ESRI Inc., USA), after the digitized land cover mapping was
completed. Thereafter, the percentage of each type of land cover and vegetation
structure was quantified, and thus could be used for indicator ratings in the assessment
of regulating services.
3.2.4 Questionnaires
A questionnaire survey was conducted within urban parks for the purpose of
collecting data from persons who were asked randomly to take part; this approach was
chosen due to the non-accessibility of mailing addresses, telephone numbers and
household visits. Questionnaires permit closer interaction between researchers and
research participants. The method provides a way in which to ascertain how well an
individual’s needs and desires are supported by their environment, and requires a
relatively short time to cover a broad range of populations with different socio-
economic backgrounds (Bell et al. 2004).
3.2.4.1 Sample Size of the Questionnaire Survey
While it is known that the sample size can affect the representativeness of results
obtained by way of a questionnaire (Kornblum 2001), there is no agreement on the
optimal size of a sample in the literature. A sample size of 500-800 people has been
suggested as adequate from a statistical perspective (Kornblum 2001). However, in
practice, deciding on a sample size typically depends on multiple factors, such as
financial funding and time. The sample sizes of many questionnaires investigating the
ecosystem services of urban green spaces were less than 500 people. For example, the
number of interviewees was in the range of 200-350 participants in the research of
Sanesi et al., 2006; Jim et al., 2006; Arnberger and Eder, 2011; Gerhardt, 2010 and
Neuvonen et al., 2006.

In this study, a random sample of 500 questionnaires was conducted during two
workdays and two weekends between June and August 2012 and May and June 2013,
60
between 7:00 am and 7:00 pm. The sites investigated were the 22 parks in Xi’an city
(see Table 3.5). The number of questionnaires to be completed was decided upon by
the area of each park, with larger parks eliciting a greater number of completed
questionnaires than smaller parks. Each questionnaire respondent was selected
randomly by picking one out of three visitors encountered by the surveyor in
sequence. Children below 15 years of age were excluded due to their limited ability to
understand the questions. Moreover, considering the limited knowledge of some
respondents, the survey attempted to make all questions easily understandable by
typical citizens of Xi’an city; this was achieved by discarding specialized terms and
engaging in oral communication or complementary explanations. The respondents
completed the questionnaire independently, and only when assistance was requested
did the interviewers provide further explanations.

Table 3.5 Questionnaire samples at the 22 selected urban parks

Research area Recorded area Total Completed


(ha) questionnaires questionnaires
Tu Men Road Park 2.81 10 8
Children's Park 3.00 10 9
Fang Zhi Park 3.70 10 9
Ci En Si Yi Zhi Park 4.50 10 7
Xi Qu Da Guan Yuan Park 4.20 10 10
Min Su Park 5.21 15 14
Xin Ji Yuan Park 6.00 15 13
Mu Ta Si Park 6.78 15 12
Mu Dan Yuan Park 6.98 15 14
Lian Hu Park 7.15 15 13
Wen Jing Park 7.60 15 14
Lao Dong Park 7.70 15 15
Ge Ming Park 10.4 20 16
Huan Cheng Xi Yuan Park 10.3 20 18
Yong Yang Park 12.00 20 17
Chang Le Park 19.60 25 19
Feng Qing Park 21.10 30 24
Qu Jiang Tang City Wall Park 31.00 40 34
City Sports Park 35.40 40 33
Tang Yan Road Tang City Wall Park 44.4 50 43
Xing Qing Park 50.20 50 42
Qu Jiang Yi Zhi Park 53.00 50 44
Total 353.03 500 428

61
3.2.4.2 Question Design
Undoubtedly, a critical step in the process of administering questionnaires is to
establish reasonable and effective questions and answers. The questionnaire used in
this research was developed mainly based on Oguze, 2000, Payne et al., 2002,
Chiesura, 2004, Jim and Chen, 2006, Sanesi and Chiarello, 2006, Shan, 2009,
Qureshiei et al., 2010, Gurung et al., 2011, Lo and Jim, 2012 and Casado-Arzuaga et
al., 2013. The main aims of this survey were to determine citizens’ use habits and
preferences, as well as their recreational activities and their perceptions and
expectations of the ecosystem services of green spaces.

A duration of no more than 20 minutes to complete a questionnaire is recommended,


in consideration of the patience of typical respondents (Lohr et al. 2004). Therefore,
combining the aims of the questionnaire, the context of Chinese society, and the
respondents’ comprehension and patience, 11 closed questions with single choices
were eventually decided upon (see Appendix 5).

Firstly, respondents’ personal information, including gender, age group, family status
and vocation, was elicited as these data would aid in the assessment of the existence
of any patterns by variable socioeconomic groups (e.g. respondents’ expectations of
green spaces or their use preferences by different age groups, vocation or family
status) (Garrod and Willis 1999 ; Tyrväinen 2001) (see Appendix 5). Apart from the
personal information, each question allowed the respondents to script their own
answers if the provided answers were not a good fit.

Secondly, respondents’ use behaviors were explored by way of six questions relating
to visit preferences and motivations, including preferred green space type, how much
time is spent on the way to the green space, use frequency, time and duration of visits,
and preferred recreational activities in green spaces (see Appendix 5). For use habits,
the optional answer categories for time spent on the way to the green space were: less
than 5 min; 5-15 min; 16-30 min and over 30 min. The available use frequencies of
green spaces were: daily; several times per week; 1-3 times per month; monthly;
several times per year and less. Answers in connection with visiting time varied
between morning, afternoon and evening, and also between weekdays and weekends.
Possible answers in relation to the duration of visits were: less than one hour; one to
two hours; half a day and nearly a whole day. Moreover, two questions were designed
with three options: “very often”, “often” and “occasionally”. One such question
concerned the respondents’ preferred type of green space while the other related to the
time of their visits. Lastly, the questions of preferred recreational activities were
designed with four options: “very important”, “important”, “not important” and
“neither important nor unimportant”. The offered recreational activities included both
popular active and passive activities (see Section 2.4), such as sports, use of
recreational facilities, relaxation, having fun with friends or playing with family and
enjoying scenery, fresh air and cooling off in summer (e.g. Dunnett, 2002; Özgüner

62
and Kendle, 2006; Tzoulas and James, 2010b and Grahn and Stigsdotter, 2010).

Five questions aimed to probe respondents’ perceptions, assessments and demands of


the ecosystem services of urban parks. In this section, two questions asked
respondents to select an answer from the four options: “very important”, “important”,
“not important” and “neither important nor unimportant”. One question concerned the
importance of green spaces within neighboring residences or work places and schools,
big city parks and parks in the outskirts. The other question was related to the
importance of 17 specific ecosystem services. As shown in the theoretical part (see
Section 2.4), the ecosystem services of green spaces related to human well-being
mainly include recreational services, microclimate regulation, air quality
improvement, water infiltration and biodiversity conservation, mainly in the forms of
shading, cooling in summer, fresh air, number of plants and animals, wildlife habitats,
water-soil conservation, places for recreational activities, more contact with nature
and social-neighbor interactions (see Appendix 5).

Two questions were designed with five options: “excellent”, “good”, “fair”, “unclear”
and “poor”; these enabled respondents to gauge the quality of urban green spaces in
Xi’an city and to nominate which features they were most satisfied with. As shown in
the theoretical part of this thesis, natural features (e.g. vegetation, water and wildlife),
recreational facilities (e.g. sports, resting and comforts), environmental quality (e.g.
quietness), aesthetics (e.g. scenic beauty), management (e.g. cleanliness), safety, and
easy access were seen as the features that high-quality green spaces should have (see
Section 2.4). Therefore, at city level, features such as the amount of green areas,
ecological functions and management of green spaces were asked to be assessed. For
specific green spaces, features such as microclimate, air quality, tree shade, scenic
beauty, the number of resting facilities and the amount of vegetation were assessed
(see Appendix 5).

An additional four options - “strongly need”, “need”, “do not care” and “do not need”
- were chosen for respondents in order to express their expectations for green spaces,
namely what the respondents would like to see improved, out of the given 12 items.
As mentioned in the theoretical part (see Sections 2.3 and 2.4), green spaces moderate
the local climate mainly through evaporative cooling, shading, wind speed control and
thermal capacity (Akbari et al. 2001; Andrade and Vieira 2007; Volker Heidt 2008).
Moreover, they purify the air via O2 release by vegetation, especially trees, and by
their trapping, absorption and degradation of air pollutants (Beckett et al. 1998;
Paoletti et al. 2011). Attitudes towards shade, wind-protected sites and water bodies
were, therefore, inquired to reveal respondents’ demands for regulating services.

According to existing research (see Section 2.4), apart from the desire for the natural
appearance of trees, flowers and water features (Yuen et al. 1999), and easy
accessibility (Jim and Chen 2006b; Sanesi and Chiarello 2006), citizens’ recreational
requirements for urban green spaces are mainly focused on features such as peace,
63
quiet, cleanliness (Jim and Chen 2006b; Tyrväinen et al. 2007; Yuen et al. 1999),
children’s playgrounds, athletic fields and shelters (Özgüner and Kendle 2006), as
well as enough recreational facilities such as tables, benches and drinking water
(Herzele and Wiedemann 2003; Oguz 2000). Therefore, in this study, a set of
questions related to opportunities for contact with nature, flowers, quiet areas, sports
and resting facilities, and children’s play areas were asked to evaluate the citizens’
recreational demands for urban green spaces.
3.2.4.3 Data Processing and Analysis
All the questionnaire data were compiled and inputted into Microsoft Excel XP
Professional Edition 2010 (Microsoft Corporation, USA), in order to develop a
research database. A series of statistical analyses was accomplished with the help of
SPSS 19.0 (IBM Corporation, USA) and Microsoft Excel XP Professional Edition
2010. General results were produced by way of descriptive statistics. Respondents’
personal information was analyzed using the Chi-squared test. The relationships
between respondents’ use behaviors / recreational activities / perceptions /
assessments / demands of green spaces and the socioeconomic / park variables were
analyzed using Student’s t-test and single-factor analysis of variance (one-way
ANOVA). Moreover, the results were also analyzed by significance analysis and
correlation analysis in order to reveal relationships such as the ones between
regulating services and recreation services or between specific ecosystem services and
green space site characters.
3.2.5 Document analysis
Documents containing text and images that have been recorded without a researcher’s
intervention (Bowen 2009), such as agendas, manuals, maps, charts and various
public records are found in libraries, newspaper archives, historical society offices,
and organizational or institutional files.

Documents provide background information and historical insights which can help
researchers understand the historical roots of specific issues and indicate the
conditions that produced the phenomena under investigation. Documents also provide
supplementary research data as a knowledge base and a means of discovering changes
and developments.

In this study, in order to obtain a better understanding of urban green space planning
and development in Xi’an and to help answer the research questions (especially the
last two), documentation and literature on planning and development were collected
and analyzed (see Table 3.6). At the city level, the documents included master plans
and green space system plans (see Table 3.6). Since the study specifically looked at
the 22 urban parks of Xi’an city, the documents at the park level mainly included park
maps and master plans for the specific parks (see Table 3.6). The review of
documents concentrated on extracting the main objectives and strategies for building
green space systems and specific green spaces such as urban parks.

64
At the same time, literature focused on green space planning and development in
Xi’an city both at the city level and park level, especially in the past ten years, were
reviewed (see Table 3.6). The review of literature was conducted mainly in order to
ascertain the development, achievements and status of the green space system and
individual green spaces, especially urban parks.

Table 3.6 Government documents on urban green spaces in Xi’an city

Document Remarks Key literature


City level Master plan of Xi’an city 1). Before 1995, there was Xueping Wu, 2012;
(1953-1972) no special urban green Yu et al., 2006;
Master plan of Xi’an city system plan for Xi’an city. Ting Wang, 2007;
(1980-2000) The master plans include Bo Liu, 2007;
Master plan of Xi’an city paragraphs concerning green Xiaoyan Zhao, 2007;
(1995-2010) spaces; Shuo Liu, 2013;
Master plan of Xi’an city 2). The master plans include Jing Feng, 2011;
(2008-2020) green space distribution Hongxing He, 2010
Urban green space system maps and land use maps
plan of Xi’an (1995-2010)
Urban green space system
plan of Xi’an (2008-2020)
Site (park) Park maps of the Yuan Li, 2014;
level investigated 22 parks; Master Hang Shi, 2013;
plan of the investigated 22 Dong Cui, 2011;
parks Lei Wang, 2010;
Ying Yan,2008;
Qin et al., 2006;
Shuo Liu, 2013;
Jing Feng, 2011;

65
4 Planning and Development of Green Spaces
in Xi’an
Overview of the planning documents
4.1.1 Emergence of a Green Space Structure Plan (1953-
1972)
Before 1949, China experienced years of wars and social transformation. The Second
World War and Civil War damaged the majority of urban greenery. During that time,
the development of urban green spaces in Xi’an had nearly stagnated, even though the
first modern park - Lian Hu Park, was built on the site of Qin Wang Garden in 1922.
Up to 1949, only a few green spaces and palaces were preserved in Xi’an.
Nevertheless, in subsequent years great changes were brought about.

With the establishment of the People’s Republic of China in 1949, the modern Xi’an
city was established. At that time, its area encompassed 131 km2 with a population of
one million. The city structure maintained the ancient chessboard shape and its grid of
streets. By 1949, the majority of examples of ancient landscape architecture and green
spaces in Xi’an had been destroyed in wars, notable exceptions being the Big Wild
Goose Pagoda, the Small Wild Goose Pagoda a and Qin Wang Garden (XUPB 1980).
The total greenery was reduced to 22 ha of green spaces and no more than 2500 street
trees in Xi’an (XUPB 1980).

The development of the green space structure of Xi’an was initiated in 1953, but
without specific and detailed discussions and planning maps (see Figure 4.1). The
existing urban green spaces were protected and maintained by the local government.
In the period of reconstruction, urban planning approaches were mainly influenced by
the former Soviet Union, including the detailed planning norms and regulations, as
well as urban planning and regional plans. The contents with regard to green spaces
were arranged together with the planning of the road system, housing and squares.
“Public urban green spaces” were put forward and underlined.

The 1953 version of the master plan with regard to green spaces focused on public
parks. The plan suggested that parks at the city level should provide various
recreational opportunities for the public, such as relaxation, sports, and meeting
friends, as well as cultural, scientific and educational activities. The design of the
parks were to consider different functional areas aiming at different groups of users,
including playgrounds, quiet areas, sporting fields, outdoor theatres, clubs, and
exhibitions with displays related to science and culture. Similar with the plan for
municipal parks, it was suggested to distribute 24 district parks in high-density
residential areas. In addition, 54 evenly distributed roadside green spaces were
planned for the whole city. Specifically, children’s parks were to be designed near
municipal park or other suitable places.
66
Figure 4.1 Master plan of Xi’an city (1953-1972) (source: Xi'an Urban Planning
Bureau)

4.1.2 The Green Space Structure Plan from the 1980s (1980-
2000)
In 1978, the built-up area of Xi’an was enlarged to 162 km2 with a population of 1.3
million (XUPB 1980), and the first urban green space plan was initiated (1980-2000).
In this plan, the goal was for the total area of green spaces to reach 1100 ha by the
year 2000, amounting to 10 m2 per person.

The 1980 version of the master plan (1980-2000) announced the establishment of an
easily accessible green space system encompassing municipal parks, district parks,
roadside green spaces, attached green spaces and street trees (see Figure 4.2). It
included planting of street trees and enlargement of pre-existing parks, as well as the
development of new public parks, roadside green spaces, squares and attached green
spaces. Street trees were connected to the road system and urban parks were to be
linked by continuous stretches of the street-side trees.

The 1980 version of the master plan also proposed the concept of “green space system”
and the principle of an interconnected green space structure. The main focus of urban
green spaces was still on urban parks and the construction of parks for the protection
of historical sites.

67
Figure 4.2 Master plan of Xi’an city (1980-2000) (source: Xi'an Urban Planning
Bureau)

4.1.3 The Green Space Structure Plan from the 1990s (1995-
2010)
Since the 1990s, Xi’an city has experienced an explosion of economic and urban
development. The bulit-up area of Xi’an increased to 275 km2 (XUPB 1995). The core
of the city has expanded to areas between the Second and Third Ring Road. The
budget for developing urban green spaces contiunued to grow and their development
was promoted (see Figure 4.3). Bulding on demolished areas in the city was
prohibited, and they were instead used for public green spaces. The main focus of
urban greening was on developing urban parks and roadside greening. During this
period, the urban planning law was issued and implemented by the goverment. The
norms governing the definition and classification of urban green spaces and urban
green space strucutre were published by the Construction Ministry of the P. R. China
in 1993 (see Table 2.1 and 2.2).

The 1995 version of the master plan made the first “urban green space structure plan”.
It outlined the green space structure with more detailed informaiton, including the
amount of vegetation cover, plant species and design approaches. It suggested that
green spaces should form a network interconnected via linear green spaces, such as
street trees (green belt), green buffers and waterside green spaces, to conncet the
green patches such as public parks, attachted and residential green spaces and
nurseries.

68
Figure 4.3 Master plan of Xi’an city (1995-2010) (source: Xi'an Urban Planning
Bureau)

4.1.4 The Green Space Structure Plan from the 2000s (2008-
2020)
By 2000, the built-up area had risen to 395 km2 and the population living in the built-
up urban area increased to 4 million (XUPB 2010). Since 2000, urban greening in
Xi’an received special attention in the context of urban development because the city
functions have evolved away from industry and agriculture, towards services.
Moreover, urban planning development has changed its focus from the central zone of
the city to the broader region (see Figure 4.4). The city started to develop “satellite
towns” in order to alleviate land use pressures and urban sprawl.

The overall strategy for urban green spaces gradually emerged based on the goals of
sustainable development and an ecologically sound environment, better living
conditions and increased tourism. It linked ecology to concepts of protecting natural
ecological resources, minimizing resource costs and damage, and constructing a
balanced urban ecological environment. To enable truly sustainable development, a
city should maintain clean air, water and environment, as well as a harmonious
society. The ecological principles for urban green space planning include using
waterside vegetation to help the restoration of rivers and wetlands, establishing green
buffers to screen air pollutants emanating from the local industry, using street trees
and other linear green spaces as green corridors for the movement of wild animals and
plants, and exploring historical relics to promote the city’s culture and public image.

At the same time, urban greening still focused on urban parks, residential green
spaces, roadside green spaces and street trees. At the city scale, the aim was to create
69
a system of various green patches interconnected with linear green spaces. At the
regional scale, the green network would integrate green (vegetation), blue (wetlands
and rivers) and mountain areas.

Figure 4.4 Master plan of Xi’an city (2008-2020) (source: Xi'an Urban Planning
Bureau)

In this plan, the specific planning of green space size, plant choice and design was
also addressed. It was proposed that all city dwellers should be able to reach
neighborhood greenery within a distance of maximum 300 m. The district and
municipal parks should be designed with maximum 500 m and 1000 m catchment
areas, respectively. Plant configurations should show flower displays in spring,
summer and autumn and keep a green appearance all year round. The rules suggested
that flowers and evergreen species should be planted more. Specifically, the design of
parks, attached green spaces, street trees, nurseries and green buffers was based on the
green structure plan of 1995, emphasizing the strength of ecological services of green
spaces. However, the government support for the achievement of green space plans
was stressed due to economic factors and land use pressures.
4.1.5 Summary of the Green Space Plan
A review of the different versions of the urban green space plan for Xi’an shows that
the plans have changed over time (see Table 4.1), and an overall planning concept and
strategy for urban green spaces has been formed. The images of Xi’an city as a
historical tourist city has been promoted. “Sustainable development” and natural
resource preservation are the currently dominant discourses in planning. The plans for
urban green spaces have moved from public green space development at a small scale
towards green structure development at a larger scale.

70
The planning of green spaces for social benefit was paid great attention since the 1953
version of the plan, including recreational activities (in the 1953 version), citizens’
living environment, local history and culture, aesthetics and city image (in the 1990
and 2000 versions). The suggested recreational activities mainly include relaxing,
socializing, exercise and cultural / historical / education activities. The principle that
parks should be divided into different areas for different recreation aims was
emphasized. Also, the accessibility of green spaces was stressed in the 2000 version
of the plan.

Table 4.1 History of the structure of green spaces in Xi’an

Green spaces structure Goals / principles of plan


th
Imperial period (11 Private gardens, Meeting the recreational and
century BC – 1912) monasteries spiritual demands of royalty
and the rich
Wartime period (1912- Private gardens, Meeting the recreation
1949) monasteries, and three demands of royalty and the
public parks rich; providing recreation to
the general public
Early construction period Few public parks Providing recreation
(1949 – 1978) opportunities for the public
with a consideration of
different use groups, such as
relaxation, sports, meeting
friends, or educational
activities related to culture
and science
Modern exploration period Public parks focusing on Forming a network of green
(1978 – 2008) historical sites, street trees, space structures by
roadside green spaces and connecting green patches via
attached green spaces green belts
New century (2008 - 2020) Public parks, attached green Preserving and optimizing
spaces, street trees, roadside the ecological value of green
green spaces, productive spaces; achieving the
green spaces, green buffers sustainable development of
etc. urban areas and protection of
natural resources;

The ecological benefits of green spaces have been increasingly stressed in the 1990
and 2000 versions of the plan. To preserve and optimize the ecological environment
has become one of the most important goals of urban green space planning (the 2000
version of the plan). The green areas in the city gradually increased. Many mature
trees present today in the city were planted during the period of the 1980s (Yan 2008).

71
The greening activities during the 1990s produced many of the green resources
available today, such as street trees, green buffers and woodlands in parks (Yan 2008).

The 2000 plan explains the ecological perspective of urban green space planning. It
aims to improve the environmental quality of the city and sustain the high-quality
development of urban ecosystems, i.e. “ecological functions” at both the park and city
level. An important improvement in the 2000 version of the plan is the proposition of
ecological principles for urban green space planning. At the same time, the plan calls
for the protection of natural ecological resources, the minimization of resource cost
and damage, and the construction of a balanced urban ecological environment.
The Development of Urban Green Spaces in
Practice
At the city level, the local government and administration play the most important
roles in the overall process of urban green space planning, management and
development. The main government organizations are the Planning Committee,
Planning Bureau, Park Administration and Forestry Bureau. In the domain of private
green spaces, an increasing number of institutions, enterprises and real-estate
developers have expressed an interest in greening their own areas, and initiated the
greening process. They are actors that directly influence the development of green
spaces. Academics and experts are also involved in the planning and development
process at both the city and local levels. Overall, the process of planning and
development of urban green spaces involves the interaction of these actors, mainly
within the public sector. However, other stakeholders and the public are far from
having an influence on decision making.

During the past years, the City Government of Xi’an has increasingly invested in the
development of urban green spaces (see Table 4.2). As shown in Table 4.2, the annual
budget has increased over a dozen times within the last years. Especially after 2011,
there has been a surge in the annual budget. At the city scale, the budget was allocated
to different types of urban green spaces, with the vast majority of it being invested in
the development of urban parks, squares and roadside greening. Urban parks account
for over half of the investment (statistical year book of investment in Chinese cities
2010-2014). Accompanied with the increasing investment, the area of green spaces
increased markedly, so that in 2010, the area of green spaces was 12140 ha, and four
years later it had risen to 18914 ha (see Table 4.2).

At the city scale, an analysis of green space accessibility based on Landsat images of
Xi’an city revealed that over half (52.22%) of the public green spaces can be reached
by foot in thirty minutes, and all the public green spaces can be accessed by bus and
automobile (Cui 2011). The accessibility of green spaces has therefore been
guaranteed by a convenient traffic system.

72
Table 4.2 Annual budgets for urban green spaces in Xi’an over five years
(Source: statistical year book of investment in Chinese cities 2010-2014)

Year
2010 2011 2012 2013 2014
Number of 66 68 72 81 85
municipal parks
and specialized
parks
Area of parks (ha) 1335 1478 1529 2406 2484
Area of green 12140 13680 15196 17751 18914
spaces (ha)
Annual budget 6,306,280 2,106,180 17,800,000 12,800,000 18,196,520
(thousand RMB)

In addition, the amount, distribution and characteristics of green spaces have been
investigated by a number of researchers. The main observations can be summarized as
follows: (1) There are fewer areas of green spaces within the second ring road than
within the third ring road (see Figure 4.5 and 4.6); (2) Green spaces within the second
ring road lack good connections due to the smaller amounts of roadside green spaces,
street trees and green belts (Cui 2011); (3) The distribution of green spaces is uneven
both within the urban and the regional area. The amount of green space in the
southern area is significantly larger than in the northern area (Cui 2011; Shi 2013;
Yan 2008; Yuan 2014 ). The main reason for this is the concentration of historical
sites and natural resources in the southern area, due to which many parks were built
especially in the south-east area of Xi’an city (Shi 2013; Wang 2007 ) (see Figure 4.4,
4.5 and 4.6). (4) A large area of green patches such as green buffers, woodlands and
nurseries is concentrated in the suburbs and regional areas (Cui 2011). For example,
woodlands are found only in very few larger parks, such as Xing Qing Park and Qu
Jiang Yi Zhi Park. In the main urban areas, fragmented green patches (e.g. urban
parks, residential and attached green spaces) and intermittent greening (e.g. street
trees, squares and roadside green spaces) are the general formats of the green space
system due to severe loss of greenery and intense land use pressures (Cui 2011; Shi
2013); (5) The green spaces such as residential and attached green spaces with smaller
areas are usually composed of only a few plant species that are repeated many times.
By contrast, the urban parks include greater numbers of species (Yan 2008; Yuan
2014 ); (6) Vegetation layers and compositions showed more configurations in urban
parks than in other types of green spaces, especially than in roadside plantings and
green buffers (Yuan 2014 ).

73
Figure 4.5 The regional development of the green space system of Xi’an city
based on the plans from the 1980s, 1990s and 2000s (Source: Yuan, 2014).

Figure 4.6 Locations of urban parks within the areas of Xi’an city inside the
third ring road in the year 2009 (source: Wang, 2010)

74
The development of urban parks can be stratified by park age. There were only seven
city parks in Xi’an before the 1980s. By 1990, the number of parks increased to 15,
reaching an approximate total area of 311 ha (XUPB 1980). Since then, the city
government has introduced stronger actions to develop public parks and squares. In
the context of increased budgets and investment, as well as stronger city policy for
improving the environment, at least one park has been constructed almost every year
(Xi’an Statistic Year Book 1990-2010). The most obvious progress in the greening of
Xi’an was made in the last two decades. By 2012, the number of parks increased to
55, with a total area of 1529 ha (Xi’an Statistic Year Book 2012), which was almost 4
times the number and 5 times the area from 1990.

The development of parks involves extensive preparation procedures and activities,


including improving existing green spaces, reclaiming land from demolished old
houses and expanding city borders. During the first stage of green space development
(before 1980), urban green spaces including parks concentrated in the city core areas,
i.e. inside the ancient city walls. The rebuilding of historical sites as public urban
parks was the main greening activity, such as in Lian Hu Park and Xing Qing Park.
Between 1980 and 2000, the greening activities focused on the areas within the
second ring road. Urban parks were developed mainly around historic sites, such as in
the case of Ci En Si Yi Zhi Park and Xi Qu Da Guan Yuan. After the year 2000, urban
parks experienced sharp development in demolished areas and new district areas.
Especially, the southern areas experienced increasing development, including
wasteland reclamation and demolishing of buildings to expand the green spaces,
which was done in large part following the establishment of the High-tech Zone and
Qu Jiang District in the early 1990s (Shi 2013). Subsequently, the green spaces were
extended to the areas between the Second and Third Ring Roads. Among the 22 parks
investigated in this study, twelve were built between the years 2000 and 2009,
including the majority of larger parks, such as Qu Jiang Tang City Wall Park, City
Sports Park and Qu Jiang Yi Zhi Park. Most parks such as the City Sports Park, Wen
Jing Park and Huan Cheng Xi Yuan were built on undeveloped land before the
expansion of the city. By contrast, Qu Jiang Tang City Wall Park, Tang Yan Road
Tang City Wall Park and Qu Jiang Yi Zhi Park were built around historic relics.
Challenges in the Planning of Urban Green
Spaces
First and foremost, the goal regarding the per-capita amount of green spaces set in the
master plans was not fulfilled. Despite the increasing amount of green spaces in Xi’an
as such, the per-capita green area is still unsatisfactory due to the immense land use
pressures and large population. In 2010, the green space area per person had risen to
5.16 m2 (He 2010). Thus, the amount of green space per person is below the national
standard of 7 m2. The per-capita green space area in Xi’an is obviously lower than the
11.2 m2 in Beijing and 12.5 m2 in Shanghai, (data from the local Statistic Bureau
website, Xi’an Statistical Yearbook 2012, http://tjj.xa.gov.cn/ptl/index.html), not to
mention the green space area in highly developed countries (XUPB 2010).
75
Secondly, there is an obvious lag in the structural planning of urban green spaces (Hu
and Duan 2004; Yan 2008). Moreover, the goals of urban green space planning
regarding their functional provision were not always fully understood and
implemented in practice. At project scale, the non-professional participants can
understand and realize the recreation values of green spaces much more than their
ecological values (Hu and Duan 2004; Yan 2008). The ecological functions of green
spaces were considered in all versions of the plans. However, during the process of
project construction, the plans with regard to ecological aspects of green spaces were
sometimes changed for various reasons, including the shortage of ecological
knowledge and insufficiencies of time, technical know-how and budget. During the
design of urban green spaces, ecological considerations included increasing the
vegetation cover at the expense of pavement, establishing multi-layered vegetation,
planting more trees, increasing the number of plant species and using native species.
Only the use of multi-layered vegetation and native species were followed through in
some green spaces of Xi’an city (Chen and Zhang 2005; Shi 2013; Xu and Zhang
2006; Yan 2008).

Thirdly, comprehensive green space design and planning is still deficient. For
instance, due to the limitations of natural greening recourses and inconsideration of
the distribution of green spaces, the provision of green spaces is uneven among
different areas, leading to unbalanced development in Xi’an (Hu and Duan 2004). As
shown in figures 4.5 and 4.6, green spaces are mainly distributed in the south-west
and south-east, especially in the areas between the second and third ring road. In fact,
the total area of green spaces in the southern area of Xi’an is three times higher than
in the north, and 10 times higher than in the west of the city (Shi 2013). This
unbalanced distribution has made the southern area much more attractive for living
and visiting (Shi 2013). Thus, how to promote the greening of northern areas and the
whole city is a problem being discussed at the moment.

In addition, technical support has not kept pace with the improved understanding of
ecology. In some cases, the planners wanted to introduce natural features such as
natural river banks and wild plants in green spaces, but these designs were replaced
by man-made features (Zhao et al. 2004). For example, the river banks and bottoms
were lined with concrete rather than natural materials, primarily due to technical
difficulties in managing the drainage and water levels (Zhao et al. 2004). Trees and
shrubs were replaced by lawns due to budget constraints. However, while the cost of
lawns seems lower at the beginning, the maintenance later on is normally expensive in
terms of labor, tools and water (Zhao et al. 2004). The challenges thus mainly lie in
the lack of investment in the early stages of green space development.

Furthermore, even though the planners knew the importance and value of urban green
spaces at city scale, faced with intense land use pressures and economic benefits, it
was difficult to control the land use of planned green spaces because there are no strict
borderlines between the different space uses (Yuan 2014 ). Therefore, the space
76
reserved for urban green spaces was occupied by commercial or other land use types
in some cases. These non-green lands negatively influenced the connections of
individual green spaces and hence their structural functions (Yuan 2014 ).

Generally, the main challenges facing urban green space planning are the
implementation of ecological principles in specific greening projects, the balanced
distribution and development of green spaces, sufficient investments and technical
support, strict rules and laws for protecting and constructing green spaces, and the
clarification of the responsibilities of individual government departments, especially
regarding land use patterns and investment in urban green spaces.

77
5 Types of Land Cover and Vegetation
Structures within Green Spaces of Xi’an
Land Cover Types within Urban Parks of Xi’an
According to the field investigation of 22 urban parks in Xi’an, six major land cover
categories were present: (1) vegetation, (2) pavements with scattered vegetation, (3)
pavements, (4) playgrounds, (5) buildings, and (6) water bodies (Table 5.1 and Annex
1).

As shown in Table 5.1, except for Qu Jiang Yi Zhi Park, the majority of parks were
primarily covered with vegetation, with the percentage ranging from 43.5% to 85.9%
(Table 5.1). Other categories included (in descending order in terms of area):
pavements with scattered vegetation, water bodies, pavements, buildings, and
playgrounds. The pavements were mainly present in the form of squares. The
percentage area of pavements was significantly lower than that of vegetation, with the
highest coverage being 29.7% and the lowest 2.0%. Furthermore, the area cover of
buildings accounted for a very low percentage of the total area, ranging from 0.2% to
16.5% (Table 5.1). The majority of the buildings contained offices and businesses,
such as restaurants, bars and shops. Elements of garden architecture that provide
shading, rest and aesthetics were also found in over half of the parks, albeit at lower
area percentages (Table 5.1).

The remaining three categories of land cover were not present within every park.
Pavements with scattered vegetation were mainly squares covered with sparse
vegetation such as trees, or trees covering shrubs. Xin Ji Yuan Park and Wen Jing Park
had no pavements with scattered vegetation, while in the other 20 parks its percentage
ranged from 0.1% to 27% of total area (Table 5.1). Water bodies in the parks of Xi’an
city mainly included lakes and fountains. In the majority of parks, water surfaces were
lakes. Min Su Park and Tang Yan Road Tang City Wall Park had no bodies of water.
The remaining parks had water coverage ranging from 0.7% in Tu Men Road Park to
54.9% in Qu Jiang Yi Zhi Park (Table 5.1). Within the investigated 22 parks, the least
represented type with the smallest percentage area of land cover was playgrounds. In
fact, half of the parks had no specifically designated and designed playgrounds. In the
parks having playgrounds, the percentage area ranged from a low of 1.1% in Lao
Dong Park to a maximal 10.4% in Children’s Park, Tu Men Road Park and Chang Le
Park (Table 5.1).

78
Table 5.1 Percentages of different land cover types within the 22 investigated urban parks of Xi’an city

Children's Park

Tu Men Road Park

Fang Zhi Park

Ci En Si Yi Zhi Park

Xi Qu Da Guan Yuan

Lian Hu Park

Min Su Park

Xin Ji Yuan Park

Wen Jing Park

Mu Ta Si Park

Mu Dan Yuan

Lao Dong Park

Ge Ming Park

Huan Cheng Xi Yuan

Yong Yang Park

Chang Le Park

Feng Qing Park

Wall Park
Qu Jiang Tang City

City Sports Park

City Wall Park


Tang Yan Road Tang

Xing Qing Park

Qu Jiang Yi Zhi Park


Parks

Land cover types


Vegetation 48.8 66.3 79.9 56.9 62.9 47.2 79.8 85.2 84.0 61.0 84.0 63.6 65.6 64.1 43.5 55.9 69.1 59.1 56.0 85.9 61.3 31.7
Pavements with
14.2 15.8 8.8 20.1 27.0 12.2 3.3 9.2 2.7 12.5 14.0 2.8 24.8 17.9 1.0 3.7 18.0 0.1 3.1 3.4
scattered vegetation
Pavements 9.0 4.3 7.3 2.0 6.5 4.0 6.0 7.8 6.1 21.8 9.9 2.0 7.1 29.7 6.4 2.4 6.4 27.7 10.7 12.7 4.3 7.2
Playgrounds 10.4 10.4 7.9 4.3 1.1 5.8 10.4 3.4 4.1 1.2 7.3
Business
14.7 2.2 0.6 15.8 1.8 0.7 0.2 0.2 2.1 0.2 0.4 8.2 1.7 10.4 4.1 0.4 0.9 5.8 1.3 3.8 0.4
buildings
Buildings
Garden
0.4 0.9 0.7 0.1 10.7 0.3 1.3 2.3 0.9 2.4 4.7 0.6 0.3 0.8 2.4
architecture
Water Lakes 2.8 0.7 2.6 4.5 1.8 30.3 6.9 3.2 7.8 1.6 10.2 5.2 1.0 14.8 9.3 14.7 3.8 8.3 19.5 54.9
bodies Fountains 0.4 0.2
Total (ha)
Vegetation 1.03 1.85 2.74 2.51 2.80 2.24 4.80 5.15 5.25 3.85 5.84 4.66 5.08 7.02 4.89 9.59 15.56 18.89 18.94 34.84 30.02 16.05
Pavements with
0.30 0.44 0.30 0.89 1.20 0.70 0.20 0 0 0.58 0.19 0.92 1.08 0.31 2.79 3.07 0.23 1.17 6.08 0.04 1.53 1.72
scattered vegetation
Pavements 0.19 0.12 0.25 0.09 0.29 0.23 0.36 0.47 0.38 1.38 0.69 0.15 0.55 3.26 0.72 0.41 1.43 8.86 3.61 5.17 2.11 3.65
Playgrounds 0.22 0.29 0 0 0 0.45 0 0 0.27 0 0 0.08 0.45 0 0 1.78 0.76 1.32 0.42 0 3.55 0
Buildings 0.31 0.07 0.06 0.74 0.07 0.04 0.65 0.01 0.16 0.01 0.12 0.77 0.19 0.26 1.17 0.71 1.15 0.46 1.98 0.52 2.23 1.43
Water bodies 0.06 0.02 0.09 0.20 0.08 1.59* 0 0.42 0.20 0.49 0.11 0.75 0.40 0.11 1.67 1.6 3.39 1.29 2.81 0 9.55 27.82
Area 2.11 2.79 3.44 4.43 4.44 5.25 6.01 6.05 6.25 6.31 6.95 7.33 7.75 10.96 11.24 17.16 22.52 31.99 33.84 40.57 48.99 50.67

* 0.47 ha of water plants were present within the 1.59 ha of water area in Lian Hu Park

79
Using the chi-squared test to analyze the relationships between land cover types and
park characteristics, it was found that park age and location had significant
relationships with the presence of playground features. In this study, a park’s age was
determined by the year it was built. The parks that were built before 1990 were
termed older parks, and those built after 1990 were termed newer parks. Among the
seven older parks, six had area percentages of playgrounds from 0-25%. By contrast,
among the 15 newer parks, only five provided playgrounds at all (X2 =5.238, P =
0.022) (Tables 3.1 and 5.1). While only three out of twelve parks that are located
between the second and third ring roads had playgrounds, eight out of ten parks
located within the second ring road had playgrounds. (X2 = 6.600, P = 0.010) (Figure
3.1 and Table 5. 1).
Vegetation Structure Types in Urban Parks of
Xi’an
Since the ecosystem services of green spaces are mainly generated by vegetation and
influenced by its structure, knowing the variations of vegetation structure in parks is
important for understanding their provision of ecosystem services. According to the
field survey, there were a total of eleven different layers and compositions of
vegetation structure within the 22 parks (Table 5.2 and Annex 1). These vegetation
structure types were: (1) lawns, (2) flower beds, (3) water plants, (4) bamboos, (5)
shrubs, (6) woodlands, (7) trees covering shrubs, (8) trees covering lawns, (9) shrubs
covering lawns, (10) trees covering shrubs and lawns, and (11) multi-layered trees
covering shrubs and lawns.

In all of the 22 urban parks, multi-layered trees covering shrubs and lawns accounted
for the highest proportion of the total area among all the types of vegetation structure,
followed by trees covering shrubs and lawns, deciduous broadleaf woodlands, and
trees covering lawns (Table 5.3). All the parks had vegetation with three and more
layers, with the coverage ranging from 16.2% to 80.1%, and reaching more than 50%
in over half of the parks. Eight parks had trees covering lawns, and their percentage
ranged from 2.9% to 20.2%. Nine parks had deciduous broadleaf woodlands, with
area percentages ranging from 1.7% to 32.7%.

The proportion of evergreen coniferous woodlands, shrubs covering lawns and trees
covering shrubs came next. Shrubs and lawns accounted for a lower proportion of
coverage, while the proportion of bamboos, water plants, flower beds and evergreen
broadleaf woodlands accounted for the lowest coverage. For example, seven parks
had evergreen coniferous woodlands, with area percentages ranging from 0.6% to
15.8% (Table 5.3). Ten parks had lawns, and their percentage ranged from 0.8% to
6.5%, with a majority around 2.0% (Table 5.3). Flower beds, water plants, bamboos
and shrubs were found in only a few parks, and with very low percentages of
coverage (Table 5.3).

The chi-squared test showed that vegetation structures had significant relationships
80
with park area and location. In this study, considering the variations of their total area,
parks were divided into four groups: small area (0-5 ha), small to medium area (5-10
ha), medium area (10-30 ha), and large area (30-60 ha). There were lawns in all five
parks with areas of 30-60 ha, while there were no lawns in any of the parks with areas
of 10-30 ha (X2 =11.110, P = 0.011) (Table 3.1 and 5.3). Considering the parks’
locations, the majority of those located within the third ring road had lawns. By
contrast, only one of the parks that had lawns was located within the second ring road
(X2 = 9.295, P = 0.002) (Figure 3.1 and Table 5.3). Trees covering shrubs were mainly
found in parks with areas from 10-30 ha (X2 = 8.192, P = 0.042), and only one small
to medium (5-10 ha) and one large (30-60 ha) park contained the vegetation type of
trees covering shrubs (Tables 3.1 and 5.3). In addition, seven parks had evergreen
coniferous woodlands, six of which were located within the third ring road (X2 =
4.023, P = 0.045) (Figure 3.1 and Table 5.3).

Table 5.2 Vegetation structures within urban parks of Xi’an city

Number of Type Sub-types


layers
1 Lawns
Flower beds
Water plants (Reed/Lotus)
Bamboos
Shrubs (including trimmed Evergreen coniferous shrubs
shrubs and hedges) Deciduous coniferous shrubs
Evergreen broadleaf shrubs
Deciduous broadleaf shrubs
Woodlands Evergreen coniferous woodlands
Deciduous coniferous woodlands
Evergreen broadleaf woodlands
Deciduous broadleaf woodlands
2 Trees covering lawns
Trees covering shrubs
Shrubs covering lawns
3 Trees covering shrubs and lawns
>3 Multi-layered trees covering
shrubs and lawns

The majority of parks investigated in this study had three to six types of vegetation
structure. Xing Qing Park contained the greatest number of types of vegetation
structure (nine): evergreen coniferous woodlands (0.6%), lawns (1.4%), evergreen
broadleaf woodlands (3.1%), flower beds (3.3%), trees covering shrubs (3.7%), trees
covering lawns (4.8%), deciduous broadleaf woodlands (6.5%), multi-layered trees
covering shrubs and lawns (12.2%), and trees covering shrubs and lawns (25.9%)

81
(Table 5.3). Xin Ji Yuan Park followed with eight types (Table 5.3). However, the
majority of the parks (20) were covered by a lower variety of vegetation structures
(less than seven types). These included Xi Qu Da Guan Yuan (six), Ci En Si Park
(five), Chang Le Park (four) and Fang Zhi Park (three) (Table 5.3). Two parks had
only two types of vegetation structure and three parks had only one (Table 5.3).

The smallest parks: Children’s park and Tu Men Road Park only had multi-layered
vegetation. Moreover, they had a relatively large cover of pavements with scattered
vegetation, and the highest percentage coverage of playgrounds. Min Su Park also
only contained multi-layered vegetation. Interestingly, having the largest percentage
of garden architecture (10.7%) of all the investigated parks is another feature of Min
Su Park.

Qu Jiang Yi Zhi Park had the largest coverage of lakes (54.9%). In addition to lakes,
the park was mainly covered by lawns, and trees covering shrubs and lawns.
Additionally, Lian Hu Park also had a large coverage with lakes (30.3%), though trees
covering shrubs and lawns (37.4%) accounted for the largest percentage if its total
area. Water plants (8.2%) were a prominent feature of Lian Hu Park. Wen Jing Park
also had 0.2% of water plants. Moreover, Wen Jing Park had 9.1% of bamboos, 8% of
evergreen coniferous woodlands, 0.2% of flowerbeds, and 3% of lawns. In spite of
this diversity of vegetation types, the 63.6% of trees covering shrubs and lawns
represented the overwhelming majority of surface cover in this park.

Extensive shrub cover was found in Yong Yang Park, Feng Qing Park and Tang Yan
Road Tang City Wall Park (Table 5.3). Yong Yang Park had the largest percentage
cover of pavements with scattered vegetation (24.8%), followed by 14.8% and 10.4%
of water and buildings, respectively.

There were significant relationships between park characteristics and total number of
types of land cover and vegetation structure, as determined using the chi-squared test.
However, such relationships were not found between park characteristics and either
the types of land cover or of vegetation structure. Among the seven older (< 1990s)
parks, six had over eight types of land cover and vegetation structure, in comparison
to nine out of the 15 newer parks (X2 = 19.695, P = 0.012). The parks located within
the second ring road were mostly older parks (Table 3.1 and Figure 3.1). When
stratified by park age, park location showed a similar influence on the types of
vegetation structures (X2 = 17.160, P = 0.028).

82
Table 5.3 Percentages of different vegetation structures within the 22 studied urban parks of Xi’an city

Children's Park

Tu Men Road Park

Fang Zhi Park

Ci En Si Yi Zhi Park

Xi Qu Da Guan Yuan

Lian Hu Park

Min Su Park

Xin Ji Yuan Park

Wen Jing Park

Mu Ta Si Park

Mu Dan Yuan

Lao Dong Park

Ge Ming Park

Huan Cheng Xi Yuan

Yong Yang Park

Chang Le Park

Feng Qing Park

Park
Qu Jiang Tang City Wall

City Sports Park

Wall Park
Tang Yan Road Tang City

Xing Qing Park

Qu Jiang Yi Zhi Park


Parks

Vegetation structures

Lawns 1.5 1.8 3.6 2.2 3.0 6.5 3.0 0.8 1.4 2.7

Flower beds 0.2 0.8 3.3

Water plants 8.2 0.2

Bamboos 9.1

Shrubs 7.7 13.9 2.9

Evergreen broadleaf woodlands 3.1

Evergreen coniferous woodlands 2.5 15.8 13.7 6.5 8.0 7.2 0.6

Deciduous broadleaf woodlands 5.6 13.9 3.9 30.8 9.0 6.4 1.7 32.7 6.5

Trees covering shrubs 4.1 6.1 14.3 11.9 3.7

Trees covering lawns 2.9 9.9 10.7 3.7 5.1 17.7 20.2 4.8

Shrubs covering lawns 12.2 3.8 11.1 1.0 7.0 2.6 5.9 4.8 2.5

Trees covering shrubs and lawns 3.2 37.4 42.5 26.5 63.6 22.2 55.3 50.7 16.2 20.1 30.0 12.6 29.4 25.9 29.0
Multi-layered trees covering
48.8 66.3 66.2 45.8 24.8 1.6 23.7 10.3 61.0 80.1 10.6 34.9 16.4 37.9 12.2
shrubs and lawns

83
Discussion
5.3.1 Comparison of Land Cover / Vegetation Structure
Types in Xi’an city with those Reported in Similar Studies
There were six types of land cover and eleven types of vegetation structure within the
urban parks of Xi’an city (see Tables 5.1 and 5.2). Layers and compositions of
vegetation structure varied from simple to relatively complex (see Table 5.3, Figure
5.1 and Annex 2). Perhaps unsurprisingly, vegetation accounted for the highest
percentage of area in the majority of parks. Multi-layered trees covering shrubs and
lawns, and trees covering shrubs and lawns were the dominant types of vegetation
structure (see Table 5.3).

Considering the categories of land cover and vegetation, older parks showed relatively
higher numbers of types of land cover and vegetation structures than newer parks.
Green spaces in Xi’an city were built successively outward. Therefore, older parks in
most cases are the parks located within the second ring road. The older parks had a
long time to form the mature appearance of their vegetation, and therefore benefited
from the maintenance of the multi-layered vegetation and compositions of different
plant communities.

Figure 5.1 Comparison of land cover and vegetation structure between


Children’s Park (left) and Xing Qing Park (right)

By implementing a common classification, specific types of land cover and vegetation


structure within urban green spaces can be compared. It is notable that open soils, ice
houses, car parks, walls, rough grass, infested grass, pastures, hay meadows and
agricultural crops were not found in the parks of Xi’an, in contrast to findings from
studies in other countries (e.g. Pauleit and Duhme, 2000; Young and Jarvis, 2001;
Cornelis and Hermy, 2004 and Gao et al., 2012). These results may be explained by
the fact that in Chinese cities, as highly artificial urban ecosystems, urban green
spaces usually lack natural elements and original types of vegetation. Pavements often
replace open soils. Mowed lawns, shrubs and hedges have replaced naturally growing
84
vegetation (e.g. rough grass, pastures and hay meadows). Agricultural crops and other
economic plants were also not present due to the limited palette of plant species (Xu
and Zhang 2006). The predominant design styles of parks obviously acted to limit
variations in land cover and vegetation structures.

The percentage area of pavements in the urban parks of Xi’an was relatively higher
than in the parks from studies by Pauleit and Duhme (2000) and Young and Jarvis
(2001). For example, in parks in Munich, pavements accounted for 3% of the total
area, while among the 22 investigated parks in Xi’an, ten had over 10% of their area
in pavements, even reaching 29.7% in Huan Cheng Xi Yuan. The increased amount of
sealed surfaces can change the ecological performance of parks in terms of decreasing
the parks’ impact on air temperature regulation because of reduced transpiration
(Lhomme and Monteny 2000; Zhang et al. 2014). However, this relatively high
proportion of pavements within green spaces can provide more space for holding
public meetings, public events, and recreational activities, by being able to admit
large numbers of citizens without damaging the vegetation.

In addition, the characteristics of vegetation in Xi’an differed in some areas from


those described in other studies. For example, there were no large areas of lawns in
the green spaces of Xi’an. In fact, the percentage area of lawns ranged from 0.8% to
6.5% (see Table 5.3). By comparison, lawns occupied around 80% of the total park
areas in Munich (Pauleit and Duhme 2000), and grasslands accounted for almost
100% in some parks of Black Country in the UK (Young and Jarvis 2001). By
contrast, the trees cover was found to be relatively low in these studies. For example,
in the study conducted in Munich, the percentage of trees was 15% of the total area of
the investigated parks (Pauleit and Duhme 2000), whereas in Xi’an, tree cover
accounted for at least 26%, and in some cases for over half of the park area (see
Figure 7.3). This lack of lawns may lead to a decrease of open scenery and limit some
recreational activities. However, the higher coverage of trees is beneficial for the
enhancement of regulating services, because trees (and especially mature trees) are
considered the most important factor in decreasing air temperature and air pollution
(Nowak 2002; Nowak and Dwyer 2007; Picot 2004). The emphasis on trees rather
than lawns may reveal a focus on regulating services of green spaces in Xi’an city due
to very tense land use pressures and limited areas of green spaces.
5.3.2 The Relationships between Land Cover / Vegetation
Structure and Park Variables
The chi-squared test revealed that some types of land cover and vegetation structures
were influenced by the variables of parks. For example, playgrounds had significant
relationships with park age and location, while park location and area influenced the
distribution of lawns.

In the 22 investigated parks, all of the five largest, with areas of 30-60 ha, had open
lawns. However, the four medium parks with areas of 10-30 ha had no open lawns at
85
all. Under the severe land use pressures in Xi’an, park designers evidently first
considered the benefits which are connected with citizens’ well-beings, such as
regulating services, as has been stressed by the green space plans. Compositions of
groups of trees and shrubs rather than open lawns possibly provide more regulating
services because of their stronger transpiration, shading and absorption of air
pollutants (see Section 2.3.2.2). Therefore, the vegetation structures with multiple
layers and large canopies are considered more suitable for parks with limited area. As
a result, the larger parks can therefore provide open lawns due to their sufficiently
large areas.

However, compared to the small-sized parks, the shortage of open lawns in medium
sized parks may be caused by the individual parks’ features and design styles. Every
park has its own compositions of land cover and vegetation structures, which makes it
difficult to explain all the reasons for their creation. In the four medium-sized parks,
multi-layered vegetation occupied the largest percentage of area. Additionally, two of
them had over 10% of water surface cover, and three had over 20% of pavements and
pavements with scattered vegetation (see Table 5.1). For example, Huan Cheng Xi
Yuan had the largest percentage of pavements (29.7%). Yong Yang Park had the
largest percentage of pavements with scattered vegetation (24.8%), in addition to its
14.8% of water surface, 7.7% of shrubs and 7.2% of broadleaf evergreen woodlands.
Chang Le Park had the largest percentage of playgrounds (10.4%) and 17.9% of
pavements with scattered vegetation. Feng Qing Park had 13.9% of shrubs and 14.7%
of water surface (see Tables 5.3 and 5.1). Thus, since there are many other types of
land cover and vegetation in a limited area, it is difficult to design special open lawns
in such parks.

While having no lawns, the medium-sized parks (10-30 ha) usually had the vegetation
structure of trees covering shrubs. These parks also had higher area percentages of
impervious surfaces (see Table 5.1). The pavements in these parks replaced the lawns
or grasses. Therefore, trees covering shrubs was a typical vegetation structure in this
case. Generally, as discussed above, the medium-sized parks were usually
characterized by trees covering shrubs, pavements, water bodies, and no lawns.

Considering the parks’ locations, it was found that the majority of parks with lawns
were located between the second and third ring roads, with only one lawn-containing
park located inside the second ring road. The parks located inside the third ring road
were always newer, and a part of them were large. Traditional design styles in Chinese
parks are characterized by abundant vegetation layers, winding roads, and dedicated
rock formations (Lou 2003). Moreover, the vegetation and buildings should set each
other off (Lou 2003). Nowadays, park designers may try to present open scenery and
playing fields via lawns in parks, as is traditional for western gardens.

In addition, seven parks had evergreen coniferous woodlands, six of which were
located within the third ring road. Since the parks located inside the third ring road are
86
usually the newer parks, this observation may indicate that, in recent years, park
designers have started to realize the importance of evergreens (see Section 2.3.2.2).
They therefore appear to have chosen more evergreen coniferous trees in order to
meet the demand for regulating functions from the viewpoint of ecology, and to keep
the parks green all year round from the viewpoint of aesthetics, as the green space
plan suggested.

However, the older parks located within the second ring road are more likely to
provide playgrounds. This is because the provision of recreational opportunities such
as playgrounds in parks was one of the most important objectives in the pre-1990s
versions of the master plan and green space plans in Xi’an (see Section 4.1).
Therefore, under the direction of master plans from that period, park designers usually
considered the provision of playgrounds. In later years, the focus on providing special
playgrounds in parks was reduced due to the higher emphasis on ecological functions
of parks and increased opportunities for entertainment in various other venues in the
city.

In summary, the urban parks of Xi’an city have varied layers and compositions of
land cover and vegetation, and are characterized by fewer lawns and more pavements
than their western counterparts. The majority of the parks have large covers of
vegetation with more than two layers, and great expanses of pavements. A continuous
cover of flower beds, bamboos or water plants is rarely seen. Park age, area and
location were the factors that influenced the types of land cover and vegetation
structure within parks. Among the 22 investigated parks, older parks usually showed
more variations of land cover and vegetation structure. The larger parks located within
the third ring road all had lawns. However, special playgrounds can be found in the
older parks located inside the second ring road. In addition, evergreen coniferous
woodlands were frequently found in the parks that were located inside the third ring
road, which are mostly newer. Although it is difficult to explain all the attributes of
land cover and vegetation types in the parks, the main results can be explained by the
prescriptions made in the master plans of green spaces during the time of each park’s
construction, in addition to the size and design style of the park in question.

To our best knowledge, this is the first study that attempts to investigate and classify
the types of land cover and vegetation structure using a large number of public parks
in Xi’an city. It compares the types of green spaces in Xi’an city with other studies,
and explores the factors influencing the types of land cover and vegetation structure.
Other studies focused on one or several green spaces (e.g. Qin et al., 2006), or on
green space as a whole (e.g. Shi, 2013), but did not provide land cover maps including
the vegetation information both in vertical and horizontal dimension, as is the case in
this study. However, this study does not provide more detailed information, such as
the exact plant species use, tree diameter at breast height, and shapes of different
patches of surface cover within green spaces. More detailed information may
therefore merit further investigation in the future.
87
6 Citizens’ Use, Perceptions, Assessments and
Demands for Green Spaces in Xi’an
Response Rate and Respondents’ Characteristics
Of the 500 administered questionnaires, 428 were returned with complete answers.
Among these, the numbers of male and female respondents were equal (Table 6.1).
The ages of respondents were mainly between 26 and 65. The age group 26-45
accounted for 42.5%, followed by 46-65 (34.3%) and 15-25 (15.9%). The senior >65
group only occupied 7.2%. Over half of respondents (61.9%) had at least one child
(aged 0-14). Couples and singles accounted for 29.2% and 8.9% of the sample,
respectively. In addition, the majority of respondents (61.4%) were ordinary working
people. Additionally, 11.7% were retirees and 10.7% were self-employment. Very
limited proportions comprised students and unemployed persons.

Table 6.1 Respondents’ personal information

Personal Category Percentage of Percentage of Chi-squared p-value


information respondents census test
Gender Male 50.0 51.3 0.273 0.601
Female 50.0 48.7
Age 15-25 15.9 78.7 2.974 0.085
26-45 42.5
46-65 34.3
>65 7.2 8.5
Children No 38.1 12.0 275.72 <0.001

Yes 61.9 88.0

Vocation Student 9.4 22.9 108.31 <0.001


Employed 61.4 61.9
Self-employed 10.7 5.4
Unemployed 6.8 4.0
Retiree 11.7 5.8
Data source: Census from Xi’an Statistic Yearbook (2011)
Significant at 0.05 level

Socioeconomic profiles were compared with the 2011 population census (Xi’an
Statistic Yearbook 2011). The chi-squared test showed that gender and age matched
the census data well (P > 0.05). However, fertility status and vocation deviated from

88
the census data (P < 0.05). Compared with the census, fewer respondents had
children.

This result may be due to lower participation of visitors accompanied by children


because they could not leave their children alone during the time needed to complete
the questionnaire. The percentage of employed respondents was nearly equal to that in
the census. Students were markedly less represented than in the census because
students were defined in the census as all people from nurseries, primary schools,
high schools and universities, while in this survey, the category only included those
over 15 years old due to their ability to understand this type of questionnaire. More
retirees than in the census may indicate that retirees have more spare time to go to
parks and/or were more inclined to take part in the survey.

Citizens’ Use Behaviors of Green Spaces


6.2.1 Use Types and Variations among Social Groups
Public parks were chosen as the most often used type of green spaces by residents in
Xi’an. In fact, an overwhelming majority comprising 80.8% of the respondents stated
that they most frequently used the public parks (Table 6.2). Residential green spaces
(75.7%) and green spaces near schools or work places (21.3%) were their second and
third choices. Roadside green spaces and squares were seldom chosen (11.0%). The
analysis of statistical significance of the connection between often-used types of green
spaces and socioeconomic variables showed that vocation had a significant
relationship with the preference for roadside green spaces and squares (p < 0.05).
More employed respondents (22.4%) were frequent users of roadside green spaces
and squares than other groups, with the percentage ranging from 1.4% among students
to 3.6% among the self-employed (p < 0.05). This may be due to the simple fact that
employed respondents often pass through roadside green spaces or squares on their
way to work. The other three socioeconomic factors had no statistically significant
association with often-used types of green spaces (p > 0.05).

Table 6.2 Types of often-used green spaces (percentage)

Type of green space Very often Often Occasionally


Public parks 80.8 17.3 1.9
Residential green spaces 75.7 22.7 1.6
Roadside green spaces and
11.0 48.6 40.4
squares
Green spaces near work
21.3 43.0 35.7
places or schools
89
6.2.2 Citizens’ Use Frequencies, Times and Durations of
Visits, and Variations among Social Groups
Over half of the respondents (54.0%) stated that they travelled 6-15 minutes to their
often-used green spaces (Table 6.3). 22.0% and 20.0% of respondents usually spent
no more than 5 minutes and 16-30 minutes on the road, respectively. Only a small
minority of visitors (4.0%) needed more than 30 minutes’ travel time. Regarding the
frequency of visits to green spaces, 48.6% of respondents stated that they visited
green spaces several times per week. The daily users accounted for 30.4%, followed
by weekly (14.7%), 1-3 times per months (4.9%) and monthly users (1.4%). It can
therefore be seen that the vast majority of respondents were active in using green
spaces. However, the duration of a typical visit was not long, often involving a stay of
one to two hours (52.1%) or less than on hour (30.1%). Only 15.7% and 2.1% stayed
half a day or nearly one day, respectively. In addition, there was no big difference
between the most frequent visits on weekdays (36.7%) and on weekends (43.7%),
even if the visits were slightly more numerous on weekends than on weekdays.
Moreover, visits by respondents from the “very often” category were typically
conducted in the evening (42.5%) and morning (39.0%).

In order to simplify the analysis of the statistical relationsships with the attributes of
park users, a classification representing the use frequency and duration was applied.
For example, the responses regarding use frequency were scored as daily (6), several
times per week (5), weekly (4), 1-3 times per month (3), monthly (2), several times
per year (1) and less (0), and the same principle as used to score the duration of a
typical use. The associations of respondents’ socioeconomic characteristics with the
use frequency and duration per visit were measured using F or t-tests via the average
scores (Table 6.4 and 6.5). In general, the respondent’s age, family status and vocation
influenced their use habits the most. Age showed a significant relationship with use
frequency and duration. The majority of active users wo reported visiting daily and
several times a week belonged to the 46-65 (85.7%) and > 65 age groups (77.4%)
(Table 6.4). Also, most respondents from the 46-65 (73.5%) and > 65 age groups
(80.6%) expressed that they stayed more than two hours, and some even half a day
per visit (Table 6.5). Children were another factor infuencing the use frequency as
revealed by the fact that respondents who had children visited green spaces more
actively than their childless peers (82.3% versus 73.6%). In addition to age, vocation
was also a factor affecting the stay time per visit in green spaces. Retirees prefered to
stay significantly longer per visit than the other respondents, and 42.9% stated that
they spent nearly half a day in green spaces at a time (Table 6.5). Most of the
90
respondents in the >65 group were retirees. In summary, the major user groups of
green sapces in Xi’an were older respondents, as well as unemployed or retired
females who have children living with them.

In addition, although over half of the respondents (262 versus 428) spent less than 15
minues on their way to the green space and simultaneously used the green space
several times a week or even daily, no significant relatioship was found between the
distance and use frequency using the chi-suqared test (x2 = 9.877, P > 0.05). Over half
of the repondents (218 vs. 428) stayed one to two hours and even longer per visit in
the short-distance parks (< 15 minutes away). However, there was also no significant
relationship between them (x2 = 14.821, p > 0.05).

Table 6.3 Use frequencies, times and durations of visits to green spaces

Category Survey results (%)


Time spent on the way ≤5 min 22.0
6-15 min 54.0
16-30 min 20.0
> 30 min 4.0
Use frequency Daily 30.4
Several times per week 48.6
Weekly 14.7
1-3 times per month 4.9
Monthly 1.4
Several times per year 0
Less 0
Duration of use Less than one hour 30.1
One to two hours 52.1
Half a day 15.7
Whole day 2.1
Very often Often Occasionally
Time of the day In the morning (before 12:00) 39.0 54.4 6.5
In the afternoon (12:00-18:00) 7.5 37.9 54.7
In the evening (after 18:00) 42.5 45.6 11.9
Time of the week Weekdays 36.7 56.8 6.5
Weekends 43.7 50.7 5.6

91
Table 6.4 Use frequency by socioeconomic variables

Variable Category Average score of F or t-statistic p-value


use frequency
Gender Male 4.96 t = -1.2638 0.2070
Female 5.07
Age 15-25 4.91 F = 3.26 0.0214*
26-45 4.91
46-65 5.19
>65 5.06
Children No 4.89 t = -2.3492 0.0193*
Yes 5.09
Vocation Student 4.88 F = 0.87 0.4816
Employed 4.99
Self-employed 5.06
Unemployed 5.21
Retiree 5.12
Average score: daily = 6, several times per week = 5, weekly = 4, 1-3 times per month
= 3, monthly = 2, several times per year = 1, less = 0;
Significant at 0.05 level (*p < 0.05)

Table 6.5 Duration per visit by socioeconomic variables

Variable Category Average score of F or t-statistic p-value


duration per visit
Gender Male 1.87 t = -0.7944 0.4274
Female 1.93
Age 15-25 1.79 F = 3.62 0.0132*
26-45 1.80
46-65 2.03
>65 2.03
Children No 1.88 T = -0.4419 0.6588
Yes 1.91
Vocation Student 1.85 F = 3.61 0.0066*
Employed 1.85
Self-employed 1.85
Unemployed 1.83
Retiree 2.26
Average score: less than one hour =1, one to two hours = 2, half a day =3, nearly a
whole day = 4;
Significant at 0.05 level (*p < 0.05)

92
6.2.3 Citizens’ Use Frequency and Duration of Visits, and
Variations among Park Variables
Similar with the analysis of social groups, the associations of park characteristics with
the use frequency and duration per visit were also measured by F or t-tests via the
average scores (Tables 6.6 and 6.7). Generally, park area, age and location did not
have a significant influence on people’s use frequency and stay time. However, tree
cover and evergreen species cover showed significant correlations with people’s use
frequency (see Table 6.6). Multi-layered vegetation cover had a significant influence
on peoples’ stay time per visit (see Table 6.7). The parks having 26-50% of tree
coverage were used often by 83.2% of respondents. The percentage of frequent use
decreased with increasing coverage of trees. Only 64.5% of respondents stated to
frequently use the parks with 76-100% of tree cover (X2 = 13.120, p = 0.041). The
variations of use frequency in connection with the coverage by evergreen species
were very small, even if there were significant relationships between them and high
use frequency in the parks with 26-50% of evergreens. In addition, 28.7% of the
respondents declared that they would stay more than two hours when they visited the
green spaces with 51-75% of multi-layered vegetation. The percentage of such
respondents decreased to 13.7% and 14.4% in the parks with 76-100% and 26-50% of
multi-layered vegetation, respectively (X2 = 13.248, p = 0.039).

93
Table 6.6 Use frequency by variables of park characteristics

Variable Category Average score of F or t-statistic p-value


use frequency
Park area (ha) 30-60 5.04 F = 0.146 0.932
10-30 4.96
5-10 5.01
0-5 5.02
Park age >1990s 5.05 t = 1.755 0.186
<1990s 4.93
Location Within the third 5.07 t = 1.797 0.181
ring road
Within the second 4.95
ring road
Vegetation and 51-75% 4.99 t = 0.168 0.682
water 76-100% 5.03
Multi-layered 26-50% 5.07 F = 2.542 0.080
vegetation 51-75% 4.85
76-100% 5.07
Tree cover 26-50% 5.11 F = 4.575 0.011*
51-75% 4.99
76-100% 4.77
Mature tree 0-25% 5.05 F = 2.846 0.059
cover 26-50% 5.12
51-75% 4.78
Evergreen 0-25% 4.84 F = 4.656 0.003*
cover 26-50% 5.18
51-75% 4.92
76-100% 4.73
Average score: daily = 6, several times per week = 5, weekly = 4, 1-3 times per month
= 3, monthly = 2, several times per year = 1, less = 0;
Significant at 0.05 level (*p < 0.05)

94
Table 6.7 Duration per visit by variables of park characteristics

Variable Category Average score of stay F or t-statistic p-value


time per visit
Park area (ha) 30-60 1.94 F = 1.96 0.119
10-30 2.00
5-10 1.79
0-5 1.77
Park age >1990s 1.87 t = 1.634 0.202
<1990s 1.96
Location Within the third 1.87 t = 0.563 0.453
ring road
Within the second 1.93
ring road
Vegetation and 51-75% 1.89 t = 0.045 0.832
water 76-100% 1.90
Multi-layered 26-50% 1.86 F = 3.975 0.019*
vegetation 51-75% 2.06
76-100% 1.82
Tree cover 26-50% 1.89 F = 0.622 0.537
51-75% 1.95
76-100% 1.84
Mature tree 0-25% 1.87 F = 2.206 0.111
cover 26-50% 2.10
51-75% 1.88
Evergreen 0-25% 1.95 F = 0.611 0.608
cover 26-50% 1.91
51-75% 1.84
76-100% 2.00
Average score: less than one hour = 1, one to two hours = 2, half a day = 3, nearly a
whole day = 4;
Significant at 0.05 level (*p < 0.05)

95
Citizens’ Recreational Activities in Green Spaces
6.3.1 Citizens’ Recreational Activities in Green Spaces and
Socioeconomic Differences
In order to investigate the respondents’ recreational activities in green spaces, the
degree of activities was measured by a simple ranking exercise (Table 6.8 and Figure
6.1). The responses to each item in the survey were ranked with scores of very
important (2), important (1), neither important nor unimportant (0), and not important
(-1). In general, the average scores of recreational activities ranged from 0.07 to 1.45
(Table 6.8 and Figure 6.1). The top three most important activities were all passive in
nature, and included enjoying the scenery (1.46), spending time near large greenery
(1.38) and contact with nature (1.30). To enjoy the weather and get fresh air (1.22), to
do sports (1.18), to relax (1.16), to spend time in quiet and peaceful areas (0.9) and to
enjoy the cool in summer (0.8) were ranked as less important. The average scores of
the last two activities dropped dramaticly, to 0.16 (to walk a dog) and 0.07 (to use
recreational facilites).

The proportion of the “very important” answer for each item was less than 50%, with
the highest for beautiful views (47.4%). Only 7.0% of respondents considered the use
of recreational facilities, and 6.8% walking a dog as very important. However, ten
activities were rated as important by over half of respondents - all but the two least
important ones (see Table 6.8).

The importance of recreational activities in green spaces was measured by summing


the scores of the 12 items and comparing them with socioeconomic variables using F
or t-tests (Table 6.9). In general, males, older citizens, retirees and respondents who
have children considered all the recreational activities more important than other use
groups. Age and vocation showed a significant relationship (Table 6.9). For example,
the use of recreational facilities accounted for 80.6% of the > 65 age group, but only
32.3%, 37.4% and 44.9% in the 15-25, 26-45 and 46-65 age groups, respectively (x2 =
144.476, p < 0.001). In addition, the middle-aged respondents (25.2%) paid more
attention to quiet and peaceful spaces than their younger counterparts (16.2%) (x 2 =
27.177, p = 0.001).

96
Table 6.8 Main reasons to use green spaces for recreation (percentage)

Reason Very Important Not Neither Average Standard


important important important nor score deviation
unimportant
To enjoy the 47.4 51.4 0.5 0.7 1.46 0.76
beautiful views
To enjoy large 43.2 53.0 1.6 2.1 1.38 0.77
greenery
To get into 39.3 54.4 2.6 3.7 1.30 0.74
contact with
nature
To enjoy the 32.2 61.7 3.7 2.3 1.22 0.72
weather and get
fresh air
To do sports 33.6 56.3 5.8 4.2 1.18 0.75
To relax / to 33.2 56.3 6.5 4.0 1.16 0.75
reduce the stress
To have fun with 32.0 55.6 6.8 5.6 1.13 0.76
friends or play
with family
To learn about 29.0 58.6 5.6 6.8 1.11 0.75
plants and
animals
To spend time in 26.6 50.2 13.6 9.6 0.90 0.82
quiet and
peaceful areas
To enjoy the 21.0 52.8 15.4 10.7 0.80 0.81
cooler air in
summer
To walk a dog 6.8 37.9 35.0 20.3 0.16 0.86
To use the 7.0 35.3 42.3 15.4 0.07 0.83
recreational
facilities (e.g.
fitness
equipment)
Average score: very important = 2, important = 1, neither important nor unimportant =
0, not important = -1

97
To use the recreational facilities
To walk a dog
To enjoy the cooler air in summer
To spend time in quiet and peaceful areas
To learn about plants and animals
To have fun with friends or play with family
To relax / to reduce the stress
To do sports
To enjoy the weather and get fresh air
To get into contact with nature
To enjoy large greenery
To enjoy the beautiful views

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Figure 6.1 Mean rankings of main recreational activities in green spaces


Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1

Table 6.9 Activities in green spaces by respondents’ socioeconomic variables

Variable Category Importance of aims F or t-statistic P-value


by average score
Gender Male 12.01 t = 0.9759 0.3297
Female 11.72
Age 15-25 11.12 F = 3.28 0.0209*
26-45 11.69
46-65 12.24
>65 12.81
Children No 11.76 t = -0.5724 0.5673
Yes 11.93
Vocation Student 11.00 F = 2.89 0.0222*
Employed 11.71
Self-employed 12.52
Unemployed 11.76
Retiree 12.88
Average score = sum of the score of 12 activities;
Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1;
Significance at 0.05 level (*p < 0.05)
98
6.3.2 Citizens’ Recreational Activities in Green Spaces and
Differences of Park Variables
The associations of recreational activities and park characteristics were also analyzed
using F or T-tests. Park area had a significant relationship with respondents’
recreational activities, including the preferences to do sports, to relax, to learn about
plants and animals and to walk a dog (see Table 6.10). The percentages of
respondents’ preferences for the above activities in the larger parks were larger than in
the smaller parks. For example, in the parks with areas of 30-60 ha, 94.7% of
respondents considered that doing sports was a very important or important
recreational activity. Accordingly, this percentage decreased to 78.2% in the parks
with areas of 10-30 ha (x2 = 28.416, p = 0.001). Similarly, more respondents (78.5%)
from the parks with areas of 30-60 ha considered that enjoying cool air in summer
was an important aspect of green spaces, while only 46.5% of respondents from
smaller parks (0 -5 ha) (x2 = 32.322, p <0.001) considered the same. Additionally,
23.1% of respondents preferred to walk a dog in the bigger parks with 30-60 ha of
area. This percentage was obviously larger than in the smaller parks with 0-5 ha of
area (4.2%) (x2 = 17.435, p = 0.042).

Park age also had a significant relationship with respondents’ activities in parks, such
as to enjoy peace and quiet, to enjoy larger greenery, to be in contact with nature, to
learn about plants and animals and to walk a dog (see Table 6.10). In the newer parks
built after the 1990s, 80.1% of the respondents deemed having peace and quiet as
important. By comparison, 70% held the same view in the older parks, which were
built prior to the 1990s (x2 = 11.779, p = 0.008). In the newer parks, 92.1% of the
respondents wanted to learn about plants and animals, compared to only 77.9% in
older parks (x2 = 21.284, p <0.001). In addition, more respondents preferred to walk a
dog in newer parks (49.3%) than in older parks (34.5%) (x2 = 11.982, p = 0.007), and
they considered contact with nature as more important in the newer parks (95.2%)
than in the older parks (90.4%). However, all the respondents considered that
enjoying large greenery is an important activity in the older parks, while the
percentage in the newer parks was 98.3%.

Moreover, the percentage of multi-layered vegetation and evergreen species had a


significant influence on respondents’ preferred recreational activities including the
preference to do sports, to relax, to have fun with family, to enjoy nice weather and
get fresh air, to enjoy cool air in the summer and to walk a dog (see Table 6.10). For
example, in the parks with 26-50% of multi-layered vegetation, 94% of the
99
respondents preferred to do sports. The percentage decreased to 82.4% in the parks
with 51-75% of multi-layered vegetation (x2 = 12.850, p = 0.045). 96.2% of
respondents preferred to relax in the parks with 26-50% of multi-layered vegetation,
compared to only 86.2% in the parks with 76-100% of multi-layered vegetation (x2 =
12.659, p = 0.049). In addition, 59.8% of the respondents preferred to walk a dog in
the parks with 26-50% of multi-layered vegetation. This percentage decreased to
38.3% and 37% in the parks with 76-100% and 51-75% of multi-layered vegetation,
respectively (x2 = 21.560, p = 0.001).

Table 6.10 Respondents’ recreational activities by variables of park


characteristics

Variable Category Average score of F or t-statistic p-value


recreational
activities
Park area (ha) 30-60 12.54 F = 10.195 <0.001*
10-30 11.03
5-10 11.78
0-5 10.26
Park age >1990s 12.2 t = 6.502 0.011*
<1990s 11.12
Location Within the third ring 12.16 t = 1.786 0.182
road
Within the second 11.48
ring road
Vegetation and 51-75% 12.09 t = 1.787 0.182
water 76-100% 11.72
Multi-layered 26-50% 12.60 F = 3.804 0.023*
vegetation 51-75% 11.31
76-100% 11.65
Tree cover 26-50% 11.94 F = 2.171 0.115
51-75% 12.33
76-100% 10.89
Mature tree 0-25% 11.97 F = 1.347 0.261
cover 26-50% 11.84
51-75% 11.31
Evergreen 0-25% 12.63 F=9.135 <0.001*
cover 26-50% 12.59
51-75% 10.78
76-100% 11.36
Average score: sum of the scores of 12 activities;

100
Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1;
Significance at 0.05 level (*p < 0.05)

With regard to the influence of evergreen species, it was found that more respondents
preferred to do sports, relax, and enjoy cool air in the summer and to walk a dog in the
parks with 26-50% of evergreens than in parks with different corresponding area
percentages (see Table 6.10). For example, 96.1% of respondents preferred to do
sports in the parks with 26-50% of evergreens. The percentage of active respondents
decreased to 91.4% and 78.1% in the parks with 0-25% and 76-100% of evergreens,
respectively (x2 = 18.943, p = 0.004). The respondents’ preferences for the other
recreational activities had very similar variation (Table 6.10).

Citizens’ Perceptions of Green Spaces


6.4.1 Citizen’ Perceptions of Green Space Types and
Socioeconomic Differences
When assessing the importance of different types of green spaces without comparing
them to each other, 50.2% of respondents first chose the green spaces near their home,
school or work place as very important (Table 6.11). 37.4% and 33.4% regarded big
suburban parks or parks outside the city areas as very important. Even when first
considering their accessibility, a large number of respondents (70.8%) did not avoid
big parks except for those that are the farthest.

The analysis of socioeconomic variables showed that older respondents had a


statistical association with the perceptions of green space types (Table 6.11). By
contrast, the responses of male and female park visitors were nearly the same. Self-
employed men rated all types of green spaces as more important than other groups
(Table 6.12). The 46-65, 26-45 and 15-25 age groups assessed all types of green
spaces as highly important. Nearby green spaces were considered important by all age
groups. However, the older age group deemphasized the importance of big city parks
and outskirt parks. The “important” option for big city parks was chosen by 64.4% of
the respondents > 65, compared to 93.2%, 89% and 79.4% of the 26-65, 26-45, and
15-25 age groups, respectively (x2 = 32.588, p < 0.01). In addition, 74.2% of the
respondents > 65 regarded outskirt parks as important, compared to 93.1% of the 46-
65, 92.8% of the 26-45 and 85.2% of the 15-25 group (x2 = 20.230, p < 0.05).
Therefore, unlike with other age groups, the elderly only considered the nearby green
spaces as important.

101
Table 6.11 Importance of different types of green spaces (percentage)

Green space Very Important Not Neither important


important important nor unimportant
Near my residence (or work 50.2 43.9 3.3 2.6
place, school)
Big city parks, even if farther 37.4 49.8 6.8 6.1
away
Parks outside the city 33.4 57.0 4.2 5.4

Table 6.12 Importance of different types of green spaces by socioeconomic


variables

Variable Category Important types by F or t-statistic p-value


average score
Gender Male 3.76 t = -0.3029 0.7621
Female 3.80
Age 15-25 3.34 F = 6.06 <0.0015*
26-45 3.87
46-65 3.98
>65 3.29
Children No 3.77 t = -0.0751 0.9402
Yes 3.78
Vocation Student 3.51 F = 1.3 0.2675
Employed 3.83
Self-employed 4.02
Unemployed 3.52
Retiree 3.71
Average score = sum of scores of the 3 types of green spaces;
Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1;
Significance at 0.05 level (*p < 0.05)

102
6.4.2 Citizen’ Attitudes Towards the Importance of
Ecosystem Services of Green Spaces and Socioeconomic
Differences
To inquire the visitors about the importance of ecosystem services provided by green
spaces, a simple ranking exercise was used as described above. A majority of the
respondents chose “very important” or “important”, with some services rated notably
higher (Table 6.13 and Figure 6.2). The average scores of 0.58-1.48 indicated general
importance of all 17 ecosystem services, with a majority (11) ranging from 0.6-1.2.
The top five services were shading (1.48), O2 release (1.43), places for recreation
(1.32), more contact with nature (1.22) and aesthetic improvement (1.09). By
comparison, water-soil conservation (0.85), neighbor-social interaction (0.83),
economic value (0.67), wind protection (0.58) and noise reduction (0.58) were ranked
the lowest.

The overall perceived importance was measured by summing the scores of the 17
ecosystem services and comparing them with socioeconomic variables using F or t-
tests (Table 6.14). The socioeconomic variables did not show a significant association
with the importance of any of the ecosystem services. In general, all respondents had
similar perceptions regardless of their personal socioeconomic attributes, except for
individual services.

Nevertheless, some diffrent perceptions were induced by socioeconomic variables.


For example, 22.9% of male respondents deemed species conservation as very
important, compared to 28.0% of females (x2 = 11.963, p = 0.008). Stratified by age
group, the assessment of water-soil conservation as “very important” accounted for
25.0% of the respondents from 15-25, in contrast with 15.9%, 14.3% and 12.9% of
the 26-45, 46-65 and >65 age groups, respectively (x2 = 19.163, p = 0.024). Vocation
was another factor that influenced the perception of water-soil conservation. 26.8% of
students chose “very important”, compared to 18.3% of the employed, 14.3% of
retirees, 8.7% of the self-employed, and 3.5% of the unemployed respondents (x2 =
59.306, p <0.001). Therefore, young students valued the importance of water-soil
conservation more than other groups. Considering neighbor-social interaction, more
respondents who have children (21.9%) tended to choose “very important” than those
without children (13.5%) (x2 = 11.647, p = 0.009).

103
Table 6.13 Perceptions of the importance of ecosystem services generated by
green spaces (percentage)

Ecosystem Service Very Important Not Neither Average Standard


important important important nor score deviation
unimportant
Shading 50.0 48.4 0.7 0.9 1.48 0.76
O2 release 44.6 54.0 0.2 1.2 1.43 0.75
Places for 40.4 53.5 2.6 3.5 1.32 0.78
recreational
activities
More contact with 35.0 56.3 4.4 4.2 1.22 0.78
nature
Aesthetic 29.0 59.1 7.7 4.2 1.09 0.78
improvement
Wildlife habitats 29.4 54.0 7.2 9.4 1.06 0.81
Species 25.5 59.4 5.6 9.6 1.05 0.79
conservation
CO2 sequestration 23.1 63.1 6.3 7.5 1.03 0.77
Increasing air 25.0 59.4 6.5 9.1 1.03 0.79
humidity
Cultural and 22.9 58.4 10.3 8.4 0.94 0.80
educational benefits
Air quality 21.8 57.0 11.2 10.1 0.89 0.81
improvement
Lower air 23.1 54.9 14.8 7.2 0.86 0.82
temperature
Water-soil 16.6 58.2 6.5 18.7 0.85 0.80
conservation
Neighbor-social 18.7 57.0 12.9 11.5 0.83 0.81
interaction
Economic value 14.7 54.2 18.0 13.1 0.67 0.83
(e.g. cultivation of
wood and fruits)
Wind protection 11.0 54.0 18.0 17.1 0.58 0.83
Noise reduction 12.4 53.5 19.9 14.3 0.58 0.83

Average score: very important = 2, important = 1, neither important nor unimportant


=0, not important = -1

104
Noise reduction
Wind protection
Economic value
Neighbor-social interaction
Water-soil conservation
Lower air temperature
Air quality improvement
Cultural and educational benefits
Increasing air humidity
CO2 sequestration
Species conservation
Wildlife habitats
Aesthetic improvement
More contact with nature
Places for recreational activities
O2 release
Shading
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6

Figure 6.2 Mean rankings of importance of ecosystem services


Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1

Table 6.14 Perceptions of the importance of ecosystem services by respondents’


socioeconomic variables

Variable Category Importance of ecosystem F-statistic or t- p-value


services by average score statistic
Gender Male 16.94 t = 0.3254 0.7450
Female 16.81
Age 15-25 16.81 F = 0.16 0.9246
26-45 16.96
46-65 16.9
>65 16.42
Children No 17.0 t = 0.3782 0.7054
Yes 16.82
Vocation Student 17.08 F = 0.66 0.6194
Employed 17.35
Self-employed 16.0
Unemployed 16.87
Retiree 17.14
Average score: sum of scores of the 17 ecosystem services;
105
Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1;
Significance at 0.05 level (*p < 0.05)

In addition, age and vocation also affected the respondents’ perceptions of neighbor-
social interaction and economic value. Young students more often disregarded the
importance of neighbor-social interaction, while retired people underestimated the
economic value of green spaces much more. Although considerable differences and
some uncertainty regarding the importance of individual ecosystem services was
present, the same perspectives were stated by the overwhelming majority of
respondents.

6.4.3 Citizen’ Perceptions of the Importance of Ecosystem


Services of Green Spaces and Differences of Park Variables
Park area showed a significant relationship with the respondents’ perceptions of the
importance of ecosystem services, as calculated using the F-test (Table 6.15). For
example, park area influenced the respondents’ perception of the importance of O2
release, CO2 sequestration, wind protection, lower air temperature, noise reduction
and aesthetic improvement. Respondents regarded these ecosystem servies as more
important in the parks with areas of 10-30 ha and 5-10 ha than in the other parks. For
example, 97.5% of respondents thought that aesthetics was important in the parks
with areas of 10-30 ha, while this percentage decreased to 76.8% in the parks with
areas of 0-5 ha (x2 = 18.540, p = 0.029).

Except for the cover of vegetation and water, the other four types of vegetation
structure had significant relationships with respondents’ perceptions of the importance
of ecosystem services. The percentage of multi-layered vegetation influenced the
respondents’ perception of CO2 sequestration, wind protection, increased humidity,
lower air temperature, and air quality improvement. These ecosystem services were
considered as increasingly important with the increasing percentage of multi-layered
vegetation cover. For example, in the parks with 26-50% of multi-layered vegetation,
65.2% of the respondents considered that lower air temperature is important,
compared to 85.2% and 83.0% in parks with 51-75% and 76-100% of multi-layered
vegetation, respectively (x2 = 22.987, p = 0.001). Similarly, 86.2% of the respondents
in parks with 76-100% of multi-layered vegetation considered that air quality
improvement is important. By contrast, only 75.8% and 69.4% of respondents in
parks with 26-50% and 51-75% of multi-layered vegetation positively perceived the
importance of air quality improvement, respectively (x2 = 16.207, p = 0.013).
106
Table 6.15 Respondents’ perceptions of ecosystem services by variables of park
characteristics

Variable Category Importance of ecosystem F or t-statistic p-value


services by average score
Park area (ha) 30-60 16.20 F = 3.190 0.024*
10-30 16.88
5-10 18.44
0-5 16.60
Park age >1990s 17.39 t = 0.314 0.576
<1990s 15.79
Location Within the third 17.16 t = 0.157 0.692
ring road
Within the second 16.53
ring road
Vegetation and 51-75% 17.21 t = 1.333 0.249
water 76-100% 16.68
Multi-layered 26-50% 15.62 F=15.191 <0.001*
vegetation 51-75% 15.67
76-100% 18.46
Tree cover 26-50% 16.96 F=3.585 0.029*
51-75% 16.43
76-100% 17.33
Mature tree 0-25% 17.21 F =11.026 <0.001*
cover 26-50% 14.29
51-75% 17.31
Evergreen 0-25% 13.95 F = 3.058 0.028*
cover 26-50% 16.94
51-75% 17.66
76-100% 16.73
Average score: sum of scores of the 17 ecosystem services;
Very important = 2, important = 1, neither important nor unimportant = 0, not
important = -1;
Significance at 0.05 level (*p < 0.05)

Tree cover and mature tree cover both obviously influenced the respondents’
perceptions of the importance of ecosystem services such as CO2 sequestration,
shading, noise reduction and aesthetic improvement. Ecosystem services were
considered more important in parks with a higher cover of trees (76-100%) and
mature trees (51-75%). For example, aesthetic improvement was considered more
important in the parks with 76-100% than in the parks with 26-50% of trees (100%

107
versus 84.1%) (x2 = 17.463, p = 0.008). All respondents valued the importance of
shading in the parks with 26-50% and 76-100% of trees, compared to 94.3% in parks
with 51-75% of trees (x2 = 23.967, p = 0.001). Similarly, all respondents regarded
shading as important in the parks with 51-75% of mature tree cover. By comparison,
this percentage decreased to 92.1% in the parks with 26-50% of mature trees (x2 =
15.817, p = 0.015).

Similarly, the percentage of evergreen species within the investigated parks also
influenced the respondents’ perceptions of the importance of ecosystem servcies such
as CO2 sequestration, wind protection, lower air temperature, air quality improvement
and aesthetic improvement. More respondents considered these services as important
in the parks with 51-75% of evergreens.

Citizens’ Assessments of Green Spaces


6.5.1 Citizens’ Assessments of Green Spaces and
Socioeconomic Differences
The degree of satisfaction with the investigated 22 parks in Xi’an city was ranked by
a simple exercise. The responses to each item were ranked with scores of excellent
(3), good (2), fair (1), unclear (0) and poor (-1). For the entirety of green spaces at city
level, the first-ranked attribute was management of the green spaces (1.56), followed
by the amount of green spaces (1.54), species conservation (1.21), recreational
opportunities (1.21) and ecological functions (0.86) (Table 6.16). For all items, more
than one fifth, and sometimes even half of respondents chose the “good” option, but
less than 11.2% chose “excellent”. Considering the proportions of the five options,
“good” and “fair” were much more present than “excellent”.

The assessment of the entirety of green spaces in Xi’an was measured by comparing
the average assessment scores with socioeconomic variables using F or t-tests (Table
6.17). In general, males, older, retired, and respondents who have no children assessed
the green spaces more positively. Age showed a significant relationship. For example,
32.3% of the respondents > 65 considered the amount of green spaces excellent, but
only 9.52% and 8.79% of the 46-65 and 26-45 groups held the same positive
assessment, falling down to zero percent for the 15-25 group, (x2 = 52.725, p < 0.05).
However, no > 65 respondents assessed the ecological functions of green spaces as
“excellent” (x2 = 32.103, p < 0.01). This reveals that the elderly hold significantly
different perceptions than the other groups.

108
Table 6.16 Assessments of the condition of the entirety of green spaces in Xi’an
city (percentage)

Condition of green Excellent Good Fair Unclear Poor Average Standard


spaces score deviation
Management of green 9.3 47.9 37.2 0.5 5.1 1.56 0.81
spaces
Amount of green areas 9.3 47.7 36.7 0.2 6.1 1.54 0.81
Species conservation 3.7 38.8 44.4 0.9 12.2 1.21 0.82
(vegetation and animals)
Recreational 11.2 35.5 33.6 2.6 17.1 1.21 0.87
opportunities
Ecological functions 1.6 26.2 48.8 3.3 20.1 0.86 0.83
(e.g. decreased
temperature, lower air
pollution)
Very important = 2, important = 1, unclear = 0, not important = -1

Table 6.17 Assessment of the condition of the entirety of green spaces in Xi‘an by
respondents’ socioeconomic variables (percentage)

Variable Category Average score over the F or t-statistic p-value


entirety of green spaces
Gender Male 6.47 t = 0.5979 0.5502
Female 6.29
Age 15-25 5.66 F = 10.52 <0.0010*
26-45 6.5
46-65 6.02
>65 8.97
Children No 6.54 t = 0.8630 0.3886
Yes 6.28
Vocation Student 6.20 F = 1.73 0.1419
Employed 6.26
Self-employed 6.04
Unemployed 6.59
Retiree 7.39
Average score: sum of scores of the 5 aspects;
Very important = 2, important = 1, unclear = 0, not important = -1;
Significant at 0.05 level (*p < 0.05)

109
For the green spaces used daily by the citizens, the average score of assessment
ranged from 0.15 to 1.98 (Table 6.18 and Figure 6.3). Half of the ratings were in the
range of 1.0-1.5. Vegetation coverage received the highest assessment (1.98),
followed by microclimate (1.84) and overall performance (1.63). The lowest was air
quality (0.74), water area (0.73) and water quality (0.15).

Similar with the entirety of green spaces, the amount of green areas and management
of the green spaces used daily by the study participants also received higher
assessments than the recreational opportunities, species conservation, and ecological
functions. Only 6.8% of respondents had the impression that their daily-used parks
performed excellently as a whole. When viewing each specific item, the highest
percentage of “excellent” assessments was of vegetation coverage, which accounted
for no more than 23.8%. By contrast, “good” assessments occupied the majority. Over
half of the respondents ranked vegetation coverage (55.4%), cleanliness (55.4%) and
microclimate (52.8%) as good. Especially, the overall performance of the parks
received good assessments in 55.8% of the cases. On the other end of the spectrum,
water quality received poor assessments in 46.7% of the cases. Dissatisfaction with
the area of water bodies also accounted for 34.6%. In addition, sporting facilities were
viewed as being in poor condition by 27.1% of respondents. Air quality and the
number of wild animals were considered poor by 24.1% and 22.2% of respondents,
respectively.

The assessments of all the 22 parks were analyzed by summing up the average scores
of 20 assessment aspects and comparing them with socioeconomic variables using F
or t-tests (Table 6.19). There was no significant difference between the assessments
based on gender / age group / vocation. However, family status showed statistical
significance. For example, 87.11% of the respondents without children positively
assessed the vegetation cover of green spaces, compared to only 74.34% of those with
children (x2 = 15.166, p = 0.004). Positive assessments for playgrounds from the
respondents with children were more prominent than from those without children
(46.01% and 37.74%) (x2 = 22.410, p <0.001). In addition, only 18.87% of
respondents who have children positively assessed air quality, compared to 28.22% of
their childless peers (x2 = 12.295, p = 0.002).

110
Table 6.18 Assessments of green spaces used daily by the citizens (percentage)

Excellent Good Fair Unclear Poor Average Standard


score deviation
Management of the green space
Cleanliness 9.4 55.4 29.0 0.5 5.8 1.62 0.80
Safety 3.0 38.1 48.1 1.4 9.4 1.24 0.81
Amount of green areas
Vegetation cover ratio 23.8 55.4 18.2 0.5 2.1 1.98 0.80
Species conservation
The number of vegetation types 9.4 43.0 35.1 3.5 9.1 1.40 0.84
The number of wild animal species
(e.g. birds or insects) 1.9 22.4 48.6 4.9 22.2 0.77 0.83

Recreational opportunities
Scenic beauty 10.3 39.0 41.1 1.6 7.9 1.42 0.83
The number of sporting facilities 4.0 23.6 44.4 0.9 27.1 0.77 0.84
The location of sporting facilities 4.4 35.7 48.6 1.2 10.1 1.23 0.81

The maintenance of sporting 2.6 35.5 46.7 1.4 13.8 1.12 0.82
facilities
The number of resting facilities (e.g. 0.5 28.7 52.3 1.6 16.8 0.94 0.81
tables or benches)
The location of resting facilities (e.g. 1.4 37.2 52.1 1.4 7.9 1.23 0.79
tables or benches)
The maintenance of resting facilities 2.6 42.1 47.9 0.2 7.2 1.33 0.79
(e.g. tables or benches)

The area of playgrounds 3.5 37.4 43.9 0.9 14.3 1.15 0.82
The location of playgrounds 3.0 39.0 47.0 0.5 10.5 1.24 0.81
The maintenance of playgrounds 2.6 47.0 44.4 0.7 5.4 1.41 0.79
Ecological functions (e.g. decrease
of temperature, lower air pollution)
Microclimate (e.g. cooling, tree 18.0 52.8 26.6 0 2.6 1.84 0.81
shade, humidity and wind)

Air quality 0.5 22.0 52.6 0.9 24.1 0.74 0.81


The area of water bodies 8.2 26.4 30.1 0.7 34.6 0.73 0.87
Water quality (if water bodies are 0.5 8.2 43.9 0.7 46.7 0.15 0.79
present)
Overall assessment of the site 6.8 55.8 34.4 0 3.0 1.63 0.78
Excellent = 3, good = 2, fair = 1, unclear = 0, poor = -1

111
Water quality (if water bodies are present)
The area of water bodies
Air quality
The number of sporting facilities
The number of wild animal species
The number of resting facilities
The maintenance of sporting facilities
The area of playgrounds
The location of resting facilities
The location of sporting facilities
The location of playgrounds
Safety
The maintenance of resting facilities
The number of vegetation types
The maintenance of playgrounds
Scenic beauty
Cleanliness
Overall assessment of the site
Microclimate
Vegetation cover ratio
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

Figure 6.3 Mean rankings of the assessments


Excellent = 3, good = 2, fair = 1, unclear = 0, poor = -1

Table 6.19 Assessments of green spaces used daily by the respondents based on
socioeconomic variables

Variable Category Average score of daily F or t-statistic p-value


used green spaces
Gender Male 24.11 t = 0.5656 0.5720
Female 23.73
Age 15-25 24.37 F = 0.86 0.4641
26-45 24.09
46-65 23.26
>65 25.10
Children No 24.96 t = 2.4152 0.0161*
Yes 23.28
Vocation Student 26.27 F = 1.97 0.0982
Employed 23.36
Self-employed 23.06
Unemployed 24.24
Retiree 25.08
112
Average score: sum of scores of the 20 aspects;
Excellent = 3, good = 2, fair = 1, unclear = 0, poor = -1;
Significant at 0.05 level (*p < 0.05)

6.5.2 Citizens’ Assessments of Green Spaces and Differences


of Park Variables
Respondents’ assessments were significantly influenced by park characteristics, as
calculated using the F or t-tests (see Table 6.20). Considering the park area, it was
shown that more positive assessments were given for the parks with areas of 10-30 ha
and 0-5 ha, followed by the parks of 30-60 ha and 5-10 ha (see Table 6.20). For
example, 88.5% of the respondents were satisfied with the vegetation cover in the
parks having 10-30 ha of area, followed by 86% in the parks with 0-5 ha. This
percentage decreased to 83.7% and 72.2% in the parks with areas of 5-10 ha and 30-
60 ha, respectively (x2 = 25.451, p = 0.013). Similarly, in the parks with areas of 10-
30 ha, 60.3% of the respondents felt satisfaction with the scenery. By contrast, in
parks with areas of 5-10 ha, the satisfaction dropped to 29.6% (x2 = 36.317, p
<0.001).

Park age also significantly influenced many of the respondents’ assessments of


features such as vegetation cover, microclimate, the number of vegetation types,
scenic beauty, the number of resting facilities, the location of sporting facilities,
resting facilities and playgrounds, and the maintenance of sporting facilities and
playgrounds. Usually, the percent of positive assessments increased with the
increasing age of parks. For example, 72.1% of respondents deemed that the number
of types of vegetation was sufficient in the older parks, which were built prior to the
1990s. However, only 43.2% of the respondents held the same position in the newer
parks (x2 = 41.083, p <0.001). Similarly, respondents’ assessments of the above-
referred items were very similar to the assessments of vegetation types.

The parks’ location also significantly influenced the respondents’ assessments of


green spaces, including features such as the vegetation cover, microclimate, the
number of vegetation types and animal species, scenic beauty and the location and
maintenance of sports facilities. In most cases, the respondents much more positively
assessed the related items in parks that were located within the second ring road than
in the parks that were located within the third ring road. For example, 86.4% of the
respondents positively assessed the vegetation cover of the parks located inside the
second ring road, compared to 73.4% of those located inside the third ring road (x 2 =
22.383, p <0.001).
113
Vegetation structure types including multi-layered vegetation, trees, mature trees and
evergreen species had significant relationships with respondents’ assessments of green
spaces. Usually, the more positive assessments for urban parks, such as the vegetation
cover, microclimate, the vegetation types, scenic beauty, safety, the number of sports
facilities, and the areas and location of playgrounds were correlated with 51-75% of
multi-layered vegetation. For example, in the parks with 51-75% of multi-layered
vegetation, 95.4% of respondents were satisfied with the cover of vegetation. By
comparison, the satisfaction decreased to 81.9% and 62.1% in the parks with 76-
100% and 26-50% of multi-layered vegetation, respectively (x2 = 65.424, p <0.001).
Similarly, in the parks with 51-75% of multi-layered vegetation, 74.1% of respondents
highly assessed the scenery. However, such assessments dropped to 34.9% in the
parks with 26-50% of multi-layered vegetation (x2 = 72.816, p <0.001).

Table 6.20 Respondents’ assessments by variables of park characteristics

Variable Category Average score of F or t-statistic p-value


respondents’
assessments
Park area (ha) 30-60 23.81 F = 6.886 <0.001*
10-30 26.45
5-10 21.88
0-5 24.58
Park age >1990s 22.89 t = 20.472 <0.001*
<1990s 26.14
Location Within the Third Ring Road 22.85 t = 11.660 0.001*
Within the Second Ring Road 25.26
Vegetation and 51-75% 24.61 t = 1.656 0.199
water 76-100% 23.51
Multi-layered 26-50% 22.14 F = 30.621 <0.001*
vegetation 51-75% 28.09
76-100% 22.78
Tree cover 26-50% 22.66 F = 9.403 <0.001*
51-75% 24.64
76-100% 26.46
Mature tree 0-25% 23.36 F = 3.590 0.028*
cover 26-50% 25.92
51-75% 25.11
Evergreen 0-25% 19.93 F = 22.188 <0.001*
cover 26-50% 22.33
51-75% 27.14
76-100% 23.94

114
Average score: sum of scores of the 20 aspects;
Excellent = 3, good = 2, fair = 1, unclear = 0, poor = -1;
Significant at 0.05 level (*p < 0.05)

It was also obvious that parks with 76-100% of trees and 26-50% of mature trees
received more positive assessments regarding the above-referred features. For
instance, in the parks with 76-100% of trees, 92.4% of the respondents were satisfied
with the vegetation cover, compared with only 71.8% of satisfaction in the parks with
26-50% of trees (x2 = 28.162, p <0.001). Similarly, 93.8% of the respondents were
satisfied with the vegetation cover in the parks with 51-75% of mature trees. By
contrast, the percentage of satisfaction dropped to 74.4% in the parks with 0-25% of
mature trees (x2 = 20.991, p = 0.007). However, the highest assessment of scenery
was 74.5% in the parks with 26-50% of mature trees, and the lowest was 35.9% in the
parks with 76-100% of mature trees (x2 = 39.718, p <0.001).

Furthermore, evergreen species influenced the respondents’ assessments of all the


above-mentioned features in urban parks. The parks with a medium percentage (51-
75%) of evergreens usually received more positive assessments than the other parks.
For example, 65.1% of positive assessments of the scenery were given for parks with
51-75% of evergreens, while only 30.2% and 21.2% of positive assessments were
given for parks with 0-25% and 76-100% of evergreens, respectively (x2 = 69.771, p
<0.001).

Citizens’ Demands for Green Spaces


6.6.1 Citizens’ Demands for Green Spaces and
Socioeconomic Differences
Rank scores were given according to the assessments: strongly need (2), need (1),
neither need nor not need (0), and do not need (-1) in order to investigate the demand
for specific features in green spaces. The top demands for urban green spaces with
average scores ≥ 1 were more shade (1.19), more opportunities to watch nature (1.14),
more flowers (1.06) and better management (1.02) (Table 6.21 and Figure 6.4). More
facilities for sports and exercise (0.92), more play areas for children (0.89), more
resting facilities (e.g. benches/tables) (0.88), more quiet places (0.69), more water
bodies (0.68), as well as more convenient and safer transportation (0.64), were ranked
in the class of less important demands. Very few respondents felt they needed more
cafes or restaurants (0.37) and quiet places (0.16).

115
Specifically, over half of the respondents declared that they strongly needed more
shade and more opportunities to watch nature (Table 6.21). Additionally, over half of
the respondents expressed moderate needs for more flowers (55.6%), more play areas
for children (56.7%), more sporting facilities (56.6%), more resting facilities (54.2%)
and better management (59.8%) (Table 6.21).

Table 6.21 Respondents’ demands for specific features in urban green spaces
(percentage)

Ecosystem service Strongly Need Do not Do not Average Standard


need need care score deviation
More shade 50.5 22.0 4.4 23.1 1.19 0.84
More wind-protected sites 23.4 2.6 33.4 40.7 0.16 0.86
More water bodies 36.2 12.9 17.5 33.4 0.68 0.90
More opportunities to watch 50.5 20.8 7.7 21.0
nature (birds, plants, etc.) 1.14 0.84

More flowers 48.1 17.5 8.2 26.2 1.06 0.85


More quiet places 40.0 8.2 19.2 32.7 0.69 0.88
More facilities for sports and 46.3 10.3 10.5 32.9 0.92 0.85
exercise
More play areas for children 45.3 11.4 13.3 29.9 0.89 0.86
More resting facilities (e.g. 43.9 10.3 9.8 36.0 0.88 0.85
benches / tables)
More cafes / restaurants 29.7 3.7 26.2 40.4 0.37 0.87
More convenient and safer 37.1 5.6 15.7 41.6
transportation to the green 0.64 0.85
spaces
Better management (e.g. 49.5 10.3 7.5 32.7
better maintenance of
plantings, facilities / less 1.02 0.83
litter)

Strongly need = 2, need = 1, do not care = 0, do not need = -1

116
More wind-protected sites

More cafes / restaurants


More convenient and safer transportation to the
green spaces
More water bodies

More quiet places

More resting facilities

More play areas for children

More facilities for sports and exercise

Better management

More flowers

More opportunities to watch nature

More shade

0 0.2 0.4 0.6 0.8 1 1.2

Figure 6.4 Mean rankings of the survey respondents’ demands for specific
features in urban green spaces
Strongly need = 2, need = 1, do not care = 0, do not need = -1

The association between demands for specific features in green spaces and
socioeconomic variables was measured by F or t-tests, whereby the 12 aspects were
summed using their average scores (Table 6.22). In general, young, male students had
stronger demands for the referred twelve features in green spaces. Moreover, age and
vocation showed significant statistical associations with demands. For example, more
quiet places were strongly needed by 32.3% of the > 65 respondents, but only 16.2%,
6.8% and 6.6% of the 15-25, 26-45 and 46-65 age groups, respectively (x2 = 43.914, p
<0.001). In addition, the middle-aged (22.0% and 32.0%) and older respondents
(29.0%) wanted more children’s play areas than did the young (8.8%) (x 2 = 39.419, p
<0.001). Notably, with increased age, respondents’ strong need for cafes and
restaurants decreased from 23.5% (15-25) to 0% (> 65) (x2 = 116.462, p <0.001). A
larger percentage of employed respondents (14.83%) wanted cafes and restaurants
than those belonging to the other groups (x2 = 59.466, p <0.001). The largest
proportion (22.5%) of strong need for more quiet spaces came from retired
respondents (x2 = 26.689, p = 0.009). At the same time, strong need for resting
facilities was mainly expressed by retired (22.45%) and unemployed respondents

117
(13.79%) (x2 = 26.738, p = 0.008). In addition, students expressed a much stronger
need for sporting facilities (56.1%) than other vocations (x2 = 79.449, p <0.001).
Perhaps unsurprisingly, they expressed the least need for children’s play areas (7.3%)
(x2 = 25.951, p = 0.011). The results revealed that the demands of young students
were different from those of the other respondents.

Table 6.22 Respondents’ demands by socioeconomic variables

Variable Category Average score of daily- F or t-statistic p-value


used green spaces
Gender Male 10.44 t = 0.3824 0.7023
Female 10.28
Age 15-25 11.38 F = 4.85 0.0025*
26-45 10.81
46-65 9.71
>65 8.52
Children No 10.41 t = 0.1964 0.8444
Yes 10.32
Vocation Self-employed 9.61 F = 2.82 0.0248*
Employed 10.53
Unemployed 9.34
Retiree 9.39
Student 11.98
Average score: sum of scores of the 12 demand categories;
Strongly need = 2, need = 1, do not care = 0, do not need = -1;
Significant at 0.05 level (*p < 0.05)

6.6.2 Citizens’ Demands for Green Spaces and Differences of


Park Variables
The relationships between citizens’ demands and park characteristics were also
analyzed using F or t-tests (Table 6.23). It was found that park age and location
significantly influenced the respondents’ demands for specific features in green
spaces, in addition to the cover of multi-layered vegetation, trees and evergreen
species.

In the older parks located within the second ring road, the respondents voiced more
demands for ecosystem services provided by green spaces. 76.2% of the respondents
thought that the older parks needed to be managed better. The percentage of
respondents who raised the same demands decreased 10% in the newer parks (x2 =
9.814, p = 0.020). While in the newer parks which were built after the 1990s, 76.4%
118
of the respondents expressed the need for more shade, this percentage was 64% in the
older parks (x2 = 19.299, p <0.001). In addition, the respondents expressed more need
for safe spaces in the parks located within the second ring road than in those inside the
third ring road (80.1% versus 62.8%) (x2 = 8.554, p = 0.036).

Table 6.23 Respondents’ demands by variables of the parks’ characteristics

Variable Category Average score of F or t-statistic p-value


respondents’ demands
Park area (ha) 30-60 7.59 F = 0.191 0.663
10-30 6.51
5-10 9.96
0-5 7.02
Park age >1990s 7.71 t = 4.890 0.028*
<1990s 8.24
Location Within the third ring 7.83 t = 6.420 0.012*
road
Within the second 7.94
ring road
Vegetation and 51-75% 7.31 t = 0.419 0.518
water 76-100% 8.22
Multi-layered 26-50% 8.56 F = 4.432 0.012*
vegetation 51-75% 7.04
76-100% 7.88
Tree cover 26-50% 8.03 F = 8.655 <0.001*
51-75% 6.70
76-100% 9.28
Mature tree 0-25% 7.50 F = 1.029 0.358
cover 26-50% 8.18
51-75% 9.50
Evergreen 0-25% 6.47 F = 4.666 0.003*
cover 26-50% 8.88
51-75% 6.80
76-100% 8.61
Average score: sum of scores of the 12 demand categories;
Strongly need = 2, need = 1, do not care = 0, do not need = -1;
Significant at 0.05 level (*p < 0.05)

Moreover, respondents from the parks with a lower cover of multi-layered vegetation
and evergreens, and higher cover of trees, expressed different demands for the
referred ecosystem services. For example, in the parks with 26-50% of multi-layered
119
vegetation, 75.7% of the respondents wished there were more shade, in comparison to
71.8% and 69.4% of respondents from the parks with 76-100% and 51-75% of multi-
layered vegetation, respectively (x2 = 15.439, p = 0.017). Evergreen cover also
influenced the respondents’ need for shading. 73.8% of the stronger need came from
the parks with 0-25% of evergreens. The need decreased to 64.7% in the parks with
26-50% of evergreens (x2 = 20.465, p = 0.015).

The cover of trees influenced the respondents’ needs for shading, wind-protected
sites, quiet spaces and park management. For example, 73.5% of the respondents from
the parks with 76-100% of trees expressed a stronger wish for more flowers. This
preference decreased to 68.2% and 55.8% in the parks with 26-50% and 51-75% of
trees, respectively (x2 = 18.696, p = 0.005). Respondents from the parks with 0-25%
of mature trees wanted more shading (73.8%) than those from parks with 26-50% of
mature trees (64.7%) (x2 = 27.249, p <0.001).

Discussion
6.7.1 Discussion of the Respondents’ Green Space Use
Behaviors
6.7.1.1 Comparison of Green Space Use Behaviors with Other

Studies
In this study, public parks and residential green spaces were the most often visited
types of green spaces in Xi’an (Table 6.2). Similarly, public urban parks were the
preferred types of outdoor places in a study of Santa Cruz, Bolivia (Wendel et al.
2012). In Guangzhou, the majority of citizens preferred to use the public urban green
spaces and newly developed urban green spaces (Shan 2009). Public parks were the
most preferred green spaces in different social contexts, probably due to their multi-
functional provisions, better maintenance and more amenities.

Respondents in Xi’an were very active in visiting urban green spaces. The majority of
respondents visited urban green spaces at least once a week (93.7%) but with a shorter
stay - less than one to two hours (82.2%) (Table 6.3). Similarly, a study in Denmark
showed that 92% of residents visited a green space at least once a week (Schipperijn
et al. 2010). Some other studies also found that citizens visited green spaces more
frequently but spent a shorter time on each visit. For example, an intercontinental
study conducted in Germany, Chile and Spain found that although residents visitied
green spaces frequently, only around half of them stayed more than one hour (Priego
120
et al. 2008). By contrast, the frequency of visits to green spaces was very low in
Karachi, and more than half of the respondents stated that they rarely or never visit
green spaces in the city (Qureshi et al. 2010). However, although the frequency of use
as a whole was low, the duration of visits was fairly long, often involving a stay of
multiple hours (Qureshi et al. 2010).

6.7.1.2 The Influence of Socioeconomic Variables on

Respondents’ Use Behaviors


The differences of use habits could be partly due to socioeconomic differencs in
attitudes towards green spaces (Oguz 2000). The frequency of visits to urban green
spaces of Xi’an was significantly affected by respondents’ age and family status.
Older respondents and those who lived with children were more likely to use green
spaces frequently. Pincetl et al. (2003) also found that respondents with children were
considerably more likely to visit parks. A possible reason is that green spaces may be
a suitable place to play with the active children. Many studies have indicated that park
use was negatively correlated with age, with older residents reporting less use (Payne
et al. 2002; Pincetl 2003; Pincetl and Gearin 2005; Rossi et al. 2016). However, this
study and the study of Shan (2009) showed a positive relationship between age and
urban green space visitation.

The stay time in green spaces per visit was significantly afftected by age and
vocation. Older respondents and retirees stayed markedly longer than the other
groups. In Chinese cities, the retirement age is 50-55 for females and 55-60 for males,
which means retirees are also older people. The findings that aged respondents visited
green spaces more frequently and stayed longer per visit may be because they have
more spare time to spend and participate in local lesuire activities. Moreover, older
respondents may be constrained by limited recreational activities and lifestyle
choices, leaving green spaces as preferred places to spend their spare time. People
nowadays have more alternatives to triditional parks, such as fitness centres, cafes and
karaoke venues, and these leisure activites are popular with the younger generations.

Furthermore, in this study, no siginificant relationship was found between distance


and use frequency or duration. The results may therefore imply that the effect of
distance on green space use in Xi’an is less strong than other studies suggested. For
example, in Denmark, a detailed comparison between the changes of citizens’ use
frequency and the distance to nearest green spaces from the home found that the
longer the distance, the lower the frequency of use, albeit distant parks with good
quality amenities attracted more visitors than closer but simpler parks (Schipperijn et
121
al. 2010). The nearest green space being more than 300 m away was a factor that
significantly diminished the citizens’ daily use. An investigation of the variations
among socioeconomic groups revealed that younger respondents (16-24) were more
likely to go to green spaces within a distance of less than 300 m than the age group of
45-65 years (Schipperijn et al. 2010). In this study, the lack of significance of distance
may be mainly due to the constrained questionnaires, since 76% of respondents
denoted that they usually visited a green space within a distance of 15 minutes from
their homes. Moreover, the majority of respondents were active users of green spaces.
The results are in agreement with a study by Schipperijn et al, which found that within
a relatively near distance, a majority of respondents visited the green spaces
frequently (Schipperijn et al. 2010).

In addition, there were other influencing factors such as the facilities and accessibility
of parks. In the UK, a study suggested that visitors were attracted by the high quality
of green spaces rather than a lack of alternative spaces within a smaller catchment
area of visitors’ homes (Lafortezza et al. 2009). The relationships between the quality
of green spaces in Xi’an city and their use frequency or duration can be discussed in
the future, according to research such as the maximum distance that visitors are
willing to accept to get to a high-quality green space. Moreover, the relationship of
visitors’ use frequency and stay time in a higher quality but slightly more distant
green space compared to a lower quality but closer green space merits further
investigation to better inform the allocation of resources.

6.7.1.3 The Influence of Park Variables on Respondents’ Use

Behavior
Considering park characteristics, respondents’ use habits were not influenced by the
variables of park area, age and location, but were influenced by the variables of
vegetation structure (see Tables 6.6 and 6.7). Parks with a lower cover of trees and
evergreens attracted more visits, while those with medium percentages of multi-
layered vegetation had longer stay times. These results may indicate that visitors in
Xi’an prefer to use parks with larger open spaces and to stay longer in parks with
abundant layers of vegetation with a relatively low density. This corroborates the
findings of previous studies which showed that moderately open scenes and an
intermediate density of vegetation cover are preferred in green spaces (Bjerke et al.
2006). However, due to the limited number of questionnaires at specific parks, it is
impossible to know if the parks with low tree cover and medium cover of multi-
layered vegetation meet the use habits of different socioeconomic groups, such as
older respondents, females, or students.
122
6.7.2 Discussion of Respondents’ Recreational Activities in
Green Spaces
6.7.2.1 A Comparison of Respondents’ Recreational Activities

with Similar Studies


Citizens of Xi’an embraced multiple recreation activities in urban green spaces. To
enjoy the scenery and large greenery and to have contact with nature constituted the
top three most important activities (Figure 6.1). By contrast, the use of recreational
facilities and walking a dog were the least important (Figure 6.1). Similarly, a study in
Guangzhou indicated that cleanliness and beautiful views were the reasons that
attracted the most citizens, while seeing birds, other wildlife and water bodies were
the least attractive (Chen 2006). In a later study, also in Guangzhou, to breathe fresh
air and to enjoy beautiful scenery were the most frequent recreational activities in
green spaces, while observing wildlife and taking part in cultural activities were the
least popular (Shan 2009). In a number of studies, to walk and to relax and enjoy the
scenery were the preferred activities (Casper et al. 2013; Derkzen 2012; Rall et al.
2017). Similar with many other studies, enjoyment of nature and beautiful views was
also aspired to by citizens of Xi’an, and the results suggest that the recreational
activities and amenities are the primary ecosystem services that the citizens would
like to derive from green spaces (Chiesura 2004; Oguz 2000; Sanesi and Chiarello
2006). In this study, over half of the preferred recreation activities in green spaces
were of the passive type. A study in Berlin and Salzburg also found that the majority
of park visitors tended to do passive activities such as reading, sunbathing, and
enjoying the flora and fauna, rather than active activities, such as walking and
walking dogs, when they used the parks (Voigt et al. 2014).

On the other hand, the lower inclination to be active, such as to walk, to walk a dog,
to do sports or to use recreational facilities, may indicate a shortage of corresponding
provisions such as sporting facilities, recreational facilities, playing fields and
playgrounds. In Chinese cities, the recreational facilities include some free amenities
such as public fitness equipment, sandpits, seesaws and swings, as well as for-pay
options such as rides, ferries wheels, boating, and fishing. The low percentage of use
of recreational facilities may indicate the poor condition or limited usefulness of the
existing facilities. Finally, the fact that walking a dog was not a main recreational
activity might simply be due to sample bias, and that not many citizens who have a
dog took part in the questionnaire survey.

123
6.7.2.2 The Influence of Socioeconomic Variables on

Respondents’ Recreational Activities


People with distinct socioeconomic backgrounds have different preferred recreational
activities. In this study, age significantly affected the choice of recreational activities.
The older respondents preferred to use the recreational facilities in green spaces more
than other groups. The results may again prove that the lifestyle choices of the elderly
are limited to the traditional and local recreation opportunities, rather than new
popular activities. Age-related differences were also found in a study by Chiesura
(2004). Doing sports and meeting friends were more popular among the young.
Relaxing and staying with children seem to be preferred by adults and the elderly. In a
study from Brazil, older people preferred to walk through the park, middle aged
people engaged more in playing with children, and young people were found to prefer
team sports (Derkzen 2012). In Guangzhou, the gender and marital status affected the
respondents’ use aims, such as relaxing and breathing fresh air (Shan 2009). In
general, the older respondents in this study enjoyed different activities in green spaces
than the other age groups. They tended to take part in more passive activities and use
more of the provided facilities in parks than the younger generations. Additionally, the
variety of recreation activities mirrors the citizens’ demands and expectations of green
spaces (Chiesura 2004). This information can be used to develop strategies to meet
recreational needs of the public.

6.7.2.3 The Influence of Park Variables on Respondents’

Recreational Activities
Respondents’ initiative to take part in recreational activities in green spaces increased
within the larger and newer parks. This may be due to the fact that the larger and
newer parks provide enough space and well-kept facilities for visitors to relax, do
sports, enjoy peace and quiet, etc. Recently, park designers have paid more attention
to providing various ecosystem services, and especially recreational services. The
larger parks can easily achieve these aims due to their larger area. The newer parks
have newer facilities and require less effort to achieve satisfactory maintenance.

Additionally, the respondents’ preferences for recreational activities were closely


connected with the parks which have 26-50% of multi-layered vegetation and
evergreen species. A study of green spaces in Iran also found that the majority of
recreational activities take place in green spaces with medium density (40-60%) and
single-layered vegetation (Aminzadeh and Ghorashi 2007). These results may indicate
that respondents prefer to do recreational activities in green spaces with simple
124
vertical structures. Multi-layered vegetation is composed of trees, shrubs and lawns.
This composition directly reduces the available space for activities such as sporting,
resting and playing with children, because of the barrier presented by shrubs. By
contrast, a piece of ground covered by trees or trees over lawn would provide more
space for visitors to do different kinds of activities. Generally, this study investigated
the respondents’ preferred recreational activities, but did not inquire which types of
recreational activities are preferably done in which types of vegetation structure.

6.7.3 Discussion of Respondents’ Perception of Green Spaces


6.7.3.1 Comparison of Respondents’ Perceptions with Similar

Studies
In Xi’an, over half of respondents deemed the green spaces near their home as very
important (Table 6.11). This finding was in agreements with studies by Shan (2009),
Schipperijn et al., (2010) and Li et al., (2011). In this study, the socioeconomic
analysis found that age significantly affected this perception. Elderly respondents
usually considered the nearby green spaces as more important than their younger
counterparts. This observation most likely means that distance would restrict the
green space use of older citizens more than in the case of other age groups.
Nevertheless, younger citizens in Denmark preferred the nearby green spaces more
than their older counterparts (Schipperijn et al. 2010). The results of this study
revealed that citizens usually preferred the green spaces near their homes but the
socioeconomic variables varied locally.

The questionnaires concerning the ecosystem services usually mentioned cultural


services (e.g. tourism, recreation, and aesthetics), regulating services (air purification,
climate regulation, water regulation and soil formation) and provisioning services
(e.g. water provision and food and material provision) (Casado-Arzuaga et al. 2013;
Jim and Chen 2006a). In this study, microclimate regulation (shading and O2 release)
and recreation (places for recreational activities, more contact with nature and
aesthetic improvement) were rated as most important (Figure 6.2). These findings
conformed to those from other studies. For example, microclimate was ranked as the
most important ecosystem service provided by green spaces in Phnom Penh,
Cambodia (Yen et al. 2016). By contrast, the aesthetic value of green spaces was
regarded as the most important service in Hong Kong, China (Lo and Jim 2010b) and
Rotterdam, the Netherlands (Buchel and Frantzeskaki 2015). Additionally, the
contribution of green spaces to contact with nature was stressed most in a study by
Pincetl and Gearin (2005).

125
However, these findings also showed some differences compared with similar studies.
Studies conducted in various countries such as Uganda (Hartter 2010), Uruguay
(Vihervaara et al. 2012), Spain (Casado-Arzuaga et al. 2013; Castro et al. 2011) and
cities such as Hong Kong (Lo and Jim 2010a; Lo and Jim 2010b), revealed that air
purification was considered to be one of the most important ecosystem services. In the
research of Tyrväinen (2001), the benefits related to nature were rated as most
important. The presence of biodiversity was perceived to be the most important green-
space service in the research of Vihervaara et al. (2012) and Izaskun et al. (2013). The
differing focus on ecosystem services may indicate that respodents from various
cultural contexts hold varying perceptions.

In this study, it was notable that the importance of noise reduction, wind protection,
economic value, neighbor-social interaction and water-soil conservation were ranked
at the bottom of the importance hierarchy. Similarly, property values, neighbor-social
interaction, noise abatement and soil erosion prevention were ascrbed the least
importance in another Chinese city - Hong Kong (Lo and Jim 2011). Noise abatement
and economic value were considered the least important in a study by Tyrväinen
(2001), while economic value and the benefits related to environment quality such as
control of wind and stormwater were ranked low in the research of Lorenzo (2000).
These results may indicate that respondents were not aware of the importance of these
indirect benefits derived from green spaces.

Several factors could have jointly contributed to the discrepancies of repondents’


perceptions of ecocystem services in Xi’an city. Firstly, local enviromental conditions
inevitably influence the perception of green spaces. For example, shading was ranked
as the most important service in many countries, possibly due to the strong sunshine
in summer, but not in Nordic countires like Finland (Tyrväinen et al. 2007), where
summers are not so hot and sunshine is welcomed. Air quality improvement was not
considered as a very important service like in other countries, although environmental
problems are more acute in Chinese cities, possibly due to limited abatement of air
pollution provided by green spaces (Nowak et al. 2006), in addition to a relatively low
percentage area of green spaces, and an overload of exhaust fumes and other air
pollutions in Xi’an. Moreover, in the predominantly high-rise city, the wind protection
functions of urban green spaces are limited in most places.

Secondly, urban living is superior to living in rural areas in the perception of most
Chinese people (Chen and Pang 2013; Li 2013). Urban areas usually provide more job
opportunities, higher incomes, better living accommodation such as convenient
transportation, more hospitals and schools (Chen and Pang 2013; Li 2013). Benefits
126
related to nature are therefore not considered as important as those that bring direct
material improvement to living conditions.

Thirdly, this study found that the most important services were those related to
personal comfort and recreation, which can be directly or easily percevied and are
tangible. On the other hand, less conspicuous ecological services such as water-soil
conservation received low importance ratings. Stormwater and wastewater is usually
collected and managed through the urban sewage systems rather than green spaces
(Lou 2003). Therefore, citizens were mostly ignorant of the fact that green spaces also
have an effect on water-soil conservation. In addition, in Chinese cities, green spaces
seldom contribute economically improtant plants such as timber and fruits. This is
perhaps unsurprising since traditionally, green spaces were understood as places of
recreation and aesthetics (Lou 2003). Their economic value was underestimated
mainly due to the lack of understanding how green spaces affect the prices of the
neighboring real-estate. The limited understanding of such benefits implies that the
possession or lack of relevant knowledge could affect perceptions. Therefore, to
improve the recognition of the ecosystem services provided by urban green spaces,
appropriate information could be included in public education and publicity programs
for communication through formal and informal channels (Costanza et al. 1997;
Lewan and Söderqvist 2002).

6.7.3.2 The Influence of Socioeconomic Variables on

Respondents’ Perceptions of Ecosystem Services


This study found that there was no significant correlation between respondents’
perceptions of urban green spaces and their socioeconomic backgrounds. However,
there were considerable differences in their ranking of individual services. This
widespread recognition of ecosystem services will hopefully help in the
implementation of urban policies and planning projects that promote ecosystem
services.

6.7.3.3 The Influence of Park Variables on Respondents’

Perceptions of Ecosystem Services


Park area and vegetation structure significantly influenced the respondents’
perceptions of ecosystem services of green spaces (see Table 6.15). Respondents from
the small to medium and medium-sized parks (5-10 ha and 10-30 ha, respectively)
tended to consider the ecosystem services of green spaces as more important.
Moreover, respondents also perceived the ecosystem services as more important in the
parks with relatively higher percentages of multi-layered vegetation, trees and mature
127
trees, and medium percentages of evergreen species. Based on a defined piece of
ground, abundant layers of vegetation increase the green biomass and hence benefit
ecosystem services such as CO2 sequestration, air humidiy and lower air temperature.
Moreover, the muliti-layered structure helps block the wind and alleviate air pollution
due to the resulting interruption of airflow and deposition of particulates. A higher
cover of trees, and especially mature trees, can directly provide shading by their large
canopy. These vegetation characteristics, which determine ecosystem services, were
likely considered as important simply because they can be perceived directly by the
respondents.

6.7.4 Discussion of the Respondents’ Assessments of Green


Spaces
6.7.4.1 Comparison of Respondents’ Assessments with Similar

Studies
The results of assessments both at the city- and the park level indicated that the
satisfaction of Xi’an’s citizens with the quality of urban green spaces still needed to
be improved. Generally, no more than 57.2% of the respondents assessed the entirety
of green spaces as positive or very positive (Table 6.18). In a study from Porto and
Lisbon, Portugal, a similar ratio of 61% and 62% of respondents were satisfied with
the cities’ green spaces, respectively (Madureira et al. 2015).

Among all the 20 assessment aspects, only 5 were assessed as positive or very
positive by over half of respondents in the often-used urban parks. These included
vegetation coverage, microclimate, overall assessment, cleanliness and scenic beauty
(Table 6.18 and Figure 6.3). The study of Qin et al. (2013) also found that overall
satisfaction with green spaces is higher than the satisfaction with specific parks.
Among the specific aspects, pleasant landscape, visual elements, proximity to water
and peaceful spaces were rated as the primary factors of satisfaction by park users in
Ankara (Oguz 2000). In Finland, opportunity for activities, beautiful landscapes and
freedom were the top three reasons for positive assessment (Tyrväinen et al. 2007). In
Guangzhou, vegetation conditions, vegetation cover ratio and air quality were
assessed very positively (Shan 2009). In general, the vegetation coverage and scenery
were much more satisfactory in the above studies. The observation that items related
to vegetation meet the demands of more respondents regarding use and aesthetics may
indicate that the vegetation cover, condition and scenery have been paid more
attention and enjoyed good development in green spaces. In fact, according to the
field observation, most of the 22 investigated parks had over 50% of vegetation (see
128
Table 6.1). The high cover of vegetation can be viewed as the proximal reason for the
respondents’ satisfaction with vegetation cover.

Water quality and area, air quality, the number of sports facilities and the number of
wild animals were rated the lowest, which may be attributed to the small size,
insufficient management and heavy use of green spaces in China. It was notable that
the neutral assessments accounted for relatively higher percentages, reaching even
50% for some aspects such as ecological functions at city level, as well as air quality
and location of resting facilities at park level (Tables 6.18 and 6.19). In fact, the
turbidity of water, especially of man-made lakes, can be seen directly in many parks
as a deficiency of cleanliness. The improvement for air and water quality often goes
beyond the department of urban green spaces and requires effective coordination
mechanisms with other government departments e.g. the department of environmental
protection.

6.7.4.2 The Influence of Socioeconomic Variables on

Respondents’ Assessments
Perhaps unsurprisingly, the respondents’ family status influenced their assessments of
the functions and facilities designed for children. For example, respondents living
with children assessed playgrounds more positively but air quality more negatively
than their childless peers. These findings may imply that respondents living with
children are concerned with the quality of green spaces primarily from the perspective
of children. Children are the main users of playgrounds, and park designers should
consider their specific use by children in order to satisfy their parents. The reduced
satisfaction with air quality from respondents who have children may indicate that
they have higher requirements for air quality. This may be due to the effect of air
pollution on children, who possibly have more pollution-related health problems due
to their still underdeveloped organs.

By comparison, in the study of Qin et al. (2013), young and middle-aged respondents
were more satisfied with the overall green spaces than the elderly (Qin et al. 2013). In
a study by Shan (2009), marital status played a significant role in the assessment of
urban green spaces - married respondents made consistently more positive
assessments than singles in all aspects. Taken together, these findings demonstrate
that the respondents’ assessments are easily influenced by their socioeconomic status,
though the concrete influences differ between social contexts.

129
6.7.4.3 The Influence of Park Variables on Respondents’

Assessments
Considering the influence of the parks’ characteristics, older parks located within the
second ring road, and those with areas of 10-30 ha and 0-5 ha received relatively
higher positive assessments from respondents (see Table 6.20). The results therefore
revealed that, from the perspectives of respondents, not only do the bigger parks
perform well in terms of the provision of ecosystem services, but the smaller parks
were also able to provide satisfactory ecosystem services. Similar research in
Guangzhou showed that municipal parks were assessed highly while small gardens
were largely neglected (Shan and Yu 2014).

Moreover, the parks with medium cover of multi-layered vegetation and evergreens,
as well as those with higher cover of trees but lower cover of mature trees, were
assessed more positively than other types. These findings were in agreement with
research conducted in Guangzhou, where park users presented more positive
assessments in municipal parks, which had better vegetation conditions than the
district parks (Shan and Yu 2014). In research from Iran, respondents most highly
assessed the scenery that was characterized by 20-60% of multi-level plantings,
mainly including various trees (Aminzadeh and Ghorashi 2007). In this study, the
results highlighted the fact that increased tree cover can help increase the satisfaction
with parks. Trees can directly benefit the quality of scenery in green spaces as its
integral components. Although tree cover cannot change the conditions of sporting
facilities, resting facilities and playgrounds, their contribution to environmental
improvement of parks may indirectly improve the perception of the overall green
space and hence also of the other items of ecosystem and leisure services. For
example, respondents may feel more comfortable when they have an opportunity to
rest in the shade of trees, and they may prefer to do sports or participate in activities in
green spaces with abundant trees. Moreover, the parks with medium cover of multi-
layered vegetation, mature trees and evergreen trees may provide balanced shading,
sporting spaces, beautiful scenery and temperature regulation due to the larger
canopies, relatively open spaces and scenery, stronger transpiration and shielding of
pollutants, and therefore, easily received positive assessments from respondents.

Generally, respondents’ assessments of the majority of the above-mentioned aspects


of green spaces were less positive. Moreover, positive assessments were more
common in the parks with a higher cover of trees but lower cover of mature trees and
medium cover of multi-layered vegetation and evergreens. The modest satisfaction
130
with urban green spaces indicates that there is a noticeable gap between current status
of green spaces and citizens’ expectations. This implies the necessity for substantial
improvement of urban green spaces in Xi’an. Despite the lack of analogous local
studies to compare, the findings from this investigation of 22 urban parks provide
some important guidelines for the improvement and development of green spaces in
the future.

6.7.5 Discussion of Respondents’ Demands for Green Spaces


6.7.5.1 Comparison of Respondents’ Demands with Similar

Studies
Generally, the respondents in this study voiced active demands for the referred items
provided by green spaces. More shade, more opportunities to observe nature, more
flowers and better management were the primary demands for urban green spaces
from respondents in Xi’an city (Figure 6.4). Some of the respondents indicated that
they needed more recreational opportunities, e.g. more facilities for sports and
exercise, more play areas for children and more resting facilities, but less wind
protection and cafes or restaurants (Figure 6.4). In spite of their different cultural
contexts, people share some common desires for features of urban green spaces. For
example, a study from Hong Kong indicated that the residents’ wanted more greenery,
and especially trees (Lo and Jim 2011). A survey in East midlands, UK showed that
the respondents most sought green spaces with a natural appearance (Bell et al. 2004).
In Sapporo, Japan, a survey suggested that flowers were the most preferred among
possible elements for the space beneath street trees (Todorova et al. 2004).

Usually, expectations of urban green spaces are predominantly associated with natural
features such as natural appearance, trees, water, flowers and grasses (Yuen et al.
1999). The need for nature contact may be prompted by the acute shortage of urban
greenery and the highly regimented and manicured design of traditional urban green
spaces. In practice, the predominant stereotypic and old-fashioned park design fails to
meet the users’ demands and is detached from the different demands among age
groups, vocations and incomes. For example, Access to managed lawns and vegetated
areas is prohibited in the majority of urban parks in Xi’an. This rule constrains the
users’ contact with nature and leads to unmet demands.

131
6.7.5.2 The Influence of Socioeconomic Variables on

Respondents’ Demands
In this study, the respondents’ demands varied with different park age and location.
Respondents from different age- and vocational groups presented different demands.
The elderly demanded more quiet spaces, larger play areas for children, more resting
facilities and less cafes and restaurants than the other age groups. The demands for
play areas and resting facilities may be related to the care for children. It can be
understood that elderly demanded more play areas for children because it is a
widespread social phenomenon in China that the elderly look after their
grandchildren. With aging, the need for quiet and resting facilities increases due to
decreased vigor. Similarly, in the study from Hong Kong, age influenced the residents’
anticipations, and older users needed more seats in green spaces (Lo and Jim 2011). In
Guangzhou, the married more strongly preferred naturalness, facilities and
accessibility than singles (Shan 2009). Bringing children to green spaces for play
usually constitutes a predominant reason for the visits of families with children, and
meeting the children’s demands is their primary consideration.

6.7.5.3 The Influence of Park Variables on Respondents’

Demands
Respondents from the older parks voiced a stronger need for improving many kinds of
ecosystem services locally. This result may indicate that the facilities and
management of older parks are unsatisfactory due to their long history of use and their
outdated design. Respondents in the parks with lower cover of multi-layered
vegetation and evergreens usually wanted more shade, wind protection, sporting
facilities and water bodies. By contrast, the respondents from the parks with higher
cover of trees and mature trees wanted better management, more quiet places and
more flowers. This shows that respondents’ demands varied with different park
characteristics, and their needs are thus connected with their direct visual and use
experiences in the parks. These subjective perspectives can work well in most cases.
For example, in the parks with higher cover of trees, there were more demands for
flowers, while in the parks with 0-25% of mature trees there were more demands for
shading because the lower cover of mature trees provided less shading.

Conversely, the results also suggest that the urban environments may influence the
citizens’ perspectives and interactions with urban nature. Hence, it would be
reasonable to establish corresponding public policies. However, green space planners
and managers often hold different views than users (Burgess 1988; Coles and Bussey
132
2000). The traditional dominance of professional inputs tends to neglect users’ views
and demands. Public participation and engagement in green space design remain
ineffective. This study demonstrates again that the public’s views and demands should
be considered and adopted in the process of green space planning and development.
Survey-based communication between the public / park users and park managers /
planners is an effective method to fill the gaps between provisions and demands, and
thus to avoid the construction of underused or unwanted green spaces.

133
7 Assessment of Ecosystem Services
Generated by Urban Parks in Xi’an
Regulating Services of Urban Parks in Xi’an
7.1.1 Indicator Ratings of Regulating Services of Urban
Parks in Xi’an
The vegetation structures and types discussed in section 2.3.2 were used as indicators
to assess the regulating services of green spaces in this study. The first four indicators
revealed the effects of microclimate regulation. In addition to these, the cover of
evergreen species was supplemented to better assess the green spaces’ effects on air
pollution.

Vegetation and Water (1.1)


All of the 22 investigated parks in Xi’an were covered with over 50% of vegetation
and water, with thirteen having over 75% (see Figure 7.1). The highest proportion of
total vegetation and water was 92.1% in Xin Ji Yuan Park, and the lowest was 62.6%
in Children’s Park (see Figure 7.1). Therefore, all the parks were ranked as having
medium or large covers of vegetation and water.

Xin Ji Yuan Park 92.1


Qu Jiang Yi Zhi Park 87.4
Wen Jing Park 87.2
Fang Zhi Park 86.6
Mu Dan Yuan 86.3
Tang Yan Road Tang City Wall Park 85.9
Feng Qing Park 84.6
Xing Qing Park 82.0
Min Su Park 81.5
Ge Ming Park 81.0
Lao Dong Park 80.1
Lian Hu Park 79.6
Ci En Si Yi Zhi Park 76.3
Chang Le Park 73.7
Yong Yang Park 73.4
Xi Qu Da Guan Yuan 71.6
City Sports Park 71.5
Mu Ta Si Park 71.2
Tu Men Road Park 71.0
Huan Cheng Xi Yuan 66.5
Qu Jiang Tang City Wall Park 64.0
Children’s Park 62.6
0 10 20 30 40 50 60 70 80 90 100
Figure 7.1 Cover ratio of vegetation and water in the 22 investigated parks in
Xi’an

134
Multi-layered Vegetation (1.2)
Multi-layered vegetation structures, having at least two layers, were found in all of the
22 investigated parks (see Figure 7.2). Moreover, the vegetated areas in over half of
the parks were covered with over 75% of multi-layered vegetation. Especially in Mu
Ta Si Park, Mu Dan Yuan and Tu Men Road Park, the relative percentages of multi-
layered vegetation reached as high as 96.3%, 94.6% and 94.4%, respectively (see
Figure 7.2). According to the rating rules, it can be seen that these parks attained the
high rank (≥76%) (see Figure 7.2). Six parks were in the medium rank (51-75%), with
percentage areas of multi-layered vegetation ranging from 52.5% in Feng Qing Park
to 73.1% in Ge Ming Park (see Figure 7.2). Five parks were in the low rank (26-50%).

Mu Ta Si Park 96.3
Mu Dan Yuan 94.6
Tu Men Road Park 94.4
Qu Jiang Yi Zhi Park 89.1
Lian Hu Park 86.4

Children’s Park 81.7


Min Su Park 81.2
City Sport Park 79.9
Fang Zhi Park 78.5
Huan Cheng Xi Yuan 77.4
Wen Jing Park 76.0
Ge Ming Park 73.1
Ci En Si Yi Zhi Park 63.6
Xing Qing Park 60.9
Chang Le Park 54.3
Feng Qing Park 52.5
Qu Jiang Tang City Wall Park 50.1
Lao Dong Park 47.0
Xin Ji Yuan Park 43.2
Xi Qu Da Guan Yuan 40.1
Tang Yan Road Tang City Wall Park 34.2
Yong Yang Park 27.6

0 10 20 30 40 50 60 70 80 90 100

Multi-layered vegetation Two-layered vegetation one-layered vegetation

Figure 7.2 Relative proportions of vegetation layers in the 22 investigated parks


in Xi’an.

Tree Cover (1.3)


Tree cover ranged from 26% to 92.2% in the 22 investigated parks, with twelve parks
having over 50% of tree cover. Six parks had over 75% of tree crown coverage in
their vegetated areas, with the percentage areas ranging from 75.5% in Ci En Si Yi
135
Zhi Park to 79.9% in Ge Ming Park, and rising to the highest percentage of 92.2% in
Lao Dong Park. These parks were hence rated as having the high rank (see Figure
7.3). Additionally, six parks had tree covers between 50% and 75%, with the
proportion ranging from 51.9% in Mu Ta Si Park to 74.4% in Min Su Park (see Figure
7.3). Therefore, these six parks were rated as having the medium rank. The other ten
parks were rated in the low rank since their relative proportions of trees were in the
rage of 26-50% (see Figure 7.3).

Mature Tree Cover (1.4)


The relative proportion of mature trees was low in all of the investigated parks, and
the vegetation cover comprised over 50% mature trees in only four. Chang Le Park
had the highest relative coverage of mature trees at 66.5%, and therefore reached the
medium rank of regulating impact (see Figure 7.4). Two parks had relative mature tree
coverages between 25% and 50%. One was Xing Qing Park (35.8%) and the other
was Children’s Park (46%) (see Figure 7.4). The remaining 16 parks all ranked very
low, whereby the percentage area of six parks ranged from 21.6% to 24.9%, while
eight parks had no cover of mature trees at all (see Figure 7.4).

Lao Dong Park 92.2


Ge Ming Park 79.9
Childern's park 79.4
Lian Hu Park 78.7
Chang Le park 75.7
Ci En Si Yi Zhi Park 75.5
Min Su Park 74.4
Xing Qing Park 64.6
City Sports park 56.3
Xin Ji Yuan Park 53.5
Tu Men Road Park 53.1
Mu Ta Si Park 51.9
Yong Yang Park 49.1
Tang Yan Road Tang City Wall Park 49.1
Huan Cheng Xi Yuan 48.5
Feng Qing Park 47.3
Qu Jiang Yi Zhi Park 47.2
Xi Qu Da Guan Yuan 45.0
Fang Zhi Park 44.6
Qu Jiang Tang City Wall Park 38.1
Wen Jing Park 31.7
Mu Dan Yuan 26.0
0 10 20 30 40 50 60 70 80 90 100

Figure 7.3 Relative proportions of tree cover in the 22 investigated parks in


Xi’an.

136
Chang Le Park 66.5
Min Su Park 59.6
Ge Ming Park 57.3
Lao Dong Park 54.1
Children’s Park 46.0
Xing Qing Park 35.8
Fang Zhi Park 24.9
Mu Ta Si Park 24.0
Tu Men Road Park 24.0
Xi Qu Da Guan Yuan 23.9
Ci En Si Yi Zhi Park 23.6
Lian Hu Park 21.6
Xin Ji Yuan Park 3.1
Feng Qing Park 1.9
Qu Jiang Yi Zhi Park 0.0
Tang Yan Road Tang City Wall Park 0.0
City Sports Park 0.0
Qu Jiang Tang City Wall Park 0.0
Yong Yang Park 0.0
Huan Cheng Xi Yuan 0.0
Mu Dan Yuan 0.0
Wen Jing Park 0.0
0 10 20 30 40 50 60 70

Figure 7.4 Relative proportions of mature trees in the 22 investigated parks in


Xi’an.

Evergreen Species Cover (1.5)


All the investigated parks were designed to contain evergreen woody plants, including
many kinds of trees and shrubs. According to the rating rules, two parks were ranked
as having a high regulating level due to their evergreen coverage, which was in excess
of 75%. The coverage reached 80.6% in Min Su Park and 87.0% in Chang Le Park
(see Figure 7.5). Accordingly, ten parks were ranked as medium impact (51-75%),
with relative evergreen covers ranging from 52.2% in Ci En Si Yi Zhi Park to 70.9%
in Tu Men Road Park (see Figure 7.5). Nine parks were ranked as low due to their 25-
50% of evergreen cover. Especially, Tang Yan Road Tang City Wall Park had only
13.3% of evergreen species’ cover and therefore was ranked as having a very low
impact on the regulation of air pollutants (see Figure 7.5).

137
Chang Le Park 87.0
Min Su Park 80.6
Tu Men Road Park 70.9
Feng Qing Park 70.2
Fang Zhi Park 69.9
Lian Hu Park 64.9
Huan Cheng Xi Yuan 64.8
Wen Jing Park 63.8
City Sport Park 63.5
Ge Ming Park 54.6
Xi Qu Da Guan Yuan 54.2
Ci En Si Yi Zhi Park 52.2
Xin Ji Yuan Park 48.1
Mu Dan Yuan 47.4
Qu Jiang Yi Zhi Park 47.1
Xing Qing Park 45.9
Children’s Park 40.5
Qu Jiang Tang City Wall Park 35.7
Lao Dong Park 35.4
Yong Yang Park 34.1
Mu Ta Si Park 27.8
Tang Yan Road Tang City Wall Park 13.3
0 10 20 30 40 50 60 70 80 90 100

Figure 7.5 Relative proportions of evergreen species in the 22 investigated parks


in Xi’an.

7.1.2 Assessment of Integrated Indicators of Regulating


Services of Urban Parks in Xi’an
In this study, regulating services were mainly investigated in terms of microclimate
regulation and air pollution improvement. The values of regulating services of the 22
investigated parks were assessed based on aggregations of the specific assessment
indicators (see the detailed method in Section 3.2.2.4). Overall, four parks had high
values of air quality regulation, fifteen parks had medium values, and three had low
values (see Table 7.1). With the exception of Lian Hu Park, the three other high-value
parks - Min Su Park, Ge Ming Park and Chang Le Park, all had medium and high
impacts on all the assessed indicators. The low valued parks - Yong Yang Park, Qu
Jiang Tang City Wall Park and Tang Yan Road Tang City Wall Park had low values of
mature tree cover and evergreen species cover.

138
Table 7.1 Assessment of integrated indicators of regulating services

Children’s Park

Tu Men Road Park

Fang Zhi Park

Ci En Si Yi Zhi Park

Xi Qu Da Guan Yuan

Lian Hu Park

Min Su Park

Xin Ji Yuan Park

Wen Jing Park

Mu Ta Si Park

Mu Dan Yuan

Lao Dong Park

Ge Ming Park

Huan Cheng Xi Yuan

Yong Yang Park

Chang Le Park

Feng Qing Park


Park
Qu Jiang Tang City Wall
City Sports Park
City Wall Park
Tang Yan Road Tang
Xing Qing Park
Qu Jiang Yi Zhi Park
Park

Score of
assessed
indicators

Score of 1.1 3 3 4 4 3 4 4 4 4 3 4 4 4 3 3 3 4 3 3 4 4 4
Score of 1.2 4 4 4 3 2 4 4 2 4 4 4 2 3 4 2 3 3 2 4 2 3 4
Score of 1.3 4 3 2 4 2 4 3 3 2 3 2 4 4 2 2 4 2 2 3 2 3 2
Score of 1.4 2 1 1 1 1 1 3 1 1 1 1 3 3 1 1 3 1 1 1 1 2 1
Total score for microclimate
13 11 11 12 8 13 14 10 11 11 11 13 14 10 8 13 10 8 11 9 12 11
regulation
Assessment of microclimate
H M M M L H H M M M M H H M L H M L M M M M
services
Score of 1.5 2 3 3 3 3 3 4 2 3 2 2 2 3 3 2 4 3 2 3 1 2 2
Total score for air pollution
15 14 14 15 11 16 18 12 14 13 13 15 17 13 10 17 13 10 14 10 14 13
removal
Final assessment of air pollution
M M M M M H H M M M M M H M L H M L M L M M
removal

H: high; M: medium; L: low; VL: very low

139
However, the value of the individual parks’ microclimate regulation showed some
differences (see Table 7.1). According to the rating rules, among all the parks, six had
high values of microclimate regulation, compared to thirteen with medium values and
three with low values. Generally, the parks provided similar regulation of microclimate
and air quality. Nevertheless, some cases were exceptions. For example, Xi Qu Da Guan
Yuan had a low value of microclimate regulation but medium value of air pollution
reduction, and Tang Yan Road Tang City Wall Park had a medium value of microclimate
regulation but a low value of air pollution reduction. Children’s Park and Lao Dong Park
showed a high value of microclimate regulation, but provided only a medium value of air
pollution reduction.

Considering the varying park area, age and location, the relationship between these
variables and air pollution removal / microclimate regulation were analyzed using the
Chi-squared test. Somewhat surprisingly, it was found that there were no significant
relationships between the value of air pollution removal and park variables. However, the
results showed that the majority of the parks with high values of regulating services were
small to medium-sized parks, which were located within the second ring road and built
prior to the 1990s, such as Lian Hu Park and Ge Ming Park. Using a similar analysis, it
was revealed that microclimate regulation of the parks showed significant relationships
with park variables. Of the six high-value parks regarding microclimate regulation, five
were built prior to the year 1990 (X2 = 10.358, p = 0.006). Furthermore, five were located
within the second ring road (X2= 6.229, p = 0.044).
7.1.3 Citizens’ Assessments of Regulating Services of Urban
Parks in Xi’an
In accordance with the descriptions set out in Chapter 6 (see Table 6.18), 70.8% of the
respondents assessed the overall microclimate regulation of the parks as “good” or
“excellent”, compared to 22.7% who stated the same for air quality. According to the
rating rules, the entirety of parks as a whole was considered to have a medium value of
microclimate regulation, but a low value of air quality improvement.

The associations between respondents’ assessments of microclimate / air quality and


socioeconomic variables were analyzed using the Chi-squared test. Children were a factor
which influenced the respondents’ assessment of air quality (X2 = 12.2946, p = 0.015), so
that more respondents who had no children positively assessed the air quality (28.22%
versus 18.87%).

Considering park characteristics, it was found that the park age and location had
significant relationships with respondents’ assessments of microclimate using the Chi-
squared tests. In the older parks, 78.7% of respondents positively assessed the
microclimate (X2 = 7.939, p = 0.047), and this positive assessment was 11.6% higher

140
than in the newer parks. 72.3% of respondents from the parks located inside the second
ring road considered the microclimate to be satisfactory, compared to 64.1% from parks
located inside the third ring road (X2 = 13.201, p = 0.004).

Using the same methods to analyze the associations between vegetation characteristics
and respondents’ assessments of microclimate and air quality, it was found that multi-
layered vegetation and evergreen species had a significant influence. 81.1% and 72.5% of
respondents from the parks with 26-50% and 51-75% of multi-layered vegetation
positively assessed the local microclimate (X2 = 19.560, p = 0.003). However, the
positive assessment of air quality was dramatically lower. The highest percentage of
positive assessments reached only 34.3% in the parks with 51-75% of multi-layered
vegetation (X2 = 21.885, p = 0.005). Moreover, there was a highly statistically significant
association between evergreen cover and microclimate assessment (X2 = 22.925, p =
0.006). More respondents (78.0% and 69.8%) in the parks with relatively lower cover of
evergreens (26-50% and 0-25%, respectively) positively assessed the microclimate than
their peers in other parks.
Recreational Services of Urban Parks in Xi’an
7.2.1 Citizens’ Assessments of Recreational Services of Urban
Parks in Xi’an
As described in section 6.5, 46.7% of the respondents assessed the overall recreational
services of all the green spaces in Xi’an city as either “excellent” or “good” (see Table
6.16), which means that less than half of the assessments were positive. According to the
rating rules, the entirety of green spaces was thus considered to have low value of
recreational services. For the individual parks, the respondents’ assessments of
recreational services were represented by the indicators of scenic beauty (2.1), sporting
facilities (2.2-2.4), resting facilities (2.5-2.7) and playgrounds (2.8-2.10).

Scenic Beauty (2.1)


Even though it is one of the most important reasons for citizens to visit green spaces,
scenic beauty received only 49.3% of “excellent” or “good” assessments from
respondents in this study (see Table 6.18). Moreover, 41.1% of respondents deemed the
scenery to be fair in their often-visited parks. According to the rating rules, the parks as a
whole were therefore considered to have a low value of scenic beauty.

Sporting Facilities (2.2-2.4)


Summing the assessments of sporting facilities provided by parks, 27.6% of respondents
considered the number of sporting facilities to be sufficient in their daily-used green
spaces (see Table 6.18). Meanwhile, 40.1% and 38.1% considered the location and
maintenance of sporting facilities to be suitable (see Table 6.18). Although more

141
respondents were satisfied with the location and maintenance, than with the number of
sporting facilities, the percentage of positive assessments was less than half. According to
the rating rules, the parks had a low value regarding their provision of sporting facilities.

Resting Facilities (2.5-2.7)


Overall, positive assessments reached only 29.2% for the number of resting facilities,
compared to 38.6% and 44.7% for the location and maintenance of resting facilities (see
Table 6.18). According to the rating rules, the parks as a whole were considered to have a
low value of resting facilities.

Playgrounds (2.8-2.10)
Overall, 40.9% of respondents thought that there were adequate areas of playgrounds in
their often-used green spaces (see Table 6.18), 42.0% and 49.6% of which felt that the
location and maintenance of playgrounds was suitable, respectively (see Table 6.18).
Generally, less than 50% of the respondents positively assessed the three items pertaining
to playgrounds. Therefore, according to the rating rules, the parks were considered to
have low values regarding their provision of playgrounds.
7.2.2 Integrated Assessment of Recreational Services of Urban
parks in Xi’an
The values of recreational services of urban parks were summarized by adding together
all the impact scores of the ten discussed indicators. When the 22 investigated parks were
taken as a whole, the final assessment was low (see Table 7.2). For the individual parks,
the assessments did not vary by different park groups and vegetation structures. Taking
the variable of park area as an example, parks in all the different area groups had low
values for total recreational services (see Table 7.2).

Nevertheless, the Chi-squared test revealed that some park variables showed significant
relationships with some specific items of recreational services. For example, park area
influenced the respondents’ assessment of scenery (X2 = 11.167, p = 0.011). Respondents
tended to positively assess the scenic beauty in the parks with areas of 10-30 and 0-5 ha.
In addition, scenery in the parks built prior to the 1990s satisfied more respondents than
the scenery in parks built after the 1990s (52.8% versus 47.9%).

Considering the influence of vegetation structure, a twice higher percentage of


respondents (31%) positively assessed the scenery in parks with 76-100% of vegetation
and water (X2 = 9.576, p = 0.048). The positive assessments of scenery were also
increased by the increasing coverage of multi-layered vegetation (X2 = 72.816, p <
0.001). However, this positive assessment decreased with the increased cover of trees and
mature trees. The highest positive assessment was recorded in the parks with 26-50% of
trees and 0-25% of mature trees (X2 = 18.546, p = 0.017, X2 = 39.718, p < 0.001). Parks

142
with a relatively high cover (51-75%) of evergreens received more positive assessments
than the others (X2 = 69.771, p < 0.001).

Respondents’ assessments of the number and location of sporting facilities varied


significantly by park area. Satisfaction with the location of sporting facilities reached
60.5% in parks with areas of 0-5 ha, followed by 43.6% and 40.3% in parks with areas of
10-30 ha and 30-60 ha, respectively (X2 = 12.783, p = 0.005). However, satisfaction with
the number of sporting facilities was 41.9% in the parks with areas of 0-5 ha and only 9%
in the parks with areas of 5-10 ha (X2 = 27.403, p < 0.001). Moreover, in the parks built
prior to the 1990s, 48% of respondents were satisfied with the location of sporting
facilities, compared to 37% in newer parks (X2 = 4.347, p = 0.037).

The cover of vegetation, water and evergreens had a significant influence on the number,
location and maintenance of sporting facilities. The parks with 75-100% of vegetation
and water, and 51-75% of evergreens, received more positive assessments than the other
parks. Interestingly, the cover of multi-layered vegetation, trees and mature trees also
influenced some items related to sporting facilities. Parks with 0-25% of mature trees
were considered more satisfactory regarding the number and maintenance of sporting
facilities, whereas the location and maintenance was deemed better in the parks with 0-
25% of trees.

In addition, park area, age and location had a significant relationship with the
respondents’ assessments of resting facilities. The smaller, older parks located inside the
second ring road received slightly more positive assessments, while parks with increasing
cover of multi-layered vegetation received higher positive assessments of the
maintenance of resting facilities. Interestingly, the parks with increasing tree cover and
mature tree cover had a decreased satisfaction rating of the number and location of
resting facilities. However, parks with 76-100% of vegetation and water were more
positively assessed regarding the number, location and maintenance of resting facilities.

Lastly, the parks with areas of less than 30 ha received more positive assessments of the
number of playgrounds. The highest percentage of satisfaction was 59.0% for the parks
with areas of 10-30 ha (X2 = 18.163, p < 0.001). Meanwhile, 55.1% and 53.5% of
positive assessments of the location of playgrounds came from parks with areas of 10-30
ha and 0-5 ha, respectively (X2 = 11.859, p = 0.008). The percentage of positive
assessments decreased for the other parks.

Park age and location also influenced the respondents’ assessments of the number of
playgrounds. In the parks built prior to the 1990s, 51.2% of respondents declared that
they are satisfied with the number of playgrounds, compared to only 36.7% in newer

143
parks (X2 = 7.623, p = 0.006). Similarly, 54.4% of satisfaction was recorded in parks
located within the second ring road, which was higher than for parks located within the
third ring road (X2 = 23.343, p < 0.001).

Considering the vegetation structure, it was found that the parks with 76-100% of
vegetation and water were positively assessed regarding the number and location of
playgrounds. Additionally, a reduced cover of trees raised the respondents’ positive
assessments of the number, location and maintenance of playgrounds.

144
Table 7.2 Integrated assessment of recreational services by park area

Often-used parks Park area (number of parks)


as a whole
(22) 0-5 ha (5) 5-10 ha (8) 10-30 ha (4) 30-60 ha (5)

Score of 2.1 2 1 1 1 1
Score of 2.2 2 1 2 1 1
Score of 2.3 2 1 1 1 1
Score of 2.4 2 2 2 2 3
Score of 2.5 2 1 1 1 1
Score of 2.6 2 2 1 1 1
Score of 2.7 2 1 1 1 1
Score of 2.8 2 1 1 1 1
Score of 2.9 2 1 1 1 1
Score of 2.10 2 1 1 1 1
Total score of
20 12 12 11 12
recreational services
Final assessments of
L L L L L
recreational services

H: high; M: medium; L: low; VL: very low

145
Discussion
7.3.1 The Influence of Park Variables on the Assessment of
Regulating Services
The majority parks investigated in this study had medium to high percentages of
vegetation and multi-layered vegetation, but low percentages of mature trees.
Consequently, they had a medium impact on the regulation of microclimate and air
quality (see Table 7.1). Nevertheless, the percentage of trees and evergreens varied
obviously among the parks, and four parks provided a high value of air quality
regulation. By comparison, more parks (six) provided a high value regarding the
regulation of microclimate. Three parks showed very limited values of air quality
removal / microclimate regulation.

Generally, the older parks with 5-10 ha of area, which were located within the second
ring road, usually provided high values of microclimate regulation and air pollution
removal. However, similar studies of Chang (2007) and Andrade and Vieira (2007)
showed that bigger parks had higher impacts on temperature regulation. In their
studies, air temperature was lower on average in larger parks than in smaller ones
(Andrade and Vieira 2007; Chi-Ru Chang et al. 2007). Notwithstanding, it is
impossible to establish an accurate linear relationship between the size of parks and
the temperature variation associated with them (Andrade and Vieira 2007; Chi-Ru
Chang et al. 2007), because the temperature change also highly depends on the
structure of vegetation, the topography and the characteristics of the parks’
surroundings (Andrade and Vieira 2007; Givoni 1991b).It should also be noted that
the range of the effect of parks on the climatic conditions within the surrounding
built-up areas is rather limited, even in the case of very large ones, or areas downwind
of the parks (Givoni 1991a). However, nearly all the larger parks investigated in this
study were built after the 1990s. These modern parks provided open lawns, and their
trees had not developed their optimal crowns yet because of their young age. The
presence of lawns and lower cover of trees, and especially mature trees, in these parks
lead to a vegetation structure which cannot provide a high value of microclimate
regulation.

The influence of park age was essentially mediated by the vegetation structure, since
it is self-evident that the relative proportion of mature trees is determined by the age
of the parks. All the parks in this study with medium proportions of mature trees were
older parks built prior to the 1990s, such as Children’s Park, Lao Dong Park and Ge

146
Ming Park (see Tables 3.1 and Figure 7.4). Trees typically require 20-30 years to grow
to a fully functional size. As such, older parks usually have more large trees, thereby
enhancing regulating services. Similarly, an investigation of park trees in Bangalore,
India, showed that park history had an impact on the distribution of trees (Henrique
Andrade and Vieira, 2007). Unsurprisingly, the size of trees (diameter at breast height
and height) in older parks (built in the 1970s) was significantly greater than of those
established in more recent years (built in 1990s). Therefore, older parks are likely to
provide stronger air quality and air temperature regulation than newer parks.

Accordingly, it would follow that the parks located within the second ring road, near
the city center, very likely provided high values of regulating services (see Table 7.2).
In reality, this relationship was not mediated by the distance to the city center, but
rather by park age, and hence vegetation structure types. Accompanying urbanization,
Xi’an city continued to expand outwards. Therefore, parks located in or near the core
city areas were constructed prior to the more distant parks (e.g. they were built before
the 1990s as opposed to later), and thus had sufficient time to form a large canopy
cover of trees and mature trees.

The value of microclimate regulation is the basis of the assessment of air pollution
abatement. The majority of parks showed correlating values of microclimate
regulation and air pollution removal. However, there were still some differences
between them, which were derived from the differences in the cover of evergreen
species (see Table 7.1). For example, Xi Qu Da Guan Yuan showed a low value of
microclimate regulation, but had a medium value of air pollution removal, due to the
medium cover of evergreen species. Children’s Park and Lao Dong Park showed high
values of microclimate regulation but medium values of air pollution removal,
because these two parks had a low proportion of evergreen species. The results thus
revealed that evergreen species are an important factor which influences air
purification. It therefore follows that if there is a lower cover of evergreen species, the
parks that provide medium to high values of microclimate regulations will provide
lower values of air pollution removal. A balanced proportion of evergreen species is
thus a useful method to promote the reduction of air pollution.

From the perspective of the respondents, it can be seen that more individuals
positively assessed microclimate than air quality. As described in the theoretical
chapter, microclimate regulation in this study mainly focused on the decreased air-
and surface temperature in summer. The removal of air pollution included the
absorption, decomposition and settlement of particulate and gaseous pollutions. In
view of the severe levels of air pollution in Xi’an, which greatly exceed the norms
147
prescribed by the World Health Organization (Chai et al. 2014), it is entirely possible
that the majority of park visitors were not able to clearly perceive the air improvement
induced by green spaces. This may be not only due to the small effect in light of the
great amount of pollution, but also because many air pollutions cannot be observed
directly. The respondents may thus primarily consider the limited improvement of air
quality that is directly observable, such as the deficiency of fresh air everywhere,
especially with consideration of children’s health. Therefore, a majority assessed the
air quality negatively. By contrast, most respondents can feel that the air is cooler in
parks than in roadsides or in buildings without air conditioning.

From the respondents’ viewpoint, older parks located in and near the city center and
with relatively lower cover (26-50%) of multi-layered vegetation and evergreens had
higher regulating services. This assessment was in agreement with the results of
indicator assessment, and therefore revealed that the respondents can indeed perceive
some ecosystem services, such as microclimate regulation. As for the influence of
vegetation structure, the respondents may consider that the thermal comfort of parks
is always connected to the shade of trees. Accordingly, the positive assessment of
microclimate regulation increased with increased cover of trees and mature trees,
though there were no significant associations between them. In fact, a majority of the
22 investigated parks had over 50% of tree cover. The relatively high cover of trees
and the limited variations among parks may have prevented the respondents from
perceiving the differences of microclimate regulation among parks. In addition,
respondents likely could not perceive the microclimate benefits derived from
increased multi-layered vegetation and evergreens, although a higher percentage of
multi-layered vegetation will produce more biomass and higher transpiration, and
hence provide measurably better temperature regulation (Armson et al. 2012).

Moreover, large shaded areas in summer are mainly provided by mature broadleaf
trees (Armson et al. 2012). However, the number of evergreen species in the parks of
Xi’an city is very limited, usually including a few types of trees and shrubs, such as
species from the family Pinaceae, and broadleaf evergreens such as M. grandiflora,
Ligustrum lucidum, Photinia serrulata and Buxus sinica (Xu and Zhang 2006; Zhao et
al. 2004). The leaf forms and crown shapes of these evergreen species normally do
not provide a large canopy. Shrubs cannot provide shading either. Therefore, the
respondents may think that a large cover of evergreens does not contribute much to
cooling.

In summary, indicator assessment system implies that vegetation structure is the key
factor deciding the regulating services of parks. The highly valued parks usually had a
148
higher cover of trees, and especially mature trees and evergreens. However,
vegetation characteristics are very local and context-dependent. It is thus difficult to
compare the results with studies from other countries. In addition, park age directly
influenced the vegetation structure, since trees typically need a long time to form their
optimal crown.

In this study, the value of regulating services was assessed without complicated
procedures. While very practical, this kind of indicator assessment has some
shortcomings. The types of vegetation structure that were used as indicators to assess
the value of parks’ regulating services concentrated on the cover, layering and
composition of vegetation, but omitted further details. However, the regulating
impacts of parks depends not only on the vegetation structure such as tree cover, plant
density and length of in-leaf season, but also on the surrounding conditions such as
built-up areas and concentration of pollutants (Nowak et al. 2006; Yin et al. 2011).
Built-up areas generate warm and unstable air (Andrade and Vieira 2007), and hence
require more transpiration from nearby parks to counteract their negative effects.
Since pollution is caused mainly by traffic emissions and is attenuated with distance,
parks that are located near major roads are exposed to more serious air pollutants due
to heavier traffic flows (Yin et al. 2011).

In addition, this study attempted to explore the associations between park area / park
age / park location and the provisions of regulating services. However, the locations
of parks were only divided by the distance to the city center along the ring roads, so
that the surrounding circumstances, such as the distance to main roads and heavy
traffic, were not considered. In the future, the exploration of more detailed
information should focus on the interactions between specific regulating services (e.g.
cooling effects and reduction of air pollution) and park characteristics (e.g. the
biomass structure in the parks and the surrounding land use of parks).

Lastly, the respondents’ assessments can be used as evidence of the improvement of


regulating services. However, among all of the 22 investigated parks, only six
received more than twenty questionnaires (see Table 3.5). The small numbers of
questionnaires at the individual park level thus limits the accuracy of the statistical
analysis needed to evaluate the respondents’ assessments.

149
7.3.2 Discussion of the Assessment of Recreational Services
7.3.2.1 Comparison of the Assessment of Recreational Services

with Similar Studies


Generally, all of the 22 parks were considered to provide low values of recreational
services as a whole. Among the specific items of recreational services, scenery was
positively assessed by no more than half of the respondents in this study. By contrast,
the study of Özgüner and Kendle (2006) noted that scenery of green spaces was
positively assessed by the majority of respondents (more than 80%) at investigated
sites in the UK. In Hong Kong, positive appraisals of scenic beauty in green spaces
exceed 52.6% (Shan 2009). Comparisons therefore show that the positive assessment
of scenery is lower in Xi’an than in other cities, including another Chinese city - Hong
Kong. The scenery in the parks of Chinese cities is typically artificial (Qiu 2009).
From the perspective of respondents’ needs in this study (see Chapter 6.6), there were
stronger desires for increased nature, peaceful atmosphere and shade in parks.
Therefore, there is an urgent need to improve the scenery of parks in an effort to meet
the citizens’ requirements.

The positive assessments of the number of sporting- and resting facilities were
relatively low. Similarly, Shan (2009) and Lo and Jim (2011) both found that resting
and sporting facilities received relatively low positive assessments in Hong Kong. In
fact, over half of the respondents considered sporting and resting facilities (e.g. seats
and pavilions in green spaces) to be insufficient (Lo and Jim 2011; Shan 2009). These
deficiencies reflect the inadequate per-capita share of urban green space areas
compared to the high population densities in modern megacities. While park visitors
tend to seek recreation (e.g. peace, relaxation, beauty and exercise) in the green
spaces, the design of green spaces has not notably focused on fulfilling this
requirement. Furthermore, there are insufficient numbers of playgrounds in the
majority of parks and some parks have no special playgrounds for visitors (see Table
4.1). The playgrounds in the parks of Chinese cities are the main places that provide
many kinds of entertainment facilities, such as carousels, Ferris wheels and miniature
trains. These entertainment facilities are always based on charging a fee. However,
park users can use any safe and open spaces to play games. This study suggests that
sufficient and suitable resting and sporting facilities and playgrounds should be
considered during the design process of urban parks to enhance the parks’ use by
citizens in their daily lives, though this is a difficult issue when faced with a huge
population and a relative shortage of green spaces.

150
7.3.2.2 The Influence of Park Variables on the Assessment of

Recreational Services
The small parks (0-5 ha) built prior to the 1990s and located inside the second ring
road, with higher cover of vegetation, water and evergreens, but lower cover of trees
and mature trees, typically elicited slightly positive responses on the specific items of
recreational services. The older parks near the city core areas usually encompassed
higher volumes of established sporting and resting facilities and playgrounds,
compared to the newer parks. The older parks also elicited more satisfaction with
scenery. This may be because the appearances of plant communities in older parks
had enough time to reach balance and optimal size. Moreover, according to the green-
space plans, the provisions of many kinds of recreational facilities were more stressed
in the last versions of the plans. Therefore, within this context, park designers paid
more attention to providing playgrounds and sporting areas. Another factor is that the
older parks in Xi’an city usually covered smaller areas, and smaller parks received
more positive assessments of the number and location of sporting facilities, resting
facilities and playgrounds. This may indicate that smaller areas can also provide
satisfactory recreational services, from the respondents’ perspective.

Moreover, parks with areas of 5-10 ha usually received relatively low positive
assessments of recreational services, especially regarding the number of playgrounds,
sporting- and resting facilities. The parks with areas of 5-10 ha are residential parks
according to the classification of Chinese parks. The results therefore may indicate
that the designers of residential parks lack a consciousness of the importance of the
recreational services of parks and therefore did not stress the provision of recreational
facilities.

According to the respondents’ assessment of recreational services, the cover of


vegetation, multilayered vegetation, water and evergreens were positive factors. Thus,
increasing the coverage of vegetation, water and evergreen species will increase the
recreational satisfaction, especially with the scenery. However, a higher cover of trees
and mature trees decreased the respondents’ recreational satisfaction. The respondents
thus may prefer to enjoy scenes that have abundant layers and are abundantly green
during all four seasons, but prefer to rest and do sports in open spaces. Large crowns
of trees and mature trees, such as woodlands, may interrupt the sight and sunlight.
Therefore, multi-layered parks with some evergreen species were preferred by the
respondents.

151
In general, this study provides a great deal of information on how to improve the
recreational services of urban parks in Xi’an. However, due to the length limits of the
questionnaire, detailed questions were not inquired, such as what types of scenery,
sporting facilities and resting facilities were preferred by the respondents and what are
the respondents’ recreational preferences in individual parks.

152
8 Final Discussion and Conclusions
Discussion of the Results
In order to assess the ecosystem services of urban parks in Xi’an, this study developed
a rapid non-monetary assessment method encompassing questionnaires and indicator
assessments. The assessed indicators included vegetation, water, multi-layered
vegetation, trees, mature trees and evergreen species. The rating rules of the indicators
were composed of scales based on indicator proportions and simple aggregations.
This simple method can be understood without the help of professionals. All the data
needed for this method can be easily collected in different study areas. In China, few
attempts have been made to use such methods, giving this study a certain pioneering
character. Additionally, few public comments have been taken into consideration.
Therefore, this method can be applied to other Chinese cities.

Using indicator assessments based on the types of vegetation structure, over half of
the parks were found to have a medium value of regulating services. More parks
showed high values of microclimate regulation than air quality regulation. Using
questionnaires, it was found that the respondents considered that urban parks have a
higher impact on microclimate regulation than on air quality improvement. Therefore,
indicator assessments and respondents’ assessments showed accordant results
regarding regulating services. However, respondents considered that urban parks
provided a low value of recreational services. Thus, the urban parks of Xi’an city as a
whole provided a higher value of regulating services than recreational services.

Vegetation structure was an important determinant of both the respondents’ subjective


assessments and objective indicator assessments. Among the 22 investigated public
urban parks, most had a limited number of types of land cover and vegetation
structure, in conjunction with a large cover of impervious and less cover of mature
trees and evergreens. This led to fewer parks having high-value ecosystem services,
and many respondents wanted more ecosystem services in the parks. The vegetation
characteristics and appearances were designed during the first stage of park
construction, based on the directions of the master plan and the documentation of the
green space plan. The older parks, providing special playgrounds and large areas of
pavements, revealed a focus on recreational services during their time of construction.
In fact, in the green spaces plans from the 1980s and 1990s, recreational services and
playgrounds were paid more attention, while in the later versions of the plans the
multifunctional green spaces and the sustainable development of the environment
were brought into the foreground.
153
The green space plan of Xi’an city stressed the importance of recreational services
and pointed out the necessity to provide recreational facilities, sporting facilities and
playgrounds in all types of urban parks. However, the fact that the respondents were
not satisfied with the recreational services of urban parks may indicate that it is not
easy to achieve the goals set out for recreational services within the limited surface
area of inner-city urban parks, especially in situations where urban parks are supposed
to provide regulating services as well.

In the last two decades, the obviously increased number and area of urban parks
demonstrate the attention paid to the development of green spaces by the government
of Xi’an city. However, the very tense land use pressures limited the quantity and area
of green spaces, and hence the speed of green-space development, compared with
other cities. In Xi’an city, urban green spaces usually are built on vacant land or
around historical relics, as stated in the master plans. At the city level, the southern
area, and especially the southeast area, has a greater number and total area of parks
than the other areas (see Figure 3.1). The unbalanced distribution of land resources
between the southern and northern areas directly led to an unbalanced distribution of
urban parks. In addition, over half of the urban parks have smaller areas of less than
10 ha. These smaller parks are always the older parks located within the second ring
road. The other twelve parks, and especially the bigger ones, were built after the year
2000 and are located between the third and second ring roads. The older parks had
various vegetation layers and compositions, with higher cover of multi-layered
vegetation, trees, mature trees, and evergreens. Therefore, the urban parks in Xi’an
city have unbalanced qualities. Although the number and areas of urban parks
decreased toward the city center (Table 3.1), individual parks near the central areas
showed increased provisions of regulating services and recreational services.

The limited area of urban parks was therefore not in conflict with the provision of
ecosystem services in the parks investigated in this study. The small and medium-
sized parks provided higher regulating and recreational services, because these
smaller parks are usually older and hence have large tree canopies, abundant
vegetation layers, and special playgrounds. By comparison, although larger parks
have relatively larger areas of lawns and water, they could not provide a high value of
ecosystem services, at least in part due to their much younger trees.

Among the socioeconomic variables assessed in this study, the visitors’ age was a
significant factor. Older visitors were the predominant users of urban parks. They
often visited more often and stayed longer in green spaces than the other visitors.
They positively took part in activities in green spaces and held different perceptions,
154
assessments and demands for green spaces than the other age groups. It is easy to
understand the obvious differences between the younger and the older respondents,
due to the physical limits and educational backgrounds of the senior citizens. These
findings suggest that future green-space plans and designs should pay more attention
to the older age group. Their use habits, perceptions and demands should be a priority,
since they are the most active users and have different demands for green spaces.

In addition, park age and location influenced the respondents’ assessments and
demands for green spaces. Respondents from the older parks and the parks located
within the second ring road expressed more positive assessments and anticipations for
ecosystem services. As discussed above, the older parks and the parks near the inner-
city areas definitely had different types of land cover and vegetation structures than
the other parks. These parks often had a higher cover of various vegetation structures
and playgrounds, and hence produced higher values of both regulating and
recreational services. The results demonstrated that respondents’ positive assessments
can reveal the objective basis of the supply of ecosystem services, and their
suggestions can therefore be used to plan green spaces.

Considering the influence of vegetation structure, it was found that respondents


preferred to stay longer in the parks with 51-75% of multi-layered vegetation and
most frequently used the parks with 26-50% of trees and evergreens. They preferred
to do sports and relax in the parks with 26-50% cover of multi-layered vegetation and
evergreens. Moreover, respondents highly perceived the importance of ecosystem
services in the parks which were covered by 76-100% of multi-layered vegetation and
trees, as well as those with 51-75% of mature trees and evergreens. They also tended
to assess the ecosystem services more positively in the parks with 76-100% of trees
and 51-75% cover of multi-layered vegetation and evergreens. At the same time, they
expressed stronger demands for individual items of ecosystem services in the parks
with 26-50% of multi-layered vegetation.

These findings showed that respondents’ use habits and recreational activities, as well
as their perceptions, assessments and demands for ecosystem services of green spaces
were connected to different vegetation structures. However, it is difficult to find
rational explanations for these connections. Nevertheless, a high cover (76-100%) of
trees consistently had a positive influence on respondents’ perceptions, assessments
and demands. The results may indicate that the respondents can intuitively understand
the benefits derived from a high tree cover.

155
Generally, the links among these findings are more complex. For example,
respondents assessed the scenery as slightly positive in the parks with higher cover of
vegetation and multi-layered vegetation, and expressed more demands for flowers in
the parks with a higher cover of trees. This may indicate that respondents prefer
scenery with layered vegetation, and the establishment of flowerbeds or scattered
populations of wildflowers should be stressed in the parks that have a high cover of
trees.

In the parks with higher cover of multi-layered vegetation, trees and evergreens,
respondents perceived the importance of shading as high. They also expressed more
need for shading in the parks with a higher cover of trees and mature trees. Such
findings may indicate that shading is a very important service from the respondents’
perspective, even in the parks that already have enough shading. However, the
respondents’ recreational activities were not influenced by shading (tree and mature
tree cover). This may indicate that, compared to the recreational facilities, shading is
not an important factor influencing the respondents’ activities in green spaces.

In addition, the respondents generally voiced more positive assessments and more
demands in the older parks located within the second ring road. This is likely because
in those parks, the respondents were satisfied with the ecosystem services, such as
vegetation types, scenery, and vegetation cover, but still anticipated better
management and more safety.

Suggestions for Improving the Green Spaces of


Xi’an
The results presented in chapters 4, 5, 6 and 7 can answer all the research questions
proposed in Chapter 2. The assessment of ecosystem services generated by urban
parks in Xi’an has enabled the recognition and understanding of the importance of
vegetation within urban areas. The questionnaires enhanced the residents’ awareness
of the value of urban green spaces and encouraged them to take part in urban greening
activities. This study has also demonstrated that the types of vegetation structure not
only directly influence the provisions of regulating services, but also significantly
influence the respondents’ preferences for individual recreational activities, as well as
their use habits, perceptions, assessments and demands for urban parks. These
research findings thus contribute to the understanding of the multi-functionality of
urban green spaces and enable their better planning and design. From the perspective
of planning approaches, the study provides a possibility to incorporate concepts of
ecosystem services into the design of individual green spaces and green-space
156
structure planning.

Based on the results of this study, several general strategies can be proposed to sustain
urban green spaces and thus maximize their ecosystem services, not only at the local
level but also at city level:

(1) Maintaining Sufficient Green Spaces Requires Limiting the Land Use
At the scale of the whole city, urban planning can play a key role in establishing
strategies to conserve existing green spaces, develop new green spaces, and connect
isolated fragments (Lovell and Johnston 2009). Compared to other countries, Chinese
cities are typically very densely populated and crowded with buildings and roads. For
example, in Xi’an city, only Wen Jing Park and City Sport Park are located in the
northern areas between the second and third ring roads (see Table 3.1). Clearly, it is
impossible to rely on these two parks to satisfy the respondents’ needs for recreational
activities and environmental regulation in the surrounding areas. The government
should thus plan more green areas during the urban development of this district in the
future. There are still potential places, even if few, for green spaces. It is feasible to
transform abandoned lands and closed factories into green spaces and to stipulate the
greening of lands in the renewal of old districts. Furthermore, it is possible to green
all the scattered public open lands such as backyards, car parks, squares and streets.
Transforming such lands to other uses should be prohibited.

(2) To Maintain the Number of Vegetation Structure Types and Plant Species
Various types of land cover and vegetation structure could contribute more
environmental benefits. For example, multi-layered vegetation could increase the
stability of plant communities and maintain multi-aged vegetation structures to
provide continuous ecosystem services over time. However, current urban green
spaces in Xi’an are characterized by simple land cover and vegetation structure types.
Overall, urban parks are dominated by simply-structured vegetation, pavements with
scattered vegetation, and plain pavements. Vegetation types were unevenly distributed
among the parks, and mainly included multi-layered trees covering shrubs and lawns,
trees covering shrubs and lawns, and lawns at the time of this study. By contrast,
evergreen broadleaf woodlands, shrubs, bamboos and flower beds were seldom found.
In the future, a broader range of vegetation structures, and especially natural
vegetation (e.g. grasslands, un-trimmed shrubs and trees) should be emphasized in the
design of green spaces.

The types of vegetation structure within the urban green spaces of Xi’an are not
optimal for microclimate regulation and air pollution removal, primarily due to their
157
low proportion of mature trees. Trees, and especially mature trees can provide large
canopies and hence large amounts of cool air in the summer. As discussed in the
studies of McPherson (1994), C.L (2002) and Yin et al. (2011), large trees can also
remove more pollutants from the air. In the studies of Jonsson (2004), Potchter et al.
(2006), Andrade and Vieira (2007), and Tzoulas and James (2010), the large cooling
effects of high and wide-canopied trees were demonstrated. In this study, trees in the
parks located inside the third ring road were much younger than those inside the
second ring road, because most of the green spaces located inside the third ring road
were built after the 1990s. This problem may simply take time to resolve by itself, as
the already planted trees inevitably mature and increase in size. However, the
vegetation composition of green spaces in Xi’an is not ideal for air pollution removal
due to the limited cover of evergreens. In some parks, e.g. Tang Yan Road Tang City
Wall Park and Mu Ta Si Park, evergreen species should be emphasized much more,
not only for the generation of regulating services, but also in order to provide green
scenery in winter.

Selecting appropriate species and configuring vegetation in patterns that are unique to
different functional purposes in urban parks could maximize their ecosystem services.
For future planning, the choice of vegetation, and especially trees, should be based on
their suitability for the urban environment and their ability to remove air pollutants.
The criteria used for tree selection should consider: (1) Evergreen trees have a higher
capacity for removing air pollutants due to longer foliage retention (Beckett et al.
1998; Nowak et al. 2006); (2) Fast growing trees usually can provide a surface for the
adsorption of air pollutants soon after establishment (Nowak and Crane 2002); (3)
The leaf characteristics will influence the deposition of air pollutants on the leaf
surface. It stands to reason that hairy, resinous, scaly, and coarse surfaces can capture
more particles than smooth leaves (Beckett et al. 1998); (4) If a tree is sensitive to
certain pollutants, it cannot be used at a site close to the source of such air pollutants
(Beckett et al. 1998); (5) Trees with high BVOC (biogenic volatile organic
compounds) and pollen emission rates should be avoided in planning to improve the
net air pollution reduction benefit of the green spaces (Haider 1996; Nowak 2000).

(3) To Balance the Provision of Regulating Services and Recreational Services


The design of ecosystem services in green spaces is a complex process. It should
consider the provision of different ecosystem services, characteristics of the local- and
surrounding environment, and different requests of users. In practice, the detailed
information of individual parks would be investigated beforehand, including location,
original vegetation and resources, main user groups, and surrounding environment.

158
In this study, parks that had a higher proportion of vegetation and water usually also
had high values of ecosystem services. Water can provide strong evaporation and thus
reduce the air temperature in summer. It also can provide the opportunity for many
kinds of activities such as boating and fishing. However, in many cases it is not
feasible to increase the amount of water bodies in parks because they are very costly
to construct, especially in the case of large artificial lakes, pools, streams or fountains.
Therefore, the construction of water bodies should depend on the local site conditions.
For example, a naturally low-laying site with an abundance of natural water sources
can be used to form a water body (Qiu 2009). Moreover, the costs of maintaining
water bodies are also obviously higher than those of vegetation (Qiu 2009). This may
explain why the water bodies in Xi’an are muddy and hence unsuitable for a lot of
possible water-related activities such as swimming and playing in the water.

It seems contradictory to simultaneously seek optimal regulating- and recreational


services. A higher cover of vegetation and water, multi-layered vegetation, trees,
mature trees and evergreens will increase the provision of regulating services.
However, from the respondents’ perspective, the recreational services were the most
satisfactory in the parks with a higher cover of vegetation, water and evergreens but
lower cover of trees and mature trees. From the perspective of the specific parks, we
can try to balance the regulating and recreational services with considerations of the
surrounding land-use types. For example, parks that are located in areas with higher
population densities will face severe pressures of visitors’ daily use. By incorporating
population statistics within the service areas of parks, a large number of residents can
be used as an indicator of a stronger need for recreational services. Therefore,
additional recreational facilities, such as benches, sporting facilities and playgrounds,
will be required to satisfy the residents’ daily needs. At the same time, these parks
should provide microclimate regulation and air purification. Shrub cover should be
reduced in order to create more useable spaces. Lawns and open soil can replace
impervious surfaces. The increased area of lawns and open soils not only provides
regulating services for air quality and temperature, but also offer spaces and fields for
play and other leisure activities. In order to satisfy the visitors’ use habits and
aesthetic preferences, a high cover of dense trees should be avoided, and sparse
woodlands can be an alternative. The thick foliage and underlying branches of mature
trees, especially at heights lower than 5 m, should be clipped to maintain open sight
lines and provide sufficient light. Resting and sporting facilities and playgrounds can
be placed under the tree canopies or near the trees. All these approaches can be used
to provide a balance of both recreational and regulating services in parks.

In addition, local environmental factors are always complex. For example, if a park is
159
located near main roads, more cover of evergreen species and multi-layered
vegetation, and even green belts, should be chosen in order to block air pollution and
reduce noise. However, the green area in question may be a small park located near
main roads in between high-density housing. It is thus difficult to answer the question
which ecosystem services should be the focus of the design in such cases. Park
designers should therefore consider which service is more important and urgently
needed. Based on the questionnaire results from this study, the respondents usually
prefer to use urban parks to enjoy scenery and greenery and to obtain a certain extent
of contact with nature. Therefore, recreational services should be planned and
emphasized first. Considering that the predominant users of urban parks in Xi’an are
the elderly and children, quiet and peaceful places for relaxation and play should be
designed, and designated playgrounds can be designed for children.

(4) To Incorporate Citizens’ Suggestions into the Process of Green Space Planning
The results of the questionnaire survey indicate that the residents of Xi’an have a
positive attitude towards urban green spaces and a widespread recognition of
ecosystem services generated by green spaces, which shows that there are potential
motives for public participation in urban greening projects. Future urban green spaces
urgently need to place a special emphasis on qualified recreational services. For
specific parks, user surveys should encourage more citizens to participate in the
process of planning, design and management of their often-used green spaces. More
detailed questions should be asked in the questionnaires, including the respondents’
preferred types of scenery, resting and sports facilities, and their concrete suggestions
for these items with details such as choice of plant species, plant configurations and
the location of resting and sporting facilities and playgrounds. For different user
groups, questions should be asked regarding what features of green spaces, e.g.
percentage of trees, vegetation types and scenery styles, could increase their use
frequencies and stay times?

(5) To Consider the Provision of Ecosystem Services at Individual-Park and City


Level
Parks in large cities do not exist individually. They are connected with each other and
with other types of green spaces, and hence form higher-order green space structures,
which are part of the green-space system of the city as a whole. Therefore, it is
necessary to consider the connection of different green spaces and to design
individual parks within the context of the entire structure of green spaces.

At the city scale, all types of green spaces should be connected through the systems of
main roads and streets. Urban parks provide ecosystem services for a large number of
160
citizens at the scale of an individual district or even the whole city. By contrast,
residential green spaces provide ecosystem services mainly at the scale of a single
neighborhood. Parks and residential green spaces usually have larger areas and hence
can be seen as green patches. These green patches are interconnected by street trees
and roadside green spaces to form a green-space system. The integrated system can
increase the provision of total ecosystem services if it is well-connected and
distributed throughout the city.

(6) To Develop a System for Better Management


Better management is critically important in order to increase the benefits provided by
existing green spaces in Xi’an. This means that it is necessary to select appropriate
species and maintain the health of the vegetation, as well as to maintain a suitable tree
density and complex vegetation structure. Existing trees should be preserved rather
than transplanted or removed as is currently common practice in Xi’an. Developers
should therefore be required to incorporate existing trees into development plans.

More financial support should be provided for the development and management of
urban green spaces. The shortage of financial support may lead to insufficient
development of urban green spaces and by extension an insufficient supply of
ecosystem services. It is therefore necessary for the government to increase the
greening budget to alleviate the chronic economic stresses in order to improve
ecosystem services supplied by urban green spaces. Moreover, more effective
institutions and regulations should be established to enable better management of
urban green spaces. Although a number of related legislations and ordinances have
been promulgated in Xi’an, there is a lack of enforcement. There is a
misunderstanding that urban green spaces are state-owned land resources and can be
used by government institutions without legal or administrative permission, and the
Xi’an Landscape Bureau has little legal or administrative power to protect urban
green spaces. A set of clear and equitable legal judgment mechanisms has to be
established to guide the scientific management of urban green spaces.

In summary, the design of urban green spaces should consider their specific purposes
such as environmental regulation, recreation, etc. An optimal quality of urban green
spaces could be achieved through the design of multifunctional urban landscapes
based on an appropriate vegetation structure, choice of species, and functional
divisions. Public involvement in urban greening projects could be a major driving
force, in addition to the political support, for the green structure planning of urban
green spaces in Xi’an. It would be helpful if the public could accurately assess the
ecosystem services. Finally, more opportunities should be provided for the residents
161
to become involved in the planning, maintenance and conservation of urban green
spaces.

Limits of This Study and Suggestions for Future


Research
8.3.1 Limits and Suggestions for the Non-monetary
Assessment Method
Traditional assessment methods used for regulating services usually rely on
mathematical models, which require complex processing not only during the
measurement and collection of indicators, but also during data computation. Moreover,
suitable models need to be established and adjusted beforehand. This study attempted
to develop a simple assessment method that can be used under the conditions of
limited resources such as limited time and finances, or a lack of professionally trained
personnel.

Generally, the assessment method of regulating services used in this study is based on
a range of indicators, indicator ratings and aggregations. The main advantage of this
method is its simple and rapid operation. The indicators included vegetation types,
vegetation composition and biological characteristics (e.g. evergreen or deciduous).
The values of the individual indicators were calculated visually based on their area of
coverage. Therefore, the vegetation information can be easily collected without the
need for complex processes or professional knowledge.

The indicator ratings used the ranges of individual indicators’ coverage based on their
positive influence on regulating services. The final assessment was aggregated using
simple addition of the rating results. The complete assessment process is therefore
simple and does not rely on complex models. However, the coarse rating rules could
not provide numerical results and assessments.

In the future, this assessment method needs further information to make it more
accurate while keeping it simple. The survey of detailed information related to
vegetation / environment of green spaces and plant species should be covered,
including the plant species, the height of the vegetation and the diameter at breast
height (DBH) of trees. These data would help to assess the ecosystem services more
accurately, since different plant species have different regulating ability of
temperature and air pollutants. The species, the height of the vegetation and the DBH
of trees can help to calculate the total leaf areas, and by extension the green biomass
162
quantitates, and can hence be used to determine the regulating services in a
quantitative way.

Furthermore, it is also possible to include environmental factors in the assessment of


ecosystem services provided by green spaces. The local population and the land use
types surrounding green spaces in a defined buffer zone, as well as the distance to the
main roads, are important factors that influence the local demand for ecosystem
services. Moreover, the density of population can influence citizens’ demands for
recreational services. The surrounding land use types can influence the balance of
ecosystem services of green spaces. For example, downtown areas need more cooling
from nearby green spaces, while industrial areas need more air purification by
neighboring vegetation. It will be beneficial to assess these environmental factors in
the future, in order to plan and design specific green spaces to provide maximal,
balanced ecosystem services.

8.3.2 The Limits and Suggestions for the Questionnaires


In spite of their simplicity, efficacy and usefulness, there are some deficiencies in the
questionnaires used in this study. Based on the questionnaire results, it is clear that
more thorough questionnaires including a range of further questions can be designed
in the future. Firstly, the number of questionnaire surveys in specific parks was
limited due to time constraints. In fact, the number of questionnaires in the majority of
individual parks was less than 20, and therefore lacked statistical weight. In future
studies, the questionnaires conducted in individual parks should encompass more
participants and more focused questions.

Secondly, more detailed questions should be asked in order to further investigate the
respondents’ perceptions and attitudes. Some general questions can be asked in any
type of green space. For example, respondents can always be asked if an increase of
the quantity and quality of water bodies, provision of clean air, more recreational
facilities, or a higher number of plants and animals will enhance their satisfaction with
the investigated green spaces. It is also possible to ask the respondents about their
preferred vegetation types and configurations. These questions could help park
designers to find out which type of scenery is welcomed. Some questions should help
to optimally distribute the sporting and resting facilities in green spaces, such as
asking about the respondents’ favorite sports, or where benches should be located.
Furthermore, questions aimed at specific green space design can be considered.

This study has demonstrated that different social groups, park characteristics and
vegetation structures lead to different user behaviors, as well as to different
163
assessments, perceptions and demands for ecosystem services of green spaces. In
order to satisfy different user groups, focused questions should be asked, such as what
is the respondents’ favorite type of green space. Answers can be offered in the form of
pictures, such as green spaces located away from roads, or spaces enclosed by high
trees or shrubs, etc. The answers to these questions can reveal the preferred types of
scenery in green spaces. Also, for individual parks, we can take pictures of different
vegetation structures and ask visitors which best represents their favorite scenery and
which types would provide better regulating services in their opinion.

8.3.3 Suggestions for Future Studies


This study is a pioneering attempt at exploring the multi-functionality of urban
ecosystems and assessing the ecosystem services provided by urban green spaces. As
such, it is limited to the assessment of regulating- and recreational services of green
spaces. Other ecosystem services, such as psychological benefits, water infiltration
and biodiversity, which are also important for the sustainable development of urban
ecosystems and the quality of urban life, were not included. In the future, more
comprehensive studies that include different kinds of ecosystem services generated by
urban green spaces could be conducted.

Moreover, additional integrated methods, such as using questionnaires and


observations to support or provide demonstrations, can be used in the future. These
approaches can help us to advance the provision of ecosystem services in useful ways.
For example, at the level of specific parks, we can ask users about their favorite
recreational activities in green spaces. On the other hand, we can observe or track
users’ physical activities directly at certain times (Cohen et al. 2011 and Fjortoft and
Sageie 2000), such as in the morning (7:00- 9:00) and evening (18:00-21:00). In order
to investigate the provision of recreational facilities, we can ask visitors about their
preferred places to have a rest in green spaces, and observe where visitors prefer to
stop and sit.

Last but not least, in order to achieve the aims surrounding the provision of optimal
ecosystem services by green spaces, and to cater to the different types of users, we
should consider not only the visitors’ socioeconomic backgrounds but also their use
groups. The user groups can be divided by their frequent activities in green spaces,
mainly into active (e.g. walking, walking a dog, exercising, jogging, cycling and
playing games) and passive users (e.g. breathing fresh air, reading, sightseeing,
relaxing, social contact and interaction, or enjoying and watching nature). Based on
this, the relationships between the respondents’ perceptions / assessments / demands
for ecosystem services of green spaces and their use groups can be analyzed. We can
164
therefore hypothesize that, for example, respondents who think that shading is the
most important service belong to user groups that prefer relaxing or reading,
respondents who want more recreational facilities belong to user groups who enjoy
exercising, and respondents who regard aesthetics as important or want more
vegetation belong to groups who prefer relaxing or sightseeing.

Taken together, this pioneering study thus demonstrates the feasibility of using simple
questionnaire- and mapping-based surveys to elucidate the provision and demand for
ecosystem services in urban green spaces, and paves the way for similar, but also
more detailed analyses in many Chinese cities in the future.

165
Bibliography
Aarts, M.J., Vries, S.I.D., Oers, H.A.V., & Schuit, A.J. (2012). Outdoor play among
children in relation to neighborhood characteristics: a cross-sectional neighborhood
observation study. International Journal of Behavioral Nutrition & Physical Activity, 9, 1-11
ADB (2003 ). SUMMARY ENVIRONMENTAL IMPACT ASSESSMENT - XI'AN
URBAN TRANSPORT PROJECT. In, Technical Assistance to the People's Republic of China
for Preparing the Xi'an Urban Transport Project (p. 31). Xi'an
Akbari, H., Pomerantz, M., & Taha, H. (2001). Cool surfaces and shade trees to reduce
energy use and improve air quality in urban areas. Solar Energy, 70, 295-310
Akbari H., D.S., Dorsano S., Huang J. and Winnett, S. (1992). Cooling our communities: A
guidebook to tree planting and light colored surfacing. U.S. E.P.A.,Office of Policy Analysis,
Climate Change Division
Aminzadeh, B., & Ghorashi, S. (2007). Scenic landscape quality and recreational activities
in natural forest parks, Iran. International Journal of Environmental Research, 1, 5-13
Amiri, R., Weng, Q., Alimohammadi, A., & Alavipanah, S.K. (2009). Spatial–temporal
dynamics of land surface temperature in relation to fractional vegetation cover and land
use/cover in the Tabriz urban area, Iran. Remote Sensing of Environment, 113, 2606-2617
Andrade, H., & Vieira, R. (2007). A climatic study of an urban green space:the Gulbenkian
Park in Lisbon (Portugal). Henrique Andrade e Rute Vieira, 84, 46
Armson, D., Stringer, P., & Ennos, A. (2012). The effect of tree shade and grass on surface
and globe temperatures in an urban area. Urban Forestry & Urban Greening, 11, 245-255
Arnold, J.C.L., & Gibbons, C.J. (1996). Impervious surface coverage, the emergence of a
key environmental indicator. Journal of the American Planning Association, 62, 15
Ayman Hassaan Ahmed, M. (2011). Analysis of the microclimatic and human comfort
conditions in an urban park in hot and arid regions. Building and Environment, 46, 2641-2656
Balram, S., & Dragicevic, S. (2005). Attitudes toward urban green spaces: integrating
questionnaire survey and collaborative GIS techniques to improve attitude measurements.
Landscape and Urban Planning, 71, 147-162
Barradas, V. (1991). Air temperature and humidity and human comfort index of some city
parks of Mexico City. International Journal of Biometeorology, 35, 24-28
Batista, M.H., & Haertel, V. (2010). On the classification of remote sensing high spatial
resolution image data. International Journal of Remote Sensing, 31, 5533-5548
Baycan-Levent, T., R. Vreeker, & Nijkamp, P. (2009). A Multi-Criteria Evaluation of Green
Spaces in European Cities. European Urban and Regional Studies, 16, 20
Bayer, P., Keohane, N., & Timmins, C. (2009). Migration and hedonic valuation: The case
of air quality. Journal of Environmental Economics and Management, 58, 1-14
Beckett, K.P., Freer-Smith, P.H., & Taylor, G. (1998). Urban woodlands: their role in
166
reducing the effects of particulate pollution. Environmental Pollution, 99, 347-360
Bell, S., Morris, N., Findlay, C., Travlou, P., Montarzino, A., Gooch, D., Gregory, G., &
Ward Thompson, C. (2004). Nature for People: The Importance of Green Spaces to East
Midlands Communities. In
Berg, P.G., Granvik, M., & Hedfors, P. (2012). Functional Density: A conceptual
framework in a townscape areas context
Bjerke, T., Østdahl, T., Thrane, C., & Strumse, E. (2006). Vegetation density of urban parks
and perceived appropriateness for recreation. Urban Forestry & Urban Greening, 5, 35-44
Blair, R.B. (1999). BIRDS AND BUTTERFLIES ALONG AN URBAN GRADIENT:
SURROGATE TAXA FOR ASSESSING BIODIVERSITY? Ecological Applications, 9, 164-
170
Bolger, D.T., Scott, T.A., & Rotenberry, J.T. (2001). Use of corridor-like landscape
structures by bird and small mammal species. Biological Conservation, 102, 213-224
Bolund, P., & Hunhammar, S. (1999). Ecosystem services in urban areas. Ecological
Economics, 29, 8
Bonan, G.B. (1997). Effects of Land Use on the Climate of the United States. Climatic
Change, 37, 449-486
Bonsignore, R. (2003). The Diversity of Green Spaces. In, Design Brief No. 2: Design
Center for American Urban Landscape, University of Minnesota, Minneapolis, MN
Bowen, G.A. (2009). Document Analysis as a Qualitative Research Method. Qualitative
Research Journal, 9, 27-40
Buchel, S., & Frantzeskaki, N. (2015). Citizens’ voice: A case study about perceived
ecosystem services by urban park users in Rotterdam, the Netherlands. Ecosystem Services,
12, 169-177
Burgess, J., S. M. Hassison, and M. Limb (1988). People, parks and the urban green: A
study of popular meanings and values for open spaces in the city. Urban studies, 25, 18
Buyadi, S., Mohd, W., & Misni, A. (2014a). The Effects of Vegetation Growth on the
Microclimate of Urban Area: A Case Study of Petaling District. In A. Abdul Rahman, P.
Boguslawski, F. Anton, M.N. Said, & K.M. Omar (Eds.), Geoinformation for Informed
Decisions (pp. 151-161): Springer International Publishing
Buyadi, S., Mohd, W., & Misni, A. (2014b). Impact of vegetation growth on urban surface
temperature distribution. In, IOP Conference Series: Earth and Environmental Science (p.
012104): IOP Publishing
C.L, B. (2002). Pollution mitigation and carbon sequestration by an urban forest.
Environmental Pollution, 116, Supplement 1, S195-S200
C.S.B. Grimmond, & T.R.Oke (1991). An evapotranspiration-interception model for urban
areas Water Resourses Research, 27, 16
Cardelino, C.A., & Chameides, W.L. (1990). NATURAL HYDROCARBONS,
URBANIZATION, AND URBAN OZONE. J. Geophys. Res., 95, 13971-13979
167
Cardinale, B.J., Duffy, J.E., Gonzalez, A., Hooper, D.U., Perrings, C., Venail, P., Narwani,
A., Mace, G.M., Tilman, D., Wardle, D.A., Kinzig, A.P., Daily, G.C., Loreau, M., Grace, J.B.,
Larigauderie, A., Srivastava, D.S., & Naeem, S. (2012). Biodiversity loss and its impact on
humanity. nature, 486, 59-67
Carlos Smaniotto Costa, Daniela Morgenstern, Hans Kasperidus, Juliane Mathey, Kathleen
Thiele, Maria Caserio, Nick Grayson, Richard Coles, Sandra Wohlleber, Sigrun Farcher, &
Rößler, S. (2004). MAKING GREENER CITIES:A PRACTICAL GUIDE. In, URGE –
Development of Urban Green Spaces to Improve the Quality of Life in Cities and Urban
Regions (p. 120). Leipzig, Germany
Casado-Arzuaga, I., Madariaga, I., & Onaindia, M. (2013). Perception, demand and user
contribution to ecosystem services in the Bilbao Metropolitan Greenbelt. Journal of
Environmental Management, 129, 33-43
Casper, J.M., Harrolle, M.G., & Kelley, K. (2013). Gender differences in self-report
physical activity and park and recreation facility use among Latinos in Wake County, North
Carolina. Annals of Behavioral Medicine, 45, 49-54
Castro, A., Martín-López, B., García-Llorente, M., Aguilera, P., López, E., & Cabello, J.
(2011). Social preferences regarding the delivery of ecosystem services in a semiarid
Mediterranean region. Journal of Arid Environments, 75, 1201-1208
Cavanagh, J.-A.E., Zawar-Reza, P., & Wilson, J.G. (2009). Spatial attenuation of ambient
particulate matter air pollution within an urbanised native forest patch. Urban Forestry
&amp; Urban Greening, 8, 21-30
Chai, F., Gao, J., Chen, Z., Wang, S., Zhang, Y., Zhang, J., Zhang, H., Yun, Y., & Ren, C.
(2014). Spatial and temporal variation of particulate matter and gaseous pollutants in 26 cities
in China. Journal of Environmental Sciences, 26, 75-82
Chameides, W., Lindsay, R., Richardson, J., & Kiang, C. (1988). The role of biogenic
hydrocarbons in urban photochemical smog: Atlanta as a case study. Science, 241, 1473-1475
Che Huizheng, Z., et al (2005). Climate change in the latest fifty years in Xi'an city, China
Journal of Arid land resouce and environment, 19, 4
Chen, F., & Pang, L. (2013). Investments and Consumes of Sporting Facilities between
City and Country SHANDONG SOCIAL SCIENCES, 166-169
Chen, H., & Zhang, X. (2005). The status and improvement of plant configuration in
landscape architecture in Xi'an Acta Agriculturae Boreali-occidentalis Sinica, 14, 2
Chen, X.-L., Zhao, H.-M., Li, P.-X., & Yin, Z.-Y. (2006). Remote sensing image-based
analysis of the relationship between urban heat island and land use/cover changes. Remote
Sensing of Environment, 104, 133-146
Chen, Y. (1990). Dendrology for Landscape architecture. Beijing: Chinese Forestry Press
Chen, Y.W. (2006). Assessing the services and value of green spaces in urban ecosystem: a
case of Guangzhou City. In. Hong Kong: The University of Hong Kong
Chi-Ru Chang, Ming-Huang Li, & Chang, S.-D. (2007). A preliminary study on the local
168
cool-island intensity of Taipei city parks. Landscape and Urban Planning, 80, 386-395
Chiesura, A. (2004). The role of urban parks for the sustainable city. Landscape and Urban
Planning, 68, 129-138
Cianga, N., & Popescu, A. (2013). Green Spaces and Urban Tourism Deveopment in
Craiova Municipality in Romania. European Journal of Geography, 4, 34-45
CMC (2007). Guidelines for Urban Green (space) System Planning (trial edition:2002) In
Ministry of Housing and Urban-Rural Development P.R. China (Ed.)
Cohen, P., Potchter, O., & Schnell, I. (2014). The impact of an urban park on air pollution
and noise levels in the Mediterranean city of Tel-Aviv, Israel. Environmental Pollution, 195,
73-83
Coles, R.W., & Bussey, S.C. (2000). Urban forest landscapes in the UK — progressing the
social agenda. Landscape and Urban Planning, 52, 181-188
Cornelis, J., & Hermy, M. (2004). Biodiversity relationships in urban and suburban parks in
Flanders. Landscape and Urban Planning, 69, 385-401
Costanza, R., D'Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K.,
Naeem, S., O'Neill, R.V., Paruelo, J., & others (1997). The value of the world's ecosystem
services and natural capital. nature, 387, 253-260
Costanza, R., Fisher, B., Mulder, K., Liu, S., & Christopher, T. (2007). Biodiversity and
ecosystem services: A multi-scale empirical study of the relationship between species richness
and net primary production. Ecological Economics, 61, 478-491
Costanza, R., Wilson, M.A., Troy, A., Voinov, A., Liu, S., & D'Agostino, J. (2006). The
value of New Jersey's ecosystem services and natural capital
Cui, D. (2011). Green Space System's Spatial Pattern and Ecosystem Service in The Center
Area of Xi'an City In, Urban Planning (p. 87). Xi'an: Xi'an Architecture & Technology
University
Dansereau, P. (2011). Description and Recording of Vegetation Upon a Structural Basis.
Ecology, 32, 57
Davidson, M.D. (2013). On the relation between ecosystem services, intrinsic value,
existence value and economic valuation. Ecological Economics, 95, 171-177
Dawes, R.M. (1972). Fundamentals of Attitude Measurement. New York: Wiley
De Groot, R.S., Wilson, M.A., & Boumans, R.M.J. (2002). A typology for the
classification, description and valuation of ecosystem functions, goods and services.
Ecological Economics, 41, 393-408
Derkzen, M.L. (2012). Experiencing the Urban Green Space - An exploratory study of
visiting behaviour, perceptions and preferences in the urban green spaces of São Paulo,
Brazil. Open Access
Di Gregorio, A., Jansen, L.J.M., (2000). Land Cover Classification System(LCCS):
Classification Concepts and User Manual. In
Dunnett, N., Swanwick, C., Wooley, H., (2002). Improving Urban Parks, Play Areas and
169
Green Spaces. In. London, England: Department for Transport
Fisher, B., Turner, R.K., & Morling, P. (2009). Defining and classifying ecosystem services
for decision making. Ecological Economics, 68, 643-653
Forsyth, A. (2003). People and Urban Green Areas: Perception and Use. In, Design Brief
No. 4,. Design Center for American Urban Landscape, University of Minnesota,
Minneapolis,MN.
Fowler, D., Cape, J.N., Unsworth, M.H., Mayer, H., Crowther, J.M., Jarvis, P.G., Gardiner,
B., & Shuttleworth, W.J. (1989). Deposition of Atmospheric Pollutants on Forests [and
Discussion]. Philosophical Transactions of the Royal Society of London. Series B, Biological
Sciences, 324, 247-265
Fuller, R.A., Irvine, K.N., Devine-Wright, P., Warren, P.H., & Gaston, K.J. (2007).
Psychological benefits of greenspace increase with biodiversity. Biology letters, 3, 390-394
Gómez-Baggethun, E., & Barton, D.N. (2013). Classifying and valuing ecosystem services
for urban planning. Ecological Economics, 86, 235-245
Gallagher, M.W., Beswick, K.M., Duyzer, J., Westrate, H., Choularton, T.W., &
Hummelshøj, P. (1997). Measurements of aerosol fluxes to speulder forest using a
micrometeorological technique. Atmospheric Environment, 31, 359-373
Gao, T., Qiu, L., Hammer, M., & Gunnarsson, A. (2012). The importance of temporal and
spatial vegetation structure information in biotope mapping schemes: a case study in
Helsingborg, Sweden. Environmental Management, 49, 459-472
Gardiner, M.M., Burkman, C.E., & Prajzner, S.P. (2013). The value of urban vacant land to
support arthropod biodiversity and ecosystem services. Environmental Entomology, 42, 1123-
1136
Garrod, G., & Willis, K.G. (1999 ). Economic valuation of the environment: Methods and
case studies. Cheltenham, UK: Edward Elgar
Georgi, J.N., & Dimitriou, D. (2010). The contribution of urban green spaces to the
improvement of environment in cities: Case study of Chania, Greece. Building and
Environment, 45, 1401-1414
Gerd, B., Wänke, M., (2002). Attitudes and Attitude Change. New York: Psychology Press
Gilbert, N.L., Woodhouse, S., Stieb, D.M., & Brook, J.R. (2003). Ambient nitrogen dioxide
and distance from a major highway. Science of The Total Environment, 312, 43
Gilbert, O.L. (1989). The ecology of urban habitats.
Gill, S.E., Handley, J.F., Ennos, A.R., & Pauleit, S. (2007). Adapting cities for climate
change: The role of the green infrastructure. Built Environment, 33, 115-133
Gillespie, R.D.B.a.T.J. (1995). Microclimatic landscape design: creating thermal comfort
and energy efficiency In (p. 177). New York John Wiley & Sons, Inc.
Givoni, B. (1991a). IMPACT OF PLANTED AREAS ON URBAN ENVIRONMENTAL
QUALITY: A REVIEW Atmospheric Environment, 25B, 10
Givoni, B. (1991b). Impact of planted areas on urban environmental quality: A review.
170
Atmospheric Environment. Part B. Urban Atmosphere, 25, 289-299
Givoni, B., La Roche, P., (2000). Indirect Evaporative Cooling with an Outdoor Pond.
Cambridge, UK: James & James
Grahn, P., & Stigsdotter, U. (2003). Urban Forestry & Landscape planning and stress.
Urban Greening, 2, 17
Grahn, P., & Stigsdotter, U.K. (2010). The relation between perceived sensory dimensions
of urban green space and stress restoration. Landscape and Urban Planning, 94, 264-275
Granzin, K.L., & Williams, R.H. (2012). Patterns of behavioral characteristics as indicants
of recreation preferences: a canonical analysis. Research Quarterly, 92, 615-624
Grimm NB, F.S., Golubiewski NE (2008). Global change and the ecology of cities
Haase, D., Larondelle, N., Andersson, E., Artmann, M., Borgström, S., Breuste, J., Gomez-
Baggethun, E., Gren, Å., Hamstead, Z., & Hansen, R. (2014). A quantitative review of urban
ecosystem service assessments: Concepts, models, and implementation. Ambio, 43, 413-433
Haider, T. (1996). Modeling impacts of increased urban vegetation on ozone air quality in
the South Coast Air Basin. Atmospheric Environment, 30, 3423-3430
Haines-Young, R., & Potschin, M. (2010). The links between biodiversity, ecosystem
services and human well-being. D. Raffaelli/C. Frid (Hg.) Ecosystem Ecology: a new
synthesis. BES Ecological Reviews Series. Cambridge: Cambridge University Press (iE)
Hansen, R., Frantzeskaki, N., Mcphearson, T., Rall, E., Kabisch, N., Kaczorowska, A.,
Kain, J.H., Artmann, M., & Pauleit, S. (2015). The uptake of the ecosystem services concept
in planning discourses of European and American cities. Ecosystem Services, 12, 228-246
Haq, S.M.A. (2011). Urban green spaces and an integrative approach to sustainable
environment. Journal of Environmental Protection, 02, 601-608
Hartter, J. (2010). Resource use and ecosystem services in a forest park landscape. Society
and Natural Resources, 23, 207-223
He, H. (2010). To constrcute a low-carbon, greening and cultural Xi'an city In X.a.U.P.
Bureau (Ed.). Xi'an Xi'an Urban Plannning Bureau
Heidt, V., & Neef, M. (2008). Benefits of Urban Green Space for Improving Urban
Climate. In M.M. Carreiro, Y.-C. Song, & J. Wu (Eds.), Ecology, Planning, and Management
of Urban Forests (pp. 84-96): Springer New York
Herzele, A.V., & Wiedemann, T. (2003). Amonitoring tool for the provision of accessible
and attractive urban green spaces. Landscape Urban Plan, 63, 17
Hiroaki T. Ishii, T.M., Keitaro Ito, Naoko Fujit, Ayumi Imanishi,Daisuke Hashimoto, Ayako
Iwasaki (2010). Integrating ecological and cultural values toward conservation and utilization
of shrine/temple forests as urban green space in Japanese cities. Landscape Ecol Eng, 6, 8
Ho, C.-h., Sasidharan, V., Elmendorf, W., Willits, F.K., Graefe, A., Godbey, G. (2005).
Gender and ethnic variations in urban parks preferences, visitations, and perceived benefits. .
Journal of Leisure Research, 37, 25
Hu, Q., Wu, W., Xia, T., Yu, Q., Yang, P., Li, Z., & Song, Q. (2013). Exploring the Use of
171
Google Earth Imagery and Object-Based Methods in Land Use/Cover Mapping. Remote
Sensing, 5, 6026-6042
Hu, Y., & Duan, H. (2004). Role of urban green system in the construction of urban
ecology Research of Soil and Water Conservation, 11, 3
Huang, L., Li, J., Zhao, D., & Zhu, J. (2008). A fieldwork study on the diurnal changes of
urban microclimate in four types of ground cover and urban heat island of Nanjing, China.
Building and Environment, 43, 7-17
James, P., Tzoulas, K., Adams, M.D., Barber, A., Box, J., Breuste, J., Elmqvist, T., Frith,
M., Gordon, C., Greening, K.L., Handley, J., Haworth, S., Kazmierczak, A.E., Johnston, M.,
Korpela, K., Moretti, M., Niemelä, J., Pauleit, S., Roe, M.H., Sadler, J.P., & Ward Thompson,
C. (2009). Towards an integrated understanding of green space in the European built
environment. Urban Forestry & Urban Greening, 8, 65-75
Jim, C., & Chen, W. (2006a). Perception and Attitude of Residents Toward Urban Green
Spaces in Guangzhou (China). Environmental Management, 38, 338-349
Jim, C.Y., & Chen, W.Y. (2006b). Impacts of urban environmental elements on residential
housing prices in Guangzhou (China). Landscape and Urban Planning, 78, 422-434
Jim, C.Y., & Chen, W.Y. (2008a). Assessing the ecosystem service of air pollutant removal
by urban trees in Guangzhou (China). J Environ Manage, 88, 665-676
Jim, C.Y., & Chen, W.Y. (2008b). Pattern and divergence of tree communities in Taipei's
main urban green spaces. Landscape and Urban Planning, 84, 312-323
Jim, C.Y., & Chen, W.Y. (2009). Diversity and distribution of landscape trees in the
compact Asian city of Taipei. Applied Geography, 29, 577-587
Jim, C.Y., Chen, S. (2003). Comprehensive greenspace planning based on landscape
ecology principles in compact Nanjing city, China. Landscape and Urban Planning, 65, 21
Jin, Y.J., & Hong, J.Y. (2015). Classification of urban park soundscapes through
perceptions of the acoustical environments. Landscape & Urban Planning, 141, 100-111
Jonsson, P. (2004). Vegetation as an urban climate control in the subtropical city of
Gaborone, Botswana. International Journal of Climatology, 24, 1307-1322
JR, P. (1989). The impact of vegetation on air conditioning consumption. Controlling
summer heat island. In G.K. In: Akbari H, Martien P (eds.) (Ed.), Proceedings of the
workshop on: Saving energy and reducing atmospheric pollution by controlling summer heat
island, University of California, Barkley, California,
Kühn, I., Brandl, R., May, R., Klotz, S., (2004). The flora of German cities is naturally
species rich. Evolutional Ecology Research, 6, 15
Kaliszuk, E., & Szulczewska, B. (2005). 'Green structure' - the term discussion In B.D.
A.C. Werquin, G. Lindholm, B. Oppermann, S. Pauleit, S. Tjallingii (Ed.), Final Report of
Green Structure and Urban Planning COST Action C11 (pp. 132-216)
Kassem, A.-w.S. (1994). Energy and water management in evaporitive cooling systems in
Saudi Arabia. Resources, Conservation and Recycling, 12, 135-146
172
Katayama, T., Ishii, A., Hayashi, T., & Tsutsumi, J.-i. (1993). Field surveys on cooling
effects of vegetation in an urban area. Journal of Thermal Biology, 18, 571-576
Kikegawa Y, G.Y., Kondo H, Hanaki K (2006). Impacts of Cityblock Scale
Countermeasures against Urban Heat Island Phenomena upon a Building's Energy
Consumption for Airconditioning. Appl. Energy, 19
King, A.C., Castro, C., Wilcox, S., Eyler, A.A., Sallis, J.F., & Brownson, R.C. (2000).
Personal and environmental factors associated with physical inactivity among different racial–
ethnic groups of US middle-aged and older-aged women. Health psychology, 19, 354
Kjelgren, R., & Montague, T. (1998). Urban tree transpiration over turf and asphalt
surfaces. Atmospheric Environment, 32, 35-41
Kornblum, W., Hayes, C., Allen, R., (2001). Public Use of Urban Parks: A Methods Manual
for Park Managers and Community Leaders In, Environmental Pollution (pp. 23-31).
Washington, DC.: The Urban Institute
Kowarik, I. (2005). Wild urban woodlands: towards a conceptual framework. Berlin and
New York: Springer
Krenichyn, K. (2004). Women and physical activity in an urban park: Enrichment and
support through an ethic of care. Journal of Environmental Psychology, 24, 117-130
Kuo, B.-Y. (2000). An analysis of microclimates in urban parks—a case study in Taiwan.
In, Institute of Architecture: National Cheng Kung University
Kuttler, W., & Strassburger, A. (1999). Air quality measurements in urban green areas–a
case study. Atmospheric Environment, 33, 4101-4108
Kyrkou, D., & Karthaus, R. (2011). Urban sustainability standards: predetermined
checklists or adaptable frameworks? Procedia Engineering, 21, 204-211
Lafortezza, R., Carrus, G., Sanesi, G., & Davies, C. (2009). Benefits and well-being
perceived by people visiting green spaces in periods of heat stress. Urban Forestry & Urban
Greening, 8, 97-108
Layke, C. (2009). Measuring nature’s benefits: a preliminary roadmap for improving
ecosystem service indicators. World Resources Institute: Washington
Layke, C., Mapendembe, A., & Brown, C. (2012). Indicators from the global and sub-
global Millennium Ecosystem Assessments: An analysis and next steps. Ecological
Indicators, 17, 77–87
Leitner, M.J., & Leitner, S.F. (1996). Leisure Enhancement (2nd Ed.). New York: Haworth
Press
Leo, N.D., Escobedo, F.J., & Dubbeling, M. (2016). The role of urban green infrastructure
in mitigating land surface temperature in Bobo-Dioulasso, Burkina Faso. Environment,
Development and Sustainability, 18, 373-392
Lewan, L., & Söderqvist, T. (2002). Knowledge and recognition of ecosystem services
among the general public in a drainage basin in Scania, Southern Sweden. Ecological
Economics, 42, 459-467
173
Lhomme, J.P., & Monteny, B. (2000). Theoretical relationship between stomatal resistance
and surface temperatures in sparse vegetation. Agricultural and Forest Meteorology, 104,
119-131
Li, F., Wang, R., Paulussen, J., & Liu, X. (2005). Comprehensive concept planning of
urban greening based on ecological principles: a case study in Beijing, China. Landscape &
Urban Planning, 72, 325-336
Li, J., and Sun, Gnenian (2002). vegetation adjusting temerature and humidity in Xi'an
during midsummer Journal od Arid land resouce and environment, 16, 4
Li, S. (2013). Differences of Environment Protection between City and Country
SHANDONG SOCIAL SCIENCES, 283-284
Lin, L. (1995). Landscape architectures. Taipei: Dijing Press
Liu, Z.a.Y., Xiuju (2008). The analysis of urban heat island and climatic paremeters of
Xi'an Journal od Arid land resouce and environment, 22, 4
Lo, A.Y., & Jim, C. (2010a). Willingness of residents to pay and motives for conservation
of urban green spaces in the compact city of Hong Kong. Urban Forestry & Urban Greening,
9, 113-120
Lo, A.Y., & Jim, C. (2010b). Differential community effects on perception and use of urban
greenspaces. Cities, 27, 430-442
Lo, A.Y.H., & Jim, C.Y. (2011). Citizen attitude and expectation towards greenspace
provision in compact urban milieu. Land Use Policy
Lohr, V.I., Pearson-Mims, C.H., Tarnai, J., & Dillman, D.A. (2004). How urban residents
rate and rank the benefits and problems associated with trees in cities. Journal of
Arboriculture, 30, 28-35
Lorenzo, A.B., Blanche, C.A., Qi, Y., Guidry, M.M., (2000). Assessing residents’
willingness to pay to preserve the community ur ban forest: a small city case study. Journal of
Arboriculture, 26, 6
Lou, Q.X. (2003). Chinese gardens. Beijing: China Intercontinental Press
Lovell, S.T., & Johnston, D.M. (2009). Creating Multifunctional Landscapes: How Can the
Field of Ecology Inform the Design of the Landscape? Frontiers in Ecology & the
Environment, 7, 212-220
Lu, X., & Peng, C. (2006). Urbanization and ecological environment problems of Xi'an
Journal of Arid Land Resources and Environment, 20, 5
Luo, T., & Li, W. (2015). A demand-oriented examination of local park services: Park
types, sizes, features and actual use in the City of Los Angeles
M. Feliciano, A.F., L. Nunes, T. Nunes (2008). The role of urban green spaces in air quality.
In, Landscape Ecology and Forest Management Conference. Chengdu, Sichuan, China
MA (2003). Ecosystems and human well-being: a framework for assessment. Washington,
DC: Island Press
MA (2005). Ecosystems and human well-being: biodiversity synthesis. Washington DC:
174
World Resources Institute
Madureira, H., Nunes, F., Oliveira, J.V., Cormier, L., & Madureira, T. (2015). Urban
residents’ beliefs concerning green space benefits in four cities in France and Portugal. Urban
Forestry & Urban Greening, 14, 56-64
Maes, J., Zulian, G., Thijssen, M., Castell, C., Baró, F., Ferreira, A.M., Melo, J., Garrett,
C.P., David, N., & Alzetta, C. (2016). Mapping and Assessment of Ecosystems and their
Services: Urban ecosystems
Manes, F., Incerti, G., Salvatori, E., Vitale, M., Ricotta, C., & Costanza, R. (2012). Urban
ecosystem services: tree diversity and stability of tropospheric ozone removal. Ecological
Applications, 22, 349-360
Manning, W.J., Feder, W.A., (1980). Biomonitoring Air Pollutants With Plants. London.:
Applied Science Publishers
Marcus CC, a.W.C., Insley E, Francis C (1998). Neighborhood Parks. In C.C.M.a.C.
Francis (Ed.), People Places, Design Guidelines for Urban Open Space (pp. 85-144). New
York: John Wiley & Sons, Inc.
Martina Petralli, L.M.a.S.O. (2009). Air temperature distribution in an urban park:
differences between open-field and below a canopy. In, The seventh International Conference
on Urban Climate. Yokohama, Japan
McDonald, A.G., Bealey, W.J., Fowler, D., Dragosits, U., Skiba, U., Smith, R.I., Donovan,
R.G., Brett, H.E., Hewitt, C.N., & Nemitz, E. (2007). Quantifying the effect of urban tree
planting on concentrations and depositions of PM10 in two UK conurbations. Atmospheric
Environment, 41, 8455-8467
McIntyre, N.E., Knowles-Yánez, K., & Hope, D. (2000). Urban ecology as an
interdisciplinary field: differences in the use of “urban” between the social and natural
sciences. Urban Ecosystems, 4, 5-24
McKinney, M.L. (2002). Urbanization, Biodiversity and Conservation. BioScience, 52,
883-890
McPherson, E.G., Nowak, D., Heisler, G., Grimmond, S., Souch, C., Grant, R., &
Rowntree, R. (1997). Quantifying urban forest structure, function, and value: the Chicago
Urban Forest Climate Project. Urban Ecosystems, 1, 49-61
McPherson, E.G., Nowak, David J., Rowntree, Rowan A. (1994). Chicago's Urban Forest
Ecosystem: Results of the Chicago Urban Forest Climate Project. In (p. 201). Radnor: USDA
McPherson, E.G., Scott, K.I., Simpson, J.R. (1998). Estimating cost effectiveness of
residential yard trees for improving air quality in Sacramento, California, using existing
models. Atmospheric Environment Health Today, 32, 9
Memon, R.A., Leung, D.Y.C., Liu, C.H., & Leung, M.K.H. (2011). Urban heat island and
its effect on the cooling and heating demands in urban and suburban areas of Hong Kong.
Theoretical and Applied Climatology, 103, 441-450
Mengtao, L.I., & Zhou, Z. (2016). Positive and negative ecosystem services evaluation and
175
its spatial pattern analysis on urban landscape:A case study of Xi'an City. Acta Geographica
Sinica
Mering, C., Baro, J., & Upegui, E. (2010). Retrieving urban areas on Google Earth images:
application to towns of West Africa. International Journal of Remote Sensing, 31, 5867-5877
Millard, A. (2000). The potential role of natural colonisation as a design tool for urban
forestry — a pilot study. Landscape & Urban Planning, 52, 173-179
Millard, A. (2010). Cultural Aspects of Urban Biodiversity Oxford, UK Wiley-Blackwell
Miller, J.R., Hobbs, R.J., (2002). Conservation where people live and work. conservation
biology, 16, 7
Miller, R.W. (1988). Urban forestry: planning and managing urban greenspaces. Prentice
Hall
Min, L., Fangying, G., Jiawei, F., Meixuan, S., & He, Z. (2011). The sustainable approach
to the green space layout in highdensity urban environment: a case study of Macau peninsula.
Procedia Engineering, 21, 922-928
More, T.A. (1985). Central City Parks: A Behavioral Perspective. In, Natural Resources.
Burlington: University of Vermont
Nasar, J.L. (1997). New developments in aesthetics for urban design. In G.M. Moor, R.
(Ed.), Advances in environment, behavior, and design. New York Plenum Press
Nicol, C., & Blake, R. (2000). Classification and Use of Open Space in the Context of
Increasing Urban Capacity. Planning Practice & Research, 15, 193-210
Nielsen, T.S., & Hansen, K.B. (2007). Do green areas affect health? Results from a Danish
survey on the use of green areas and health indicators. Health Place, 13
Niemelä, J., Saarela, S.R., Söderman, T., Kopperoinen, L., Yli-Pelkonen, V., Väre, S., &
Kotze, D.J. (2010). Using the ecosystem services approach for better planning and
conservation of urban green spaces: A Finland case study. Biodiversity and Conservation, 19,
3225-3243
Nishimura, N., Nomura, T., Iyota, H., & Kimoto, S. (1998). NOVEL WATER FACILITIES
FOR CREATION OF COMFORTABLE URBAN MICROMETEOROLOGY. Solar Energy,
64, 197-207
Nowak, D.J. (2002). The Effects of Urban Trees on Air Quality. In. Syracuse,NY
Nowak, D.J., & Dwyer, J. F. (2000). The Urban Forest Effects (UFORE) Model:
Quantifying urban forest structure and functions. In T.B. M. Hansen (Ed.), IUFRO
Conference (pp. 714-720). St. Paul, MN U.S. Department of Agriculture, Forest Service,
North Central Research Station
Nowak, D.J., & Crane, D.E. (2002). Carbon storage and sequestration by urban trees in the
USA. Environmental Pollution, 116, 381-389
Nowak, D.J., Crane, D.E., & Stevens, J.C. (2006). Air pollution removal by urban trees and
shrubs in the United States. Urban Forestry & Urban Greening, 4, 115-123
Nowak, D.J., & Dwyer, J.F. (2007). Understanding the Benefits and Costs of Urban Forest
176
Ecosystems Urban and Community Forestry in the Northeast. In J.E. Kuser (Ed.) (pp. 25-46):
Springer Netherlands
NRC (2001). Grand Challenges in Environmental Sciences. Washington, DC.: National
Academy Press
Oguz, D. (2000). User surveys of Ankara's urban parks. Landscape and Urban Planning,
52, 165-171
Oke, T.R. (2011). Urban heat islands In I. Douglas (Ed.), The Routledge handbook of urban
ecology. Routledge: London
Ooi, G.L. (2009). Challenges of sustainability for Asian urbanisation. Current Opinion in
Environmental Sustainability, 1, 187-191
Ouyang, J. (2016). The indication of air pollution by plants Journal of Green Science and
Technology
Özgüner, H., & Kendle, A.D. (2006). Public attitudes towards naturalistic versus designed
landscapes in the city of Sheffield (UK). Landscape and Urban Planning, 74, 139-157
Paoletti, E., Bardelli, T., Giovannini, G., & Pecchioli, L. (2011). Air quality impact of an
urban park over time. Procedia Environmental Sciences, 4, 10-16
Pataki, D.E., Carreiro, M.M., Cherrier, J., Grulke, N.E., Jennings, V., Pincetl, S., Pouyat,
R.V., Whitlow, T.H., & Zipperer, W.C. (2011). Coupling biogeochemical cycles in urban
environments: ecosystem services, green solutions, and misconceptions. Frontiers in Ecology
and the Environment, 9, 27-36
Pauleit, S., & Breuste, J.H. (2011). Land use and surface cover as urban ecological
indicators. Handbook of Urban Ecology, 19-30
Pauleit, S., & Duhme, F. (2000). Assessing the environmental performance of land cover
types for urban planning. Landscape Urban Plan, 52, 20
Pauleit, S., Duhme, F. (2000). Assessing the environmental performance of land cover
types for urban planning. Landscape Urban Plan, 52, 20
Payne, L.L., Mowen, A.J., & Orsega-Smith, E. (2002). An Examination of Park Preferences
and Behaviors Among Urban Residents: The Role of Residential Location, Race, and Age.
Leisure Sciences, 24, 181-198
Peschardt, K.K., Schipperijn, J., & Stigsdotter, U.K. (2012). Use of small public urban
green spaces (SPUGS). Urban Forestry & Urban Greening, 11, 235-244
Pickett, S.T.A., Cadenasso, M.L., Grove, J.M., Groffman, P.M., Band, L.E., Boone, C.G.,
Burch, W.R., Grimmond, C.S.B., Hom, J., & Jenkins, J.C. (2008). Beyond Urban Legends:
An Emerging Framework of Urban Ecology, as Illustrated by the Baltimore Ecosystem Study.
BioScience, 58, 139-150
Pickett, S.T.A., Cadenasso, M.L., Grove, J.M., Nilon, C.H., Pouyat, R.V., Zipperer, W.C.,
& Costanza, R. (2001). Urban Ecological Systems: Linking Terrestrial Ecological, Physical,
And Socioeconomic Components Of Metropolitan Areas. Urban Ecology, 32, 127-157
Picot, X. (2004). Thermal comfort in urban spaces: impact of vegetation growth: Case
177
study: Piazza della Scienza, Milan, Italy. Energy and Buildings, 36, 329-334
Pincetl, S. (2003). Nonprofits and Park Provision in Los Angeles: An Exploration of the
Rise of Governance Approaches to the Provision of Local Services*. Social Science
Quarterly, 84, 979-1001
Pincetl, S., & Gearin, E. (2005). The reinvention of public green space. Urban Geography,
26, 19
Potchter, O., Cohen, P., & Bitan, A. (2006). Climatic behavior of various urban parks
during hot and humid summer in the mediterranean city of Tel Aviv, Israel. International
Journal of Climatology, 26, 1695-1711
Pretty, J., Peacock, J., Sellens, M., & Griffin, M. (2005). The mental and physical health
outcomes of green exercise. International journal of environmental health research, 15, 319-
337
Priego, C., Breuste, J., & Rojas, J. (2008). Perception and value of nature in urban
landscapes: a comparative analysis of cities in Germany, Chile and Spain. Landscape Online
Qian, W., Quan, L., & Shi, S. (2002). Variations of the Dust Storm in China and its
Climatic Control. Journal of Climate, 15, 1216-1229
Qin, J., Zhou, X., Sun, C., Leng, H., & Lian, Z. (2013). Influence of green spaces on
environmental satisfaction and physiological status of urban residents. Urban Forestry &
Urban Greening, 12, 490-497
Qin Yaomin, L.K., Wang Lijun (2006). the study on ecological funcrtions of green space in
Xi'an Chinese Journal of Ecology, 25, 4
Qiu, H. (2009). Study on Recreation and Renewal of Chinese Comprehensive Parks In,
Landscape Architecture (p. 198). Beijing, China Beijing Forestry University
Qiu, L., & Nielsen, A.B. (2015). Are Perceived Sensory Dimensions a Reliable Tool for
Urban Green Space Assessment and Planning? Landscape Research, 1-21
Qureshi, S., Breuste, J., & Lindley, S. (2010). Green Space Functionality Along an Urban
Gradient in Karachi, Pakistan: A Socio-Ecological Study. Human Ecology, 38, 283-294
Rall, E., Bieling, C., Zytynska, S., & Haase, D. (2017). Exploring city-wide patterns of
cultural ecosystem service perceptions and use. Ecological Indicators, 77, 80-95
Redman, C.L. (1999). Human Dimensions of Ecosystem Studies. Ecosystems, 2, 296-298
Robitu, M., Musy, M., Inard, C., & Groleau, D. (2006). Modeling the influence of
vegetation and water pond on urban microclimate. Solar Energy, 80, 435-447
Rosenfeld, A.H., Akbari, H., Romm, J.J., & Pomerantz, M. (1998). Cool communities:
strategies for heat island mitigation and smog reduction. Energy and Buildings, 28, 51-62
Rossi, S.D., Pickering, C.M., & Byrne, J.A. (2016). Not in our park! Local community
perceptions of recreational activities in peri-urban national parks. Australasian Journal of
Environmental Management, 1-20
Rydberg, D., & Falck, J. (2000). Urban forestry in Sweden from a silvicultural perspective:
a review. Landscape and Urban Planning, 47, 1-18
178
Sander, H., Polasky, S., & Haight, R.G. (2010). The value of urban tree cover: A hedonic
property price model in Ramsey and Dakota Counties, Minnesota, USA. Ecological
Economics, 69, 1646-1656
Sandström, U.G., Angelstam, P., & Mikusinski, G. (2006). Ecological diversity of birds in
relation to the structure of urban green space. Landscape and Urban Planning, 77, 39-53
Sanesi, G., & Chiarello, F. (2006). Residents and urban green spaces: The case of Bari.
Urban Forestry &amp; Urban Greening, 4, 125-134
Schipperijn, J., Ekholm, O., Stigsdotter, U.K., Toftager, M., Bentsen, P., Kamper-
Jørgensen, F., & Randrup, T.B. (2010). Factors influencing the use of green space: Results
from a Danish national representative survey. Landscape and Urban Planning, 95, 130-137
Shan, X. (2009). Urban green spaces in Guangzhou (China) : attitude, preference, use
pattern and assessment. In. Hong Kong: The University of Hong Kong
Shan, X.Z., & Yu, X. (2014). Citizen Assessment as Policy Tool of Urban Public Services:
Empirical Evidence from Assessments of Urban Green Spaces in China. Sustainability, 6,
7833-7849
Shashua-Bar, L., & Hoffman, M.E. (2000). Vegetation as a climatic component in the
design of an urban street: An empirical model for predicting the cooling effect of urban green
areas with trees. Energy and Buildings, 31, 221-235
Shi, H. (2013). Green Space system of Gardening City - Xi'an In, Landscape Architecture
(p. 54). Yang Ling: North-West A&F University
Shuqing Zhao, L.D., Zhiyao Tang, Hejun Fang, Kun Song, and Jingyun Fang (2006).
Ecological consequences of rapid urban expansion: Shanghai, China Front Ecol Environ, 4, 5
Smith, W.H. (1990). Air pollution and forests. New York: Springer-Verlag
Souch, C.A.S.a.C. (1993). The effect of trees on summertime below canopy urban
climates:a case study Bloomington,Indiana. Journal of Arboriculture, 19
Sparks, T.H., Butchart, S.H., Balmford, A., Bennun, L., Stanwell-Smith, D., Walpole, M.,
Bates, N.R., Bomhard, B., Buchanan, G.M., & Chenery, A.M. (2011). Linked indicator sets
for addressing biodiversity loss. Oryx, 45, 411-419
Spronken-Smith, R., & Oke, T. (1998). The thermal regime of urban parks in two cities
with different summer climates. International Journal of Remote Sensing, 19, 2085-2104
Spronken-Smith, R.A., Oke, T.R., & Lowry, W.P. (2000). Advection and the surface energy
balance across an irrigated urban park. International Journal of Climatology, 20, 1033-1047
Stern, T.N. (2010). Bewertung der sozio-ökologischen Funktionen von Grünflächen-
dargestellt am Beispiel der Megastadt Shanghai. In, FB Geographie und Geologie (p. 168):
Paris-Lodron Universität Salzburg
Sun, R., & Chen, L. (2012). How can urban water bodies be designed for climate
adaptation? Landscape and Urban Planning, 105, 27-33
Swanwick, C., Dunnett, N., & Woolley, H. (2003). Nature, role and value of green space in
towns and cities: an overview. Built Environment, 29, 12
179
Taha HG, A.H., RosenfeldAH (1988). Vegetation canopy micro-climate: a field project in
Davis, California. Berkley,CA: Lawrence Berkeley Laboratory, University of California-
Berkeley
TEEB (2010). The Economics of Ecosystems and Biodiversity Ecological and Economic
Foundations. In P.K. Earthscan (Ed.). London and Washington
Tjallingii, S. (2005). Green structure and urban planning -- general outcomes of COST C11
In B.D. A.C. Werquin, G. Lindholm, B. Oppermann, S. Pauleit, S. Tjallingii (Ed.), Final
Report of Green Structure and Urban Planning COST Action C11 (pp. 15-37)
Todorova, A., Asakawa, S., & Aikoh, T. (2004). Preferences for and attitudes towards street
flowers and trees in Sapporo, Japan. Landscape and Urban Planning, 69, 403-416
Toftager, M., Ekholm, O., Schipperijn, J., Stigsdotter, U., Bentsen, P., Grønbæk, M.,
Randrup, T.B., & Kamper-Jørgensen, F. (2011). Distance to green space and physical activity:
a Danish national representative survey. Journal of Physical Activity & Health, 8, 741-749
Tong, C., Feagin, R.A., Lu, J., Zhang, X., Zhu, X., Wang, W., & He, W. (2007). Ecosystem
service values and restoration in the urban Sanyang wetland of Wenzhou, China. Ecological
Engineering, 29, 249-258
Tratalos, J., Fuller, R.A., Warren, P.H., Davies, R.G., & Gaston, K.J. (2007). Urban form,
biodiversity potential and ecosystem services. Landscape and Urban Planning, 83, 308-317
Tyrväinen, L. (2001). Economic valuation of urban forest benefits in Finland. Journal of
Environmental Management, 62, 17
Tyrväinen, L., Mäkinen, K., & Schipperijn, J. (2007). Tools for mapping social values of
urban woodlands and other green areas. Landscape and Urban Planning, 79, 5-19
Tyrväinen, L., & Miettinen, A. (2000). Property Prices and Urban Forest Amenities.
Journal of Environmental Economics and Management, 39, 205-223
Tyrväinen, L., Pauleit, S., Seeland, K. & de Vries, S. (2005). Urban forests and trees in
Europe: A reference book Berlin Heidelberg: Springer-Verlag
Tzoulas, K., & James, P. (2010a). Making biodiversity measures accessible to non-
specialists: an innovative method for rapid assessment of urban biodiversity. Urban
Ecosystems, 13, 113-127
Tzoulas, K., & James, P. (2010b). Peoples’ use of, and concerns about, green space
networks: A case study of Birchwood, Warrington New Town, UK. Urban Forestry & Urban
Greening, 9, 121-128
Tzoulas, K., Korpela, K., Venn, S., Yli-Pelkonen, V., Kazmierczak, A., Niemela, J., &
James, P. (2007). Promoting ecosystem and human health in urban areas using Green
Infrastructure: A literature review. Landscape and Urban Planning, 81, 167-178
UN (2010). Urbanization in China- Unitaed Nations Report. In
UN (2014). World Urbanization Prospects: The 2013 Revision. In. New York: United
Nations
Vihervaara, P., Marjokorpi, A., Kumpula, T., Walls, M., & Kamppinen, M. (2012).
180
Ecosystem services of fast-growing tree plantations: a case study on integrating social
valuations with land-use changes in Uruguay. Forest Policy and Economics, 14, 58-68
Voigt, A., Kabisch, N., Wurster, D., Haase, D., & Breuste, J. (2014). Structural Diversity: A
Multi-dimensional Approach to Assess Recreational Services in Urban Parks. Ambio, 43, 480
Volker Heidt, M.N. (2008). Benefits of Urban Green Space for improving Urban Climate.
(p. 12)
Wang, T. (2007 ). Eco-city Construction in Xi'an In, Management of Science and
Engineering (p. 76). Xi'an Xi'an Architecture & Technology University
Wang, Y., Groot, R.D., Bakker, F., Wörtche, H., & Leemans, R. (2016). Thermal comfort in
urban green spaces: a survey on a Dutch university campus. International Journal of
Biometeorology, 1-15
Wendel, H.E.W., Zarger, R.K., & Mihelcic, J.R. (2012). Accessibility and usability: Green
space preferences, perceptions, and barriers in a rapidly urbanizing city in Latin America.
Landscape & Urban Planning, 107, 272-282
Weng, Q. (2001). A remote sensing-GIS evaluation of urban expansion and its impact on
surface temperature in the Zhujiang Delta, China. International Journal of Remote Sensing,
22, 1999-2014
Weng, Q., & Yang, S. (2004). Managing the adverse thermal effects of urban development
in a densely populated Chinese city. Journal of Environmental Management, 70, 145-156
Whitford, V., Ennos, A.R., & Handley, J.F. (2001). "City form and natural process"--
indicators for the ecological performance of urban areas and their application to Merseyside,
UK. Landscape and Urban Planning, 57, 91-103
Whitford, W., de Soyza AG, Van Zee JW, Herrick, J., & Havstad, K. (1998). Vegetation,
soil, and animal indicators of rangeland health. Environmental Monitoring and Assessment,
51, 200
WHO (2002). The World Health Report 2002:Reducing Risks, Promoting Healthy Life. In
W.H. Organization (Ed.). Geneva: World Health Organization
Woolley, H. (2006). Freedom of the city: Contemporary issues and policy influences on
children and young people's use of public open space in England. Children S Geographies, 4,
45-59
Wu, F., Li, S., & Liu, J. (2006). Effects of Urban Green Space with Different Vegetation
Quantity on Temperature and Humidity. LA Investigation and Research, 4
XAALADI (2004). Master Plan of Xi'an City (2004-2020). In Xi'an Ancient Landscape
Architecture Institute (Ed.). Xi'an
XAALADI (2009). Urban Green Spaces System Plan of Xi'an City (2008-2020). In Xi'an
Ancient Landscape Architecture Institute (Ed.). Xi'an: Xi'an Ancient Architecture and
Landscape Architecture Design Institute
XAEPB (2000). The tenth five years plan of environmental protection of Xi'an In. Xi'an
Xi'an Environmental Protention Bureau
181
Xu, M., Zhou, S., Yu, A., & Lin, T. (2013). Plant Resistance to Urban Air Pollution Forest
Engineering, 29, 49-54
Xu, W., & Zhang, Y. (2006). The Application Problems of Greening Tree in Xi'an Acta
Agriculturae Boreali-occidentalis Sinica, 15, 5
XUPB (1980). Master Plan of Xi'an City (1980-2000). In X'an urban Planning Bureau
(Ed.). Xi'an Xi'an Urban Planning Bureau
XUPB (1995). Master Plan of Xi'an City (1995-2010). In X'an urban Planning Bureau (Ed.)
XUPB (2010). Master Plan of Xi'an City (2008-2020). In X'an urban Planning Bureau (Ed.)
Yan, Y. (2008). Green space system analysis of Xi'an In, land use management Xi'an
Changan University
Yang, z. (2007). greening in the condition of air pollutants SCI-TECH INFORMATION
DEVELOPMENT & ECONOMY, 17, 277-278
Yeh, C.-T., Huang, S.-L. (2009). Investigating spatiotemporal patterns of landscape
diversity in response to urbanization. Landscape and Urban Planning, 93, 11
Yen, Y., Wang, Z., Shi, Y., & Soeung, B. (2016). An Assessment of the Knowledge and
Demand of Young Residents regarding the Ecological Services of Urban Green Spaces in
Phnom Penh, Cambodia. Sustainability, 8
Yijun, S. (2011). Ecological City and Urban Sustainable Development. Procedia
Engineering, 21, 142-146
Yin, S., Shen, Z., Zhou, P., Zou, X., Che, S., & Wang, W. (2011). Quantifying air pollution
attenuation within urban parks: An experimental approach in Shanghai, China. Environmental
Pollution, 159, 2155-2163
Yin Shan, C.J., Chen Liping, Shen Zhemin, Zou Xiaodong, Wu Dan, Wang Wenhua
(2007 ). Effects of vegetation status in urban green spaces on particle removal in a street
canyon atmosphere ACTA ECOLOGICA SINICA, 27, 5
Yli-Pelkonen V, N.J. (2005). Linking ecological and social systems in cities: urban
planning in Finland as a case. Biodiversity and Conservation, 14, 20
Yongming, H., Peixuan, D., Junji, C., & Posmentier, E.S. (2006). Multivariate analysis of
heavy metal contamination in urban dusts of Xi'an, Central China. Science of The Total
Environment, 355, 176-186
Young, C., & Jarvis, P. (2001). Assessing the structural heterogeneity of urban areas: an
example from the Black Country (UK). Urban Ecosystems, 5, 49-69
Young, R.F. (2010). Managing municipal green space for ecosystem services. Urban
Forestry & Urban Greening, 9, 313-321
Yu, K., Li, D., & Li, N. (2006). The evolution of Greenways in China. Landscape and
Urban Planning, 76, 223-239
Yuan, F., Bauer, M.E. (2007). Comparison of impervious surface area and normalized
difference vegetation index as indicators of surface urban heat island effects in Landsat
imagery. Remote Sens, 106, 11
182
Yuan, L. (2014 ). Ecological green space pattern and development strategy research in
Xi'an city In, Urban Planning&Design (p. 90). Xi'an: Xi'an Architecture and Technology
Univerisy
Yuen, B., Kong, L., & Briffett, C. (1999). Nature and the Singapore resident. GeoJournal,
49, 8
Yufeng Liu, Zhihua Yuan, Wei Kong, & Sun, B. (2015). The Changing Trend of Heat
Island Intensity and Main Influencing Factors during 1993-2012 in Xi'an City. JOURNAL OF
NATURAL RESOURCES, 30, 12
Zhang, B., Xie, G.-d., Gao, J.-x., & Yang, Y. (2014). The cooling effect of urban green
spaces as a contribution to energy-saving and emission-reduction: A case study in Beijing,
China. Building and Environment, 76, 37-43
Zhang, W., Yang, J., Ma, L., & Huang, C. (2015). Factors affecting the use of urban green
spaces for physical activities: Views of young urban residents in Beijing. Urban Forestry &
Urban Greening, 14, 851-857
Zhao, X., Wu, X., & Chen, X. (2004). Urban vegetation investigaiton in Xi'an city Journal
od Arid land resouce and environment, 18, 5
Zhou, W., Huang, G., & Cadenasso, M.L. (2011). Does spatial configuration matter?
Understanding the effects of land cover pattern on land surface temperature in urban
landscapes. Landscape and Urban Planning, 102, 54-63

183
Appendixes and Annexes
Appendix 1: Profiles of the six urban parks used for pre-testing the methodology

Area Year
Park name Location in the city Park type
(ha) built
Lian Hu Park 5.25 1916 City center, inside First Residential park
Ring Road
Yong Yang Park 11.24 2006 South-west, inside Third District
Ring Road comprehensive park
Feng Qing Park 22.52 2004 South-west, inside Second Municipal
Ring Road comprehensive park
City Sports Park 33.84 2006 North, inside Third Ring Sports park
Road
Xing Qing Park 48.99 1958 South-east, inside Second Municipal
Ring Road comprehensive park
Qu Jiang Yi Zhi Park 50.67 2008 South-east, inside Third Municipal
Ring Road comprehensive park

184
Appendix 2: Survey sheet 1 – pre-test table of land cover within urban parks
(sources: Sekliziotis, 1980; Pauleit and Duhme, 2000; Hermy and Cornelis, 2000
and Cornelis and Hermy, 2004)

Land cover types Subdivisions of land covers Park name:_______


Built / Buildings
Roads
Car parks
Pavements / concrete surfaces
Open soils
Vegetation Forest
Deciduous wood
Coniferous wood
Mixed wood
Trees
Tree rows
Shrubs
Hedges
Lawns
Mown grassland as sports fields
Turf grass
Rough grass
Infested grass
Pastures
Grasslands
Hay meadows
Tall herb vegetation
Flower beds
Vegetables
Arable crops
Water Water / freshwater
Pools
Brooks
Supplementary types

(Use a √ to mark the blank box under the park name column. If the land cover type is
not shown in the given types, it should be recorded as a supplementary type)

185
Appendix 3: Survey sheet 2 – pre-test table of vegetation structure within urban
parks

Patch Vegetation Vegetation Height of vegetation


number layers composition MT: ≥10m;
(1/2/3/>3) T: 5-10 m;
S:< 5 m;
H: <10cm / 10-20cm / 20- 50cm / 0.5-1m / 1-2m
1 MT:
T:
S:
H:
2 MT:
T:
S:
H:
3 MT:
T:
S:
H::
… MT:
T:
S:
H:

Note:
1) For types of land cover and vegetation, refer to Appendix 2;
2) Vegetation layers included one layer, two layers, three layers and multilayer;
3) Apart from vegetation layers, vegetation composition was recorded as found on-
site, such as multi-layered groups of trees covering shrubs and lawns, trees
covering groups of shrubs and lawns, trees covering groups of shrubs, woodlands
or lawns;
4) Abbreviations: MT: mature trees; T: trees; S: shrubs; H: herbs.

186
Appendix 4: Survey sheet 3 – investigative table of land cover and vegetation structure within urban parks

Park name:_____________
Date: ________________
Vegetation layers Evergreen broadleaf tree cover Shrub cover and evergreen shrub Herb cover Other
Patch Land Low tree cover Tall tree cover
and compositions Evergreen coniferous tree cover cover (<5m) (lawn <10 cm information
number covers (5-10m) (tree height ≥ 10m)
(1/2/3/>3) Deciduous broadleaf tree cover /grassland <20cm) (P/S/SF/RF)
EBC1 EBC2 EBC3 EBC4
ESC1 ESC2 ESC3 ESC4
1 C1 C2 C3 C4 C1 C2 C3 C4 ECC1 ECC2 ECC3 ECC4 C1 C2 C3 C4

DBC1 DBC2 DBC3 DBC4 SC1 SC2 SC3 SC4

EBC1 EBC2 EBC3 EBC4


ESC1 ESC2 ESC3 ESC4
2 C1 C2 C3 C4 C1 C2 C3 C4 ECC1 ECC2 ECC3 ECC4 C1 C2 C3 C4

DBC1 DBC2 DBC3 DBC4 SC1 SC2 SC3 SC4

EBC1 EBC2 EBC3 EBC4


ESC1 ESC2 ESC3 ESC4
3 C1 C2 C3 C4 C1 C2 C3 C4 ECC1 ECC2 ECC3 ECC4 C1 C2 C3 C4

DBC1 DBC2 DBC3 DBC4 SC1 SC2 SC3 SC4

EBC1 EBC2 EBC3 EBC4


ESC1 ESC2 ESC3 ESC4
4 C1 C2 C3 C4 C1 C2 C3 C4 ECC1 ECC2 ECC3 ECC4 C1 C2 C3 C4

DBC1 DBC2 DBC3 DBC4 SC1 SC2 SC3 SC4

EBC1 EBC2 EBC3 EBC4


ESC1 ESC2 ESC3 ESC4
5 C1 C2 C3 C4 C1 C2 C3 C4 ECC1 ECC2 ECC3 ECC4 C1 C2 C3 C4

DBC1 DBC2 DBC3 DBC4 SC1 SC2 SC3 SC4

(Use a √ to mark the answer under the fourth to ninth column; the second, third and tenth column should be recorded as observed)

187
Explanations and abbreviations:
1) Land cover types: there are six optional types of land cover based on the pre-test
sheet (V: vegetation; P: pavements; PV: vegetation over pavements;
PG: playgrounds; B: buildings; W: water bodies);
2) Lawn means frequently mown grass whereas grassland means naturally growing
grass;
3) Abbreviations: C: coverage; SC: coverage of shrubs; ESC: coverage of evergreen
shrubs; EBC: coverage of evergreen broadleaf trees; ECC: coverage of evergreen
coniferous trees; DBC: coverage of deciduous broadleaf trees; P: pavement; S:
soil surface; SF: sports facilities; RF: recreational facilities;
4) Each vegetated patch was assigned to class 1 (0-25%), 2 (26-50%), 3 (51-75%) or
4 (76-100%). To abbreviate the four classes for different layers of vegetation, the
following acronyms were used: C1,SC1,ESC1,EBC1,ECC1,DBC1
/C2,SC2,ESC2,EBC2,ECC2,DBC2 / C3,SC3,ESC3,EBC3,ECC3,DBC3 /
C4,SC4,ESC4,EBC4,ECC4,DBC4.

188
Appendix 5: Questionnaire
Park name:
Date:
Investigator:
Note: Please place √ in the □ at the beginning of the choice answer.

Part I Personal Information


Sex: □female □male
Age: □15-25□26-45□46-65□>65
Family status: □single □couple □3(parents and a child) □>3 (parents and more than
one child)
Vocation: □student □employed □self-employed □unemployed □retiree

Part II Use of Urban Green Spaces


1. Where do you go for your daily recreation (please tick as appropriate)?
Green spaces Very often Often Occasionally
1 Public parks
2 Residential green spaces
3 Roadside green spaces and squares
4 Green spaces near work place or school
5 Other (please state)

2. How long do you spend travelling to the green spaces you usually frequent (please
tick as appropriate)? (min=minutes)
6 <5 min
7 5-15 min
8 16-30 min
9 >30 min

3. How often do you visit these green spaces (please tick as appropriate)?
10 Daily
11 Several times per
week
12 Weekly
13 1-3 times per month
14 Monthly
15 Several times per year
16 Less

189
4. At what time do you use these green spaces (please tick as appropriate)?
Time of use Very often Often Occasionall
y
17 Time of the In the morning (before 12:00)
day
In the afternoon (12:00-
18:00)
In the evening (after 18:00)
18 Time of the Weekdays
week
Weekends

5. How long do you normally remain in these green spaces (please tick as
appropriate)?
19 Less than one hour
20 One to two hours
21 Half a day
22 Nearly a whole day

6. What is your main recreational activity when you use green spaces (please tick as
appropriate)?
Very Important Not Nether
important important important
nor
unimportant
23 To do sport
24 To use the recreational facilities (e.g.
fitness equipment)
25 To relax / reduce stress
26 To spend time in a quiet and peaceful
area
27 To have fun with friends or play with
family
28 To enjoy the beautiful views
29 To enjoy the surrounding greenery
30 To be in contact with nature
31 To learn about plants and animals
32 To enjoy the weather and get fresh
air
33 To enjoy the cooler climate in
summer
34 To walk the dog
35 Other (please state)
190
Part III Perspectives and Attitudes towards the Ecosystem Services of Urban
Green Spaces
7. What type of urban green space is most important for you (please tick as
appropriate)?
Very Important Not Nether important nor
important important unimportant
36 Near my residence
(or work place, or
school)
37 Big city parks, even
if further away
Parks outside the
38
city
39 Other (please state)

8. How do you feel about the services provided by urban green spaces (please tick as
appropriate)?
Very Important Not Nether
important important important
nor
unimportant
40 O2 release
41 CO2 sequestration
42 Wind protection
43 Increased air humidity
44 Lower air temperature
45 Shade
46 Air quality improvement
47 Noise reduction
48 Wildlife habitats
49 Species conservation
50 Water-soil conservation
Places for recreational
51
activities
52 Aesthetic improvement
Cultural and educational
53
benefits
54 More contact with nature
55 Neighbor-social interaction
Economic value (e.g.
56 cultivation of wood and
fruits)
57 Other (please state)
191
9. How do you feel about the urban green spaces of Xi’an city (please tick as
appropriate)?
Excellent Good Fair Unclear Poor
58 Amount of green areas
59 Vegetation diversity
60 Recreational opportunities
Ecological functions (e.g. decreased
61
temperature, lower air pollution)
62 Management/maintenance of the green spaces
63 Other (please state)

10. What is your assessment of the quality of frequently used green spaces (please
tick as appropriate)?
Excellent Good Fair Unclear Poor
Management of green spaces
64 Cleanliness
65 Safety
Amount of green areas
66 Vegetation cover ratio
Species conservation (vegetation and animals)
67 The number of plant types
The number of wild animal types (e.g. birds
68
and insects)
Recreational opportunities
69 Scenic beauty
70 The number of sporting facilities
71 The location of sporting facilities
72 The maintenance of sporting facilities
The number of resting facilities (e.g. tables
73
and benches)
The location of resting facilities (e.g. tables
74
and benches)
The maintenance of resting facilities (e.g.
75
tables and benches)
76 The area of playgrounds
77 The location of playgrounds
78 The maintenance of playgrounds
Ecological functions (e.g. decreased
temperature, lower air pollution)
Microclimate (e.g. cooling, tree shade,
79
humidity and wind)
80 Air quality

192
Excellent Good Fair Unclear Poor
81 The area of a body of water
Water quality (if in connection with a water
82
body)
83 Overall assessment of this site

11. What would you like to see improved in your frequently used green spaces (please
tick as appropriate)?
Strongly Need Do not Do not need
need care
84 More shade
85 More wind-protected sites
86 More quiet places
87 More facilities for sports and exercise
88 More play areas for children
89 More resting facilities (e.g. benches / tables)
90 More cafes / restaurants
91 More bodies of water
92 More flowers
More opportunities to watch nature
93
(vegetation, birds, etc.)
Better management (e.g. better maintenance of
94
plantings, facilities / less litter)
More convenient and safer transportation to
95
the green spaces
96 Other (please state)

193
Annex 1 Land cover types, characteristics and examples within urban parks of Xi’an
Land cover type Sub-types Characteristics Examples
Vegetation Lawns Continuous large fields of mown grass,
with height of less than 10 cm

Flower beds Continuous fields of flowers,


with height varying from 20-100 cm

Water plants Continuous fields of aquatic plants


living in or near water, with height
varying from 20-100 cm

Bamboos Continuous fields of bamboo,


with height varying from 1-2 m

194
Land cover type Sub-types Characteristics Examples
Vegetation Shrubs Continuous groups or rows of
shrubs trimmed or growing naturally,
with heights of less than 5 m

Evergreen broadleaf woodlands Continuous tree canopy coverage


dominated by evergreen broadleaf trees,
with heights of more than 5 m

Evergreen coniferous woodlands Continuous tree canopy coverage


dominated by evergreen coniferous trees,
with heights of more than 5 m

Deciduous broadleaf woodlands Continuous tree canopy coverage


dominated by deciduous broadleaf trees,
with heights of more than 5 m

195
Land cover type Sub-types Characteristics Examples
Vegetation Trees covering shrubs A mixture of groups or rows of trees and
shrubs

Trees covering lawns A mixture of groups or rows of trees


covering lawns

Shrubs covering lawns A mixture of groups or rows of shrubs


covering lawns

Trees covering shrubs A mixture of groups or rows of trees,


and lawns shrubs and lawns

196
Land cover type Sub-types Characteristics Examples
Vegetation Multi-layered trees A multi-layered mixture of groups or
covering shrubs and lawns rows of trees, shrubs and lawns

Pavements with Sealed surface (≥50% of the field area)


scattered vegetation covered with scattered vegetation;
the dominant feature is the sealed surface

Pavements All sealed surfaces that are not part


of the road system, including squares,
sealed sport courses or enlarged road areas

Playgrounds Open fields with recreational facilities

197
Land cover type Sub-types Characteristics Examples
Buildings Business buildings Buildings for commercial use
(e.g. offices, public toilets and other facilities)

Garden architecture Buildings with recreational and


aesthetic functions

Water bodies Lakes Large water areas (e.g. manmade lakes)

Fountains Water jets with a purely recreational


and aesthetic purpose

198
Annex 2: Land Cover and Vegetation Structures Within the 22 Investigated
Urban Parks of Xi’an

Figure 1 Land cover and vegetation structure in Children’s Park

Figure 2 Land cover and vegetation structure in Tu Men Park


199
Figure 3 Land cover and vegetation structure in Fang Zhi Park

Figure 4 Land cover and vegetation structure in Ci En Si Yi Zhi Park


200
Figure 5 Land cover and vegetation structure in Xi Qu Da Guan Yuan

Figure 6 Land cover and vegetation structure in Lian Hu Park (water plants
distributed within the water area)
201
Figure 7 Land cover and vegetation structure in Min Su Park

Figure 8 Land cover and vegetation structure in Xin Ji Yuan Park


202
Figure 9 Land cover and vegetation structure in Wen Jing Park

Figure 10 Land cover and vegetation structure in Mu Ta Si Park


203
Figure 11 Land cover and vegetation structure in Mu Dan Yuan

Figure 12 Land cover and vegetation structure in Lao Dong Park


204
Figure 13 Land cover and vegetation structure in Ge Ming Park

Figure 14 Land cover and vegetation structure in Huan Cheng Xi Yuan

205
Figure 15 Land cover and vegetation structure in Yong Yang Park

Figure 16 Land cover and vegetation structure in Chang Le Park

206
Figure 17 Land cover and vegetation structure in Feng Qing Park

Figure 18 Land cover and vegetation structure in Qu Jiang Tang City Wall Park

207
Figure 19 Land cover and vegetation structure in City Sports Park

Figure 20 Land cover and vegetation structure in Tang Yan Road Tang City Wall
Park
208
Figure 21 Land cover and vegetation structure in Xing Qing Park

Figure 22 Land cover and vegetation structure in Qu Jiang Yi Zhi Park


Abbreviations of Figure 5.1-5.22: P: Pavements; P_V: Vegetation covering
209
pavements; PG: Playgrounds; B: Buildings; B_LA: Landscape architecture; W: Water
bodies; V_L: Lawns; V_F: Flower beds; V_W: Water plants; V_B: Bamboos; V_S:
Shrubs; V_EBW: Evergreen broadleaf woodlands; V_ECW: Evergreen coniferous
woodlands; V_DBW: Deciduous broadleaf woodlands; V_TS: Trees covering shrubs;
V_TL: Trees covering lawns; V_SL: Shrubs covering lawns; V_TSL: Trees covering
shrubs and lawns;V_MTSL: Multi-layered trees covering shrubs and lawns.

210
CURRICULUM VITAE

Personal data Tong Cai


Date of Birth: June First, 1981
Place of Birth: Shaanxi, China
Nationality: Chinese

Educations
9/2005-6/2008 Major in Forest Genetic and Breeding for master degree at
Northwest A&F University, Yangling, China

9/1999-7/2003 Major in Landscape Architecture for Bachelor degree at


Northwest A&F University, Yangling, China

9/1996-9/1999 26th Middle School of Xi’an, Shaanxi, China

9/1993-9/1996 90th junior Middle School of Xi’an, Shaanxi, China

Work Experiences
10/2003-4/2005 Work at Greening Center in Xi’an Engineering University
9/2005-9/2006 Part-time teacher in Yangling vocational & technical college

Papers
CaiTong, Guo-Junzhan, RuanYu (2008). Investigation on resource and gardening
application of colorful plants in Xi’an. Journal of Northwest College of Forestry, 23(4)
196-199

RuanYu, CaiTong et al.(2007). Study on the development of green space in


residential areas of cities. Journal of Yangling Vocational & Technical College, 6(3)
38-41

Conferences
Informing Decisions in a Changing World-2012 US, IALE Symposium, April 8-12,
2012, Newport, Rhode Island

Megacity Shanghai – Development of Sustainable of Urban Environment, 8th IALE


World Congress, August 13-16, 2011

211

You might also like