CONFLICTING PRIORITIES, CONFLICTING SCALES: URBAN FORESTRY IN ACCRA

BETWEEN HOUSEHOLD (RE)PRODUCTION AND GLOBAL SUSTAINABILITY

by

LYN-KRISTIN HOSEK

A thesis submitted to the University of Birmingham for the degree of

DOCTOR OF PHILOSOPHY

Department of African Studies and Anthropology

School of History and Cultures

College of Arts and Law

University of Birmingham

July 2021
UNIVERSITYDF
BIRMINGHAM

University of Birmingham Research Archive

e-theses repository

This unpublished thesis/dissertation is copyright of the author and/or third

parties. The intellectual property rights of the author or third parties in respect
of this work are as defined by The Copyright Designs and Patents Act 1988 or
as modified by any successor legislation.

Any use made of information contained in this thesis/dissertation must be in

accordance with that legislation and must be properly acknowledged. Further
distribution or reproduction in any format is prohibited without the permission
of the copyright holder.
 ABSTRACT
Accra Metropolitan Assembly intends to plant 100,000 trees to mitigate the effects of global cli-

mate change. The Environmental Protection Agency investigates the durability of tree-shaded

road surfaces to alleviate the national budget. John Yetsowodo, an Accra resident, is selling coco-

nuts to put cash in his pocket. This thesis is about both the blessing and curse of multi-functional

urban trees that ameliorate urban living conditions in Accra, but also cause conflict over man-

agement priorities and actual decision-making. When stakeholder aims diverge, decision-making

power becomes decisive. Such power produces urban forests that are neither random nor neutral,

and tree species and planting site choices which more often reflect national budgetary and global

sustainability concerns, than a preoccupation with the lives and livelihoods of city inhabitants.

Yet, incorporating public participation in urban forest interventions, especially by private land

and tree owners, facilitates more effective, better integrated and more equitable city-wide plan-

ning processes and management outcomes. As part of a mixed methods research design, I com-

bined tree inventories, archival research, stakeholder interviews, a sample household survey and

selected household case study ‘biographies’ (Feb. 2016–Aug. 2017) to analyse Accra’s urban for-

est changes over time and across space. I further examine current urban forestry management

systems and assess planned and proposed future interventions. Centring residents’ use strategies

and priorities within an emancipatory approach, the thesis demonstrates how a focus on individ-

ual and household level benefits of urban trees can contribute often absent but potentially indis-

pensable knowledges to urban forest planning and management.

In particular, I outline the lasting imprints on Accra’s urban forest left by historical events and

processes. As such outcomes cannot be changed without understanding the factors which con-

tribute to their existence in the first place, I also analyse the effects of previous land use or city

planning decisions as part of my scrutiny of current and proposed interventions for their likely
socio-environmental impacts. While numerous public stakeholders plan and manage (parts of)

Accra’s forest, their effectiveness is impeded by a lack of resources and written documentation

for assigning tasks and responsibilities in clear and unequivocal ways. Additionally, selected key

government stakeholders continue to focus on educating and sensitising the public, resting on the

misguided belief that Accra’s residents are neither aware of, nor particularly interested in, the

trees’ benefits. However, based on my household survey and case study, I refute this assumption

as respondents undoubtedly understand and appreciate the city trees’ multiple functions. At the

individual and household level, tree products are consumed for subsistence and sold for cash.

These products, when gifted or exchanged, also create and reinforce social and cultural capital,

while tree shade provides a pleasant outdoor environment for enacting the interactions in ques-

tion.

My research thus furthers our understanding of how natural vegetation resources in cities con-

tribute to livelihoods, highlighting how diversifying management objectives can benefit the wider

community, especially those living in arduous conditions. Nonetheless, current urban tree-related

events in Accra continue to prioritise higher scale benefits, exacerbating existing injustices regard-

ing available and accessible trees and their benefits, with various government stakeholders failing

to utilise urban forestry’s potential to promote secure livelihoods, especially for marginalised and

disadvantaged city residents.

In sum, I contribute to a people-centred, emancipatory and explicitly political reading of urban

forestry, which challenges simplistic explanations that assign blame to seemingly disinterested and

ignorant city residents for low and/or diminishing tree cover. In place of the latter narrative, I

emphasise the pervasive structural mechanisms that affect how urban forests are configured and

consider possibilities for incorporating meaningful popular participation in redressing the struc-

tural imbalances implied.

 DEDICATION

~ Für Oma und Opa, weil ihr meine Vergangenheit seid ~

~ Für Liliana, Rafael und Miriam, weil ihr meine Zukunft seid ~
 ACKNOWLEDGEMENTS
I am deeply thankful to all the people in Accra who so kindly gave their time, shared their

knowledge and information with me, and helped me with data collection. This includes everyone

who allowed me to measure trees on their land, patiently responded to my questions, or opened

the doors of their home to me. It also includes Dr Joseph Teye, who significantly helped with the

organisational aspects of data collection and the Department of Botany, University of Ghana,

that kindly helped identifying tree species.

I am deeply thankful for the guidance and support of my supervisor Dr Reginald Cline-Cole. So

many years ago, I approached him with an idea and now, thankfully, this finalised thesis looks

nothing like it. It will be a strange feeling to no longer find his e-mails in my inbox, especially

since it has been 15 years since I first met him during my undergraduate studies. Farewell, Reg!

I am deeply thankful for the help of my second supervisor, Dr Kate Skinner, who provided many

helpful comments and suggested readings, which made me feel more knowledgeable.

I am deeply thankful for the support of my husband and children through all these years. With-

out their love and patience, I would never have been able to finish this.

I am deeply thankful to my parents who helped with childcare and never held me back but en-

couraged me instead.

I am deeply thankful to my father and his wife, who visited many weekends and took care of the

children, cooked and baked cake, giving me extra time to work. Your support was invaluable and

I cannot adequately express how much it means to me.

I am deeply thankful for the help of Molin Hosek and Stephanie Reimann who proofread my

thesis for spelling, punctuation, and consistency.

This PhD project is supported by the Royal Geographical Society (with IBG) with a Postgraduate

Research Award.
 TABLE OF CONTENTS

1 INTRODUCTION 1

1.1 Introduction 1

1.2 Problem Statement 5

1.3 Aims and Objectives 6

1.4 Research Questions 8

1.5 Research Design 9

1.6 Original Contribution 10

1.7 Thesis Structure 12

2 CONCEPTUAL AND THEORETICAL FRAMEWORK AND LITERATURE

REVIEW 13

2.1 Urban Forestry 14

2.2 Zooming in 15
2.2.1 The Urban Forest 15
2.2.2 From Tree to Value 16

2.3 Zooming out 20

2.3.1 Actors and Their Interventions 20
2.3.2 Acting within Context 22
2.3.3 Inheriting (from) the Past 23

2.4 The Larger Picture – Political Ecology 25

2.5 Focus on the Framework’s Components 28

2.5.1 Context SpecifiCITY 28
2.5.2 Urban Forest Structure 30
2.5.3 Path Dependencies and Legacy Effects 31
 2.5.4 Urban Forestry – Planning and Managing the Urban Forest 34
2.5.5 Repeating the Process, Looking to the Future 41

2.6 Chapter Conclusion 42

3 METHODOLOGY 43

3.1 Research Design 43

3.1.1 Mixed Methods Research 43
3.1.2 The Transformative Paradigm 45
3.1.3 A Design for Mixed Methods Research 47
3.1.4 Positionality 49
3.1.5 Ethical Considerations 51

3.2 Study Site 53

3.3 Data Collection 55

3.3.1 Tree Inventory 55
3.3.2 Household Survey 62
3.3.3 Documentary Research 66
3.3.4 Stakeholder Interviews 68
3.3.5 Household Case Study 73

3.4 Chapter Conclusion 77

4 ACCRA’S URBAN FOREST – STRUCTURE OVER TIME AND ACROSS

SPACE 78

4.1 Human–Nature Interactions and Their Legacy Effects 79

4.1.1 Accra and Trees – The Pre-Colonial Past 80
4.1.2 Accra and Trees – The Colonial Past 85
4.1.3 Accra and Trees – The Post-Independence Past 94

4.2 Accra and Trees – The Present 101

4.2.1 City-Level 103
 4.2.2 Intra-City Variation 112

4.3 Chapter Conclusion 118

5 MANAGING AND USING THE URBAN FOREST – STAKEHOLDER

OPINIONS AND INTERACTIONS 119

5.1 Establishing Stakeholding in Accra’s Urban Forest 120

5.1.1 Stakeholders – The Theory 120
5.1.2 Stakeholders in Accra – The Practice 121

5.2 Urban Forestry in Accra 126

5.2.1 Organisational Structure and Functioning 126
5.2.2 Further Stakeholders and Interviewees 127

5.3 Stakeholder views – Strengths, Weaknesses, Opportunities, Threats 131

5.3.1 Strengths, Weaknesses, Opportunities, Threats – The Theory 131
5.3.2 Strengths, Weaknesses, Opportunities, Threats – The Practice 133

5.4 From SWOT to Themes 138

5.4.1 Theme 1 – The Little That We Have, the Lot That We Have to Do 139
5.4.2 Theme 2 – The Long Road from Awareness to Action 156

5.5 Chapter Conclusion 165

6 URBAN TREES – URBAN LIVELIHOODS 166

6.1 Urban Lives 167

6.2 Livelihoods – Urban Livelihoods – Sustainable Urban Livelihoods 169

6.3 Household Case Study 171

6.3.1 Household 1 – Nima 176
6.3.2 Household 2 – Mamobi 178
6.3.3 Household 3 – Nima 182
6.3.4 Household 4 – Kokomlemle 184
6.4 Tree Use and Care in the Case Study Households 186
6.4.1 Trees as Assets 186
6.4.2 Tree Propagation, Care and Management 202

6.5 Chapter Conclusion 206

7 URBAN FORESTRY INTERVENTIONS: THEIR LIKELY SOCIO-

ENVIRONMENTAL EFFECTS 209

7.1 The Natural Resource Base – Development of Tree Stock 210

7.2 Current and Proposed Interventions and Their Possible Impacts 212
7.2.1 Urban Forest Inventories and Management Plans 212
7.2.2 Re-Evaluation of Planning Schemes and Participation 216
7.2.3 Tree Laws and Tree Law Changes 226
7.2.4 Tree Planting and Other Environmental Initiatives 230
7.2.5 Achimota Forest Turns Eco-Park 234

7.3 Facilitating Tree Planting 238

7.4 Chapter Conclusion 242

8 CONCLUSION 244

8.1 Summary and Synthesis 245

8.2 Thesis Contributions 248

8.3 Recommendations 252

8.3.1 Recommendation One: Diversify the Knowledge and Experience Base 253
8.3.2 Recommendation Two: Develop an Urban Forest Management Plan 258
8.3.3 Recommendation Three: Re-Evaluate Species and Planting Site Selection 259
8.3.4 Recommendation Four: Focus on Disadvantaged Neighbourhoods 261
8.3.5 Recommendation Five: Fund the Department of Parks and Gardens 263

8.4 Looking Backward and Looking Forward 263

8.4.1 Hindsight 264
 8.4.2 Foresight 265

8.5 Chapter Conclusion 267

LIST OF REFERENCES 268

APPENDICES 307

1 Participant Information Sheets 307

1.1 Participant Information Sheet for Survey Respondents 307
1.2 Participant Information Sheet for Minors – Household Survey 308
1.3 Participant Information Sheet for Stakeholder Interviews 308
1.4 Participant Information Sheet for Household Case Study 309
1.5 Participant Information Sheet for Minors – Case Study 311

2 Consent Forms 312

2.1 Consent Form for Survey Respondents 312
2.2 Consent Form for Minors – Household Survey 313
2.3 Consent Form for Stakeholder Interviews 314
2.4 Consent Form for Household Case Study 315
2.5 Consent Form for Minors – Case Study 316

3 Neighbourhood Poverty Ratings 317

4 Tree Inventory Form 318

5 Household Questionnaire 321

6 List of Newspaper Articles 330

7 Guide for Semi-Structured Stakeholder Interviews 336

8 List of Trees in Schumacher’s Book 338

9 Neighbourhood Land Cover Values with Standard Errors 344

10 Terms for Interactions between Stakeholders 346

11 Information on All Household Trees 348

 LIST OF FIGURES
Figure 1 Freshly harvested coconuts in the trunk of a taxi in Mamobi 1

Figure 2 Location of Volta Region (grey) 4

Figure 3 Conceptual framework 13

Figure 4 Framework – Part I 15

Figure 5 Framework – Part II 15

Figure 6 Framework – Part III 17

Figure 7 Framework – Part IV 17

Figure 8 Framework – Part V 18

Figure 9 Framework – Part VI 19

Figure 10 Framework – Part VII 20

Figure 11 Framework – Part VIII 23

Figure 12 Framework – Part IX 25

Figure 13 Location of Kumasi 29

Figure 14 Procedural diagram 48

Figure 15 Extract from my research journal 51

Figure 16 Location of Ghana (grey) and Accra 53

Figure 17 Neighbourhood boundaries 54

Figure 18 Neighbourhood canopy cover classes 58

Figure 19 Household survey neighbourhoods 62

Figure 20 Ayawaso sub-metropolitan areas 63

Figure 21 Location of Fort James, Ussher Fort, and Fort Christiansborg (left to right) 81

Figure 22 Location of Axim and Cape Coast 84

Figure 23 Places listed as needing a town planning advisor 93

Figure 24 Location of Accra, Akosombo, and Tema 100

Figure 25 Previously European and African neighbourhoods 103

Figure 26 Land-use classification of sampled trees 109

Figure 27 Number of trees in public and private ownership divided by species 110

Figure 28 Tree height distribution of inventoried trees 111

Figure 29 Diameter distribution of inventoried trees 112

Figure 30 Neighbourhood locations 113

Figure 31 Trees per hectare in each poverty class 116

Figure 32 Number of species in each poverty class 116

Figure 33 Location of Dansoman 128

Figure 34 Location of Amasaman and Pokuase 130

Figure 35 Compound layout (Household 1) 176

Figure 36 Compound layout (Household 2) 179

Figure 37 Location of Lapaz 180

Figure 38 Location of Korle Bu and Korle Gonno 181

Figure 39 Compound layout (Household 3) 183

Figure 40 Compound layout (Household 4) 185

Figure 41 Location of Teshie Nungua 242

Figure 42 ‘The Unusual Garden’ – I 255

Figure 43 ‘The Unusual Garden’ – II 256

Figure 44 ‘The Unusual Garden’ – III 256

Figure 45 Location of places mentioned in Schumacher (1827) 343

 LIST OF TABLES
Table 1 Examples of ecosystem functions and resulting services and benefits 19

Table 2 Involvement in urban forestry based on actor type and land ownership 21

Table 3 The Transformative Paradigm: Ontology, Epistemology, Methodology, and Axiology 46

Table 4 Number of neighbourhoods in each poverty class 55

Table 5 Examples of different land cover categories 56

Table 6 Number of neighbourhoods in each canopy cover class 57

Table 7 Number of plots per stratum 59

Table 8 Collected plot and individual tree data 60

Table 9 Replaced plots per stratum 61

Table 10 Equipment and producer information 61

Table 11 Searched newspapers and their characteristics 67

Table 12 Interviewees and dates of interviews 69

Table 13 Interviewees associated with the case study households 75

Table 14 Recorded species and number of trees per species 105

Table 15 List of species commonly planted by DPG 107

Table 16 Comparison between iTree Canopy values for selected neighbourhoods 114

Table 17 Average canopy cover per poverty class 114

Table 18 Average plantable and non-plantable space per canopy cover class 115

Table 19 Trees per poverty class 115

Table 20 Trees in public and private ownership per poverty class 117

Table 21 Stakeholder types and examples for each type 122

Table 22 Potential stakeholders, their interests, importance and influence 123

Table 23 SWOTs identified by the stakeholder interviewees 134

Table 24 Interview text examples for different types of resources 140

Table 25 Examples of phrases used to describe interactions or their lack 152

Table 26 People’s possible positions regarding the purchase of a product or service 157

Table 27 Hierarchy of effects model with an urban forestry example 160

Table 28 Summary information on the four participating households 173

Table 29 Questionnaire respondents who mentioned food-producing tree benefits 193

Table 30 Tree-related social interactions 199

 LIST OF ACRONYMS
AMA Accra Metropolitan Assembly
DBH Diameter at Breast Height
DPG Department of Parks and Gardens
EPA Environmental Protection Agency
FC Forestry Commission
FORIG Forestry Research Institute of Ghana
GPS Global Positioning System
GYEM Ghana Youth Environmental Movement
(H)HH (Head of) Household
IMF International Monetary Fund
LUPMP Land Use Planning and Management Project
MEA Millennium Ecosystem Assessment
MESTI Ministry of Environment, Science, Technology and Innovations
MLGRD Ministry of Local Government and Rural Development
MMR Mixed Methods Research
NGO Non-Governmental Organisation
PPI Progress out of Poverty Index
PPP Public-private Partnerships
PRAAD Public Records and Archives Administration Department
SAP Structural Adjustment Program
SD Standard deviation
SGBA Sustainable Greening and Beautification of Accra
SL Sustainable Livelihoods
SWOT Strengths, Weaknesses, Opportunities, Strengths
TA Thematic Analysis
TCPD Town and Country Planning Department
UK United Kingdom
UN United Nations
US(A) United States (of America)
USDA United States Department of Agriculture
Y4GG Youth4GreenGhana
YEA Youth Employment Agency
 1 INTRODUCTION

“I think there is a paradigm shift now worldwide, and we are talking about
globalisation. So Accra is not an isolated place, we are battling with resilience
[…], and greenery is an important aspect of it. […] the green is very, very im-
portant to us and we want to really revitalise Accra with respect to greenery,
and I have told you the mayor’s vision is to make sure that all the open spaces
we develop it into greenery and also trying to plant and nurture trees along the
streets in the communities. So that is the vision. So greenery is very, very im-
portant.”

Mr Yeboah, Department of Town and Country Planning (TCPD)1

1.1 Introduction
Freshly harvested coconuts have been loaded into the trunk of a taxi cab in Accra, the capital city
of the Republic of Ghana in West Africa (Fig. 1).

Figure 1 Freshly harvested coconuts in the trunk of a taxi in Mamobi

Own picture

1 See Section 3.3.4 for a complete list of interviewees and dates.

1
Neither the coconuts nor the cab in and of themselves would normally attract much attention;

both are familiar sights in the city. Their concurrence in place and time, however, embodies the

duality of cityscapes, in particular, how ‘nature’ assumes an intriguing double meaning in refer-

ences to ‘the nature of cities’. The image is both thought-provoking and serves as a visual intro-

duction to, and representation of, the subject matter of this thesis: urban society-nature relations

and the role of context that both shapes and is in turn shaped by these relations. This introducto-

ry chapter thus presents the rationale for this project; outlines its aims, objectives and research

questions; and discusses the overall research approach. Finally, it summarises the original contri-

butions of this work.

First, however, I narrate how the coconuts ended up in the taxi’s trunk as but one example of

stories I was told and events that I witnessed. These anecdotes should be given due space as they

illustrate otherwise easily overlooked society-nature interactions in cities and the urban trees’ mul-

tiple functions. Coconut palms are of slender growth but can easily reach heights of 20 metres or

more. In a low-rise neighbourhood like Mamobi, located in Accra’s centre, they are visible from

afar as they grow above even the highest walls and fences. I spotted the man in the palm tree be-

fore walking through the gate leading into the compound of one of the households (HH) that

had agreed to participate in my case study (see Section 3.3.5). I paused and admired how the man

was able to hold onto the tree trunk with just his bare hands and feet while comfortably cutting

off ripe coconuts with his cutlass several metres above the ground. I watched him harvest coco-

nuts from a few more trees as the tree owner, John, also observed from the foot of the tree (see

Section 6.3.2).

By sheer good fortune, I had arrived on a day when it was possible to observe coconut harvesting

in progress and talk about it with the tree owner and the coconut harvester/roadside seller. In

addition to the fruits, the coconut picker also cut off some of the old, dry fronds which he threw,

2
just like the coconuts, to the ground where John, who is also the compound’s landlord, assem-

bled them in piles. The fruit picker stopped when about 35 fruits lay on the ground and, having

returned safely to the ground, started to trim bits of stalk from the tops of the fruits. John left

but returned shortly after with a large metal washtub in which he placed coconuts he had selected

to keep for himself. After the harvest was divided up between the tree owner and fruit picker, the

latter disappeared, and John sat down on a bench to provide more information about his palms

and the coconut sale.

Coconut sellers regularly visit his compound to obtain fruits which they then subsequently sell to

buyers along the roadside. While three of his clients are repeat buyers, most of the people he sells

to are one-off visitors who enquire about the possibility of buying fruits when they notice the

palm trees in his compound as they walk by. Some of the roadside coconut sellers who buy from

him harvest the fruits themselves, while others hire experienced harvesters. Either way, the owner

told me, the decision about which fruits were ready for harvesting was entirely up to the coconut

picker, who is experienced enough to identify ripe fruits. He explained how it had been his father

who originally planted the land with about 90 coconut seedlings, not all of which have survived.

His father had brought them from his home village in the Volta Region, where his grandfather

grew palms on cleared forest land (Fig. 2). His father transported the seedlings on a bus, wanting

to raise them to produce coconut oil which his wife, the landlord’s stepmother, extracted for sale.

Additionally, the palms were helpful in establishing claim to, and demonstrating ownership of

land, while the fronds were used as fencing material. In the late 1990s, after inheriting the land

from his father, he cut down many of the palms to make space for accommodation that he want-

ed to rent out. He and his wife would also discontinue the production and sale of coconut oil as

selling the coconuts as fresh fruit was more convenient. They also installed a low-maintenance

metal fence to replace the palm frond fencing a long time ago.

3
 Figure 2 Location of Volta Region (grey)
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

A couple of minutes after we finished talking, the coconut picker returned in a taxi and filled its

trunk with the fruits he had harvested earlier. The car drove off and took him to his vending cart

from which he sells the coconuts to the city’s inhabitants. This story not only illustrates the mul-

tiple types of human–tree interactions that we can encounter in urban settings but hints at the

various functions of trees and their benefits as an essential component in many human–human

interactions as well. Furthermore, this example also highlights the importance of concepts like

‘time’ and ‘space’ for understanding ‘the nature of cities’.

4
1.2 Problem Statement
The world’s urban population is projected to increase from 55 % in 2018 to 68 % by 2050 (Unit-

ed Nations (UN), 2019). As a growing proportion of people reside in town and city environ-

ments, it is essential to understand urban living conditions and experiences. Trees have always

been a component of human settlements (Konijnendijk et al., 2006; Hauer et al., 2017). Like trees

in rural areas, urban trees are valued for their multiple ecosystem services and the resulting bene-

fits that humans can obtain from them (Nowak and Crane, 2002; Sullivan et al., 2004; Tyrväinen

et al., 2005; Yang et al., 2005; Nowak et al., 2006; Lin and Lin, 2010). They help to reduce heat

stress, air pollution, or flooding events (Mage et al., 1996; Gabriel and Endlicher, 2011; Rain et al.,

2011). These problems are often especially pronounced in cities with their specific configurations

and characteristics, such as compacted and sealed soils or exhaust fumes from vehicles and facto-

ries. Research predicts climate change’s adverse effects to worsen (Intergovernmental Panel on

Climate Change, 2013), creating increasingly challenging urban living conditions, especially for al-

ready disadvantaged groups who often reside in the most hazardous areas (Dionisio et al., 2010;

Amoako and Inkoom, 2018; Mehrotra et al., 2018). While urban trees may provoke allergic reac-

tions or litter the ground with their leaves (Lohr et al., 2004; Escobedo et al., 2011; Gómez-

Baggethun and Barton, 2013; Lyytimäki, 2014), careful decision-making and appropriate species–

site matching can reduce or altogether avoid these disservices. Maintaining existing trees and in-

corporating them into future developments is, therefore, a viable option in attempts to create liv-

ing conditions conducive to the mental and physical health of city residents.

As a global scale concern, the ecosystem services of urban trees, which can mitigate climate

change’s adverse effects, have essentially rendered invisible the trees’ (re)productive functions at

the more local scale, like the provision of food and medicine. There is thus a definite need to in-

vestigate how urban trees and their benefits fit into the broader system of individual and house-

hold production and social reproduction. This perspective draws attention to the provisioning

5
benefits of a city’s trees and their potential value, especially to residents of lower socio-economic

status, a topic which has recently received attention in South Africa (Shackleton et al., 2014; Ka-

oma and Shackleton, 2015; Shackleton et al., 2015). Generally, a more detailed understanding of

tree usage priorities and management approaches related to the privately-owned proportion of

the urban forest, which often constitutes the larger part of the total urban forest ‘estate’, is desir-

able (City of Alexandria, 2009; City of New Westminster, 2015). While not empirically or ‘scien-

tifically’ established during this study, extensive anecdotal evidence obtained during data collec-

tion in Accra indicates an ongoing reduction in the city’s green cover and the associated decline

in its tree population. It is, therefore, necessary to situate urban forestry within broader processes

of urban and regional planning and development and to analyse its relation to neoliberalism,

globalisation and climate change adaptation in recent decades.

1.3 Aims and Objectives

In this research project, I aim to generate a comprehensive account of the history of urban for-

estry in Accra, investigating the form and functions of the city’s trees over time. While I identi-

fied first descriptions of Accra’s trees dating back to the early 18th century, information for the

19th century and later is more widespread and, therefore, the focus of my work.

I further seek to challenge current policies and associated planning and management priorities,

using a scale- and place-sensitive approach that demonstrates how present activities reinforce ex-

isting social disparities. Thereby, urban forestry in Accra forgoes the opportunity to more effec-

tively utilise urban trees’ multiple benefits to simultaneously improve individuals’ well-being while

addressing pressing issues of global scope, such as the harmful effects of climate change.

To achieve these aims, my objectives were as follows:

6
1. Analyse archival documents to produce a coherent account of the history of urban forest poli-

cy and practice in Ghana, with particular reference to Accra.

2. Conduct a tree inventory and canopy cover analysis to document the structure of Accra’s con-

temporary urban forest and its connections to both historical and current events and processes.

3. Interview and survey key actors and stakeholders to analyse planning and management ap-

proaches and their effectiveness concerning identified strengths, weaknesses, opportunities, and

threats (SWOTs).

4. Observe and enquire into human–tree relations to document the well-being and livelihood

functions of urban trees in present-day Accra, with particular reference to residents in neigh-

bourhoods of modest means or low(er) socio-economic status.

5. Analyse contemporary urban forestry interventions for their (likely) effects on socio-

environmental conditions.

This research will generate previously unavailable information on neighbourhood-level canopy

cover and the urban forest of Accra that can act as baseline information in future planning and

management activities. In exposing the legacy effects of historical events and processes on the

present-day structure of the city’s urban forest, I hope to achieve two connected goals. First, ini-

tiate a discussion that will lead to a re-evaluation of the perpetuated narrative that marginalised

population groups are unaware of and responsible for existing intra-city variations of canopy

cover and its ongoing decline. Secondly, I suggest that practical urban forestry activities should be

adapted to address how current approaches systematically disadvantage particular categories of

individual citizens and groups. A more nuanced understanding of the multiple functions of urban

trees, including for human well-being and livelihoods, is not just relevant to Accra but also to the

broader discussion of urban lives and the role that natural elements and resources might play in

it.

7
1.4 Research Questions
In line with the above aims and objectives, the research questions that I address focus on four

broad themes/topics. First, I investigate the contextual background of urban forestry in the city

to answer the following question:

1. How have contextual factors affected the structure of Accra’s urban forest over time and

across space?

This section consists of an analysis of the key historical and contemporary state and non-state

policy interventions, as well as other social, legal, political, environmental, and technical variables,

including both one-time events and longer-term processes. As the process of ‘doing urban forest-

ry’ is integral to urban/city place-making, I will locate this analysis within the wider context of

urban and regional planning and development. This information can then be related to data col-

lected on the structure of Accra’s urban forest, including its spatial and temporal variation.

Second, I focus on currently dominant views on urban tree management and use, as well as the

policy and other interventions they inform, in answering the following research question:

2. How do individual actors and different stakeholder groups use and manage Accra’s trees, with

a view to identifying strengths, weaknesses, opportunities, and threats?

This section starts by identifying major actors and stakeholders and eliciting their views on the

purpose(s) of planning, managing, and ultimately using the urban forest. It compares and con-

trasts the opinions and actions of these groups and individuals, in addition to identifying the na-

ture of the interactions between them. Central to this part is an explicit focus on the (unequal)

distribution of decision-making power and its effects on urban forestry and its practitioners in the

city.

Third, I concentrate on and emphasise the perspective of individuals and households in investi-

gating the question:

8
3. To what end do residents in unplanned high-density, low-income neighbourhoods manage and

use public and private trees?

Here, I analyse the multiple well-being and livelihood functions of Accra’s trees for the city’s res-

idents, one of the principal groups of stakeholders and, where they grow trees on their private

land, not just stakeholders but also actors. To answer this research question, I investigate both

their perceptions on management and use as well as practical activities, mirroring research ques-

tion two. The section also contains a comparison between four case study stakeholders/actors,

drawing attention to the heterogeneous nature of this group. Findings from the previous section

will be integrated to highlight discrepancies of opinions and approaches.

The first research question investigates how past events and processes have shaped Accra’s con-

temporary urban forest structure. In this section, I seek to close the circle and investigate con-

temporary and proposed interventions and their potential future impacts. Thus, I ask:

4. How will current and proposed interventions and developments affect people’s ability to use

urban trees within a livelihood context?

After introducing five of these current and proposed interventions, I discuss their likely impacts

on individual and household capacity to benefit from a changing urban forest. I purposefully se-

lected these interventions to cover a range of actors and stakeholders, tree types, and stages in the

planning and management cycle. Based on the projected effects, I then offer suggestions as to

how tree planting might be facilitated and done in a way that may reduce tree cover disparities

and provide accessible trees to all, rather than further intensifying intra-city disparities.

1.5 Research Design

Based on the above-stated aims, objectives, and research questions, I have opted for a mixed

methods approach that combines qualitative and quantitative data collection and methods of

9
analysis in a single study (Tashakkori and Teddlie, 2003; Creswell and Plano Clark, 2007). I use

qualitative information from stakeholder interviews, archival research, and a household case

study component to complement quantitative data from canopy cover analysis, a tree inventory,

and household questionnaire survey. I align my project with Donna Mertens’ transformative par-

adigm (2008) in which the axiological belief – “enhancement of social justice, furtherance of hu-

man rights, and respect for cultural norms” (Mertens, 2010, p. 470) – is the foundation upon

which ontology, epistemology, and methodology are built. As such, the research explicitly focuses

on uneven power relationships that can be related to economic status, race/ethnicity, gender or

age (Mertens, 2012) and how they affect whose reality is privileged or normalised.

According to Mertens, the task of the researcher is not just to recognise existing inequities and in-

justices, but “to challenge the status quo” (Mertens, 2007, p. 212), a position which is reflected in

my intention to critically examine both the dominant view(s) of urban forestry and their differen-

tiated material and other outcomes in Accra. Mertens’ transformative paradigm, combined with

the conceptual framework I develop in the next chapter, made me adopt urban political ecology

as the main theoretical framework for this thesis. Through this transdisciplinary ‘lens’, researchers

seek to investigate “who produces what kind of socio-ecological configurations for whom”

(Heynen et al., 2006, p. 2) to challenge common, apolitical narratives (see Chapter 2, especially

Section 2.4).

1.6 Original Contribution

In this research, I move beyond intra-city comparisons of green space or tree cover and outline

the history of urban forestry in Accra. I thereby identify the lasting effects of previous events and

processes in creating current unequal spatial and other configurations in Accra’s treescapes. As

such, the study’s findings raise important questions about the adequacy of the currently favoured

10
narrative of some key state actors. Several of these actors suggest that canopy cover variations be-

tween neighbourhoods simply reflect how people ‘value’ urban trees – a ‘cultural’ preference that

can be influenced or countered via education and persuasion. However, combining several meth-

ods of data collection, I generate results, which suggest the need for re-contextualising such

views. I demonstrate how residents in lower socio-economic neighbourhoods are neither una-

ware nor lack knowledge of the benefits and importance of trees. Thus, they are not indifferent

to planting, protecting, and maintaining them, but have often found themselves having to priori-

tise other livelihood pursuits instead. As a result, this research challenges both common stereo-

types, as well as the efficiency, appropriateness and responsiveness of government institutions’

interventions designed to focus inordinately on ‘public’ education and information campaigns.

The findings from my case study with four households further extend our knowledge of urban

human–tree interactions and the well-being and livelihood functions of the city’s urban forest.

Tangible tree products not only strengthen human capital, such as through the health benefits of

consuming fresh fruits or accessible medicinal products but also play a central role in gifting and

exchange situations. As part of these, tree products are an integral element in creating and main-

taining social capital, a type of resource that can serve to increase livelihood security. These tan-

gible products, however, are not the only tree benefit of use in producing social capital as the

plants themselves, and most commonly the pleasant environment they create through their

shade, act as ‘place-makers’ in which social interactions can take place. I suggest that these in-

sights add to our understanding in a novel way and require further investigation in diverse con-

texts to generate information capable of challenging basic tenets and widely-held assumptions,

which have long functioned as drivers of urban forest planning and management.

11
1.7 Thesis Structure
In the next chapter, I use the existing literature to construct a conceptual framework and intro-

duce the relevant theories which have informed my analysis. This framework then serves as a

starting point to review previous research, demonstrating the gaps in our current knowledge and

the research questions that I, therefore, ask to address these gaps. Chapter 3 outlines my research

methodology, demonstrating why, based on the research questions, I have opted for a mixed

methods research (MMR) approach. This chapter also includes a comprehensive discussion of

the individual methods adopted. The discussion Chapters 4 to 7 are devoted successively to the

four different research questions identified earlier (see Section 1.4). In Chapter 4, I trace the his-

tory of urban forestry in Accra, linking it to the present-day structure of the city’s trees. This

chapter is a modified and extended version of a previously published paper (Hosek, 2019). The

following chapter focuses on how various actors utilise and manage Accra’s forest, using an anal-

ysis of identified strengths, weaknesses, opportunities and threats to create a more comprehen-

sive and connected picture of urban forestry in Ghana’s capital. Chapter 6 includes a detailed

analysis and discussion of individual and household-level tree interactions and relations. In Chap-

ter 7, I introduce five current and proposed urban forestry-related activities and their expected ef-

fects on the city’s residents. The concluding chapter integrates the insights from earlier chapters,

highlighting the need for further research and proposing an actionable plan to evaluate current

urban forest planning and management activities and potential adjustments.

12
 2 CONCEPTUAL AND THEORETICAL FRAMEWORK AND LITERA-
TURE REVIEW

“[T]raditionally, the focus of Ghana’s Forestry was on timber production. And

much of the focus was on forest in the hinterlands, in the forest reserves, in
the off reserves. However, the trends are changing, and people are beginning
to study other components of the environment, other than just timber, and so
urban forestry is fairly new in the system.”

Researcher, Forestry Research Institute of Ghana (FORIG)

In this chapter, I develop a conceptual framework that illustrates how events and processes at

multiple scales directly and indirectly affect urban forestry (Fig. 3, left side of diagram). The re-

sulting urban forest structural and functional configurations then determine who benefits where,

when and how, as the right half of Fig. 3 indicates.

Figure 3 Conceptual framework

It differs from other frameworks used to discuss ecosystem services in two ways. First, it is urban

forestry specific and, second, explicit about ecosystem management’s human dimension. Starting

13
at ‘urban forestry’, I trace the path from urban trees, an inherently natural resource, to their socie-

tal benefits. After that, I again start at ‘urban forestry’, this time to focus on management aspects.

Here, I consider actors and stakeholders, their roles, and the broader context in which they oper-

ate. I then combine both components to illustrate why a political ecology lens is suitable for this

project. Finally, I utilise five components of the framework to review the relevant literature, thus

identifying gaps in our knowledge and demonstrating how my research questions address these.

2.1 Urban Forestry

Urban forestry is no curious contradiction in terms but recognises the hybrid nature of cities

where trees are as present as brick and mortar. Referring to the process of managing urban trees,

its professional origin has been traced back to the late 1800s (Konijnendijk et al., 2006) though

trees have been planted in cities since ancient times (Hauer et al., 2017). Historical overviews exist

for countries such as the United States (US) (Johnston, 1996), Ireland (Johnston, 1997a), Britain

(Johnston, 1997b; Johnston, 1997c), Canada (Rosen, 2015), Singapore (Webb, 1998), or regions

like Europe (Lawrence, 1993; Forrest and Konijnendijk, 2005). Though non-exhaustive, this list

reflects the geographical foci of urban forestry research. Different local approaches and profes-

sionals from various backgrounds have caused inconsistent terminology (Konijnendijk et al.,

2006), thus generating multiple definitions of ‘urban forestry’ (Hauer, 2005; Brown, 2007). A

popular definition considers it to be the “art, science, and technology of managing trees and for-

est resources in and around urban community ecosystems for the physiological, sociological,

economic, and aesthetic benefits trees provide society” (Helms, 1998, p. 193). I use this definition

as it refers to both the theoretical and applied components of urban forestry, its natural and so-

cial science base, and arboricultural maintenance. To achieve holistic planning and management,

these elements need to complement each other. Furthermore, it asserts the existing relationship

between trees and people, with the provision of services to society regarded as the fundamental

14
purpose. This link must be examined in detail to specify what ‘managing’ the trees involves (Fig.

4).

Figure 4 Framework – Part I

2.2 Zooming in
2.2.1 The Urban Forest
What constitutes an urban forest has been defined in various ways, not least due to the ambigu-

ous nature of the two contributing words. Some entertain a broader approach, defining the urban

forest to consist of all open and green spaces containing woody vegetation (Pütz et al., 2015).

Others contain a stronger tree focus, suggesting it includes forested stands like the ones in rural

areas, but also “trees found along streets, in residential lots, in parks, and in other land uses”

(Nowak, 2016, p. 37). Though in both perspectives, the urban forest is formed by the same trees,

I suggest that the definitions differ and likely evoke distinct management approaches and re-

search questions. Trees in cities, undoubtedly, form part of a comprehensive green network, in-

cluding grasslands, herbaceous, and other vegetation covers. In this study, however, a tree-

focused approach permits investigating their specific roles and functions, many of which other

forms of greenery do not offer (Fig. 5).

Figure 5 Framework – Part II

The urban forest manager or researcher faces no shortage of tree-centred definitions of what ex-

actly they try to manage or research. I chose the definition of Miller et al. (2015, p. 4) as it illus-

trates two critical aspects. They state that the urban forest is “the sum of street trees, residential

trees, park trees, and greenbelt vegetation”, but specify that “[i]t includes trees on unused public

15
and private land, trees in transportation and utility corridors, and forests on watershed lands”.

This highlights that urban trees grow on both public and private lands, a conceptually vital dis-

tinction (see Section 2.3.2). Furthermore, “[s]ome of these trees and forests were willfully planted

and are carefully managed by their owners, while others are accidents of land-use decisions, eco-

nomics, topography, and neglect” (Miller et al., 2015, p. 4). The urban forest, therefore, is created

by multiple interacting variables. In this research, I use the term ‘tree’ in a deliberately broad

sense to include plants with a tree-like habit, such as Arecaceae (palms). Though they differ ana-

tomically, palms commonly grow in Mediterranean and (Sub)tropical cities and are functionally

similar enough to actual trees to be included in this project.

2.2.2 From Tree to Value

Urban trees are elements of a socio-ecological system. A biophysical structure or process (such as

an urban forest) performs specific functions, thereby creating services that benefit people and can

be economically valued (Potschin and Haines-Young, 2011). This connection is generally accept-

ed but deceptive in its seeming simplicity. The underlying terminology is debated as key words

are redefined, combined, split up, or replaced by alternatives (Balmford et al., 2008; Fisher et al.,

2009; de Groot et al., 2010a). Similarly, some criticise widely used typologies for “mix[ing] pro-

cesses (means) for achieving services and the services themselves (ends) within the same classifi-

cation category” (Wallace, 2007, p. 236). Examples of this include pollination, water regulation or

photosynthesis, which rarely are management objectives and usually just serve to obtain ‘actual’

services like food or potable water. From an economic perspective, this constitutes a valid argu-

ment as double-counting and false trade-offs in decision-making should be avoided (Boyd and

Banzhaf, 2007). However, I agree with those favouring coexisting classification systems (Con-

stanza, 2008; Fisher and Turner, 2008) as they need to be specific to the investigated ecosystem

and the context in which it is studied (Fisher et al., 2009). To illustrate the path from tree(s) to

16
value, I now describe, define and link the key urban forestry-related concepts, creating an appro-

priate framework for this project.

Urban forest managers require detailed information about the current tree stock to develop strat-

egies, monitor the process, and evaluate its outcomes. Attributes such as “number of trees, spe-

cies composition, tree size, health, tree location” (Nowak et al., 2008, p. 347) and diameter at

breast height (DBH) or crown dimensions (Nowak, 2008) describe a city’s forest structure and

the spatial distribution of individual trees. This knowledge forms the basis of decision-making,

and these decisions will, in turn, affect both specific trees and the forest as a whole (Fig. 6).

Figure 6 Framework – Part III

The trees then perform several ecosystem functions, which are the “physical, chemical and bio-

logical processes occurring in ecosystems that are necessary for its self-maintenance” (Dobbs et

al., 2011, p. 196). Among others, this includes nutrient and carbon cycling or the regulation of the

atmosphere, microclimate, carbon emission, and the hydrological cycle (Ordóñez and Duinker,

2012) (Fig. 7).

Figure 7 Framework – Part IV

These functions, in turn, produce ecosystem services that link the trees’ natural processes to hu-

mans and their well-being. While not a new concept, it started to receive increased attention

when the Millennium Ecosystem Assessment (MEA) was published in the 2000s. In this report,

ecosystem services are, rather vaguely, defined as “the benefits people obtain from ecosystems”

(Millennium Ecosystem Assessment, 2003, p. 53) and classified into four categories based on

17
function. The supporting services include soil formation, nutrient cycling, and primary produc-

tion. The other three types – provisioning, regulating, and cultural services – produce various

tangible and intangible benefits, such as food or recreational places. Like other common defini-

tions (Constanza et al., 1997; Daily, 1997), the MEA also illustrates the difficulties of clearly sepa-

rating ecosystem functions, services, and benefits and providing self-contained descriptions for

each. As I agree with Brown et al. (2007, p. 332), who call the difference between function and

service “more than semantic”, I follow their approach in distinguishing between these two con-

cepts (Fig. 8). Their definition of function concurs with the above-cited by Dobbs et al. –- “phys-

ical, chemical and biological processes occurring in ecosystems that are necessary for its self-

maintenance” – while ecosystem services are “the specific results of those processes that either

directly sustain or enhance human life” (Brown et al., 2007, p. 332).

Figure 8 Framework – Part V

Besides the MEA’s typology, ecosystem services are categorised according to functional (de

Groot et al., 2002), organisational (Norberg, 1999), or descriptive groups (Moberg and Folke,

1999). For this study, I adapt MEA’s scheme and consider what they call supporting services to

be ecosystem functions instead. Despite this change, I consider the remaining groups – provi-

sioning, regulating, and cultural services – adequate categories for this research. However, I want

to highlight Norberg’s (1999, p. 185) typology criterion that considers the “level of ecological hi-

erarchy” at which services are produced and maintained. These levels range from the individual

through to species, functional group, community, ecosystem, landscape and to the global cycle.

Planning the urban forest requires awareness of the trade-offs between these scales and their rela-

tion to management objectives.

18
The urban forest’s provisioning, regulating, and cultural services produce numerous tangible and

intangible benefits that help to maintain or increase human well-being. Specific benefits include

the provision of food or fuelwood, positive health impacts from reduced temperatures or air pol-

lution, as well as spaces for recreation, education, or religious activities (Fig. 9).

Figure 9 Framework – Part VI

To illustrate the differences and relationships between function, service, and benefit, Escobedo et

al. (2011, p. 2080) provide a helpful table from which I present some examples (Table 1).

Table 1 Examples of ecosystem functions and resulting services and benefits

Ecosystem Function Ecosystem Service Benefit

Primary productivity, biodi-

Natural areas for human use Outdoor recreation
versity

Atmospheric deposition, fil-

Reduced mortality and mor-
tering and interception of Air quality improvements
bidity related to air quality
pollutants

Primary productivity, nutrient

Production of grains, fruits,
cycling, pollination, soil
nuts, and seeds; water availa- Food harvests
productivity, disease regula-
bility
tion
Adapted from Escobedo et al., 2011, p. 2080

My framework so far replicates the ‘cascading model’, initially proposed by Haines-Young and

Potschin (2010), which depicts the same path from ecosystem structure to benefit. In this model,

and its adaptations (de Groot et al., 2010b), ecosystem services are located between what is called

‘ecosystems and biodiversity’ and ‘human well-being’. While the first includes ecosystem structure

19
and functions, the second refers to benefits and value. It illustrates the more than semantic dif-

ference between ‘function’ and ‘service’, but I disagree with the similarly rigid separation between

‘service’ and ‘benefit’. The anthropocentric position from which both ‘service’ and resulting ‘ben-

efit’ are defined (see Table 1 above) closely links the two. An ecosystem service is a ‘service’ pre-

cisely due to the benefit that humans can obtain from it, and therefore, I suggest, conceptually

more related to the socio-cultural component of the scheme, if not a part of it. This anthropo-

centric basis has been criticised for disregarding nature’s intrinsic value, instead commodifying it

to calculate its economic worth (Schröter et al., 2014) (Fig. 10).

Figure 10 Framework – Part VII

2.3 Zooming out

2.3.1 Actors and Their Interventions
In addition to zooming in from ‘urban forestry’ to ‘value’, we can also zoom out from ‘urban for-

estry’, considering the broader planning and management context of this natural resource. Miller

et al.’s definition has already suggested that some urban trees are ‘carefully managed’, implying

that individuals and groups exist who are involved in ‘doing’ urban forestry. These include both

state actors, like government departments and agencies, and non-state actors. According to the

African, Caribbean, and Pacific–European Union Partnership Agreement 2000/483/EC, also

known as the ‘Cotonou Agreement’ (2000, Article Six), non-state actors include the private sector

and civil society like religious groups, sport and cultural organisations, Non-Governmental Or-

ganisations (NGOs), trade unions, research institutions, or the media. For this project, I also clas-

sify individuals and households as non-state actors. While all other actor types consist of these

individuals, the capacity in which they serve concerning a specific issue functionally differentiates

them. In urban forestry, actors like NGOs provided saplings and tree planting support to indi-

20
viduals and households (Lipkis and Lipkis, 1990) or local governments (Pincetl, 2010). The media

was involved in raising awareness and communicating events (Ottitsch and Krott, 2005), while

municipal governments collaborated with the private sector to plan and conduct city greening ini-

tiatives (McPhearson et al., 2010). These examples demonstrate how state and non-state actors in-

tervene in urban forestry directly (i.e. planting, maintaining, utilising, or felling trees) and indirect-

ly (e.g. communicating planting events; formulating regulations, laws, plans and policies; or form-

ing partnerships).

Returning to the definition of ‘urban forest’, we can recall that managed trees are located on both

private and public lands. This requires a conceptual and practical separation of how various ac-

tors can participate in urban forestry on land under different ownership. While state actors can,

directly and indirectly, intervene in public urban forestry, they can affect private urban forestry

only indirectly through laws and regulations. Land owning individuals and households, i.e. non-

state actors, are usually the only type that can intervene directly in private urban forestry. Other

non-state actors can indirectly intervene in both public and private urban forestry while their di-

rect engagement in either is contingent on permission from governmental organisations or pri-

vate individuals (Table 2). Numerous actors combined with varying land ownership types then

cause the complex nature of holistic, city-wide urban forest planning and management.

Table 2 Involvement in urban forestry based on actor type and land ownership

Public Land Private Land

State actors direct and indirect indirect (and direct1)

Non-state actors – groups and direct1 and indirect direct1 and indirect
organisations

Non-state actors – individuals direct1, 2 direct

and households
1 with permission and/or depending on laws, 2 depending on activity, direct involvement may be illegal

21
The question of land ownership, and often by extension tree ownership, is essential to urban for-

estry. Though varying between locations, trees on private lands contribute substantially to a city’s

green cover. In various British towns, an iTree Eco survey (see Section 3.3.1.2) estimated that be-

tween 25 % and 79 % of trees were in private ownership (Vaz Monteiro et al., 2020). According

to iTree Canopy assessments (see Section 3.3.1.1), private lands contained 61 % of total canopy

cover in New Westminster (Canada) (City of New Westminster, 2015) and even 90 % in Alexan-

dria (United States of America) (City of Alexandria, 2009). As private property contains a large

proportion of urban trees, state actors require efficient systems to indirectly intervene in private

urban forestry to meet city-wide management objectives. Stronger coercive and punishing ele-

ments might likely be required the more their objectives diverge from or interfere with non-state

actors’ views, especially those of individuals and households. These complexities surrounding

land/tree ownership, actor types, and modes of involvement require being explicit about what

trees, actors, and actions are under investigation.

2.3.2 Acting within Context

Urban forestry is practised within a specific environmental, legal, technological and socio-political

context. These factors are both internal to the actors, e.g. the number of employees or available

technology, and external, such as soil or climatic conditions. Additionally, decisions made and ac-

tions taken by seemingly uninvolved groups can have unintended or undesirable effects on plan-

ning and management or the trees directly. As urban forestry is embedded within the broader

context of urban development, it also competes with city planning’s other objectives. Common

aims include efficient cities that can ensure economic growth, feature inclusive communities and

conserve natural, built and historic resources (Campbell, 1996; Couch, 2016). These complemen-

tary goals are often contradictory in terms of their implementation, thus requiring trade-offs

(Campbell, 1996). Road constructions that compact soils and trenching for underground utility

22
pipes can impede tree root health (Jim, 2003; Day et al., 2010). Furthermore, urban growth re-

quires an adequate housing supply that can reduce space for other land cover types (Pauleit et al.,

2005). The effects of events and processes at the regional, national or global scale, such as new

laws or technological innovations, may also trickle down and, directly or indirectly, alter a city’s

urban forest (Fig. 11).

Figure 11 Framework – Part VIII

2.3.3 Inheriting (from) the Past

Urban forest structure is not just affected by processes at different scales but also by historical

human–nature interactions. Any investigation aiming to understand current configurations

should evaluate these feedback mechanisms. The study of such interplays started in the 18 th cen-

tury, formalised as the discipline of historical ecology in the 1960s (Szabó, 2015). Lunt and

Spooner (2005, p. 1860) describe it as “a new paradigm in which ecologists view ecosystems as

historically and spatially influenced non-equilibrium systems that are complex and open to human

inputs”. Balée (2006), in turn, considers these past interactions not just crucial for our under-

standing of contemporary landscapes but also for their effects on the formation of culture. His-

torical ecology, however, not only facilitates an understanding of the present, as contemporary

23
landscapes are not endpoints but rather intermediate points in a continuing development. Evalu-

ating how the past has influenced the present facilitates anticipating the future. This ‘predictive

endeavor’, as Christensen (1989) calls it, often includes developing models that provide scenarios

for ecosystem changes or future land-use patterns based on historical information. These predic-

tions also “explain or [to] forecast spatial differences in system structure or function” (Pickett et

al., 1994, p. 139), merging space and time and illustrating their interdependencies.

Early interest in human–environment interactions was especially strong in forestry as “foresters

deal with living organisms that have life cycles much longer than humans” (Agnoletti, 2000, p. 2),

and past conditions play “a significant and continuing role in the structure of most forest ecosys-

tems” (Christensen, 1989, p. 116). Though rural trees tend to outlive their urban counterparts

that grow in more challenging conditions, city trees nonetheless grow for extended amounts of

time. While their life spans vary based on planting site (Roman and Scatena, 2011), past events

and processes affect current configurations. The legacy effects of human and biophysical factors

have recently been reviewed (Roman et al., 2018, p. 160), corroborating that “historical processes

must be considered in any investigation that seeks to explain how urban forest structure and spa-

tial patterns emerged”. My framework, therefore, needs to reflect the importance of understand-

ing past developments and path dependencies (Fig. 12).

24
 Figure 12 Framework – Part IX

2.4 The Larger Picture – Political Ecology

My framework illustrates how two seemingly distinct parts are connected. The framework’s right

side focuses on trees and their characteristics, natural elements in a biological and ecological

sense. On the left side, however, socio-political and historical themes dominate, constituting the

context within which the former is planned and managed. The far-right side, where ecosystem

functions turn into human benefits, further emphasises how urban trees are both natural and so-

cial elements. To examine Accra’s trees in context, I adopt the lens of political ecology, “a trans-

disciplinary research field addressing nature-society interrelations” with an explicit focus on social

inequality and asymmetries of power (Karlsson, 2015, p. 350). Political ecologists challenge

common narratives, focus on the social production of the environment and “identify broader sys-

tems rather than blaming proximate and local forces” (Robbins, 2012, p. 13). In apolitical expla-

nations, these local forces are often marginalised groups whose behaviour is regarded as irrational

or ignorant. Environmental degradation or specific spatial structures of animal biodiversity, forest

cover or land use are then blamed on their activities, disregarding alternative interpretations.

While political ecology has been accused of “lacking a coherent theory” (Karlsson, 2015, p. 350),

eclectically borrowing from e.g. (Neo-)Marxism, feminist studies, post-structuralism, or post-

25
colonial studies, it offers an alternative reading of environmental management and transfor-

mations.

Political ecology initially focused on rural areas due to its intellectual origins in human and cultur-

al ecology, their proponents “[t]ypically working in rural and agrarian Third World societies”

(Peet and Watts, 1993, p. 239). However, an urban version has since emerged, considering cities

to be “dense networks of interwoven socio-spatial processes that are simultaneously local and

global, human and physical, cultural and organic” (Heynen et al., 2006, p. 1). Inequalities in cities,

therefore, arise from unjust processes caused by unequal distributions of power and reveal them-

selves both in the presence and location of ‘the bad’ and ‘the good’. Thus, in Accra, poorer peo-

ple inhabit the area of Agbogbloshie, which contains a toxic e-waste dump for equipment origi-

nating from all over the world (Akese and Little, 2018). Meanwhile, a functioning sewage system

predominantly serves households in affluent neighbourhoods (Boadi and Kuitunen, 2005).

Urban political ecologists have researched city trees and their management, acknowledging “that

the urban forest is a political and social construct, a creation of people’s preferences and choices”

(Sandberg et al., 2015, p. 6). McLain et al. (2012) expressed similar views, arguing that urban forest

configurations are never neutral but exhibit how dominant voices and interests prevail over oth-

ers. Urban forest structure, because of human decision-making, is inherently political. If the

structure is political, so are the resulting functions, services, and benefits. Within cities, structural

elements, like canopy cover or species diversity, often correlate with socio-economic variables,

like household income (Pedlowski et al., 2002; Kuruneri-Chitepo and Shackleton 2011; Schwarz et

al., 2015). While residents of affluent neighbourhoods benefit from the surrounding trees’ ser-

vices, inhabitants of less wealthy areas are disadvantaged twice. Besides fewer available ecosystem

services, they also suffer from the harmful effects of poorer urban living conditions, residing in

hotter (Mehrotra et al., 2018) and more polluted areas (Dionisio et al., 2010) or neighbourhoods

26
prone to flooding (Amoako and Inkoom, 2018). Urban political ecologists, especially (Neo-)

Marxist ones, consider such manifestations to embody neoliberal socio-structural realities like in-

come inequality and resource ownership. In doing so, they insert ‘neoliberalism’ into the equa-

tion.

Neoliberalism, a political and economic theory and associated practices, entails “an institutional

framework characterised by strong private property rights, free markets and free trade” (Harvey,

2005, p. 2). Beyond maintaining or creating the necessary political-economic and other structures

to facilitate its operation in the way described, direct state intervention or participation in the

economy is kept to a minimum. Neoliberalism has “become hegemonic as a mode of discourse”

(Harvey, 2005, p. 3), having shaped global developments. Advocates of neoliberalism work in

many institutions, including the World Bank and International Monetary Fund (IMF), and have

disseminated their views widely, often at considerable expense (Cline-Cole and Harrison, 2010).

In 1983, Ghana launched the Structural Adjustment Program (SAP), introducing neoliberal re-

forms linked to loans from these lenders of last resort. While inflation rates reduced and the

gross domestic product grew, within-country inequalities increased and further marginalised al-

ready disadvantaged groups, including the urban poor, notably women (Brydon, 1999; Konadu-

Agyemang, 2000; Obeng-Odoom, 2012).

My framework demonstrates how neoliberalism and urban forestry connect as tree benefits turn

into (financial) value on the far-right side. Dollar prices are assigned to tonnes of stored carbon

dioxide (Nowak and Crane, 2002; Liu and Li, 2012) and garden or neighbourhood trees increase

property values (Anderson and Cordell, 1988; Sander et al., 2010; Escobedo et al., 2015). Addi-

tionally, “economic valuation provides monetary values for non-priced urban-forest amenities”

(Tyrväinen, 2001, p. 76) such as its recreational use (Dwyer et al., 1989; Tyrväinen and Väänänen,

27
1998). This process inserts nature into a neoliberal market structure where a tree benefit’s finan-

cial value is often a key consideration when actors prioritise management objectives.

2.5 Focus on the Framework’s Components

I now focus on the framework’s components to review the pertinent literature, thereby identify-

ing knowledge gaps while rooting the research questions within an appropriate theoretical and

conceptual structure. Though I favour comprehensibility and discuss individual elements sepa-

rately, the framework itself substantiates how such divisions are artificial.

2.5.1 Context SpecifiCITY

State and non-state actors plan and manage the urban forest within specific environmental, social,

legal, political and technological settings. Urban forestry is, therefore, highly context-specific and

practical experiences or study outcomes not necessarily transferable (Shackleton, 2012). Urban

forestry research in Ghana is limited, despite slight increases in the past five years. A non-

systematic literature search indicates a low double-digit number of publications, recording more

than five journal papers for 2017 and 2018 only. A sub-set of this research focuses on the pro-

motion and destruction of urban vegetation (Mensah, 2014a; Mensah, 2014b; Oduro-Ofori et al.,

2014; Arku et al., 2016; Nero, 2017; Owusu, 2018), some of which connected to intra-city vari-

ance and the population’s socio-economic characteristics (Stow et al., 2013; Yeboah, 2016). Oth-

ers collected species-level and individual tree information in Kumasi, about 250 km northwest of

Accra (Uka and Belford, 2016)(Fig. 13), calculated carbon stocks (Nero et al., 2017; Nero et al.,

2018a) or estimated the nutrients provided by food trees (Nero et al., 2018b). Researchers also

collected information on animals inhabiting the urban forest (Deikumah and Kudom, 2010). De-

28
spite the breadth of the covered topics and employed methods, the overall low publication num-

ber limits the depth of our understanding.

Figure 13 Location of Kumasi

Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

Urban forestry research in Ghana focuses on Kumasi, probably due to its historical ‘Garden City’

designation and as home to the country’s Forestry Research Institute (FORIG). However, some

publications devoted to Accra cover unique subjects, investigating food and nutrients sourced

from the city’s trees (Nero et al., 2018b) or the Achimota urban forest as a religious space

(Okyerefo, 2014; Okyerefo and Fiaveh, 2017). Other topics mirror the Kumasi-based research

and include changing land use, varying intra-city forest cover or neglected public parks (Stow et

al., 2013; Arku et al., 2016, Yeboah, 2016; Owusu, 2018). While population size and climate in

Kumasi and Accra resemble each other, the Köppen-Geiger classification assigning a tropical sa-

29
vannah/dry-winter climate to both, the cities are located in different agro-ecological zones (Ab-

bam et al., 2018) with different histories and cultures (Owusu and Agyei-Mensah, 2011). Setting

my research in Accra, I not only redress the current imbalance in Kumasi’s favour but also add to

a field where outputs on North American and European countries dominate (Roy et al., 2012;

Shackleton, 2012).

2.5.2 Urban Forest Structure

Urban foresters require basic information about the resource they manage, including the total

number of trees, their locations, species, size, and health. They collect this data during complete

or sample inventories, their scope defined by the requirements and available resources. Tree

stock is assessed through field-based bottom-up approaches, ‘remote’ top-down ones that use

aerial or satellite images, or a combination of both types. Forest structure has been studied all

over the world, from Nordic countries (Gundersen et al., 2005) to South Africa (De Lacy and

Shackleton, 2017) and from Chile (Escobedo et al., 2006) to Australia (Dobbs et al., 2013). Besides

their geographic range, these studies also utilised various methods, including aerial photographs

(Melbourne), postal questionnaires (Northern Europe) and field-based sampling (Grahamstown

and Santiago). Additionally, research may encompass all public and private trees, like in Australia,

or focus on public trees (Chile), sacred sites (South Africa) or woodlands (Nordic countries). Fur-

thermore, these four examples investigated diverse structural variables like canopy cover, species

composition and richness, tree density, stem diameter, or leaf area index. Indeed, these few se-

lected studies display the range of approaches available to investigate urban forest structure.

Most studies, however, not only assess forest structure but also investigate its correlation at sub-

city level with socio-economic variables like race/ethnicity, income, or educational level (Heynen

and Lindsey, 2003; Hope et al., 2003; Martin et al., 2004; Landry and Chakraborty, 2009; Lubbe et

30
al., 2010; Kuruneri-Chitepo and Shackleton, 2011; Schwarz et al., 2015). Generally, these structur-

al and socio-economic variables correlate, thereby producing an uneven urban forest. Though ef-

fects were weak in Kumasi (Nero, 2016), areas of higher socio-economic status in Accra feature

denser vegetation cover (Stow et al., 2013; Yeboah, 2016). Whereas structural causes may system-

atically (dis)advantage certain neighbourhoods, private land-use decisions, e.g. based on cultural

preferences or available resources (Lubbe et al., 2010), also affect a city’s urban forest.

In this project, I produce baseline information on Accra’s current urban forest structure. Previ-

ous research utilised satellite images to measure vegetation cover but also included bushes, hedg-

es, or grass areas (Stow et al., 2013; Owusu, 2018). These top-down approaches study the city’s

green space though not individual trees. Yeboah (2016), who researched how urban forest and

residential class interrelate, conducted a field-based assessment but only counted trees within

random plots. This approach fails to include essential components of a tree inventory like diame-

ter at breast height (DBH) or tree size (Leff, 2016). For a holistic picture, I combine desk- and

field-based data collection methods to analyse Accra’s urban forest structure. My neighbourhood

boundaries match those of an earlier ‘poverty mapping exercise’ (CHF International and AMA,

2010) which rated areas based on population and housing density or access to toilets and fresh

water (see Section 3.2). Combining these ratings with my urban forest structure data then allows

investigation of how urban forest structure and socio-economic variables correlate in Accra.

2.5.3 Path Dependencies and Legacy Effects

Contemporary urban forest structure reflects past events and processes. As trees are “long-lived

organisms that require decades to achieve mature size” (Roman et al., 2018, p. 158), their socio-

spatial distribution is clearly linked to the past (Boone et al., 2010; Grove et al., 2014). These lega-

cy effects “have too often been overlooked in analyses of urban forest spatial patterns” (Roman

31
et al., 2018, p. 166). When researchers offer historically contextualised explanations, their detail

and depth vary. Cursory statements observe how Accra lost much of its vegetation “when Ghana

embarked on [the] Economic Recovery Program (ERP) and its associated trade liberalization”

(Owusu, 2018, p. 43). A more nuanced analysis relates contemporary plant diversity to a steep so-

cio-economic gradient caused by the apartheid regime in South Africa (Lubbe et al., 2010, p.

2908). The third type, though, consists of comprehensive narratives, often covering US cities

(Nowak, 1993; McPherson and Luttinger, 1998; Merse et al., 2009). I now discuss examples from

Nairobi and Kumasi in detail to illustrate the complexities of path dependency.

Njeru (2010; 2013) examines how donor-driven neoliberal processes affected the Karura urban

forest in Nairobi. I selected this example because Njeru uses an approach they label ‘political

ecology’ in investigating the effects of SAP on urban forestry. Njeru argues that powerful interna-

tional donors forced both democratisation processes and the SAP on the country, thereby influ-

encing political and economic developments. Kenya depended on IMF loans, while the Moi re-

gime “creatively used the discourse of privatization (and its associated free market ideology) to

mask patronage connections underpinning the selling of the public forest reserve to private inter-

ests” (Njeru, 2010, p. 339). The funds raised were spent on a re-election campaign as donors had

forced Moi to adopt a multiparty system. A report by the succeeding government suggested re-

claiming the sold parts and protecting the forest, prohibiting human activity within its bounda-

ries. However, this disadvantaged the surrounding inhabitants of low-income housing areas

whose livelihoods depended on the forest, where they collected firewood, medicinal plants and

fodder. Despite their active protests against privatisation in the first place, they were later exclud-

ed in the name of environmental protection.

The second example is a recent investigation of Kumasi’s development ‘From the seat of a tradi-

tional Kingdom to a Garden city’ (Amoako and Adom-Asamoah, 2019). The authors investigate

32
how the city’s socio-political developments interact with green space management and attribute

changing landscapes to a conflict between what they call ‘traditional set-up’ and ‘state official-

dom’. The traditional set-up refers to the “well-developed socio-political structure with the office

of the Asantehene commanding absolute power and loyalty” (Amoako and Adom-Asamoah,

2019, p. 312). State officialdom, on the other hand, was introduced when the British defeated the

Asantes in the late 19th century and changed the city according to their needs. However, chiefs re-

turned most land to its original form and function when land ownership was restored to the

Asante Kingdom in the 1920s. British ad hoc city planning in the 1940s then created an ‘un-

planned hybridity’ where “[a] haphazard colonial urbanism overlay an older order, masking but

not erasing it” (McCaskie, 2000, p. 205). Furthermore, the planners took the Asantes’ support for

granted, overlooking that people would try to assert ownership over plots that could potentially

become valuable building land.

After independence, Kumasi City Council encountered similar problems as “chiefship and its

many allies worked to retain historic rights of disposition over land” (McCaskie, 2000, p. 230).

The dual management system, composed of customary and state institutions, remained in place,

the latter group trying to manage green spaces “which are usually lands created by the state but

owned by the powerful customary leadership.” (Amoako and Adom-Asamoah, 2019, p. 314).

State departments then cannot address encroachment and might be pressured into re-zoning

open areas into commercial or residential use when traditional authorities ally themselves with

prospective developers. Similar to Njeru, the authors comment on how ideas of preserving public

open spaces lacked the city residents’ inputs and “has been an imposition by state institutions

with little or no consultations” (Amoako and Adom-Asamoah, 2019, p. 323). Since 1980, Kumasi

has lost over half its open and green spaces, caused by conflicting needs and lack of cooperation

between the ‘traditional set-up’ and ‘state officialdom’. Despite the differences between Kumasi

33
and Accra, similar legacies might also affect the capital’s urban forest and its historical develop-

ment.

I now combine context specificity, urban forest structure and its path dependencies to form the

first research question. Urban forestry research has grown over the past 20 years but remains

dominated by European and North American study locations (Shackleton, 2012) and researchers

(Bentsen et al., 2010). As my framework illustrates, it is situated within a wider context, defined by

place-specific environmental, legal, political, economic, or social variables with feedback loops

across multiple scales. Despite “common threads in legacy effects across many cities for similar

historical time periods, neighborhood forms, and bioregional contexts, there can be idiosyncra-

sies in urban forest patterns and trajectories” (Roman et al., 2018, p. 165). These context-specific

characteristics require us to “recognize the crucial role of local sociopolitical history” (Roman et al., 2018,

p. 165). We, therefore, need to obtain baseline information on Accra’s urban forest structure and

examine how biophysical and human legacies have shaped it over time. Accordingly, the first re-

search question I address is:

1. How have contextual factors affected the structure of Accra’s urban forest over time and

across space?

2.5.4 Urban Forestry – Planning and Managing the Urban Forest

Urban trees naturally produce ecosystem services, but planning and management processes affect

their location, quality and quantity. Urban forestry links management and maintenance to desired

benefits and the required urban forest structure. ‘Sustainable management’, i.e. “maintaining

healthy and functional vegetation and associated systems that provide long-term benefits desired

by the community” (Dwyer et al., 2003, p. 49), is a core component of this process. Dwyer et al.

propose a sustainable management model with five relevant factors. These include the social con-

text, management goals and objectives, management outcomes, inventory and survey data, as well

34
as the “vegetation structure and/or management programs that have been identified as necessary

to sustain desired urban forest benefits” (p. 52). They recognise the central role of people in this

process, such as stakeholders’ participation and partnerships among decision-makers. However,

they conceptually disconnect urban forest managers from ‘the community’ for whose benefit the

managers develop strategies that are then being ‘administered’.

The model proposed by Dwyer et al. (2003) resembles ‘governance by government’ approaches

(Lawrence et al., 2013), where official bodies seek limited inputs from the community, thus over-

looking its central role. Urban forestry, though, is moving towards ‘governance with government’

(Lawrence et al., 2013; Ugolini et al., 2015). Environmental governance research, however, has re-

ceived limited attention despite investigating these “formal and informal institutions, rules, mech-

anisms and processes of collective decision-making that enable stakeholders to influence and co-

ordinate their interdependent needs and interests and their interactions with the environment at

the relevant scales” (Tacconi, 2011, p. 240). Reflecting this approach, Lawrence et al. (2013) de-

veloped a framework where detailed narrative information complements summary outlines of any

urban forest project. Combined with the ‘sustainable management’ model, this framework can

guide the analysis of urban forest management processes. Besides identifying stakeholders and

their relationships, it draws attention to the institutional and discursive circumstances in which

they define management priorities or monitor and evaluate outcomes.

2.5.4.1 Who?
The diverse nature of actors and stakeholders complicates identifying all involved groups and in-

dividuals, but always includes the general public, some individuals only as stakeholders, others as

both actors and stakeholders when they own land with trees. Depending on place and project,

further stakeholders may include governmental departments, environmental groups, researchers,

35
and arboriculturists, “all of whom tend to have different points of view, goals, perspectives, skills,

and means, which in many cases inhibit productive collaboration among them” (Ugolini et al.,

2015, p. 366). Besides generic stakeholder classifications (Miles, 2017) or simply distinguishing

between state and non-state actors, researchers have also developed context-specific ones. These

can be broad – Ugolini et al. (2015) identified three actor types (academics, public administrators,

and practitioners) – or detailed, a Swiss study identifying 13 stakeholder groups (Wilkes-Allemann

et al., 2015). Typical research topics include how specific actor/stakeholder types plan and man-

age the urban forest, interact with each other, or a combination of both.

Focussing on municipal urban foresters, Ordóñez et al. (2019) reviewed the literature to identify

their concerns and priorities. Out of 60 publications, however, only two were from an African

country (South Africa). Research has also investigated how private individuals view and maintain

their household trees (Summit and McPherson, 1998; Kirkpatrick et al., 2012; Kirkpatrick et al.,

2013a; Shakeel and Conway, 2014; Hungerford and Moussa, 2017). Private individuals also partic-

ipate in non-state and state actors’ urban forestry activities (Austin, 2002; Greene et al., 2011;

Gulsrud et al., 2018). Using Melbourne as a case study, Gulsrud et al. (2018) present an example

of successful and reflexive urban forest co-governance. Interested residents can become volun-

teer ‘citizen foresters’, participate in workshops or access tree data. When Melbourne faced water

shortages, residents dissuaded the city from removing trees to reduce irrigation needs, instead

prioritising their long-term benefits. Key to such co-governance is “coordination by local authori-

ties that recognizes the autonomy of the individual and their situated expertise in social-ecological

resilience” (Gulsrud et al., 2018, p. 165). While government departments are central to urban for-

estry (van der Jagt and Lawrence, 2019), they ultimately operate in a multi-stakeholder environ-

ment.

36
2.5.4.2 What for?
Urban tree benefits have received considerable attention, but a growing body of research also

points out potential disservices. How trees affect property values has been investigated widely

(Anderson and Cordell, 1988; Dombrow et al., 2000; Conway et al., 2010; Sander et al., 2010;

Dimke et al., 2013; Pandit et al., 2013), most studies concluding that house prices/rents positively

correlate with green cover. However, (densely) vegetated areas can increase fear of crime (Talbot

and Kaplan, 1984), but findings are mixed and actual crime rates often rather decreased (Kuo et

al., 1998; Troy et al., 2012; Gilstad-Hayden et al., 2015; Locke et al., 2017). Empirical evidence for

energy savings from trees’ cooling effects is less ambiguous but requires “planting the right trees

in optimal locations” to maximise this benefit (Ko, 2018, p. 324). This study, however, focused

on single-family, air-conditioned homes in North America, again underscoring how context mat-

ters in urban forestry.

The list of tree services further includes reduced groundwater run-off (Coder, 1996) and noise

(Dwyer et al., 1992), the provision of recreational space (Tyrväinen, 2001), air pollution intercep-

tion (Yang et al., 2005), or carbon sequestration (Nowak and Crane, 2002). They reduce tempera-

tures and heat stress (Akbari et al., 2001; Yu and Hien, 2006) but may also cause allergic reactions

(Cariñanos and Casares-Porcel, 2011) or attract dangerous and disease-bearing wildlife

(Lyytimäki, 2014). Trees damaging infrastructure, fertilisers and pesticides harming the environ-

ment and resources spent on maintenance operations are further disservices (Lohr et al., 2004;

Escobedo et al., 2011; Gómez-Baggethun and Barton, 2013; Lyytimäki, 2014). While cost-benefit

analyses could support decision-making, comprehensiveness is not easily achieved (Song et al.,

2018). Such calculations demand location-specific inputs (Williams, 2002; Avolio et al., 2015) and

multiple temporal and scalar trade-offs when selecting valuation methods and examined services

and disservices.

37
Urban tree benefits have been researched globally (Lohr et al., 2004; Schroeder et al., 2006; Peck-

ham et al., 2013; Camacho-Cervantes et al., 2014; Ordóñez and Duinker, 2014), but I now focus

on African countries to counter the existing geographical bias. Researchers inventoried urban

forests in Nigeria (Kayode, 2010; Borokini, 2012), Togo (Raoufou et al., 2011), and Ghana (Uka

and Belford, 2016). Uka and Belford (2016) identified 70 street tree species in Kumasi, fewer than

reported for Lomé (Raoufou et al., 2011). In Ekiti state (Kayode, 2010) and Ibadan (Borokini,

2012), species numbers ‘in and around households’ resembled those from Kumasi and respond-

ents in both Nigeria and Ghana listed fruits, medicine, fuel, shade, wind control, and cash income

as important tree benefits. Kayode (2010, p. 483), therefore, concluded that “the products deriv-

able from the trees influenced the choice of species planted by the respondents”. While Uka and

Belford (2016) recommend planting large trees in Kumasi to remove pollutants and sequester

carbon, Kayode (2010) emphasises how better meeting people’s needs could avoid repeating pre-

vious failures of urban tree planting exercises.

A South African photorealistic visualisation study (Donaldson-Selby et al., 2007) concurs with

Kayode’s (2010) conclusion, albeit based on opposite results. Residents in a low-cost, high-

density neighbourhood assigned lower ratings to landscapes with tangible benefits, like vegetable

gardens or fruit trees. Instead, they favoured flower gardens, shade trees and sports fields, which

surprised the researchers “given the socio-economic nature of the housing settlement” (Don-

aldson-Selby et al., 2007, p. 11). However, the participant number was low (n = 20) and the

study’s aim imprecise, not specifying if people were asked to rate the landscapes’ aesthetic appeal

or the benefits they would provide. Another South African study also illustrates how research de-

sign affects results. Socio-economically diverse respondents valued their towns’ green spaces for

recreation, job opportunities and human well-being, these intangible benefits probably dominat-

ing as green spaces include “golf courses, sport fields and other open nature areas” (Shackleton

and Blair, 2013, p. 106). Furthermore, responses to open-ended questions were classified post-

38
hoc and ‘job opportunities’ or ‘human well-being’ could include tangible products as research has

documented their importance for South Africans (Kaoma and Shackleton, 2014; Kaoma and

Shackleton, 2015; Shackleton et al., 2015; Gwedla and Shackleton, 2019).

Research on urban tree benefits is ubiquitous but geographically and topically incomplete. With

some exceptions (e.g. McLain et al., 2012; Clark and Nicholas, 2013; Poe et al., 2013; McLain et al.,

2014; Tiwary et al., 2020), research in Europe and North America has focused largely on intangi-

ble tree benefits. When tangible products are studied, it usually is in the context of public rather

than private trees. Almost a decade ago, Shackleton (2012) suggested investigating tree products

for their potential role in alleviating urban poverty, which requires a place-specific understanding

of the trees’ individual and household level functions. While Shackleton and colleagues have

started to investigate this aspect, their studies often focus on quantitative research approaches

(Kaoma and Shackleton, 2014; Kaoma and Shackleton, 2015; Shackleton et al., 2015; Ward and

Shackleton, 2016). This preference for the quantitative pervades research on urban tree benefits

and costs more generally as a review of 115 papers demonstrated that no study “used observa-

tions, content/text analysis or focus groups” (Roy et al., 2012, p. 356).

2.5.4.3 Under What Circumstances?

City residents’ views on tree values often differ from those of municipal and state actors (Hunter,

2001; Hofmann et al., 2012). While some cities have developed urban forest management plans,

they can fail to identify planning objectives (Ordóñez and Duinker, 2013; Gibbons and Ryan,

2015). Where these documents are unavailable, like in 93 % of surveyed South African municipal-

ities (Chishaleshale et al., 2015), urban forestry lacks transparency. Few studies (e.g. Kirkpatrick et

al., 2013b; Gwedla and Shackleton, 2015) have investigated tree professionals’ and government

departments’ management priorities, despite their views eventually being reflected in policies.

39
More commonly, researchers investigate the challenges these actors encounter. While Swedish

green space managers feel sufficiently funded (Randrup et al., 2017), they are an exception

(Stobbart and Johnston, 2012; Driscoll et al., 2015; Kronenberg, 2015). Facing resource con-

straints, managers forego regular tree monitoring (Ordóñez et al., 2019) and focus on reactive ar-

boricultural activities rather than pro-active schemes (Groniger et al., 2002) thus impeding long-

term sustainable urban forestry.

Besides limited budgets, stakeholder interactions, or the lack thereof, also complicate urban for-

estry (Ordóñez et al., 2019). As Dwyer et al. (2003, p. 51) argue, “[i]f management activities are

administered independently of community goals, they are not likely to represent the convergence

of what is socially desirable and ecologically possible”. Official bodies and the public are but two

groups whose relationships have been studied (Janse and Konijnendijk, 2007; Ugolini et al., 2015;

Kozová, 2018), researchers investigating differing priorities and the cost-benefit analyses they are

based upon. In Accra, heat, pollution, and flooding have caused health problems and deaths (Di-

onisio et al., 2010; Rain et al., 2011). I, therefore, expect government departments to focus on uti-

lising trees to reduce indoor and outdoor temperatures, stormwater run-off, and traffic-derived

pollutants and particulate matter. I further assume that official actors regard the benefits as ex-

ceeding the costs, though species-site matching and maintenance affect both. I propose the fol-

lowing research questions to investigate how stakeholders value, manage, and use the city’s

greenery:

2. How do individual actors and different stakeholder groups use and manage Accra’s trees, with

a view to identifying strengths, weaknesses, opportunities, and threats?

3. To what end do residents in unplanned high-density, low-income neighbourhoods manage and

use public and private trees?

40
2.5.5 Repeating the Process, Looking to the Future
Path dependencies have shaped contemporary urban forests, and current events and processes

will shape the forests of tomorrow. The global discourse on Millennium and Sustainable Devel-

opment Goals centres cities and their nature, the trees’ ecosystem services expected to mitigate

problems associated with population growth, urbanisation, and climate change (Endreny, 2018).

However, these same processes also threaten the urban forest, creating harsher growing condi-

tions while land-use conflicts can cause tree damage or removal. Additionally, underfunded urban

forestry programmes may “lack an overall vision about the values to be achieved by the tree

planting, and the urban forest design best fit to achieve these values” (Barron et al., 2016). Barron

et al.’s decision support framework breaks this vision down into specific strategies with aligned

indicators and targets but reflects how quantitative and positivist approaches dominate urban

forestry research (Ostoić and Konijnendijk, 2015), limiting individual or household level interpre-

tation and input.

When we predict future developments, proposed scenarios will tend towards the generic due to

urban forestry’s complexity; but to increase social equality, we need to be specific about who is

involved and how they are affected. Hale et al. (2015) propose a structured process that begins

with the desired tree benefits and then identifies under which conditions they are produced (e.g.

water supply or canopy cover level). After that, researchers or practitioners can assess how future

scenarios might affect these conditions and, thereby, the available benefits. While developed in a

United Kingdom (UK) context, the approach can be adapted to other places. I suggest, though,

that it currently fails to address how changes affect people differently, related, for example, to

their socio-economic status. Urban trees’ proposed role in sustaining human well-being and alle-

viating poverty is based at the individual and household scale. However, analysis of the effects of

current and planned urban forestry interventions often ignores this scale. I, therefore, pose the

following research question:

41
4. How will current and proposed interventions and developments affect people’s ability to use

urban trees within a livelihood context?

2.6 Chapter Conclusion

In this chapter, I have developed a time-, scale-, and place-explicit framework that illustrates how

different urban forestry-related concepts connect. I defined these components and introduced

further relating classifications, typologies, and models. Through a stepwise process, urban forest-

ry activities determine structure, functions, services, benefits, and values of city trees. Urban for-

estry itself occurs in a complex, multi-stakeholder setting that can be analysed in terms of its po-

litical, environmental, social, technological, and legal structure. Furthermore, events and process-

es at various scales and historical human–environment linkages also affect contemporary urban

forest structure. Urban trees are both natural and social elements as discourses and power distri-

bution shape urban forest planning and management of both public bodies and private individu-

als. These processes can be analysed through the lens of (urban) political ecology, which is crucial

as they create, legitimise and thereby reproduce unequal availability and access opportunities. Fo-

cusing on specific framework parts, I reviewed the pertinent literature, thereby identifying

knowledge gaps and the respective research questions that I am addressing in this project.

42
 3 METHODOLOGY

“Aside that, it’s only individuals like your type who come here looking for in-
formation, they come for interview, after that, when they finish their disserta-
tion, they put it on their libraries or whatever it is, and that ends it.”

Mr Awidi, Department of Parks and Gardens (DPG)

In this chapter, I present the methodological approach taken to answering the research questions

identified in the previous chapter. To investigate these diverse, multi-dimensional issues in Accra,

I adopted a mixed methods research (MMR) approach, collecting both quantitative and qualita-

tive data. For this project’s purpose – an in-depth understanding of stakeholder interactions and

the trees’ livelihood functions – I prioritised the qualitative component. To reflect political ecolo-

gy’s focus on power relationships, I chose to work with Mertens’ (2007; 2008; 2010) transforma-

tive-emancipatory paradigm. Mixed methods research that includes marginalised groups often us-

es this paradigm due to its core concern for social justice. Furthermore, it requires examining the

researcher’s positionality and the project’s ethical considerations in detail. After introducing my

study site, I then focus on the data collection and analysis process of the tree inventory, the

household survey and case study, the documentary research, and the stakeholder interviews.

3.1 Research Design

3.1.1 Mixed Methods Research
Mixed methods research has become popular in a variety of disciplines since its introduction dur-

ing the 1960s. Such a study design is defined as research “in which the investigator collects and

analyzes data, integrates the findings, and draws inferences using both qualitative and quantitative

approaches or methods in a single study or a program of inquiry” (Tashakkori and Creswell,

43
2007, p. 4). Its advantages and disadvantages have been discussed within different contexts

(Ivankova and Stick, 2007; Guerra et al., 2011), including natural resource management in Ghana

(Teye, 2012). These studies require being familiar with several data collection and analysis meth-

ods and often need more time and financial resources. However, their main advantage is often

described as providing a more comprehensive understanding with qualitative and quantitative in-

formation complementing each other (Johnson and Onwuegbuzie, 2004), offering the ‘best of

both worlds’.

These two worlds refer to positivism and interpretivism, common paradigms in social science re-

search. Until the early 20th century, positivists dominated and conducted ‘value-free’ research to

discover the objectively existing truth (Onwuegbuzie and Leech, 2005). Proponents of interpre-

tivism, however, argued for a constructed and subjective reality. While positivism developed into

post-positivism in the 1950s, rejecting the former’s rigid claims about value-neutrality, the two

worlds’ ontological, epistemological, and methodological views were considered irreconcilable. In

the 1990s, MMR became increasingly popular and challenged these claims (Howe, 1988). Re-

searchers interpret ‘paradigm’ in four nested ways, from an all-encompassing worldview to a nar-

row description of how research is done in a given field (Masterman, 1970; Morgan, 2007), the

chosen definition determining if and how data can be mixed. However, the methods themselves

are essentially a-theoretical and a-methodological though often associated with specific paradigms

and their methodologies that include certain preferred forms of data analysis (Giddings and

Grant, 2007).

Mixed methods researchers developed several approaches to accommodate collecting and analys-

ing various data types. An a-paradigmatic stance ignores potential incompatibility and assumes

that methodology is independent of epistemology (Patton, 1990), an approach often rejected as

“no research is paradigm free” (Hall, 2013, p. 5). Multi-paradigmatic approaches keep methods

44
separate, mix assumptions, or use the project’s specific design to determine the most appropriate

paradigm (Tashakkori and Teddlie, 2003; Cameron, 2011; Hall, 2013). Additionally, there are sin-

gle-paradigmatic approaches for mixed methods research, such as the pragmatic, (critical) realist,

or transformative ones. However, paradigms themselves are constructed by humans (Guba and

Lincoln, 1994) and “what counts as a paradigm and how the core content of a paradigm is por-

trayed involves a series of ongoing struggles between competing interest groups” (Morgan, 2007,

p. 61). I do not suggest that any classification is meaningless but instead agree with Onwuegbuzie

and Leech (2005, p. 275), who argue in favour of re-conceptualising dichotomies as lying on a

continuum instead.

3.1.2 The Transformative Paradigm

Mertens’ transformative paradigm centres social justice and power issues, and while constructivist

or pragmatic research can also accommodate these, I chose Mertens’ more radical approach for

this project. Its axiological objectives of “enhancement of social justice, furtherance of human

rights, and respect for cultural norms” (Mertens, 2010, p. 470) directly affect ontological, episte-

mological and methodological views. Though I did not fully adhere to its methodological ap-

proach, as I did not interact with the participants in defining the problem (Table 3), I take this

explicitly non-neutral stance for three reasons. First, solely on principle; second, as appropriate

due to the environmental equity issues identified in the literature review; and third, as a relevant

overarching structure for a political ecology lens. Another paradigm often used in MMR is the

outcome-oriented pragmatism (Johnson and Onwuegbuzie, 2004) that selects the most suitable

methods for particular research questions (Feilzer, 2010). Mertens (2003), however, challenges

the approach’s prioritisation of practicality, questioning who it is practical for and for what pur-

pose. Additionally, pragmatism has been described as a ‘trojan horse’, often inserting (post-)

45
positivist views into the research (Giddings and Grant, 2007) and the mixed methods discourse

(Denzin, 2010).

Table 3 The Transformative Paradigm: Ontology, Epistemology, Methodology, and Axiology

Belief System Transformative Paradigm

“There are multiple realities that are socially constructed, but it is neces-
sary to be explicit about the social, political, cultural, economic, ethnic,
racial, gender, age, and disability values that define realities. Different real-
ities can emerge because different levels of unearned privilege are associ-
Ontology
ated with characteristics of participants and researchers. Transformative
researchers need to be aware of societal values and privileges in determin-
ing the reality that holds potential for social transformation and increased
social justice.” (Mertens, 2007, p. 216)

“To know realities, it is necessary to have an interactive link between the

researcher and the participants in a study. Knowledge is socially and his-
Epistemology
torically located within a complex cultural context. Respect for culture
and awareness of power relations is critical.” (Mertens, 2007, p. 216)

“A researcher can choose quantitative or qualitative or mixed methods,

but there should be an interactive link between the researcher and the
participants in the definition of the problem, methods should be adjusted
Methodology
to accommodate cultural complexity, power issues should be explicitly
addressed, and issues of discrimination and oppression should be recog-
nized.” (Mertens, 2007, p. 216)

“Three basic principles underlie regulatory ethics in research: respect, be-

neficence, and justice. The transformative axiological assumption pushes
these principles on several fronts. Respect is critically examined in terms
of the cultural norms of interaction within a community and across
Axiology
communities. Beneficence is defined in terms of the promotion of human
rights and an increase in social justice. An explicit connection is made be-
tween the process and outcomes of research and furtherance of a social
justice agenda.” (Mertens, 2007, p. 216)

46
3.1.3 A Design for Mixed Methods Research

With rising popularity, the number of mixed methods research projects increased, and so did the

desire to classify them. MMR studies vary regarding the prioritised data type (qualitative or quan-

titative), the timing of data collection (e.g. concurrent or sequential), and how and when research-

ers combine data. Classification schemes have been developed to group designs (Morse, 1991;

Greene and Caracelli, 1997; Tashakkori and Teddlie, 1998), some, however, criticised as too

complicated, too simplistic, or inconsistent (Leech and Onwuegbuzie, 2009). Creswell et al. (2003)

identified six major but flexible designs, adaptable to specific projects. In their typology, my de-

sign corresponds to a ‘sequential transformative’ one, where either qualitative, quantitative, or

both data types can be prioritised, either one collected first while data is integrated at the inter-

pretation phase. This type’s more distinctive feature is the researcher’s explicit advocacy or em-

powerment perspective. While I first collected quantitative data, I prioritised qualitative infor-

mation which corresponds to a ‘partially mixed, sequential, dominant status’ design in Leech and

Onwuegbuzie’s (2009) scheme. To illustrate my project’s design, I provide a procedural diagram

drawn according to Ivankova and Stick’s (2007) guidelines (Fig. 14).

47
Figure 14 Procedural diagram

48
3.1.4 Positionality
The multiple contexts in which I found myself allow no simple answer as to how my age, race,

class, or gender attributes affected this research. All interviewees from governmental organisa-

tions and sub-metropolitan assemblies were male. Men also founded and led the two NGOs,

though one woman participated in one of these interviews. I also never encountered a female

herbalist or a woman selling fresh coconuts or tree seedlings and saplings by the roadside. Apart

from the Achimota visitor whom I recruited during my visit to the forest, no woman could have

replaced any of the male interviewees. In several instances, I felt I was not being taken seriously

as a woman talking about (urban) forestry, particularly being significantly younger than some of

the interviewed senior government officials. However, this might have worked in my favour as

respondents possibly shared information more freely with me, being perceived as neither threat-

ening nor overtly critical in conversation (Gurney, 1985; McDowell, 1988; England, 1994).

Being a white female undoubtedly facilitated access in neighbourhoods of higher socio-economic

status, where walls and security guards protected the houses. I believe that my race and gender at-

tributes made me appear non-intimidating and guards more likely to forward my request for par-

ticipation to homeowners. Additionally, the notion of a shared class or socio-economic back-

ground might well have acted in my favour in securing interview permission with governmental

actors. A static insider/outsider dichotomy, therefore, “ignores the dynamism of positionalities in

time and through space” (Mullings, 1999, p. 340). Throughout data collection, I shifted between

positions located more towards either side of an insider-outsider continuum. However, position-

ality might be better conceptualised through multiple, simultaneously interacting attributes (e.g.

age or gender) for which location on the insider-outsider continuum can vary in any given situa-

tion.

My personal views, values and experiences also affected the study’s design and data analysis. The

transformative paradigm and the political ecology lens reflect my aim to uncover alternative reali-

49
ties and knowledges that dominant groups or influential individuals tend to mute. However, I

have never experienced either absolute or relative poverty in the study context and, therefore,

might not be the most suitable person to examine urban tree livelihood functions and roles in

promoting human well-being. As a European, Ghana’s culture, history, and climate were unfamil-

iar to me; and while this might enhance the inclination to question the unknown, it also brings

with it the risk of naivety. In hindsight, I remember several PhD supervisions where my supervi-

sor asked about tree benefits other than the tangible ones I so insisted on focusing on. It required

first-hand experience of Accra’s climate to understand how tree shade can indeed top the list of

desirable services. Both my personal experiences and the Euro- and Americentric nature of urban

forestry research (Bentsen et al., 2010; Roy et al., 2012), therefore, influence my assumptions and

approaches.

Besides positionality, the researcher’s personality also affects a study’s design and data analysis

process (Moser, 2008). Moser argues that positionality attributes, like gender, race, or class, are

superficial and a person’s character more closely related to what kind of data collection ap-

proaches might be favoured. I relate to this view, preferring the position of the (quiet) observer,

taking notes, eventually inquiring about what I see, or working in an archive. While observation

and listening skills are also core abilities in interviews, these additionally require ‘thinking on the

spot’ and more direct inter-personal social skills. At the beginning of the project and long before

data collection, I wrote a reflexivity memo, questioning my ability to conduct interviews effective-

ly (Fig. 15). While increasing experience reduced my pre-interview anxiousness, which varied de-

pending on interviewee and setting, these personality traits are less affected by practice. As “we

bring different internal qualities and various emotional abilities to our fieldwork that have an im-

pact on the knowledges we create” (Moser, 2008, p. 390), research is never ‘un-personal’ and the

researcher’s traits are integral elements in each step of the process.

50
 25. February 2016: What worries me more is, however, my personality, which I con-
sider to be ‘wrong’ for doing interviews well. These kinds of things do not come
naturally to me.
Figure 15 Extract from my research journal

3.1.5 Ethical Considerations

The University of Birmingham’s Humanities and Social Sciences Ethics Committee approved this

research. All potential participants were introduced to the study’s objectives, the specific data col-

lection tool, and their expected role using the respective participant information sheet or oral ex-

planations in English or Twi (Appendix 1). Informed consent was obtained if the individual had

no further questions and agreed to participate (Appendix 2). Besides these standard procedures, I

now discuss three other ethical considerations. The first two concern participants’ contributions

and include the topics of confidentiality and power relationships, while the third one considers

the environmental impacts of ‘fieldwork abroad’.

‘Good practice’ in the social sciences often involves guaranteeing participants confidentiality.

However, some types of research may contain visual data like photographs (Miller, 2015) or re-

quire describing the researched, their lives or inter-personal relationships in detail (Surmiak,

2018). This in-depth information can cause deductive disclosure, where the combination of per-

sonal attributes can identify a participant. De-identifying data (e.g. reporting age ranges instead of

a specific number) mitigates this problem but “can alter or destroy the original meaning” (Kaiser,

2009, p. 1635). I informed interview and case study participants of this risk and selectively de-

identified some data in my discussion chapters. While confidentiality may be the gold standard,

some participants want to be identified and thereby recognised for their contributions (Miller,

2015). Especially in emancipatory and participatory approaches, the notion of partnership be-

tween researcher and researched implies “that confidentiality should not be imposed but negoti-

ated with participants” (Surmiak, 2018, no page number). Accordingly, I offered my participants

51
the opportunity to link their names to their contributions, and one household and several inter-

viewees opted to do so (see Sections 3.3.4, 3.3.5, and 3.6.2).

The transformative paradigm and political ecology are explicit about power relationships. How-

ever, participants volunteer their time and knowledge with potentially no or limited benefit for

themselves, as the chapter’s introductory quote illustrates. If the interviewee’s statement creates

an uncomfortable feeling, it does so because it demonstrates that “recognizing or even being sen-

sitive to these power relations does not remove them” (England, 1994, p. 85). Even in an eman-

cipatory-transformative approach, which conceptualises research as a partnership, some people

are more equal than others. Interviewing, observing, or otherwise engaging individuals constitutes

a “purposeful disruption of other people’s lives” (England, 1994, p. 85), and consent does not

eliminate these unequal power relationships. Though a research focus might be determined in

collaboration with the to-be-researched, the investigator ultimately creates these spaces and in-

vites others into them.

The environmental impacts of frequent research-related (air) travels and their ethical implications

have received limited attention so far. Few universities or other research institutions have calcu-

lated the environmental footprint caused by the mobility of both staff and students (Stohl, 2008;

Arsenault et al., 2019). While all researchers should consider the adverse effects of (air) travel, it is

essential for climate researchers whose credibility declined with increased flying (Attari et al.,

2016). The ethical approval process of any research should include reflecting on travel behaviour

and reducing trips (Stohl, 2008; Le Quéré et al., 2015). Throughout this project, I took multiple

intercontinental flights from Germany to Ghana. Starting the PhD, I lacked concern for the envi-

ronmental impacts of flying, which only developed more recently. In hindsight, my behaviour is

ethically questionable, especially since social justice is central to my research and climate change,

to which air travel contributes, disproportionately affects lower-income countries and their most

52
marginalised populations (Hallegatte et al., 2016; Islam and Winkel, 2017; Diffenbaugh and Burke,

2019).

3.2 Study Site

Accra is Ghana’s capital city and located at about 5°33’ N 0°11’ E, directly on the Atlantic Ocean

(Fig. 16). It is the country’s largest city, both in terms of population and area, with approximately

2 million inhabitants as of 2019.2 While reported land area varies, often excluding information on

the boundaries used or the year of reference, my study site covered 156 km2. The city experiences

two rainy seasons (March to June and September to November) and has a tropical savannah cli-

mate, an Aw type according to the Köppen-Geiger climate classification scheme. This climate is

characterised by the driest month having both less than 60 mm precipitation and less than

(Peel et al., 2007).

Figure 16 Location of Ghana (grey) and Accra

Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

2https://www.statsghana.gov.gh/regionalpopulation.php?population=MTM0NTk2MjQzOS4yMDE1&&Greater
%20Accra&regid=3 (Accessed: 27.01.2020)

53
My city and neighbourhood boundaries coincide with those of the Accra Poverty Map (CHF In-

ternational and AMA, 2010), which, for lack of access, excluded the Airport area, Burma Camp

(military zone), and Korle lagoon (Fig. 17). Using these boundaries, I can correlate my canopy

cover assessment and the map’s poverty ratings.

Figure 17 Neighbourhood boundaries

Shapefiles from EPA, map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

01 – Kisseman 02 – Legon 03 – East Legon Ext. 04 – Akweteyman 05 – Alogboshie

06 – Abofu 07 – Achimota College 08 – North Dzorwulu 09 – South Legon 10 – East Legon
11 – Kwashiman 12 – Lapaz 13 – Abeka 14 – Old Tesano 15 – Abelenkpe
16 – Dzorwulu 17 – Airport West Res. 18 – North Odorkor 19 – Darkuman 20 – New Fadama
21 – Tesano 22 – North Alajo 23 – Roman Ridge 24 – Airport Residential 25 – South Odorkor
26 – Bubuashie 27 – North Kaneshie 28 – North Industrial 29 – Avenor 30 – Alajo
31 – Kpehe 32 – Kokomlemle 33 – Kotobabi 34 – New Town 35 – Mamobi
36 – Nima 37 – Kanda 38 – Dansoman 39 – West Abossey-Okai 40 – Mateheko
41 – Abossey-Okai 42 – Kaneshie 43 – Awudome 44 – South Industrial 45 – Adabraka
46 – Asylum Down 47 – North Ridge 48 – Cantonments 49 – La Dadekotopon 50 – Russia
51 – Sukura 52 – Lartebiokorshie 53 – Sabon Zongo 54 – Zoti 55 – Tudu
56 – West Ridge 57 – Ringway Estate 58 – Labone 59 – Sempe 60 – Old Mamprobi
61 – Korle Bu 62 – Korle Dudor 63 – Accra Central 64 – Ministerial Area 65 – Osu
66 – Nyaniba Estate 67 – South Labadi 68 – La 69 – Mpoase 70 – Gbegbeyise
71 – Mamponse 72 – Mamprobi 73 – Chorkor 74 – Korle Gonno 75 – Jamestown
76 – Ussher Town

54
In 2010, Accra Metropolitan Assembly (AMA), together with an NGO, conducted the poverty

mapping exercise and rated the city’s neighbourhoods on eight variables (population density, in-

come level, housing density, room occupancy, nature of access to water, domestic wastewater,

mode of solid waste disposal, and nature of access to toilet facilities). Each variable was scored

on a five-point scale (non-poverty, low poverty, moderate poverty, high poverty, very high pov-

erty) and combined to an ‘overall poverty rating’ for each neighbourhood (Table 4; see Appendix

3 for individual neighbourhood ratings).

Table 4 Number of neighbourhoods in each poverty class

Poverty Class Number of Neighbourhoods

1 – Non-poverty 07

2 – Low poverty 14

3 – Moderate poverty 25

4 – High poverty 28

5 – Very high poverty 02

3.3 Data Collection

3.3.1 Tree Inventory
3.3.1.1 iTree Canopy
Between February and April 2016, I used the iTree Canopy programme3, an online tool provided

by the United States Department of Agriculture’s (USDA) Forest Service, to determine tree cover

of Accra and its neighbourhoods. I selected this tool because it requires neither technical

knowledge nor extensive training and is freely available. On the website, the user uploads shape-

files of their study area, which are overlaid on Google Maps aerial photographs. The programme

then displays random points that the user classifies according to previously defined land cover

3 https://canopy.itreetools.org/ (Accessed: 16.03.2021)

55
categories. As the standard error decreases with an increasing number of points, between 500 and

1,000 should be categorised. For this study, I classified 500 points for Accra as a whole and for

each of its neighbourhoods separately, using the categories ‘tree’, ‘non-tree plantable’, and ‘non-

tree non-plantable’ (Table 5). The aerial photographs’ low quality impeded a more detailed divi-

sion as even distinguishing between ‘tree’ and ‘non-tree’ proved difficult in some areas. After the

user has classified the desired amount of points, the tool generates a report, summarising the cat-

egories’ per cent values, the corresponding land area, and the respective standard error (reported

in brackets in the discussion section).

Table 5 Examples of different land cover categories

Land Cover Category Examples

Tree Tree

Backyards

Non-tree plantable Roadsides

Open spaces

Buildings

Non-tree non-plantable Water bodies

Roads

To facilitate data visualisation, I transformed the continuous iTree Canopy values into categorical

data, which further ensures comparability with the poverty map that also utilised (five) ordinal

classes. As the place-specific nature of urban tree cover precludes applying values from other cit-

ies to Accra, I used Jenks’ natural breaks classification method to obtain five canopy classes. This

method, also known as Jenks optimisation or Jenks-Caspall algorithm, clusters data in a way that

minimises the average deviation of the class mean while maximising each class’s deviation from

the other means (Jenks and Caspall, 1970). The Achimota College neighbourhood, because of the

56
3.6 km2 Achimota Forest, has a canopy cover of 53.6 %, significantly higher than the second-

highest value of 33.6 % in Legon. Accra’s average of 14 % thus ended up in the second-lowest

canopy class, which ranged from 8.4 to 15.2 % canopy cover. As the city average should be a ref-

erence point when creating classes that are to show intra-city variation, I excluded Achimota Col-

lege neighbourhood as an outlier, resulting in the city average being located in class 3, now rang-

ing from 12.4 to 17.4 % (Table 6).

Table 6 Number of neighbourhoods in each canopy cover class

Class Canopy Cover Range % No. of Neighbourhoods

1 26.2 to 33.6 07

2 19.4 to 24.8 09

3 12.4 to 17.4 14

4 7.4 to 11.8 17

5 0.2 to 6.6 28

Using these canopy cover classes, I coloured the neighbourhood map accordingly, with lighter

green for lower and darker green for higher canopy cover (Fig. 18). I used the software ‘R’ (R

Core Team, 2015) for statistical analyses, employing the Fisher-Freeman Halton test to investi-

gate correlations between poverty and canopy classes. This test is suitable for 5 × 5 contingency

tables where observed numbers include zeros or expected numbers are smaller than five in more

than 20 % of the cells. For computational reasons, I used the test with the Monte Carlo method

(1,000,000 replicates).

57
 Canopy cover class

Figure 18 Neighbourhood canopy cover classes

Shapefile from EPA, map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
See Fig. 16 for neighbourhood numbering

3.3.1.2 iTree Eco

Besides the desk-based iTree Canopy cover analysis, I also conducted a plot-based sample inven-

tory stratified by neighbourhood poverty class. For this, I used the iTree Eco tool4, another free

software application from the USDA Forest Service. Users can either conduct a complete inven-

tory of all trees within their study site or collect data in randomly located plots. The programme

then runs a model that quantifies the forest’s structure – species condition, distribution, biomass,

4 https://www.itreetools.org/tools/i-tree-eco (Accessed: 16.03.2021)

58
leaf area – and its functions, including pollution removal, carbon sequestration or avoided run-

off. The exact output, however, depends on the information collected and further inputs such as

air pollution and meteorological data. In designing the project, I followed the user manual’s

guidelines and recommendations (USDA Forest Service, 2016), which suggest a total of 200 cir-

cular plots with at least 20 per stratum. Based on the combined land area of the neighbourhoods

in each poverty class, I calculated the number of plots per stratum using a log formula (Table 7).

Table 7 Number of plots per stratum

Stratum Combined Land Area (ha) No. of Plots

1 (No poverty) 1,138.05 41

2 4,236.05 46

3 6,270.56 47

4 3,811.61 46

5 (Very high poverty) 188.89 20

Data collection was based on the field manual but excluded some of the non-required variables

that lacked relevance in relation to my research questions (see Appendix 4 for data collection

form). As the iTree Eco approach requires considerable resources, I sought to maximise efficien-

cy by trading off costs and the potential added value of further data inputs (Table 8). Additional-

ly, some of the application’s outputs necessitate information, such as annual hourly amounts of

certain pollutants, that was unavailable for Accra and collecting the related ground-based data

would have been redundant. The data collected included:

59
 Table 8 Collected plot and individual tree data

Plot Individual Tree

ID ID

Stratum Species

Global Positioning System (GPS)

Land use
coordinates of centre

% of plot measured DBH

% of plot tree cover Crown condition

Trees present y/n Total tree height

Land use Live crown height

% Land use of plot Crown base height

Ground cover Crown width

% Crown missing

Crown light exposure

Street tree y/n

Public tree y/n

Tree GPS coordinates

Maintenance recommended y/n

Sidewalk conflict y/n

Utility conflict y/n

Two two-member teams visited the 200 plots between August and November 2016. I had to re-

place 22 plots that were located on school grounds or private property to which owners did not

grant access (Table 9).

60
 Table 9 Replaced plots per stratum

Stratum (Poverty Rating) No. of Plots Replaced

1 6

2 8

3 5

4 2

5 1

Before starting data collection, I trained the research assistants in a theoretical class, after which

we visited three plots to gain practical experience. For fieldwork, each team was equipped with

the required forms and tools, including a GPS device, measuring tape, and clinometer (Table 10).

Additionally, they took a tree identification guide and an information sheet that introduced the

project to homeowners when a plot was located on their private property. If the trees could not

be identified in situ, sample leaves were collected in plastic bags and given to the Department of

Botany at the University of Ghana for identification. While most trees could be identified to spe-

cies level, some could be identified to genus level only or not at all. The iTree Eco application

then allows for entering a tree simply as hardwood or softwood. After I entered all paper-based

data into the computer programme, a model was run and a comprehensive report provided.

Table 10 Equipment and producer information

Equipment Type Producer

Brüder Mannesmann
Measuring tape 20 m, Glasfaser-Bandmaß Werkzeuge, Remscheid,
Germany

Clinometer PM-5/360 PC Suunto, Vantaa, Finland

GPS device GP-102 G-Porter Canmore, Zhubai, Taiwan

61
3.3.2 Household Survey
Between October and December 2016, I conducted a cross-sectional household survey with 100

participants in five of Accra’s neighbourhoods that I chose based on three criteria. First, poverty

map rating (Dzorwulu – 1; Kokomlemle - 2; Alajo – 3; Mamobi – 4; Nima – 5); second, their in-

ner-city location; and third, proximity to each other (Fig. 19), all located in the Ayawaso sub-

metropolitan areas (Fig. 20).

Figure 19 Household survey neighbourhoods

Shapefile from EPA, map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
16 – Dzorwulu 30 – Alajo 32 – Kokomlemle 35 – Mamobi 36 – Nima

62
 Figure 20 Ayawaso sub-metropolitan areas
Shapefile from EPA, map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
Green – Ayawaso West Blue – Ayawaso Central Orange – Ayawaso East

To account for potentially different opinions in relation to canopy cover, I used a stratified sam-

pling approach. Stratification, however, was not proportionate as I lacked access to census data

listing every household in the study sites. Instead, I adopted a disproportionate approach, with 20

respondents per neighbourhood. Initially, I intended to utilise a three-step process to identify re-

spondents, starting with a random point (provided by iTree Canopy), then enumerating all

households within a plot surrounding the point, and finally randomly selecting one from the list.

However, I had misjudged the situation, and this approach would have been overly resource-

consuming and was complicated by inadequate road naming and house numbering. As the dis-

63
proportionate sampling approach already limited external validity, I resorted to a two-step pro-

cess where a random point served as the starting location for a convenience sample approach,

contacting the closest available household. This offered a feasible approach considering the given

circumstances.

Designing the questionnaire, I used various guidelines covering theoretical and practical aspects

involved in developing a household survey (Grosh and Glewwe, 2000; Statistics Canada, 2003;

United Nations, 2005a; United Nations 2005b). My finished questionnaire contained six sections

with mostly close-ended questions followed by open-ended ones so respondents could explain

their views (Appendix 5). Section A, the household roster, followed the Ghana Living Standard

Surveys in terms of identifying household members and options for pre-coded answers. Section

B consisted of multiple-choice questions to determine the household’s Progress out of Poverty

Index (PPI®)5. It calculates the household’s likelihood of falling below a certain poverty line

based on ten country-specific questions.6 I added this section to account for potential intra-

neighbourhood variation of socio-economic status. However, within-household differences of

income or views on trees were covered in the household case study (see Section 3.3.5). Sections C

and D contained general questions about the benefits and maintenance of public and private

trees, while the remaining sections investigated tree product use (E) and sale (F) in the previous

year. Pre-administration, I discussed the questionnaire with the research assistants and tested tim-

ing and clarity on three randomly selected households. Minor adjustments were made to improve

the wording of some questions, after which a research assistant translated them into Twi.

Household questionnaires were administered and resembled structured interviews as self-

administration could have lowered response rates due to the population’s characteristics (e.g. high

illiteracy rate). Depending on the given answers and the related skip patterns of irrelevant follow-

5 since rebranded as the Poverty Probability Index

6 The higher the PPI score, the lower the household’s likelihood to fall below the poverty line

64
up questions, administering each questionnaire required between 30 and 60 minutes. Administer-

ing questions to a multi-person household in a group setting may provide differing opinions and

potentially more holistic results. However, some individual members’ views may dominate and

arranging appointments and repeated visits is time and cost-intensive. The questionnaire was,

therefore, conducted with any available adult household member, this convenience sample ap-

proach including instances of adult children representing their households and 52 female re-

spondents. As 68.2 % of households in the Greater Accra Metropolitan Area are headed by men

(Ghana Statistical Survey, 2014a), a group setting would potentially have made women’s respons-

es less visible. After data collection, I entered the paper-based information into a spreadsheet and

used basic descriptive statistics for analysis.

Non-sampling errors are common in surveys and include respondents providing false infor-

mation, intentionally or unintentionally, and issues related to administering the questionnaire.

There is evidence that at least one respondent provided wrong information on their household

composition, directly affecting PPI scores. This index also included potentially sensitive questions

as people could avoid reporting not owning certain assets, like mobile phones, or be unaware of

another household member owning them. Asking respondents to report information from a par-

ticular previous period can introduce recall bias as “there is an inverse relationship between the

length of time over which subjects are asked to recall prior use and the accuracy of the reported

estimates” (Clarke et al., 2008, p. 1275). Researchers, therefore, face a trade-off between the in-

formation to be obtained and potential errors caused by inadequate recall periods. In my survey, I

asked respondents about tree parts and their use(s) in the previous twelve months, a relatively

long timeframe. This recall period, however, covered trees’ seasonality in flowering/fruiting and

uncommon events like decorating with palm leaves on Palm Sunday. While I used a matrix to fa-

cilitate remembering less common occasions, participants may have unintentionally misreported

on these interactions, many of them ordinary and lacking any memorable characteristics.

65
Certain problems can arise when individuals are tasked with administering a questionnaire they

did not design. While I was familiar with how different questions or skip patterns related due to

constructing the survey myself, the research assistants had to accustom themselves to the ques-

tionnaire and process of administration. Before starting data collection, I introduced the assis-

tants to my project in general and the survey’s purpose more specifically. After that, we discussed

any existing questions and the technical aspects of administering the questionnaire. I then ac-

companied them into the field, and they interviewed two to four participants in my presence.

Though there were no further concerns when these training sessions ended, a lack of attention

seems to have caused some problems when research assistants went out alone in the following

days. In several cases, relevant boxes were left unmarked, and while responses could be deduced

from those given to the connected open-ended follow-up questions, these occurrences potentially

reduced the survey’s quality.

3.3.3 Documentary Research

3.3.3.1 Online Newspaper Article Research
Despite a recent increase, published information about urban forestry in Ghana is still limited,

and I only identified a few relevant articles during my initial literature search. In April and May

2017, I searched the online presences of the country’s four highest circulated daily newspapers

for further information on forest planning and management, especially in Accra. The newspapers

included the Daily Graphic, the Daily Guide, the Ghanaian Times, and The Chronicle, all super-

regional but differing in ownership, political preferences and number of copies (Kuehnhenrich,

2012)(Table 11). I searched each newspaper’s website for four terms: ‘Parks and Gardens’, ‘Tree

AND Accra’, ‘Forest AND Accra’, and ‘Environmental Protection Agency’. Rather than produc-

ing a systematic review on urban forestry, I employed this scoping review to identify possible

stakeholders and current issues before commencing interviews and the case study. The search

66
yielded 114 articles (Appendix 6) which I categorised according to the topic(s) discussed. As I en-

countered issues with the searchability of the websites, including unprocessed requests due to

timeouts and what appeared to be a limited time scope, I likely did not find all relevant articles.

Table 11 Searched newspapers and their characteristics

Newspaper Ownership Political Position Copies

above party lines, slightly towards

Daily Graphic Public 100,000
the National Democratic Congress

Daily Guide Private New Patriotic Party close 22,000

National Democratic Congress

Ghanaian Times Public 80,000
close

above party lines, slightly towards

The Chronicle Private 45,000
New Patriotic Party
According to Kuehnhenrich, 2012

3.3.3.2 Archival Work

In April 2017, I conducted research in the archive of the Public Records and Archives Admin-

istration Department (PRAAD) in Accra. As documents had not been classified according to the

topic of urban trees and forestry, my approach was exploratory and focused on obtaining de-

scriptions of tree-related activities in Accra’s past. After summarising my project, an employee

suggested potentially relevant group codes, including departmental reports, minutes of Town

Council meetings, and documents related to city planning and development. I entered each file

that contained pertinent information into a spreadsheet, including its classmark, the record’s title

and date of reference, a summary of its content, access date, direct quotes as well as further

comments and observations. I also searched for historic photographs and other depictions which

I planned to describe and analyse. However, photographs mostly lacked any reference to location

or used colonial road and place names, which I could not match to their respective present loca-

67
tions. While an option for future research, it requires cooperating with someone knowledgeable

about Accra’s past. As many of the retrieved documents were in a poor state, I refrained from

studying some of them to avoid further damage, limiting the work’s scope.

3.3.4 Stakeholder Interviews

Between April and August 2017, I interviewed 13 individuals from governmental agencies and

departments, sub-metropolitan officers, NGO leaders, and people otherwise participating in ur-

ban forestry-related activities in Accra (Table 12). My first list of potential interviewees included

those stakeholders whose names appeared during the literature review and online newspaper

search. This list mostly covered governmental organisations like the Department of Parks and

Gardens (DPG), the Town and Country Planning Department (TCPD), the Environmental Pro-

tection Agency (EPA), or the Forestry Commission (FC). To obtain multiple perspectives, I also

wanted to include non-state actors, but the internet search suggested only two relevant NGOs,

and neither interviewee suggested other relevant organisations that could have offered further in-

sights. While inventorying trees, I noticed individuals such as herbalists, and roadside tree or co-

conut sellers whom I added to the list of potential participants. During each interview, I also

questioned the respondents about other relevant groups or individuals, thereby introducing the

element of chain referral. This resulted in a long list including, e.g. the Department of Urban

Roads, the National Disaster Management Organisation, or the Ghana Garden and Flower Show.

Due to limited resources, I could not interview every possible respondent and prioritised accord-

ing to their involvement’s intensity, the centrality of urban trees in their work, and new angles

they might offer.

68
 Table 12 Interviewees and dates of interviews

Interviewee Name(s) Position/Role Date

Reginald Quaye, Amakoue

Four members of the
Youth4GreenGhana (Y4GG) Ayih, Solomon Affah, 09.04.17
NGO
Afua Owusu-Ansah

Self-employed herbal-
Herbalist / 07.06.17
ist with roadside shop

Pastor who regularly

Achimota Forest User/Pastor / 08.06.17
visits Achimota Forest

Self-employed road-
Roadside coconut seller Samuel Kwakye 21.07.17
side coconut seller

Department of Parks and Assistant Horticultural

Ebenezer Awidi 31.07.17
Gardens Officer

Town and Country Planning Metropolitan Director

Kwadwo Yeboah 02.08.17
Department of Town Planning

Head of EPA’s Natu-

Environmental Protection
Carl Fiati ral Resource Depart- 04.08.17
Agency
ment in Accra

Ghana Youth Environmental

Gideon Commey Founder 04.08.17
Movement (GYEM)

Environmental health
Ayawaso West (Sub-metro) / 07.08.17
officer

Environmental health
Ayawaso Central (Sub-metro) L. Sampson 09.08.17
officer

Environmental health
Ayawaso East (Sub-metro) / 09.08.17
officer

Self-employed road-
Roadside tree seller Anasawudu 10.08.17
side tree seller

Forestry Research Institute of

/ Researcher at FORIG 11.08.17
Ghana (FORIG)

69
Interviews were semi-structured and consisted of three components, however, the respondents’

backgrounds varied considerably, and I adjusted questions accordingly (see Appendix 7 for an in-

terview guide example). I first asked respondents to introduce their organisation/job and describe

their activities, responsibilities and professional background. I then inquired about planning and

management objectives and tree maintenance where relevant. In this second part, I also focused

on the strengths and weaknesses of their work and that of their organisation, as well as on oppor-

tunities and threats they identified in relation to urban forestry, both in the country and in Accra

specifically. Interviews lasted between 30 and 70 minutes and were audio-recorded when con-

ducted in English. In the case of Twi speaking respondents, a research assistant was present for

the interview and translated my questions and the interviewee’s responses, which I then wrote

down in English.

Before analysing interviews, I first transcribed the audio recordings for which various approaches

exist, each with its advantages and disadvantages. I selected an approach that focused on content

and transcribed interviews verbatim using Braun and Clarke’s orthographic approach (2013, p.

165-166), a simplified version of Jefferson’s system (1984) widely used in conversation analysis.

While I added overlaps, pauses, laughter, and uncertain hearings, my interviews were focused on

obtaining topical information about different stakeholders and their views rather than on linguis-

tic subtleties. My transcriptions meet the minimum requirements that Braun and Clarke (2006, p.

88) specify for their version of thematic analysis (TA), including “a rigorous and thorough ‘or-

thographic’ transcript – a ‘verbatim’ account of all verbal (and sometimes nonverbal – eg,

coughs) utterances”. However, transcribing is an interpretive act itself (Bird, 2005) as even decid-

ing on punctuation affects perceived meaning (Braun and Clarke, 2006). Additionally, the re-

searcher chooses what is represented how or what is excluded and why (ten Have, 1990). Where I

include direct interview quotes, I have lightly edited them for speech repetitions and fill words to

facilitate easy understanding of the content.

70
Using NVivo for data management, I analysed the interview transcripts using Braun and Clarke’s

(2006) reflexive thematic analysis, which differs from commonly used coding reliability and code-

book approaches. In their opinion, positivist assumptions underpin coding reliability approaches,

such as by Boyatzis (1998), which claim to find ‘reality’ in the data, a view they oppose. They

consider codebook approaches, such as framework or template analysis (Ritchie and Spencer,

2002; King and Brooks, 2017), to be paradigmatically closer to their approach. However, these

structured codebooks organise data in a way that themes are (also) analytical inputs rather than

just outputs. I chose Braun and Clarke’s approach as they emphasise how the researcher’s posi-

tionality, the respondents and their data, and the research context co-produce themes, which do

not simply ‘emerge’ from the data. In reflexive thematic analysis, there is, therefore, no ‘accurate’

way of coding and analysing data, an approach in accordance with the general assumptions un-

derlying this research project.

Braun and Clarke’s TA involves a six-step process from data familiarisation to coding, generating

initial themes, reviewing these themes, defining and naming them, and writing up the findings.

Data familiarisation is essential and includes reading through the transcripts several times. In the

second step, the researcher codes selected data segments, organising them into meaningful

groups. I combined ‘descriptive’ and ‘versus’ coding, the former summarising “in a word or short

phrase – most often as a noun – the basic topic of a passage” (Saldaña, 2009, p. 71) while the lat-

ter looks for binary divisions “that suggest strong conflicts within, among, and between partici-

pants” (Saldaña, 2009, p. 94). Versus coding constitutes a more interpretive and latent approach

suited to my aim of identifying strengths, weaknesses, opportunities, and threats. Combining two

approaches can cause simultaneous coding, which is “the application of two or more different

codes to a single qualitative datum” (Saldaña, 2009, p. 62), warranted if coding both at the de-

scriptive and inferential level (Miles et al., 2014). The researcher then develops the codes into

themes that capture “something important about the data in relation to the research question,

71
and represents some level of patterned response or meaning within the data set” (Braun and

Clarke, 2006, p. 82). These themes should not be confused with ‘domain summaries’ (DeSantis

and Ugarriza, 2000; Sandelowski and Leeman, 2012; Connelly and Peltzer, 2016) that simply

summarise the content of a group of codes.

Various problems can arise when planning, conducting and analysing interviews. Rather than fo-

cusing on the general limitations associated with interviews as a research method, I want to focus

on an aspect specific to this project, which is that of language. My initial list of potential stake-

holders mainly included government employees whom I planned to interview in English. During

the data collection process, however, I came across individuals, such as the coconut seller or the

herbalist, with whom I did not share a common language to the degree that we could easily con-

verse. While I could have audio-taped these interviews, with a research assistant translating an-

swers and questions back and forth, I instead decided on note-taking only. Thus, the research as-

sistant translated my questions into Twi, and I handwrote the back-translated responses down

with the translator interpreting both the interviewees’ and my words. These transcripts, therefore,

differ from those I conducted in English.

When researcher and participants do not share a common first language, and sometimes none at

all, translation becomes a core component of the work. Just like transcription, it contains multiple

interpretive acts as “communication across languages involves more than just a literal transfer of

information” (Temple and Edwards, 2002, p. 2). Some researchers refrain from using translators

(Filep, 2009) to avoid ‘triple subjectivity’, referring to the interactions between participants, inter-

preter, and researcher (Temple and Edwards, 2002, p. 6). However, as I am a native German

speaker conducting research in English, avoiding any form of translation, even if only internal,

was impossible. Rather than employing a professional interpreter, I had a research assistant trans-

late who also participated in the tree inventory and survey, being familiar with the work and its

72
objectives. As a geography graduate, they had extensive fieldwork experience and served as a type

of ethnographic informant explaining and clarifying cultural-specific aspects. I consider it worth-

while for future research to ‘formalise’ this additional role of a translator and interview them, as

suggested by Temple and Edwards (2002), about their views and backgrounds in relation to the

research topic.

3.3.5 Household Case Study

In June and July 2017, four surveyed households also participated in my case study to investigate

human–tree interactions in Accra and the trees’ livelihood functions. For this purpose, a case

study approach is particularly suitable as it answers ‘why’ and ‘how’ questions when the research-

er investigates a “contemporary phenomenon within a real-life context” over which they exercise

limited or no control (Yin, 2009, p. 2). A case study is not a data collection tool as such, but ra-

ther a research strategy that utilises contemporary documents and archival material, interviews,

physical artefacts, or direct and participant observation. While I focused on interviews and obser-

vations, documents and physical artefacts were also included in this study. Three writers, whose

approaches differ notably, have influenced case study research. For Yin’s post-positive position,

case studies are designed to falsify hypotheses and use qualitative and quantitative data collection

and analysis methods to triangulate information and minimise levels of subjectivity (Harrison et

al., 2017). Stake and Merriam, however, take a constructivist/interpretivist approach and empha-

sise the researcher’s role in interpreting subjective realities (Yazan, 2015; Harrison et al., 2017). I

partly followed Yin’s more structured approach for design choices and terminology, though I

take an explicitly constructivist stance.

In case study research, individuals, organisations, or specific events or programmes can constitute

the case(s) under investigation. In my project, I define the four households as cases and their in-

73
dividual members as embedded subcases. Including household members as subcases allows in-

vestigating intra-household dynamics and differences, as they are heterogeneous units of analysis.

My descriptive-explanatory case study (Lijphart, 1971; Stake, 1995; Yin, 2003) uses multiple cases

“because it is believed that understanding them will lead to better understanding, and perhaps

better theorizing, about a still larger collection of cases” (Stake, 2005, p. 446). The descriptive

component provides an in-depth understanding of how and why households and individuals in-

teract with city trees and fills an existing knowledge gap. My case study’s findings offer a founda-

tion on which to start building theories around urban human–tree interactions and their broader

context and relevance.

I purposefully selected the four cases to represent a diverse range in terms of tree interactions

and their household composition. I first eliminated those survey respondents who did not own a

tree on their private property. Though valuable information, investigating how city residents in-

teract with public trees in a case study approach could produce limited insights, especially on tree

care and management. I then grouped the remaining tree-owning households according to loca-

tion, type and amount of used tree benefits, reported maintenance, and whether or not they had

sold tree products in the previous twelve months. I further considered household size, age struc-

ture and gender composition, and while I also intended to select households based on PPI scores,

none had a high likelihood of falling under the income poverty line. Though dividing PPI score-

cards by urban and rural sites did not improve accuracy in other countries (Schreiner, 2015), Ac-

cra’s characteristics as a primate city could potentially affect households’ scores. The final decid-

ing factor remained the household members’ willingness to participate in time-consuming and in-

trusive research. While the participating households might not cover an ‘ideal’ range (see Section

6.3), case study research does not aim for statistical generalisation as cases are not used to “ex-

trapolate probabilities” (Yin, 2014, p. 21).

74
I spent about two weeks with each household, staying in rented accommodation and visiting par-

ticipants during the day. I interviewed household members and related individuals, such as other

compound residents or a neighbour that had planted trees in the participant’s yard (Table 13). In-

terviews were semi-structured and included questions about the respondent’s life, their views on

tree benefits and disservices, and tree maintenance. In most cases, the research assistant who had

previously participated in data collection translated these interviews. Additionally, I observed how

people interacted with trees and recorded these fieldnotes in a notebook, typing them up in more

detail afterwards. While writing notes might seem intuitive, Spradley (2016, p. 82-83) developed a

descriptive question matrix that suggests collecting data on spaces, objects, acts, activities, events,

times, actors, goals, and feelings, thereby helping to structure observations. For data analysis, I

used framework analysis, a structured version of thematic analysis developed by Ritchie and

Spencer (2002) for applied policy research but since used in various other contexts as well (Gale

et al., 2013).

Table 13 Interviewees associated with the case study households

Associated Case
Interviewee Name
(see Section 6.3)

Tenant / 1 – Nima

Four men regularly sitting under-

/ 1 – Nima
neath a household owned tree

Tenant and friend of the partici-

Fred Agbobli 1 – Nima
pant

Tenant / 2 – Mamobi

Neighbour/friend who planted

moringa trees in the household’s David Mamattah 4 – Kokomlemle
compound

75
In framework analysis, the researcher produces a matrix with cases and topics/codes, entering

the relevant summarised data into the cells. Data can then be compared and contrasted across

cases and, in this project, also within cases, i.e. between individual household members. I com-

bined deductive and inductive coding and theme-generating approaches, drawing on both the lit-

erature review and my research questions. Additionally, I aimed “to leave space to discover other

unexpected aspects of the participants’ experience or the way they assign meaning to phenome-

na” (Gale et al., 2013, no page number). Data analysis is similar to Braun and Clarke’s thematic

analysis and includes transcription, data familiarisation and coding based on which the researcher

develops an analytical framework. Relevant information is then entered into a spreadsheet, where

summarising and thereby reducing the data is an integral part of the process. The obtained matrix

is “visually straightforward and can facilitate recognition of patterns in the data […], including

through drawing attention to contradictory data, deviant cases or empty cells” (Gale et al., 2013,

no page number).

To move towards theory building, I would have liked to increase case number, including house-

holds that only reported using public trees’ benefits or recalled few interactions. I obtained a con-

siderable amount of data during the two weeks, but spending more time with each household and

using a more participatory form of observation, could yield further insights. Additionally, certain

tree products are only seasonally available and care activities, such as pruning or planting, con-

ducted at specific times of the year and on few occasions. A longitudinal design might therefore

be more appropriate to investigate certain human–tree interactions. The above-described limita-

tions of working with a translator are equally relevant in this case study, potentially even more so

due to the intimate everyday settings. While stakeholder interviews tended to be topically fo-

cused, the case study aimed to obtain an in-depth understanding of people’s views and choices,

making a shared language both parties feel comfortable conversing in even more preferential.

76
3.4 Chapter Conclusion
This chapter outlined my research design choices and how they relate to the identified research

questions. To investigate the complexities of urban forest planning and management in Accra, I

opted for a mixed methods approach. While collecting and analysing both qualitative and quanti-

tative data, I focused on the former. A visual diagram of the study’s design summarises the more

detailed descriptions of the data collection and analysis process. The literature review demon-

strated how the distribution of power affects urban forest structure, the trees being both natural

and social elements. I, therefore, chose to work within Mertens’ transformative paradigm that in-

cludes social justice issues as a core component. I summarised the paradigm’s ontological, epis-

temological, methodological, and axiological views and discussed my positionality and my pro-

ject’s ethical implications in detail. I further introduced the study site and described each of my

data collection methods – tree inventories, documentary research, stakeholder interviews, and a

household survey and case study – in detail. This also included outlining my data analysis ap-

proaches and the methods’ disadvantages and problems.

77
4 ACCRA’S URBAN FOREST – STRUCTURE OVER TIME AND ACROSS
SPACE

“[T]here may be more trees but now, as development is going on, that is where
my fear is because the majority [of] trees are now being replaced with buildings
as development is going on.”

Environmental Health Officer, Ayawaso East

~~~~~

“Many ornamental and shade trees in towns and residential areas are being de-
stroyed or mutilated by bad cutting. Much of this damage is being done in
connection with development works, improvement to roads, new buildings etc.
and can be avoided.”

PRAAD, RG 3/5/349 – 1950, p. 12

In this chapter, I outline how human–environment interactions have shaped Accra’s urban forest

from pre-colonial to present times. I first detail how pre-colonial, colonial, and post-

independence events and processes affected the city’s tree cover. Here, I consider both biophysi-

cal legacies, such as earthquakes or tree pests and diseases, as well as human legacies, such as state

and non-state interventions. Following, I present the findings of my desk-based and field-based

tree inventories to describe Accra’s current urban forest structure both at city and sub-city level.

Over the centuries, numerous legacies have interacted with each other in complex ways, creating

a city that, today, is characterised by sparsely populated, greener and ‘richer’ neighbourhoods on

one end of the continuum and densely populated, less green and ‘poorer’ neighbourhoods on the

other end. Throughout this chapter, I intersperse descriptive summaries of historic developments

pertaining to issues of land ownership and use, central to city planning and, by extension, to ur-

ban forestry. The following parts should be read as an example of how some selected factors

78
have shaped Accra’s urban forest rather than as an all-encompassing explanation. While archival

records provide information on past events and activities, this will be incomplete due to the mul-

tifaceted ‘silence of the archive’. The archive as an institution, and the records within archives,

hold considerable ‘power over memory’ (Schwartz and Cook, 2002). Despite best efforts to in-

clude alternative knowledges and non-events, i.e. those that do not appear in archival documents,

this information was difficult to access, posing limitations to and creating biases in this chapter.

4.1 Human–Nature Interactions and Their Legacy Effects

In this section, I examine the historical spatio-temporal dynamics and their legacy effects on Ac-

cra’s forest. These legacies are “impacts that previous events, processes, and phenomena have on

current properties or processes” (Roman et al., 2018, p. 159) and facilitate an understanding of

contemporary patterns (Szabó, 2010). Several authors have recognised the importance of history

in ecology (Cook et al., 2012; Bürgi et al., 2017; Roman et al., 2018) and created different frame-

works that demonstrate how society and environment interconnect in and over time. I define his-

tory to include both past human and natural activities, which Roman et al. (2018) term ‘human’

and ‘biophysical’ legacies. The former includes civic beautification movements, socio-economic

changes or national and regional identity as “species composition of many urban forests is

strongly influenced by the selection of species that held strong cultural or political meaning in the

past” (Roman et al., 2018, p. 161). The latter, among others, subsumes ecological disturbances like

extreme weather events and disease outbreaks. These legacies act within the bioregional context,

i.e. climate, native biome, or regional species pool and “co-produce spatio-temporal dynamics of

urban tree canopy cover, species composition, and diversity” (Roman et al., 2018, p. 158).

79
4.1.1 Accra and Trees – The Pre-Colonial Past
The origin of the Gas, the founders of Accra, is a matter of diverging opinions (Reindorf, 1895;

Bruce-Myers, 1927; Boahen, 1966; Odotei, 1991) but of secondary importance as, in this case, the

relevant time begins with their establishment along the shore in the 15th century. Due to “[t]he

growing importance of European trade along the coast, the gradual degradation of the soil, and

some rivalry between the Gas and their neighbours” (Alhassan, 2006, p. 22), they left their inland

location some kilometres away and founded numerous coastal towns. While Gas and Europeans

had traded since the second half of the 16th century, the English, Portuguese, Dutch and Danish

only started building (trading) forts in the 17th century. Despite the then permanent European

presence and the forts’ association with the Ga towns, “the Gã king and people were clearly the

sovereign power. They chose which Europeans to trade with, and they decided, by treaty, which

nations could establish forts and settlements in their territory” (Osei-Tutu and von Hesse, 2018,

p. 211). Additionally, Europeans had to pay lump sums and rent to the king, who also taxed the

goods arriving in Accra (Odotei, 1991).

Barbot, a French man who travelled to Guinea twice (1678/1679 and 1681/1682), describes the

‘kingdom of Acra [sic]’ as being composed of the towns of Soko (also known as Ngleshi/Ŋléshì,

Jamestown), Little Acra/Accra (Kinka, Usshertown) and Orsaky (Osu). The first associated with

the English Fort James, the second with the Dutch Fort Crèvecœur (later Ussher Fort) and the

last with Fort St. Francis Xavier (Fig. 21). When he visited, the latter belonged to the Portuguese

but was shortly after (1683) returned to the Danish and henceforth known as Fort Christiansborg

(Hair et al., 1992). Ngleshi and Kinka, two towns in the Europeans’ eyes, were, however, “seven

geographically and politically defined divisions (akutsei, sing. akutso)” in the Ga’s eyes (Kropp

Dakubu, 1997, p. 43). Within these quarters, extended families/lineages (wei, sing. we) lived in gat-

ed, rectangular homesteads, “at the centre of which is a shade giving tree, under which all family

gatherings […] take place” (Quarcoopome, 1998, p. 136). Barbot estimated that Soko (Ngleshi)

80
and Little Accra (Kinka) consisted of less than 100 and about 60 houses, respectively (Hair et al.,

1992, p. 430). However, the population was likely low as the Akwamu defeated the Gas in a war

in 1678, driving them away to surrounding areas, depopulating their coastal towns (Anquandah,

2006).

Figure 21 Location of Fort James, Ussher Fort, and Fort Christiansborg (left to right)
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
Forts marked in red, current-day Accra highlighted in grey for reference

Customary and social norms embedded in religious beliefs guided Ga pre-colonial notions of

land ownership (Quarcoopome, 1992). The Ga Mashie, Accra’s indigenous inhabitants, believed

that “the ultimate owners of their lands are the gods” who inhabited the lagoons (Quarcoopome,

1992, p. 41). The towns’ inhabitants managed the lands they occupied and could farm unoccu-

pied land that belonged to their quarter. Allodial rights were vested in clans, families and individ-

uals acting as corporate groups and could be obtained through conquest, sale, ritual transfer and

81
the occupation of unused land (Sackeyfio, 2012). To build and settle on a plot required a ritual to

be performed and a small fee to be paid, after which the land was allocated by corporate groups

like clans or families. However, these were usufructuary rights as the land could not be alienated

(Quarcoopome, 1992).

With time, population growth reduced available land and while this transformed ideas of owner-

ship, “[s]uch rights were regulated by codes of conduct rather than proprietary laws that connot-

ed notions of individual ownership” (Sackeyfio, 2012, p. 297). Indeed, a family or clan head could

not sell/alienate land without his group’s approval which, however, occurred as early as the 1830s

in the Ga towns (Sackeyfio, 2012). In the 1870s, unoccupied land between Kinka and Osu (‘New

Site’), an area located within current-day Accra, “was selected as a convenient location for the of-

fices of the growing colonial bureaucracy” (Sackeyfio, 2012, p. 298). The chief of Accra and other

similar authorities transferred this land to the British, further increasing the number of groups

and individuals who could lay some form of claim to usage rights in certain parts of the growing

city.

Over the years, as trade intensified with Europeans, the coastal towns grew with people immi-

grating in search of income opportunities. Historical records show “that the Ga, Akwamu,

Akyem, Asante and Fante as entrepreneurs in collaboration, or competition, with Europeans,

played no small role in the rise of Accra to the pinnacle of international commerce in the 17th

and 18th centuries” (Anquandah, 2006, p. 4). In Accra, European, African (Ga) and also increas-

ingly Euro-African merchants amassed great riches, serving as middlemen in trade between the

Europeans and groups from further inland (Winsnes, 2013; Paller, 2019). They did so in both the

slave trade and after its abolition in the early 19th century when the export of other goods, espe-

cially agricultural produce, replaced the slave trade. However, some of these products, like oil

from the African Oil Palm (Elaeis guineensis), had already been traded before (Lynn, 1997).

82
Many of the Africans and Euro-Africans were educated in fort and missionary schools, while

some were also sent to Europe for educational purposes (Quarcoopome, 1998; Winsnes, 2013).

The relevancy of these developments resides in their associated changes to the growing town.

These “prominent citizens and wealthy trading merchants all constructed houses heavily influ-

enced by European architectural style” (Labi, 2006, p. 132) quite distinct to and separated from

the more densely populated and unplanned areas of Accra. Those who could afford it left Central

Accra and settled in the new suburbs, forming part of the city’s urban elite (Quarcoopome, 1998).

The distinction between often unplanned and over-populated areas inhabited by the African

population and the less densely populated areas inhabited by a predominantly European popula-

tion is, therefore, a simplified one as the richer African and Euro-Africans also resided in the lat-

ter type.

Current-day Accra’s original vegetation cover is thought to have been a dry semi-deciduous forest

that human impact turned into a savannah grassland type (Simmering et al., 2013; Stow et al.,

2013; Owusu, 2018). Though some early European visitors described the area’s floristic composi-

tion and the Natives’ tree uses, these accounts often discussed the Gold Coast in general or plac-

es other than Accra (Cruickshank, 1853; Bosman, 2011; Burton and Cameron, 2011). While Bur-

ton and Cameron collected plants in Axim in 1881, Burton had also travelled further east two

decades prior (1863). Travelling from Cape Coast to Accra by ship, he could “easily distinguish

from the quarter-deck acacias and mimosas, wild dates, adansonias, and guinea palms” (Burton,

2011, p. 132)(Fig. 22). He then described Accra as “dotted with green” (p. 133), “wind-wrung co-

coa [coconut], forming a natural vane” (p. 134) around James Fort while “a stratum of equally

bright green verdure” (p. 134) surrounded Christiansborg Castle. These accounts help picture

early Accra and its surroundings. However, botanists, those with an explicit interest in the disci-

pline, and individuals who lived in Accra for an extended time provide more detailed infor-

mation.

83
 Figure 22 Location of Axim and Cape Coast
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

Two Danish chaplains who resided at Christiansborg Castle one hundred years apart documented

their stays and commented on trees in Accra and its immediate surroundings. Johannes Rask left

for the Gold Coast in 1708 and described the plants growing in the Christiansborg Castle garden,

which included pawpaws and palm trees (Winsnes, 2009a, p. 85-86). More importantly, however,

among the examined accounts, his is among the few providing information on the trees in the ar-

eas surrounding Christiansborg that were inhabited by the African population. Rask narrates that

“[a]long all the streets green trees have been planted, […] which provide a comfortable shade and

a lovely appearance, so that every [settlement7] looks like a small green grove” (Winsnes, 2009a,

pp. 156-157). His comment indicating that Gas actively greened their environs in the early 18th

century. H.C. Monrad, stationed in the fort from 1805 to 1809, dedicated a chapter’s section to

7 I replaced the original (Danish) term used at that time due to its resemblence with current racist terms

84
the area’s flora but only listed and described tree species and their uses (Winsnes, 2009b, pp. 154-

159). Although neither Rask nor Monrad were explicitly interested in botany, their accounts re-

main valuable as early first-hand descriptions of trees in and around Accra/Osu.

In the nearly one hundred years separating the stays of the two chaplains, two botanists, P. E.

Isert (1783-1786) and P. Thonning (1799-1802), travelled to the Gold Coast. Both sent their col-

lected specimens to Copenhagen, where Prof. Schumacher published a book in 1827 describing

almost 500 plant species of the Gold Coast area (Schumacher, 1827). Schumacher “also acquired

Thonning’s notebooks including the detailed and accurate field descriptions of the living plants”

(Adams, 1957, p. 31), and incorporated some of his comments in the 1827 book. While only a

minority of the plants identified are trees, the book provides the most detailed information on

the Gold Coast flora, including several plants described for the first time. In addition to Latin

plant descriptions, Schumacher’s book contains comments in Danish on the trees’ uses and

growth sites.8 Though most trees were observed outside the towns proper, Schumacher mentions

that Dialium guineense Willd. grows in Accra (Schumacher, 1827, p. 18). This species was also pre-

sent in my tree inventory (see section 4.3.1) and remains common in the city’s peri-urban loca-

tions (Simmering et al., 2013). The book is another of the few sources providing information on

the trees in the African towns, mentioning that the natives commonly plant Ficus umbellata Vahl

for its fast growth and shade (see Appendix 8 for a list of all trees in Schumacher’s book).

4.1.2 Accra and Trees – The Colonial Past

In this section, I discuss tree planting as well as town planning and development in colonial Ac-

cra. These two are implicated in co-producing the spatio-temporal dynamics of the city’s urban

forest and continue to affect its structure until today. Official annual government reports permit

8 Part of the almost 500 species descriptions were also translated into German, particularly those that included either
the local name and/or comments on its use in the original version (Ascherson, 1879).

85
tracing town development and the long history of tree-related activities in Accra. The city’s Town

Council was another interest group involved in these activities. Established in 1898 with the

“purpose of training the inhabitants of this country in local self-government” (PRAAD, CSO

20/1/3a, p. 3), it consisted of African and European councillors and derived its income from li-

cences, land rates, and a government grant (PRAAD, ADM 5/1/27). In addition to general

monthly meetings, the council also discussed specific issues such as the destruction of seedlings

planted along roads (PRAAD, CSO 20/1/3b). The multiple urban forestry actors were not just

financially connected – the Town Council providing a yearly budget to the Department of Agri-

culture for tree planting and maintenance of ornamental gardens (PRAAD, CSO 20/1/7a) – but

also debated responsibilities and questions of ownership (PRAAD, CSO 20/1/7b).

In 1862, an earthquake destroyed large parts of Accra, which allowed the colonial power to re-

build the space according to their wishes (Grant and Yankson, 2003). When the British moved

headquarters from Cape Coast to Accra in 1877, the town started growing due to increased

commerce. As Accra grew, so did the demand for land by the government, firms, and individuals

to construct buildings and associated infrastructure. In 1876, the Public Lands Ordinance was

enacted, regulating “the acquisition of land required for public service of the colony and protect-

ed territories as well as the manner in which such lands would be held” (Sackeyfio, 2012, p. 299).

It allowed property to be valued as a commercial asset, and while Africans could assert their

ownership rights, the British written documentation process often superseded oral evidence

(Sackeyfio, 2012). In 1894, the British attempted to introduce the Crown Lands Bill, which would

have vested all unoccupied ‘waste lands’ in the Crown (Omosini, 1972). Chiefs and the educated

African elite, however, protested this bill, arguing that “there was not an inch of land without an

owner under customary law” (Omosini, 1972, p. 458). In their opinion, the Public Lands Ordi-

nance already “gave the colonial government adequate allowance to acquire land for legitimate

public uses” (Nti, 2012, p. 7).

86
Three years later, in 1897, the British attempted to introduce a bill that sought to provide the

Crown with administrative rights rather than actual ownership. What appeared like a scaled-down

version was, however, more elaborate and expansive, including not just ‘waste lands’ but all land

in the country. Opposing the bill, chiefs, elites, and British commercial interest groups formed

the Gold Coast Aborigines’ Rights Protection Society (Gocking, 2005). It argued that “the con-

version of customary title into English certificates of individual ownership would destroy the

chiefs’ and community rights in land” (Omosini, 1972, p. 463). In 1898, a deputation of the socie-

ty met Colonial Secretary Chamberlain to voice their objections and to propose amendments to

the bill. Surprisingly, he conceded all points, and in 1900 the Concessions Ordinance formalised

the principle “that all lands were vested in African communities and chiefs and not in the crown

or governor” (Omosini, 1972, p. 467).

Based on the evidence of early colonial interest in city trees, I now discuss two interrelated core

aspects of urban forestry: the rationale behind what trees (species selection) are planted where

(site selection). Species selection determines what types of benefits are produced, archival records

suggesting that the colonial government focused on three services: trees for shade, beautification,

and economic development. While some tree benefits are not location-specific, many of them

are. A tree offers shade, where it is located, beautifies its immediate surroundings, and reduces

wind and noise in its vicinity. Therefore, a tree’s location, and often by extension ownership sta-

tus, affects who can benefit from its services. In colonial Accra, site-selection heavily favoured

the city’s European areas constituting, along with town planning characteristics, one of the most

fundamental legacies to impact contemporary urban forest structure. I now discuss the elements

of species and site selection in more detail.

87
4.1.2.1 Shading Accra
Prioritising shade trees is a logical response to Accra’s bioregional context, which the British

hoped to ameliorate through tree planting. In 1895, the Botanical Garden Department cleared

some land at the Governor’s Lodge to establish a tree nursery “to raise different kinds of shade

and other trees for planting in and around Accra” (PRAAD, ADM 5/1/51, p. 57). Two years lat-

er, 800 shade trees had been raised, now large enough to be transplanted during the next rainy

season (PRAAD, ADM 5/1/53). Additionally, 400 seedlings were brought to the city from Aburi

station, located 40 km north of Accra and the Botanical Garden Department’s head station.

However, many of these did not successfully establish in Accra’s arid climate and died during the

dry season. The British planted these seedlings in the Club House, the Colonial Secretary’s bun-

galow, and the Inspector General’s quarters (PRAAD, ADM 5/1/53). Throughout the years,

shade trees were commonly planted in Accra, the records mentioning 3,000 in 1912 (PRAAD,

ADM 5/1/68) and 4,500 in 1922 (PRAAD, ADM 5/1/79), for example.

The preference for shade trees has survived into the present, prioritised by DPG and EPA (Mr

Awidi, DPG; Mr Fiati, EPA) and the most frequently named benefit in my household survey.

While planting shade trees is a reaction to the area’s bioregional context, the climatic conditions

themselves constitute a biophysical legacy. Though the British recognised that “young plants will

require a lot of water, especially in the dry season” (PRAAD, ADM 5/1/51, p. 57), they more

than once lost trees to drought (PRAAD, ADM 5/1/58; PRAAD, ADM 5/1/59; PRAAD,

ADM 5/1/60) and mentioned the positive impact of unusually heavy rains (PRAAD, ADM

5/1/74). Scarce rainfall tends to influence urban forest spatio-temporal dynamics on a shorter-

term basis only, as smaller seedlings are especially susceptible to drought but can, theoretically, be

easily replaced. However, urban forest planners and managers should not underestimate how in-

sufficient and infrequent precipitation impedes tree establishment. One interviewed stakeholder,

88
an NGO representative, reported losing all their planted trees, partly because they lacked the re-

sources to water them regularly (Interview Y4GG).

4.1.2.2 Beautifying Accra

Colonial greening efforts also focused on growing decorative plants. In the 1920s, the British

planted them in significant numbers, mostly along avenues in their neighbourhoods and on gov-

ernment compounds (PRAAD, ADM 5/1/78; PRAAD, ADM 5/1/79; PRAAD, ADM 5/1/88;

PRAAD, ADM 5/1/89). The Agricultural Department provided the seedlings and assisted the

European population in designing their bungalow grounds (PRAAD, ADM 5/1/73, PRAAD,

ADM 5/1/75). These efforts indicate the value the British attached to living in vegetated and

green surroundings, a human legacy factor (Roman et al., 2018). The British also supplied trees to

the African population (PRAAD, ADM 5/1/77; PRAAD, ADM 5/1/79), encouraging “the

growing of flowering and decorative plants by natives in the vicinity of their homesteads”

(PRAAD, ADM 5/1/73, p. 16). Additionally, chiefs were urged to promote tree planting, the

British offering free seeds, trees, and advice (PRAAD, ADM 5/1/79). However, provided seed-

lings were few and limited planning interventions caused crowded and increasingly overcrowded

African neighbourhoods (Larbi, 1996). As shelter is not only a basic need but also a critical in-

come-generating asset in the urban context (Moser, 1998), people likely prioritised housing over

tree planting.

The Town Council meetings’ minutes reveal that tree planting in the 1930s focused on High

Street and other European dominated areas like Post Office Square or Christiansborg Cross

Roads, where trees were planted to beautify Accra (PRAAD, CSO 20/1/7a). In 1936, the Town

Council appointed a special committee tasked to control and coordinate arboricultural and horti-

cultural works on Crown land in Accra. The group was responsible for hiring staff and super-

89
vised tree planting and cutting on municipal areas and government residential land (PRAAD, RG

3/5/349). The committee considered hiring a full-time officer from Kew Botanic Gardens as

“the work to be carried out was of a difficult nature, and could not be satisfactorily executed

without a large measure of European supervision” (PRAAD, CSO 20/1/66, p. 1). However, they

apparently abandoned the plan due to the involved costs. Departmental reports and meetings’

minutes suggest that the European neighbourhoods tended to be fairly green during colonial

times. Most African neighbourhoods, though, were more densely populated, with 16 to 24 inhab-

itants per hectare compared to two per hectare in the European areas in the 1940s (PRAAD,

CSO 20/12/20). Restricted planting space combined with the British government’s limited plant-

ing efforts in these areas likely caused African neighbourhoods to be less green.

These planning and planting patterns are frequently visible in cities with a colonial history, where

segregation created distinct European and native areas. Pryor (2015, p. 36), who investigated co-

lonial Hong Kong’s tree planting history, reported that in the early period, “[n]o trees appear to

have been planted […] within the Chinese Quarters”, despite widespread planting elsewhere in

the town. Likewise, greening in Delhi benefitted “specifically that part of the city which was in-

habited by the white colonizers” (Mann and Sehrawat, 2009, p. 553). In previously French towns,

such as Dakar, similar developments played a “crucial role in the creation of racially polarized co-

lonial urban environments” (Bigon, 2013, p. 477). In many of these places, “street trees came to

emphasise the informal racial segregation of the city” (Pryor, 2015, p. 40). Acknowledging urban

forestry’s political nature avoids creating and reproducing oversimplified explanations. While rac-

ist reasoning is no longer employed to explain Accra’s tree cover variation, narratives that suggest

a causal relationship with socio-economic status have often replaced them, a narrative for which I

did not find any support (see Chapters 5 and 6).

90
4.1.2.3 Monetising Urban Land
In the early 20th century, the British attempted to commercialise Accra’s land and planted eco-

nomically valuable trees. These activities constitute a geographical extension of previous colonial

plantation agriculture, which, as ‘legitimate commerce’, was supposed to replace the slave trade

(Austin, 2013). In 1901, they sowed 13,000 coconuts at Christiansborg Castle Garden, later trans-

ferring these to a 90 ha plantation site located between the castle and Labadi, located just 4 km to

the east, as part of efforts to develop the country’s copra industry. While 9,000 coconuts germi-

nated, many others were stolen (PRAAD, ADM 5/1/56) or killed by droughts and insect attacks

(PRAAD, ADM 5/1/58; PRAAD, ADM 5/1/59). In 1905 alone, beetles destroyed more than

4,000 coconut seedlings (PRAAD, ADM 5/1/60) and, as later reports contain no further refer-

ences to the plantation, the British would appear to have given up on this endeavour. Still, the in-

tervention is a reminder that a key aim behind the establishment of Aburi station was to “teach

the natives to cultivate economic plants in a systematic manner for purposes of export”

(PRAAD, ADM 5/1/76, p. 5). However, beetle pest infestation and challenging climatic condi-

tions, both biophysical (legacy) factors, limited the achievement of this goal, while still affecting

tree distribution at the time.

By 1908, the Botanical Garden Department had become part of the Agricultural Department,

and the reports’ style shifted to a more scientific and technical tone. Their content now focused

on exportable cash crops like cotton, oil palm, kola nuts, and rubber (PRAAD, ADM 5/1/63),

and it took several years before a Division of Horticulture was created. The division became re-

sponsible for “the upkeep of public gardens and the provision of assistance in avenue planting in

towns” (PRAAD, ADM 5/1/88, p. 12), but archival records provide limited information on its

work. In the following years, the Agricultural Department expanded its research on economically

attractive plants and stated in 1915 that, due to the “necessity for experimental work on agricul-

tural crops, little opportunity occurs at any of the departmental Stations for horticultural effort”

91
(PRAAD, ADM 5/1/72, p. 17). From the 1930s on, urban trees disappear from the reports and

cocoa production and plant pathology dominate (PRAAD, ADM 5/1/94; PRAAD, ADM

5/1/101), reflecting this cash crop’s economic importance.

4.1.2.4 Colonial Town Planning

The city’s designation and subsequent growth as colonial headquarters reflected its newfound

importance and associated cumulative advantages. For the British who now lived and worked

there, its development became a central interest. Some general principles, including physically

separating European and African areas, guided early colonial town development (Grant and

Yankson, 2003). However, despite the city’s expansion in the late 19th and early 20th century,

town, regional, and country planning formally became an issue only in the 1930s. At this time,

Lord Passfield, the Secretary of State for the Colonies, was a pioneer in recognising the need for

effective town and country planning overseas. Arguing that “[n]othing is more expensive than

haphazard or narrowly conceived development which will later involve the costly undoing of ear-

lier mistakes” (PRAAD, CSO 20/12/3, p. 1), he suggested appointing a Regional and Town

Planning Officer. The Director of Public Works and the Governor, however, offered resistance,

delaying implementation for over a decade. The British only took further steps in 1944, when

they officially listed Accra, Sekondi-Takoradi, Kumasi, Cape Coast, and Tarkwa as places that

needed a Town Planning advisor’s services (Fig. 23).

92
 Figure 23 Places listed as needing a town planning advisor
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

In 1945, the British enacted the Gold Coast Town and Country Planning Ordinance (CAP 84),

which granted a Town and Country Planning Board the power to establish ‘planning areas’. In

these sites, the Board controlled development, granting building permits and preventing unau-

thorised construction. Effectively, the British had gradually undermined pre-colonial norms of

land ownership and use, which had “shifted from Ga traditional authorities to the central gov-

ernment” (Quarcoopome, 1992, p. 42) (see Sections 4.1.1 and 4.1.2). In Accra, an increasing de-

mand for land, and rising prices paid for it, facilitated alienation. Furthermore, natural resources

offered no alternative path to wealth, and the population was obliged to allow the government to

93
develop stool and private land for public use (Quarcoopome, 1992). By the 1950s, large parts of

Accra were no longer under Ga ownership, hindering the younger generation to “benefit from

their family land holdings as a matter of birthright” (Quarcoopome, 1992, p. 46). The Ga youth,

living in now densely populated Usshertown or Jamestown, could no longer move as the peri-

urban land had been alienated. The resulting mosaic of tenure affected and continues to affect

Accra’s development, including urban forestry, where the government lacks ownership and land

disputes obstruct holistic planning.

4.1.3 Accra and Trees – The Post-Independence Past

With independence in 1957, the colonial government reports ceased, but urban forestry in Accra

experienced further profound changes in the following decades. I discuss three aspects in more

detail, including continued town planning activities and the role of trees within this, the Depart-

ment of Parks and Gardens and its work, as well as socio-economic developments in the country.

Their separate examination does not imply independence of each other as the previously devel-

oped framework demonstrates how actors, such as the Department of Parks and Gardens, oper-

ate within a wider context, like the economic situation of a region or nation or more local pro-

cesses, e.g. in the form of town planning schemes.

4.1.3.1 Tree Planting and Town Planning Post-Independence

In 1958, the Government prepared a ‘Plan for the Town’, revising a first plan from 1944. Clerk

and Fry, the latter Town Planning Advisor to West Africa, had developed the earlier plan, which

promoted a ‘garden city’ approach. These towns feature “neighbourhoods connected by an outer

ringroad with each area served by its own shops, community centres, schools and parks” (Jack-

son, 2019, p. 103) and gardens separating the settlements (Jackson and Holland, 2014). The pair

94
also developed Kumasi’s planning schemes, which were “rhapsodic over future possibilities […]

but sober in its detailing of basic enablements” (McCaskie, 2000, p. 205), potentially also the case

in Accra’s plan. The earlier plan never became a statutory one, but the continuous planning ef-

forts supposedly created favourable preconditions for its successor (PRAAD, RG 8/2/779). The

1958 plan describes the city as “fortunate in so far as existing and potential open spaces are con-

cerned” as the Government “has had the foresight to retain extensive building free zones”

(PRAAD, RG 8/2/779, p. 23). These open spaces receive praise for “providing both amenity

and opportunities for relaxation and recreation”, recognising, however, that the “distribution of

the open space throughout the town is important” to maximise benefits for the residential areas

(PRAAD, RG 8/2/779, p. 23). These excerpts demonstrate, once more, how location and spe-

cies choice (i.e. the desired benefits) are two central aspects of urban forestry.

Town planning in Accra, loosely following ‘garden city’ principles, reinforced the city’s spatial dif-

ferences. The ‘green wedges’ radiating through the town would separate “the residential areas

from each other and thus providing them with a physical boundary which helps to give a special

character to each area” (PRAAD, RG 8/2/779, p. 24). Supplying each neighbourhood with its

own facilities reduces travel time and increases access to schools or shops but also preserves co-

lonial green space inequalities. With independence, political power had shifted to an indigenous

elite, yet poorer people and new immigrants still settled in the cheaper, high-density areas.

Though parts of the Ghanaian middle and upper class already lived in less populated places (see

Section 4.1.1), they now also moved into the spacious and green areas that used to be European

(Grant and Yankson, 2003). Early Ghanaian town planning, therefore, retained and reinforced in-

tra-city variations of population density or green cover that originated in colonial times.

The British focus on tree planting for shade and beautification also survived the transition to in-

dependence. Town gardens, street trees and parks featured prominently in the 1958 planning

95
documents, the former envisaged as recreational sites for workers, shoppers, and visitors or fram-

ing notable buildings (PRAAD, RG 8/2/779). Street trees, in turn, were valued for their beauty

and “the vitally important question of shade which makes all the difference between pleasant

conditions and acute discomfort” (PRAAD, RG 8/2/779, p. 30). The four areas that were desig-

nated to become parks were located in central Accra and totalled 60 ha of land. They were con-

sidered vital “as a place where one may walk in quiet and pleasant surroundings and where chil-

dren may play” (PRAAD, RG 8/2/779, p. 24). The 1958 plan, however, was not adhered to

(Arku et al., 2016), as undeveloped open spaces are often considered free to draw upon for con-

struction projects (PRAAD, RG 8/2/779). Today, these sites feature buildings, albeit with a fair

amount of trees, being located in the greener neighbourhoods of Asylum Down and North

Ridge. Until today, parks in Accra are few, and the existing ones often difficult to access or poor-

ly maintained (Arku et al., 2016).

Like its predecessor, the post-independence government prioritised selected public tree benefits,

disregarding their provisioning services despite the city’s expansion into previously rural locations

where trees fulfil this role (Gakou et al., 1994; Kaimowitz, 2003; Kalaba et al., 2009). Though the

government recognised the need for food production in urban areas, it delegated this responsibil-

ity to privately owned food growing gardens (PRAAD, RG 8/2/779, p. 5). Until today, these

benefits, which primarily accrue on the individual to household scale, occupy little space on offi-

cial agendas (e.g. interviews with DPG and EPA). Instead, government departments often priori-

tise services that are important on a national scale, e.g. avoided spending on road maintenance

through tree shade, while mitigating climate change’s adverse effects even matters at a global

scale.

Town planning in Accra not only guided development within existing neighbourhoods but also

attempted to regulate the city’s outward growth. Accra’s continuous expansion incorporated land

96
into the city, which was previously located outside its boundaries. In the early 1920s, the British

established Achimota Forest as a firewood plantation north of the city (PRAAD, ADM 5/1/80),

the forest now part of Accra metropolis. Described in 1958 as “not only a useful source of fire-

wood but also an important landscape feature” (PRAAD, RG 8/2/779), it is no longer used for

the former, but its importance concerning the latter has grown. Today, the Achimota Forest is a

vital greenbelt and recreation area (Interview TCPD, Interview GYEM), but its history continues

to affect current-day management. Though DPG is responsible for maintaining Accra’s trees, the

Forestry Commission is accountable for Achimota Forest, creating dual responsibility for the

city’s tree cover. Additionally, FC disagrees with the Town and Country Planning Department

about the forest’s purpose and its management. While the former argues that it should be trans-

ferred to more profitable land use, which supposedly would also protect it against encroachment,

the latter favours retaining Achimota as part of Accra’s greenbelt (Mr Yeboah, TCPD).

Disagreement over Achimota includes other groups, which contest the forest’s current ownership

as the various co-existing land tenure types in Accra raise questions about the limits of, and over-

laps between, usufructuary and ownership rights. In the 2010s, the Osu division’s chief filed to

reclaim about 70 hectares of the Achimota forest as it was not utilised for the purpose for which

the colonial government compulsorily acquired it. The Chief also backed a private developer who

claimed part of the land, and both the police and FC officials became involved when the latter

erected a fence to protect the land against the developer (The Chronicle, 2014a; The Chronicle,

2014b; The Chronicle, 2014c; The Chronicle, 2014d; Daily Graphic, 2014a; Daily Graphic,

2014b). Furthermore, the La Stool argued that the disputed land was theirs and did not belong to

the Osu Stool (The Chronicle, 2014e). Disagreements caused by multiple land tenure systems and

competing interest groups can have direct and lasting effects on land use. In the case of Achimo-

ta, these conflicts, among other reasons, reduced its size from 494.95 hectares in 1930 to 341.28

hectares in 2015 (Tuffour-Mills et al., 2020).

97
4.1.3.2 The Department of Parks and Gardens
Post-independence, the Department of Agriculture, the Planning Department, and the Arboricul-

tural and Horticultural Committee carried out “[m]uch valuable tree planting work along main

roads and on open spaces” (PRAAD, RG 8/2/779, p. 30). In 1961, the Department of Parks and

Gardens was established, and four years later, Nkrumah supported a proposal to upgrade it to the

Ministry of Parks and Gardens and Tourism. After the First Republic’s fall in the following year,

the Ministry was demoted to a department under the Forestry Department. It was then trans-

ferred to the Ministry of Works and Housing, the Ministry of Local Government and Rural De-

velopment (MLGRD), the Ministry of Environment and Science and back to the MLGRD,

where it is still located today. When government priorities change, urban forestry can become

overlooked or (un)intentionally sidelined, either when personnel is substituted or in the process

of bureaucratic and administrative restructuring. Therefore, some state and non-state actors wish

to stay ‘politically neutral’ (Interview Y4GG, Interview FORIG), avoiding too close an associa-

tion with politics to minimise the risk of losing funding or support when urban forestry is reor-

ganised.

Despite Nkrumah’s early interest in urban greenery, the Department of Parks and Gardens today

suffers from a lack of (financial) support and prioritisation of its work. At the district level, the

DPG is merged with TCPD to form the Physical Planning Department. However, interviewees

suggested a merger in name only and limited practical operational details in place. Recently, a

study that aimed to identify factors that differentiate ‘good’ and ‘poor’ performing public organi-

sations in Ghana classified DPG as a ‘poor performer’ (Owusu, 2006). The author started by

compiling a list of 47 institutions and then tasked 23 ‘knowledgeables’ to rate them based on the

ability to perform their respective functions. These ‘knowledgeables’ included individuals from a

range of backgrounds, working in “government agencies, bilateral and multilateral agencies, non-

governmental organizations, academic institutions, research organizations and the private sector”

98
(Owusu, 2006, p. 476). Creating mean scores for each organisation, the 13 highest and lowest

ranking ones were then classified as ‘good’ or ‘poor’ performers. While the author acknowledges

possible limitations due to the sample size, several interviewees (Interview Y4GG, Interview

TCPD) and newspaper articles (Ghanaian Times 2015; Daily Graphic 2017a; Daily Guide 2017)

also commented on DPG’s poor performance.

4.1.3.3 The Socio-Economic Context

Both town planning and DPG’s work take place in a specific socio-economic context. At inde-

pendence, Accra’s economy was non-manufacturing and dominated mainly by foreign export

trading firms (Grant and Yankson, 2003). To facilitate economic growth, Nkrumah intended to

shift emphasis away from commerce, which had been the focus during colonial times, to import

substitution industrialisation. He envisioned a new ‘Industrial Triangle’ formed by Accra, Tema

and Akosombo (Ahlman, 2017). While Tema, located just east of Accra, was the proposed site

for the new industrial harbour, Akosombo was strategically located at the Volta river to the

North (Fig. 24). Nkrumah’s policies attracted a growing amount of industries to Accra, and peo-

ple followed, searching for jobs and drawn by services that were unavailable elsewhere in the

country (Adarkwa, 2012). Accra’s high population growth rate destroyed the government’s ambi-

tious plans to provide adequate housing to all inhabitants. New immigrants, therefore, settled at

the city’s outskirts where slums and unplanned satellite towns developed to the vegetation’s det-

riment (Government of Ghana, 1964).

99
 Figure 24 Location of Accra, Akosombo, and Tema
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

By the early 1980s, “economic mismanagement, over-regulation, inability to control inflation,

productivity disincentives, an over-bloated and mismanaged public sector, over-subsidized social

services (education and health in particular), overvalued currency, political instability, corruption,

inept leadership, acts of God, unfavorable terms of trade, and clientelism” (Konadu-Agyemang,

2000, p. 473) had brought Ghana close to bankruptcy. To save the economy, the country imple-

mented the Structural Adjustment Program (SAP) and its neoliberal, ‘free-market’ policies in

1983. Between 1975 and 1997, Accra’s area increased by 318 %; however, most of this expansion

occurred in the post-SAP years and took the form of “encroaching on rural settlements or into

green spaces between rural settlements” (Yeboah, 2000, p. 70). Additionally, the post-SAP years

also coincided with the period of greatest vegetation loss within Accra (Owusu, 2018). According

100
to Owusu (2018), increased urbanisation and green space conversion into residential and com-

mercial land directly relate to the SAP and its predecessor, the Economic Recovery Program.

Structural adjustment indirectly affected urban and peri-urban tree cover through various mecha-

nisms. SAP’s changes to the finance and transportation system facilitated urban growth and ur-

banisation (Briggs and Yeboah, 2001). More liberal currency transaction laws caused increased

remittances from expatriates while reduced import/export controls aided capital accumulation,

often invested in real estate. Additionally, deregulated public transportation improved access to

peri-urban zones, increasing development in these fringe areas (Briggs and Yeboah, 2001). As

people anticipated “infrastructure and services like roads, drains, electricity, water, and telephone

services” (Yeboah, 2000, p. 72), the capital’s surroundings densified and houses replaced vegeta-

tion. Structural adjustment also cut state expenditures on social welfare and public services

(Konadu-Agyemang, 2000), disproportionately affecting society’s more vulnerable groups, includ-

ing the urban poor, whose numbers rose from 9 to 23 % (Grant and Yankson, 2003). A sizeable

proportion of Accra’s population rents accommodation (UN-HABITAT, 2011), especially in the

densely populated, lower-income neighbourhoods (Yankson, 2012). Interviewed tenants often

expressed little interest in planting trees in their compounds, even if allowed by their land-

lords/landladies, as they expect to move before being able to reap the benefits. Thus, urban

compounds often lack trees unless the landowners plant and maintain them (see Chapter 6).

4.2 Accra and Trees – The Present

In the late 1950s, the colonial ‘green wedges’ gradually disappeared as the government allowed

residents to encroach upon them, primarily in low-density areas with sufficient open spaces

(PRAAD, RG 8/2/779). Until today, though, neighbourhoods continue to differ in terms of

greenery. If legacy effects exist, formerly European areas should tend to be greener and less pop-

101
ulated, while higher population densities in previously ‘native’ areas would limit canopy cover.

However, these two configurations constitute a continuum’s two ends (see Section 4.1.1), and

neighbourhoods themselves can be heterogeneous like Nima, its southern parts greener and less

built-up (Cobbinah et al., 2021). Previously European areas include Cantonments, Ridge (presum-

ably North and West Ridge) (Rain et al., 2011), Kanda, Ringway Estate, and Airport Residential

(Quayson, 2014), while Nima, Jamestown (Andersson, 2016), and Sabon Zongo (Pellow, 2001)

were African areas (Fig. 25). As neighbourhood boundaries and names change, contemporary

residential areas not necessarily conform to historical ones. Additionally, many current outskirt

neighbourhoods used to be rural hinterlands in colonial times, located well outside the city’s

boundaries. While colonialism is but one legacy factor, it can leave unmodifiable marks due to its

strong relationship with city planning and development. In this section, I first describe Accra’s

urban forest at city-level before discussing how it varies at sub-city level.

102
 Figure 25 Previously European and African neighbourhoods
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
24 – Airport Residential 36 – Nima 37 – Kanda 47 – North Ridge 48 – Cantonments
53 – Sabon Zongo 56 – West Ridge 57 – Ringway Estate 75 – Jamestown

4.2.1 City-Level
Estimates of Accra’s green cover are limited and vary considerably. According to my desk-based

iTree Canopy analysis, the city’s tree cover totals 14.0 % (± 1.55). However, the ground-based

iTree Eco approach suggests a canopy cover of only 8.2 %, translating to 391,600 trees in the

study site. As the aerial photographs’ low quality impeded the iTree Canopy analysis, I might

have misclassified bushes or other green areas as trees, potentially overestimating Accra’s canopy

cover. In comparison, Stow et al. (2013) reported a vegetation cover of 29.2 % in 2010, consider-

ably higher than the 13.45 % estimated by Owusu (2018) for the end of 2013. Both of these stud-

103
ies used satellite images and determined not just the tree cover but the amount of all green, vege-

tated land, including grasslands or bushes. Additionally, the study site boundaries in these earlier

studies do not coincide with the ones used for this study. Stow et al. included only about two-

thirds of Accra Metropolitan Assembly (AMA), while Owusu measured vegetation cover in areas

that I had excluded. As neither the Department of Parks and Gardens nor the Environmental

Protection Agency could provide further information, I assume that Accra’s canopy cover is

somewhere between 8.2 % and 14 % (plantable space according to iTree Canopy equals 40.4 % ±

2.19, non-plantable space 45.6 % ± 2.23).

Researchers and practitioners have estimated canopy cover values for cities all around the world,

but these figures should not be directly compared. First, research methods and definitions of

what constitutes a ‘tree’ may vary. More importantly, “there is no set canopy cover level that

would be considered ‘optimal’ everywhere – or perhaps even in a single location” (Leff, 2016, p.

12), thus other cities’ values cannot serve as reference points. Instead, place-specific inquiries

must determine the amount of tree cover that is desirable, attainable and sustainable considering

a location’s area, climate, original vegetation, urban density, and financial resources. My research

provides current baseline information on Accra’s tree canopy, which can be used to analyse and

monitor future changes.

Though tree cover should not be compared between cities, Accra has experienced land-use

changes similar to many other urban agglomerations. Between 1986 and 2013, the city’s vegetated

area reduced from about 38 % to 13.6 %, with most green cover lost in the northern part where

previously open land became built-up (Owusu, 2018). Similarly, Stow et al. (2013) report a de-

crease of almost 6 % between 2002 and 2010, caused by vegetation being converted into residen-

tial and commercial land use. Both in Kumasi (Mensah, 2014a; Nero, 2017) and Sekondi-

Takoradi (Mensah et al., 2018) developments were similar. Despite Tema’s built-up area increas-

104
ing between 1990 and 2007, green space also increased temporarily due to rising rainfall and an

afforestation programme (Mariwah et al., 2017). While this tree planting exercise first increased

vegetation cover, it decreased again when the trees were felled for sale, demonstrating urban for-

estry’s impact on canopy levels.

Across the 200 iTree Eco plots, I recorded a total of 190 trees, considerably fewer than Morgen-

roth et al. (2020) inventoried in 38 US cities using the same approach. Sixteen trees remain uni-

dentified, and 20 were only identified to genus level. Trees identified to species level belonged to

27 species, the most common ones neem (Azadirachta indica A. Juss) with 32 individuals, mango

(Mangifera indica L.) with 16 trees and African oil palm (Elaeis guineensis Jacq.) with 14 plants. I re-

corded eleven individuals each of coconut palm (Cocos nucifera L.), Indian Cork tree (Millingtonia

hortensis L.f.) and white lead tree (Leucaena leucocephala (Lam.) de Wit) while papaya (Carica papaya

L.) and royal palm (Roystonea regia (Kunth) O.F.Cook) were present with eight plants each. Addi-

tionally, there were five avocado (Persea Americana Mill.), false ashoka (Polyalthia longifolia Sonn.),

and moringa trees (Moringa oleifera Lam.). Species with less than five individuals include, among

others, cacao (Theobroma cacao L.), soursop (Annona muricata L.), ackee (Blighia sapida K.D.Koenig),

lime (Citrus aurantiifolia (Christm.) Swingle), and cashew (Anacardium occidentale L.)(Table 14). While

another study in six of Accra’s neighbourhoods recorded more species, neem, mango, oil palm

and coconut were also the most common ones (Nero et al., 2018b). Overall, my sampled trees

were in good health and vigour, and only 25 individuals received a crown condition rating of less

than 90 %, only six of which I rated as ‘dead’.

Table 14 Recorded species and number of trees per species

Species Number of Trees

Neem Tree (Azadirachta indica) 32

Mango (Mangifera indica) 16

African oil palm (Elaeis guineensis) 14

105
Coconut palm (Cocos nucifera) 11

White lead tree (Leucaena leucocephala) 11

Indian cork tree (Millingtonia hortensis) 11

Florida royal palm (Roystonea regia) 8

Papaya (Carica papaya) 8

Avocado (Persea americana) 5

False ashoka (Polyalthia longifolia) 5

Moringa (Moringa oleifera) 5

Cacao (Theobroma cacao) 3

Soursop (Annona muricata) 3

Ackee (Blighia sapida) 2

Key Lime (Citrus aurantiifolia) 2

Senegal mahogany (Khaya senegalensis (Desr.) A. Juss.) 2

Yellow flametree (Peltophorum pterocarpum (DC.) Backer ex K.Heyne) 2

Velvet tamarind (Dialium guineense Willd.) 2

Madagascar almond tree (Terminalia mantaly H.Perrier) 2

Millettia thonningii (Schumach.) Baker 2

Rain tree (Albizia saman/Samanea saman (Jacq.) Merr.) 2

Albizia ferruginea (Albizia ferruginea (Guill. & Perr.) Benth.) 1

Cashew (Anacardium occidentale) 1

Roughbark lignum-vitae (Guaiacum officinale L.) 1

Boundary tree (Newbouldia laevis (P.Beauv.) Seem. ex Bureau) 1

Indian almond tree (Terminalia catappa L.) 1

Brimstone tree (Morinda lucida Benth.) 1

106
Tree and palm species commonly planted by the Department of Parks and Gardens (Table 15)

were underrepresented in the sample, potentially due to its limited planting activities, both in

numbers and covered area.9

Table 15 List of species commonly planted by DPG

Botanic Name Common Name Function

Blighia sapida Ackee apple Foliage with fruit

Cananga odorata Lavender tree Fragrance

Caryota urens Toddy palm Foliage

Cassia fistula Golden showers Yellow flowering (beautification)

Chrysalidocarpus lutescens Cane palm Foliage

Cocos nucifera Coconut palm Fruiting

Delonix regia Flame of the forest Red flowering (beautification)

Eucalyptus multifida Blue gum Foliage for sound absorption

Jacaranda mimosaefolia Jacaranda Blue flowering (beautification)

Queen flower/ Pride of

Lagerstroemia speciosa Flowering (beautification)
India

Lignum vitae Tree of Life Mauve flowering (beautification)

Michellea champaca Champac tree Fragrance

Millettia thonningii Millettia Flowering

Millingtonia hortensis India cork tree White flowering

Mimusops elengii Mimosups Foliage/Shade

Phitecellobium saman Rain tree Foliage/Shade

White/yellow flowering (beauti-

Plumeria alba/rubra Frangipani
fication)

9 Botanic/common names may differ from those in Table 14, as I used them as provided by DPG

107
 Roystonia regia Royal palm Palm for avenue

Tabebuia rosea Tabebuia Pink flowering (beautification)

Terminalia ivoriensis Umbrella tree Foliage/Shade

Thuja orientalis Oriental arbor vitae Foliage

Source: Mr Awidi, DPG

Furthermore, most inventoried trees grew on residential land, 97 on single-family and 18 on mul-

ti-family property (Fig. 26), the 28 trees belonging to eleven different species. Nero et al. (2018b)

described similar high levels of species richness and abundance in home gardens in Accra. Four-

teen of the 15 individuals growing on vacant land were neem trees. Neem’s potential uses and

benefits have led to the tree being described variously as ‘green treasure’ (Girish and Shankara,

2008), ‘nature’s drugstore’ (Hashmat et al., 2012), and ‘a tree for solving global problems’ (Na-

tional Research Council, 2002). However, the species has also been declared a pest/weed in some

parts of the world, including in Australia, where it “has a reputation for surviving in harsh, dry,

infertile soils” (State of Queensland, 2016, p. 8). Together with Leucaena leucocephala and Millington-

ia spp., also recorded in my tree inventory, neem has been listed among the top 20, respectively

30, invasive alien plants in Ghana (CAB International, 2004; Boy and Witt, 2013). Considering its

dominance on vacant land, there is a risk of neem replacing native species, especially when man-

agement is lacking or absent. While city residents plant trees to claim land ownership, the neem

trees’ growing locations, such as in the middle of building ruins, suggest that some plants likely

resulted from natural regeneration instead.

108
 Land-Use Classification of Sampled Trees
1% 1% 2%
Cemetery
3% Park
3%

7%
Commercial/Industrial
Transport

8% Wetland
Institutional
51%
Vacant
10%
Multi-family residential
Agricultural
14% Residential

Figure 26 Land-use classification of sampled trees

A tree’s ownership status is closely related to the land use on which it grows. As part of the in-

ventory, I classified all trees within the 200 plots as either ‘public’ or ‘private’. Similarly to classify-

ing the land use, identifying tree ownership is prone to mistakes, and I might have misclassified

some trees. However, species composition of Accra’s private and public urban forest parts differs

(Fig. 27), suggesting that private individuals’ selection criteria and those adopted by government

bodies for public lands also differ. Several species appeared only in public ownership, such as In-

dian cork or almond tree, while soursop, cacao and avocado were all encountered on private land.

Even where a species grew on both land types, it often predominated in one as in the case of

neem, mango or papaya. With the exception of ackee, all other food trees were either exclusively

privately owned or more common in this land type than on public ones. I discuss species selec-

tion criteria and urban forest management objectives and priorities of both public and private ac-

tors in more detail in Chapters 5 and 6.

109
 Number of Trees in Public and Private Ownership Divided
by Species
25

20
Number of trees

15

10

5
Private
0 Public

Species

Figure 27 Number of trees in public and private ownership divided by species

Tree height and diameter at breast height (DBH) are two commonly recorded inventory varia-

bles, which describe both the current urban forest structure and serve to predict future develop-

ments. The inventoried trees’ average height was 6.2 m, and more than half were up to 5 m high

(Fig 28). Lower trees dominate in Accra’s urban forest for two likely reasons. First, species selec-

tion often focuses on small to medium-sized trees, reflecting limited growing space in cities. Se-

cond, biotic and abiotic factors create harsh growing conditions, which can shorten the trees’ life

spans. As the iTree Eco definition of ‘tree’ requires a diameter at breast height of at least 2.54 cm,

no tree up to 1 m was included in this inventory, although individuals in this size class grew in the

sampled areas.

110
 Height Distribution of Sampled Trees
35
30
25
No. of trees

20
15
10
5
0
≤ 2 >2 >3 >4 >5 >6 >7 >8 >9 >10 >11 > 12 >13 >14 > 15 > 16 >17 > 18 >19 >20
and and and and and and and and and and and and and and and and and and and
≤3 ≤4 ≤5 ≤6 ≤7 ≤8 ≤9 ≤10 ≤11 ≤12 ≤13 ≤14 ≤15 ≤16 ≤17 ≤18 ≤19 ≤20 ≤21
Height (m)
Figure 28 Tree height distribution of inventoried trees

The DBH distribution mirrors the height distribution, with almost half of the trees having a di-

ameter10 of up to 10 cm (the average equals 19 cm). To analyse diameter distribution, I excluded

palms and other tree-like species such as papaya, which do not have woody stems, thus remaining

with 147 trees. The observed curve (Fig. 29) is a ‘reverse J’ type, also called a ‘youthful’ popula-

tion “characterised by a high proportion of young, small trees” (Morgenroth et al., 2020, p. 2).

Though researchers have suggested ideal distributions for specific places and/or sub-types of ur-

ban trees (Richards, 1983; Millward and Sabir, 2010), DBH/age correlation varies between spe-

cies and preferred distribution would depend on the defined management objectives (Morgen-

roth et al., 2020). A preference for planting smaller species, comparatively high natural regenera-

tion of these trees and increased planting activities in recent times can result in a large proportion

of trees in smaller diameter classes (Morgenroth et al., 2020). Information to determine how these

factors interplay in Accra is insufficient, but this distribution type requires a short-term focus on

management and maintenance to ensure that small trees successfully establish and mature.

10 In line with iTree Eco guidelines, I measured the largest stem of multi-stemmed individuals

111
 Diameter Distribution of Inventoried Trees
70
60
50
No. of trees

40
30
20
10
0
≤10 >10 >20 >30 >40 >50 >60 >70 >80 > 90 >100 >110
and and and and and and and and and and and
≤20 ≤30 ≤40 ≤50 ≤60 ≤70 ≤80 ≤90 ≤100 ≤110 ≤120
Diameter at breast height (cm)

Figure 29 Diameter distribution of inventoried trees

4.2.2 Intra-City Variation

In Accra, neighbourhoods’ tree covers range from 0.2 % (± 0.2) in Chorkor to 53.6 % (± 2.23) in

Achimota College, the latter including the previously-mentioned Achimota Forest, a roughly 3.6

km2 urban forest. Plantable space is lowest in Ussher Town (10.4 % ± 1.74) and other densely

populated inner-city neighbourhoods like Chorkor, Mamobi, and Nima, which, by the same to-

ken, have significant proportions of non-plantable land (e.g. Ussher Town with 87.0 % ± 1.5 or

Nima with 78.0 % ± 1.85). Free space is higher in neighbourhoods on the peri-urban fringes of

Accra like East Legon Extension (49.8 % ± 2.24) and La Dadekotopon (67.2 % ± 2.1), or in the

less densely populated inner-city neighbourhoods like Dzorwulu (41.4 % ± 2.2) and Roman

Ridge (46.4 % ± 2.23) (Fig. 30, see Appendix 9 for further values).

112
 Figure 30 Neighbourhood locations
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
03 – East Legon Extension 07 – Achimota College 16 – Dzorwulu 23 – Roman Ridge 35 – Mamobi
36 – Nima 49 – La Dadekotopon 73 – Chorkor 76 – Ussher Town

Nero et al. (2018b), who also conducted an iTree Canopy analysis in their six study neighbour-

hoods (Jamestown, Nima, Asylum Down, Kanda, East Legon, and Roman Ridge), report signifi-

cantly higher numbers (Table 16) than mine. Their high standard errors suggest that they classi-

fied fewer points per neighbourhood than I did, potentially even less than the recommended

number. They likely over-estimate tree canopy cover, which exceeds the broader green cover val-

ues determined for Accra (Stow et al., 2013; Owusu, 2018).

113
 Table 16 Comparison between iTree Canopy values for selected neighbourhoods

Neighbourhood Nero et al. 2018b This Study

Jamestown 10.2 % ± 4.15 1.2 % ± 0.49

Nima 14.0 % ± 5.29 4.4 % ± 0.92

Asylum Down 30.0 % ± 10.0 12.6 % ± 1.48

Kanda 37.5 % ± 7.65 21.4 % ± 1.83

East Legon 46.7 % ± 6.44 14.4 % ± 1.57

Roman Ridge 60.9 % ± 5.88 24.8 % ± 1.93

Various elements of urban forest structure, like canopy cover, leaf area or species diversity, show

pronounced intra-city spatial variation that correlates with socio-economic variables in locations

around the world (Pedlowski et al., 2002; Landry and Chakraborty, 2009; McConnachie and

Shackleton, 2010; Kuruneri-Chitepo and Shackleton, 2011; Seburanga et al., 2014; Schwarz et al.,

2015). In Accra, the average canopy cover declines with increasing poverty class (Table 17,

Standard deviation (SD) in brackets). While non-poverty neighbourhoods have an average cover

of 23.1 %, this declines to 3.5 % in the highest poverty class. Examining the relationship between

poverty and canopy classes, lower-poverty neighbourhoods fall into higher canopy cover classes

and those with higher poverty ratings into lower canopy classes significantly more often than ex-

pected by chance (N = 76, p < .01). These results correspond to those by Stow et al. (2013), who

reported higher vegetation cover for neighbourhoods of higher socio-economic status, even

while using different neighbourhood boundaries and methods from those used in the present

study.

Table 17 Average canopy cover per poverty class

Poverty Class Non Low Moderate High Very High

Average Canopy Cover 23.1 % (6.3) 13.8 % (7.0) 12.4 % (8.8) 8.1 % (4.8) 3.5 % (0.9)

114
With increasing canopy cover, plantable space also increases while non-plantable space decreases

(Table 18). This illustrates how canopy cover tends to be higher in less densely populated and

built-up neighbourhoods, structural factors that can often be traced back to early (colonial) town

planning activities.

Table 18 Average plantable and non-plantable space per canopy cover class

Canopy Class 5 (low) 4 3 2 1 (high)

26.9 % 34.6 % 36.8 % 37.8 % 42.6 %

Average plantable space
(SD 7.7) (SD 6.1) (SD 9.3) (SD 6.4) (SD 7.2)

68.5 % 59 % 53.2 % 46.4 % 31.3 %

Average non-plantable space
(SD 10.7) (SD 6.3) (SD 9.6) (SD 8.1) (SD 6.9)

Based on the inventoried trees in the 200 plots, iTree Eco estimates the total number of trees in

each stratum (Table 19).

Table 19 Trees per poverty class

Poverty Class Number of Trees

1 (non) 35,432

2 118,896

3 182,293

4 52,384

5 (very high) 2,567

However, these absolute numbers are misleading as trees/ha values in non- to medium-poverty

neighbourhoods (classes one to three) exceed those in poverty classes four and five (Fig. 31), the

first three above the city’s average of 25 trees/ha, the other two below. This density pattern also

115
exists in Kigali (Rwanda), where richer, more ‘Western-style’ settlements feature more trees (Se-

buranga et al., 2014).

Trees per Hectare in Each Poverty Class

35
30
25
Trees/ha

20
15
10
5
0
1 2 3 4 5
Neighbourhood poverty class

Figure 31 Trees per hectare in each poverty class

Similarly, species richness, i.e. the number of different species, also decreases with increasing

poverty class (Fig. 32), as could also be observed in a Brazilian city (Pedlowski et al., 2002).

Number of Species in Each Poverty Class

20
18
16
14
Number of species

12
10
8
6
4
2
0
1 2 3 4 5
Neighbourhood poverty class

Figure 32 Number of species in each poverty class

116
Though determining a tree’s ownership status may include a certain level of misclassification, it

provides information on who is planting and where. In Accra, I categorised 81.6 % of all trees in

the non-poverty stratum as ‘public’, while this applied to roughly half of the trees in stratum 2

and 3. In the higher poverty strata, however, a majority of trees were classified as ‘private’ (Table

20). This pattern could indicate that ample growing space in sparsely populated neighbourhoods

allows for higher levels of naturally regenerating trees in open spaces. However, it more likely

points to government agencies focusing their planting and maintenance endeavours in these resi-

dential areas. In turn, inhabitants of the poorer neighbourhoods must rely on privately planted

trees and, as discussed above (see Section 4.2.1), their landowners’ interest and/or permission

when renting accommodation.

Table 20 Trees in public and private ownership per poverty class

Poverty Class 1 (non) 2 3 4 5 (very high)

Public trees 81.6 % 45.1 % 42.6 % 8% 27.3 %

Private trees 18.4 % 54.9 % 57.4 % 92 % 72.7 %

Although urban vegetation and socio-economic characteristics do not always correlate (Talen,

1997; Nero, 2016), urban green divides exist in most places. As within-city variations are perva-

sive, they point to structural issues and underlying mechanisms and processes that create and

maintain them. Current canopy cover varies significantly between richer and poorer neighbour-

hoods in Accra, resulting in uneven access to trees and their benefits. These ecosystem goods and

services can help to compensate for problems associated with rapid urban population growth and

climate change, thereby improving people’s well-being. While urban trees do have costs, outputs

can be optimised by selecting appropriate trees for designated planting spaces and implementing

adequate maintenance schemes to enhance their survival and growth. Some trade-offs – in terms

of who can benefit, how and to what extent – are unavoidable due to the complex interplay of

117
services and disservices at various scales. The decision of what to plant where and for what pur-

pose is, in fact, a fundamental conundrum of urban forest management as “planting a finite

number of trees to create an equitable distribution for a city’s inhabitants might, in the end, re-

duce the overall efficiency (defined as the greatest benefit provided for the most people) of the

urban forest” (Heynen, 2003, pp. 989-990).

4.3 Chapter Conclusion

In this chapter, I discussed how some historical events and processes affected Accra’s develop-

ment and, directly or indirectly, its urban tree cover. Exclusively European areas in colonial times

were well planned, with tree planting an integral part of their design. Many African areas, howev-

er, suffered from unplanned growth and increasingly high population densities as people migrat-

ed to Accra in search of income opportunities. My desk-based iTree Canopy assessment and the

ground-based iTree Eco inventory demonstrate how this structure has been preserved over time

and continues to be visible in current day Accra. Here, factors related to historical land owner-

ship, city planning and tree planting patterns have left lasting imprints, moulding the basis for

contemporary urban forestry. History, however, does not just help explain the present but can

and should be used to analyse how today’s urban forest might develop under current political,

economic, and social events and processes. Assessing these potential effects requires not just a

deep understanding of the past but a substantial amount of foresight. While not a precise predic-

tion, the “effects of past management systems can provide an indication of how current ecosys-

tems may react to management efforts” (Szabó, 2010, p. 383). As such, we can utilise an under-

standing of historical human–environment interactions to enhance future outcomes.

118
 5 MANAGING AND USING THE URBAN FOREST – STAKEHOLDER
OPINIONS AND INTERACTIONS

“[F]or pruning we don’t restrict them too much because we even encourage
them to do. Now felling, you need to take permit.”

Mr Sampson, Environmental Health Officer, Ayawaso Central

~~~~~

“Yes, you pay for pruning because we still lose value. You have taken off some
of the branches, and the services that those branches are to provide for the
general good of the public has been taken away by one person. So you must
compensate.”

Mr Fiati, Head of the Natural Resource Department, EPA

In this chapter, I analyse how urban forestry stakeholders in Accra act and interact, demonstrat-

ing how inadequate resources, conflicting priorities and an ill-defined allocation of tasks and re-

sponsibilities impede efficient tree management. I thereby challenge the prevailing assumption

that holds private individuals, especially in lower-income neighbourhoods, primarily responsible

for declining tree cover. After reviewing stakeholder theory, I apply it to urban forestry in Accra

and introduce the interviewed stakeholders – including government actors, NGOs and private

individuals connected to the city’s trees – and their organisational structure. I then summarise

what strengths, weaknesses, opportunities, and threats (SWOTs) the interviewed stakeholders

identified regarding their urban forestry-related activities, experiences and views. These explicitly

mentioned SWOTs form the basis of my thematic analysis, where I categorise and interpret these

statements. I identified two themes, ‘The little that we have, the lot that we have to do’ and ‘The

long road from awareness to action’, the first consisting of two further sub-themes. While the

119
first theme focuses on actors’ activities and priorities within the context of resource availability

and interactions among these stakeholders, the second one traces the steps involved in the pro-

cess from ‘becoming aware’ of a topic to ‘taking action’. Using a hierarchy of effect model, I sug-

gest that (especially) government organisations need to re-evaluate their views on the general

public’s position.

5.1 Establishing Stakeholding in Accra’s Urban Forest

5.1.1 Stakeholders – The Theory
In 1984 Freeman proposed that ‘stakeholders’ include “any group or individual who can affect or

is affected by the achievements of the firm’s objectives” (p. 46). Since then, researchers and prac-

titioners have modified his definition or created new ones. As the chosen definition impacts

study sample criteria and selection (Mitchell et al., 1997), I opted for an inclusive and broader one

to avoid overlooking stakeholders (Bryson, 2004; Lewis and Gillman, 2005), recognising how

more stakeholders might increase (management) complexity. Additionally, I selected a definition

less focused on researching a commercial company, defining stakeholders in this research as “in-

dividuals or groups who have an interest or some aspect of rights or ownership in the project,

and can contribute to, or be impacted by, the outcomes of the project” (Bourne and Walker,

2006, p. 6). Two questions then arise: first, what constitutes ‘the project’, and second, how do we

identify stakeholders and their interests to systematically characterise them and their (potential)

roles in the project? In this section, I briefly discuss these two aspects in general and theoretical

terms before applying them specifically to the case of urban forestry in Accra in the following

one.

The Project Management Institute defines a project as “a temporary endeavor undertaken to cre-

ate a unique product, service, or result” (Project Management Institute, 2013, p. 3). Every project

involves stakeholders whose influence, needs and expectations need to be understood as

120
“[f]ailure to do so can lead to delays, cost increases, unexpected issues, and other negative conse-

quences including project cancellation” (Project Management Institute, 2013, p. 31). Stakeholder

analysis utilises various tools to identify these individuals and groups, commencing either with a

brainstorming approach (Smith, 2000) or a more structured one that examines the political, eco-

nomic, social/cultural or legal contexts (Manchester Metropolitan University, no date). Addition-

ally, one can consult theme-specific lists of potential stakeholders, existing for both forestry

(Mayers, 2005) and urban forestry (Leff, 2016; see below for more detail). Furthermore, identified

stakeholders may identify others while stakeholders can also self-identify in response to public

announcements (Mayers, 2005). Identified stakeholders can then be listed, summarised or classi-

fied in tables, matrices and graphs to understand their interests, impact levels, relative priorities,

types of participation in different project stages and/or their potential influence with regards to

achieving project aims. Based on these factors, project managers can then determine the best

management strategies for each involved stakeholder.

5.1.2 Stakeholders in Accra – The Practice

Applying the above theory and blueprint matrices, graphs and tables to Accra first requires iden-

tifying the specific project under investigation. However, three major state actors, the Environ-

mental Protection Agency, the Department of Parks and Gardens, and the Town and Country

Planning Department, participate in Accra’s urban forest planning and management but seem to

operate lacking a comprehensive plan. My suggestion of what the ‘project’ can be understood as,

from their perspective, therefore, remains vague, focusing on the most basic common goal: plan

and manage the urban forest sustainably to produce a range of benefits. To identify further

stakeholders, I conducted a web and literature search before commencing data collection and also

followed up on individuals and groups mentioned during stakeholder interviews and the house-

hold case study. Additionally, Leff’s (2016, p. 34) list of possible stakeholders and collaborators in

121
urban forest planning and management provided a starting point for further brainstorming and

online search activities (Table 21). Though he developed the list within the context of US-

American urban forestry, several of these stakeholder types both actively participate or are pas-

sively affected in Accra as well.

Table 21 Stakeholder types and examples for each type

Type Examples

Forestry, Natural resources, Housing, Transporta-

Municipal departments and agencies
tion, Public utilities, Fire, Education

Public health agencies, Regional planners, Park dis-

Regional and state agencies
tricts, Forest preserves

Residential homeowners, Homeowner associations,

Landholders
Institutional/Commercial/Industrial

Colleges/Universities with tree or horticultural pro-

Academia grammes, Colleges/Universities where students and
faculty seek research

Tree advocacy groups, Watershed partnerships, En-

NGOs
vironmental justice organisations

General public, Faith groups, Community and

Public individuals/groups
neighbourhood groups, Tree-planting volunteers

Developers, Landscape architects, Arborists and

Private individuals/groups
tree-care companies, Small business associations
According to Leff (2016)

As available resources determined how many stakeholders I could interview, I tried to identify

their interests and assess their importance and influence before travelling to Accra. Influence “in-

dicates a stakeholder’s relative power over and within a project”, and one with high influence

“would control key decisions within the project and have strong ability to facilitate implementa-

tion of project tasks and cause others to take action” (Smith, 2000, p. 5). Importance, however,

122
“indicates the degree to which the project cannot be considered successful if needs, expectations,

and issues are not addressed” (Smith, 2000, p. 5). Besides focusing on whose views and/or ac-

tions seemed most pertinent, I also interviewed groups and individuals who could offer addition-

al insights, thereby including participants from diverse backgrounds. However, I judged this pos-

sible relevancy based on literature and web searches which often offered limited information on

potential stakeholders. The table below (Table 22) depicts my understanding of urban forestry in

Accra prior to data collection, resulting from the available academic literature, newspaper articles

and government departments’ websites. It describes the current situation rather than an ideal

scenario of how influence and importance should be attributed to create (greater) benefits for a

more diverse group of stakeholders.

Table 22 Potential stakeholders, their interests, importance and influence

Potential Stakeholders Stake Influence Importance

Implement national environ-

Environmental Protection Agen- mental policy; manage, protect,
High High
cy and enhance the country’s nat-
ural resources

Department of Parks and Gar- Grow, plant, and maintain trees Low to
High
dens on public lands medium

Develop planning schemes for

Town and Country Planning De- Medium
the city, including areas provi- High
partment to high
sioned for green/trees

Plan and develop road infra-

Department of Urban Roads structure; integration Low Medium
of/conflict with roadside trees

Manage the Achimota urban Medium to

Forestry Commission Medium
forest high

Department of Physical Planning Prepare land use/physical plans Low High

Legal Department ? ? ?

123
Sub-metropolitan offices ? ? ?

National Disaster Management

? Low Low
Organization

Educate people on tree-related Low to

School of Horticulture ?
topics medium

Offer a module for young peo-

Youth Employment Agency Low to
ple to be trained in green- ?
(YEA) medium
ing/planting/agriculture

Developers (e.g. Wonda World,

? ? ?
Clifton Homes)

An NGO that has previously

Youth4GreenGhana Low Medium
planted trees in Accra

Ghana Youth Environmental An NGO that works in the en-

Low Medium
Movement vironmental area

Communication agency that

Stratcomm Africa, Ghana Gar- started the Ghana Garden and
Low Low
den and Flower Movement Flower Movement with an an-
nual Garden and Flower Show

University of Ghana, Legon Bo- Research projects on urban

Low Low
tanical Gardens trees; children’s playground

NGO that created a children’s

Mmofra Foundation Low Low
park in Accra

Ghana Climate Innovation Cen- Business incubator focusing on

? ?
tre the ‘green economy’

Environmentally focused
Centre for Environmental Re- NGO that promotes sustaina-
? ?
search and Policy Analysis bility through research and ed-
ucation

Medium to
General public Low
high

124
 Tree/Tree-part salespeople (e.g. Depend on trees and tree parts
Low Low
seeds, fruits, medicine) for their livelihoods

Faith groups (e.g. in Achimota Meet and pray in prayer camps Low to me-
Low
Forest) in the forest dium

Own part of the city’s land

Homeowners/Landowners with trees; grow, plant, and Low High
maintain trees

I interviewed representatives of the major government departments and agencies working in ur-

ban forestry, including the Environmental Protection Agency, the Department of Parks and Gar-

dens and the Town and Country Planning Department. I could not interview the Forestry Com-

mission about Achimota Forest due to a ban on discussing its transformation into the Accra Eco-

Park (see Chapter 7). These four government departments plan, manage, and maintain the city’s

urban forest but operate under different ministries. While they divide some forestry-related activi-

ties between them (see Section 5.2.1), other tasks are potentially duplicated as both EPA and

DPG plan to or currently develop educational campaigns to raise the public’s awareness of urban

tree benefits. Additionally, I interviewed a representative of the Forestry Research Institute of

Ghana (FORIG) and three sub-metropolitan assemblies’ environmental officers (Ayawaso East,

Central, and West). I also talked to two non-governmental organisations, Youth4GreenGhana

and Ghana Youth Environmental Movement and interviewed four Accra inhabitants who de-

pend to varying degrees on the urban forest’s natural resources for their livelihoods or activities

(a coconut seller, a tree seedling seller, a herbalist, and a pastor). To elicit a picture of the general

public’s views, I further conducted a household questionnaire in five neighbourhoods.

125
5.2 Urban Forestry in Accra
5.2.1 Organisational Structure and Functioning
The Town and Country Planning Department was established in 1945 and plans the city’s growth

and development. The employees design and prepare planning schemes demarcating areas for use

as open and green spaces. The Department of Parks and Gardens then ensures that these green

areas are established (through tree planting) and maintained. Under the country’s decentralisation

policy, these two departments merged their activities at the district level, forming the Department

of Physical Planning. Previous attempts at such integration have, however, been unsuccessful and

remain limited up to today. As of July 2017, 177 people worked in DPG’s Accra location, about

100 less than three years before due to a government freeze on employment. Apart from about

ten office and managerial staff, everyone participates in growing, planting, and maintaining the

trees on the department’s grounds and across the city. They water planted trees, occasionally fer-

tilise them and prune or fell them, but their small budget and lack of necessary equipment limits

these activities. The interviewee, Mr Awidi, stated a departmental interest in forming public-

private partnerships (PPP), a national policy where the private sector partner could support them

with funding or technology. Furthermore, DPG also supports individuals, institutions, and other

groups with tree planting and maintenance, advising on suitable species and recommended tree

care.

The third major actor, the Environmental Protection Agency (est. 1994), emerged from the Envi-

ronmental Protection Council (est. 1974). Like TCPD, the agency operates under the Ministry of

Environment, Science, Technology and Innovations (MESTI) and oversees the implementation

of the country’s environmental policy. EPA ensures that environmentally responsible behaviour

and resource use are mainstreamed into Ghana’s development at every level, from community to

nation. The interviewee, Mr Fiati, the head of EPA’s Natural Resource Department, reported

that the agency seeks solutions to natural resource management problems which, regarding Ac-

126
cra’s trees, means regulating removal and pruning through a permit system. The Forestry Com-

mission’s work focuses on trees outside urban areas but, due to the historical reasons discussed in

the previous chapter, includes Accra’s Achimota Forest.

Besides these government bodies, several further actors participate in Accra’s urban forestry ac-

tivities. The Accra Metropolitan Assembly (AMA) constitutes one of the 260 Metropolitan, Mu-

nicipal and District Assemblies across the country. In 2017, Mohammed Adjei Sowah became the

new mayor, the Metro Chief Executive, to whom the city’s 16 Statutory Departments report.

AMA is further divided into sub-metropolitan assemblies, the boundaries and numbers of which

have changed several times in the past. I conducted the household survey and case study compo-

nents in five neighbourhoods spread across Ayawaso East (Nima, Mamobi), Ayawaso Central

(Kokomlemle, Alajo), and Ayawaso West (Dzorwulu) sub-metros. In each of these assemblies, I

interviewed an environmental health officer about their tree-related work, which includes pro-

moting and encouraging tree planting, popularising the trees’ benefits and minimising their dis-

services by advocating pruning and clearing fallen leaves. Though located near Kumasi, I also in-

terviewed a researcher of the Forestry Research Institute of Ghana (FORIG), an institute operat-

ing under the Council for Scientific and Industrial Research. While FORIG lacks a division de-

voted explicitly to urban forestry, researchers from various divisions (e.g. Forest and Climate

Change; Governance and Livelihoods; Biodiversity Conservation and Ecosystem Services) work

on this topic, although they have, by their own admission, focused on Kumasi’s trees.

5.2.2 Further Stakeholders and Interviewees

An internet search prior to data collection revealed two NGOs which had worked in urban for-

estry in Accra. The youth-based NGO Youth4GreenGhana (Y4GG) was established in 2011 and

seeks to give back to society and Mother Earth. Focusing on environmental conservation activi-

127
ties, they planted coconut palms along the beach and 100 seedlings in the Dansoman neighbour-

hood in 2013 (Fig. 33). Youth4GreenGhana reported that they had received these seedlings for

free from the Department of Parks and Gardens, but they had forgotten the trees’ species. De-

spite losing all seedlings due to lack of maintenance and browsing animals, they want to continue

their tree planting project in the Dansoman area. They further pursue other related activities and

aim to open an organic food shop.

Figure 33 Location of Dansoman

Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
38 – Dansoman

The second NGO, Ghana Youth Environmental Movement (GYEM), started in 2012 as an envi-

ronmental advocacy and campaign group that aims to solve Ghana’s sanitation and climate

128
change problems. They train and educate young people in school clubs, through workshops, con-

ferences, and a book written by the founder. The NGO’s steering group decides on campaign de-

sign and strategies and then forwards these to members on the ground and an estimated 5,000

online supporters. They ran two campaigns on trees, one in Aburi Garden, where their work

stopped tree felling in a reserve. Their other tree-related campaign aimed to protect Accra’s only

greenbelt, the Achimota Forest, from being transformed into the Eco-Park. However, the gov-

ernment did not react to their petition on change.org and parts of the general public also seemed

to support the government’s plans. As progress regarding Achimota’s redevelopment is slow –

the interviewee, Mr Commey, even stating that he doubted it would ever get started – GYEM

was in a waiting position as of August 2018.

Further interviewees included a herbalist in Nima, who has obtained part of his raw materials

from public and private city trees for about five years. However, the majority (about 95 %) of his

supplies originate from rural areas. He estimates that he buys about 70 % of his urban supplies

and collects the remaining 30 % for free, calling the easier accessibility of urban sources ‘conven-

ient’. For his medicines, he uses neem tree leaves and noni tree fruits (Morinda citrifolia L.) from

both rural and urban areas. He disclosed collecting some of these fruits and leaves from Pokuase

and Amasaman, towns adjacent to Accra and part of the Greater Accra Region, but located out-

side the study city’s boundaries to the northwest (Fig. 34). The herbalist appeared somewhat hesi-

tant to disclose (further) urban collection and purchase locations as he mentioned competing for

these resources with his colleagues. He further stated that he believes the medicine from rural

products to be of better quality since garbage dumped on urban trees can ‘contaminate’ them.

Additionally, he mentioned that trees from very public and busy areas like Nima could be ‘spirit-

ually contaminated’ by people with ‘evil eyes’ or used by malams, spiritual Muslims for their pur-

poses, making them ineffective for the herbalist’s medicines.

129
 Figure 34 Location of Amasaman and Pokuase
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
Accra highlighted in grey for reference

In addition to the herbalist, I also interviewed a pastor who visits Achimota Forest to meet and

pray with some of his church members, and a roadside tree seedling seller. The pastor has been

going to the forest with fellow churchgoers for about ten years, leading his own congregation for

five. His group constitutes one of the many groups that regularly meet in the forest and for which

FC has cleared undergrowth, thereby creating prayer camps sheltered by trees. He visits Achimo-

ta on a fixed day each week, as well as at other unscheduled times, and considers Achimota an

environment conducive to prayer and spirituality. His church assembly is not involved in any

form of forest protection or tree maintenance, as these tasks reside with FC’s forest guards. The

tree seedling seller, Anasawudu, has been producing and selling fruit trees like mango (Mangifera

indica L.) and avocado (Persea americana Mill.), fruit/ornamental species like Indian almond (Termi-

nalia catappa L.) and ornamental palms, such as Roystonea regia for some 15 years. He grows his

seedlings from seeds recovered from fruits that he eats and obtains additional seeds from DPG

130
when demand is high. He tends his plants using poultry manure and wastewater harvested from

the gutter that runs through his roadside shop. AMA assigned this place to him, and he paid

them 200 cedis for a permit to sell his plants.

The interviewed coconut seller, Mr Kwakye, has been engaged in his trade for six to seven years,

selling on a bridge in the Nima/Mamobi area. While typically buying his coconuts from sur-

rounding villages, he does obtain some coconuts from urban palms when rural supply is low in

June/July or when he sells more than expected and needs to re-stock urgently. He has bought

twice (210 coconuts total) from one of my case study households, something he would repeat

when necessary. While urban palms allow him to obtain some coconuts quickly, his regular stock

exceeds the amount available from nearby city trees. I further base my analysis on information

collected during the household survey in Nima, Mamobi, Alajo, Kokomlemle and Dzorwulu.

While methodically different from the stakeholder interviews, these views of the general public

supplement both the government agencies’ ‘institutional perspective’ and that of NGOs, herbal-

ist, pastor, tree and coconut seller. Successful urban forest planning and management require the

public’s involvement, but households’ views may vary depending on size, economic status or lo-

cation. Additionally, this stakeholder group includes both people who own trees and those that

do not, while some even sell tree products. In reality, therefore, several sub-groups with different

views and priorities can be identified in this group.

5.3 Stakeholder views – Strengths, Weaknesses, Opportunities,

Threats
5.3.1 Strengths, Weaknesses, Opportunities, Threats – The Theory
Strengths, weaknesses, opportunities, threats analysis (SWOT) is a planning tool that helps to an-

alyse positive and negative internal and external factors to understand the context in which a

company or organisation operates. Strengths and weaknesses relate to the organisation’s internal

131
functioning and include human, financial and physical resources, as well as past experiences. Op-

portunities and threats, in turn, refer to the external environment and include the economy, legis-

lation or demographics. As such, the analysis of the external situation resembles a ‘PESTLE anal-

ysis’, or one of its derivatives, which considers what macro-environmental factors a company

works in, analysing its political, economic, social, technological, legal, and environmental context.

SWOT analysis is a tool often used in business management but also applies to analysing indus-

tries, countries, or not-for-profit organisations (Helms and Nixon, 2010). While recommended

for “its simplicity and value in focusing attention on key issues” (Pickton and Wright, 1998, p.

102), some researchers and practitioners criticise the tool. Users may misclassify threats as oppor-

tunities (Morris, 2005) or misuse generic checklists without considering context (Valentin, 2001).

Additionally, critics have called SWOT analysis an atheoretic classification system (Grant, 2008).

Morris (2005) suggested identifying ‘options’ rather than ‘opportunities’ as identifying the latter

should be the analysis’ final objective. Others have argued that SWOT items lack hierarchy which

hinders prioritisation (Helms and Nixon, 2010) and leads to the production of simple or incom-

plete lists with no further decision-making value (Kangas et al., 2003); or to long and descriptive,

irrelevant lists with general and ambiguous terms that detract from what is crucial (Hill and

Westbrook, 1997). Further criticisms of SWOT analysis include difficulties in distinguishing in-

ternal from external factors (Helms and Nixon, 2010), too much emphasis placed on exploring

the positive aspects while failing to address the negative ones related to incompetence and weak-

nesses (Drago and Folker, 1999), and its questionable usefulness in a fast-paced information and

technology age different from when the tool was first developed and utilised (Hill and West-

brook, 1997).

Despite the tool’s potential shortfalls, I use it as a first step to obtain an overview of what various

stakeholders think, capitalising on its ability to simplify a complex situation. As users conduct a

132
SWOT analysis at a certain point in time, it needs to be repeated periodically to monitor evolving

circumstances (Helms and Nixon, 2010). Here, the approach serves as a starting point, providing

baseline knowledge against which future developments can be compared. To off-set some of the

problems and improve the utility and practical value of SWOT as a tool, researchers suggested

combining SWOT with other approaches (Lee and Ko, 2000; Valentin, 2001; Hai and Tsou,

2009) or using it as “a precursor to strategy formulation, managerial decision making and action”

(Pickton and Wright, 1998, p. 102). I, therefore, utilise the strengths, weaknesses, opportunities

and threats explicitly identified by the interviewees as a precursor for my analysis of urban forest-

ry in Accra. Based on my findings, I suggest strategies and potential actions for improvement in

the concluding chapter.

5.3.2 Strengths, Weaknesses, Opportunities, Threats – The Practice

When analysing interview transcripts for SWOTs, two types of SWOTs need to be identified.

First, an interviewee may explicitly state a SWOT, reflecting how the person interprets a given

situation, in this case, their urban forestry-related activities and the context in which they operate.

The second type results from how I categorise and interpret these responses. My assessment uti-

lised a guide “designed primarily to help municipalities assess the state of their urban forest, iden-

tify management concerns, and chart a path forward, step by step, toward long-term sustainabil-

ity” (Leff, 2016, p. 4). In this document, Leff proposes targets/criteria with key objectives and re-

lated performance indicators to assess a city’s forest planning and management activities in three

categories. These include the vegetation resource itself, stakeholder engagement and the resource

management approach like plans, practices and policies. As my analysis combines interview re-

sponses and the sustainable urban forest guide, my interpretations might or might not agree with

those of the interviewees. I first summarise some SWOTs that have been explicitly stated by in-

terviewees (first type) to offer examples that cover positive and negative aspects, including disa-

133
greements and opposing views (Table 23). Though I took these statements from the interviews,

how I assigned them to one of the categories involved a level of subjectivity as previously dis-

cussed (see Section 5.3.1).

Table 23 SWOTs identified by the stakeholder interviewees

S W O T

Urban forestry
may become more
They have been
All divisions of important in the No national policy
unsuccessful in
FORIG are inter- future: different on urban forestry
FORIG getting funding for
ested in Urban people/groups are is part of urban
urban forestry re-
Forestry. starting to become land planning.
search projects.
interested in the
topic.

Sometimes people
put benefits higher
than potential dis-
He thinks he is
services (e.g. do
well informed People know they
not want to prune
about tree-related need felling per-
Ayawaso to maximise me-
things and is told mits and act ac-
Central dicinal product
everything he cordingly: no ille-
yield, but over-
needs to know to gal felling.
grown trees attract
do his work.
potentially dan-
gerous animals like
snakes).

Their work has

The majority of
had an effect: they
They lack re- A lot of peo- trees are replaced
convinced people
Ayawaso sources – e.g. no ple/groups do by buildings (“de-
who wanted to fell
East car, so they need something about velopment”) –
a tree, not to do so
to walk to Nima. the environment. numbers are re-
because of the
ducing.
benefits of trees.

134
 Ayawaso West in-
habitants are
Working relation- knowledgeable
ship with EPA about trees, aware Some parts of the
Ayawaso and DPG general- of tree-related laws sub-metro are los-
West ly good. He thinks and regulations or ing trees to con-
they are well- open to learning. struction.
informed. People follow ad-
vice on tree man-
agement.

Difficult to get
They are deter-
people’s attention People showed in- There are no simi-
mined to continue
(little visibility/ terest when they lar groups that
Y4GG their work, plant
presence): website were planting they could work
more trees, pursue
down, no money trees. with.
their dreams.
for offices.

Through existing
PPPs take a long
partnerships (Gar- Number of em- Decentralisation is
time to set up,
den and Flower ployees reducing, beneficial: em-
some departments
Show, Embassies, many are too old ployees located at
DPG do their work
YEA), they get (government assemblies where
without including
workers, tools, and freeze on recruit- they can focus on
DPG or including
education oppor- ment). greening the area.
them too late.
tunities.

Space and shelter

Backed by laws in economically
(e.g. severe fines depressed com-
for illegal felling; munities is a prior-
EPA plant two trees ity – little space
when felling one; for trees (over-
required number crowded) and
of trees per plot). people focused on
other things.

135
 They are known
enough to have
e.g. Gesellschaft
People tend to be
He recruits and für Internationale
busy in Accra due
trains many peo- Zusammenarbeit,
GYEM to the urban life-
ple, but few then British Council, or
style – no time to
take action. WaterAid ap-
join campaigns.
proach them and
support/finance
specific projects.

The new mayor of

No real coopera- Accra is interested
tion with DPG, in trees and wants Implementation is
everybody does to develop all difficult because
TCPD
their own thing as open spaces – of lack of land
DPG is not re- tasked TCPD to ownership.
sourced at all. identify them and
acquire them

The quantity he
has been taking
(for five years) is
reducing – thinks
Being able to get because of de-
some products mand from other
from the city is herbalists.
convenient as he Some trees are not
Herbalist
does not have to suitable as sources
travel far to the of medicinal
villages, reducing products because
acquisition costs. they grow in un-
hygienic condi-
tions (illegal gar-
bage dumps or
urinals).

136
 He did not resist
price changes of
Lack of direct of-
entrance fees or
ficial communica-
the proposed con- Achimota offers a
tion: not informed
version to the peaceful atmos-
Achimota when entrance fee
Eco-Park, because phere to pray –
visitor/ changed, not in-
he, and others like fresh air, less pres-
pastor formed about
him, are not in a sure, less crowd-
Eco-Park plans
position to change ing.
(just signs/radio).
anything/have the
power to influence
such decisions.

He has a working
maintenance
Tree seller scheme, and less
than 5 % of plant-
ed seeds die.

When coconut
supply is short in
Most places in the
the villages, or
Getting coconuts city are too small
there is a great
Coconut from the city can to have a large
demand, the nuts
seller be convenient if enough number of
from the city can
close by. coconut palms in
help during these
one place.
temporary short-
falls.

The 100 survey participants’ responses complement these perspectives on urban trees and forest-

ry in Accra. Fifty-seven respondents mentioned public trees like neem, coconut, mango, moringa

and pine that had caught their attention. They mentioned twelve reasons for noticing these trees,

four of them negative (unpleasant scent, causing accidents, fallen leaves, habitat for undesirable

137
animals11). The most common positive reasons included food, medicine, shade, and beautification

and apart from four people, all respondents argued in favour of planting more trees. A third of

the participants considered Accra’s public trees to be in ‘a bad state’, mainly because they were

not pruned or otherwise maintained. Only one respondent thought that the city authorities do

not allow unregulated tree cutting and ensured regular pruning. Forty-nine respondents did not

think that the city authorities took good care of public trees (35 ‘yes’, 14 ‘don’t know’, 2 non-

responses), identifying poor maintenance and cutting without replacement planting as a signifi-

cant problem. In a position of influence, several people would enact tree protection laws and en-

force them. Out of the 64 tree-less households, 42 said they would like to own trees, primarily for

shade, fruits, and beautification, the same reasons for which questionnaire respondents would al-

so want more trees to be planted on public land.

5.4 From SWOT to Themes

Comparing, contrasting, and relating these stakeholder views allows identifying ‘themes’ or “pat-

terns of shared meaning underpinned or united by a core concept” (Braun and Clarke, 2019, p. 593,

their emphasis). Thus examining the above interview responses, I identified two main themes,

one with two sub-themes, all of which are “produced at the intersection of the researcher’s theo-

retical assumptions, their analytic resources and skill, and the data themselves” (Braun and

Clarke, 2019, p. 594). Taken together, they offer creative and interpretive stories about the da-

taset in question. Unlike the SWOT table above, the following analysis is based on how I inter-

preted the available data (the second type of SWOTs), which I distinguish from the strengths,

weaknesses, opportunities and threats directly and explicitly stated by the interviewees.

11As these were responses in a questionnaire, no follow-up questions could be asked as to the types of animals or
the source of unpleasant scent.

138
5.4.1 Theme 1 – The Little That We Have, the Lot That We Have to Do
5.4.1.1 The Little That We Have…
Interviewees commonly discussed resources – or their lack or absence – identifying their availa-

bility as a strength and, in turn, their lack as a weakness. Resources mentioned included monetary

funds, personnel (both its quantity and quality), as well as equipment and material/supplies. In

most cases, participants explicitly identified these resources as absent or unavailable in the re-

quired amounts or conditions. The DPG interviewee, for instance, stated that “staff are ageing

and there is urgent need to replace them or to inject young people who have physical strength”

(Mr Awidi, DPG). In addition to these direct statements, respondents provided information that

I interpreted as either positive or negative, despite not specifically mentioned as such by the in-

terviewee. This includes text passages like the following:

Researcher: How many members do you have at the moment?

Respondent 1: About twenty.

Researcher: Twenty, and how many did you start with?

Respondent 1: Erm, we were fortunate to have most of the guys turn up

day one.

(Interview Y4GG)

While the respondent considered it positive to have started with 20 members/volunteers, my in-

terpretation differed. I considered a constant number of volunteers and the associated inability to

mobilise more people, a weakness as they explicitly stated requiring further individuals to support

their work and future tree planting activities. Besides the more apparent resources like money and

workforce, I identified other inputs which determine urban forestry dynamics in Accra. These re-

sources include connections to other stakeholders, time, or visibility/presence, among others. Be-

low, I provide interview text passages to offer examples and to illustrate my coding approach

139
(Table 24). The chosen text excerpts reflect how crucial resources are, especially for NGOs, and

illustrate how each stakeholder works with a particular set of inputs. Through the process of re-

source management within each organisation or by each individual, these inputs lead to certain

outputs. Here, outputs refer to the urban forestry-related plans and actions that each stakeholder

designs or implements to achieve their desired overall objectives. The interviewees, both gov-

ernment and non-government bodies alike, shared the sentiment that resources were generally in-

adequate to accomplish their respective goals.

Table 24 Interview text examples for different types of resources

Resource Type Text Examples from Interviews

“so we made it a serious issue, we ran a campaign, we took a lot of pho-

tos, we gave it to the media, the media ran a lot of stories, and I think it
Connections
ceased”

(Mr Commey, GYEM)

“For instance, the public-private partnership policy takes a long process.”

Time
(Mr Awidi, DPG)

“we had a lot of people who disagreed with us because they thought the
Trustworthiness/ government was going to do something good because the government’s
credibility/ idea looks good on paper and our idea sounded, sounded very radically
[radical], okay?”
legitimacy
(Mr Commey, GYEM)

“So we operate from home as of now. But we believe that if we should

Visibility/ have an office in town that can also help a lot because like operating from
home, a lot of people have not even heard about us.”
presence
(Interview Y4GG)

“But, you know, once we have given ourselves to this, we will keep
Perseverance/ fighting till one day our dreams become a reality, yeah, so that is it, we will
not give up.”
vision
(Interview Y4GG)

140
In many places around the world, a lack of available inputs, especially finance and adequate staff-

ing, threaten sustainable urban forest planning and management. Insufficient budgets hinder ur-

ban forestry in European countries (Konijnendijk, 2003) and in Portland (USA)(Driscoll et al.,

2015). In Mississippi (Grado et al., 2006) and Pennsylvania (Elmendorf et al., 2003), a lack of fi-

nancial resources and staffing prevent successful urban forestry. Researchers identified similar

problems in New Zealand (Stobbart and Johnston, 2012) and in South Africa, where, in addition

to money and staff, equipment was also lacking (Gwedla and Shackleton, 2015). In Kumasi, in-

sufficient funding also threatens the sustainability of urban green spaces (Mensah et al., 2017).

Not only do widely dispersed cities share these problems, but researchers have commented on

them for a long time (Johnson, 1982). Increasing budgets for the necessary activities, workforce,

and equipment could solve these problems. However, as these issues seem common through

time and across space, it might be useful to consider if this discrepancy between actual and re-

quired inputs constitutes a symptom of another underlying phenomenon.

I suggest that the resource issue has two components: availability and accessibility. While availa-

bility determines the general ‘pot’ from which actors may draw, interviewees from government

bodies indicated how accessibility is a more determining factor. The Department of Parks and

Gardens is, according to its own view and that of other stakeholders, extremely underfunded. For

each financial period, DPG forwards its budget estimate to its parent ministry, which then de-

cides on the actual monetary allocation. Thus, while the ministry has a certain budget available,

DPG cannot necessarily access enough of this to meet its needs in full. The DPG interviewee de-

scribes the reason as follows:

Respondent: So if you take the budget, you look at the sectors and

their allocations, and we happen to fall under a Ministry

that has to deal with district assemblies. Their major

141
 challenge is how to deal with sanitation. Okay, so chunk

of their budget […] is trying to put sanitation measures

into place.

(Mr Awidi, DPG)

While there is, theoretically, more money available, the party that decides on how to allocate re-

sources has an order of priorities which puts sanitation measures above trees and green spaces in

Accra. The reason for this is also explained by the interviewee:

Respondent: Because […] what are the immediate returns they are going

to get? If somebody picks a container full of rubbish,

you can see it. Okay, or if there should be a public

place of say convenience built for the people within a

certain community, that one, the benefits are there and

immediately you can see. But if you say you want to spend

so much on planting trees which nobody is really going to

benefit from it now, (I mean) the effect, long-term is

what we are looking at.

(Mr Awidi, DPG)

Tree planting and care do not offer immediately visible benefits, lessening their attractiveness for

decision-makers who prefer concrete evidence of their work. Influential individuals and groups

often determine the accessibility of tangible resources like equipment or money. Additionally,

they also design the processes and related legal requirements which others need to follow to ob-

tain these resources. One NGO stated how they had to go through a very time-consuming series

of actions to obtain a tree planting permit, unable to speed up the process as they could not in-

142
fluence the pace of government decision-making. Thus, as a result of their dependence, they had

to invest an increased amount of the resource ‘time’ necessary for the successful pursuit of the

organisation’s urban forest/try goals.

Establishing vertical ties with those who can, directly and indirectly, affect resource allocation

may be crucial to increasing one’s efficiency. Interviewees in Accra identified political support or,

more generally, the backing of powerful individuals and groups, as indispensable to successful

urban forestry work, a factor also recognised in other studies (Elmendorf et al., 2003; Driscoll et

al., 2015). However, those in dependent positions in such relationships also often fear aligning

themselves too closely with the personnel, policy and other intervention priorities of a governing

party to avoid being perceived as biased in the event of a change of government. The NGO

Y4GG specifically stated that they wanted to stay ‘politically neutral’:

Respondent: So we don't want to even touch, you know, it [politics]

in some way. We want to be neutral as much as possible

because it was not in the era of this government that

this NGO was set up. It was in the previous government.

So we don't want some political [something political/some

politician] to block our way. So anything that is politi-

cal related, we actually not touch it or talk about it.

(Interview Y4GG)

For similar reasons, FC declined my request for an interview to discuss Achimota Forest/Eco-

Park, and other respondents, such as the individual from the government’s research organisation

FORIG, were reluctant to comment on specific questions which they felt were too political:

143
Researcher: So there is this Kumasi Urban Forestry Project. Do you

know anything about this?

Respondent: I don't wanna talk too much because there is a lot of

politics.

[…]

Respondent: There is a lot of politics in it, and I wouldn't want to

commit to any political issues, but yes, there is an ur-

ban forestry project in Kumasi.

(Researcher, FORIG)

In Accra, some interviewees tried to overcome the potential drawbacks of existing vertical ties by

establishing (more) horizontal substitutes to promote interactions of greater interdependence ra-

ther than debilitating dependence. Both DPG and TCPD commented negatively on the public

sector freeze on employment, a requirement for IMF assistance. Despite a large pool of well-

trained and employable individuals, they could not hire the needed personnel. DPG tries to com-

pensate for the lack of personnel, funding and equipment by establishing links with embassies,

collaborating with the Youth Employment Agency (YEA), and trying to initiate public-private

partnerships. While some of these interactions are less hierarchical than budget allocations be-

tween departments and their respective ministry, and ultimately the government, these relation-

ships are still a response to, or requirement of, the political and policy context. They allow access

to alternative sources of funding, personnel or expertise and can potentially increase a depart-

ment’s capability and efficiency. However, there remains the need for caution, given how even

the seemingly more horizontal of such ties can be affected by unequal power relations, as the re-

spondent from DPG reminds us:

144
Respondent: For instance, the public-private partnership policy takes

a long process. So first, you need to identify a particu-

lar project, then when the project is identified, you

have to register the project with the Ministry of Fi-

nance, the unit that is responsible for that […]. You

have a unit that also analyses the detailed report that

you have sent and see whether what you are asking for is

feasible. If it's not feasible, then they will look for

alternatives.

(Mr Awidi, DPG)

Stakeholders who directly use tree products in their work, such as the herbalist or coconut seller,

seemed unconcerned regarding the ongoing availability of and accessibility to these items. They

would simply replace the tree products currently acquired within the city with those from rural

areas where plentiful supplies remain, and the bulk of their products already originate from.

Thus, while they consider sourcing some inputs from the city convenient, saving travel time and

money, these supplies constitute an occasional option to be capitalised on when possible. While

‘The little that we have’ is most evident in the presence or absence of physical, human, or finan-

cial assets, it is ultimately only the visible manifestation of the underlying structures and processes

which mediate access to such assets in the first place.

5.4.1.2 …the Lot That We Have to Do

The theme’s second part has two further components. The first concerns stakeholder specific, in-

ternal views (‘What matters?’), while the second (‘134 shades of stakeholder interactions’) illus-

trates what happens where different parties interact or, indeed, do not interact. The unequally dis-

tributed power prevents a situation where each stakeholder has access to the required money,

145
personnel, and equipment to meet their objectives. However, this resource scarcity is often artifi-

cially created (e.g. by budget allocations), favouring the work of those who can affect resource

distribution and may do so for their own benefit or in the pursuit of their specific priorities. As

priorities concerning urban forest management differ between stakeholders, there is no agree-

ment on what actually should be achieved with the available resources. Thus the views and objec-

tives of comparatively resource-richer groups/individuals and those in decision-making positions

are likely to dominate. This lack of between-stakeholder agreement on urban forestry aims cre-

ates a situation in which ‘the lot that we have to do’ is not a single goal pursued by all. Instead, a

range of priorities and objectives exist, with those backed by more resources and ultimately more

powerful individuals and groups, more likely to actually be achieved. The following two sub-

themes consider these issues in more detail.

5.4.1.2.1 Sub-Theme 1 – What Matters? On Views, Objectives, and Priorities

As part of each interview, I asked the respondents to provide a summary of their tree-related

tasks and objectives. The answers show a general agreement on the premise that it is desirable to

plant, maintain, and protect trees in Accra and that the many benefits provided by the urban for-

est justify its establishment and management:

Respondent: As a youth-based non-profit organisation, we are into

giving back to Mother Earth. You know, we believe that we

have received so much, and it is time for us to give

back. […] So our mission actually is to become an indomi-

table environmental conservative force in Africa and the

world at large. […] They are cutting our trees down,

things are going so bad in this society, when there is

rain there is flood here and there. So which organisation

146
 is there on the ground that can help us to prevent or

stop all these things? So that is what we stand for.

(Interview Y4GG)

Respondent: Our task is to lead in the conservation of biological di-

versity and to support efforts to improve human habita-

tion. So, conserving biological diversity, you have urban

biodiversity, and one of the prominent things you see is

the trees. Yeah, the trees provide significant part of

the habitat for urban biodiversity. So apart from provid-

ing habitat for biodiversity, it has other direct and in-

direct benefits for human settlement. So, what we do is

to try to manage the urban tree cover within Accra.

(Mr Fiati, EPA)

Respondent: [Y]ou know, trees have environmental purpose, so they are

very good to have. We encourage tree planting.

(Mr Sampson, Ayawaso Central)

While they superficially agree on the value of urban trees, a more detailed analysis of implicit and

explicit stakeholder views suggest differing opinions as to the trees’ role and how they should be

planned, used and managed. The following excerpts illustrate some of these conflicting opinions

between different stakeholders as even when they agree about the overarching objective, there

can be considerable disagreement over how to achieve the objective(s) in question:

Respondent: And the second one had to do with Achimota Forest. We

heard that the government of Ghana wanted to do an Eco-

Park. We didn't understand what that was, but we need the

147
 Achimota Forest, it was the only greenbelt in Accra, and

we wanted the integrity of that to be preserved.

(Mr Commey, GYEM)

Respondent: Yeah, they [Forestry Commission] are saying about it, but

what is Eco-Park? Eco-Park for what?

(Mr Yeboah, TCPD)

These two respondents disagree with FC’s plans to turn Achimota Forest into the Accra Eco-

Park. The Forestry Commission, in turn, considers the development necessary to protect the re-

maining forest from further gradual but steady encroachment. While all parties aim to preserve

the forest area, their views on how best to achieve this diverge. GYEM, TCPD and FC, however,

are not the only involved stakeholders, but others feel ill-informed, like the interviewed pastor.

He stated that FC did not inform his or any other prayer group, let alone consult them, about the

proposed development of which they became aware when a billboard was put up on the site and

information broadcast over the radio. He expects FC to plant more trees, which he supports, but

also worries that the Commission will deny religious groups access to the new Eco-Park. Thus,

instead of considering urban forestry to be a city-level and city-wide endeavour, it consists of

many urban forest ‘producers’, aiming to create a structure that best meets their needs and re-

quirements. This situation is summarised by Mincey et al. (2013, p. 561) as the “multifaceted ma-

trix of property rights and management strategies subject to a myriad of actors and their associat-

ed governance regimes”.

Arguing that stakeholder priorities differ may seem obvious yet important for two reasons. First,

stakeholders at the national to international scale seem to agree on the importance of global sus-

tainability. For this objective, (city) green constitutes a vital component for its ability to mitigate

148
some of the adverse effects of climate change – such as reducing air pollution, storing carbon,

decreasing water runoff or reducing ambient temperatures – thereby helping to produce more

‘resilient’ cities (see Section 7.2.4). Such a scalar focus, however, renders invisible the many other

benefits which urban trees produce at smaller scales, primarily concerned with household

(re)production and individuals’ livelihood strategies. Second, the concept of ‘sustainability’, re-

gardless of scale, is broad, including not just an environmental component but also an economic

and social one. The term on its own is, therefore, too unspecific to deduce concrete urban forest

planning and management actions, with priorities and paths to reach related goals being actor-

specific rather than generalisable. The conflict of scales and priorities then creates a complex sys-

tem involving numerous actors/stakeholders who follow potentially conflicting strategies both at

different stages of the planning and management process and for various land ownership types.

Taken together, there are myriad spaces of interaction or possible interaction where these stake-

holders disagree, can or should collaborate, or monopolise these spaces.

5.4.1.2.2 Sub-Theme 2 – 134 Shades of Stakeholder Interactions

Trees in cities are purposefully planted and managed for several reasons. To achieve these objec-

tives, “there needs to be an effective and integrated working relationship across public, private,

voluntary and community sectors” (Jones et al., 2005 p. 187). But how can this relationship look

like if producers prioritise different uses and operate at various scales while the urban forest

transcends the boundary between the public and private domains? In addition, insufficient re-

sources, limited political support, and a lack of interagency or between-stakeholder interaction all

explain limited urban forestry success in Ghana and elsewhere. Indeed, better communication

and increased cooperation are desirable for and indispensable to positive outcomes (Stobbart and

Johnston, 2012; Mensah et al., 2017). Coordination and collaboration, “the collective actions of

individuals” (Mincey et al., 2013, p. 555), ensure an urban forest’s robustness and constitute core

149
elements of approaches and frameworks that have been applied to urban forest management

(Lawrence et al., 2013; Mincey et al., 2013; Leff, 2016). These relationships happen within a com-

plex and diversified context of varied stakeholders and resources, contrasting views and priorities,

and policies and power relationships which together mediate outcomes. In Accra, several inter-

viewees felt that stakeholder interaction was inadequate:

Respondent: Yeah, so from Town and Country, I think they have been

practising urban forestry, and that tells you the situa-

tion, right. That there was never any integration, and

that's why maybe Forestry Commission is not even taking

part because they haven't been involved […].

(Researcher, FORIG)

Respondent: Yes, areas that we think have less canopy, certainly we,

once we demonstrate to the public that these are the ben-

efits from the study, the public should be able to coop-

erate to improve the canopy.

(Mr Fiati, EPA)

Researcher: Do you feel well informed about what is expected of you

in terms of work around trees?

Respondent: It is not so much. Information on trees are not so much.

(Environmental Health Officer, Ayawaso East)

While this view dominated, some individuals (and the groups they represent) felt adequately in-

formed about current developments and their role in urban forestry:

150
Researcher: Do you think that you have all the information necessary?

Do they tell you everything that you need to know for you

to be able to do your work?

Respondent: Oh, yeah, yeah, yeah.

(Mr Sampson, Ayawaso Central)

However, this last sentiment was voiced by the same interviewee, who was also unaware of the

need for individuals to obtain pruning permits for trees on their private land. Though he might

feel well informed, the available information suggests otherwise, further supported by his state-

ment that Ayawaso Central used to receive seedlings from the government until about ten years

ago but could not provide information on why this had ceased.

Some interviewees explicitly stated that improving working relationships between government

departments, consisting e.g. of the exchange of ideas or actual collaboration, would be beneficial

and indeed central to sustainable urban forestry in the city:

Respondent: […] but under the decentralisation programme, we are sup-

posed to merge with Parks and Gardens, but that one has

not been materialised. We are still battling with it. […]

But maybe as time goes on, this thing, we will be able to

realise, and Parks and Gardens will be part of the Physi-

cal Planning Department.

Researcher: Do you think that would be a good thing?

Respondent: I think that will be a very good thing, that will be a

very good thing. You prepare the schemes, you earmark a

place for greenery. Which department is responsible for

that? Parks and Gardens. So that they do it along with

151
 the scheme that we have prepared, and if they are some-

where, you do your thing, and they also do their own

thing and then it becomes very fragmented.

(Mr Yeboah, TCPD)

In Accra, the lack of coordination creates various outcomes. Thus crucial information either

passes through a long and resource-consuming chain of command or nowhere at all, with sub-

metropolitan assemblies, for example, receiving information through their head offices rather

than from DPG directly. Alternatively, incorrect information is disseminated (Ayawaso East

sending people to DPG for permits when the responsibility lies with EPA), efforts potentially

duplicated (educational campaigns by EPA and DPG), or conflicting advice given (EPA stating

the need for a pruning permit while Ayawaso Central is unaware of this rule and encourages the

sub-metropolitan residents to prune their trees). Each of these situations reduces efficiency and

the chance of producing an ‘overall’ urban forest that can provide varied benefits at different

scales. I, therefore, suggest considering the nodes of interaction in more detail, as stakehold-

ers/actors and their varied interests and aspirations meet at these locations. As a starting point, I

discuss the sizeable number and range of terms that interviewees used to describe interactions or

their lack (Table 25, see Appendix 10 for complete list).

Table 25 Examples of phrases used to describe interactions or their lack

Phrases

to (not) comply with something

to (not) be part of something

to (not) encourage someone to do something

to (not) control someone/something

to (not) work closely with someone

152
 to (not) educate someone

to (not) channel your ideas into someone else’s work

to (not) correct the attitude of someone

to (not) have your work undone

I identified a total of 134 terms and purposefully selected the above examples to discuss three

points: first, these phrases are often ill-defined; second, they contain ‘action directions’; and,

third, imply different levels of ‘coercive strength’. Interviewees talked about stakeholder interac-

tion at various management stages but often used vague words, which did not offer concrete

guidance on what precisely is expected of either stakeholder partnerships or partners. While sub-

metropolitan officers stated that it was their job to ‘encourage’ tree planting by the general public,

this consisted mainly of environmental health officers talking to people about tree benefits, wher-

ever and whenever seemed appropriate. As only one of many similar examples, these ambiguous

terms allow for multiple interpretations or even one of insignificance that may result in these ac-

tivities being disregarded altogether, with actors focusing on more explicitly stated (job) require-

ments.

In Accra, official documentation such as an urban forest management plan, which defines ac-

tor/stakeholder rights and responsibilities in unambiguous terms, remains unavailable. Instead,

especially EPA and TCPD have apparently misclassified the role and position of the admittedly

heterogeneous stakeholder group ‘the public’. My analysis suggests that these agencies consider

the city inhabitants, specifically the sub-group residing in sparsely greened areas, as (secondary)

‘observers’ or tertiary stakeholders only, rather than as primary, contributing ones. While second-

ary stakeholders’ “acceptance or compliance is required to sustain the activity”, tertiary ones only

“use the output of the activity” (McGrath and Whitty, 2017, p. 732). If planning and management

153
are based on such a (mis)classification, even if just implicit or subconscious, urban forestry un-

likely produces varied benefits and simultaneously runs the risk of increasing social injustices.

The selected terms also highlight how ‘direction’ and ‘coercive strength’ of interaction combine

and relate. During the coding process, I noticed how actor relationships fall into three types:

partner and cooperate; advise, support and assist; as well as instruct and inform. ‘Instruct and in-

form’ consists of a one-directional flow of information, or directive for action, with limited to no

further interaction between stakeholders. Here, the sending actor already decided on a course of

action which the receiver is expected to execute, perform, or simply take notice of (and potential-

ly incorporate into future actions). The level of ‘coercive strength’ varies from low to high, as

sub-metropolitan officers might just highlight the trees’ benefits which justify their planting or in-

struct tree owners to prune overgrown trees on threat of prosecution for failure to comply. As

with ‘instruct and inform’, stakeholder interactions of the ‘advise, support or assistance’ kind are

uni-directional, with knowledge or some other resource being transferred from one stakeholder

to the other one. However, the receiving partner also benefits from whatever resource is trans-

ferred. In this scenario, the donor has no further involvement following the transfer, which is

thus characterised by little or no ‘coercive strength’. The third type, ‘partner and cooperate’, con-

cerns interactions where at least two stakeholders work jointly to pursue a shared objective, with

each actor both contributing to and benefitting from the exchange. Also, as stakeholders often

collaborate at different stages of the management process, this interaction tends to be longer-

lived than the other types. As partners share a vision characterised by common priorities and ap-

proaches, ‘coercive strength’ should be minimal at most.

Dissecting stakeholder interactions draws attention to the various components that can affect

these exchanges’ outcomes. Each stakeholder enters the interaction with their specific back-

ground of resources, views, and priorities, i.e. the expected or wanted outcome. The power they

154
have to affect results varies; while some can draw on laws (e.g. EPA), others rely on creating their

legitimacy in other ways, e.g. through producing a situation in which they draw on morality (e.g.

GYEM). Also, as some interactions may exist over more extended periods as stable partnerships,

I suggest that their context reflects their gradual evolution over time. Overall, and taken together,

interactions shape urban forest planning and management in Accra and, by extension, the urban

forest’s structure and benefits while also impacting forest users and their livelihoods. Nonethe-

less, these interaction spaces require further examination, given that their functioning and dynam-

ics are central to a proper understanding of the strengths, weaknesses, opportunities and threats

of and to urban forestry in Accra.

To ensure that my analysis was based on these SWOTs, I used a versus-coding approach which

highlighted the importance of two types of interaction spaces: ‘empty’ and ‘colonised’. These

spaces most likely emerge or might indeed only become visible from a non-neutral position that

explicitly considers justice and equity as a desirable urban forestry outcome and input. An interac-

tion space remains ‘empty’ when there is no or only one stakeholder present who needs and/or

wants to interact with others. While the NGO Y4GG wanted to work with another NGO, they

could not find a suitable partner. Often, spaces remain ‘empty’ when groups or individuals are

not recognised as stakeholders either due to oversight or because their influence and/or im-

portance are (falsely) judged as low. In ‘colonised’ spaces, interactions should take place, but a

powerful stakeholder prevents others from entering the node, thereby eliminating the space and

assuming decision-making power for themselves. The Forestry Commission not consulting the

prayer groups about the Eco-Park plans constitutes an example of a ‘colonised’ space. As the in-

terviewed pastor asked: “Who are we? If Forestry Commission decides to do something, then we

have no power to do anything about it” (Interview Achimota visitor/pastor). In all instances,

analysing how power dynamics create a node and determine the rules of engagement therein is

central, an argument also made in participation research (Cornwall, 2002).

155
5.4.2 Theme 2 – The Long Road from Awareness to Action
In this section, I outline why some interaction spaces in Accra’s urban forestry remain ‘empty’ or

are ‘colonised’. To do so, I use a model common in marketing research that belongs to the ‘hier-

archy of effects’ models, describing how a (potential) customer passes through several steps be-

fore eventually purchasing a product. I decided to use this model when I analysed interview tran-

scripts and noticed some frequently used words. The earliest examples of such models appeared

at the beginning of the 20th century, theorising that “salespeople had to attract attention, maintain

interest, and create a desire to be successful” (Barry, 1987, p. 252). This original Attention, Inter-

est, Desire, Action model has since been modified and developed further, with authors adding

steps or replacing them (Barry, 1987). In 1961, Lavidge and Steiner initiated what Barry terms the

‘modern development phase’ of these models, proposing six steps: awareness, knowledge, liking,

preference, conviction, purchase. They further relate their six steps to three major functions of

advertising and behaviour (Table 26). These distinctions are “more than a semantic issue, because

the actions that need to be taken to stimulate or channel motivation may be quite different from

those that produce knowledge. And these, in turn, may differ from actions designed to produce

favorable attitudes toward something” (Lavidge and Steiner, 1961, p. 60-61).

156
 Table 26 People’s possible positions regarding the purchase of a product or service

Functions of Adver- Dimensions of Be-

Lavidge and Steiner’s Steps
tising haviour

1. “Near the bottom of the steps

stand potential purchasers who are
/ /
completely unaware of the existence of
the product or service in question.”

2. “Closer to purchasing, but still a

long way from the cash register, are
those who are merely aware of its ex- Information or ideas Cognitive (intellectual,
istence.” about the product mental, rational state)
3. “Up a step are prospects who know
what the product has to offer.”

4. “Still closer to purchasing are those

who have favorable attitudes to-
ward the product – those who like
Create favourable atti-
the product.” Affective (feeling or
tudes or feelings to-
5. “Those whose favorable attitudes emotional state)
wards the product
have developed to the point of pref-
erence over all other possibilities are
up still another step.”

6. “Even closer to purchasing are

consumers who couple preference
with a desire to buy and the convic-
tion that the purchase would be Conation (striving or
Produce action
wise.” behavioural states)
7. “Finally, of course, is the step
which translates this attitude into
actual purchase.”
According to Lavidge and Steiner (1961, p. 59, their emphasise)

157
In the case of urban forestry, the ultimate objective is not for another party to purchase a product

but to figuratively buy into a ‘product’ or proposed action or strategy. This may include support-

ing an NGO’s work (with the organisation and its work as the product), not encroaching on land

demarcated for greenery in planning schemes, or applying for a permit before felling a tree. The

keywords that make this model relevant revolve around awareness, knowledge, interest and pref-

erence, which, as the excerpts below show, interviewees used in several instances. Though they

sometimes mentioned these words directly and explicitly, sometimes I had to interpret the text

passages and their context before they became pertinent:

Respondent: Oh, cool! Cool, cool, cool. Dansoman [neighbourhood of

Accra], we wanna make Dansoman greener, that’s what we

want. So right now, er, anybody who wanna come on board

and support us to give that kind of face to Dansoman is

welcome, is welcome.

(Interview Y4GG)

Researcher: So there are a lot of these plans? Plans like this with

green space and trees?

Respondent: […] The whole place is supposed to be a green area, see

the green? Green, green, green, it's supposed to be a

green area, see the football pits? Yes, it's supposed to

be green, the whole area is green area, the green is tak-

ing about eighty per cent of it. But you go there now,

and it's a different, a completely different thing. So it

tells you the difference between the implementation and

the scheme. We prepared the scheme all right.

(Mr Yeboah, TCPD)

158
Researcher: So from your experience, people in neighbourhoods or

parts of the city with more trees are more likely to care

about them and protect them?

Respondent: Yes. Exactly. Exactly, exactly. So, middle-class are more

likely to take care of the trees and value the trees, the

lower class not so much. Yes, particularly in depressed

communities, the most important thing is to have your

shelter and the space. For them, space is so critical,

and so you see buildings flock together: Nima, Mamobi,

but as you move out you begin to see trees, of course,

you find some avenue trees along the streets, yes, you

can find some, but within the community, in the yards,

the density is quite low.

(Mr Fiati, EPA)

In these examples, one can identify desired results (the ‘purchases’) such as ‘take care of the trees

and value the trees’, or, respectively, the absence of these (‘But you go there now and it’s a differ-

ent, a completely different thing’). Additionally, some terms refer to (the absence of) interest

(‘anybody who wanna come on board’) or liking/preference (‘the most important thing is to have

your shelter’). Applying this model to urban forestry, we then have a range of ‘vendors’ (such as

government departments or NGOs) who try to ‘sell’ their ‘products’ to other actors and stake-

holders. If the ‘vendor’ correctly identifies the current positions of the potential ‘customers’, ac-

tivities and initiatives can be tailored to move them up the steps towards purchase. However, and

I suggest this is common in urban forestry in Accra, if ‘vendors’ misclassify ‘customers’ regarding

their current position, the former may waste resources or create interaction spaces, which remain

‘empty’ or end up being ‘colonised’. As an example, I use the very generic ‘maintaining trees for

their benefits’ as an end purchase which government bodies like EPA, TCPD and DPG want the

159
general public to buy into (Table 27). The text passages illustrate at which steps various actors be-

lieve the general public, or a sub-group of it, to be currently located at.

Table 27 Hierarchy of effects model with an urban forestry example

Steps Interview Excerpts
Awareness “[B]ut the awareness creation is ongoing. We have some fly-
ers that we have prepared that we issue out to educate peo-
ple about the importance of trees.”
↓ (Mr Fiati, EPA)
Knowledge
“[…] those who are not aware we are making effort to sensi-
tise them.”
(Mr Awidi, DPG)
↓
Liking “Now speaking and researching and working in the setting, I
realised that what people really care about is their socio-
economic status, like how they can make ends meet. If they
are not able to do that, other issues hardly strike them,
↓ you know, like they hadn't taken notice of other issues
Preference when they haven't been able to feed themselves, haven't
been able to take care of their children.”
(Mr Commey, GYEM)

↓ “The other day somebody, we were going down this way around
Conviction Dzorwulu, and somebody was like ‘you know, some time ago
this whole area used to be bush and now the area is so
beautiful, nice buildings’. They say ‘wow, it's beautiful’
and I say ‘yes, but nice buildings without vegetation
doesn't make it beautiful’. But anyway, that’s it.”
(Researcher, FORIG)
↓
‘Purchase’ Maintain trees for their benefits and services

Some interviewees, especially from government departments, located a large proportion of the

general public in the first stages, suggesting that people need to be educated and informed on ur-

160
ban trees and forests. The interviewee from EPA discussed how his department, together with

the Environmental Education department, was preparing flyers as part of an education campaign.

Such campaigns may have the desired effect if people are indeed unaware or uninformed. How-

ever, they unlikely produce positive results if other stakeholders, like the diverse ‘general public’,

simply prioritise differently and focus on feeding their families or paying their children’s school

fees, as suggested by the interviewee from an NGO (Mr Commey, GYEM).

Comparing stakeholder interviews with the household survey reveals an intriguing situation in

Accra: the survey (and household case study) suggests the general public to be aware and well-

informed on urban tree benefits, which explicitly includes residents of high-density and less green

areas. Another recent study investigated ‘slum’ residents’ perceptions of urban green spaces in

Nima (and Agbogbloshie) and corroborated my findings, concluding that “results from the

household survey show respondents demonstrated a reasonable understanding of green spaces

and their importance” (Cobbinah et al., 2021, p. 6). Yet, many government bodies seem to focus

on flyer campaigns or one to one conversations to ‘educate’ and ‘sensitise’ the city’s residents on

the merits of urban forests. The disparity between actual stakeholder position within the hierar-

chy of effects model and assumed position as seen by the ‘vendors’, likely reduces the latter’s

ability to sell their ‘product’ with potentially negative effects on sustainable urban forestry.

Not all public actors/stakeholders, however, support this view of the widespread need for further

educating the city’s inhabitants:

Researcher: So when you talk to people about trees, do you feel that

they are aware of the benefits of trees?

Respondent: Yes.[…]Sometimes, when you ask some of them to even trim

their trees, then they [don’t want to]. Medicinal purpos-

161
 es and other things, yes, it’s true, but it doesn’t mean

it should be overgrown.

(Mr Sampson, Ayawaso Central)

Researcher: And when you do that, do you feel that people are aware

of the benefits of trees?

Respondent: Oh, people, those that we talk to, know very well that

trees are great to plant.

Researcher: So people are not resistant to listening to you talking

about tree benefits?

Respondent: No, no, no, no, no.

(Environmental Health Officer, Ayawaso East)

My household survey, which elicited the public’s views on urban trees and their functions on

public and private lands (see Section 3.3.2), reflects the experiences of these two sub-

metropolitan officers. Right at the beginning of the questionnaire, and before the following ques-

tions would suggest any possible answers, I asked survey participants to list urban tree benefits.

Every household could name at least one benefit, with many people listing significantly more, on

average three. Furthermore, people were also aware of maintenance activities and some even of

the more common laws and rules surrounding urban forestry.

Why, despite evidence to the contrary, does this narrative of the unaware and unknowledgeable

public prevail with activities of government organisations primarily tailored at educating and sen-

sitising people? My data suggests three possible explanations. First, it might be an innocent over-

sight, potentially due to a lack of data, which creates ‘empty’ or ‘colonised’ interaction spaces as

162
the city’s population remains unacknowledged as a potential partner. Second, educating the pub-

lic could be a makeshift solution as most government departments lamented the lack of re-

sources. Flyer campaigns and other sensitisation campaigns offer a comparatively cheap option to

give the appearance of activity and involvement and might receive less (overt) resistance than

more extensive interference in people’s lives. Furthermore, government employees themselves

are not generally able to affect the available budget and might feel disempowered, lacking finan-

cial means for alternative approaches. However, a third explanation involves power dynamics and

the ability to create and enact the rules which generate and shape interaction spaces. When the

general public is constructed as ignorant, there is little or no justification for recognising this

stakeholder as a legitimate potential interaction partner, except as one in need of edification via

the simple remedy of flyers and other forms of public ‘education’.

Should actors like EPA, TCPD or DPG acknowledge that private individuals’ decisions result

from other priorities rather than reflecting unawareness, colonising supposed spaces of interac-

tion would be morally less defensible. These decisions include both the direct and active partici-

pation in tree-related activities as well as seemingly unrelated actions which, however, ‘replace’

the former (e.g. using resources to obtain money to pay school fees rather than planting trees).

Ultimately, individuals compare their options’ opportunity costs and potential benefits. Especially

resource-poor groups will more commonly decide against the tree-related option. It might be

more useful, though, to re-frame this as not so much a decision against active tree care but as

one for other activities with lower opportunity costs and more immediate benefits. Such an ap-

proach could start a constructive dialogue focused on devising strategies that incentivise resi-

dents’ participation in urban forestry but requires engaging meaningfully with all stakeholders. It

would, however, require acknowledging that decisions regarding tree management largely result

from complex and dynamic structures and processes of socio-economic and environmental

change. These processes have made living conditions and livelihood circumstances less secure for

163
many individuals who have to invest available resources first and foremost into meeting basic

sustenance needs, rather than in urban tree care for the medium to long term, even though the

two are not necessarily mutually exclusive.

Though I adopted and applied Lavidge and Steiner’s model in response to the interview tran-

scripts, it still leaves the question of whether the model is sensitive enough to the variety of be-

haviour this chapter tries to explain. For, as the text makes clear, even though all forest manage-

ment institutional actors seek to convince Accra inhabitants to maintain and/or propagate trees,

the specific interventions they promote to pursue this goal in the short term vary noticeably.

While they share tree protection goals, the means and methods for achieving these differ. In Ac-

cra, the various forestry institutions encourage people to ‘buy in’ to slightly different options ra-

ther than simply acquiring ‘end products’. The relative importance accorded to considerations

like awareness, knowledge or motivation in such rankings might not necessarily correspond with

their position in the model hierarchy. This is explained by the variety of activities being ‘bought’

and ‘sold’ and by their diversity which, in practice, means ‘purchasing’ decisions are not necessari-

ly based on like-for-like comparisons.

Thus in promoting urban forest conservation, people are sometimes actively encouraged to maintain

trees by undertaking particular tasks. At other times, to prevent damage to or threaten the surviv-

al of trees, they are explicitly discouraged from undertaking particular tasks, including tree felling. On

yet other occasions, the requirement is for people to go beyond protecting and maintaining to ac-

tually propagating or planting seedlings. Similarly, encouraging a landowner to prune trees in their

spacious private garden requires a different ‘sales pitch’ from that aimed at a squatter in their self-

built shack needing to be ‘nudged’ into accepting the legal requirement to pay a fee for a permit.

Its overall utility notwithstanding, Lavidge and Steiner’s market model’s capacity to cater for

complex non-market considerations, including trade-offs and opportunity costs between different

164
forest management options and for different demographics and neighbourhoods, appears some-

what limited. Equally justifiably, one might argue that such a capability was incorporated into nei-

ther the model’s initial design nor its original purpose. Either way, this could be a ready-made

subject for future research.

5.5 Chapter Conclusion

In this chapter, I introduced the organisational structure of Accra’s urban forestry, outlining the

roles and management objectives of various stakeholders. A SWOT analysis served as the starting

point for considering tree-related planning and management in the city, showing how some prob-

lems are commonly experienced across various stakeholder groups (e.g. lack of resources). Other

strengths, weaknesses, opportunities, and threats are unique or experienced more strongly by

specific actor types. The chapter illustrated how formal and informal rules and mechanisms pro-

duce spaces of interaction, a key element through which one can try to understand the complexi-

ties of urban forest structure and its dynamics over time and across space. Additionally, these

spaces also allow scrutiny of the management process and how to improve it both in Accra and

elsewhere. If the government wants to increase people’s involvement in tree planting and care, it

needs to recognise that these activities are best understood as part of complex and dynamic indi-

vidual and group livelihood calculations, including the use of private land and trees/shrubs as

livelihood resources. Seen this way, multi-purpose urban forestry becomes integral to processes,

which aim to diversify and thereby render urban livelihoods more sustainable.

165
 6 URBAN TREES – URBAN LIVELIHOODS

“Menni adeɛ pɔtee ɛntia mani gye ɛndua ho, ɛye adeɛ a mani gye hoo kɛkɛ.”

“I just can’t say why I particularly like trees... I just like them.”

John Yetsowodo (Household 2, Person 1)

In this chapter, I discuss the various livelihood roles and functions of urban trees and their prod-

ucts. Rather than simply quantifying how trees contribute to urban livelihoods, the household

case study elicits an in-depth understanding of human–tree interactions in urban settings. Among

others, this includes information on why and how people use tree benefits and their choices re-

garding tree protection, propagation and maintenance. Centring city residents’ lived experiences,

who are generally found at the receiving end of policies, laws and rules made by EPA, DPG or

TCPD, it complements the previous chapter, which focused on government departments and

NGOs. After introducing important concepts and their definitions, such as ‘urban’, ‘household’,

and ‘livelihood’, I then discuss the Sustainable Livelihoods (SL) Approach. Subsequently, I focus

on the four case study households, for which I provide a tabular overview for quick reference

and more detailed, descriptive information. I then discuss two topics in more detail. First, I con-

sider three different livelihood functions of urban trees, i.e. trees for subsistence use, as a source

of cash income and in the production of social capital. Second, I discuss tree caring activities,

such as propagation, fertilisation or pruning, both topics also offering insights into intra-

household dynamics.

166
6.1 Urban Lives
Investigating urban trees’ functions for city residents must commence with a definition of ‘ur-

ban’, a core component of both ‘urban forestry’ and ‘urban lives and livelihoods’. However, dif-

ferent countries define ‘urban’ in varying ways, thereby complicating cross-country comparisons

(Satterthwaite, 2010). In Ghana, localities with at least 5,000 inhabitants are officially defined as

urban (Ghana Statistical Service, 2014b). Such a threshold is considered rather low and may be

“less reliable to distinguish between ‘rural’ and ‘urban’ settlements in an economic functional sense”

(Potts, 2018, p. 968, their emphasis). Thus Ghana’s Statistical Service notes that “an area could

easily attain the 5,000 threshold population to qualify as urban but still maintain its rural outlook

based on the economic activities the residents may be engaged in” (Ghana Statistical Service,

2014b, p. 4). While the Statistical Service suggests reviewing the classification for future censuses,

with several million inhabitants, these definitional changes will not affect Accra’s classification as

‘urban’.

Despite existing definitions, some researchers argue against viewing ‘urban’ and ‘rural’ as two dis-

tinct concepts (Wratten, 1995; Cohen, 2004; Potts, 2017). Wratten (1995, p. 20), for instance, ar-

gues that “[a] strong case can be made for treating the urban–rural divide as a continuum rather

than as a rigid dichotomy. […] The cut-off point for any such division is bound to be chosen ar-

bitrarily”. Migration, seasonal labour, or family support networks link rural and urban areas and

“[i]ntervention in one part of the system will have a range of repercussions, affecting other parts

of the ‘same sector’ as well the ‘other sector’ (Wratten, 1995, p. 20). The interviewee from the

Department of Town and Country Planning also commented on these interdependencies, argu-

ing that Accra’s growth had changed peri-urban areas. In these places, the expanding city con-

strains crop cultivation and farmers are then “forced to come to the city to sell on the streets and

the rest. So if we don’t strategise very well, then we always have a compounded situation in Accra

with respect to problems” (Mr Yeboah, TCPD). Various social, economic, ecological, and politi-

167
cal linkages, therefore, connect different settlement types and affect urban lives and livelihoods.

However, some specific traits do distinguish cities from smaller, rural settlements, especially

when looking at Accra, with a population significantly exceeding the definitional threshold.

Urban and rural areas may differ in some regards, but neither is homogeneous. While cities’ aver-

age population densities exceed rural ones, some neighbourhoods may be well above and others

well below this value. Similar spatial variation can exist for income, land use, and neighbourhood

formality (Farrington et al., 2002) or, as already seen, in terms of tree cover. Wratten (1995, p. 21)

suggests that “[o]nce the city is studied as a series of interrelated but heterogeneous neighbour-

hoods or districts, the dichotomous categories of urban and rural become less relevant”. Instead,

she proposes to use different categories to classify, and to illustrate, variation at sub-city level. In

the case of Accra, its early development from Ga villages into seven city quarters (eleven when

including Osu/Danish Accra/Christiansborg) preserved the areas’ political, ethnic, and kinship-

related fragmentation for some time (Parker, 2000). Additionally, the city’s continuous expansion

incorporated rural fringe areas into the growing town, and these places often retained some char-

acteristic features of non-urban lands (Doan and Oduro, 2012; Gaisie et al., 2019). Thus, when

discussing urban lives, livelihoods, views on, and uses of urban trees, possible intra-city differ-

ences should be considered.

Regardless of sub-city variation, some settlement and lifestyle characteristics are more prevalent

in cities than in the countryside. These often include certain types of modern infrastructure and

facilities in the areas of education, health, energy or environmental services (Ghana Statistical

Service, 2014b). However, housing costs, even in unplanned residential areas with limited access

to these services, can account for a large proportion of household income, sometimes combined

with insecure tenure. Furthermore, higher crime rates and social diversity and fragmentation can

also characterise urban living (Wratten, 1995). Cities “attract rural migrants and refugees with dif-

168
ferent ethnic, cultural and linguistic origins”, thereby creating ‘heterogeneous melting pots’ (Wrat-

ten, 1995, p. 23). This diversity can create tensions and impersonal relationships as “[l]ifestyles,

kinship and neighbourhood support networks are different from those in rural areas” (Wratten,

1995, p. 24). Additionally, more immediate contact with state agents and policies, which can be

oppressive (Wratten, 1995; Beall and Fox, 2007), can make urban residency more demanding.

Both these positive and negative features can affect people’s lives and livelihoods and the role of

trees within.

6.2 Livelihoods – Urban Livelihoods – Sustainable Urban Livelihoods

In the last three decades, much research and many policy, programme, and project interventions

have targeted livelihoods (Small, 2007; Zhang et al., 2019). While exact terms and definitions vary,

a livelihood is a means to a living (Chambers and Conway, 1992; Scoones, 1998). People pursuing

a livelihood utilise their tangible and intangible assets, sometimes also called livelihood capitals,

which can take the form of stores, resources, claims or access to these. Types of capital (or asset

categories) typically include natural, economic/financial, human, social, physical and political cap-

ital (Scoones, 1998; Krantz, 2001; Farrington et al., 2002) on which people draw to make a liveli-

hood. A ‘sustainable’ livelihood, in turn, can “cope with and recover from stress and shocks,

maintain or enhance its capabilities and assets, and provide sustainable livelihood opportunities

for the next generation” (Chambers and Conway, 1992, p. 6). While Chambers and Conway add a

further sustainability criterion, a livelihood contributing “net benefits to other livelihoods at the

local and global levels and in the short and long term” (Chambers and Conway, 1992, p. 6), this is

both difficult to assess and achieve (Scoones, 1998).

Two elements are central to sustainability. A livelihood can be environmentally sustainable “in its ef-

fects on local and global resources and other assets” and socially sustainable, i.e. “able to cope with

169
stress and shocks, and retain its ability to continue and improve” (Chambers and Conway, 1992,

p. 9). While the first aspect evaluates a livelihood’s external impacts, the second one considers its

internal capacity to cope with pressures from the outside (Chambers and Conway, 1992). Social

sustainability can be either reactive or proactive and dynamic “in adapting to, exploiting and cre-

ating change, and in assuring continuity” (Chambers and Conway, 1992, p. 10). To cope with

stresses and shocks, people employ different strategies, including stinting, hoarding, protecting,

depleting, diversifying, claiming, and moving (Chambers and Conway, 1992). A diversified asset

portfolio and response repertoire then allow households (or other study units) to “respond more

effectively and with less loss” (Chambers and Conway, 1992, p. 11) to changing circumstances.

Furthermore, external interventions, such as public actions (e.g. flood reduction), can also reduce

livelihood vulnerability.

Most SL approaches use households as their unit of analysis, which can mask an individual’s la-

bour or how resources, or access to them, differ within. As with ‘urban’, different definitions of

what constitutes a household exist (Beaman and Dillon, 2012; OECD, 2013). Fundamentally,

though, it is a (political) economic concept that “implies a domestic unit with decision-making

autonomy about production and consumption” (Guyer, 1981, p. 89). However, decades of re-

search in African countries have demonstrated the concept’s problem “of designating complex

collectivities as units” (Guyer, 1981, p. 104) that jointly decide on resources and their allocation.

In the 1980s, (rural) forestry and agroforestry research showed how “complex structures and

processes were governing the gendered division and sharing of resources” (Rocheleau and Ed-

munds, 1997, p. 1352). Similar dynamics might also affect urban forestry, though ‘age’ or ‘inter-

est’ could be further variables to consider in addition to ‘gender’. The Ghana census defines a

household as “a person or group of related or unrelated persons who live together in the same

housing unit, sharing the same housekeeping and cooking arrangements and are considered as

one unit, who acknowledge an adult male or female as the head of the household” (Ghana Statis-

170
tical Service, 2014a, p. 4). While my household questionnaire aggregated at this level, I specifically

focused on intra-household differences in the case study.

Although livelihood thinking initially developed in relation to rural society, economy and nature,

researchers and practitioners also applied its insights to urban contexts. As the models are con-

ceptually broad and context-specific (Farrington et al., 2002), they can be adapted to reflect the

different conditions in city environments. Thus, urban-focused versions of SL approaches (e.g.

Moser 1998; Meikle et al., 2001) highlight access to housing or the greater reliance on cash (Sat-

terthwaite and Tacoli, 2002). The approach’s strength lies in how it demonstrates that people

employ different strategies – located within the political, social, economic, environmental, or legal

context – to make a living. It illustrates that poverty is not just a problem of low income (Wrat-

ten, 1995) but rather requires a holistic view to understand what resources people draw on to

construct and maintain a living. The following part, my household case study, now considers the

roles of urban trees and their benefits within such a framework. Descriptive accounts of who

does what, how, where and when are followed by an analysis of the trees’ multiple livelihood

functions, conflict and cooperation evolving around them as well as their network-creating po-

tential.

6.3 Household Case Study

A single older man, two nuclear families with smaller children, and a woman with her teenage and

young adult sons living with her extended family, all of these households pursue a livelihood in

Accra. Benefitting from the city’s opportunities, they simultaneously find themselves largely de-

pendent on cash and market exchanges, limited by land shortages, and potentially disconnected

from the wider family and other social ties. To make a living, they draw on various capi-

tals/assets, including their compounds’ trees. However, not every individual participates in any or

171
all tree-related activities, either because other household members exclude them or simply be-

cause they lack interest. At certain times, they might cooperate and divide work along gender

lines. At other times, some tasks, often the more cumbersome ones like picking up and disposing

of leaves and branches, might be laid on younger children. These same children though, as teen-

agers and young adults, can become the sole responsible party for tree propagation and care,

both grown out of an interest in the environment and economic considerations. Dynamics evolv-

ing around tree care and their benefits are not just limited by household affiliation but can extend

to other family members, neighbours, friends and even strangers. Social interactions in the trees’

shade or gifts, sales and exchanges of tree products create networks among city residents, some-

times with branches spreading to nearby peri-urban and rural areas.

The four selected households (HH) are located in Nima, Mamobi, and Kokomlemle, which re-

ceived a very high (class 5), high (class 4), and low (class 2) poverty rating, respectively. However,

a household’s specific socio-economic conditions can differ significantly from these neighbour-

hood ratings. As the table below illustrates, the Mamobi household had a considerably higher PPI

score than the Kokomlemle one (Table 28). The table also summarises information both on par-

ticipants and their trees, while Appendix 11 provides further tree data, such as height, DBH, or

crown spread. Besides describing each household, I also provide compound maps, which indicate

locations of buildings, trees, and other relevant elements. I have attempted to balance detail with

the need to protect participants’ identities, as there was only one household where all members

opted to have their names attached to their data (HH 2).

172
 Table 28 Summary information on the four participating households
Household 1 Household 2 Household 3 Household 4

Area Nima Mamobi Nima Kokomlemle

PPI 60 81 63 67

Person 1 Person 1 (John) Person 1 Person 1

 Head of Household  HHH  HHH  HHH

(HHH)  male  male  female
 male  40 to 45 years  35 to 40 years  45 to 50 years
 65 to 70 years  electrical technician,  sells beverages
 born and raised in Accra sells water and coco- Person 2  cooks for the whole
 used to work as an elec- nuts, rents out accom- household in the eve-
modation  Wife of HHH
trical engineer, now re- nings
tired  30 to 35 years
Members
 2 children in their 20s  takes care of children
Person 2 (Christine) Person 2 and 3
who live outside Accra and house duties
 wife of HHH  cleaning job once/week  children of person 1
 30 to 35 years  other days fries and sells  15 to 20 years
 takes care of children yam, also produces and  attend school and uni-
and house duties sells bead articles versity
 operated a grocery store
before having children
Person 3, 4 and 5 Person 4
and plans to do so again
when they are grown  children of person 1 and  brother of person 1

173
  sells water if husband is 2
not around  between 4 and 10 years Person 5 and 6

 sisters of person 1
Person 3 and 4
 own and sell in the shop
 children of person 1 and next to the compound
2
 under six and newborn Person 7 and 8

 children of person 6
 under 5

Person 9

 mother of person 1, 4, 5,
6
 70 to 75 years
 landlord of the com-  person 1 is landlord of  live in compound where  compound with mostly
pound the compound all people are part of ex- family members
 used to be father’s land  land originally bought by tended family  some trees planted by
 mother lives in same father when he retired  everybody in compound late father of person 1
Living
compound from the military can take from trees
situation
 7 other households in  8 other households in
the compound compound
 everybody in compound
can take from trees

174
 8 trees 43 trees 3 trees 20 trees

 mango (1)  coconut (24)  avocado (1)  coconut (8)

 avocado (1)  oil palm (3)  moringa (2)  oil palm (2)
 Terminalia catappa (1)  papaya (1)  mango (1)
 Newbouldia laevis (5)  neem (1)  Annona squamosa (1)
 Newbouldia laevis (14)  moringa (8) – planted by
Trees  Ficus capensis (1) Mr Mamattah, a neigh-
7 saplings
 Hippocratea sp. (2) bour/friend with no
 mango (1) space in his own com-
 soursop (4) pound
10 saplings
 velvet tamarind (1)
 coconut
 unknown species (1)
3 seedlings

 lemon (3)

Own Own Own Own

 medicine  food  medicine  food

 spiritual cleansing  medicine  food supplement  medicine
 shade  spiritual cleansing  food  shade
Uses  windbreak
 food Other Other Other

 sell coconuts  gifting  sell mangos

Other
 sell seedlings  sell coconuts
 gifting  gifting  gifting

175
6.3.1 Household 1 – Nima
The first case study household in Nima is located at one of the major roads and approximately

600 m2 large (Fig. 35). The participant (HH 1, P 1), a retired male and the compound’s landlord,

shares the plot with seven other households. While he mentioned a wife and two children in the

questionnaire, his adult son and daughter had moved out, and he avoided discussing his wife. He

resides in his late father’s place, and his mother, whose poor health impeded interviewing her, al-

so lives in the compound. A wall of about 1.6 m in height, built from purple-grey stones, sur-

rounds the compound. Though a small gate on the compound’s west side opens directly to the

main road, both visitors and residents commonly enter through the red south gate that connects

to a smaller foot sand path.

Figure 35 Compound layout (Household 1)

Buildings: A – Houses12 B – Shop C – Bathrooms
Trees: 1 – Mango 2 – Avocado 3 to 7 – Newbouldia laevis 8 – Umbrella tree (Terminalia catappa)
Saplings: 1 – Mango 2 to 5 – Soursop 6 – Velvet tamarind (Dialium guineense) 7 – Unknown
Others: i – Benches

12 case study households in dark grey

176
Entering the compound, one immediately faces a large mango tree (1) whose branches extend

over both the surrounding houses and the wall, casting shade on to large, self-built benches just

on the wall’s outside. Initially, the participant had two such benches within the compound and

one outside, but as more and more people came to rest on the outside bench, he added another

one to give more space. As these benches were already old, he eventually replaced them with

those from within the compound and dismantled the broken outside ones, giving the wood to a

neighbour who used it for her food preparation business. As the participant’s father and brother

both worked as carpenters, he had learned some basic skills, and his late father had left him his

tools. Most of the wood he had acquired for free from carpenter friends in the area or taken from

construction left-overs. The participant constructed these benches as he felt that the air inside the

compound was too warm and not fresh, and he now spends a lot of his leisure time resting, eat-

ing, and talking to people on these benches. Often, other compound residents, neighbours,

friends, and even random passers-by also rest or socialise on these benches. Underneath the

bench is a little seedling, possibly an avocado, growing from one of the many seeds discarded by

the landlord when eating fruits while sitting on his bench.

According to another compound resident, Mr Agbobli, an old friend of the landlord, his father

planted the mango about 80 years ago. Up to about 1.8 m in height, there are signs of pruning on

the mango’s stem where bigger branches had been removed to allow people to pass through. The

landlord, and later a friend, removed branches about three to four times per year to prevent caus-

ing damage to the surrounding houses. However, five years ago, the friend died, and the landlord

neither owned the required equipment nor felt healthy enough to prune the tree himself. Further

up, the mango spreads into two larger stems, one of them growing towards the house where

people made a hole into the metal sheet roofing for the branches to grow through. Residents

have attached several lines to the tree, some to hang washing while others presumably seem to be

electricity wires. While compound residents do not regularly water the mango, I observed how

177
wastewater from household use was disposed of on the tree’s base on more than one occasion,

one of the few open spaces in an otherwise overbuilt place. Every compound resident may col-

lect fruits from the tree, and when passers-by inquire, they are also allowed to pick some mangos.

The landlord’s accommodation is separated from the other houses by a low wooden fence locat-

ed on the right-hand side when entering the compound. In front of his entrance door grows an

avocado (2) which he planted about 12 years earlier, hoping to obtain some avocado fruits. So

far, the tree did not bear fruits, even when he tried to prune it, and he now considers replacing it

with a hopefully more productive seedling. There are also three Newbouldia laevis trees (3 to 5) and

several saplings of various species around which he built some planters of 0.5 m by 1.2 m and 0.5

m height to protect the roots from exposure. He planted all these seeds less than a year ago, and

now all plants have grown to between 0.8 m and 1.2 m in height. Two Newbouldia laevis trees are

also found at the compound’s opposite end (6 and 7), growing behind one of the houses. Just

outside the compound wall, located at the main road but on his land, is a large umbrella tree (8)

which the participant planted in 1979.

6.3.2 Household 2 – Mamobi

The second case study household is located in Mamobi, a neighbourhood directly north of Nima.

The household’s head is John Yetsowodo (HH 2, P 1), a middle-aged man who lives with his

wife Christine (HH 2, P 2) and two young children under six. They live in a large compound of

about 1,660 m2 (Fig. 36) that is visible from afar due to the many high coconut palms which stand

out in an area otherwise sparsely covered by trees. The other accommodations are rented out to

eight households to which they are not related.

178
 Figure 36 Compound layout (Household 2)
Buildings: A – Houses D – Shack E – Water tank
Trees: 1 to 24 Coconut 25 to 27 – Oil palm 28 – Papaya 29 – Neem
30 to 43 – Newbouldia laevis 44 to 45 – Hippocratea sp. 46 – Ficus capensis
Saplings: 1 to 10 – Coconut

179
John grew up about 200 m away from his current place with his parents and two siblings. When

he was still young, his father bought the land John currently resides on to farm and garden on it.

As his father lacked the money to build a house, he fenced his land with coconuts and planted

about 90 of them, which he brought from his own father’s village in the Volta Region. After

school, John and his two siblings would help their father weed, plant, and maintain the plantains,

bananas, corn, and oil palms he also planted. Around 1998, John started to cut down some palms

to make space for rental accommodation, and when he and his wife expected their second child,

they also moved to the new compound. John’s father has since died, and his (step)mother lives in

Lapaz (Fig. 37).

Figure 37 Location of Lapaz

Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
12 – Lapaz

180
John works as an electrical technician but pursues a range of other income-creating activities. Be-

sides renting out accommodation on his land, he sells water from a large tank in his compound

and another one at a marketplace. He also regularly sells coconuts and runs some unspecified er-

rands. Christine usually sells groceries but stopped working for some time to tend to their new-

born. When John is out, she sells the water but does not participate in the coconut sale. While

John used to be part of a neighbourhood support group in the 1990s, these activities stopped

when the group collapsed. He belongs to the Ethiopian World Foundation but has not attended

any of their meetings in the past three years. However, in 2011 members of a US Rasta Union

came to Ghana and collaborated with his local group in some tree planting activities around

Korle Gonno and Bu (Fig. 38) in which John also participated.

Figure 38 Location of Korle Bu and Korle Gonno

Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL
61 – Korle Bu 74 – Korle Gonno

181
Entering the compound through the main gate, one immediately faces a large water tank, and

during data collection, several people came to buy water from him. Initially, his father had dug a

well from which he sold the water before John dug a mechanised borehole and added pumps and

pipes. Today, the Ghana Water Company’s pipes reach their place and provide the water. On the

left-hand side of the gate, there is a high wall that only has a small passage to the other side of the

compound, right next to the water tank. Several trees/palms (10, 44 to 46) are overgrowing the

wall, the large Ficus capensis (46), a species that can be deciduous or evergreen, however, had shed

most of its leaves at the time of data collection. The head of this case study household reported

that both the Ficus and the Hippocratea (44 and 45) trees had not been planted but regenerated

naturally. They were left standing as people from the area asked for leaves and bark for medicinal

and spiritual purposes, which is how the landlord and his father learned about the uses of these

specific species. The majority of the compound’s trees are located in the bottom left part, where

coconuts, neem, oil palm and Newbouldia laevis form an almost closed canopy, the latter (32 to 41)

planted along another fence which separates the left bottom and left back part of the compound.

In the bottom-left part of the compound, the landowner also grows several coconut saplings. He

had also once planted mango and plantain, which were eaten by goats from another house that

entered through the side gate and came to the bottom left part as the wall at that time did not

have a door.

6.3.3 Household 3 – Nima

The third case study household also lives in Nima and consists of a young married couple with

three younger children between the ages of four and ten. The compound covers approximately

580 m2, and the other residents are mostly relatives. The woman (HH 3, P 2) was born in Nima

and lived nearby until 2008 when she moved to her current compound to join her husband. She

usually performs all household duties and prepares the children for school, but one day per week,

182
she has a cleaning job in Madina, just north of Accra. On the remaining days, she fries and sells

yam in front of her house or beads sandals or bracelets for sale. Her husband works long hours

and does not participate in decision-making around the compound’s trees. Entering through the

gate, one faces the large avocado tree (1) around which the houses are arranged (Fig. 39). When

the fruits are ripe, every compound member may collect what they want and, according to the

participant, there have never been conflicts concerning the fruits’ distribution. Before she joined

the compound, the residents had used building bricks to form a circle (diameter: 2 m) around the

avocado’s stem to avoid extensive sweeping in that area which might lead to soil erosion and ex-

pose the roots. When she joined the compound and realised that some blocks had broken, she

replaced them with stones from the neighbours who were just rebuilding their wall and had some

leftovers.

Figure 39 Compound layout (Household 3)

Buildings: A – Houses D – Shack
Trees: 1 – Avocado 2 to 3 – Moringa

183
At the back of the houses, a former resident and family member planted three moringa trees

about six years earlier to use their leaves for medicinal purposes and as a food supplement. He

did not obtain permission from the other compound residents, but the moringa trees were com-

mon property, and anybody could use them as they wished. In 2016, however, one resident cut

one of the trees to allow more light into her building. Additionally, she also preferred to plant

plantain for direct consumption, and while she approached each compound household and in-

formed them of her decision, she did not specifically ask for permission. The two remaining

moringa trees had been topped off at about 1.5 m height and were now resprouting, providing

the compound residents with leaves. Generally, the available growing space between the backside

of the house and the compound wall leaves little space for the trees to mature.

6.3.4 Household 4 – Kokomlemle

The fourth case study household is located in the Kokomlemle neighbourhood, bordering Nima

on the west. Household members include a middle-aged female head of household (HH 4, P 1),

her two children (a teenager and a young adult), as well as her mother, brother and two sisters,

one of whom has two small children of her own. Their compound’s size equals about 1,000 m2

and is surrounded by a high wall (Fig. 40). The household’s head usually sells beverages around

the Accra Technical Training Centre in Kokomlemle but has been unable to exercise her profes-

sion for about a month due to ill-health. Her sisters, also traders, operate a small shop just out-

side the compound walls where they sell a variety of everyday items, including bread and pre-

served food and drinks. The younger children visit school, while the older son (P 2) attends uni-

versity. The household head’s late father bought the land several decades ago and started planting

trees on it until he could afford to construct accommodation. Each time the man’s wife gave

birth, he also planted another tree on the land, and 30 years ago, the family moved there.

184
 Figure 40 Compound layout (Household 4)
Buildings: A – Houses B – Shop D – Shack
Trees: 1, 2, 4, 14 to 17, 20 – Coconut 3, 5 – Oil palm 6 to 13 – Moringa
18 – Annona squamosa 19 – Mango
Seedlings: 1 to 3 – Lemon

The small proportion of built-up land within the compound creates an ‘open’ and spacious feel-

ing though old cars in various stages of repair take up some of the compound’s area. The house-

hold has rented out one of their buildings to a household with no family connection to them.

The large mango tree (19) towards the compound’s centre is in the process of dying and has not

given fruits in the past seven years. Along the walls, the residents have planted coconut and oil

palms but, due to their height, they provide limited shade to the compound. In the Annona squa-

mosa’s (18) shade, the household head’s children nurse lemon seedlings (see Section 6.4.2), three

of which they have already planted out. Two avocado trees also used to grow in the compound,

185
but one had died due to old age or possibly termite damage, while the other was removed when a

former tenant complained about its shade depriving his accommodation of natural light. In the

compound’s top right part, a neighbour/friend of the head’s brother has planted eight moringa

trees (4 to 13)(see Section 6.4.1). He has planted them in his friend’s compound as he is only

renting his own place and does not want to plant on land he does not own. Additionally, there is

limited space to plant in his rental accommodation, and the landlord’s previous actions did not

suggest that they were particularly interested in trees.

6.4 Tree Use and Care in the Case Study Households

Over the course of about two weeks, I spent several hours with the above-introduced house-

holds, interviewing individuals and observing their tree-related activities. To compare and con-

trast household and intra-household views, activities and dynamics, I summarised information on

various topics (e.g. tree care, sale, gifting and different tree benefits). These topics are reflected in

the semi-structured interview guide and draw on the literature about urban tree use and aspects in

the sustainable livelihood framework. I now discuss two topics in more detail. First, I outline the

various ways in which household-owned trees and their products serve as assets, including their

domestic use, as a source of income and as part of the process of creating or maintaining social

capital. After that, I compare and contrast tree propagation, care and management of the four

case study households, including observed intra-household dynamics.

6.4.1 Trees as Assets

Within a sustainable livelihood approach, trees constitute a form of natural capital/asset and pro-

vide various benefits. In Accra, city residents use tree products such as leaves or fruits for medic-

inal and alimentary purposes. A more diverse diet and improved health, in turn, strengthen the

186
workforce, i.e. the human capital, which possibly expands production or enhances productivity.

Additionally, tree owners sell products, generating financial capital to spend on physical capital

like construction timber, building poles, furniture wood and a range of household and other

utensils and tools. As fuelwood and charcoal, wood is also a common source of domestic energy

in Accra, like elsewhere in African cities (Zulu, 2010; Kimemia and Annegarn, 2011; Cline-Cole

and Maconachie, 2016). The natural asset ‘urban tree’ finally also creates social capital, both as

tangible products are gifted to family, friends, and even strangers, and as a provider of shelter and

shade underneath its crown, which often serves as a pleasant meeting place for social interaction

and/or exchange. The respondents’ views suggest that they mostly perceive compound trees as

positive and desirable.

6.4.1.1 Trees for Subsistence and Own Domestic Use

Urban inhabitants obtain medicine, food, fuelwood, fodder, and materials for decoration, con-

struction or spiritual and religious activities from both public and private trees in places as varied

as South Africa (Kaoma and Shackleton, 2014; Shackleton et al., 2015), Nigeria (Kayode, 2010;

Agbelade et al., 2016), and US America (Poe et al., 2013; Synk et al., 2017). In Accra, questionnaire

respondents mentioned these same benefits and 92 out of 100 participants reported having used

products from public or private trees within the previous 12 months. While the previously listed

studies discuss and quantify tree product use or private households’ species compositions, quali-

tative and in-depth information on how humans and trees interact in urban settings remains lim-

ited. My household case study offers this detailed information on why and how people consume

tree products, use their benefits and protect, propagate, and maintain them. Not only did I con-

verse with household members of different gender and age but also observed their every-day and

tree-related activities over the course of two weeks. For further context, I also interviewed other

187
compound residents and associated individuals to create a more holistic understanding of the

complex tree-related networks between the city’s inhabitants.

Respondents from all four case study households reported that they used trees for medicinal

purposes or as part of their diet, acquired tree parts from both public and private trees, and listed

a wide range of specific applications for these products. Avocado seeds were used to reduce high

blood pressure, mango leaves for sores and wounds, Terminalia seeds for eye diseases, mango tree

bark for dental problems or roots, bark, and leaves to treat fever, typhoid, malaria, stomach or

joint pain. Some parents had treated their children with these herbal medicines, a common prac-

tice identified by Asase and Kadera (2014), who listed 42 medicines of these types, which parents

used to treat 20 child health conditions in Accra. In their study, the most frequently used tree

species included neem, citrus aurantiifolia (lime) and mahogany, while papaya and oil palm were in-

cluded but considered of lesser importance. Leaves were the most commonly used medicinal tree

product identified by Asase and Kadera. As the tree species in question grow on public lands

from which EPA does not sanction the harvesting of small quantities of leaves for personal use,

they can be collected by individuals who do not own trees and/or lack access to trees and tree

products on private land.

Family members, friends, neighbours, and even strangers widely exchange herbal medicine relat-

ed knowledge. Indeed, the interviewed herbalist confided that for simple treatments, he simply

explains to his clients what products to obtain and how to prepare and use them. Nonetheless,

some respondents used such traditional pharmacopoeia rarely, only twice in 30 years (HH 4, P 1,

P 5 and P 6). However, others valued tree medicine, taken against hypertension, for supposedly

fewer side effects like headaches or dizziness (HH 4, P 9), a common benefit identified by other

users of alternative medicine in Accra (Aziato and Antwi, 2016). Regardless of how widely

household members used such ‘traditional medicine’, they also visited physicians and took

188
‘pharmacy drugs’, especially if symptoms were unspecific (HH 2, P 1) or severe and people ap-

prehensive about treating themselves (HH 1, P 1). Commonly, younger individuals relied exclu-

sively on pharmacy drugs (HH 3, P 2), a tendency corroborated by the interviewed herbalist. His

typical clientele tends to be over 30 years old, while younger people prefer what he calls ‘Europe-

an’ medicine. While some individuals would combine tree and pharmacy medicine (HH 4, P 1),

the herbalist advises against this approach. One respondent (HH 3, P 2), however, described how

she follows pharmacy medicine with herbal one to ‘finish off’ the last remaining symptoms and

conclude the treatment, an approach widespread in the Greater Accra Region (Agyepong and

Manderson, 1994). Thus, using free and accessible plant materials for medicinal purposes can

help to sustain human capital while reducing monetary outlay, which market purchases would

otherwise require.

All four case study households lived in compounds with fruit and other food-producing trees,

although frequency and intensity of reported use/consumption varied widely. Some respondents

consumed non or few fruits either because they disliked particular food items (HH 2, P 2) or be-

cause they lacked the equipment and skills to harvest them, such as coconuts from tall palms

(HH 2, P 1; HH 4, P 2). Alternatively, products like oil palm fruits, an ingredient of palm nut

soup, were underutilised because they need to be processed before they can be cooked (HH 2,

HH 4). Alternatively, households readily consumed avocados or mangos, which require little to

no preparation (HH 1, HH 3). As trees often produce more fruits than an individual (HH 1) or

household (HH 3) can consume, respondents often shared left-overs with other compound resi-

dents (HH 1, HH 2, HH 3) and family members (HH 4). No participant reported any conflict

over access to and distribution or consumption of these resources as they perceived it as shared

(HH 1) and equitable (HH 4). While potentially reluctant to discuss quarrels with a stranger, they

could have preferred the socially accepted and non-committal response. However, their com-

189
ments suggested that they felt to have both unimpeded and sufficient access to the trees’ prod-

ucts to meet their immediate needs, as well as surpluses for sharing with other individuals.

Tree product distribution depends, among others, on the products themselves, household com-

position, and living arrangements. In HH 1, the respondent explained that any compound resi-

dent could freely collect mangos. In Mamobi (HH 2), where professional harvesters climbed the

coconut palms to obtain fruits for sale, John kept some fruits for his family and distributed fur-

ther fruits among the compound occupants. He usually gifts about three coconuts to individuals

and small households, while those with at least five members receive about ten. Exact numbers,

however, depend on how many ripe fruits were harvested. Household 4, also owning coconut

palms, equally distributes those fruits not for sale among its members, including the older chil-

dren. With palm nuts, though, the household head first processes them into concentrate before

sharing it, including with the wider family. Unlike in rural areas, where trees can be a central live-

lihood asset, urban residents with limited land tend to accommodate them around more essential

infrastructure like houses. While city trees provide fruits or medicine, they are not cash crops like

cocoa and sold, if at all, in small quantities (but compare to HH 2). When private landowners’

tree-related needs are met, they seem to willingly share ‘left-overs’ as the trees’ main values seem

to be subsistence use and their role in creating social capital (see Section 6.4.1.3).

Fruit-bearing and otherwise edible tree species were widespread in the four study households.

These species choices mirror findings from Kumasi (Nero et al., 2018a), where edible species like

oil palm, mango, avocado, coconut, and orange constituted the five most common species in ur-

ban home gardens. Elsewhere in Accra, Nero et al. (2018b) reported that 66 % of trees in home

gardens in six study neighbourhoods cutting across income zones were edible. Here, the most

common food included mango, oil palm, coconut, papaya, Indian almond, and Moringa oleifera, the

last one described as a leafy vegetable or food supplement in stews and soups (HH 3). House-

190
holds in Accra ration food as a coping strategy when financial shortfalls limit their purchasing

abilities (Maxwell et al., 2000). Having access to free fruits can thus positively impact human capi-

tal by contributing to a more diverse and healthier diet. As a proactive rather than reactive strate-

gy, obtaining free fruits can reduce spending in this area, freeing financial capital for expenditure

on other essential needs or investments. This appeared to be the logic driving two members of

HH 4 who decided to tend lemon trees to obviate the need for their mother to spend money

buying lemons in the market.

Though the potential effects of public and private food trees might seem limited at first, Accra’s

households spend on average 54.5 % of their households’ budget on food, the single largest item

(Maxwell et al., 2000). “Urban food expenditure patterns in Accra are characterized by heavy de-

pendence on purchased food commodities, including a substantial amount of street foods”

(Maxwell et al., 2000, p. 59). While richer households’ diets were diversified and included fruits,

vegetables, meat, fish, eggs, dairy, and non-alcoholic drinks, the diets of lower-income ones relied

primarily on staples like grains and cereals (Maxwell et al., 2000). Relatedly, 70 % of the sampled

low- to middle-income households in Accra were mildly to severely food insecure (Tuholske et

al., 2020). While household trees can mitigate some of these issues, access to public fruit trees

could benefit a larger share of the city’s residents and improve their diet, especially as fruits sell at

comparatively high market prices (Codjoe et al., 2016; Stevano et al., 2020).

The government departments responsible for managing public lands do not favour fruit-bearing

species. While EPA currently investigates the trees’ shade effects on increased road longevity and

reduced energy spending, DPG focuses on selecting species for shade and beautification. Though

naturally regenerating fruit trees grow on public lands and are, according to EPA, free to be har-

vested by anyone provided the plant remain intact, the government does not prioritise food trees

on public lands. While DPG does suggest fruit trees for use in private homes, this creates two

191
problems. First, it disadvantages residents of densely populated neighbourhoods with limited

outside or compound spaces for tree plantings. Second, the large amount of renters depends en-

tirely on their landowners’ goodwill, notably their attitudes towards tree planting, but also their

preferred species. Interviews with case study household associates living in compounds 1 and 2

revealed how these tenants often refrain from even considering planting trees. Not only do they

assume that they would be forbidden to plant, but they were convinced to move away before be-

ing able to reap the benefits of the investment that tree planting and tending would represent.

The time lag between tree planting and produced benefits, however, might only be a partial ex-

planation. The interviewed tenants had lived in their accommodation for ten and 20 years, re-

spectively, and the city’s severe housing shortage “limits access to both homeownership and rent-

al tenancy” (Arku et al., 2012, p. 3178). Land in and around Accra sells at a premium, and even if

renters should obtain a plot of their own, they would still have to invest in accommodation.

However, savings are often scarce as landlords and -ladies charge high advance rates, even for

low-quality housing. Additionally, tenants make heavy “advance lump-sum rental payments cov-

ering a period of about two to five years” (Arku et al., 2012, p. 3178) when entering a rental con-

tract. Even when landowners allow renters to plant (HH 1) or consider their tenants’ wishes,

such as HH 4, who removed a tree upon a renter’s request, tenants remain at the owner’s good-

will, themselves lacking control over the resource. When control over land is lacking, some indi-

viduals find alternatives, such as the tree-planting enthusiast whose landlord did not desire trees

in his compound. Instead, the tenant approached his friend/neighbour (HH 4, P 4) and obtained

permission to plant on his land. While such arrangements do not grant him ownership rights over

the trees he planted, he benefits from close to unlimited access or usufructuary rights to the

moringa trees for as long as the friendship between landowner and tree planter lasts.

192
Besides the four case study households and their members, most questionnaire respondents also

exhibited an interest in urban trees as food sources. While case study participants often preferred

planting fruit-bearing species if space were available (HH 1, 2, and 4), most household question-

naire respondents shared these views. Out of 100 respondents, 54 mentioned these potential ali-

mentary benefits in the introductory section, where I asked them to list all tree benefits they

could think of. In another survey section, I also asked if they had noticed particular trees, both in

their neighbourhoods or elsewhere in Accra, if they wanted to increase the city’s green cover or

own trees or land with trees. In each instance, respondents cited the food-producing potential of

urban trees (Table 29). However, not all respondents’ answers indicated why they noticed or

wanted trees as they simply referred to their wide range of benefits. Within a survey model, how-

ever, I was unable to ask follow-up questions. The numbers below, therefore, potentially under-

estimate people’s values attached to trees’ food-producing abilities. In essence, the government’s

(species) decision-making criteria seem to differ noticeably from how important many city resi-

dents perceive a range of further specific tree functions.

Table 29 Questionnaire respondents who mentioned food-producing tree benefits

No. of respondents

Noticed public trees in their neighbourhood due to the 18

fruits/food they produce

Noticed public trees elsewhere due to the fruits/food they pro- 08

duce

Named fruits/food as reason to want more public trees in Accra 08

Named fruits/food as reason why they would like to own land 14

with trees/trees

193
6.4.1.2 Trees as a Source of Income
In addition to contributing to subsistence consumption and use, trees can also function as a

source of income. Two households (HH 2, HH 4) had reported during the survey that they sell

tree products, and I further enquired about these activities. John’s compound (HH 2) in Mamobi

is large with ample space to plant trees, in this case mostly coconuts. As I reported in the intro-

ductory chapter, I was able to observe coconut harvesting in person. While I did not see a coco-

nut sale in HH 4, its members reported that sellers enquire about purchasing ripe fruits, signalling

their interest to harvest them. However, their compound is smaller than compound 2 with fewer

palms, possibly reducing how frequently coconut harvesters/sellers approach the household

about the fruits. John (HH 2) reported catering to regular coconut buyers and random opportun-

istic passers-by looking to acquire fruits for resale at least three times per month. In contrast, HH

4 sells much smaller amounts to a transient clientele monthly to bi-monthly. Other than the chil-

dren, all household members can negotiate with harvesters/traders at their own discretion, while

in John’s household, his wife Christine is no longer allowed to sell coconuts due to an incident

where coconut harvesters deceived her and claimed that they had already paid her husband.

Household 4 usually also retains some harvested coconuts for household consumption, com-

monly earmarking about three to four fruits per person. While both households initially charged

0.5 cedis/coconut, they doubled their prices when roadside coconut sellers increased theirs from

1 to 2 cedis. While John estimated earning around 150-200 cedis/coconut palm/year, contrib-

uting about 10 % to his annual income, HH 4 could not provide such estimates. In their house-

hold, all coconut sale income goes towards person 9, the head’s mother, for whom this is an ir-

regular but welcome source of cash. Household members support this distribution as all other

members have their own/other sources of income. While both households also sell/sold addi-

tional tree products, including coconut seedlings (HH 2) and mangos (HH 4), the obtained in-

come was negligible.

194
In both households (HH 2, HH 4), members sell trees and tree products as one of many income-

earning strategies. Nonetheless, coconut sales alone contribute about 10 % to John’s income,

suggesting that it plays a non-negligible role in subsidising individual and group livelihoods. Yet,

as with all livelihood activities, John (HH 2) constantly reviewed the trees’ role and their relative

importance. To obtain a tenth of individual livelihood income from coconut sales requires copi-

ous fruiting palms that, despite their slender growth habit, take up valuable compound land. As

the landlord already rents out accommodation, he knows about the higher income-earning poten-

tial of property rental over selling coconuts. Lacking the means to construct further houses for

tenants, he plans to lease out portions of his land for other people to build their houses on. As

accommodation in Accra is highly sought after, leasehold rent would provide him with a higher

and more secure income than coconut sales. However, re-structuring the compound would re-

quire felling a number of palms and other trees to clear land for accommodation. His expressed

care for trees and their multiple benefits would not deter him from replacing them with alterna-

tive land uses that generate more income.

Several households had cleared trees in the past to provide construction land or reduce some of

the trees’ disadvantages, like unwanted shade. Similar to HH 2, HH 1 also desired to increase its

rental accommodation, for which the household’s head would also consider removing compound

trees. Like John (HH 2), he also lacked the required large-up front investments and seemed to

have no concrete plans for raising such funds. However, recent Google Earth imagery indicates

that he not just felled the avocado that failed to fruit but that further ongoing developments alter

his compound. In 2018, the large umbrella tree outside the compound’s wall was heavily pruned

and then removed after. Images from 2020 show about one-third of the road-facing compound’s

side to be under construction. Due to the images’ relatively low quality, I cannot comment on the

exact changes taken place. These examples, however, illustrate landowners’ varied and complex

decision-making processes regarding the compounds’ land use and their associated vegetation.

195
They also suggest that, while cash returns can be limited, investing in single or a few trees can

constitute a promising strategy, particularly where financial resources for potentially more profit-

able investments are lacking. Furthermore, the case studies indicate that taste and preference,

combined with means and opportunity, are likely to have a more significant effect on private ur-

ban forest structure in Accra than (a lack of) community or popular awareness and knowledge, as

many official institutions and representatives insist.

6.4.1.3 Trees and Their Products as Social Capital

Trees and their benefits can affect a livelihood’s base in a third form. Both tangible products like

fruits, leaves and bark and intangible ones such as a tree crown’s shade can create and maintain

social capital, which, in turn, can potentially sustain or enhance other livelihood assets. Within a

livelihood approach, social capital describes networks, relationships, affiliations, or social claims,

which people utilise to pursue their selected livelihood strategies (Scoones, 1998). Social capital is

not just fundamental in sustainable livelihood frameworks but also in economics and other social

sciences, creating various definitions. These multiple interpretations have questioned the con-

cept’s meaningful use (Robison et al., 2002). Researchers also debated in how far social capital is

‘capital’ in the word’s economic sense (Arrow, 1999; Solow, 1999; Adler and Kwon, 2002) and

argued about its origins and development to clarify its meaning (Harriss and de Renzio, 1997;

Woolcock, 1998). After outlining some definitional basics, I return to those discussions relevant

to this research.

Commonly, individuals do not own social capital, which only comes into existence through the

relationship of at least two people, thus being a relational construct (Robison et al., 2002; Vemuri

et al., 2011). Definitions in political science, sociology, and anthropology vary but often describe

social capital as “the set of norms, networks, and organizations through which people gain access

196
to power and resources that are instrumental in enabling decision-making and policy formula-

tion” (Serageldin and Grootaert, 1999, p. 45). Others centre the sentiments that underlie these re-

lationships and the access they facilitate, seeing social capital as “a person’s or group’s sympathy

toward another person or group” (Robison et al., 2002, p. 19). While some even name it ‘good-

will’ – “the sympathy, trust, and forgiveness offered us by friends and acquaintances” (Adler and

Kwon 2002, p. 18) – it fundamentally “highlights the importance of nonmarket social interactions

in socioeconomic outcomes” (Quibria, 2003, p. 20). In what follows, I draw attention to the mul-

tiple ways in which trees and their benefits form part of social interactions in an urban setting,

highlighting what can be easily missed due to the unremarkable, everyday nature of many of these

activities.

Like other widely used concepts, researchers have criticised ‘social capital’ for its perceived short-

comings. Some object to the use of the economic term ‘capital’ though consider the underlying

ideas relevant (Solow, 1999). Others suggest that ‘social capacity’ (Smith and Kulynych, 2002) or

‘social networking’ (Taormina et al., 2012) could replace the economically connoted term ‘capital’

or simply recognise it as a metaphorical rather than a literal concept (Adler and Kwon, 2002; Up-

hoff, 1999). Researchers and practitioners have further criticised social capital’s often overlooked

capacity to both include and exclude individuals and groups, thus potentially limiting access to

material and non-material livelihood assets and choices for the excluded. As Portes (2014, p.

18407) has observed, such ‘excessive communitarianism’, or social capital’s anti-social tendency,

may extend to other areas with the potential to “produce harm instead of welfare” (Ostrom,

1999, p. 198).

Ben Fine, one of social capital’s most outspoken critics, questions whether “social capital add[s]

anything other than a new name to old analysis (how individual and social variables impact upon

the political process)”, wondering about the legitimacy of “such an approach to the political pro-

197
cess given its bias towards being insensitive to context, content and contingency?” (Fine, 2002,

no page number). Fine objects to the term glossing over and replacing a more nuanced under-

standing of the underlying complex issue of social stratification (according to, e.g. class, race, eth-

nicity, gender, or age), a blanket approach that contains a “remarkable absence of power and poli-

tics” (Navarro, 2002, p. 427). For Fine (2002), therefore, any capital, including physical one, is a

relationship rather than a resource and thus all forms of capital are inherently social, and he thus

argues for abolishing the concept. While I disagree with the idea of abolishing the concept of ‘so-

cial capital’, it requires being located within a broader context that recognises power dynamics’

relevancy. However, most common SL approaches seem, at least theoretically, sufficiently atten-

tive to such contextualisation, which is why I use the idea of urban tree-related ‘social capital’ in

this section.

As Fine suggests, ‘social capital’ can mask other variables of stratification and in Accra, politicians

and voters use indigenous exclusion/inclusion norms to “mobilize voters and select their political

candidates” (Paller, 2019, p. 43). Paller illustrates how clientelism, often involving members of

the same ethnic group, can guide city development. Though much Ga land (see Section 4.1.2)

had been alienated over the centuries, several ownership claims over now centrally located and

developmentally attractive land remain with members of this group. Paller’s ‘Everyday Politics’ il-

lustrates how “Ga dominate Accra Metropolitan Assembly (AMA)”, forming “a strong interest

group when it comes to development policy and urban development” (Paller, 2019, p. 159). An-

other study’s interview respondents, implying the influence of nepotism, noted that “[a]ll the de-

partmental heads come from the Ga areas” (Nathan, 2019, p. 257). When specific interest groups

dominate government organisations and their members live close to each other, such configura-

tions can affect intra-city resource allocation and prioritisation, directly and indirectly, to the cu-

mulative benefit of specific neighbourhoods. Regarding trees, group dominance may influence

investment patterns or the protection of and access to public urban trees and their benefits, both

198
reflecting and reinforcing opportunity inequalities between and within Accra’s neighbourhoods

and inhabitants.

In Accra, numerous interactions, events, and processes can create tree-related social capital. A

group of friends sits in a large tree’s shade and discusses religion, while someone else may ask

their neighbour for some neem leaves to make medicine. Meanwhile, a teenager offers his visiting

friend one of the coconuts that had just been harvested from a household palm. Many tree-

related interactions may seem unremarkable when regarded on their own but gain importance

when considered within the context of using one’s assets to pursue a sustainable livelihood strat-

egy. The list of ways in which tangible and intangible tree benefits sustain or enhance the social

capital base of Accra’s case study households and their members is extensive (Table 30).

Table 30 Tree-related social interactions

Ex. HH Trees and Tree Products as Social Capital or in Social Interaction

John’s compound is large, and several parts are not easily visible from his house.
In the past, some coconut sellers had convinced his wife that they had agreed
upon sale terms with John. However, they ended up defrauding them by not
paying for the fruits they had collected. The wife is no longer allowed to sell co-
conuts now, and as John has a lot of outside activities and spends little time at
1 2
home, their palms are often unsupervised. Fortunately, they have a neighbour
who keeps an eye on the sections of land that they cannot see, and she contacts
John whenever she observes coconut harvesters on his land to confirm that they
have been authorised. In return, John occasionally gifts her some compound co-
conuts, recognising her ongoing help.

When this Kokomlemle household harvests palm nuts, they divide them into
four equally sized piles. They keep the first one for the household (the head then
prepares soup for all) while they gift the remaining piles to a neighbour and two
2 4
distinct relatives living in Adenta, just north of Accra. Although these relatives
have only ever asked for some palm nuts once, they now receive gifts of palm
nuts after every harvest.

199
 The household’s landlord (Nima) owns several trees within his compound.
However, one of them also grows adjacent to the external wall, just outside his
compound. The large Terminalia tree casts shade on both his compound and
nearby public lands. When I visited the household for data collection, I often
observed a group of four to six middle-aged men sitting on chairs underneath its
3 1 crown. One of them has owned a nearby shop for about ten years, during which
time the others have regularly come to visit him. They know that the tree be-
longs to the case study participant, but about once every year, they cut back
some of the lower branches hanging over the road, posing a risk to passers-by.
As they are all Muslims, they often discuss religious issues, read the Quran to-
gether and, during Ramadan, break their fast under the tree.

Friends sometimes visit the adult woman in this other Nima household to
whom she offers free avocados from the compound’s tree (when ripe). Some
visitors enjoy the taste so much that they also want to plant seedlings in their
4 3 homes. She grows the plants from seeds in empty water sachets, regularly water-
ing them with the kitchen’s waste water. She does not charge for the seedlings as
she incurs little monetary costs in producing them and because they are destined
for friends.

About ten years ago, person 1 built a large wooden bench that he put just out-
side his compound walls where he felt the air was fresh and temperatures cooler.
He positioned the bench in the mango tree’s shade and started to sit there in his
leisure time. After a while, people from his neighbourhood joined him, some
5 1 even bringing their own chairs and benches. They first small-talked but later re-
placed the topics with discussions about politics as their friendship developed.
As most people came every afternoon, he constructed a second bench for them
to sit on. Sometimes, occasional acquaintances or strangers also stop by to rest
or participate in ongoing conversations.

As John’s wife had recently given birth, several visitors brought gifts of toiletries
and baby products like soap or diapers. To some of these visitors, they gifted
6 2
coconuts in return, and while some asked explicitly whether they could obtain a
coconut, many did not but were still offered household surpluses nonetheless.

200
These examples demonstrate urban trees’ multiple ‘social capital’ functions. Tree products (Ex-

amples 2 and 6), whole trees (Example 4), and the ‘socio-ecological’ space created by the trees’

shade (Examples 3 and 5) facilitate interactions between family members, friends, neighbours,

and even strangers. This type of access to both information and the circulation channels (Exam-

ple 1) can mediate livelihood decision-making (Coleman, 1988; Lin, 2000). As Coleman (1988, p.

104) argues, “acquisition of information is costly. At a minimum, it requires attention, which is

always in scarce supply”. Indeed, John and his wife’s vigilant neighbour relieves the owners of the

need to guard the palms and, in the process, frees them to pursue other, potentially more produc-

tive, livelihood activities. In several examples (1, 2, 4, 6), participants gifted tree products or seed-

lings to people ranging from casual acquaintances to family members. Some case study partici-

pants linked these presents to items they themselves had previously received (Example 6). Oth-

ers, however, had not reciprocated (yet), or at least nothing in ‘kind’. As Sobel (2002, p. 151)

notes, drawing on Marcel Mauss’ widely-known work on gift-giving, “offering a gift creates both

an obligation to accept and an obligation to reciprocate”. Creating and maintaining these relation-

ships between individuals and groups represents, in essence, the idea of ‘social capital’.

Tree-related social capital in Accra is not just produced by exchanging and gifting tangible prod-

ucts but by how the trees create ‘interaction spaces’. Due to the city’s climate, tree canopy shade

provides possibly the most favoured benefit. Seventy-eight of 100 survey respondents mentioned

it when asked why they would desire more trees. People congregating underneath trees when

spending time outside is a common sight, and they network, establish and strengthen connections

(Examples 3 and 5) or share intelligence and gossip. Residents in temperate places like Chicago

(Coley et al., 1997), the Netherlands (Maas et al., 2009), and Baltimore (Holtan et al., 2015) inter-

acted in similar ways, and their social activities increased due to “neighborhood common spaces

that are made more hospitable because of tree canopy” (Holtan et al., 2015, p. 518). Researchers

have documented the importance of social relationships in facilitating access to livelihood oppor-

201
tunities through knowledge disseminated via people’s networks in many urban locations (Beall

1995; Beall 2001), including in Ghana (Hanson, 2005; Oberhauser and Hanson, 2007; Yeboah,

2010). Tree-related social capital seems integral to many city inhabitants, improving access to ma-

terial and non-material livelihood capitals that may positively affect the overall asset base or live-

lihood potential of individuals and groups.

6.4.2 Tree Propagation, Care and Management

In this section, I discuss compound tree care and use, both within and between households. Par-

ticipants identified a range of tree-related management and care activities, including seed collec-

tion and planting, seedling/tree protection, watering, fertilising, pest control, and pruning. As tree

propagation, care and maintenance activities are closely interlinked, I do not attempt to discuss

them separately but loosely arrange them by how these topics interact with each other.

6.4.2.1 Growing Trees

All four households were or had been involved in producing tree seedlings, albeit for different

reasons: own use (HH 4); own use and gifting (HH 1); sale (HH 2); and gifting (HH 3). Within

households, enthusiasm for and participation in tree management varied between members and

included the heads (HH 1 and 2), the head’s wife (HH 3), or the head’s children (HH 4). Thus of

Household 4’s nine members, only the teenager and young adult produced and tended trees.

Growing lemons from the seeds of market-bought fruits, they hope to eventually replace the reg-

ular supply their mother purchases from the market. Creatively, they planted the seeds in various

available containers, including small plastic bags or buckets, but also old tyres and even discarded

shoes. While they share responsibility for tending to the seeds and planting them out in the com-

pound, the younger brother takes full responsibility, notably for daily watering, when his sibling is

202
away at university. As other household members lack interest in tree propagating, including their

mother as household head, the brothers pursue their interest independently of family supervision

or input. When some of the lemon seeds germinated and reached a certain height, they also de-

cided, without consulting others, where to plant the seedlings out in the compound.

In the two other multiple-member study households (HH 2 and 3), interest in raising trees also

varied. In household 3, where person 2 grows avocados from seed, she started her nursery opera-

tion when visiting friends enjoyed the fruits and asked for seedlings to plant in their homes. As

she had lacked experience, she refined her technique through observing the operations of road-

side tree seedling vendors, who also recycle old plastic bags as seedling containers. As she has no

further space to plant trees in her compound, she produces the seedlings entirely to order and in

small numbers. John in Mamobi (HH 2) also produced tree seedlings for which he built a nursery

planter. He, as well, did not grow the coconut seedlings for compound planting, even though he

had previously planted some of his seedlings on his land. As he already earmarked land for leas-

ing, he did not want to cover more of his compound with trees, thus producing all seedlings for

sale. However, trade has been less than brisk lately, possibly because he had recently increased his

asking price by a third, from 15 cedis to 20 cedis per seedling, a significant increase and a relative-

ly high asking price.

The first household’s head also reported lacking further planting space in his compound and the

desire to re-develop it, which trees might impede. Nonetheless, he does produce some seedlings

in a planter in front of his house. He had grown mango, avocado and soursop seedlings for a

while but more recently also started experimenting with ‘blackberry’ (velvet tamarind). Although

his acquaintances do not solely visit him to obtain seedlings, they appreciate the offer of free

plants to take home and grow in their compounds. When people visit, they do not come for spe-

cific tree species but take what the man is currently growing. Usually, they hear about him propa-

203
gating tees at social events like funerals and weddings. Recipients of his free seedlings sometimes

express their appreciation by offering him small sums of money (about five cedis) ‘to buy beer’,

which he considers compensation for, and recognition of, the time he invests in producing, ra-

ther than as payment for the plant material.

6.4.2.2 Maintaining Trees

Case study households also seek to improve all-around tree health and vitality, for example water-

ing, fertilising, or treating for pests, to maximise overall productivity. When maggots infested HH

3’s avocado tree some ten years earlier, compound residents worried about the animals spreading

to their houses. However, they only monitored the situation, waiting and hoping for the maggots

to disappear on their own. Person 2 reported that they were not concerned about the tree’s

health itself, and therefore did not act, involving potentially unnecessary outlays of time and

money. But when a chemical-selling friend visited the compound sometime after, he recom-

mended and supplied a product that exterminated the maggots within a few days of spraying the

tree. The compound residents, however, were significantly more proactive when they noticed that

the tree had produced fewer and smaller avocados than previously. After discussing amongst

themselves, the residents agreed to attempt to restore the tree’s full productivity. Their plan

would have the men loosen the soil around the trunk and add compost, while the women would

collect free cow dung from a cattle-rearer in the neighbourhood. The men would then add the

mixed compost and dung to the aerated and fertilised soil around the tree roots. While the plan

had not been executed when I collected the data, person 2 was hopeful that the planned interven-

tion would eventually take place.

Other case study participants also debated whether to fertilise (some of) their trees or spend time

and money otherwise. In household one, the head’s brother, who used to live in the compound,

204
chemically fertilised the mango tree in 2015, and the respondent noticed an increase in yield in

the subsequent year. Since then, however, fruiting vigour again declined, and the tree had not

produced any fruits during the harvest season prior to data collection. While the household head

attributes the mango fruits’ absence to a lack of pruning, he also believes that fertilising the tree

would help. However, he has delayed treatment, missing the required funds to pay for chemical

fertilisers. Though he enjoyed eating the mangos, he considers the financial investment unwar-

ranted. A large mango tree also grows in household four, in Kokomlemle, and the household’s

head reported that increasingly fewer fruits had been growing on the plant over recent years. She

attributes this decline to the tree’s advanced age and concluded that any treatment would be

pointless as the mango was in the process of dying. As the tree exhibited few leaves and consid-

erable branch dieback, her assessment appears accurate.

Especially in warmer climates, trees need to be regularly watered during their establishment

phase. In household one, the large mango tree occasionally received left-over household water,

but the landlord frequently watered the plants in front of his home’s entrance, including the seed-

lings and saplings. Additionally, he sometimes fertilised them with ash from incinerated house-

hold and compound waste. While the avocado growing lady from the third household also used

left-over water, she mentioned that some seedlings had died when she left town for several days,

and nobody paid attention to watering the plants. The two younger men in household 4 were also

well aware of the need to regularly water their lemon seedlings. However, all case study partici-

pants, in addition to stakeholder interviewees, affirmed that established trees no longer require

this watering level and ‘will just grow on their own’. In how far climate-change induced changes

in rainfall patterns may affect these assumptions should be investigated. While researchers have

developed a climate–species–matrix for Central European urban trees (Roloff et al., 2009),

drought resistance matters, especially in (sub)tropical climates. Thus selecting suitable species and

planting sites is essential to create a sustainable urban forest in Accra.

205
Pruning constitutes another essential urban tree maintenance procedure. In household one, a

carpenter, who is the landlord’s friend, pruned the mango about three to four times per year until

he died five years ago. Ever since, the owner has resorted to removing only small and very low-

hanging branches, reachable without any tools. His approach corresponds to John’s (HH 2), who

removes only dead/dying coconut fronds within his reach while the professional coconut sellers

tidy up the tall palms when harvesting the fruits. In the third household, compound residents lim-

it pruning to avocado branches that have grown too close to the ground and block people’s way.

The side on which the branches overgrew determines the responsible household (man) to cut

them off. In HH 4, coconut harvesters also prune leaves as part of their service and person 4, the

head’s brother, used to cut avocado tree branches when they interfered with house-

hold/compound life. His nephew (P 2) then picked up the branches and burned them in the

compound’s corner. These examples illustrate how maintaining household trees depends on, e.g.

physical ability and age, available equipment, and the (gendered) distribution of chores. In gen-

eral, access to and control over livelihood and other resources; the ability, knowledge and willing-

ness to perform the required tasks; perceived outcomes; and potential returns of invested time,

money, knowledge and non-material claims determine urban tree-related choices and interven-

tions.

6.5 Chapter Conclusion

Researchers have investigated trees and forest products’ roles in rural society, economy, nature

and livelihoods (Fisher, 2004; Paumgarten, 2005; Shackleton et al., 2007; Babulo et al., 2009;

Kamanga et al., 2009; Rayamajhi et al., 2012). Recently, related to a perceived growing urbanisa-

tion of poverty, interest in how city trees might or might not perform similar livelihood roles has

increased (Kaoma and Shackelton, 2014; McLain et al., 2014). Previous work, however, has fo-

cused on quantifying tree benefits’ contributions to livelihoods. My case study, though, uses a

206
qualitative approach to shed light on often mundane and overlooked urban human–tree interac-

tions. Accra residents consume tree products for subsistence or sell them for cash, two of the

major livelihood functions of rural trees (Angelsen and Wunder, 2003; Babulo et al., 2009). Addi-

tionally, these public and private urban trees create and maintain social capital. I neither want to

participate in exaggerating “claims for the universal efficacy of social capital” (Ostrom, 1999, p.

173) nor suggest it as a universal remedy (Durlauf, 1999). Instead, I propose to increase focus on

urban trees’ multiple ‘social capital’ function(s). As the concept’s critics rightly argue, this requires

a framework that recognises that access to capital “can be accounted for largely by structural con-

straints” (Lin, 2000, p. 793; also Ferragina and Arrigoni, 2017). Using insights from SL thinking

as part of a broad political ecology approach should direct attention to questions of this kind.

When compound trees are absent, renters and individuals that lack tree access through relation-

ships or networks depend on public trees to obtain their benefits. However, naturally regenerat-

ing seedlings aside, government departments choose both species and planting sites for public

trees, and their urban forestry agendas may differ from how ordinary Accra residents prioritise

tree use and care. Furthermore, all main state actors involved in Accra’s urban forestry (EPA,

DPG, and TCPD) aim to increase green cover and protect existing public trees from damage and

felling. While large-scale felling might be more commonly associated with companies/developers

rather than individual city inhabitants, people excluded from private urban forestry resources or

depending on space for their livelihood activities (e.g. roadside sellers) could (involuntarily) en-

croach upon the existing vegetation. State and parastatal agencies need to consider the full range

of trees’ social and ecological functions when assigning roles in regional planning and landscape

management to urban forests. This can only be achieved in a meaningful way when taking into

account the needs, views, and preferences of Accra’s rich, not-so-poor, poor, and very poor.

207
My second case study’s contribution lies in how it explicitly examines intra-household and com-

pound dynamics. While tree planting, care and use in relation to socio-economic variables have

been widely examined, the existing studies tend to aggregate at a neighbourhood-scale (Summit

and McPherson, 1998; Vogt et al., 2015). Though some researchers looked at finer scales (Shakeel

and Conway, 2014; Conway, 2016), their works combine at the household level and largely disre-

gard structures and dynamics within. Following Rocheleau and Edmunds (1997), who questioned

the oversimplified concept of ‘household’, I suggest investigating ‘gender’ or ‘age’ as further fac-

tors since the case study has demonstrated how engagement and interest in tree-related activities

vary between household members. Together, this and the previous chapter now form the basis

on which to read the following one that examines the potential effects of current and proposed

urban forestry-related activities on the general public’s ability to incorporate urban trees into a

livelihood strategy effectively.

208
 7 URBAN FORESTRY INTERVENTIONS: THEIR LIKELY SOCIO-
ENVIRONMENTAL EFFECTS

“And we will use our permitting system to increase the cover […].”

Mr Fiati, EPA

~~~~~

“[T]he purpose is several-fold: […] we want to use it [tree inventory] to deter-

mine what we call a green index so that we can evaluate how our permitting
process is impacting on tree cover within the city and to be able to tell whether
it’s leading to depletion or improving tree cover. So that is the primary reason
why we are taking this inventory.”

Mr Fiati, EPA

In this chapter, I discuss five current and proposed urban forestry-related events and processes.

The design and implementation of these activities illustrate how the government actors’ focus on

national to global scale tree benefits limits their functions at the individual to local scale. I suggest

that the involved state actors should rethink and rebalance their management priorities and rec-

ognise the trees’ direct livelihood role in household (re)production and in mitigating some nega-

tive effects of poverty. Centring smaller scales and focusing on social justice could also eventually

increase benefits at higher scales through a ‘trickle-up’ effect. I first discuss the current develop-

ment of Accra’s tree stock, outlining why I refrained from using iTree Eco’s ‘forecast model’. Af-

ter that, I analyse five interventions: EPA’s and DPG’s forest inventories and management plans,

TCPD’s attempt to re-evaluate planning schemes for which I combine a theoretical discussion of

participation and the new planning guidelines with information obtained from interviews, and re-

cent changes to tree-related laws. Further interventions include tree planting activities in Accra

209
and FC’s plan to turn Achimota urban forest into the Accra Eco-Park. While all discussed activi-

ties directly include urban trees, developments in seemingly unrelated sectors should not be ig-

nored. Anticipating these possible in-between sector interactions is vital though considerably

more difficult, requiring not just insight but also speculation in a dynamic and multifactorial con-

text. To conclude, I consider how state actors could more effectively incorporate urban trees into

broader poverty reduction strategies. Discussing the current and proposed activities, I adopt a

non-neutral perspective that follows current models of sustainable urban forestry, which include

environmental justice and equity, as well as citizen involvement as a core feature (Leff, 2016).

7.1 The Natural Resource Base – Development of Tree Stock

The iTree Eco tool contains a ‘forecast’ component that allows the user to run models which

predict how the urban forest structure may develop under various scenarios. In the default set-

ting, ‘forecast’ uses a 30 year period, 150 frost-free days per year and base annual mortality rates

related to tree health ratings, each manually adjustable. While running different scenarios seems

appealing, considering the limited available data on Accra’s urban forest structure, I decided

against this as the model lacks the input necessary to obtain a meaningful output. Below, I discuss

three reasons why I refrained from using ‘forecast’, as basing management decisions on unrealis-

tic and unreliable numbers may create more harm than benefit in Accra.

The first problem revolves around how the model excludes natural regeneration when predicting

future tree numbers. While users can manually add it as a form of ‘tree planting’, research on re-

generation remains limited and showed a significant variance of ‘planted vs naturally regrown’ ra-

tios between cities and land-use types within the city. As the only existing reference study re-

searched North American towns (Nowak, 2012), different climatic conditions, among other fac-

tors, limit its applicability to Accra. Second, the model currently includes few ‘extreme events’.

210
Though users can add storms and pest/disease outbreaks (Morgenroth and Armstrong, 2012;

Leksungnoen et al., 2017), adding further disturbances like earthquakes or flooding would in-

crease the model’s value for Accra. The final reason why I decided against using this application

to forecast developments in Accra is its use of mortality rates. The default values’ origins remain

somewhat unclear but seem to originate from a single study of maple street trees in one US

American city conducted by Nowak in 1986 (Roman et al., 2016; Steenberg et al., 2017). Studies

on mortality rates are both limited in number and mainly devoted to North American cities in

climate zones markedly different from that of Accra (Hilbert et al., 2019). Furthermore, mortality

rates vary widely, and an array of biophysical and human factors further influence them (Hilbert

et al., 2019). As the forecast’s accuracy depends on these numbers, the model is unlikely to pro-

vide useful information and might instead misguide planning and management decision-making.

Simulating how an urban forest might develop is nonetheless important as structural changes af-

fect the location, quality, and quantity of the trees’ ecosystem services. In place of a complex

model, I then combine information from interviews, the survey, case study, and both iTree Can-

opy and Eco with the existing literature. I thereby illustrate an overall trend of forest structure

development rather than predict exact numbers. Green space in Accra is declining (Stow et al.,

2013; Owusu, 2018) and as trees constitute one component of such spaces, tree cover, in particu-

lar, is also likely to be declining. Interview and case study respondents supported this hypothesis,

citing urbanisation and related construction and infrastructure development, lack of tree mainte-

nance and animal browsing as possible reasons. Very few people felt that the tree population was

stable, and not a single participant thought that numbers were increasing. Unless proven other-

wise, the following section assumes that current and proposed urban forest interventions are

happening within the context of declining tree cover in Accra.

211
7.2 Current and Proposed Interventions and Their Possible Impacts
Several factors influence urban forestry in Accra or could do so in the future. I selected five cur-

rent or planned events and processes to analyse how they might affect people’s ability to integrate

trees, and their use and management, into livelihood activities. These five topics emerged during

interviews and represent a range of relevant themes, perspectives and concerns, and cover inter-

ventions and initiatives at various stages of the planning and management cycle. In this section, I

first introduce EPA’s and DPG’s data collection plans and discuss possible problems. I then dis-

cuss TCPD’s attempts to re-evaluate their planning schemes, paying special attention to the ques-

tion of participation. The third topic considers recent changes to tree laws, while the fourth one

presents tree-related environmental initiatives. Finally, I look at the Forestry Commission’s plan

to turn Achimota Forest into Accra Eco-Park.

7.2.1 Urban Forest Inventories and Management Plans

Urban forest planning and management decisions require baseline information for initial assess-

ments and continuous monitoring, including data on the current tree stock. This information is

indispensable to the proper functioning of organisations responsible for urban forests. In Accra,

previous research investigated green space at the neighbourhood and city-scale (Stow et al., 2013;

Owusu, 2018), but no inventory offers tree-level data such as species, height, location or DBH.

During interviews, both EPA and DPG expressed needing this type of data. While EPA was al-

ready running a pilot inventory, DPG was hoping to collect such information in the near future.

This information would enable the department to produce a plan for re-developing the city land-

scape and creating an elaborate and proactive tree maintenance scheme, including regular pruning

and irrigation during the establishment phase and tree cover increase through afforestation. Be-

low I discuss these activities in more detail to highlight the objectives and questions they raise be-

fore suggesting how their possible impacts on human–tree interactions in the city may look like.

212
Although DPG aims to promote landscape beautification and tree maintenance, its personnel feel

unable to do so efficiently, lacking baseline information to inform a routine maintenance scheme.

Currently, DPG reacts to problems in an ad-hoc manner, such as sending units when trees have

sustained storm damage and need pruning or felling to protect property and people. However,

the interviewee mentioned that “the department is on the verge of developing a strategic plan to

take a look at the entire city landscape maybe for the next five years” (Mr Awidi, DPG). EPA, on

the other hand, was already conducting a pilot study that involved two employees geocoding each

tree’s location in Accra’s ‘Ministries’ area. During the interview, the respondent mentioned three

main purposes. First, they aim to establish a green index to evaluate the permit system’s impact

on tree cover. Second, the agency wants to evaluate how trees affect air quality and, in turn, pub-

lic health (respiratory diseases). Lastly, they want to investigate whether the presence of roadside

trees increases road longevity through the trees’ shade and associated lower temperatures. Both

interviewees recognised the need to collect further data to facilitate the development of their re-

spective management plans.

While the interviewees explicitly mentioned some problems related to their activities, I inferred

further ones from their statements. Though DPG knows about the need for a proactive mainte-

nance approach, it lacks resources for most of its tasks; thus the realisation of their proposed

plans seems unlikely. The department’s parlous funding situation is common knowledge in Accra,

and other interviewees called them “under-resourced” (Mr Yeboah, TCPD), as well as “there on

paper” and “kind of dormant” (Interview Y4GG). Additionally, journalists also share these sen-

timents and “wonder whether the Department of Parks And Gardens still exists” (Daily Guide,

2017), calling it “neglected” (Daily Graphic, 2013a) and “dead or dying” (Daily Graphic, 2017a).

The DPG interviewee repeatedly supported these statements, stressing their financial and human

resource constraints. While not explicitly framed as a problem, EPA also faces issues with their

ongoing project. Though two employees currently geocode trees, they do so for only part of the

213
working week. As they tag every tree in the pilot area, they will unlikely finish this time-

consuming process soon, further impeded as the agency shares equipment between projects,

making the GPS devices unavailable at times. The lack of resources to initiate or complete such

data collection exercises promptly hinders the production of a city-wide management plan.

Designing an urban forest management plan for Accra not only requires certain types of currently

unavailable information, but I suggest that EPA’s approach also limits the usability of the data

they currently acquire. According to the EPA interviewee, the employees only collect the trees’

GPS locations, as they had not received the requisite training to collect data on other variables

and characteristics like species composition. Even though EPA claims to know what kind of

public trees are growing where, variables like species, DBH, and tree health/condition constitute

the minimum requirements of a credible tree inventory (Leff, 2016). Currently, EPA spends

scarce resources on collecting data that fail to meet the basic standards for meaningful data analy-

sis, as provided ecosystem services depend on tree species, size, and state/health. Since establish-

ing the GPS location requires approaching each tree, employees should collect further data, alt-

hough this would indeed require further training for this expanded responsibility. If this addition-

al information cannot be obtained in the current inventory, subsequent visits should be arranged

to collect this data as the extra expense seems manageable and well worth the improved output.

Alternatively, EPA could enlist individuals with a horticultural education, trained by the Youth

Employment Agency as part of their ‘Youth in Greening Ghana’ module, or collaborate with

DPG who already have suitably trained personnel but seemed unaware of EPA’s pilot project.

To assess the possible effects that these activities might have on people’s ability to utilise trees as

part of a livelihood strategy, it is worth considering what EPA and DPG propose to use the data

for and what this might signify in terms of likely species and site selection. EPA focuses on the

trees’ health impact as they absorb pollutants and how tree shading, and reduced temperatures,

214
could increase the road surface’s longevity. These regulating ecosystem services tend to require a

scalar approach at the neighbourhood level or above, with trees established along major roads

and in larger open spaces suitable for clustered planting. Such a pattern is unlikely to significantly

increase cover in the densely populated neighbourhoods, which already tend to be devoid of

trees. Thus, EPA’s plan may have little or no positive impact on the availability and accessibility

of services produced at smaller scales for the benefit of individuals, households or compounds.

Though the EPA interviewee mentioned the need to increase canopy cover in less green areas, he

seemed sceptical about achieving this objective, identifying high rates of illegal felling due to a

perceived low local valuation of trees as a major cause of concern.

The Department of Parks and Gardens, for its part, selects species based on their shading and

beautification potential. This, again, reflects limited ambition: they prioritise a restricted set of

benefits while refraining from planting fruit-bearing species on public lands. Additionally, re-

source shortage has restricted their activities to areas in the department’s vicinity, where they

plant trees along major roads and on government lands. The DPG respondent also noted a de-

partmental objective to specifically green those areas which currently lack trees, benefitting from

the government’s decentralisation policy, because “officers who are going to be positioned at

those less green areas they can work together with the assemblies” (Mr Awidi, DPG). However,

as observed by TCPD, the structures in these places are weak and accommodating staff a major

problem. Neither EPA’s nor DPG’s plans may directly reduce canopy cover in the ‘poorer’

neighbourhoods. However, focusing on areas surrounding DPG and neighbourhoods with better

structures can nonetheless increase intra-city differences when green cover in ‘richer’ neighbour-

hoods increases further. Additionally, the proximity to trees determines the ability to benefit from

its shade or beautification potential. The proposed activities appear to maintain or reinforce both

people’s dependence on privately owned trees in the more densely populated areas and their ina-

bility to obtain access to the services of trees they might not own.

215
7.2.2 Re-Evaluation of Planning Schemes and Participation
In Accra, many planning schemes are not realised on the ground. While TCPD designates at least

17 % of Accra’s land area for public open spaces, these areas are rarely converted into fully func-

tioning green spaces. Although ‘the green’ constitutes an integral part of the department’s spatial

plans, in practice, buildings (‘the brown’) dominate the landscape. To assess this incongruence,

TCPD intends to review its schemes. The respondent identified a preference for the conventional

land-use planning approach as responsible for the mismatch between theory and practice. In this

standard model, a professional planner/expert identifies problems and devises appropriate solu-

tions without consulting the public (Mahjabeen et al., 2009). According to the interviewee, this

approach has led to residents being unaware of scheme provisions, making them (sometimes un-

knowingly) encroach on public green spaces and/or fail to report such unlawful encroachment.

However, a different form of urban planning which emerged as long ago as the 1960s recognised

that a “successful democratic planning process, like a successful democratic government, must al-

low for the representation of the interests and identities of its subcommunities” (Godschalk and

Mills, 1966, p. 86; see also Healey, 1997; Margerum, 2002). By adopting a more participatory type

of planning, which the interviewee called a ‘paradigm shift’, TCPD hopes to increase conformity

between plans and their implementation in Accra.

Over time, community participation and its synonyms turned into a core element of a diverse set

of policies and projects in and outside forestry. Governmental and non-governmental actors alike

incorporated participation in sectors ranging from education to health, conservation, agriculture,

and sanitation (Abbot, 2013). However, involving the community might not necessarily be the

most appropriate or useful approach in all cases. Researchers and practitioners have questioned

the efficiency of such projects as they tend to require more resources, both time and money, than

their non-participatory counterparts (Margerum, 2002). If donors, practitioners, or ‘beneficiaries’

measure success based on the speed of service delivery, time-consuming participation and seem-

216
ingly little visible and tangible progress can deter the project’s continuation (Botes and Rensburg,

2000). Though resources saved at the implementation and evaluation stages may offset higher ini-

tial costs, extensively involving people is often seen as “not cost-effective, because participation

in practice is always a slow and uncertain process” (Botes and Rensburg, 2000, p. 50). While I

work within a research paradigm that centres participation, its possible added value, in all con-

texts, depends heavily on how this approach is conceived and implemented, or indeed why it is

adopted.

7.2.2.1 Participation’s Context: Spatial Planning in Ghana

Participation’s ubiquity “reflect[s] the ideological range of interpretations of development and dif-

ferent approaches to planning” (Moser, 1989, p. 81). Following Moser, this raises four questions

– why participation, when participation, whose participation, how participation – which offer a

framework to explore what aspects influence participatory project design and implementation.

After outlining the history of physical/spatial planning in Ghana to provide the required context,

I discuss each of these four questions and then suggest how a more participatory approach might

affect people’s ability to utilise the city’s trees. In this section, I combine a theoretical discussion

of participation with information obtained from the TCPD interviewee and the country’s most

recent spatial planning guidelines.

As remarked upon earlier, physical planning started under colonial rule (see Section 4.4.1). The

promulgation of the Town and Country Planning Ordinance (CAP 84) in 1945 served as the ba-

sis on which independent Ghana performed “a series of plan-making projects at the national level

and for the country’s major towns and cities” (Acheampong, 2019, p. 30). Acheampong notes

that long-term planning was largely absent between 1966 and the late 1980s when mostly military

takeovers followed one another in short succession. In 1992, decentralisation introduced a form

217
of planning that “almost eliminated spatial planning from policy discourse and rendered the insti-

tutions with competences in the field weak to carry out their mandate effectively” (Acheampong,

2019, pp. 30-31). Only in 2007 did the government implement reforms pertaining to the Land

Use Planning and Management Project (LUPMP), which rekindled spatial planning in the coun-

try.

Though considering Ghana’s land-use planning in detail falls outside the scope of this section

(see Acheampong, 2019), I want to briefly note three aspects. First, early planning systems, such

as CAP 84, limited stakeholder ‘participation’, restricting it to informing citizens about declared

planning areas in newspapers and on public notice boards. Only part of the population, the one

able to read in English and understand technical documents, could access this information. Se-

cond, the shift towards decentralised governance in the late 1980s aimed to increase/facilitate

participation in development processes at the local level. However, legal and institutional restruc-

turing processes meant that “[s]patial planning was essentially neglected at the national and re-

gional levels of political administration” (Acheampong, 2019, p. 46). While planning at the local

level generated schemes for cities and neighbourhoods, they were unconnected to the larger-scale

socio-economic situation, resulting in two disconnected types of planning (Acheampong, 2019).

Though increasing participation constituted a key element of the restructuring process, it took the

form of “consultations through representation” (Acheampong, 2019, p. 159) as “heads of tech-

nical departments […], elected officials representing various communities within the assemblies

and other unelected opinion leaders would be invited on an ad hoc basis” (Acheampong, 2019, p.

158). Third, LUPMP in 2007 aimed to reform spatial planning in the country, explicitly introduc-

ing a three-tier system “based on consultative and participatory approaches” (Acheampong 2019,

p. 48). As part of this process, ‘The New Spatial Planning Model Guidelines’ (Ministry of Envi-

ronment, Science and Technology – Town and Country Planning Department, 2011) were devel-

oped and supposedly guide today’s planning approach.

218
7.2.2.2 Why Participation?
Researchers and practitioners have criticised traditional forms of project design and implementa-

tion for various reasons. Public interests can be underrepresented or not included at all, no in-

formation exchanged between professionals and the public, or strong interest groups can domi-

nate proceedings (Moote et al., 1997). Involving the community, in turn, can increase accountabil-

ity, with the public taking ownership of the project (Botes and Rensburg, 2000). Furthermore,

participation may cause the adoption of more relevant policy and incorporate local knowledge

and preferences (Rydin and Pennington, 2000). Why a participatory approach might be favoured

affects a project’s design and, relating back to Moser, also determines who can participate, how

and when. These reasons can be categorised using either Paul’s (1987) five objectives of commu-

nity involvement or by distinguishing between participation as a means or an end (Nelson and

Wright, 1995). Analysing World Bank projects, Paul identified five reasons for adopting a partici-

patory approach, including improved efficiency, cost-sharing, effectiveness, building beneficiary

capacity and empowerment. In the means/end division, Paul’s first three objectives coincide with

participation as a means, with the latter two associated with participation as an end (Moser, 1989).

These distinctions are important as they relate to the community’s level of involvement and the

question of ‘power’ and its transfer.

A project designed to involve the community to share costs will likely differ from one with em-

powerment as a core objective. While in some cases, participation includes a transfer of power

away from professionals that used to decide on their own, others consider asking for the public’s

opinions as sufficient (Habraken, 1986). Besides the five objectives of community involvement,

Paul (1987) also identified four levels of intensity. These range from information sharing and

consultation, which according to Habraken, lack a transfer of power, to the public making deci-

sions and initiating action, which involves power being transferred to the community. Similar ide-

as are also found in Arnstein’s (1969) famous ladder of citizen participation. The many possible

219
combinations of objectives and intensities illustrate the complexities of discussing ‘why participa-

tion’. Imprecisions around terminology can be consciously utilised to protect hidden agendas as

“participation is often used by governments as a means of legitimizing the political system and as

a form of social control” (Botes and Rensburg, 2000, p. 45). Or, as Cooke and Kothari (2001, p.

14) argue, the ‘tyrannical potential’ of community participation relates to its proponents often be-

ing “naive about the complexities of power and power relations.”

The Department of Town and Country Planning aims to shift towards a participatory planning

approach to increase congruency between schemes and the situation on the ground. In their new

guidelines, they argue that “plans so often fail in their realization because key stakeholders are not

adequately involved in the plan preparation process” (Ministry of Environment, Science and

Technology – Town and Country Planning Department, 2011, p. 3). I argue, however, that their

ideas may fail to empower the community as they see participation as a means rather than an end,

focusing on building beneficiary capacity to share costs. As I built my assessment on the inter-

viewee’s comments, I provide a more extensive excerpt below to help support my evaluation.

The TCPD respondent stated:

Respondent: […] the conventional approach to spatial planning, we

prepare the schemes with direct cause to the beneficiar-

ies, especially the community members, we don't consult

them, and they don't really know the provisions in the

scheme, so if I am living in a community I don't even

know that this area has been earmarked as maybe a public

park.

Respondent: So somebody will start developing there because I don't

know, I cannot report or stop the person but if we all

prepare the scheme and we all know the provisions in this

220
 scheme, and we all approve it, then, of course, everybody

can keep an eye on what is going on within the community.

Respondent: [N]ow we are trying to review the schemes, and we want to

do it in a participatory manner whereby we will go to the

communities, present to them the existing situation and

then the scheme that was prepared for the area because if

you go through Accra now, the existing situation is real-

ly at variance with the approved scheme that you have, so

we have to adapt the existing situation but we will have

to let the community know what we are going to do there

and they will give their inputs so when it is approved

then, of course, we make also copies as billboards within

the communities. So if it's an open space earmarked green

the people will know that ‘oh, this place’ and we have to

also [inaudible: one word] a signpost that this is a pub-

lic park, keep off. So the people themselves, cause we

don't have the personnel to always go around the city,

but the communities themselves can really monitor when we

involve them. And that is the paradigm shift, that is

what we want to do now.

(Mr Yeboah, DTCP)

It seems that residents are primarily involved to increase adherence to designated land uses and

have people report illegal developments. Through this form of ‘participation’, the department can

extend control where its own mechanisms are insufficient due to limited resources. While they

seek people’s inputs, they do so after preparing schemes, suggesting little actual participation in

the early phases. Rather, they attempt to “sell a prepared plan” (Godschalk and Mills, 1966, p.

86), a form of pseudo-involvement or tokenism on Arnstein’s ladder where participation is

221
“submerged in a culture of powerful political structures and institutions (Mahjabeen et al., 2009,

p. 46).

The department’s plans do not seem to fundamentally question issues of power since people’s

knowledges and lived experiences are not considered sufficiently. Instead, TCPD simply “let[s]

the community know what we are going to do there” (Mr Yeboah, TCPD). If the department on-

ly informs the community on new planning schemes and then imposes them, like previous ones,

the public’s interest to refrain from encroaching and to monitor and report misuse is unlikely to

increase. Furthermore, the guideline displays a patronising tone, stating that participation “affords

actors the opportunity to learn, and hence, own the process and break and transform past habits

in order to achieve the desired objectives of the plan” (Ministry of Environment, Science and

Technology – Town and Country Planning Department, 2011, p. 43). Such a view may put peo-

ple off from engaging with or accepting any resulting plan. Additionally, one might consider the

incongruence between plans and reality to be a symptom of current city planning – and urban

forest structure as part of it – not meeting the population’s needs. Business as usual, disregarding

people’s inputs in the planning and design stages, combined with enforcement issues, may cause

problems as it can prevent the utilisation of the urban trees’ full potential both for individu-

al/group livelihood purposes as well as in response to the global climate crisis.

7.2.2.3 When Participation?

‘Why participation’ closely relates to ‘when participation’, which refers to a project’s different

stages, such as planning, financing, implementing, monitoring or evaluating. While the communi-

ty may participate in any or all stages, the above-discussed objectives and intensities determine

the exact configuration. Based on the interview statements, TCPD seems to involve Accra inhab-

itants at a limited number of stages and mostly at the implementation/monitoring phase to share

222
costs through unpaid labour. Indeed, the planning guidelines state that involving stakeholders

“serves the purpose of ensuring that the plan reflects the stakeholders needs but also that the

stakeholders are aware of their obligation to invest in the realisation of the plan” (Ministry of En-

vironment, Science and Technology – Town and Country Planning Department, 2011, p. 37).

Such wording suggests limited community involvement in the planning or design phase beyond

being at the receiving end of information sharing.

The new guidelines recognise three types of plans, the spatial development frameworks, relevant

at the national, regional and district level, while the other two types, structure plans and local

plans, are prepared at the city to neighbourhood level. The two latter types require stakeholder

consultation at various stages, e.g. when presenting data analysis, after developing a draft plan

and at the final plan and report stage (Ministry of Environment, Science and Technology – Town

and Country Planning Department, 2011). If TCPD were to work with city occupants on a plan

to efficiently and sustainably incorporate green spaces into the cityscape, the department would

have already decided that such a plan should exist in the first place, as well as on its purpose and

extent. This limits the opportunity for more fundamental and potentially creative ways of re-

imagining urban human–tree coexistence from a bottom-up perspective. Though I do not want

to exclude the possibility that the approach might provide for community decision-making power

at various stages, the interview responses indicate otherwise.

7.2.2.4 Whose Participation

Close examination of community participation requires that I attend as much to the concept of

‘participation’ as to that of ‘community’. Involving the public in planning and decision-making

processes now often constitutes an essential component in striving towards democracy, justice,

and sustainability, both in urban planning and other areas (Mahjabeen et al., 2009, p. 46). Howev-

223
er, communities and their knowledge base are inhomogeneous and include a diverse range of

people, whose internal power relationships might exclude women, the elderly, or minority groups

more generally (Mosse, 2001). Thus, who participates affects whose knowledge becomes visible.

If design and implementation disregard community heterogeneity, only sections of such a com-

munity may be empowered or associated costs shared inequitably.

Besides intra-community power relationships, we need to consider other aspects when asking

‘whose participation’. In almost all situations, apart from when users initiate action (Paul, 1987), a

government department or a non-governmental organisation acts as a ‘provider’ of participation.

In these situations, the provider occupies a powerful position, able to affect the programme. The

provider can then construct participants as ‘beneficiaries’, ‘clients’, ‘users’ or ‘citizens’, which “in-

fluences what people are perceived to be able to contribute or entitled to know or decide, as well

as the perceived obligations of those who seek to involve them” (Cornwall, 2002, p. 8). Addition-

ally, the providing/initiating party may also define and create the space for participation which is

pervaded by power relations (Cornwall, 2002). Thus, Gaventa (2004, p. 35) argues that we need

to ask ourselves “how they [spaces for participation] were created, and in whose interest and with

what terms of engagement”.

As I lack information on how TCPD plans to inform people of these participatory approaches, I

cannot comment on how they attend to the issue of community heterogeneity. Nonetheless, I

suggest that the process will likely happen in ‘invited spaces’ (Cornwall, 2002; Gaventa, 2004), as

the department itself proposes to involve the potentially affected and/or supposedly benefitting

citizens. Even though spaces for participation offer possible places of resistance (Cornwall,

2002), ‘inviting’ participation “may have the effect of neutralising energy for engagement outside

them [spaces] and may render other spaces for voice illegitimate” (Cornwall, 2002, p. 8). I suggest

that TCPD’s plan – limited community involvement in the planning phase combined with ex-

224
pected strong presence at the implementation and monitoring stage – merely maintains the status

quo. A status currently characterised by high levels of community non-compliance within the

context of low governmental enforcement capacity.

7.2.2.5 How Participation

Lastly, ‘how participation’ considers participation’s practicalities, such as the location of meetings,

their size, time or type. These aspects closely relate to ‘whose participation’, as certain times of

the day or weekdays might exclude specific groups, while large-scale meetings might favour some

voices over others. In re-developing their planning schemes, TCPD wants to focus on low con-

formity areas. These places tend to be located in neighbourhoods of lower socio-economic status,

where people often work long hours in the informal sector and have limited time to attend meet-

ings. The guidelines envision “a departure from what in the past has been mere consultation and

information provision which often came as ‘fait accompli’” towards a situation where

“[p]articipation may involve information sharing, consultation and collaboration” (Ministry of

Environment, Science and Technology – Town and Country Planning Department, 2011, p. 43).

The planning guidelines list affected community residents as a group mandatorily to be involved

in local plans. However, there is no indication that efficient mechanisms address issues of availa-

bility or intra-community power relationships, creating a danger of ‘participation by representa-

tion’ for certain sub-groups. The currently envisioned approach seems to lack any incentive for

participation and appears unlikely to increase green cover and its protection in the city. With pub-

lic urban trees, the department further faces the problem of ‘free riding’ as non-participating citi-

zens can also benefit from collective goods like pollution reduction or shade, without incurring

any costs, such as time spent, for getting involved (Mincey et al., 2013).

225
7.2.3 Tree Laws and Tree Law Changes
In Accra, several laws and regulations13,14,15 guide tree planting, removal and maintenance. Revi-

sions to some of these laws in December 2015 changed the requirements for obtaining a permit

to fell or prune a tree. While the fees for cutting a tree amounted to two cedis for domestic pur-

poses and three cedis for commercial purposes, it now costs 80 (domestic) and 150 (commercial)

cedis to cut a large tree (70/100 cedis for medium-sized and 60/80 cedis for small trees, see Fees

and Charges Amendment Instrument, 2015 (LI 2228), p. 1516). Failure to obtain a permit and be-

ing caught felling illegally now results in punitive fines, officially designated as ‘administrative

charges’, that were absent before the laws’ revisions. These ‘charges for non-compliance’ amount

to 240/450 cedis for large, 210/300 cedis for medium and 180/240 cedis for small trees (Fees

and Charges Amendment Instrument, 2015 (LI 2228), p. 31). According to the EPA respondent,

this change simplified urban forest management, as previously people would simply cut trees not

fearing any punishment, and “those who applied paid and those who didn’t apply just fell the

tree, no charges” (Mr Fiati, EPA). However, the recent modifications likely affect people’s use of

trees due to the increased costs, the onerous process of obtaining a permit, and the attached con-

ditions. I now discuss each of these aspects in more detail, focusing on privately owned trees on

residential plots.

EPA claims that the permits’ increased price and related fines for illegal cutting also increased

their ability to manage Accra’s urban forest. However, the 40-fold rise (from two to 80 cedis) is

excessively high for part of the city’s population. As values of self-reported income often suffer

13Accra Metropolitan Assembly (Environmental Protection) Bye-Laws, 2017 – No. 8 to 11, available at:
https://ama.gov.gh/doc/bye-laws.pdf (Accessed: 12.02.2021)
14 Environmental Assessment Regulation, 1999, LI 1652, available at:

http://epa.gov.gh/ghanalex/acts/Acts/ENVIRONMENTAL%20ASSESSMENT%20REGULATION,1999.pdf
(Accessed: 12.02.2021)
15 Fees and Charges (Amendment) Instrument, 2015, LI 2228, available at:

http://www.epa.gov.gh/epa/sites/default/files/downloads/publications/Approved%20Fees%20and%20Charges_
LI%202228__0.pdf (Accessed: 12.02.2021)
16 The interviewee only mentioned 80 and 150 cedis and did not refer to tree size. I found this additional information

when consulting the LI 2228 document.

226
from measurement errors (Moore et al., 2000), previous research produced varying values for Ac-

cra. Tuholske et al. (2018, p. 6) report an average monthly household income of 890 cedis in low-

and middle-income neighbourhoods of Accra, similar to a mean monthly income of about 612

cedis17 in Accra’s Jamestown neighbourhood (Tutu et al., 2019, p. 64). These values differ signifi-

cantly from an average annual household income in the Greater Accra region of 1529 cedis,

translating to only 127 cedis per month (Rachmadyanto et al., 2016, p. 248). The region, though,

includes rural areas, which might partly explain the lower figure. Even for the highest reported

income, 80 cedis for a tree felling permit equals about 10 % of the average monthly income.

Therefore, residents in low- and middle-income neighbourhoods are unlikely to be able or willing

to pay for a license.

Private individuals can plant a tree on their property at little or no cost, but the expectation of

potentially having to pay 80 cedis for felling it might deter people from growing trees. Though it

might take decades before the fee accrues, tree pests or damage to surrounding structures could

necessitate early removal. Additionally, landowners become inflexible as to how they use their

land, as previously planted trees can impede infrastructural developments on their plots, an issue

brought up by several case study participants. Indeed, permit fees could easily reach four-digit

numbers, as in the case of John (HH 2), who owns more than 40 trees and plans to clear land and

lease it out for people to construct houses. Furthermore, payment is also due for pruning because

“[y]ou have taken off some of the branches and the services that those branches are to provide

for the general good of the public has been taken away by one person. So you must compensate”

(Mr Fiati, EPA). This maintenance operation will become necessary at certain points, most likely

earlier and more frequently in densely populated and built-up areas. Assuming that people know

about these regulations and want to adhere to them, the anticipated spending for permits might

17 The study reported a value of US$154 which was equivalent to 612 cedis when Tutu et al. conducted their research.

227
prevent cash-poor individuals from growing trees on their premises. In practice, however, illegal

pruning or felling on private property might go unnoticed due to limited official surveillance.

Contrary to expectations, the EPA interviewee reported an increase in permit applications which

he attributed to the penalties for illegal felling. Moreover, he stated that “most people prefer to

plant because there are direct benefits apart from beautification. [T]here is shading, and the aes-

thetics, and the wood. Yeah, most people would like some trees in their home” (Mr Fiati, EPA).

Though not representative, case study participants’ responses suggest that this might be a mis-

conception. One participant would refrain from further planting if these laws were properly en-

forced as the permit’s cost exceeds the monetary income she could realise from the tree products’

subsistence use or commercial/market exchange (HH 4). She questioned the system’s utility and

efficacy, given that nobody had ever tried to enforce the law in her neighbourhood. Another par-

ticipant shared her sentiments, being dismissive of the regulations and convinced they were too

impractical to be enforceable (HH 2). The new system’s effects might emerge as both more com-

plex than assumed and not entirely positive for long-term green cover sustainability. The revised

laws increase EPA’s tree management options and demonstrate the value the agency attaches to

urban trees. However, a paradoxical effect might occur if the law deters tree planting on private

property, thus negatively affecting people’s ability to include trees in their livelihood activities.

Even though the building code stipulates that each plot must contain at least three trees, the law’s

enforcement seems insufficient to counterbalance financial (and administrative) barriers to tree

planting for individuals of (very) limited economic means.

Apart from the high fees, the process of obtaining a permit could further discourage applicants.

As described by the EPA interviewee, it is time-consuming and includes several restrictions and

conditions. The procedure starts with an application form that requires information on the loca-

tion, type and number of trees to be felled and the justification for the proposed felling. An EPA

228
employee then inspects the plot within two days “to establish whether the reason for deciding to

remove the tree is valid” (Mr Fiati, EPA). If deemed invalid, the inspector might recommend

pruning instead of felling. After the site visit, the employee prepares a report which passes

through several levels of vetting until the executive director either grants or denies permission.

While the EPA interviewee considers the application form easy to complete, it might represent a

significant barrier for illiterate people faced with the option of investing monetary and other re-

sources in obtaining a permit or cutting illegally. Additionally, applicants need to travel often long

distances to the agency to acquire the relevant papers, requiring further expenses of time and

money on top of the actual felling fees. These processes effectively marginalise and undermine

the capacity of low-income earning individuals, those working long hours and/or residing in dis-

tant neighbourhoods, from productively engaging with a public institution.

EPA exerts a high level of control and decision-making power at all stages of the permit process.

As summarised by the EPA interviewee, “we do an assessment, we check the site, we check the

kind of trees, and if it is agreeable with us, then we issue the permit” (Mr Fiati, EPA). Even while

recognising that the agency’s mission to protect the environment entirely justifiably leads to re-

fusal of some permit requests, the process nonetheless disadvantages sections of Accra’s popula-

tion. When considering an application, agency officials also consider the surrounding area’s tree

density and “[i]f it’s very sparse we are less likely to grant it unless we think that there is an immi-

nent threat to life or property” (Mr Fiati, EPA). Thus, EPA’s approach reduces the chance of ob-

taining a permit in comparatively less green areas, which tend to correlate with lower socio-

economic neighbourhoods. Thereby, they introduce uncertainty over whether tree owners can le-

gally prune, pollard or completely remove their trees in the future. Furthermore, applicants have

to pledge to plant two trees for each one cut, de facto increasing the number of trees on the land.

In densely populated neighbourhoods, these conditions might further deter tree planting. While

EPA allows replacement planting on public spaces or school grounds, these arrangements require

229
further resource commitments on the part of residents to work with DPG to locate suitable sites

or places.

The recent changes have over-complicated the permit process and created a system that disad-

vantages and deters certain population groups from planting trees. As the chapter’s introductory

quotes indicate, EPA itself seems unsure of the system’s impact as they conduct their current tree

inventory to analyse if the new permit process depletes or improves tree cover. To simplify the

process and increase the likelihood of it having a positive effect on the city’s green while minimis-

ing the costs to the city’s poor, I suggest three changes. First, application forms should be obtain-

able at readily accessible locations across the city. Second, the required information should be

easy to assemble and report. Third, prices should be tiered to avoid the risk of impoverishing sec-

tions of the population or forcing them to break the law for lack of resources. Additionally, EPA

might need more personnel to process applications thoroughly and promptly to avoid delays.

This example illustrates through which mechanisms a government agency can affect land-use de-

cisions on private property. Focusing on the intended outcome – in this case, to protect and in-

crease tree cover on private land – while failing to create a feasible process to achieve this objec-

tive, however, might have unintended consequences. In the worst case, actual effects might be

opposite to the intended ones, such as a decreasing or stagnating tree cover on private land rather

than increasing planting activities.

7.2.4 Tree Planting and Other Environmental Initiatives

Large-scale tree planting exercises have recently been announced for Ghanaian cities, including

Accra. ‘The Green Republic Project’, a youth-driven grassroots effort, aims to plant 20 million

trees between 2018 and 2028. Tamale’s Metropolitan Chief Executive announced plans to add 20

million trees to the city within a single year in 2015 (Daily Graphic, 2015). The Kumasi Metropol-

230
itan Assembly aims to restore its garden city label through planting programmes, and AMA wants

to plant 100,000 trees a year, starting in 2018. Nationwide, several other projects include tree

plantings, such as the ‘Greening Ghana’ initiative, a joint effort between FC and YEA, which

aims to employ 15,000 people in its ‘Youth in Agriculture and Afforestation Module’. While simi-

lar projects elsewhere have succeeded in planting the stated number of trees (e.g. in New York),

others, like Los Angeles or Kumasi, have fallen short of their targets (Daily Graphic, 2017b). The

number of trees to be planted is often absurdly high, especially considering the lack of personnel

or volunteers. Furthermore, such programmes require suitable land and resources to maintain the

seedlings. How such projects are designed and implemented affects the types, location, quality

and quantity of produced ecosystem services. In this section, I illustrate why existing initiatives

may fail to produce individual scale, livelihood relevant benefits, specifically for low-income resi-

dents. As such, these projects often miss the opportunity to utilise urban trees as part of a pov-

erty reduction approach.

Over the years, Accra has participated in several initiatives incorporating trees and green spaces.

The BreatheLife Accra Project forms part of the Urban Health Initiative supported by the World

Health Organization. It focuses on improving air quality, including via tree planting in selected

schools. The EPA interviewee also discussed this tree benefit, as the agency’s tree inventory data

will be used to analyse the relationship between respiratory diseases and canopy cover. But Accra

is also a member of the global C40 Cities climate leadership group, which addresses climate

change and the 100 Resilient Cities (100 RC) programme that aims to increase cities’ resilience to

physical, economic, and social shocks and stresses. The TCPD interviewee stated:

Respondent: I think there is a paradigm shift now worldwide, and we

are talking about globalisation. So Accra is not an iso-

lated place, we are battling with resilience, resilient

cities, 100 Resilient Cities, and Accra is part of it.

231
 C40 is also being some projects […], and greenery is an

important aspect of it. […] the green is very, very im-

portant to us and we want to really re-vitalise Accra

with respect to greenery, and I have told you the mayor's

vision is to make sure that all the open spaces we devel-

op it into greenery and also trying to plant and nurture

trees along the streets in the communities. So that is

the vision. So greenery is very, very important.

(Mr Yeboah, DTCP)

In 2014, Accra joined the 100 RC network and developed the Sustainable Greening and Beautifi-

cation of Accra’ project (SGBA), a component of its resilience strategy. The network receives fi-

nancial support from the Rockefeller Foundation and offers its members financial and logistical

guidance to establish a Chief Resilience Officer position, support for developing a resilience

strategy, and access to partners that can help develop and implement such a strategy. In 2019,

Accra released its strategy, which also aims to contribute to achieving the UN Sustainable Devel-

opment Goals that Ghana committed to. At the national level, these global objectives form part

of several policies, including the President’s Coordinated Programme of Economic and Social

Development Policies (2017-2024) (Akufo-Addo, 2017) or the Medium-Term National Devel-

opment Policy Framework (2018-2021). The latter specifically mentions the promotion of urban

forestry as one strategy within the focus area of ‘Climate Variability and Change’ (National De-

velopment Planning Commission, 2017, p. 189). By reducing greenhouse gases, the country’s na-

tional policy aims to make human settlements inclusive, safe, resilient and sustainable. While the

documents name FC and EPA as collaborating agencies, they explicitly mention neither DPG

nor TCPD, though they might also participate, as their ministries, MLGRD and MESTI, are part

of the process.

232
As part of the resilience strategy, SGBA contributes to the broader aim to “[d]esign and adapt in-

frastructure to maximise co-benefits and simultaneously address Accra’s flooding, waste, sanita-

tion, and climate challenges” (AMA and 100 Resilient Cities, 2019, p. 38). The trees in green

spaces and parks are supposed to reduce flooding and improve air quality (AMA and 100 Resili-

ent Cities, 2019, p. 60), two problems that caused the premature death of several thousand inhab-

itants (Asumadu-Sarkodie et al., 2015; World Health Organization, 2016). This activity’s timeline

is defined as ‘short-term’, signifying a length of 1–2 years, and its status is described as ‘on-going’

(p. 60). Further information about the project’s specifications remains limited, but groups like the

Bank of Africa, Ghana Commercial Bank, Ecobank, or the Church of Jesus Christ of Latter-day

Saints have pledged their support and ‘adopted’ certain spaces as part of their (corporate) social

responsibility. The Metropolitan Assembly’s webpage18 describes the project as “aimed at beauti-

fying the metropolis”, a cultural ecosystem service, with no reference to provisioning services.

Accra’s network memberships, interview excerpts, and national policies demonstrate an orienta-

tion towards the national and international scale, prioritising related tree benefits rather than fo-

cusing on, or at least including, provisioning services for individuals to benefit from directly.

Due to urban forestry’s multipurpose nature, individuals could also benefit personally and direct-

ly from trees planted for climate change mitigation, provided they are accessible and grow in rea-

sonable proximity. However, SGBA’s geographical focus prohibits this option as it aims to

“transform the Independence Avenue through the Ridge enclave to the Ako Adjei Interchange

and the environs of the Accra Sports Stadium”19. While these are pilot areas20, they suggest a fo-

cus on major roads and on already comparatively green areas like North Ridge, with the necessary

18 https://ama.gov.gh/news-
details.php?n=cnMybzU5MnFuc3A2NDIxMDc3cDEycDJycW5xcnJxNXBzMXI5cHI4MA%3D%3D (Accessed:
10.02.2021)
19 https://ama.gov.gh/news-

details.php?n=cnMybzU5MnFuc3A2NDIxMDc3cDEycDJycW5xcnJxNXBzMXI5cHI4MA%3D%3D (Accessed:
10.02.2021)
20 https://www.todaygh.com/ama-launches-greening-accra-project/ (Accessed: 19.01.2020)

233
public spaces available. As I demonstrated in Chapter 4, levels of canopy cover and socio-

economic status tend to correlate, implying that the residents who could benefit from additional

trees will find it difficult to access the new plantings. Additionally, the project also includes

“community durbars, house-to-house sensitisation on protection of environment, promotion of

green space development and the improved awareness”21. These activities relate to my argument

in Chapter 5, where I suggested that several government actors misclassify the public’s different

priorities as a supposed lack of awareness and knowledge.

The ambitious SGBA project ties in with the mayor’s and president’s aim to green and clean Ac-

cra. However, its ability to make available and accessible tree benefits of direct livelihood im-

portance seems limited. In this project, DPG is responsible for maintaining the green spaces,

which it considers an opportunity to re-vitalise the department. Considering its resource situation,

I question the sustainability of these activities as they already lacked financial resources, employ-

ees, tools, and equipment to adequately maintain the city’s existing tree cover. Unless the gov-

ernment increases the department’s budget, it seems unlikely that ongoing and proposed inter-

ventions will create lasting effects. Furthermore, DPG’s inefficient tree propagation method,

which does not produce enough seedlings for larger-scale afforestation schemes, has so far also

impeded their desire to green Accra. Pressured by the project’s short timeline, DPG might focus

only on the selected areas, further decreasing their ability to green the city more equitably.

7.2.5 Achimota Forest Turns Eco-Park

Achimota Forest, managed by the Forestry Commission, is located in the city’s north and covers

approximately 3.6 km2. In the early 1920s, the British established the forest as a fuelwood planta-

tion, at that time located well outside Accra’s boundary. Following a series of species tests, they

21 ibid

234
restricted nursery propagation and plantation stocking to neem and Cassia siamea (PRAAD, ADM

5/1/84), as other species failed to establish due to scarce and irregular rainfall (PRAAD, ADM

5/1/80). Over the years, the forest lost its role as a peri-urban firewood plantation and became

incorporated into an expanding capital city. While Achimota previously covered almost 5 km2,

the area has gradually but steadily lost land to encroachment, a fate shared with the Kumasi For-

est Reserve (Ahmed and Puppim de Oliveira, 2017). Today, individuals and groups use the forest

for recreation and as a spiritual retreat. FC reports that Achimota hosts 20,000 visitors annually,

generating a revenue of US$60,000 from entrance fees. Recently, there has been much debate

about FC’s plans to transform Achimota Forest into something called the Accra Eco-Park. As

FC refused to be interviewed on the topic, and official information remains unavailable, I base

this section on newspaper reports and interview statements from third parties. To suggest how

the project might affect people’s ability to use the forest, I first discuss the issue of participation

before outlining how the proposed changes might also change the forest’s user base.

Over the past decades, Accra’s urbanisation caused rapid and unplanned developments, resulting

in encroachment on the forested area. In 2009 FC consulted stakeholders like the Ministry for

Lands and Natural Resources to investigate how to protect Achimota against further conversion

and degradation (Daily Graphic, 2016a). FC’s Chief Executive stated that there was a “need to

undertake innovative ecotourism development as a means of safeguarding the ecological integrity

of the forest” ((Daily Graphic, 2014c). They further stated that they “looked for success stories in

ecotourism development elsewhere” and shared their findings “with major stakeholders in a con-

sultative process that was transparent” (Daily Graphic, 2014c). Besides protecting the forest, FC

expects to create hundreds of new jobs and further revenue from tourism. In November 2011,

the Forestry Commission supposedly obtained the public’s support through a consultative pro-

cess (Daily Graphic, 2016b). The Commission agreed with a private partner, Aikan Capital Lim-

ited, on a ten-year lease, with the Ghanaian company responsible for designing, building, and op-

235
erating the park. Total project costs amount to US$1.2 billion, of which the private partner will

cover 45 % and FC the remaining 55 %. While the project’s construction, estimated to take about

seven years, supposedly took off with a sod-cutting ceremony in August 2016 (Daily Graphic,

2016c), I noticed no real progress when collecting data in 2017.

7.2.5.1 Lack of Participation

The Forestry Commission’s Chief Executive stated in a newspaper article that they had consulted

major stakeholders before drawing final design plans (Daily Graphic, 2014c). Nothing indicates,

however, that the Commission included members of the public. One interviewee, the Achimota

visitor/pastor, confirmed that FC excluded the prayer groups, which regularly frequent the forest,

from decision-making. He described how he visited Achimota one day and saw a sign at the en-

trance announcing the area’s transformation into the Accra Eco-Park. Neither the respondent

nor anyone he knows protested as they consider themselves lacking the power to do anything

against FC’s plans. Trying to resist the project, a group of four individuals took FC, Lands Com-

mission and EPA to court, as they feared the project would harm the environment, increasing

health risks and other hazards, which the city’s inhabitants would have to contend with (Daily

Graphic, 2016d). At the same time, in 2016, a newspaper article argued how “the people of Accra

will not be given the opportunity to comment, get involved, participate in, understand and bene-

fit from a project that is going to affect their lives forever” (Daily Graphic, 2016e). Based on the

available information, I conclude that FC excluded at least one major stakeholder, the general

public, from participating and that the process was intransparent.

Though several government agencies like EPA, the Tourist Board, or the Ministry of Lands and

Natural Resources support the project, others oppose the plans. The Minister of Environment,

Science, Technology and Innovation, for example, does not think the park’s supposed environ-

236
mental benefits justify its creation (Ghanaian Times, 2017). Concerns have also been voiced by

the National Development Planning Commission, which warns that the transformation may

cause increased flooding as the planned structures replace tree cover, thereby reducing the soil’s

absorptive capacity and generating greater run-off (Daily Graphic, 2016f; Daily Graphic, 2016g).

The TCPD interviewee shared these sentiments, strongly opposing the project. He described

how FC, in 2016, contacted the department and asked for a change of land-use demarcation,

which TCPD refused. While the respondent stated that FC could not develop the forest unless

his department approves of the land-use changes, several interests compete, and it remains un-

clear which interest and, therefore, vision will prevail. The latest online newspaper article on the

Eco-Park project specified that FC was currently reviewing the terms for the construction with

the investor (Daily Graphic, 2018).

7.2.5.2 Change of Users – Change of Use

Achimota forest currently houses many prayer camps, tree-enclosed areas of varying sizes where

FC has cleared the undergrowth, providing space to religious groups to hold their services. Eco-

Park plans also include a place for prayer, described as a spiritual enclave with worshipping areas,

a pavilion, and a grotto. However, plans further outline an amusement park, a cultural village, a

walkway, a drive safari (Daily Graphic, 2013b), eco-lodges, camping sites, a restaurant, and a but-

terfly sanctuary (Daily Graphic, 2013c). Considering that the available area amounts to only 3.6

km2, it remains unclear how these facilities can be accommodated, especially since one million

trees are also earmarked for planting in the park (Daily Graphic, 2018). In 2013, Papa Kwesi

Nduom, the founder of the Progressive People’s Party, the country’s third-largest but number-

wise insignificant one, urged the then-president not to allow the forest’s conversion to proceed.

He mentioned the lack of government support for other parks in the country and questioned the

237
proposed development’s feasibility. In his opinion, the development would turn the forest into an

amusement park, defeating its purpose as a greenbelt (The Chronicle, 2013a).

Besides potentially reducing tree cover, the costly Eco-Park would likely increase the current en-

trance fee of 0.5 cedis. According to the interviewed pastor, entrance previously cost 0.2 cedis

and the higher price already excluded some people. A further (drastic) rise will change the visitor

profile from a predominantly local population to a more international eco-tourism clientele with

higher purchasing power, even though FC stresses the importance of local patronage.22 In the

view of FC’s Chief Executive Officer, “families who are often stranded for lack of place for real

relaxation will soon have something to smile about” (Daily Graphic, 2013c), and the Deputy

Minister for Lands and Natural Resources added that the park would offer Accra residents an-

other weekend or holiday destination besides the beaches (The Chronicle, 2013b). The available

evidence, however, suggests that the average resident might find it difficult to afford entrance to

the forest in the future, excluding cash-poor individuals from benefitting from the cultural ser-

vices Achimota Forest currently provides. The forest’s decreasing area seems to offer a conven-

ient explanation as to why FC needs to transform it into an eco-tourism park. The interviewee

from GYEM shared this view, opining “it was just an idea to kind of commercialise the place and

make money out of the place, urbanise the place and make money out of the place” (Mr

Commey, GYEM).

7.3 Facilitating Tree Planting

The five examples of Accra’s urban forestry-related events and processes I have discussed above

illustrate how these kinds of interventions may reduce people’s ability to use trees within a liveli-

hood context. While access to privately owned trees improves an individual’s resource base,

22 https://www.fcghana.org/page.php?page=283&section=28&typ=1&subs=286 (Accessed 19.01.2020)

238
changing laws can reduce the natural capital’s usability. Besides, international institutions and

processes also affect national developments, such as the Rockefeller Foundation discontinuing its

funding to 100 RC in 2019. Accra’s membership in this and similar groups, EPA’s research inter-

ests, DPG’s species selection criteria and FC’s Eco-Park plans indicate a preference for tree bene-

fits with limited or no direct livelihood function at the micro or local scale. Instead, most gov-

ernment actors focus on the national to global scale, trying to benefit the Ghanaian economy by

commercialising urban trees, utilising them to reduce expenditures on road repairs, energy and

water infrastructure, and as a component of climate resilience strategies. Many households in

densely populated areas rely exclusively on public tree benefits and are therefore especially sus-

ceptible to, and limited by, government actors’ decisions. In such cases, there is no direct individ-

ual, household or compound control over the tree resource, whose availability and accessibility

may change unexpectedly. Thus, tree benefits might act as safety nets in times of shocks or

stresses but cannot efficiently become a part of long-term sustainable livelihood strategies capa-

ble of reducing poverty in a lasting manner.

The existing structures, processes, and interventions prioritise the larger scales and consider fur-

ther tree functions as an afterthought, if at all. To achieve Accra’s greening objectives, city resi-

dents’ support and compliance will be crucial. However, the current situation offers few incen-

tives for private tree planting and the protection of public trees and green spaces. Urban forestry,

multipurpose in nature, can address several issues simultaneously, although planning and man-

agement decisions determine species and site choices, and therefore the prioritised scale of activi-

ty. While trade-offs will always be necessary, a larger variety of configurations could be included

in the decision-making process. This entails accepting that incorporating individual scale benefits

might result in less than ideal urban forest structures for globally important ecosystem services,

such as climate regulating ones (Heynen, 2003). However, if government agencies were to focus

on tree benefits of more direct livelihood relevance, the general public might be more inclined to

239
prioritise tree protection and care, thereby simultaneously creating positive effects at larger scales

through increased tree cover and health.

Assuming relevant departments and agencies agree with my assessment, the challenge becomes

how to amend existing structures and processes. I do not want to list specific actions or devise a

road map to be followed. Rather, I suggest some general ideas that could initiate conversations

with the city’s residents who should contribute to the process. To facilitate and support planting

on private lands, especially in densely populated and often ‘poorer’ neighbourhoods, the permit

process should be simplified. Pruning constitutes a routine maintenance operation, and while

EPA correctly states that it reduces tree benefits in the short to medium term, it does not threat-

en tree survival and longevity. Furthermore, the permit application system is overly complicated,

prohibitively costly, obstructive in nature, potentially time-consuming and with limited prospects

of success. Moreover, this ‘user-unfriendly’ process ignores the complex, varied and dynamic na-

ture of human–tree interactions in Accra’s often physically constrained spaces. Reducing people’s

flexibility in making routine and necessary tree maintenance choices, as the current state man-

agement system seeks to do, could undermine their interest in and willingness to invest in tree

planting, protection or nurturing in the first place.

Instead of regulations, incentives may better help to protect and retain urban trees on private

lands (Ordóñez-Barona et al., 2021). Utilising the green index EPA wants to produce, the gov-

ernment might consider offering tax or other financial incentives for tree planting on private

property in sparsely vegetated urban areas. Where initial tree purchasing costs prevent planting

on private land, a tree purchase voucher system could benefit cash-poor individuals (Thompson,

2018). However, the potential efficiency in Accra would have to be analysed as Thompon’s case

study in Louiseville (US) suggested low initial costs as people used seeds and growing containers

they already owned, something I also observed in my case study. Nonetheless, the Louiseville

240
cost-benefit analysis demonstrated how these vouchers, given to private property owners by the

government, benefitted both parties. Such systems could also be financed through private-private

partnerships, with large companies not just ‘adopting’ open spaces but also funding seedling pro-

duction and distribution to private landowners. However, such voucher systems may not just al-

low the government to control costs but to “regulate tree species to ensure economic and envi-

ronmental impacts are realized” (Thompson, 2018, p. 9). Thus implementing an incentive ap-

proach requires attention to avoid excessive government interference with private property own-

ers’ species choices.

Involving the Youth Employment Agency might offer a feasible way to ensure adequate seedling

and sapling maintenance, as trained participants could help people who lack the required tools or

the physical ability to regularly water or prune plants. Communicating with the public should be

the first step as they might come up with creative ideas, such as the group of citizens who came

together in Teshie Nungua, just east of Accra, to build a small children’s park (Fig. 41). Here, the

local imam provided water, another resident tools and storage space for materials, and several in-

dividuals contributed money via their mobile phones to cover additional project costs for grass or

labour for planting.23 Similar citizen-driven activities involving trees might be an option elsewhere

in Accra, given government and private sector support, and subject to meaningful participatory nego-

tiation of progressive access rights, imaginative joint management arrangements and secure in-

come or funding streams.

23https://www.dw.com/en/accra-is-becoming-a-concrete-jungle-so-locals-are-building-their-own-parks/a-49730607
(Accessed: 20.01.2020)

241
 Figure 41 Location of Teshie Nungua
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

7.4 Chapter Conclusion

Prioritising individual to neighbourhood scales might create an urban forest that does not pro-

vide ‘optimal results’ for other scales (Heynen, 2003), such as the global one, preoccupied with

the trees’ role in mitigating climate change’s negative effects. In Accra, government departments’

management priorities focus on these larger scales with urban forest strategies devised to maxim-

ise benefits like health, energy savings, and produce (climate change) resilient cities. However, the

current top-down approach neither created nor maintained a sustainable urban forest that effi-

ciently provides these benefits. A bottom-up approach that centres trees’ livelihood functions for

(re)production could, in turn, incentivise people to plant on private lands and maintain and pro-

tect both public and private trees. This strategy might provide a more feasible way to increase

overall tree cover rather than the government’s current approach, which largely fails to establish

the relevance of urban forestry for people’s everyday lives. A detailed understanding of the mul-

242
tiple and complex ways in which urban forestry is implicated in livelihoods is an important first

step. This may lead to re-thinking planning and management and, eventually, developing ap-

proaches and strategies which increase the likelihood that city residents can and want to decide

for tree planting and care.

243
 8 CONCLUSION

“I think there are two ways to open the doors to every transformational
change: One, presidential leadership and two, public advocacy and action.
Presidential leadership we don’t really have so much control over it, but we
have control over public advocacy and action, and I think sometimes when the
campaign is so strong, and the media actually buys into it, and the media wants
to talk about it then it challenges government to also engage.”

Mr Commey, Ghana Youth Environmental Movement

In this concluding chapter, I propose five recommendations for urban forestry in Accra. I ar-

range these suggestions in a stepwise process, which reflects the order in which I think these rec-

ommendations should be implemented. The first step is to diversify the knowledge and experi-

ence base on which urban forestry-related actors in Accra make their decisions. In the second

step, the city should, together with relevant stakeholders, develop an urban forest management

plan. Related to this step, there are two further recommendations, which include re-evaluating

species selection criteria/species–site matching and a focus on disadvantaged neighbourhoods.

The last step I then recommend is to fund the Department of Parks and Gardens. I begin the

chapter by summarising the results of the previous ones, discussing the findings in relation to my

research questions. I then synthesise my work’s components demonstrating how they relate to

each other and what theoretical and practical implications this more holistic view generates. After

that, I discuss three ways in which my research adds to political ecology and urban forestry re-

search. Based on my assessment of current urban forestry practices, I then discuss the above-

mentioned five recommendations in more detail, which address the city’s two most fundamental

urban forestry problems. First, a common narrative that falsely blames marginalised groups for

low and/or diminishing tree cover and, second, the likely effects of current and proposed inter-

244
ventions, which fail to capitalise on urban forestry’s multi-purpose nature. Where relevant, I try

to estimate my recommendations’ level of difficulty in terms of implementing them, the necessary

contextual pre-conditions, and their likely effects. I conclude the chapter with a look both back-

wards and forwards, discussing problems encountered during data collection and the pathways

these issues suggest for future research.

8.1 Summary and Synthesis

In this thesis, I have demonstrated how historical events and processes, most notably colonial

city planning approaches, created a lasting structure of inequity that persists until today. While

other variables, such as tree pests or extreme weather events, also impacted the city’s urban for-

est, they have not as fundamentally changed forest structure as the city’s racially motivated mor-

phology. Though Achimota Forest originated as a fuelwood plantation, colonial tree planting fo-

cused on producing shade and beautifying selected parts of the town, especially those inhabited

by the British. These tree benefits remain important in present-day urban forest planning and

management, constituting the Department of Parks and Gardens’ main species selection criteria

for public trees. My analysis demonstrates how we cannot fully understand contemporary urban

forest structure without investigating the legacy effects that have contributed to its formation.

Thus, depicting currently existing intra-city disparities and correlations between forest structure

variables and neighbourhood characteristics can create simplistic interpretations of cause and ef-

fect. Failing to assess the context, which includes investigating urban forest development over

time and across space, may falsely promote a view that identifies marginalised groups as respon-

sible for the structural discrimination they experience. Such accounts need to be critically exam-

ined to expose the mechanisms and interests that both facilitate and benefit from their existence

and deployment.

245
I have further presented numerous strengths, weaknesses, opportunities, and threats that urban

forestry-related stakeholders identified regarding their use and management of the city’s trees. I

combined these individual accounts and showed how government departments, NGOs, and in-

dividuals whose work includes trees and their products currently manage Accra’s urban forest.

These actors often fail to achieve their respective objectives due to resource shortages, unneces-

sary and wasteful duplication of effort, and destructive intra-sectoral competition. A detailed ur-

ban forest management plan, whose aims would determine agreed-upon activities and a specified

course of action, could remedy the second and third problems. Creating such a plan is fraught

with potential pitfalls concerning citizen participation and inclusion of diverse knowledges. How-

ever, the current undirected situation offers little direction and no coherent response to ongoing

processes of green space loss or socio-environmental inequalities within Accra.

The Department of Parks and Gardens restricts its tree planting mostly to sites along major roads

in the headquarter’s vicinity. A shortage of GPS devices hinders EPA’s attempts to geolocate the

city’s trees, and sub-metropolitan officers have to walk to and through neighbourhoods, reducing

the time available for tree-related work. Rather than acknowledging these management shortcom-

ings and how they negatively affect the urban forest, several officials prefer to evoke the narrative

of ‘the unaware and disinterested poor destroying the environment’. Such explanations require no

acknowledgement of fault on the part of the government or its agents. Thus, residents in low

canopy cover neighbourhoods are not just blamed for current disparities in tree cover but also

for the ongoing loss of green space and the related decrease in benefits like reduced tempera-

tures, air pollution and stormwater run-off absorption.

I then focused on households’ experiences, illustrating the multiple (re)productive functions of

private and public trees in people’s lives and livelihoods. Household members differed with re-

gards to their role in the management and use of privately-owned trees. Some actively planted

246
and maintained trees, others ‘passively’ enjoyed their benefits or showed little interest in, albeit

being aware of, the trees’ multiple purposes. These positions cut across age ranges, gender, and

position, but in all four case study households, people managed their trees for various reasons,

including direct consumption of fruits and tree-based medicines or use of tree products as part of

gifting and exchange systems. Therefore, I suggested that the physical capital ‘tree’ not only af-

fects human capital (diversifying a person’s diet or improving their health) but can also constitute

an essential component in creating and maintaining social capital. Furthermore, in ameliorating

micro-climates through their shade, trees offer a space in which people can meet and interact.

The trees thus further act as vital intermediaries in the process of creating and maintaining social

capital. The household questionnaire and case study both clearly demonstrated the public’s

awareness of and knowledge about the importance of urban trees, contrary to what a major actor

like EPA claimed. As many of the benefits that households manage their trees for are not current

state policy priorities, public trees on which tenants in treeless compounds depend often fail to

meet the people’s needs and requirements.

Finally, I selected five urban forestry-related interventions to analyse their likely effects on urban

trees’ functions within a sustainable livelihood framework. These examples included current and

proposed tree inventories, a re-evaluation of city planning schemes, changing tree laws, large-

scale tree planting activities, and Achimota’s development into the Accra Eco-Park. I outlined

how these interventions might affect Accra’s urban forest and thereby the type, location, quality

and quantity of provided tree benefits. While resource shortages hinder EPA’s and DPG’s inven-

tory (plans), a lack of participation, over-complicated processes, and a restricted set of prioritised

tree benefits reduce the interventions’ ability to benefit especially marginalised population parts.

Generally, I identified little indication that the five examined events and processes could success-

fully decrease intra-city disparities and might instead widen them even further. Disregarding the

individual to local-level tree benefits might disincline the public to participate in tree propagation

247
and protection and thus decrease overall tree cover and, thereby, national to international-level

benefits. I, therefore, suggested considering a bottom-up approach that might produce larger

trickle-up effects than the current top-down approach.

Accra’s trees are distributed unevenly across the city and continue to decrease in number. Tracing

the history of both urban forestry and city planning, I argue that current intra-city green inequali-

ties cannot be understood without referring to the legacy effects of past events and processes.

Ignoring these impacts creates conditions favourable for simplistic arguments, which hold mar-

ginalised groups responsible, rather than centring discourses that emphasise the fundamental

contribution of institutional and structural discrimination. Portraying inhabitants of low canopy

cover neighbourhoods as unaware of tree benefits turns them into easy scapegoats for the con-

tinued loss of green. A problem, I maintain, more likely due to the absence of a comprehensive

urban forest management plan, a lack of resources, and missing coordination between major ac-

tors/stakeholders. My qualitative-focused, household-based analysis supports the claim that poli-

cy and decision-making state actors largely fail to recognise and acknowledge the trees’ multiple

(re)productive livelihood functions. Therefore, current and proposed developments fail to more

extensively meet people’s needs through establishing a multi-purpose urban forest that offers a

wide range of benefits simultaneously across geographical space and scales. Instead, the dominant

narrative and prevailing management priorities are more likely to widen already existing dispari-

ties, disproportionately harming marginalised population groups and further restricting their ac-

cess to urban trees.

8.2 Thesis Contributions

Quantitative data collection approaches, (post)positive views, and particularly a focus on Europe-

an and North American cities have long dominated urban forestry and cognate disciplines (Roy et

248
al., 2012; Shackleton, 2012; Ostoić and Konijnendijk, 2015; Shackleton et al., 2021). Urban forest-

ry thus lacks a comprehensive empirical base, given its limited coverage of views and experiences

of more than three-quarters of the world’s population, which has so far inhibited its ability to

produce genuinely inclusive theories and frameworks. My research has not only used a qualitative

approach, produced primary data on Accra’s forest structure, investigated its development over

time and across space, and demonstrated the effects of conflicting management priorities, but it

has also added another geographical data point to current urban forestry research coverage.

Due to urban forestry’s context- and place-specific nature (Roman et al., 2018), qualitative re-

search approaches offer valuable tools for data collection, which my findings demonstrate, nota-

bly in relation to three main points highlighted in what follows. First, the discipline requires a

stronger historical focus that moves beyond simply outlining its emergence and growth in coun-

tries or regions historically favoured by researchers, to providing detailed and contextualised city-

level analyses capable of assisting in establishing the details of often complex temporal and spatial

linkages or interactions in space/place. Second, a sustainable urban forest, large parts of which

grow on privately-owned land, requires attention to people’s emotions and affective behaviour,

and cannot be created and maintained through use of an entirely cognitive, information-

providing approach such as the EPA’s focus on education campaigns. Third, individual trees

combine to create a city’s urban forest in the same way that individual inhabitants collectively

constitute a city’s population. Thus, to understand the entirety of all views, priorities, and uses

relevant to practical urban forest planning and management, and the discipline’s theoretical as-

sumptions and frameworks, we need to consider at least four possible combinations of per-

son/society and tree/forest interactions.

My thesis has analysed various legacy effects, such as colonial city planning or tree benefit prefer-

ences, on urban forestry in Accra and used these insights to suggest likely future socio-

249
environmental developments and potential adjustments to improve outcomes. While studies on

the politics of natural resource use, management, and extraction typically privilege scale and

space, temporal dynamics have received less attention (Fent and Kajola, 2020; Kajola, 2020).

However, such dynamics provide essential information for contextualising current urban forest

structure, processes and dynamics. My analysis of Accra’s urban forest thus adds to the growing

body of political ecology-related research focusing on temporal dynamics of natural resource

management (Kojola, 2020). And, selecting the city’s trees, I also extend analyses associated with

more commonly studied resources like metals and minerals (Fent, 2020; Kojola, 2020), or water

(Braun, 2020), given that urban forest benefits are only partly tangible, whereas global develop-

ment policies and practice have facilitated the commodification of several of its intangible ser-

vices, and thus incorporated them into a capitalist framework of management, production, and

extraction.

In Accra, several state actors participate in developing urban tree-related policies, regulations, and

laws, and while these are designed to “predict or prevent a range of potential problems, such as

[…] continued effects from climate change” (Fent 2020, p. 882), they essentially reflect normative

assumptions about the urban forest’s ideal future. The actors in question thus claim a right to

govern this future in the present, to capitalise on their ability to turn their anticipation into action

by deploying the power to construct and maintain dominant narratives. My research demon-

strates how ‘past’, ‘present’, and ‘future’ do not constitute clearly delimited entities in urban for-

estry, but exist simultaneously through both tangible, non-human structures like buildings and in-

tangible variables like ‘anticipation’ or emotional memories of a past of greater tree cover to be

recovered. My analysis thus starts to question whether any urban forest can be fully understood

solely through colonial and capitalist notions of “measurable and linear time” (Fent and Kojola,

2020, p. 821).

250
The thesis further demonstrates that a sustainable urban forest cannot be planned, created and

maintained by the use of a cognitive, data-based and information-providing approach only. The

hierarchy of effects model draws attention to the role of affection, as do case study participants

comments. John Yetsowodo, for example, derives a considerable amount of income from his

trees, yet he stated that he could not describe particular reasons for liking them – he simply does.

While researchers have investigated the psychological and mental health effects of the urban for-

est (e.g. Kondo et al., 2018; Lee and Son, 2018; Chen et al., 2021), my research highlights the need

to investigate further the role of emotions and feelings for urban forestry. Understanding people’s

internal and intimate connections with trees, where natural elements are not a means to an end,

but ends in themselves which exist outside of benefit-oriented and utilitarian frameworks, is thus

fundamental to understanding how urban forests ‘become’. Such research would add to existing

approaches that, in employing questionnaires and statistical analyses to elucidate people’s beliefs,

attitudes, and preferences for certain ecosystem services and management priorities, measure

cognition rather than affection (Perrin and Benassi, 2009).

Individual trees combine to produce a city’s urban forest, and individual humans collectively con-

stitute a city’s population. Research thus needs to investigate (at least) four interaction combina-

tions (forest–society, forest–person, tree–society, tree–person) for a comprehensive understand-

ing of the socio-environmental construct ‘urban forest’. However, much recent work has centred

the forest–society link, alongside a framework, in which the city’s trees provide nature-based so-

lutions in the form of green infrastructure to create resilient urban areas. These concepts may fur-

ther neoliberal agendas (Kotsila et al., 2021), or fail to address issues of power struggles and

(in)justice (Cousins, 2021; van der Jagt et al., 2021). They also render invisible the urban forest’s

‘building blocks’, i.e. single trees and how individuals use and manage them, a subject on which

there has thus far been little data (Kirkpatrick et al., 2012), especially outside of ‘Northern’ cities.

My research adds to the body of literature that addresses these tree–person relationships, and po-

251
sitions these arboricultural and horticultural attitudes and behaviours as central to city-wide forest

planning and management, as private lands contain a significant proportion of any urban forest. I

thus suggest that currently dominant concepts, models and guidelines in strategic urban forestry

(e.g. Leff, 2016) may need to be revisited in the light of these findings.

8.3 Recommendations
Based on my analysis, I now propose five recommendations aimed at major state actors. These

suggestions include diversifying the knowledge base, developing an urban forest management

plan, re-evaluating species selection criteria and species–site matching, focusing on disadvantaged

neighbourhoods, and funding the Department of Parks and Gardens. None of the recommenda-

tions is, on its own, more important than the others, as they combine to address a single issue

from several different angles. Their order, though, is essential, which is why I have arranged them

in a step-wise process. While an alternative sequence might well increase urban forestry’s organi-

sational efficiency in Accra, it could easily fail to achieve the fundamentally important element of

social justice and equity. These recommendations, followed in the proposed step-wise order, have

the potential to improve urban forestry activities and outcomes in Accra.

My transformative political ecology view centres justice and equality, thus ‘improving urban for-

estry activities’ signifies increasing tree cover in Accra in a way that reduces intra-city variation. It

further includes taking advantage of urban forestry’s multi-purpose nature, intentionally aiming to

produce a wider range of benefits at various scales. Ultimately, I propose moving beyond the cur-

rent focus on the national and global scale, centring urban forest management on supporting in-

dividual and group livelihood opportunities. As part of a poverty reduction and resilience foster-

ing approach, this would link larger-scale concerns with local-level preoccupations. Fundamental-

ly restructuring urban forestry, involving the re-distribution of power, will likely face opposition

252
by those attempting to maintain their decision-making privileges. However, I suggest re-

prioritising scales – away from the current focus on the preferred urban tree benefits within a ne-

oliberal, global capitalist system – can create a trickle-up system that might be viewed favourably

in an outcome-oriented assessment.

8.3.1 Recommendation One: Diversify the Knowledge and Experience Base

One last time, I return to urban forestry’s definition, the “art, science, and technology of manag-

ing trees and forest resources in and around urban community ecosystems for the physiological,

sociological, economic, and aesthetic benefits tree provide society” (Helms 1998, p. 193). This

definition identifies benefits as the outcome of the management process. However, the term

‘benefits’ sounds deceptively simple and, when scrutinised, poses more questions than it answers.

What concrete benefits are we talking about? Where exactly are they to be produced and in what

amounts? How do we decide which ones are needed? Depending on how we answer these ques-

tions, a certain urban forest structure emerges as the most promising one to produce the desired

benefits in the ideal locations at the right time. However, the many involved stakeholders likely

respond differently to these questions, creating a situation in which multiple views conflict.

Whose opinions, experiences, and knowledges are prioritised then turns into a matter of being

able to make oneself heard.

Urban forestry’s multi-purpose nature constitutes its greatest challenge, constantly requiring

trade-offs, but is simultaneously its major strength, an opportunity to centre social justice when

making those (scalar) trade-offs. Based on my work, I hypothesise that orienting urban forestry

towards benefits produced and used at the individual to neighbourhood level can result in a trick-

le-up effect through an overall increased canopy cover. More city trees then also provide more

ecosystem functions and services typically associated with higher scales. Thus, a scalar bottom-up

253
approach may eventually also outperform the currently dominant top-down one with regards to

tree benefits’ quality and quantity at national or global scope. The current approach provides in-

sufficient trickle-down effects and disincentivises people from allocating resources to tree care.

However, for an effective and responsive bottom-up approach, we must understand how people

use trees within a livelihood context to centre their knowledge and experiences when re-designing

urban forestry.

Diversifying the knowledge and experience base needs to precede any other step as failing to do

so is unlikely to result in an inclusive and comprehensive urban forest management plan. This

first step may also be the most challenging one, requiring people and entities in decision-making

positions to re-distribute power and acknowledge responsibility for current shortcomings, at least

with regards to the potential effects on social justice of the proposed interventions. Additionally,

it requires very fundamental and time-consuming structural changes and an openness to further

investigate human–tree relationships in the city. A differently tailored approach to urban forestry

could help create more resilient livelihoods (with a chain of positive effects on other scales) and

benefit the national economy directly, such as the trees’ shade protecting roads from premature

deterioration. Accra’s mayor is keen on urban greening and has started to take action, also involv-

ing other state actors like TCPD, intending to increase green spaces in the city. Furthermore, Ac-

cra joined networks that promote nature-based solutions to pressing issues like the negative ef-

fects of climate change. There is, however, limited indication suggesting an explicit acknowl-

edgement of the dimension of social justice, especially regarding the often overlooked provisional

tree benefits and those able to strengthen the human and social capital base.

As quoted at the chapter’s beginning, the interviewee from the NGO GYEM suggested that

there are two paths to transform any dominant process: (presidential) leadership or public advo-

cacy and action. While mayoral interest in Accra’s urban environment is visible, state actors could

254
seek inspiration in how other cities have cooperated with their inhabitants and support ‘grass-

roots initiatives’, public action, and creative approaches, and move from infor-

mation/consultation to partnership and empowerment. In Lisbon (Portugal), the city provided

2.5 million Euros to implement the winning citizen proposal to develop public green spaces

(Ambrose-Oji et al., 2017). In Edinburgh (UK), the city council drew up a statement of support

for residents who had started to plant on public places like street corners or grass areas around

residential buildings, as they lacked access to private gardens and public green spaces (Ambrose-

Oji et al., 2017). The city of Melbourne (Australia), in turn, used online forums, pop-up face-to-

face engagement sessions, stakeholder workshops and discussion groups to engage its citizens

(Gulsrud et al., 2018). I do not suggest copying these activities but further explore such approach-

es, including urban forestry’s creative and artistic potential. As the traffic roundabout in Accra’s

North Ridge neighbourhood demonstrates, the first steps have already been taken (Figs. 42 to

44).

Figure 42 ‘The Unusual Garden’ – I

Own picture

255
 Figure 43 ‘The Unusual Garden’ – II
Own picture

Figure 44 ‘The Unusual Garden’ – III

Own picture

To diversify the knowledge and experience base, state actors need to consider the natural re-

source base as the intermediate link that connects their work with the wider society in which and

256
for which they should operate. EPA has been developing educational materials and aims to run

events where they teach citizens about the urban trees’ values and environmental protection. Re-

designing these activities could create a two-directional flow of information, actively seeking the

population’s ideas and knowledges. Such communication could occur in an informal manner that

can be implemented cheaply and quickly. However, a more formal approach might better reflect

this activity’s importance, requiring the development of short questionnaires or employing a

moderator for a focus group discussion. Utilising already existing structures or those currently be-

ing built to consult the city residents should not be confused with a re-distribution of power. I

have previously (see Chapter 5) discussed how participation inherently suffers from power imbal-

ances when certain groups create such spaces and then ‘invite’ others into them. Sub-

metropolitan environmental health officers constitute another underused source of information

as they interact with the city’s inhabitants and could provide insights on human–tree relationships

in their neighbourhoods. These sub-metropolitan offices could also serve as meeting points in

formal and informal setups.

Involving city inhabitants in top-down urban forest planning approaches requires interest and ac-

tion on the part of decision-making state actors. Where this interest remains limited, public advo-

cacy and action can increase residents’ visibility. While the previously discussed issues with partic-

ipation and representation also apply in these settings, bottom-up actions may be more likely to

include and centre on the idea of social justice. Few NGOs have previously worked in the area of

urban forestry in Accra, and their actions have been of varying success. Youth4GreenGhana

planted street trees but could not maintain them in the long term. While DPG provided saplings

to Y4GG, the lack of more extensive government organisations’ support seems like a missed op-

portunity in terms of creating lasting partnerships for mutual benefit. Additionally, NGOs and

other grassroots agencies could consider including urban trees in their agenda. This does not only

apply to environment-related groups but could, indeed, be incorporated in social movements

257
concerned with urban development, health, or livelihood resilience and poverty reduction. I do

not consider it necessary for urban trees to be at the centre of these activities, as they may just

constitute one component of a more holistic and diversified programme.

8.3.2 Recommendation Two: Develop an Urban Forest Management Plan

Once the knowledge base has been diversified, various stakeholders can collaborate to develop a

formal urban forest management plan for Accra. I base this recommendation on my analysis of

interviews with state and non-state actors from a variety of backgrounds, which indicated a se-

vere lack of interaction and collaboration. Currently, there is widespread confusion about the dis-

tribution of responsibilities, with unnecessary duplication of effort or crucial activities left un-

done. Additionally, one actor unintentionally undermined and thus rendered ineffective the work

of another one, such as road and infrastructure development occurring after roadside tree plant-

ing. Developing a management plan requires personnel to partake in repeated inventories of the

bio-physical resource base and the financial means to purchase necessary equipment. While re-

sources are scarce or rather distributed towards other government organisations, I suggest taking

advantage of existing structures and connections. Inventory personnel could be recruited from

YEA’s ‘Youth in Agriculture and Afforestation’ module or institutions of Higher Education,

both for developing an initial plan and for ongoing monitoring activities. When government

funding is insufficient, fines for illegal tree felling and/or part of the fees for felling permits could

be used to develop and follow the forest management plan.

When preparing an urban forest management plan, groups and individuals can draw on several

free online resources to help facilitate the process. An affiliate of the California Urban Forests

Council24 developed the Urban Forest Management Plan Toolkit, while the US Forest Service,

24 https://ufmptoolkit.net/ (Accessed 21.08.2020)

258
American Forests, and the National Association of Regional Councils created the Vibrant Cities

Lab Urban Forestry Toolkit25. I recommend forming a multidisciplinary team that includes,

among others, social scientists, NGO representatives and members of the general public working

alongside foresters and natural and environmental scientists. My first recommendation, to diversi-

fy the knowledge and experience base, should have already made state actors aware and apprecia-

tive of the varied viewpoints of Accra’s differentiated population. As the largest and arguably

most important group, Accra’s private citizens should play a central role in a process that needs

to pay attention to this group’s heterogeneity to ensure inclusive planning and implementation.

While funding the development of a management plan might be challenging, its value cannot be

overstated. Urban forestry in Accra could become more efficient, responsive and representative

when stakeholders work with, rather than separately from or even against, each other. To pursue

the goal of social justice, I now discuss in more detail two further recommendations closely

linked to this step, which I consider indispensable for creating an inclusive urban forest both in

Accra and elsewhere.

8.3.3 Recommendation Three: Re-Evaluate Species and Planting Site Selection

As part of developing an urban forest management plan, EPA and DPG should re-evaluate the

rationale behind their current species selection and species–site matching criteria. At the moment,

they plant and protect public trees mainly for their aesthetic value and shade. Other benefits,

such as fruits, are of no importance also due to fear of windfall fruits littering sidewalks and roads

or harvesting activities creating risk to traffic and bystanders. This approach, however, foregoes

the opportunity to capitalise on urban trees’ multiple functions. While potential traffic disruptions

cause valid concern, not all public trees are street trees and instead grow in open areas at safe dis-

tances from roads. The NGO Philadelphia Orchard Project, for example, converts underutilised

25 https://www.vibrantcitieslab.com/about/ (Accessed 21.08.2020)

259
spaces like vacant lots or church and school grounds into community-owned orchards, focusing

on low-wealth neighbourhoods in Philadelphia (USA). If food items that grow in high pollution

environments pose a health risk when consumed by humans remains unclear as it depends on the

precise interaction between location, pollutant/metal and tree species (Li et al., 2006; Oliva et al.,

2008; von Hoffen and Säumel, 2014). Rather than excluding this benefit based on general health

concerns, I suggest analysing viability in a species and place-specific manner to determine what

might be safe to consume under which circumstances in Accra.

Instead of proposing, very specifically, how to create a more multifunctional urban forest, I focus

on general ideas to consider when developing a management plan. Implementing any activity,

however, will depend on its suitability within the specific context. I suggest that stakeholders ex-

amine the two related concepts of ‘urban food forestry’ and ‘edible green infrastructure’. Accord-

ing to Clark and Nicholas (2013, p. 1652), the former is defined as “the intentional and strategic

use of woody perennial food-producing species in urban edible landscapes to improve the sus-

tainability and resilience of urban communities”. Their introduction to the topic highlights the

advantages of re-evaluating species selection, as well as the requirements for such re-evaluation,

including significant cooperation between relevant actors and stakeholders. Edible green infra-

structure relates to a broader approach that does not just include trees. It is defined as “a sustain-

able planned network of edible food components and structures within the urban ecosystem

which are managed and designed to provide primarily provisioning […] ES [ecosystem services]”

(Russo et al., 2017, p. 54). Worldwide, several initiatives and cities use urban food trees in their

work and plans (Clark and Nicholas, 2013; Kowalski and Conway, 2019). These projects could be

studied to evaluate whether and how similar strategies could become part of Accra’s urban forest

future.

260
To diversify Accra’s public urban forest, stakeholders should produce a list of climate suitable

species, which can then be examined for their potential uses, an approach similar to that in New

York (USA) (Hurley and Emery, 2018). As DPG and EPA already consider the species’ beautifi-

cation, shading and human health benefits, I recommend focusing on additional types of benefits

such as the trees’ provisioning ecosystem services. However, other ideas and needs may emerge

when diversifying the knowledge base or planning in a participatory approach. Practically, this

step requires diversifying DPG’s tree seedling nursery and increasing its capacity. The necessary

resources depend on the desired speed, future production volume, and the intensity of diversifi-

cation but could be matched to available personnel and budget. Additionally, the involved parties

need to be aware of available planting locations and their associated growing conditions and po-

tentials. This information should be collected as part of developing the management plan, and

required resources will vary based on the chosen approach, e.g. ground-based or desk-based.

Public trees would then offer more varied benefits, including those associated with the individual

or household scale, available and accessible to non-landowning individuals, i.e. a large proportion

of the city’s population. This recommendation does not demand immediate large-scale invest-

ments in or changes to the forest’s structure as it could commence in selected places and be geo-

graphically expanded over time.

8.3.4 Recommendation Four: Focus on Disadvantaged Neighbourhoods

In Chapter 4, I demonstrated how tree cover in Accra varies between neighbourhoods, positively

correlating with the areas’ socio-economic rating. Thus, the urban forest management plan needs

to be spatially explicit at the below-city level, focusing on disadvantaged neighbourhoods. As the-

se areas often feature limited plantable space, tree cover across Accra cannot be levelled out, but

the gap reduced. I recommend including both public and private lands in this step, supporting

and incentivising landowners to plant trees rather than relying on the laws to enforce these activi-

261
ties. Government actors could distribute free seedlings to interested residents, providing species

that produce desired benefits. To ensure a long growth time, these trees should require limited

maintenance, especially in terms of irrigation, and reach a size suitable for the available space.

This setting further requires solutions as to how renters can access trees, such as making com-

pound trees available for non-destructive use to all associated households and individuals. Addi-

tionally, land owned by neighbours, friends, or family may provide alternative planting locations

for non-landowners, such as observed in one of the case study households. When diversifying the

knowledge and experience base, participants in community meetings or other settings may brain-

storm other creative ideas and suggestions as to how to approach this situation.

Though shifting theoretical focus to disadvantaged neighbourhoods can be instant, the slow

growth of trees will create a noticeable effect in the medium to long-term only. In the planning

phase, this step requires a willingness to engage with residents in less green neighbourhoods to

jointly develop feasible proposals aiming to increase canopy cover. While conducting these meet-

ings and paying related personnel requires financial inputs, these will increase even further during

the practical phase when seedlings might need to be distributed, planting support given, and con-

tinued maintenance ensured. Overall, I would expect a high need for collaboration between dif-

ferent actors such as the Department of Parks and Gardens, sub-metropolitan offices, individuals

trained as part of the Youth Environmental Agency’s modules, and the general public. If success-

ful, this step has the potential to provide individual and household-scale tree benefits to a wider

population group, thereby positively affecting well-being and livelihood security, while also in-

creasing citywide canopy cover and possibly the quantity of those benefits that require larger

stands and better-connected tree patches.

262
8.3.5 Recommendation Five: Fund the Department of Parks and Gardens
Lastly, I recommend funding the Department of Parks and Gardens. DPG has been underfund-

ed for years, recognised not only by the department itself but also by other government depart-

ments and agencies, NGOs, the media, and the public. While DPG attempts to raise money and

other resources from foreign embassies or through public-private partnerships, department em-

ployees continue to lack the means to work efficiently. Recent tree planting activities have in-

creased their responsibilities (see Chapter 7), and developing and later implementing an urban

forest management plan would do so even further. I recommend considering allowing DPG to

retain a proportion of the revenue they generate and to direct a part of the permit application fees

and fines towards them. This, however, requires recognising the department’s role not just in city

greening specifically but also in sustainable and comprehensive city planning and development.

While neglected in the recent past, DPG might find an ally in Accra’s mayor who has shown in-

terest and concern regarding the city’s plants. Though the government apparently recognised the

link between city greening and planning, theoretically fusing DPG and TCPD at the local level,

visible effects are still lacking, suggesting a continued low rank on the government’s agenda.

Without this last step, even the most comprehensive management plan will fail to succeed, as

funding DPG is a basic requirement to create lasting structures of sustainably planning, monitor-

ing, and evaluating Accra’s urban forest.

8.4 Looking Backward and Looking Forward

In the methodology chapter (see Chapter 3), I discussed my work’s limitations related to lan-

guage, researcher behaviour in environment and sustainability-related studies, and the ethics of

my chosen approach. In this section, I focus on methods-related issues and the data collection

experience, discussing the mixed methods design and some of its individual components. Based

on the problems and limitations I identify, I then suggest three related topics for future research,

263
relating to household survey design, household and individual level interactions with trees, and to

the integration between urban forest research and practice.

8.4.1 Hindsight
The mixed methods approach I chose was suitable and indeed necessary to comprehensively ana-

lyse urban forestry in Accra and respond to my research questions. I want to reiterate, however,

some of the challenges Teye (2012) identified in his mixed methods study of the formulation and

implementation of forest policy in Ghana. Teye described the research process as time-

consuming and requiring both high amounts of labour and money, which resonates strongly with

my experience. To keep project demands manageable, I had to restrict the resources spent on my

study’s individual components. As statistical requirements prescribed the scope of canopy cover

analysis, tree inventory, and household survey, I had to adjust the remaining time for archival re-

search, interviews, and the case study accordingly. I, therefore, suggest examining how require-

ments related to quantitative data affect the quality and quantity of qualitative data in mixed

methods research. Being new to mixed methods research, I also followed an often-walked path

and combined common research methods. In hindsight, I wish I had chosen a more unusual ap-

proach and mixed methods more creatively.

In future projects, I would want to focus more on qualitative approaches. In hindsight, especially

the information from the survey does not seem to fully justify the resources I spent on it. Admin-

istering the questionnaire demanded excessive amounts of time, personnel, and money, some-

thing I had seriously underestimated. In a similar project, I would make major adjustments to a

survey, if including one at all, focussing on a few, fast to ask and easy to answer, closed-ended

questions with limited open-ended follow-up questions. This approach should significantly re-

duce the time necessary to administer the questionnaire, as well as the demand on respondent

264
time, attention and goodwill. I feel that the current survey’s length exceeded the concentration

and interest span of several participants, possibly reducing the quality of the obtained infor-

mation. I would then invest freed up resources into a more extensive case study component,

which would provide detailed context to closed-ended questionnaire responses.

8.4.2 Foresight
Where a sizeable proportion of the study population is illiterate and/or unreachable by mail or

online, researchers often have to administer their questionnaires. Such surveys can then become

unfeasible or negatively balanced in terms of required inputs compared with obtainable outputs.

I, thus, suggest creatively rethinking survey methodology and testing new approaches for their

appropriateness in a setting such as my study. In Burundi, Lee et al. (2016) utilised colour coding

and images in a researcher-assisted, self-completion questionnaire with a preliterate population

on a sensitive and potentially stigmatising topic. While I provided pictures to rank and rate prese-

lected urban tree benefits, I would like to investigate if using more visual images in similar ques-

tionnaires with non-literate respondents could decrease resource needs. In North America, an in-

creasingly common approach to reduce costs and non-response rates involves mixing modes of

survey data collection. Researchers offer respondents the opportunity to choose between a tele-

phone, web or postal survey and a directly administered questionnaire in the form of a face-to-

face interview (de Leeuw et al., 2008; Tijdens and Steinmetz, 2016). However, the feasibility of

such an approach has not been investigated in the context of mostly non-literate multi-lingual

study populations with limited internet access and no fully developed formal road naming and

address system.

I started to investigate urban trees’ functions within the context of household production and re-

production but within the limited scope of four cases in one city. For a more detailed under-

265
standing of individual-level human–tree interactions, this type of research has to be expanded,

both geographically and in volume. Where cities and countries are compared, the effects of the

environmental and socio-political contexts would need to assume great prominence. Comparative

research could enhance our understanding and appreciation of how political, environmental, so-

cial, technological, or legal variables shape people’s relationships with urban trees. Larger case

study designs could support theory building, thereby adding to the discipline of urban forestry in

which quantitative and positivist approaches continue to dominate. The four household cases

produced a wealth of information, significantly more than I had anticipated and despite the rela-

tively limited time I spent with each. While I observed and interacted with the case study house-

holds, using a more ethnographic ‘participant observation’ approach might noticeably extend the

limits of our current knowledge in this area. Becoming more actively involved could also be ex-

tended to other actor types, such as government bodies or NGOs. While EPA offered me the

opportunity to observe and participate in their tree-tagging activities, time constraints and prior

appointments prevented this experience.

Finally, we need to devote more time and effort to integrating urban forest theory and practice. I

established common reasons for people in Accra not to plant trees, which explain their reluctance

to invest in slow-growing and immovable plants. I also collected a list of reasons why people do

plant trees, outlining their many (re)productive functions. These constitute central aspects of ur-

ban forestry research worldwide, but applying this information in a way that has practical, prob-

lem-solving value seems to remain limited. Not only does urban forestry research often fail to

make practical impacts, but it also appears oddly detached from the practical experiences of individ-

uals and groups. Many of these groups share their experiences on the internet or social media

platforms, and a simple web search produces various initiatives similar to the Philadelphia Or-

chard Project (see Section 6.2.2.1). Their experiences, however, seem noticeably underutilised in

266
urban forestry as a research discipline. While I attempted to identify such groups in Accra prior to

and during data collection, I was unable to locate a relevant initiative.

8.5 Chapter Conclusion

In this chapter, I summarised the preceding chapters and how they responded to my research

questions. I then offered five recommendations, which not only would increase Accra’s urban

forestry efficiency but support the development of a more sustainable and socially just approach.

These recommendations, however, constitute no blueprint but should be adjusted according to

stakeholders’ inputs. Reflecting on some issues I noticed during data collection, I also suggested

related areas for future research. Urban forestry’s value is inherently linked to its perceived capac-

ity for contributing to policy implementation and the applied planning, establishment and man-

agement of a city’s forest. Thus, we need to branch out within the research discipline itself but al-

so to disciplines like urban ecology, urban planning or anthropology. At its most elementary level,

doing urban forestry, i.e. planning and managing city trees, is a matter of trade-offs. As these

trade-offs have real-world implications, the importance of asking ourselves who decides and who

does not, which experiences and knowledges count and which do not, should occupy a more

fundamental position, both in theory and implementation.

267
 LIST OF REFERENCES

Abbam, T., Johnson, F. A., Dash, J., and Padmadas, S. S. (2018) ‘Spatiotemporal variations in
rainfall and temperature in Ghana over the twentieth century, 1900–2014’, Earth and Space Science,
5(4), pp. 120-132.

Abbot, J. (2013) Sharing the City: Community Participation in Urban Management. London: Routledge.

Acheampong, R. A. (2019) Spatial Planning in Ghana: Origins, Contemporary Reforms and Practices, and
New Perspectives. Basel: Springer International Publishing.

Adams, C. D. (1957) ‘Activities of Danish Botanists in Guinea 1783-1859’, Transactions of the His-
torical Society of Ghana, 3(1), pp. 30-46.

Adarkwa, K. K. (2012) ‘The changing face of Ghanaian towns’, African Review of Economics and Fi-
nance, 4(1), pp. 1-29.

Adler, P. S. and Kwon, S. W. (2002) ‘Social capital: Prospects for a new concept’, Academy of man-
agement review, 27(1), pp. 17-40.

Agbelade, A. D., Onyekwelu, J. C., and Oyun, M. B. (2016) ‘Tree species diversity and their bene-
fits in urban and peri-urban areas of Abuja and Minna, Nigeria’, Applied Tropical Agriculture, 21(3),
pp. 27-36.

Agnoletti, M. (2000) ‘Introduction: the development of forest history research’, in Agnoletti, M.

and Anderson, S. (eds.) Methods and Approaches in Forest History. Wallingford: CABI, pp. 1-20.

Agyepong, I. A. and Manderson, L. (1994) ‘The diagnosis and management of fever at household
level in the Greater Accra Region, Ghana’, Acta tropica, 58(3-4), pp. 317-330.

Ahlman, J. S. (2017) Living with Nkrumahism: Nation, State, and Pan-Africanism in Ghana. Athens:
Ohio University Press.

Ahmed, A. and Puppim de Oliveira, J. A. (2017) ‘Integration of biodiversity in urban planning in-
struments in developing countries: the case of Kumasi Metropolitan Assembly, Ghana’, Journal of
Environmental Planning and Management, 60(10), pp. 1741-1764.

Akbari, H., Pomerantz, M., and Taha, H. (2001) ‘Cool surfaces and shade trees to reduce energy
use and improve air quality in urban areas’, Solar energy, 70(3), pp. 295-310.

Akese, G. A. and Little, P. C. (2018) ‘Electronic Waste and the Environmental Justice Challenge
in Agbogbloshie’, Environmental Justice, 11(2), pp. 77-83.

Akufo-Addo, N. A. D. (2017) Coordinated Programme of Economic and Social Development Policies (2017-
2024). Available at: https://s3-us-west-2.amazonaws.com/new-ndpc-
static1/CACHES/PUBLICATIONS/2018/04/11/Coordinate+Programme-
Final+(November+11,+2017)+cover.pdf (Accessed: 17.09.2020).

Alhassan, O. (2006) ‘Land use and environment in early Accra (1200 to 1800 AD)’, Research Review
of the Institute of African Studies, Supplement 17, pp. 21-35.

268
AMA and 100 Resilient Cities (2019) Accra Resilience Strategy. Available at:
https://ama.gov.gh/documents/Accra-Resilience-Strategy.pdf (Accessed: 17.09.2020).

Ambrose-Oji, B., Buijs, A., Gerőházi, E., Mattijssen, T., Száraz, L., Van der Jagt, A., Hansen,R.,
Rall, E., Andersson, E, Kronenberg, J., and Rolf, W. (2017) Innovative Governance for Urban Green
Infrastructure: A Guide for Practitioners, GREEN SURGE project Deliverable 6.3, Copenhagen: Uni-
versity of Copenhagen.

Amoako, C. and Adom-Asamoah, G. (2019) ‘From the seat of a traditional Kingdom to a Garden
city: the socio-spatial politics of managing green areas in Kumasi, Ghana’, African Geographical Re-
view, 38(4), pp. 310-325.

Amoako, C. and Inkoom, D. K. B. (2018) ‘The production of flood vulnerability in Accra, Gha-
na: Re-thinking flooding and informal urbanisation’, Urban Studies, 55(13), pp. 2903-2922.

Anderson, L. M. and Cordell, H. K. (1988) ‘Influence of trees on residential property values in

Athens, Georgia (USA): A survey based on actual sales prices’, Landscape and urban planning, 15(1-
2), pp. 153-164.

Andersson, H. (2016) ‘Colonial Urban Legacies: An analysis of socio-spatial structures in Accra,

Ghana’, Uppsala University. Available at: http://www.diva-
portal.org/smash/get/diva2:1073546/FULLTEXT01.pdf (Accessed: 16.09.2020).

Angelsen, A. and Wunder, S. (2003) Exploring the Forest – Poverty Link: Key Concepts, Issues and Re-
search Implications, CIFOR Occasional Paper No. 40, Bogor: Center for International Forestry Re-
search.

Anquandah, J. (2006) ‘The Accra plains c. AD 1400-1800: An overview of trade, politics and cul-
ture from the perspective of historical archaeology’, Research Review of the Institute of African Studies,
Supplement 17, pp. 1-20.

Arku, G., Luginaah I. and Mkandawire P. (2012) ‘“You Either Pay More Advance Rent or You
Move Out”: Landlords/Ladies’ and Tenants’ Dilemmas in the Low-income Housing Market in
Accra, Ghana’, Urban Studies, 49(14), pp. 3177-3193.

Arku, G., Yeboah, I. E. and Nyantakyi-Frimpong, H. (2016) ‘Public parks as an element of urban
planning: a missing piece in Accra’s growth and development’, Local Environment, 21(12), pp.
1500-1515.

Arnstein, S. R. (1969) ‘A ladder of citizen participation’, Journal of the American Institute of planners,
35(4), pp. 216-224.

Arrow, K. (1999) ‘Observations on Social Capital’, in Dasgupta, P. and Serageldin, I. (eds.) Social
Capital: A multifaceted perspective. Washington: The International Bank for Reconstruction and De-
velopment/The World Bank, pp. 3-5.

Arsenault, J., Talbot, J., Boustani, L., Gonzalès, R. and Manaugh, K. (2019) ‘The environmental
footprint of academic and student mobility in a large research-oriented university’, Environmental
Research Letters, 14(9), 095001.

269
Asase, A. and Kadera, M. L. (2014) ‘Herbal medicines for child healthcare from Ghana’, Journal of
Herbal Medicine, 4(1), pp. 24-36.

Ascherson, P. (1879) ‘Botanisch-ethnographische Notizen aus Guinea. Aus den Aufzeichnungen

von Thonning in Schumacher’s Beskrivelse af Guineiske Planter’, Zeitschrift für Ethnologie, 11, pp.
231-255.

Asumadu-Sarkodie, S., Owusu, P. A. and Rufangura, P. (2015) ‘Impact analysis of flood in Accra,
Ghana’, Advances in Applied Science Research, 6(9), pp. 53-78.

Attari, S. Z., Krantz, D. H. and Weber, E. U. (2016) ‘Statements about climate researchers’ car-
bon footprints affect their credibility and the impact of their advice’, Climatic Change, 138(1-2), pp.
325-338.

Austin, G. (2013) ‘Commercial agriculture & the ending of slave-trading and slavery in West Afri-
ca, 1780s-1920s’, in Law, R., Schwarz, S. and Strickrodt, S. (eds.) Commercial Agriculture, the Slave
Trade and Slavery in Atlantic Africa. Woodbridge: Boydell & Brewer, pp. 243-265.

Austin, M. E. (2002) ‘Partnership opportunities in neighborhood tree planting initiatives: Building

from local knowledge’, Journal of Arboriculture, 28(4), pp. 178-186.

Avolio, M. L., Pataki, D. E., Pincetl, S., Gillespie, T. W., Jenerette, G. D. and McCarthy, H. R.
(2015) ‘Understanding preferences for tree attributes: the relative effects of socio-economic and
local environmental factors’, Urban ecosystems, 18(1), pp. 73-86.

Aziato, L. and Antwi, H. O. (2016) ‘Facilitators and barriers of herbal medicine use in Accra,
Ghana: an inductive exploratory study’, BMC complementary and alternative medicine, 16(1), pp. 142.

Babulo, B., Muys, B., Nega, F., Tollens, E., Nyssen, J., Deckers, J. and Mathijs, E. (2009) ‘The
economic contribution of forest resource use to rural livelihoods in Tigray, Northern Ethiopia’,
Forest policy and Economics, 11(2), pp. 109-117.

Balée, W. (2006) ‘The research program of historical ecology’, Annu. Rev. Anthropol., 35, pp. 75-
98.

Balmford, A., Rodrigues, A., Walpole, M., ten Brink, P., Kettunen, M., Braat, L. and de Groot, R.
(2008) Review on the economics of biodiversity loss: Scoping the science. Brussels: European Commission.

Barron, S., Sheppard, S. R. and Condon, P. M. (2016) ‘Urban forest indicators for planning and
designing future forests’, Forests, 7(9), 208.

Barry, T. E. (1987) ‘The Development of the Hierarchy of Effects: An Historical Perspective’,

Current Issues and Research in Advertising, 10(1-2), pp. 251-295.

Beall, J. (1995) ‘Social security and social networks among the urban poor in Pakistan’, Habitat In-
ternational, 19(4), pp. 427-445.

Beall, J. (2001) ‘Valuing social resources or capitalizing on them? Limits to pro-poor urban gov-
ernance in nine cities of the south’, International Planning Studies, 6(4), pp. 357-375.

270
Beall, J. and Fox, S. (2007) Urban Poverty and Development in the 21st Century: Towards an Inclusive and
Sustainable World. Oxford: Oxfam.

Beaman, L. and Dillon, A. (2012) ‘Do household definitions matter in survey design? Results
from a randomized survey experiment in Mali’, Journal of Development Economics, 98(1), pp. 124-135.

Bentsen, P., Lindholst, A. C. and Konijnendijk, C. C. (2010) ‘Reviewing eight years of Urban
Forestry & Urban Greening: Taking stock, looking ahead’, Urban Forestry & Urban Greening, 9(4),
pp. 273-280.

Bigon, L. (2013) ‘Garden cities in colonial Africa: a note on historiography’, Planning Perspectives,
28(3), pp. 477-485.

Bird, C. M. (2005) ‘How I stopped dreading and learned to love transcription’, Qualitative inquiry,
11(2), pp. 226-248.

Boadi, K. O. and Kuitunen, M. (2005) ‘Environment, wealth, inequality and the burden of disease
in the Accra metropolitan area, Ghana’, International Journal of Environmental Health Research, 15(3),
pp. 193-206.

Boahen, A. (1966) ‘A New Look at the History of Ghana’, African Affairs, Vol. 65(260), pp. 212-
222.

Boone, C. G., Cadenasso, M. L., Grove, J. M., Schwarz, K. and Buckley, G. L. (2010) ‘Landscape,
vegetation characteristics, and group identity in an urban and suburban watershed: why the 60s
matter’, Urban Ecosystems, 13(3), pp. 255-271.

Borokini, I. (2012) ‘Diversity, distribution and utilization of urban trees in Ibadan metropolis,
southwest Nigeria’, Nature & Faune, 26(1), pp. 54-59.

Bosman, W. (2011) A New and Accurate Description of the Coast of Guinea: Divided into the Gold, the
Slave, and the Ivory Coasts. Cambridge: Cambridge University Press.

Botes, L. and van Rensburg, D. (2000) ‘Community participation in development: nine plagues
and twelve commandments’, Community development journal, 35(1), pp. 41-58.

Bourne, L. and Walker, D. H. (2006) ‘Visualizing stakeholder influence – Two Australian exam-
ples’, Project Management Journal, 37(1), pp. 5-21.

Boy, G. and Witt, A. (2013) ‘Invasive Alien Plants and their Management in Africa’, CABI Africa.
Available at: https://www.cabi.org/Uploads/CABI/publishing/promotional-materials/african-
invasives-book.pdf (Accessed: 03.11.2020).

Boyatzis, R. E. (1998) Transforming qualitative information: Thematic analysis and code development. Lon-
don: SAGE.

Boyd, J. and Banzhaf, S. (2007) ‘What are ecosystem services? The need for standardized envi-
ronmental accounting units’, Ecological economics, 63(2-3), pp. 616-626.

Braun, V. and Clarke, V. (2006) ‘Using thematic analysis in psychology’, Qualitative research in psy-
chology, 3(2), pp. 77-101.

271
Braun, V. and Clarke, V. (2013) Successful qualitative research: A practical guide for beginners. London:
SAGE.

Braun, V. and Clarke, V. (2019) ‘Reflecting on reflexive thematic analysis’, Qualitative Research in
Sport, Exercise and Health, 11(4), pp. 589-597.

Braun, Y. (2020) ‘Lesotho’s white gold: the political ecology of temporality and the economy of
anticipation in resource extraction and large dam infrastructural projects’, Journal of Political Ecolo-
gy, Vol. 27, pp. 853-876.

Briggs, J. and Yeboah, I. E. A. (2001) ‘Structural adjustment and the contemporary sub-Saharan
African city’, Area, 33(1), pp. 18-26.

Brown, I. K. (2007) Wisconsin Statewide Urban Forest Assessment: Development and Implementation. M.Sc.
Thesis. University of Wisconsin. Available at:
https://www.uwsp.edu/cnr/Documents/GraduateProgramDocument/Urban%20forest%20the
sis.pdf (Accessed: 11.09.2020).

Brown, T. C., Bergstrom, J. C., and Loomis, J. B. (2007) ‘Defining, Valuing, and Providing Eco-
system Goods and Services’, Natural Resources Journal, 42(7), pp. 329-376.

Bruce-Myers, J. M. (1927) ‘The Origin of the Gãs: Part I’, Journal of the Royal African Society, Vol.
27(105), pp. 69-76.

Brydon, L. (1999) ‘‘With a Little Bit of Luck...’ Coping with Adjustment in Urban Ghana, 1975-
90’, Africa, 69(3), pp. 366-385.

Bryson, J. M. (2004) ‘What to do when stakeholders matter: stakeholder identification and analy-
sis techniques’, Public management review, 6(1), pp. 21-53.

Burton, R. F. (2011) Wanderings in West Africa from Liverpool to Fernando Po, Volume 2. Cambridge:
Cambridge University Press.

Burton, R. F. and Cameron, V. L. (2011) To the Gold Coast for Gold, Volume 2. Cambridge: Cam-
bridge University Press.

Bürgi, M., Östlund, L. and Mladenoff, D. J. (2017) ‘Legacy effects of human land use: ecosystems
as time-lagged systems’, Ecosystems, 20(1), pp. 94-103.

CAB International (2004) ‘Prevention and Management of Alien Invasive Species: Forging Co-
operation throughout West Africa’, Proceedings of a workshop held in Accra, Ghana, 9-11
March 2004. Available at: http://www.issg.org/pdf/publications/gisp/resources/wafrica-en.pdf
(Accessed: 03.11.2020).

Camacho-Cervantes, M., Schondube, J. E., Castillo, A. and MacGregor-Fors, I. (2014) ‘How do

people perceive urban trees? Assessing likes and dislikes in relation to the trees of a city’, Urban
ecosystems, 17(3), pp. 761-773.

Cameron, R. (2011) ‘Mixed methods research: The five Ps framework’, Electronic Journal of Business
Research Methods, 9(2), pp. 96-108.

272
Campbell, S. (1996) ‘Green cities, growing cities, just cities?: Urban planning and the contradic-
tions of sustainable development’, Journal of the American Planning Association, 62(3), pp. 296-312.

Cariñanos, P. and Casares-Porcel, M. (2011) ‘Urban green zones and related pollen allergy: A re-
view. Some guidelines for designing spaces with low allergy impact’, Landscape and urban planning,
101(3), pp. 205-214.

Chambers, R. and Conway, G. (1992) Sustainable rural livelihoods: practical concepts for the 21st century.
IDS Discussion Paper 296, Brighton: Institute of Development Studies, University of Sussex.

Chen, K., Zhang, T., Liu, F., Zhang, Y. and Song, Y. (2021) ‘How Does Urban Green Space Im-
pact Residents’ Mental Health: A Literature Review of Mediators’, International Journal of Environ-
mental Research and Public Health, Vol. 18, 11746.

CHF International and AMA (2010) Accra Poverty Map: A Guide to Urban Poverty Reduction in Accra.
Accra: CHF International.

Chishaleshale, M., Shackleton, C. M., Gambiza, J. and Gumbo, D. (2015) ‘The prevalence of
planning and management frameworks for trees and green spaces in urban areas of South Africa’,
Urban Forestry & Urban Greening, 14(4), pp. 817-825.

Christensen, N. L. (1989) ‘Landscape history and ecological change’, Journal of Forest History, 33(3),
pp. 116-125.

City of Alexandria (2009) Urban Forestry Master Plan. Available at:

https://www.alexandriava.gov/uploadedFiles/planning/info/masterplan/UrbanForestryMasterP
lanFinal2009.pdf (Accessed: 15.09.2020).

City of New Westminster (2015) Urban Forest Management Strategy. Available at:
https://www.newwestcity.ca/database/files/library/Urban_forest_management_strategy_FINA
Lreduced_size.pdf (Accessed: 15.09.2020).

Clark, K. H. and Nicholas, K. A. (2013) ‘Introducing urban food forestry: a multifunctional ap-
proach to increase food security and provide ecosystem services’, Landscape Ecology, 28(9), pp.
1649-1669.

Clarke, P. M., Fiebig, D. G. and Gerdtham, U. G. (2008) ‘Optimal recall length in survey design’,
Journal of health economics, 27(5), pp. 1275-1284.

Cline-Cole, R. and Harrison, G. (2010), ‘It is (always) the political economy, stupid’, Review of Afri-
can Political Economy, 37(126), pp. 395-402.

Cline-Cole, R. and Maconachie, R. (2016) ‘Wood energy interventions and development in Kano,
Nigeria: A longitudinal, ‘situated’ perspective’, Land Use Policy, 52, pp. 163-173.

Cobbinah, P. B., Asibey, M. O., Zuneidu, M. A. and Erdiaw-Kwasie, M. O. (2021) ‘Accommo-

dating green spaces in cities: Perceptions and attitudes in slums’, Cities, 111, 103094.

Coder, K. D. (1996) Identified Benefits of Community Trees and Forests. University of Georgia Coopera-
tive Extension Service, Forest Resources Publication FOR96-39. Available at:

273
https://nfs.unl.edu/documents/communityforestry/coderbenefitsofcommtrees.pdf (Accessed:
15.09.2020).

Codjoe, S. N. A., Okutu, D. and Abu, M. (2016) ‘Urban household characteristics and dietary di-
versity: an analysis of food security in Accra, Ghana’, Food and nutrition bulletin, 37(2), pp. 202-218.

Cohen, B. (2004) ‘Urban growth in developing countries: a review of current trends and a caution
regarding existing forecasts’, World development, 32(1), pp. 23-51.

Coleman, J. S. (1988) ‘Social capital in the creation of human capital’, American journal of sociology,
94, S95-S120.

Coley, R. L., Sullivan, W. C. and Kuo, F. E. (1997) ‘Where does community grow? The social
context created by nature in urban public housing’, Environment and behavior, 29(4), pp. 468-494.

Connelly, L. M. and Peltzer, J. N. (2016) ‘Underdeveloped themes in qualitative research: Rela-

tionship with interviews and analysis’, Clinical nurse specialist, 30(1), pp. 52-57.

Constanza, R. (2008) ‘Ecosystem services: Multiple classification systems are needed’, Biological
Conservation, 141, pp. 350-352.

Constanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem,
S., O’Neill, R. V., Paruelo, J., Raskin, R. G., Sutton, P. and van den Belt, M. (1997) ‘The value of
the world’s ecosystem services and natural capital’, Nature, 387(6630), pp. 253-260.

Conway, D., Li, C. Q., Wolch, J., Kahle, C. and Jerrett, M. (2010) ‘A spatial autocorrelation ap-
proach for examining the effects of urban greenspace on residential property values’, The Journal
of Real Estate Finance and Economics, 41(2), pp. 150-169.

Conway, T. M. (2016) ‘Tending their urban forest: Residents’ motivations for tree planting and
removal’, Urban Forestry & Urban Greening, 17, pp. 23-32.

Cook, E., Hall, S. and Larson, K. (2012) ‘Residential landscapes as social-ecological systems: a
synthesis of multi-scalar interactions between people and their home environment’, Urban Ecosys-
tems, 15, pp. 19-52.

Cooke, B. and Kothari, U. (eds.) (2001) Participation: The new tyranny?. London: Zed books.

Cornwall, A. (2002) Making spaces, changing places. IDS Working Paper 170. Available at:
https://www.powercube.net/wp-
content/uploads/2009/11/making_spaces_changing_places.pdf (Accessed: 17.09.2020).

Cotonou Agreement (2000) ‘2000/483/EC: Partnership agreement between the members of the
African, Caribbean and Pacific Group of States of the one part, and the European Community
and its Member States, of the other part’, signed in Cotonou on 23 June 2000.

Couch, C. (2016) Urban Planning: An Introduction. London: Palgrave.

Cousins, J. (2021) ‘Justice in nature-based solutions: Research and pathways’, Ecological Economics,
106874.

274
Creswell, J. W. and Plano Clark, V. L. (2007) Designing and conducting mixed methods research. London:
SAGE.

Creswell, J. W., Plano Clark, V. L., Gutmann, M. L. and Hanson, W. E. (2003) ‘Advanced Mixed
Methods Research Designs’, in Tashakkori, A. and Teddlie, C. (eds.) Handbook of Mixed Methods in
Social & Behavioral Research. London: SAGE, pp. 209-240.

Cruickshank, B. (1853) Eighteen Years on the Gold Coast of Africa: including an account of the native tribes,
and their intercourse with Europeans. Vol. 1, London: Hurst and Blackett.

Daily, C. G. (1997) ‘Introduction: What are ecosystem services?’, in Daily, C. G. (ed.) Nature’s Ser-
vices: Societal Dependence on Natural Ecosystems. Washington: Island Press, pp. 1-10.

Daily Graphic (2013a) ‘From my Rooftop: What next after the familiarisation tours?’, by
Akordor, K., 23.04.2013.

Daily Graphic (2013b) ‘Spiritual enclave to be created in Achimota Forest’, by Enin, E.,
02.10.2013.

Daily Graphic (2013c) ‘Rescuing Achimota Forest Reserve’, no name, 18.11.2013.

Daily Graphic (2014a) ‘Forestry Commission, police in tango over Achimota Forest’, no name,
26.11.2014.

Daily Graphic (2014b) ‘Protecting Achimota Forest…Parliamentary Select C’ttee, team of po-
licemen clash’, no name, 27.11.2014.

Daily Graphic (2014c) ‘Achimota Forest to be turned into recreational park’, no name,
07.07.2014.

Daily Graphic (2015) ‘Prez Mahama inaugurates renovated Tamale Market’, by Duodu, S. and
Gyes, Z. K., 16.11.2015.

Daily Graphic (2016a) ‘Achimota Forest: Major Ecotourism destination’, no name, 17.05.2016.

Daily Graphic (2016b) ‘Development of Accra Eco Park now on Course’, no name, 22.08.2016.

Daily Graphic (2016c) ‘$1.2 billion Accra Eco-Park Project takes off’, no name, 20.08.2016.

Daily Graphic (2016d) ‘Four sue gov’t over sale of Achimota Forest’, no name, 27.04.2016.

Daily Graphic (2016e) ‘Sale of Achimota forest – four go to court’, no name, 17.05.2016.

Daily Graphic (2016f) ‘Disaster looms if Achimota Forest Ecotourism Project goes ahead –
NDPC boss’, no name, 16.06.2016.

Daily Graphic (2016g) ‘Must we cut our nose to spite our face?’, no name, 17.06.2016.

Daily Graphic (2017a) ‘Beyond the euphoria of the new area’, by Yeboah-Afari, A., 06.01.2017.

275
Daily Graphic (2017b) ‘Restoring greenery of a pounded city – Will KMA succeed this time?’, no
name, 17.07.2017.

Daily Graphic (2018) ‘Forestry Commission reviews Achimota Eco-park contract’, by Bokpe,
S.J., 03.01.2018.

Daily Guide (2017) ‘Pockets Of Indiscipline’, no name, 29.04.2017.

Day, S. D., Wiseman, P. E., Dickinson, S. B. and Harris, J. R. (2010) ‘Tree root ecology in the ur-
ban environment and implications for a sustainable rhizosphere’, Journal of Arboriculture, 36(5), pp.
193-205.

de Groot, R. S., Alkemade, R., Braat, L. Hein, L. and Willemen, L. (2010a) ‘Challenges in inte-
grating the concept of ecosystem services and values in landscape planning, management and de-
cision making’, Ecological Complexity, 7, pp. 260-272.

de Groot, R. S., Fisher, B., Christie, M., Aronson. J., Braat, L., Gowdy, J., Haines-Young, R.,
Maltby, E., Neuville, A., Polasky, S., Portela, R. and Ring, I. (2010b) ‘Integrating the ecological
and economic dimensions in biodiversity and ecosystem service valuation’, in Kumar, P. (ed.) The
Economics of Ecosystems and Biodiversity: Ecological and Economic Foundations. London: Earthscan, pp. 9-
39.

de Groot, R. S., Wilson, M. A. and Boumans, R. M. (2002) ‘A typology for the classification, de-
scription and valuation of ecosystem functions, goods and services’, Ecological economics, 41(3), pp.
393-408.

Deikumah, J. P. and Kudom, A. A. (2010) ‘Biodiversity status of urban remnant forests in Cape
coast, Ghana’, Journal of Science and Technology (Ghana), 30(3), pp. 1-8.

De Lacy, P. and Shackleton, C. M. (2017) ‘Woody plant species richness, composition and struc-
ture in urban sacred sites, Grahamstown, South Africa’, Urban Ecosystems, 20(5), pp. 1169-1179.

de Leeuw, E. D., Hox. J. J., and Dillman, D. A. (2008) ‘Mixed-mode Surveys: When and Why’, in
de Leeuw, E. D., Hox, J. J., and Dillman, D. A. (eds.) International Handbook of Survey Methodology.
London: Taylor & Francis, pp. 299-316.

Denzin, N. K. (2010) ‘Moments, mixed methods, and paradigm dialogs’, Qualitative inquiry, 16(6),
pp. 419-427.

DeSantis, L. and Ugarriza, D. N. (2000) ‘The concept of theme as used in qualitative nursing re-
search’, Western Journal of Nursing Research, 22(3), pp. 351-372.

Diffenbaugh, N. S. and Burke, M. (2019) ‘Global warming has increased global economic ine-
quality’, Proceedings of the National Academy of Sciences, 116(20), pp. 9808-9813.

Dimke, K. C., Sydnor, T. D. and Gardner, D. S. (2013) ‘The effect of landscape trees on residen-
tial property values of six communities in Cincinnati, Ohio’, Arboriculture & Urban Forestry, 39(2),
pp. 49-55.

Dionisio, K. L., Rooney, M. S., Arku, R. E., Friedman, A. B., Hughes, A. F., Vallarino, J., Agyei-
Mensah, S., Spengler, J. and Ezzati, M. (2010) ‘Within-neighborhood patterns and sources of par-

276
ticle pollution: mobile monitoring and geographic information system analysis in four communi-
ties in Accra, Ghana’, Environmental health perspectives, 118(5), pp. 607-613.

Doan, P. and Oduro, C. Y. (2012) ‘Patterns of population growth in peri‐urban Accra, Ghana’,
International Journal of Urban and Regional Research, 36(6), pp. 1306-1325.

Dobbs, C., Escobedo, F. J. and Zipperer, W. C. (2011) ‘A framework for developing urban forest
ecosystem services and goods indicators’, Landscape and urban planning, 99(3-4), pp. 196-206.

Dobbs, C., Kendal, D. and Nitschke, C. (2013) ‘The effects of land tenure and land use on the
urban forest structure and composition of Melbourne’, Urban Forestry & Urban Greening, 12(4), pp.
417-425.

Dombrow, J., Rodriguez, M. and Sirmans, C. F. (2000) ‘The market value of mature trees in sin-
gle-family housing markets’, Appraisal Journal, 68(1), pp. 39-43.

Donaldson-Selby, G., Hill, T. and Korrubel, J. (2007) ‘Photorealistic visualisation of urban green-
ing in a low-cost high-density housing settlement, Durban, South Africa’, Urban Forestry & Urban
Greening, 6(1), pp. 3-14.

Drago, W. A. and Folker, C. A. (1999) ‘Achilles revisited: the impact of incompetencies on firm
performance’, Management Research News, 22(8), pp. 18-25.

Driscoll, A. N., Ries, P. D., Tilt, J. H. and Ganio, L. M. (2015) ‘Needs and barriers to expanding
urban forestry programs: An assessment of community officials and program managers in the
Portland–Vancouver metropolitan region’, Urban Forestry & Urban Greening, 14(1), pp. 48-55.

Durlauf, S. N. (1999) ‘The case “against” social capital’, Focus, 20(3), pp. 1-5.

Dwyer, J. F., McPherson, E. G., Schroeder, H. W. and Rowntree, R. A. (1992) ‘Assessing the
benefits and costs of the urban forest’, Journal of Arboriculture, 18, pp. 227-234.

Dwyer, J. F., Nowak, D. J. and Noble, M. H. (2003) ‘Sustaining urban forests’, Journal of Arboricul-
ture, 29(1), pp. 49-55.

Dwyer, J. F., Schroeder, H. W., Louviere, J. J. and Anderson, D. H. (1989) ‘Urbanites willingness
to pay for trees and forests in recreation areas’, Journal of Arboriculture, 15(10), pp. 247-252.

Elmendorf, W. F., Cotrone, V. J. and Mullen, J. T. (2003) ‘Trends in urban forestry practices,
programs, and sustainability: Contrasting a Pennsylvania, US, study’, Arboriculture & Urban Forest-
ry, 29(4), pp. 237-248.

Endreny, T. A. (2018) ‘Strategically growing the urban forest will improve our world’, Nature
communications, 9(1), 1160.

England, K. V. (1994) ‘Getting personal: Reflexivity, positionality, and feminist research’, The pro-
fessional geographer, 46(1), pp. 80-89.

Escobedo, F. J., Adams, D. C. and Timilsina, N. (2015) ‘Urban forest structure effects on proper-
ty value’, Ecosystem Services, 12, pp. 209-217.

277
Escobedo, F. J., Kroeger, T. and Wagner, J. E. (2011) ‘Urban forests and pollution mitigation:
Analyzing ecosystem services and disservices’, Environmental pollution, 159(8-9), pp. 2078-2087.

Escobedo, F. J., Nowak, D. J., Wagner, J. E., De la Maza, C. L., Rodríguez, M., Crane, D. E. and
Hernández, J. (2006) ‘The socioeconomics and management of Santiago de Chile’s public urban
forests’, Urban Forestry & Urban Greening, 4(3-4), pp. 105-114.

Farrington, J., Ramasut, T. and Walker, J. (2002) Sustainable Livelihoods Approaches in Urban Areas:
General Lessons, with Illustrations from Indian Cases. ODI Working Paper No. 162. Available at:
https://www.odi.org/sites/odi.org.uk/files/odi-assets/publications-opinion-files/2706.pdf (Ac-
cessed: 15.09.2020).

Feilzer, M. (2010) ‘Doing mixed methods research pragmatically: Implications for the rediscovery
of pragmatism as a research paradigm’, Journal of mixed methods research, 4(1), pp. 6-16.

Fent, A. (2020) ‘The anticipatory politics of dispossession in a Senegalese mining negotiation’,

Journal of Political Ecology, Vol. 27, pp. 877-897.

Fent, A. and Kojola, E. (2020) ‘Political ecologies of time and temporality in resource extraction’,
Journal of Political Ecology, Vol. 27, pp. 819-829.

Ferragina, E. and Arrigoni, A. (2017) ‘The rise and fall of social capital: requiem for a theory?’,
Political Studies Review, 15(3), pp. 355-367.

Filep, B. (2009) ‘Interview and translation strategies: coping with multilingual settings and data’,
Social Geography, 4(1), pp. 59-70.

Fine, B. (2002) ‘It ain’t social, it ain’t capital and it ain’t Africa’, reprint without page numbers
from Studia Africana, 13, pp. 18-33.

Fisher, B. and Turner, R. K. (2008) ‘Ecosystem services: classification for valuation’, Biological con-
servation, 141(5), pp. 1167-1169.

Fisher, B., Turner, R. K. and Morling, P. (2009) ‘Defining and classifying ecosystem services for
decision making’, Ecological economics, 68(3), pp. 643-653.

Fisher, M. (2004) ‘Household welfare and forest dependence in Southern Malawi’, Environment
and Development Economics, 9(2), pp. 135-154.

Forrest, M. and Konijnendijk, C. (2005) ‘A History of Urban Forests and Trees in Europe’, in
Konijnendijk, C., Nilsson, K., Randrup, T. and Schipperijn, J. (eds.) Urban Forests and Trees. Berlin:
Springer, pp. 23-48.

Freeman, R. E. (1984) Strategic Management: A Stakeholder Approach. Boston: Pitman.

Gabriel, K. M. and Endlicher, W. R. (2011) ‘Urban and rural mortality rates during heat waves in
Berlin and Brandenburg, Germany’, Environmental pollution, 159(8-9), pp. 2044-2050.

Gaisie, E., Kim, H. M. and Han, S. S. (2019) ‘Accra towards a city-region: Devolution, spatial de-
velopment and urban challenges’, Cities, 95, 102398.

278
Gakou, M., Force, J. E. and McLaughlin, W. J. (1994) ‘Non-timber forest products in rural Mali:
a study of villager use’, Agroforestry systems, 28(3), pp. 213-226.

Gale, N. K., Heath, G., Cameron, E., Rashid, S. and Redwood, S. (2013) ‘Using the framework
method for the analysis of qualitative data in multi-disciplinary health research’, BMC medical re-
search methodology, 13(1), pp. 1-8.

Gaventa, J. (2004) ‘Towards Participatory Local Governance: Assessing the Transformative Pos-
sibilities’, in Hickey, S. and Mohan, G. (eds.) Participation : From Tyranny to Transformation? Exploring
New Approaches to Participation in Development. London: Zed Books, pp. 25-41.

Ghanaian Times (2015) ‘Parks and gardens dept. cries for assistance’, by Abdul-Rahaman, S.,
16.06.2015.

Ghanaian Times (2017) ‘Achimota Forest shouldn’t be turned into Eco Park’, by Frimpong-
Boateng, K., 13.03.2017.

Ghana Statistical Service (2014a) Ghana Living Standards Survey Round 6 (GLSS 6) Main Report.
Available at:
https://statsghana.gov.gh/gssmain/fileUpload/Living%20conditions/GLSS6_Main%20Report.
pdf (Accessed: 28.12.2020).

Ghana Statistical Service (2014b) 2010 Population and Housing Census Report: Urbanisation. Available
at:
https://www.statsghana.gov.gh/gssmain/fileUpload/pressrelease/Urbanisation%20in%20Ghan
a.pdf (Accessed: 18.09.2020).

Gibbons, K. H. and Ryan, C. M. (2015) ‘Characterizing comprehensiveness of urban forest man-

agement plans in Washington State’, Urban Forestry & Urban Greening, 14(3), pp. 615-624.

Giddings, L. S. and Grant, B. M. (2007) ‘A Trojan horse for positivism?: A critique of mixed
methods research’, Advances in nursing science, 30(1), pp. 52-60.

Gilstad-Hayden, K., Wallace, L. R., Carroll-Scott, A., Meyer, S. R., Barbo, S., Murphy-Dunning,
C. and Ickovics, J. R. (2015) ‘Research note: Greater tree canopy cover is associated with lower
rates of both violent and property crime in New Haven, CT’, Landscape and Urban Planning, 143,
pp. 248-253.

Girish, K. and Shankara, B. S. (2008) ‘Neem – a green treasure’, Electronic journal of Biology, 4(3),
pp. 102-111.

Gocking, R. S. (2005) The History of Ghana, London: Greenwood Press.

Godschalk, D. R. and Mills, W. E. (1966) ‘A collaborative approach to planning through urban

activities’, Journal of the American Institute of Planners, 32(2), pp. 86-95.

Gómez-Baggethun, E. and Barton, D. N. (2013) ‘Classifying and valuing ecosystem services for
urban planning’, Ecological economics, 86, pp. 235-245.

Government of Ghana (1964) Seven-Year Development Plan. Available at: https://s3-us-west-

2.amazonaws.com/new-ndpc-

279
static1/CACHES/PUBLICATIONS/2017/11/03/SevenYearDevtPlan.pdf (Accessed:
16.09.2020).

Grado, S. C., Grebner, D. L., Measells, M. K. and Husak, A. L. (2006) ‘Status, needs, and
knowledge levels of Mississippi’s communities relative to urban forestry’, Journal of Arboriculture,
32(1), pp. 24-32.

Grant, R. and Yankson, P. (2003) ‘Accra’, Cities, 20(1), pp. 65-74.

Grant, R. M. (2008) ‘Why Strategy Teaching Should Be Theory Based’ Journal of Management In-
quiry, 17(4), pp. 276-281.

Greene, C. S., Millward, A. A. and Ceh, B. (2011) ‘Who is likely to plant a tree? The use of public
socio-demographic data to characterize client participants in a private urban forestation program’,
Urban Forestry & Urban Greening, 10(1), pp. 29-38.

Greene, J. C. and Caracelli, V. J. (1997) ‘Defining and Describing the Paradigm Issue in Mixed-
Method Evaluation’, New Directions for Evaluations, 74, pp. 5-17.

Groninger, J. W., Close, D. D. and Basman, C. M. (2002) ‘Can small, rural communities practice
urban forestry?’, Journal of Forestry, 100(1), pp. 23-28.

Grosh, M. and Glewwe, P. (2000) Designing household survey questionnaires for developing countries: Les-
sons from 15 years of the Living Standards Measurement Study. Washington: The International Bank for
Reconstruction and Development/The World Bank.

Grove, J. M., Locke, D. H. and O’Neil-Dunne, J. P. (2014) ‘An ecology of prestige in New York
City: Examining the relationships among population density, socio-economic status, group identi-
ty, and residential canopy cover’, Environmental management, 54(3), pp. 402-419.

Guba, E. G. and Lincoln, Y. S. (1994) ‘Competing paradigms in qualitative research’, in Denzin,

N. K. and Lincoln, Y. S. (eds.) Handbook of Qualitative Research. London: SAGE, pp. 105-117.

Guerra, N. G., Williams, K. R. and Sadek, S. (2011) ‘Understanding bullying and victimization
during childhood and adolescence: A mixed methods study’, Child development, 82(1), pp. 295-310.

Gulsrud, N. M., Hertzog, K. and Shears, I. (2018) ‘Innovative urban forestry governance in Mel-
bourne?: Investigating “green placemaking” as a nature-based solution’, Environmental Research,
161, pp. 158-167.

Gundersen, V., Frivold, L. H., Löfström, I., Jørgensen, B. B., Falck, J. and Øyen, B. H. (2005)
‘Urban woodland management – the case of 13 major Nordic cities’, Urban Forestry & Urban
Greening, 3(3-4), pp. 189-202.

Gurney, J. N. (1985) ‘Not one of the guys: The female researcher in a male-dominated setting’,
Qualitative sociology, 8(1), pp. 42-62.

Guyer, J. (1981) ‘Household and Community in African Studies’, African Studies Review, Vol. 24(2-
3), pp. 87-137.

280
Gwedla, N. and Shackleton, C. M. (2015) ‘The development visions and attitudes towards urban
forestry of officials responsible for greening in South African towns’, Land Use Policy, 42, pp. 17-
26.

Gwedla, N. and Shackleton, C. M. (2019) ‘Perceptions and preferences for urban trees across
multiple socio-economic contexts in the Eastern Cape, South Africa’, Landscape and Urban Plan-
ning, 189, pp. 225-234.

Habraken, N. J. (1986) ‘Towards a new professional role’, Design Studies, 7(3), pp. 139-143.

Hai, H. L. and Tsou, M. C. (2009) ‘Strategic development for the department of information
management of Shih Chien University Kaohsiung campus in Taiwan by a quantifiable SWOT
analysis’, Journal of Information and Optimization Sciences, 30(1), pp. 87-107.

Haines-Young, R. and Potschin, M. (2010) ‘The links between biodiversity, ecosystem services
and human well-being’, in Raffaelli, D. G. and Frid, C. L. J. (eds.) Ecosystem Ecology: A New Synthe-
sis. Cambridge, Cambridge University Press, pp. 110-139.

Hair, P. E. H., Jones, A. and Law, R. (1992) Barbot on Guinea: The writings of Jean Barbot on West Af-
rica, 1678-1712, Vol. 2, Farnham: Ashgate.

Hale, J. D., Pugh, T. A., Sadler, J. P., Boyko, C. T., Brown, J., Caputo, S., Caserio, M., Coles, R.,
Farmani, R., Hales, C., Horsey, R., Hunt, D. V. L., Leach, J. M., Rogers, C. D. F. and MacKenzie,
A.R. (2015) ‘Delivering a multi-functional and resilient urban forest’, Sustainability, 7(4), pp. 4600-
4624.

Hall, R. F. (2013) ‘Mixed Methods: In Search of a Paradigm’, in Lê, T. and Lê, Q. (eds.) Conducting
Research in a Changing and Challenging World. Hauppauge: Nova Science Publisher, pp. 71-78.

Hallegatte, S., Bangalore, M., Bonzanigo, L., Fay, M., Kane, T., Narloch, U., Rozenberg, J.,
Treguer, D. and Vogt-Schilb, A. (2016) Shock Waves: Managing the Impacts of Climate Change on Pov-
erty. Washington: The International Bank for Reconstruction and Development/The World
Bank.

Hanson, K. T. (2005) ‘Landscapes of survival and escape: social networking and urban liveli-
hoods in Ghana’, Environment and Planning A, 37(7), pp. 1291-1310.

Harriss, J. and de Renzio, P (1997) ‘‘Missing link’ or analytically missing?: The concept of social
capital’, Journal of International Development, 9(7), pp. 919-937.

Harrison, H., Birks, M., Franklin, R., and Mills, J. (2017) ‘Case study research: Foundations and
methodological orientations’, Forum Qualitative Sozialforschung/Forum: Qualitative Social Research ,
18(1).

Harvey, D. (2005) A brief history of neoliberalism. Oxford: Oxford University Press.

Hashmat, I., Azad, H. and Ahmed, A. (2012) ‘Neem (Azadirachta indica A. Juss) – A nature’s
drugstore: an overview’, International Research Journal of Biological Sciences, 1(6), pp. 76-79.

281
Hauer, R. J. (2005) Urban Forestry and Urban Forest Capacity: Defining Capacity and Models of Capacity
Building. Ph.D. Thesis. University of Minnesota. Available at: https://search.proquest.com (Ac-
cessed: 15.09.2020).

Hauer, R. J., Miller, R. W., Werner, L. P. and Konijnendijk, C. C. (2017) ‘The history of trees in
the city’, in Ferrini, F., Konijnendijk, C, C. and Fini, A. (eds.) Routledge Handbook of Urban Forestry.
London: Routledge, pp. 17-32.

Healey, P. (1997) Collaborative planning: Shaping Places in Fragmented Societies. Vancouver: UBC Press.

Helms, J. A. (1998) The Dictionary of Forestry. Bethesda: Society of American foresters.

Helms, M. M. and Nixon, J. (2010) ‘Exploring SWOT analysis – where are we now? A review of
academic research from the last decade’, Journal of Strategy and Management, 3(3), pp. 215-251.

Heynen, N. C. (2003) ‘The scalar production of injustice within the urban forest’, Antipode, 35(5),
pp. 980-998.

Heynen, N. C. and Lindsey, G. (2003) ‘Correlates of urban forest canopy cover: implications for
local public works’, Public Works Management & Policy, 8(1), pp. 33-47.

Heynen, N. C., Kaika, M. and Swyngedouw, E. (2006) ‘Urban political ecology: politicizing the
production of urban natures’, in Heynen, N. C., Kaika, M. and Swyngedouw, E. (eds.) In the Na-
ture of Cities: Urban political ecology and the politics of urban metabolism. London: Routledge, pp. 1-20.

Hilbert, D. R., Roman, L. A., Koeser, A. K., Vogt, J. and van Doorn, N. S. (2019) ‘Urban tree
mortality: a literature review’, Arboriculture & Urban Forestry, 45(5), pp. 167-200.

Hill, T. and Westbrook, R. (1997) ‘SWOT analysis: it’s time for a product recall’, Long range plan-
ning, 30(1), pp. 46-52.

Hofmann, M., Westermann, J. R., Kowarik, I. and Van der Meer, E. (2012) ‘Perceptions of parks
and urban derelict land by landscape planners and residents’, Urban Forestry & Urban Greening,
11(3), pp. 303-312.

Holtan, M. T., Dieterlen, S. L. and Sullivan, W. C. (2015) ‘Social life under cover: tree canopy and
social capital in Baltimore, Maryland’, Environment and behavior, 47(5), pp. 502-525.

Hope, D., Gries, C., Zhu, W., Fagan, W. F., Redman, C. L., Grimm, N. B., Nelson, A. L., Martin,
C. and Kinzig, A. (2003) ‘Socioeconomics Drive Urban Plant Diversity’, Proceedings of the National
Academy of Sciences of the United States of America,100, pp. 8788-8792.

Hosek, L. (2019) ‘Tree Cover of Accra’s Neighbourhoods – a Green Divide’, Urban Forum, 30(3),
pp. 341-355.

Howe, K. R. (1988) ‘Against the quantitative-qualitative incompatibility thesis or dogmas die

hard’, Educational researcher, 17(8), pp. 10-16.

Hungerford, H. and Moussa, Y. (2017) ‘Seeing the (urban) forest through the trees: governance
and household trees in Niamey, Niger’, African Geographical Review, 36(3), pp. 286-304.

282
Hunter, I. R. (2001) ‘What do people want from urban forestry? – The European experience’,
Urban Ecosystems, 5(4), pp. 277-284.

Hurley, P. T. and Emery, M. R. (2018) ‘Locating provisioning ecosystem services in urban for-
ests: Forageable woody species in New York City, USA’, Landscape and Urban Planning, 170, pp.
266-275.

Intergovernmental Panel on Climate Change (2013) Climate change 2013: the physical science basis.
Cambridge: Cambridge University Press.

Islam, S. N. and Winkel, J. (2017) Climate Change and Social Inequality. DESA Working Paper No.
152, New York: UN. Available at: https://www.un.org/esa/desa/papers/2017/wp152_2017.pdf
(Accessed: 16.09.2020).

Ivankova, N. V. and Stick, S. L. (2007) ‘Students’ persistence in a distributed doctoral program in

educational leadership in higher education: A mixed methods study’, Research in Higher Education,
48(1), pp. 93-135.

Jackson, I. (2019) Sharing Stories from Jamestown: The creation of mercantile Accra. Liverpool: Liverpool
School of Architecture/TAG PRESS.

Jackson, I. and Holland, J. (2014) The Architecture of Edwin Maxwell Fry and Jane Drew. Farnham:
Ashgate.

Janse, G. and Konijnendijk, C. C. (2007) ‘Communication between science, policy and citizens in
public participation in urban forestry – Experiences from the Neighbourwoods project’, Urban
Forestry & Urban Greening, 6(1), pp. 23-40.

Jefferson, G. (1984) ‘Transcription Conventions’, in Atkinson, J. and Heritage, J. (eds.) Structures

of Social Action. Cambridge: Cambridge University Press, pp. ix-xvi.

Jenks, G. F. and Caspall, F. C. (1970) ‘Error on choroplethic maps: Definition, measurement, re-
duction’, Annals of the Association of American Geographers, 61(2), pp. 217-244.

Jim, C. Y. (2003) ‘Protection of urban trees from trenching damage in compact city environ-
ments’, Cities, 20(2), pp. 87-94.

Johnson, C. (1982) ‘Political & administrative factors in urban-forestry programs’, Journal of Arbor-
iculture, 8(6), pp. 160-163.

Johnson, R. B. and Onwuegbuzie, A. J. (2004) ‘Mixed methods research: A research paradigm

whose time has come’, Educational researcher, 33(7), pp. 14-26.

Johnston, M. (1996) ‘A brief history of urban forestry in the United States’, Arboricultural Journal,
20(3), pp. 257-278.

Johnston, M. (1997a) ‘The development of urban forestry in the Republic of Ireland’, Irish Forestry
Journal, 54(2), pp. 14-32.

Johnston, M. (1997b) ‘The early development of urban forestry in Britain: part I’, Arboricultural
Journal, 21(2), pp. 107-126.

283
Johnston, M. (1997c) ‘The early development of urban forestry in Britain: part II’, Arboricultural
Journal, 21(4), pp. 317-330.

Jones, N., Collins, K., Vaughan, J., Benedikz, T. and Brosnan, J. (2005) ‘The Role of Partnerships
in Urban Forestry’, in Konijnendijk, C., Nilsson, K., Randrup, T. and Schipperijn, J. (eds.) Urban
Forests and Trees. Berlin: Springer, pp. 187-205.

Kaimowitz, D: (2003) ‘Not by Bread Alone...Forests and Rural Livelihoods in Sub-Saharan Afri-
ca’, Forests in Poverty Reduction Strategies: Capturing the Potential. Tuusula, Finland, 1-4 October, 2002.
Joensuu: European Forest Institute. Available at:
https://efi.int/sites/default/files/files/publication-bank/2018/proc47_net.pdf#page=45 (Ac-
cessed: 16.09.2020).

Kaiser, K. (2009) ‘Protecting Respondent Confidentiality in Qualitative Research’, Qualitative

Health Research, 19(11), pp. 1632-1641.

Kalaba, F. K., Chirwa, P. W. and Prozesky, H. (2009) ‘The contribution of indigenous fruit trees
in sustaining rural livelihoods and conservation of natural resources’, Journal of Horticulture and For-
estry, 1(1), pp. 1-6.

Kamanga, P., Vedeld, P. and Sjaastad, E. (2009) ‘Forest incomes and rural livelihoods in Chirad-
zulu District, Malawi’, Ecological Economics, 68(3), pp. 613-624.

Kangas, J., Kurttila, M., Kajanus, M. and Kangas, A. (2003) ‘Evaluating the management strate-
gies of a forestland estate – the SOS approach’, Journal of environmental management, 69(4), pp. 349-
358.

Kaoma, H. and Shackleton, C. M. (2014) ‘Collection of urban tree products by households in

poorer residential areas of three South African towns’, Urban Forestry & Urban Greening, 13(2), pp.
244-252.

Kaoma, H. and Shackleton, C. M. (2015) ‘The direct-use value of urban tree non-timber forest
products to household income in poorer suburbs in South African towns’, Forest Policy and Eco-
nomics, 61, pp. 104-112.

Karlsson, B. G. (2015) ‘Political Ecology: Anthropological Perspectives’, in Wright, J. D. (ed.) In-

ternational Encyclopedia of the Social & Behavioral Sciences. Oxford: Elsevier, pp. 350-355.

Kayode, J. (2010) ‘Demographic survey of tree species in urban centers of Ekiti State, Nigeria’,
Journal of Sustainable Forestry, 29(5), pp. 477-485.

Kimemia, D. and Annegarn, H. (2011) ‘An urban biomass energy economy in Johannesburg,
South Africa’, Energy for Sustainable Development, 15, pp. 382-387.

King, N. and Brooks, J. M. (2017) Template Analysis for Business and Management Students. London:
SAGE.

Kirkpatrick, J. B., Davison, A. and Daniels, G. D. (2012) ‘Resident attitudes towards trees influ-
ence the planting and removal of different types of trees in eastern Australian cities’, Landscape
and urban planning, 107(2), pp. 147-158.

284
Kirkpatrick, J. B., Davison, A. and Daniels, G. D. (2013a) ‘Sinners, scapegoats or fashion victims?
Understanding the deaths of trees in the green city’, Geoforum, 48, pp. 165-176.

Kirkpatrick, J. B., Davison, A. and Harwood, A. (2013b) ‘How tree professionals perceive trees
and conflicts about trees in Australia’s urban forest’, Landscape and urban planning, 119, pp. 124-
130.

Ko, Y. (2018) ‘Trees and vegetation for residential energy conservation: A critical review for evi-
dence-based urban greening in North America’, Urban Forestry & Urban Greening, 34, pp. 318-335.

Kojola, E. (2020) ‘Divergent memories and visions of the future in conflicts over mining devel-
opment’, Journal of Political Ecology, Vol. 27, pp. 898-916.

Konadu-Agyemang, K. (2000) ‘The best of times and the worst of times: structural adjustment
programs and uneven development in Africa: the case of Ghana’, The Professional Geographer, 52(3),
pp. 469-483.

Kondo, M., Fluehr, J., McKeon, T. and Branas, C. (2018) ‘Urban Green Space and Its Impact on
Human Health’, International Journal of Environmental Research and Public Health, Vol. 15, 445.

Konijnendijk, C. C. (2003) ‘A decade of urban forestry in Europe’, Forest policy and Economics, 5(2),
pp. 173-186.

Konijnendijk, C. C., Ricard, R. M., Kenney, A. and Randrup, T. B. (2006) ‘Defining urban forest-
ry –A comparative perspective of North America and Europe’, Urban Forestry & Urban Greening,
4(3-4), pp. 93-103.

Kotsila, P., Anguelovski, I., Baró, F., Langemeyer, J., Sekulova, F. and Connolly, J. (2021) ‘Na-
ture-based solutions as discursive tools and contested practices in urban nature’s neoliberalisation
processes’, Environment and Planning E: Nature and Space, Vol. 4(2), pp. 252-274.

Kowalski, J. M. and Conway, T. M. (2019) ‘Branching out: The inclusion of urban food trees in
Canadian urban forest management plans’, Urban Forestry & Urban Greening, 45, 126142.

Kozová, M., Dobšinská, Z., Pauditšová, E., Tomčíková, I. and Rakytová, I. (2018) ‘Network and
participatory governance in urban forestry: An assessment of examples from selected Slovakian
cities’, Forest Policy and Economics, 89, pp. 31-41.

Krantz, L. (2001) ‘The Sustainable Livelihood Approach to Poverty Reduction: An Introduction’,

SIDA, Division for Policy and Socio-Economic Analysis 44. Available at:
http://commdev.org/pdf/publications/The-Sustainable-Livelihood-Approach-to-Poverty-
Reduction-SIDA.pdf (Accessed: 28.01.2021).

Kronenberg, J. (2015) ‘Why not to green a city? Institutional barriers to preserving urban ecosys-
tem services’, Ecosystem services, 12, pp. 218-227.

Kropp Dakubu, M. E. (1997) Korle Meets the Sea: A Sociolinguistic History of Accra. Oxford: Oxford
University Press.

Kuehnhenrich, D. (2012) Entwicklung oder Profit? Die staatliche und private Presse in Ghana. Stuttgart:
ibidem.

285
Kuo, F. E., Bacaicoa, M. and Sullivan, W. C. (1998) ‘Transforming inner-city landscapes: Trees,
sense of safety, and preference’, Environment and behavior, 30(1), pp. 28-59.

Kuruneri-Chitepo, C. and Shackleton, C. M. (2011) ‘The distribution, abundance and composi-

tion of street trees in selected towns of the Eastern Cape, South Africa’, Urban Forestry & Urban
Greening, 10(3), pp. 247-254.

Labi, K. A. (2006) ‘European descriptions of the art and architecture of early Accra ca. 1450-
1800’, Research Review of the Institute of African Studies, Supplement 17, pp.121-136.

Landry, S. M. and Chakraborty, J. (2009) ‘Street trees and equity: evaluating the spatial distribu-
tion of an urban amenity’, Environment and Planning A: Economy and Space, 41(11), pp. 2651-2670.

Larbi, W. O. (1996) ‘Spatial planning and urban fragmentation in Accra’, Third World Planning Re-
view, 18(2), 193-214.

Lavidge, R. J. and Steiner, G. A. (1961) ‘A model for predictive measurements of advertising ef-
fectiveness’, Journal of marketing, 25(6), pp. 59-62.

Lawrence, A., De Vreese, R., Johnston, M., van den Bosch, C. C. K. and Sanesi, G. (2013) ‘Urban
forest governance: Towards a framework for comparing approaches’, Urban Forestry & Urban
Greening, 12(4), pp. 464-473.

Lawrence, H. W. (1993) ‘The neoclassical origins of modern urban forests’, Forest & Conservation
History, 37(1), pp. 26-36.

Lee, H. and Son, S. (2018) ‘Qualitative Assessment of Experience on Urban Forest Therapy Pro-
gram for Preventing Dementia of the Elderly Living Alone in Low-Income Class’, Journal of Peo-
ple, Plants, and the Environment, Vol. 21(6), pp. 565-574.

Lee, J. S., Bartolomei, L. and Pittaway, E. (2016) ‘Survey research with preliterate adult popula-
tions in post conflict situations using researcher-assisted self-completion questionnaires’, Interna-
tional Journal of Social Research Methodology, 19(6), pp. 717-730.

Lee, S. F. and Ko, A. S. O. (2000) ‘Building balanced scorecard with SWOT analysis, and imple-
menting “Sun Tzu’s The Art of Business Management Strategies” on QFD methodology’, Mana-
gerial Auditing Journal, 15(1-2), pp. 68-76.

Leech, N. L. and Onwuegbuzie, A. J. (2009) ‘A typology of mixed methods research designs’,

Quality & quantity, 43(2), pp. 265-275.

Leff, M. (2016) The sustainable urban forest: a step-by-step approach. Philadelphia: Davey Institute /
USDA Forest Service. Available at:
https://www.itreetools.org/documents/175/Sustainable_Urban_Forest_Guide_14Nov2016.pdf
(Accessed: 15.09.2020).

Leksungnoen, N., Eiadthong, W. and Kjelgren, R. (2017) ‘Thailand’s catastrophic flood: Bangkok
tree mortality as a function of taxa, habitat, and tree size’, Urban Forestry & Urban Greening, 22, pp.
111-119.

286
Le Quéré, C., Capstick, S., Corner, A., Cutting, D., Johnson, M., Minns, A., Schroeder, H., Walk-
er-Springett, K., Whitmarsh, L. and Wood, R. (2015) Towards a culture of low-carbon research for the
21st Century. Tyndall Working Paper 161. Available at:
http://tyndall.ac.uk/sites/default/files/twp161.pdf (Accessed: 16.09.2020).

Lewis, C. W. and Gillman, S. C. (2005) The Ethics Challenge in Public Service: A Problem-Solving Guide.
San Francisco: John Wiley & Sons.

Li, J. T., Qiu, J. W., Wang, X. W., Zhong, Y., Lan, C. Y. and Shu, W. S. (2006) ‘Cadmium con-
tamination in orchard soils and fruit trees and its potential health risk in Guangzhou, China’, En-
vironmental Pollution, 143(1), pp. 159-165.

Lijphart, A. (1971) ‘Comparative politics and the comparative method’, The American Political Sci-
ence Review, 65(3), pp. 682-693.

Lin, B. and Lin, Y. (2010) ‘Cooling effect of shade trees with different characteristics in a sub-
tropical urban park’, HortScience, 45(1), pp. 83-86.

Lin, N. (2000) ‘Inequality in social capital’, Contemporary sociology, 29(6), pp. 785-795.

Lipkis, A. and Lipkis, K. (1990) The simple act of planting a tree: a citizen forester’s guide to healing your
neighborhood, your city, and your world. Los Angeles: Jeremy P. Tarcher.

Liu, C. and Li, X. (2012) ‘Carbon storage and sequestration by urban forests in Shenyang, China’,
Urban Forestry & Urban Greening, 11(2), pp. 121-128.

Locke, D. H., Han, S., Kondo, M. C., Murphy-Dunning, C. and Cox, M. (2017) ‘Did community
greening reduce crime? Evidence from New Haven, CT, 1996-2007’, Landscape and Urban Plan-
ning, 161, pp. 72-79.

Lohr, V., Pearson-Mims, C., Tarnai, J. and Dillman, D. (2004) ‘How urban residents rate and
rank the benefits and problems associated with trees in cities’, Journal of Arboriculture, 30(1), pp.
28-35.

Lubbe, C. S., Siebert, S. J. and Cilliers, S. S. (2010) ‘Political legacy of South Africa affects the
plant diversity patterns of urban domestic gardens along a socio-economic gradient’, Scientific Re-
search and Essays, 5(19), pp. 2900-2910.

Lunt, I. D. and Spooner, P. G. (2005) ‘Using historical ecology to understand patterns of biodi-
versity in fragmented agricultural landscapes’, Journal of Biogeography, 32(11), pp. 1859-1873.

Lynn, M. (1997) Commerce and Economic Change in West Africa: The palm oil trade in the nineteenth centu-
ry. Cambridge: Cambridge University Press.

Lyytimäki, J. (2014) ‘Bad nature: Newspaper representations of ecosystem disservices’, Urban For-
estry & Urban Greening, 13(3), pp. 418-424.

Maas, J., Van Dillen, S. M., Verheij, R. A. and Groenewegen, P. P. (2009) ‘Social contacts as a
possible mechanism behind the relation between green space and health’, Health & place, 15(2),
pp. 586-595.

287
Mage, D., Ozolins, G., Peterson, P., Webster, A., Orthofer, R., Vandeweerd, V. and Gwynne, M.
(1996) ‘Urban air pollution in megacities of the world’, Atmospheric Environment, 30(5), pp. 681-
686.

Mahjabeen, Z., Shrestha, K. K. and Dee, J. A. (2009) ‘Rethinking community participation in ur-
ban planning: The role of disadvantaged groups in Sydney metropolitan strategy’, Australasian
Journal of Regional Studies, 15(1), pp. 45-63.

Manchester Metropolitan University (n.d.) ‘Stakeholder Analysis toolkit’. Available at:

https://www.mmu.ac.uk/media/mmuacuk/content/documents/bit/Stakeholder-analysis-
toolkit-v3.pdf (Accessed: 25.11.2020).

Mann, M. and Sehrawat, S. (2009) ‘A city with a view: the afforestation of the Delhi ridge, 1883-
1913’, Modern Asian Studies, 43(2), pp. 543-570.

Margerum, R. D. (2002) ‘Collaborative Planning: Building Consensus and Building a Distinct

Model for Practice’, Journal of Planning Education and Research, 21(3), pp. 237-253.

Mariwah, S., Osei, K. N. and Amenyo-Xa, M. S. (2017) ‘Urban land use/land cover changes in
the Tema metropolitan area (1990–2010)’, GeoJournal, 82(2), pp. 247-258.

Martin, C. A., Warren, P. S. and Kinzig, A. P. (2004) ‘Neighborhood socioeconomic status is a

useful predictor of perennial landscape vegetation in residential neighborhoods and embedded
small parks of Phoenix, AZ’, Landscape and Urban Planning, 69(4), pp. 355-368.

Masterman, M. (1970) ‘The Nature of a Paradigm’, in Lakatos, I. and Musgrave, A. (eds.) Criticism
and the Growth of Knowledge. Available at:
https://cursosfilos.files.wordpress.com/2015/08/proceedings-of-the-international-colloquium-
in-the-philosophy-of-science-london-1965-volume-4-imre-lakatos-ed-alan-musgrave-ed-criticism-
and-the-growth-of-knowledge-cambridge.pdf (Accessed: 15.09.2020).

Maxwell, D., Levin, C., Armar-Klemesu, M., Ruel, M., Morris, S. and Ahiadeke, C. (2000) Urban
Livelihoods and Food and Nutrition Security in Greater Accra, Ghana. Washington: International Food
Policy Research Institute.

Mayers, J. (2005) ‘Stakeholder power analysis’. Available at: https://policy-

powertools.org/Tools/Understanding/docs/stakeholder_power_tool_english.pdf (Accessed:
25.11.2020).

McCaskie, T. C. (2000) Asante Identities: History and Modernity in an African Village 1850-1950. Lon-
don: Edinburgh University Press.

McConnachie, M. M. and Shackleton, C. M. (2010) ‘Public green space inequality in small towns
in South Africa’, Habitat international, 34(2), pp. 244-248.

McDowell, L. (1988) ‘Coming in from the dark: Feminist research in geography’, in Eyles, J. (ed.)
Research in Human Geography. Oxford: Blackwell, pp. 154-173.

McGrath, S. K. and Whitty, S. J. (2017) ‘Stakeholder defined’, International Journal of Managing Pro-
jects in Business, 10(4), pp. 721-748.

288
McLain, R. J., Hurley, P. T., Emery, M. R. and Poe, M. R. (2014) ‘Gathering “wild” food in the
city: rethinking the role of foraging in urban ecosystem planning and management’, Local Envi-
ronment, 19(2), pp. 220-240.

McLain, R., Poe, M., Hurley, P. T., Lecompte-Mastenbrook, J. and Emery, M. R. (2012) ‘Produc-
ing edible landscapes in Seattle’s urban forest’, Urban Forestry & Urban Greening, 11(2), pp. 187-
194.

McPhearson, P. T., Feller, M., Felson, A., Karty, R., Lu, J. W., Palmer, M. I. and Wenskus, T.
(2010) ‘Assessing the effects of the urban forest restoration effort of MillionTreesNYC on the
structure and functioning of New York City ecosystems’, Cities and the Environment (CATE), 3(1),
Article 7.

McPherson, E. G. and Luttinger, N. (1998) ‘From nature to nurture: The history of Sacramento’s
urban forest’, Journal of arboriculture, 24(2), pp. 72-88.

Mehrotra, S., Bardhan, R. and Ramamritham, K. (2018) ‘Urban informal housing and surface ur-
ban heat island intensity: exploring spatial association in the City of Mumbai’, Environment and Ur-
banization ASIA, 9(2), pp. 158-177.

Meikle, S., Ramasut, T. and Walker, J. (2001) Sustainable Urban Livelihoods: Concepts and implications
for policy. London: University College London, The Development Planning Unit.

Mensah, C. A. (2014a) ‘Is Kumasi still a garden city? Land use analysis between 1980-2010’, Jour-
nal of Environment and Ecology, 5(2), pp. 89-107.

Mensah, C. A. (2014b) ‘Destruction of urban green spaces: A problem beyond urbanization in

Kumasi city (Ghana)’, American Journal of Environmental Protection, 3(1), pp. 1-9.

Mensah, C. A., Antwi, K. B., Eshun, J. K. and Baidoo, P. (2017) ‘Towards sustainability: Over-
coming the physical barriers to urban green spaces in Kumasi, Ghana’, Ghana Journal of Geography,
9(2), pp. 125-150.

Mensah, C. A., Gough, K. V. and Simon, D. (2018) ‘Urban green spaces in growing oil cities: the
case of Sekondi-Takoradi Metropolis, Ghana’, International Development Planning Review, 40(4), pp.
371-395.

Merse, C. L., Buckley, G. L. and Boone, C. G. (2009) ‘Street trees and urban renewal: A Balti-
more case study’, The Geographical Bulletin, 50(2), pp. 65-81.

Mertens, D. M. (2003) ‘Mixed Methods and the Politics of Human Research: The Transforma-
tive-Emancipatory Perspective’, in Tashakkori, A. and Teddlie, C. (eds.) Handbook of Mixed Meth-
ods in Social and Behavioral Research. London: SAGE, pp. 135-166.

Mertens, D. M. (2007) ‘Transformative paradigm: Mixed methods and social justice’, Journal of
mixed methods research, 1(3), pp. 212-225.

Mertens, D. M. (2008) Transformative research and evaluation. London: Guilford Press.

Mertens, D. M. (2010) ‘Transformative mixed methods research’, Qualitative inquiry, 16(6), pp.
469-474.

289
Mertens, D. M. (2012) ‘Transformative mixed methods: Addressing inequities’, American Behavior-
al Scientist, 56(6), pp. 802-813.

Miles, M. B., Huberman, A. M. and Saldaña, J. (2014) Qualitative Data Analysis: A Methods Source-
book. London: SAGE.

Miles, S. (2017) ‘Stakeholder theory classification: A theoretical and empirical evaluation of defi-
nitions’, Journal of Business Ethics, 142(3), pp. 437-459.

Millennium Ecosystem Assessment (2003) Ecosystems and Human Well-being: A framework for assess-
ment. Washington: Island Press.

Miller, K. E. (2015) ‘Dear critics: Addressing concerns and justifying the benefits of photography
as a research method’, Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 16(3), Arti-
cle 27.

Miller, R. W., Hauer, R. J. and Werner, L. P. (2015) Urban Forestry: Planning and Managing Urban
Greenspaces. Long Grove: Waveland Press.

Millward, A. A. and Sabir, S. (2010) ‘Structure of a forested urban park: Implications for strategic
management’, Journal of Environmental Management, 91, pp. 2215-2224.

Mincey, S. K., Hutten, M., Fischer, B. C., Evans, T. P., Stewart, S. I. and Vogt, J. M. (2013)
‘Structuring institutional analysis for urban ecosystems: A key to sustainable urban forest man-
agement’, Urban Ecosystems, 16(3), pp. 553-571.

Ministry of Environment, Science and Technology – Town and Country Planning Department
(2001) ‘The New Spatial Planning Model Guidelines’. Available at:
http://www.luspa.gov.gh/files/GUIDELINES-FOR-THE-NEW-SPATIAL-PLANING-
MODEL.pdf (Accessed: 12.02.2021).

Mitchell, R. K., Agle, B. R. and Wood, D. J. (1997) ‘Toward a Theory of Stakeholder Identifica-
tion and Salience: Defining the Principle of Who and What Really Counts’, The Academy of Man-
agement Review, 22(4), pp. 853-886.

Moberg, F. and Folke, C. (1999) ‘Ecological goods and services of coral reef ecosystems’, Ecologi-
cal economics, 29(2), pp. 215-233.

Moore, J. C., Stinson, L. L. and Welniak, E. J. (2000) ‘Income Measurement Error in Surveys: A
Review’, Journal of Official Statistics, 16(4), pp. 331-361.

Moote, M. A., McClaran, M. P. and Chickering, D. K. (1997) ‘Theory in practice: Applying partic-
ipatory democracy theory to public land planning’, Environmental management, 21(6), pp. 877-889.

Morgan, D. L. (2007) ‘Paradigms Lost and Pragmatism Regained: Methodological Implications of

Combining Qualitative and Quantitative Methods’, Journal of Mixed Methods Research, 1(1), pp. 48-
76.

Morgenroth, J. and Armstrong, T. (2012) ‘The impact of significant earthquakes on Christchurch,

New Zealand’s urban forest’, Urban Forestry & Urban Greening, 11(4), pp. 383-389.

290
Morgenroth, J., Nowak, D. J. and Koeser, A. K. (2020) ‘DBH Distributions in America’s Urban
Forests – An Overview of Structural Diversity’, Forests, 11(2), 135.

Morris, D. (2005) ‘A new tool for strategy analysis: the opportunity model’, Journal of Business
Strategy, 26(3), pp. 50-56.

Morse, J. M. (1991) ‘Approaches to qualitative-quantitative methodological triangulation’, Nursing

Research, 40(2), pp. 120-123.

Moser, C. O. (1989) ‘Community participation in urban projects in the Third World’, Progress in
Planning, 32(2), pp. 71-133.

Moser, C. O. (1998) ‘The asset vulnerability framework: reassessing urban poverty reduction
strategies’, World development, 26(1), pp. 1-19.

Moser, S. (2008) ‘Personality: a new positionality?’, Area, 40(3), pp. 383-392.

Mosse, D. (2001) ‘‘People’s Knowledge’, Participation and Patronage: Operations and Represen-
tations in Rural Development’, in Cooke, B. and Kothari, U. (eds.) Participation: The new Tyranny?.
London: SAGE, pp. 16-35.

Mullings, B. (1999) ‘Insider or outsider, both or neither: some dilemmas of interviewing in a

cross-cultural setting’, Geoforum, 30(4), pp. 337-350.

Nathan, N. L. (2019) Electoral Politics and Africa’s Urban Transition: Class and Ethnicity in Ghana.
Cambridge: Cambridge University Press.

National Development Planning Commission (2017) Medium-term National Development Policy

Framework: An Agenda for Jobs: Creating Prosperity and Equal Opportunity for All (First Step) 2018-2021.
Available at: https://s3-us-west-2.amazonaws.com/new-ndpc-
static1/CACHES/PUBLICATIONS/2018/08/23/Medium-term+Policy+Framework-
Final+June+2018.pdf (Accessed: 17.09.2020).

National Research Council (2002) Neem: A Tree for Solving Global Problems. New York: Books for
Business.

Navarro, V. (2002) ‘A Critique of Social Capital’, International Journal of Health Services, 32(3), pp.
423-432.

Nelson, N. and Wright, S. (1995) Power and Participatory Development: Theory and Practice. London: IT
Publications.

Nero, B. F. (2016) ‘Urban green space dynamics and distributional equity in Kumasi, Ghana’, Liv-
ing Planet Symposium 2016. Prague, Czech Republic, 9-13 May 2016. Available at:
https://www.researchgate.net/profile/Bertrand_Nero2 (Accessed: 15.09.2020).

Nero, B. F. (2017) ‘Urban green space dynamics and socio-environmental inequity: multi-
resolution and spatiotemporal data analysis of Kumasi, Ghana’, International Journal of Remote Sens-
ing, 38(23), pp. 6993-7020.

291
Nero, B. F., Callo-Concha, D., Anning, A. and Denich, M. (2017) ‘Urban Green Spaces Enhance
Climate Change Mitigation in Cities of the Global South: The Case of Kumasi, Ghana’, Procedia
Engineering, 198, pp. 69-83.

Nero, B. F., Callo-Concha, D. and Denich, M. (2018a) ‘Structure, Diversity, and Carbon Stocks
of the Tree Community of Kumasi, Ghana’, Forests, 9, 519.

Nero, B. F., Kwapong, N. A., Jatta, R. and Fatunbi, O. (2018b) ‘Tree Species Diversity and Soci-
oeconomic Perspectives of the Urban (Food) Forest of Accra, Ghana’, Sustainability, Vol. 10,
3417.

Njeru, J. (2010) ‘‘Defying’ democratization and environmental protection in Kenya: The case of
Karura Forest reserve in Nairobi’, Political Geography, 29(6), pp. 333-342.

Njeru, J. (2013) ‘‘Donor-driven’ neoliberal reform processes and urban environmental change in
Kenya: The case of Karura Forest in Nairobi’, Progress in Development Studies, 13(1), pp. 63-78.

Norberg, J. (1999) ‘Linking Nature’s services to ecosystems: some general ecological concepts’,
Ecological economics, 29(2), pp. 183-202.

Nowak, D. J. (1986) Silvics of an urban tree species: Norway maple (Acer platanoides L.). Ph.D. Thesis.
State University of New York.

Nowak, D. J. (1993) ‘Historical vegetation change in Oakland and its implications for urban for-
est management’, Journal of Arboriculture, 19(5), pp. 313-319.

Nowak, D. J. (2008) ‘Assessing Urban Forest Structure: Summary and Conclusions’, Arboriculture
& Urban Forestry, 34(6), pp. 391-392.

Nowak, D. J. (2012) ‘Contrasting natural regeneration and tree planting in fourteen North Amer-
ican cities’, Urban Forestry & Urban Greening, 11(4), pp. 374-382.

Nowak, D. J. (2016) ‘Urban Forests’, in Robertson, G. and Mason, A. (eds.) Assessing the Sustaina-
bility of Agricultural and Urban Forests in the United States, pp. 37-52. Available at:
https://www.fs.fed.us/research/publications/FS-1067SustainabilityAgUrb.pdf (Accessed:
15.09.2020).

Nowak, D. J. and Crane, D. E. (2002) ‘Carbon storage and sequestration by urban trees in the
USA’, Environmental pollution, 116(3), pp. 381-389.

Nowak, D. J, Crane, D. E. and Stevens, J. (2006) ‘Air pollution removal by urban trees and
shrubs in the United States’, Urban Forestry & Urban Greening, 4(3-4), pp. 115-123.

Nowak, D. J., Crane, D. E., Stevens, J. C., Hoehn, R. E., Walton, J. T. and Bond, J. (2008) ‘A
Ground-Based Method of Assessing Urban Forest Structure and Ecosystem Services’, Arboricul-
ture & Urban Forestry, 34(6), pp. 347-358.

Nti, K. (2012) ‘This Is Our Land: Land, Policy, Resistance, and Everyday Life in Colonial South-
ern Ghana, 1894-7’, Journal of Asian and African Studies, 48(1), pp. 3-15.

292
Obeng‐Odoom, F. (2012) ‘Neoliberalism and the urban economy in Ghana: Urban employment,
inequality, and poverty’, Growth and Change, 43(1), pp. 85-109.

Oberhauser, A. M. and Hanson, K. T. (2007) ‘Negotiating livelihoods and scale in the context of
neoliberal globalization: perspectives from Accra, Ghana’, African Geographical Review, 26(1), pp.
11-36.

Odotei, I. (1991) ‘External influences of Ga society and culture’, Research Review, Vol. 7(1-2), pp.
61-71.

Oduro-Ofori, E., Imoro, B. and Agyeman, K. O. (2014) ‘Promoting Green Infrastructure in

Kumasi: Challenges and Strategies’, Research on Humanities and Social Sciences, 4(26), pp. 110-118.

OECD (2013) OECD Guidelines for Micro Statistics on Household Wealth. Available at:
https://www.oecd.org/statistics/guidelines-for-micro-statistics-on-household-wealth-
9789264194878-en.htm (Accessed: (18.09.2020).

Okyerefo, M. P. K. (2014) ‘“I experience serenity and convenience in the forest”-Achimota For-
est turned sacred space to confront the vicissitudes of life’, Journal of the Sociology and Theory of Reli-
gion, 3, pp. 1-30.

Okyerefo, M. P. K. and Fiaveh, D. Y. (2017) ‘Prayer and health-seeking beliefs in Ghana: under-
standing the ‘religious space’ of the urban forest’, Health Sociology Review, 26(3), pp. 308-320.

Oliva, S. R., Valdés, B. and Mingorance, M. D. (2008) ‘Evaluation of some pollutant levels in bit-
ter orange trees: Implications for human health’, Food and chemical toxicology, 46(1), pp. 65-72.

Omosini, O. (1972) ‘The Gold Coast land question, 1894-1900: Some issues raised on West Afri-
ca’s economic development’, The International Journal of African Historical Studies, 3, pp. 453-469.

Onwuegbuzie, A. J. and Leech, N. L. (2005) ‘Taking the “Q” Out of Research: Teaching Re-
search Methodology Courses Without the Divide Between Quantitative and Qualitative Para-
digms’, Quality and Quantity, 39, pp. 267-295.

Ordóñez, C. and Duinker, P. N. (2012) ‘Ecological integrity in urban forests’, Urban Ecosystems,
15(4), pp. 863-877.

Ordóñez, C. and Duinker, P. N. (2013) ‘An analysis of urban forest management plans in Cana-
da: Implications for urban forest management’, Landscape and Urban Planning, 116, pp. 36-47.

Ordóñez, C. and Duinker, P. N. (2014) ‘Urban forest values of the citizenry in three Colombian
cities’, Society & Natural Resources, 27(8), pp. 834-849.

Ordóñez, C., Threlfall, C. G., Kendal, D., Hochuli, D. F., Davern, M., Fuller, R. A., van der Ree,
R. and Livesley, S. J. (2019) ‘Urban forest governance and decision-making: A systematic review
and synthesis of the perspectives of municipal managers’, Landscape and urban planning, 189, pp.
166-180.

Ordóñez-Barona, C., Bush, J., Hurley, J., Amati, M., Juhola, S., Frank, S., Ritchie, M., Clark, C.,
English, A., Hertzog, K., Caffin, M., Watt, S. and Livesley, S. (2021) ‘International approaches to

293
protecting and retaining trees on private urban land’, Journal of Environmental Management, 285,
112081.

Osei-Tutu, J. K. and von Hesse, H. W. (2019) ‘Cosmopolitan Conundrums: Impacts of Trade

Fortresses on the Gã Space, 1450-1870’, in Osei-Tutu, J. K. (ed.) Forts, Castles and Society in West
Africa: Gold Coast and Dahomey, 1450-1960. Leiden: Brill.

Ostoić, S. K. and Konijnendijk, C. C. (2015) ‘Exploring global scientific discourses on urban for-
estry’, Urban Forestry & Urban Greening, 14(1), pp. 129-138.

Ostrom, E. (1999) ‘Social Capital: A Fad or a Fundamental Concept?’, in Dasgupta, P. and

Serageldin, I. (eds.) Social Capital: A multifaceted perspective. Washington: The International Bank for
Reconstruction and Development/The World Bank, pp. 172-214.

Ottitsch, A. and Krott, M. (2005) ‘Urban Forest Policy and Planning’, in Konijnendijk, C., Nils-
son, K., Randrup, T. and Schipperijn, J. (eds.) Urban Forests and Trees. Berlin: Springer, pp. 117-
148.

Owusu, A. B. (2018) ‘An assessment of urban vegetation abundance in Accra Metropolitan Area,
Ghana: a geospatial approach’, Journal of Environmental Geography, 11(1-2), pp. 37-44.

Owusu, F. (2006) ‘On public organizations in Ghana: what differentiates good performers from
poor performers?’, African development review, 18(3), pp. 471-485.

Owusu, G. and Agyei-Mensah, S. (2011) ‘A comparative study of ethnic residential segregation in

Ghana’s two largest cities, Accra and Kumasi’, Population and Environment, 32(4), pp. 332-352.

Paller, J. W. (2019) Democracy in Ghana: Everyday Politics in Urban Africa. Cambridge: Cambridge
University Press.

Pandit, R., Polyakov, M., Tapsuwan, S. and Moran, T. (2013) ‘The effect of street trees on prop-
erty value in Perth, Western Australia’, Landscape and Urban Planning, 110, pp. 134-142.

Parker, J. (2000) Making the Town: Ga State and Society in Early Colonial Accra. Oxford: James Curry.

Patton, M. Q. (1990) Qualitative evaluation and research methods. London: SAGE.

Paul, S. (1987) Community Participation in Development Projects: The World Bank Experience. Washing-
ton: The International Bank for Reconstruction and Development/The World Bank. Available
at: http://documents1.worldbank.org/curated/en/850911468766244486/pdf/multi-page.pdf
(Accessed: 17.09.2020).

Pauleit, S., Ennos, R. and Golding, Y. (2005) ‘Modeling the environmental impacts of urban land
use and land cover change – a study in Merseyside, UK’, Landscape and urban planning, 71(2-4), pp.
295-310.

Paumgarten, F. (2005) ‘The role of non-timber forest products as safety-nets: a review of evi-
dence with a focus on South Africa’, GeoJournal, 64(3), pp. 189-197.

Peckham, S. C., Duinker, P. N. and Ordóñez, C. (2013) ‘Urban forest values in Canada: Views of
citizens in Calgary and Halifax’, Urban Forestry & Urban Greening, 12(2), pp. 154-162.

294
Pedlowski, M. A., Da Silva, V. A. C., Adell, J. J. C. and Heynen, N. C. (2002) ‘Urban forest and
environmental inequality in Campos dos Goytacazes, Rio de Janeiro, Brazil’, Urban Ecosystems,
6(1-2), pp. 9-20.

Peel, M. C., Finlayson, B. L., and McMahon, T. A. (2007) ‘Updated world map of the Köppen-
Geiger climate classification’, Hydrology and Earth System Sciences Discussions, 4(2), pp. 439-473.

Peet, R. and Watts, M. (1993) ‘Introduction: Development Theory and Environment in an Age of
Market Triumphalism’, Economic Geography, 69(3), pp. 227-253.

Pellow, D. (2001) ‘Cultural differences and urban spatial forms: elements of boundedness in an
Accra community’, American Anthropologist, 103(1), pp. 59-75.

Perrin, J. and Benassi, V. (2009) ‘The connectedness to nature scale: A measure of emotional
connection to nature?’, Journal of Environmental Psychology, Vol. 29, pp. 434-440.

Pickett, S., Kolasa, J. and Jones, C. G. (1994) Ecological Understanding: The Nature of Theory and the
Theory of Nature. London: Elsevier.

Pickton, D. W. and Wright, S. (1998) ‘What’s SWOT in strategic analysis?’, Strategic change, 7(2),
pp. 101-109.

Pincetl, S. (2010) ‘Implementing municipal tree planting: Los Angeles million-tree initiative’, Envi-
ronmental management, 45(2), pp. 227-238.

Poe, M. R., McLain, R. J., Emery, M. and Hurley, P. T. (2013) ‘Urban forest justice and the rights
to wild foods, medicines, and materials in the city’, Human Ecology, 41(3), pp. 409-422.

Portes, A. (2014) ‘Downsides of social capital’, Proceedings of the National Academy of Sciences,
111(52), pp. 18407-18408.

Potschin, M. B. and Haines-Young, R. H. (2011) ‘Ecosystem services: exploring a geographical

perspective’, Progress in Physical Geography, 35(5), pp. 575-594.

Potts, D. (2017) ‘Conflict and Collisions in Sub-Saharan African Urban Definitions: Interpreting
Recent Urbanization Data From Kenya’, World Development, 97, pp. 67-78.

Potts, D. (2018) ‘Urban data and definitions in sub-Saharan Africa: Mismatches between the pace
of urbanisation and employment and livelihood change’, Urban Studies, 55(5), pp. 965-986.

Project Management Institute (2013) A Guide to the Project Management Body of Knowledge, Fifth Edi-
tion. Newtown Square: Project Management Institute.

Pryor, M. R. (2015) ‘Street Tree Planting in Hong Kong in the Early Colonial Period (1842–98)’,
Journal of the Royal Asiatic Society Hong Kong Branch, 55, pp. 33-56.

Public Records and Archives Administration Department, ADM 5/1/27, Annual Report on the
Social and Economic Progress of the People of Gold Coast 1931-1932.

Public Records and Archives Administration Department, ADM 5/1/51, Government of the
Gold Coast, Departmental Reports 1895.

295
Public Records and Archives Administration Department, ADM 5/1/53, Government of the
Gold Coast, Departmental Reports 1897.

Public Records and Archives Administration Department, ADM 5/1/56, Government of the
Gold Coast, Departmental Reports 1901.

Public Records and Archives Administration Department, ADM 5/1/58, Government of the
Gold Coast, Departmental Reports 1903.

Public Records and Archives Administration Department, ADM 5/1/59 Government of the
Gold Coast, Departmental Reports 1904.

Public Records and Archives Administration Department, ADM 5/1/60, Government of the
Gold Coast, Departmental Reports 1905.

Public Records and Archives Administration Department, ADM 5/1/63, Government of the
Gold Coast, Departmental Reports 1908.

Public Records and Archives Administration Department, ADM 5/1/68, Government of the
Gold Coast, Departmental Reports 1912.

Public Records and Archives Administration Department, ADM 5/1/72, Government of the
Gold Coast, Departmental Reports 1915.

Public Records and Archives Administration Department, ADM 5/1/73, Government of the
Gold Coast, Departmental Reports 1916.

Public Records and Archives Administration Department, ADM 5/1/74, Government of the
Gold Coast, Departmental Reports 1917.

Public Records and Archives Administration Department, ADM 5/1/75, Government of the
Gold Coast, Departmental Reports 1918.

Public Records and Archives Administration Department, ADM 5/1/76, Government of the
Gold Coast, Departmental Reports 1919.

Public Records and Archives Administration Department, ADM 5/1/77, Government of the
Gold Coast, Departmental Reports 1920.

Public Records and Archives Administration Department, ADM 5/1/78, Government of the
Gold Coast, Departmental Reports 1921.

Public Records and Archives Administration Department, ADM 5/1/79, Government of the
Gold Coast, Departmental Reports 1922-23.

Public Records and Archives Administration Department, ADM 5/1/80, Government of the
Gold Coast, Departmental Reports 1923-1924.

Public Records and Archives Administration Department, ADM 5/1/84, Government of the
Gold Coast, Departmental Reports 1926-1927.

296
Public Records and Archives Administration Department, ADM 5/1/88, Government of the
Gold Coast, Departmental Reports 1928-29.

Public Records and Archives Administration Department, ADM 5/1/89, Government of the
Gold Coast Colony, Departmental Reports 1929-30.

Public Records and Archives Administration Department, ADM 5/1/94, Report on the Depart-
ment of Agriculture 1934-37.

Public Records and Archives Administration Department, ADM 5/1/101, Report on the De-
partment of Agriculture 1937-39.

Public Records and Archives Administration Department, CSO 20/1/3a, Minutes of General
Monthly Meeting from Accra Town Council on 08.07.1935.

Public Records and Archives Administration Department, CSO 20/1/3b, Minutes of General
Monthly Meeting from Accra Town Council on 08.07.1935.

Public Records and Archives Administration Department, CSO 20/1/7a, Minutes of General
Monthly Meeting from Accra Town Council on 14.11.1938.

Public Records and Archives Administration Department, CSO 20/1/7b, Minutes of General
Monthly Meeting from Accra Town Council on 11.09.1939.

Public Records and Archives Administration Department, CSO 20/1/66, Minutes of the Stand-
ing Committee to Co-ordinate and Control Arboricultural and Horticultural Work in Accra on
08.06.1938.

Public Records and Archives Administration Department, CSO 20/12/3, Appointment of Re-
gional Planning Officer in connection with Town and regional planning in the Colonies, 1930.

Public Records and Archives Administration Department, CSO 20/12/20, Town Planning
Schemes, 1944-47.

Public Records and Archives Administration Department, RG 8/2/779, Accra - A Plan For The
Town: The Report for the Minister of Housing, 1958.

Public Records and Archives Administration Department, RG 3/5/349a, Tree Planting and Cut-
ting in Accra - Circular No. 49/50.

Pütz, M., Schmid, S., Bernasconi, A. and Wolf, B. (2015) ‘Urban forestry: Definition, Trends und
Folgerungen für die Waldakteure in der Schweiz’, Schweizerische Zeitschrift fur Forstwesen, 166(4), pp.
230-237.

Quarcoopome, S. S. (1992) ‘Urbanisation, land alienation and politics in Accra’, Research Review,
8(1-2), pp. 40-54.

Quarcoopome, S. S. (1998) ‘Social impact of urbanisation: The case of Ga Mashie of Accra’,

Transactions of the Historical Society of Ghana, New Series, 2, pp. 133-146.

297
Quayson, A. (2014) Oxford Street, Accra: city life and the itineraries of transnationalism. Durham: Duke
University Press.

Quibria, M. G. (2003) ‘The Puzzle of Social Capital: A Critical Review’, Asian Development Review,
20(2), pp. 19-39.

R Core Team. (2015) R: A Language and Environment for Statistical Computing. Vienna: R Foundation
for Statistical Computing. Available at: http://www.R-project.org/.

Rachmadyanto, H., Dijk, M. P. V. and Oduro-Kwarteng, S. (2016) ‘Mapping pro-poor water

supply services: what are the technical and institutional options for the utility and private provi-
sion in Accra?’, International Journal of Water, 10(2-3), pp. 246-266.

Rain, D., Engstrom, R., Ludlow, C. and Antos, S. (2011) Accra Ghana: A City Vulnerable to Flooding
and Drought-Induced Migration. Available at: http://www.unhabitat.org/grhs/2011 (Accessed:
14.09.2020).

Randrup, T. B., Östberg, J. and Wiström, B. (2017) ‘Swedish green space management – The
managers perspective’, Urban Forestry & Urban Greening, 28, pp. 103-109.

Raoufou, R., Kouami, K. and Koffi, A. (2011) ‘Woody plant species used in urban forestry in
West Africa: Case study in Lom, capital town of Togo’, Journal of Horticulture and Forestry, 3(1), pp.
21-31.

Rayamajhi, S., Smith-Hall, C. and Helles, F. (2012) ‘Empirical evidence of the economic im-
portance of Central Himalayan forests to rural households’, Forest Policy and Economics, 20, pp. 25-
35.

Reindorf, C. C. (1895) History of the Gold Coast and Asante, based on traditions and historical facts, com-
prising a period of more than three centuries from 1500 to 1860. Basel: Missionsbuchhandlung.

Richards, N. A. (1983) ‘Diversity and stability in a street tree population’, Urban Ecology, 7, pp.
159-171.

Ritchie, J. and Spencer, L. (2002) ‘Qualitative data analysis for applied policy research’, in Bry-
man, A. and Burgess, R. G. (eds.) Analyzing qualitative data. London: Routledge, pp. 173-194.

Robbins, P. (2012) Political Ecology: A Critical Introduction. Chichester: John Wiley & Sons.

Robison, L. T., Schmid, A. A. and Siles, M. E: (2002) ‘Is Social Capital Really Capital?’, Review of
Social Economy, 60(1), pp. 1-21.

Rocheleau, D. and Edmunds, D. (1997) ‘Women, Men and Trees: Gender, Power and Property
in Forest and Agrarian Landscapes’, World Development, 25(8), pp. 1351-1371.

Roloff, A., Korn, S. and Gillner, S. (2009) ‘The Climate-Species-Matrix to select tree species for
urban habitats considering climate change’, Urban Forestry & Urban Greening, 8(4), pp. 295-308.

Roman, L. A., Battles, J. J. and McBride, J. R. (2016) Urban Tree Mortality: a Primer on Demographic
Approaches. Newtown Square: U.S. Forest Service.

298
Roman, L. A. and Scatena, F. N. (2011) ‘Street tree survival rates: Meta-analysis of previous stud-
ies and application to a field survey in Philadelphia, PA, USA’, Urban Forestry & Urban Greening,
10(4), pp. 269-274.

Roman, L. A, Pearsall, H., Eisenman, T., Conway, T., Fahey, R., Landry, S., Vogt, J., van Doorn,
N., Grove, J., Locke, D., Bardekjian, A., Battles, J., Cadenasso, M., Konijnendijk van den Bosch,
C., Avolia, M., Berland, A., Jenerette, G., Mincey, S., Pataki, D. and Staudhammer, C. (2018)
‘Human and biophysical legacies shape contemporary urban forests: a literature synthesis’, Urban
Forestry & Urban Greening, 31, pp. 157-168.

Rosen, M. (2015) ‘A Brief Historical Perspective of Urban Forests in Canada’, Histoires forestières
du Québec, 7(1), pp. 27-32.

Roy, S., Byrne, J. and Pickering, C. (2012) ‘A systematic quantitative review of urban tree bene-
fits, costs, and assessment methods across cities in different climatic zones’, Urban Forestry & Ur-
ban Greening, 11(4), pp. 351-363.

Russo, A., Escobedo, F. J., Cirella, G. T. and Zerbe, S. (2017) ‘Edible green infrastructure: An
approach and review of provisioning ecosystem services and disservices in urban environments’,
Agriculture, Ecosystems & Environment, 242, pp. 53-66.

Rydin, Y. and Pennington, M. (2000) ‘Public participation and local environmental planning: the
collective action problem and the potential of social capital’, Local environment, 5(2), pp. 153-169.

Sackeyfio, N. (2012) ‘The Politics of Land and Urban Space in Colonial Accra’, History in Africa,
39, pp. 293-329.

Saldaña, J. (2009) The Coding Manual for Qualitative Researchers. London: SAGE.

Sandberg, L. A., Bardekjian, A. and Butt, S. (eds.)(2015) Urban Forests, Trees, and Greenspace: A Po-
litical Ecology Perspective. London: Routledge.

Sandelowski, M. and Leeman, J. (2012) ‘Writing usable qualitative health research findings’, Quali-
tative health research, 22(10), pp. 1404-1413.

Sander, H., Polasky, S. and Haight, R. G. (2010) ‘The value of urban tree cover: A hedonic prop-
erty price model in Ramsey and Dakota Counties, Minnesota, USA’, Ecological Economics, 69(8),
pp. 1646-1656.

Satterthwaite, D. (2010) Urban Myths and the Mis-use of Data that Underpin them. United Nations
University, World Institute for Development Economics Research. Working Paper No. 2010/28.
Available at: https://www.wider.unu.edu/sites/default/files/wp2010-28.pdf (Accessed:
18.09.2020).

Satterthwaite, D. and Tacoli, C. (2002) ‘Seeking an Understanding of Poverty that Recognizes

Rural–Urban Differences and Rural – Urban Linkages’, in Rakodi, C. and Llyod-Jones, T. (eds.)
Urban Livelihoods: A People-centred Approach to Reducing Poverty. London: Earthscan, pp. 52-70.

Schreiner, M. (2015) Simple Poverty Scorecard® : Poverty-Assessment Tool Ghana. Available at:
http://www.simplepovertyscorecard.com/GHA_2012_ENG.pdf (Accessed: 16.09.2020).

299
Schroeder, H., Flannigan, J. and Coles, R. (2006) ‘Residents’ Attitudes Toward Street Trees in the
UK and U.S. Communities’, Arboriculture & Urban Forestry, 2(5), pp. 236-246.

Schröter, M., van der Zanden, E. H., van Oudenhoven, A. P., Remme, R. P., Serna‐Chavez, H.
M., De Groot, R. S. and Opdam, P. (2014) ‘Ecosystem services as a contested concept: a synthe-
sis of critique and counter‐arguments’, Conservation Letters, 7(6), pp. 514-523.

Schumacher, C. F. (1827) Beskrivelse af Guineiske planter: som ere fundne af Danske botanikere, især af
etatsraad Thonning. Kjöbenhavn: F. Popp.

Schwartz, J. M. and Cook, T. (2002) ‘Archives, records, and power: The making of modern
memory’, Archival science, 2(1-2), pp. 1-19.

Schwarz, K., Fragkias, M., Boone, C. G., Zhou, W., McHale, M., Grove, J. M., O’Neil-Dunne, J.,
McFadden, J. P., Buckley, G. L., Childers, D., Ogden, L., Pincetl, S., Pataki, D., Whitmer, A., and
Cadenasso, M. L. (2015) ‘Trees grow on money: urban tree canopy cover and environmental jus-
tice’, PloS one, 10(4), e0122051.

Scoones, I. (1998) Sustainable Rural Livelihoods: A Framework for Analysis, IDS Working Paper 72,
Brighton: IDS.

Seburanga, J. L., Kaplin, B. A., Zhang, Q. X. and Gatesire, T. (2014) ‘Amenity trees and green
space structure in urban settlements of Kigali, Rwanda’, Urban Forestry & Urban Greening, 13(1),
pp. 84-93.

Serageldin, I. and Grootaert, C. (1999) ‘Defining Social Capital: An Integrating View’, in Dasgup-
ta, P. and Serageldin, I. (eds.) Social Capital: A multifaceted perspective. Washington: The International
Bank for Reconstruction and Development/The World Bank, pp. 40-58.

Shackleton, C. M. (2012) ‘Is there no urban forestry in the developing world?’, Scientific Research
and Essays, 7(40), pp. 3329-3335.

Shackleton, C. M. and Blair, A. (2013) ‘Perceptions and use of public green space is influenced by
its relative abundance in two small towns in South Africa’, Landscape and urban Planning, 113, pp.
104-112.

Shackleton C.M., Cilliers S.S., du Toit M.J., Davoren E. (2021) ‘The Need for an Urban Ecology
of the Global South’, in Shackleton C.M., Cilliers S.S., Davoren E., du Toit M.J. (eds) Urban Ecol-
ogy in the Global South.Cham: Springer, pp. 1-26.

Shackleton, C. M., Hebinck, P., Kaoma, H., Chishaleshale, M., Chinyimba, A., Shackleton, S. E.,
Gambiza, J. and Gumbo, D. (2014) ‘Low-cost housing developments in South Africa miss the
opportunities for household level urban greening’, Land use policy, 36, pp. 500-509.

Shackleton, C. M., Shackleton, S. E., Buiten, E. and Bird, N. (2007) ‘The importance of dry
woodlands and forests in rural livelihoods and poverty alleviation in South Africa’, Forest policy and
economics, 9(5), pp. 558-577.

Shackleton, S., Chinyimba, A., Hebinck, P., Shackleton, C. and Kaoma, H. (2015) ‘Multiple bene-
fits and values of trees in urban landscapes in two towns in northern South Africa’, Landscape and
Urban Planning, 136, pp. 76-86.

300
Shakeel, T. and Conway, T. M. (2014) ‘Individual households and their trees: Fine-scale charac-
teristics shaping urban forests’, Urban Forestry & Urban Greening, 13(1), pp. 136-144.

Simmering, D., Addai, S., Geller, G. and Otte, A. (2013) ‘A university campus in peri-urban Ac-
cra (Ghana) as a haven for dry-forest species’, Flora et Vegetatio Sudano-Sambesica, 16, pp. 10-21.

Small, L. (2007) ‘The Sustainable Rural Livelihoods Approach: A Critical Review’, Canadian Jour-
nal of Development Studies/Revue canadienne d’études du développement, 28(1), pp. 27-38.

Smith, L. W. (2000) ‘Stakeholder Analysis: A Pivotal Practice of Successful Projects’, Proceedings

of the Project Management Institute Annual Seminars & Symposium, Houston, USA, 7-16 Sep-
tember 2000.

Smith, S. S. and Kulynych, J. (2002) ‘It may be social, but why is it capital? The social construc-
tion of social capital and the politics of language’, Politics & society, 30(1), pp. 149-186.

Sobel, J. (2002) ‘Can we trust social capital?’, Journal of economic literature, 40(1), pp. 139-154.

Solow, R. M. (1999) ‘Notes on Social Capital and Economic Performance’, in Dasgupta, P. and
Serageldin, I. (eds.) Social Capital: A multifaceted perspective. Washington: The International Bank for
Reconstruction and Development/The World Bank, pp. 6-12.

Song, X. P., Tan, P. Y., Edwards, P. and Richards, D. (2018) ‘The economic benefits and costs of
trees in urban forest stewardship: A systematic review’, Urban Forestry & Urban Greening, 29, pp.
162-170.

Spradley, J. P. (2016) Participant Observation. Long Grove: Waveland Press.

Stake, R. E. (1995) The art of case study research. London: SAGE.

Stake, R. E. (2005) ‘Qualitative Case Studies’, in Denzin, N. K. and Lincoln, Y. S. (eds.) The
SAGE Handbook of Qualitative Research. London: SAGE, pp. 443-466.

State of Queensland (2016) Invasive plant risk assessment: Neem tree, Azadirachta indica. Available at:
https://www.daf.qld.gov.au/__data/assets/pdf_file/0006/63168/IPA-Neem-Tree-Risk-
Assessment.pdf (Accessed: 16.09.2020).

Statistics Canada (2003) Survey Methods and Practices. Available at:

https://www150.statcan.gc.ca/n1/pub/12-587-x/12-587-x2003001-eng.pdf (Accessed:
16.09.2020).

Steenberg, J. W., Millward, A. A., Nowak, D. J., Robinson, P. J. and Ellis, A. (2017) ‘Forecasting
urban forest ecosystem structure, function, and vulnerability’, Environmental Management, 59(3), pp.
373-392.

Stevano, S., Johnston, D. and Codjoe, E. (2020) ‘The urban food question in the context of ine-
quality and dietary change: A study of schoolchildren in Accra’, The Journal of Development Studies,
56(6), pp. 1177-1189.

Stobbart, M. and Johnston, M. (2012) ‘A survey of urban tree management in New Zealand’, Ar-
boriculture and Urban Forestry, 38(6), pp. 247-254.

301
Stohl, A. (2008) ‘The travel-related carbon dioxide emissions of atmospheric researchers’, Atmos-
pheric Chemistry and Physics Discussions, 8(2), pp.7373-7389.

Stow, D. A., Weeks, J. R., Toure, S., Coulter, L. L., Lippitt, C. D. and Ashcroft, E. (2013) ‘Urban
vegetation cover and vegetation change in Accra, Ghana: connection to housing quality’, The Pro-
fessional Geographer, 65(3), pp. 451-465.

Sullivan, W. C., Kuo, F. E. and Depooter, S. F. (2004) ‘The fruit of urban nature: Vital neighbor-
hood spaces’, Environment and behavior, 36(5), pp. 678-700.

Summit, J. and McPherson, E. G. (1998) ‘Residential tree planting and care: a study of attitudes
and behavior in Sacramento, California’, Journal of Arboriculture, 24, pp. 89-97.

Surmiak, A. D. (2018) ‘Confidentiality in qualitative research involving vulnerable participants:

Researchers’ perspectives’, Forum Qualitative Sozialforschung/Forum: Qualitative Social Research, 19(3),
Article 12.

Synk, C. M., Kim, B. F., Davis, C. A., Harding, J., Rogers, V., Hurley, P. T., Emery, M. R.,
Nachman, K. E. (2017) ‘Gathering Baltimore’s bounty: Characterizing behaviors, motivations,
and barriers of foragers in an urban ecosystem’, Urban Forestry & Urban Greening, 28, pp. 97-102.

Szabó, P. (2010) ‘Why history matters in ecology: an interdisciplinary perspective’, Environmental

Conservation, 37(4), pp. 380-387.

Szabó, P. (2015) ‘Historical ecology: past, present and future’, Biological Reviews, 90(4), pp. 997-
1014.

Tacconi, L. (2011) ‘Developing environmental governance research: the example of forest cover
change studies’, Environmental Conservation, 38(2), pp. 234-246.

Talbot, J. F. and Kaplan, R. (1984) ‘Needs and fears: The response to trees and nature in the in-
ner city’, Journal of Arboriculture, 10(8), pp. 222-228.

Talen, E. (1997) ‘The social equity of urban service distribution: An exploration of park access in
Pueblo, Colorado, and Macon, Georgia’, Urban geography, 18(6), pp. 521-541.

Taormina, R. J., Kuok, A. C. and Wei, W. (2012) ‘Social capital as dehumanizing terminology’,
Advances in Applied Sociology, 2(2), pp. 143-148.

Tashakkori, A. and Creswell, J. W. (2007) ‘Editorial: The New Era of Mixed Methods’, Journal of
Mixed Methods Research, 1(1), pp. 3-7.

Tashakkori, A. and Teddlie, C. (1998) Mixed Methodology: Combining qualitative and quantitative ap-
proaches. London: SAGE.

Tashakkori, A. and Teddlie, C. (2003) Handbook of Mixed Methods in Social & Behavioral Research.
London: SAGE.

Temple, B. and Edwards, R. (2002) ‘Interpreters/translators and cross-language research: Reflex-

ivity and border crossings’, International journal of qualitative methods, 1(2), pp. 1-12.

302
ten Have, P. (1990) ‘Methodological issues in conversation analysis’, Bulletin of Sociological Methodol-
ogy/Bulletin de Méthodologie Sociologique, 27(1), pp. 23-51.

Teye, J. K. (2012) ‘Benefits, challenges, and dynamism of positionalities associated with mixed
methods research in developing countries: Evidence from Ghana’, Journal of Mixed Methods Re-
search, 6(4), pp. 379-391.

The Chronicle (2013a) ‘Mr. President, Stop The Destruction Of Achimota Forest!’, by Nduom,
P. K., 11.11.2013.

The Chronicle (2013b) ‘Minister Outdoors Committee To Plan Eco-Park’, by Adam, F.,
24.05.2013.

The Chronicle (2014a) ‘Chief Seizes Achimota Forest…School Under Threat, As Lands Com-
mission Fails To Defend Suit’, by Akli, E., 06.06.2014.

The Chronicle (2014b) ‘Battle Of Achimota…Legal Bullets Ricochet Across Terrain’, by Akli, E.,
09.06.2014.

The Chronicle (2014c) ‘Forestry Commission, Police And Achimota Forest’, no name,
26.11.2014.

The Chronicle (2014d) ‘MPs, Police In Hot Exchanges Over FC Lands, by Attenkah, R. and Ad-
am, F., 26.11.2014.

The Chronicle (2014e) ‘Re: Chief Seizes Achimota Forest…As Forestry Commission Fails To
Defend Suit’, Adjei, N. O., 26.06.2014.

Thompson Jr, P. E. (2018) ‘Mutual Economic Incentives for Urban Tree Canopy Restoration’,
Undergraduate Economic Review, 14(1), Article 9.

Tijdens, K. and Steinmetz, S. (2016) ‘Is the web a promising tool for data collection in developing
countries? An analysis of the sample bias of 10 web and face-to-face surveys from Africa, Asia,
and South America’, International Journal of Social Research Methodology, 19(4), pp. 461-479.

Tiwary, A., Vilhar, U., Zhiyanski, M., Stojanovski, V. and Dinca, L. (2020) ‘Management of na-
ture-based goods and services provisioning from the urban common: a pan-European perspec-
tive’, Urban Ecosystems, 23, pp. 645-657.

Troy, A., Grove, J. M. and O’Neil-Dunne, J. (2012) ‘The relationship between tree canopy and
crime rates across an urban – rural gradient in the greater Baltimore region’, Landscape and urban
planning, 106(3), pp. 262-270.

Tuffour-Mills, D., Antwi-Agyei, P. and Addo-Fordjour, P. (2020) ‘Trends and drivers of land
cover changes in a tropical urban forest in Ghana’, Trees, Forests and People, 2, 100040.

Tuholske, C., Andam, K. S., Blekking, J., Evans, T. and Caylor, K. (2018) Measures and determinants
of urban food security: evidence from Accra, Ghana, Working Paper No. 50, Washington: International
Food Policy Research Institute.

303
Tuholske, C., Andam, K. S., Blekking, J., Evans, T. and Caylor, K. (2020) ‘Comparing measures
of urban food security in Accra, Ghana’, Food Security, 12, pp. 417-431.

Tutu, R. A., Gupta, S., Elavarthi, S., Busingye, J. D. and Boateng, J. K. (2019) ‘Exploring the de-
velopment of a household cholera-focused health literacy scale in James Town, Accra’, Journal of
Infection and Public Health, 12(1), pp. 62-69.

Tyrväinen, L. (2001) ‘Economic valuation of urban forest benefits in Finland’, Journal of environ-
mental management, 62(1), pp. 75-92.

Tyrväinen, L., Pauleit, S., Seeland, K. and de Vries, S. (2005) ‘Benefits and uses of urban forests
and trees’, in Konijnendijk, C., Nilsson, K., Randrup, T. and Schipperijn, J. (eds.) Urban Forests
and Trees. Berlin: Springer, pp. 81-114.

Tyrväinen, L. and Väänänen, H. (1998) ‘The economic value of urban forest amenities: an appli-
cation of the contingent valuation method’, Landscape and urban planning, 43(1-3), pp. 105-118.

Ugolini, F., Massetti, L., Sanesi, G. and Pearlmutter, D. (2015) ‘Knowledge transfer between
stakeholders in the field of urban forestry and green infrastructure: Results of a European sur-
vey’, Land Use Policy, 49, pp. 365-381.

Uka, U. N. and Belford, E. J. (2016) ‘Inventory of Street Tree Population and Diversity in the
Kumasi Metropolis, Ghana’, Journal of forest and environmental science, 32(4), pp. 367-376.

UN-HABITAT (2011) Ghana Housing Profile. Nairobi: UN-HABITAT. Available at:

https://unhabitat.org/sites/default/files/documents/2019-07/ghana_housing_profile.pdf (Ac-
cessed: 16.09.2020).

United Nations (2005a) Household Sample Surveys in Developing and Transition Countries, Part I. New
York: United Nations. Available at: https://unstats.un.org/unsd/hhsurveys/ (Accessed:
16.09.2020).

United Nations (2005b) Household Sample Surveys in Developing and Transition Countries, Part II. New
York: United Nations. Available at: https://unstats.un.org/unsd/hhsurveys/ (Accessed:
16.09.2020).

United Nations (2019) World Urbanization Prospects – The 2018 Revision. New York: United Na-
tions. Available at: https://population.un.org/wup/Publications/Files/WUP2018-Report.pdf
(Accessed: 22.06.2021).

Uphoff, N. (1999) ‘Understanding Social Capital: Learning from the Analysis and Experience of
Participation’, in Dasgupta, P. and Serageldin, I. (eds.) Social Capital: A multifaceted perspective. Wash-
ington: The International Bank for Reconstruction and Development/The World Bank, pp. 215-
252.

USDA Forest Service (2016) i-Tree Eco: User’s Manual. v6.0, 2.26.2016, Washington: USDA.

Valentin, E. K. (2001) ‘SWOT analysis from a resource-based view’, Journal of marketing theory and
practice, 9(2), pp. 54-69.

304
van der Jagt, A., Kiss, B., Hirose, S. and Takahashi, W. (2021) ‘Nature-Based Solutions or Deba-
cles? The Politics of Reflexive Governance for Sustainable and Just Cities’, Frontiers in Sustainable
Cities, Vol. 2, 583833.

van der Jagt, A. P. and Lawrence, A. (2019) ‘Local government and urban forest governance: in-
sights from Scotland’, Scandinavian Journal of Forest Research, 34(1), pp. 53-66.

Vaz Monteiro, M., Handley, P. and Doick, K. J. (2020) ‘An insight to the current state and sus-
tainability of urban forests across Great Britain based on i-Tree Eco surveys’, Forestry: An Interna-
tional Journal of Forest Research, 93(1), pp. 107-123.

Vemuri, A. W., Grove, J. M., Wilson, M. A. and Burch, W. A. (2011) ‘Two Scales: Evaluating the
Relationship Among Life Satisfaction, Social Capital, Income, and the Natural Environment at
Individual and Neighborhood Levels in Metropolitan Baltimore’, Environment and Behavior, 43(1),
pp. 3-25.

Vogt, J. M., Watkins, S. L., Mincey, S. K., Patterson, M. S. and Fischer, B. C. (2015) ‘Explaining
planted-tree survival and growth in urban neighborhoods: A social–ecological approach to study-
ing recently-planted trees in Indianapolis’, Landscape and Urban Planning, 136, pp. 130-143.

von Hoffen, L. P. and Säumel, I. (2014) ‘Orchards for edible cities: Cadmium and lead content in
nuts, berries, pome and stone fruits harvested within the inner city neighbourhoods in Berlin,
Germany’, Ecotoxicology and Environmental Safety, 101, pp. 233-239.

Wallace, K. J. (2007) ‘Classification of ecosystem services: Problems and solutions’, Biological Con-
servation, 139, pp. 235-246.

Ward, C. D. and Shackleton, C. M. (2016) ‘Natural resource use, incomes, and poverty along the
rural-urban continuum of two medium-sized, South African towns’, World Development, 78, pp.
80-93.

Webb, R. (1998) ‘Urban forestry in Singapore’, Arboricultural Journal, 22(3), pp. 271-286.

Weihrich, H. (1982) ‘The TOWS matrix – A tool for situational analysis’, Long range planning,
15(2), pp. 54-66.

Wilkes‐Allemann, J., Pütz, M. and Hirschi, C. (2015) ‘Governance of Forest Recreation in Urban
Areas: Analysing the role of stakeholders and institutions using the institutional analysis and de-
velopment framework’, Environmental Policy and Governance, 25(2), pp. 139-156.

Williams, K. (2002) ‘Exploring resident preferences for street trees in Melbourne, Australia’, Jour-
nal of Arboriculture, 28(4), pp. 161-170.

Winsnes, S. (2009a) Two views from Christiansborg Castle, Vol. 1, Accra: Sub-Saharan Publishers.

Winsnes, S. (2009b) Two views from Christiansborg Castle, Vol. 2, Accra: Sub-Saharan Publishers.

Winsnes, S. (2013) A Danish Jew in West Africa: Wulff Joseph Wulff, Biography And Letters 1836-1842.
Accra: Sub-Saharan Publishers.

305
Woolcock, M. (1998) ‘Social capital and economic development: Toward a theoretical synthesis
and policy framework’, Theory and society, 27(2), pp. 151-208.

World Health Organization (2016) World Health Statistics 2016: Monitoring health for the SDGs. Ge-
neva: World Health Organization.

Wratten, E. (1995) ‘Conceptualizing urban poverty’, Environment and urbanization, 7(1), pp. 11-38.

Yang, J., McBride, J., Zhou, J. and Sun, Z. (2005) ‘The urban forest in Beijing and its role in air
pollution reduction’, Urban Forestry & Urban Greening, 3(2), pp. 65-78.

Yankson, P. W. K. (2012) ‘Rental Housing and Tenancy Dynamics with Particular Focus on
Low-income Households in Greater Accra Metropolitan Area’, in Ardayfio-Schandorf, E., Yank-
son, P. W. K. and Bertrand, M. (eds.) The Mobile City of Accra: Urban Families, Housing and Residen-
tial Practices. Dakar: CODESRIA, pp. 183-206.

Yazan, B. (2015) ‘Three approaches to case study methods in education: Yin, Merriam, and
Stake’, The qualitative report, 20(2), pp. 134-152.

Yeboah, A. K. (2016) Urban forest and residential class in Accra - The interrelationship between Urban forest
and Residential Area Classes - Case Study of the Ayawaso Sub-metro Districts in the Accra Metropolis. M.Sc.
Thesis. University of Eastern Finland. Available at:
https://epublications.uef.fi/pub/urn_nbn_fi_uef-20160486/ (Accessed: 14.09.2020).

Yeboah, I. E. A. (2000) ‘Structural adjustment and emerging urban form in Accra, Ghana’, Africa
Today, 47(2), pp. 61-89.

Yeboah, M. A. (2010) ‘Urban poverty, livelihood, and gender: Perceptions and experiences of
porters in Accra, Ghana’, Africa Today, 56(3), pp. 42-60.

Yin, R. K. (2003) Case Study Research: Design and Methods. 3rd Edition. London: SAGE.

Yin, R. K. (2009) Case Study Research: Design and Methods. 4th Edition. London: SAGE.

Yin, R. K. (2014) Case Study Research: Design and Methods. 5th Edition. London: SAGE.

Yu, C. and Hien, W. N. (2006) ‘Thermal benefits of city parks’, Energy and buildings, 38(2), pp. 105-
120.

Zhang, C., Fang, Y., Chen, X. and Congshan, T. (2019) ‘Bibliometric Analysis of Trends in Glob-
al Sustainable Livelihood Research’, Sustainability, 11(4), 1150.

Zulu, L. C. (2010) ‘The forbidden fuel: Charcoal, urban woodfuel demand and supply dynamics,
community forest management and woodfuel policy in Malawi’, Energy Policy, 38, pp. 3717-3730.

306
 APPENDICES

1 Participant Information Sheets

1.1 Participant Information Sheet for Survey Respondents
My name is Lyn Hosek and as part of my PhD degree in African Studies at the University of
Birmingham (UK), I am carrying out research on trees and their uses in the city of Accra. I would
like to collect information about these trees to know what officials and ordinary people think
about planting, using and protecting them. I also want to know whether people like you who live
in Accra own trees, what you use them for or get from them and what you think about the value
of trees (for eg, do you like having them around? Do you think they are useful things? Would you
like to see/have more around? Who is responsible for them?).
As part of my research I am conducting a household survey to ask questions about these things
and I am inviting you and the other people in your household to participate. Your household was
randomly selected as a possible respondent. The survey consists of two parts: The first part con-
tains questions about the members of your household and I am expecting this part to take ap-
proximately 5 to 10 minutes. The second part contains questions about urban trees and their ser-
vices and products. I am expecting this part to take about 30 minutes. I would like to ask these
questions to as many members of your household as possible, in the form of a group meeting.
You and the other people in your household do not have to take part in this survey. Your partici-
pation is completely voluntary. If you agree to participate, I will ask each person taking part to
sign a consent form to say that they are willing to take part. You can withdraw from the study at
any point if you change your mind during the research, and for up to four weeks after I finish
talking to people in the household. You will not have to give any reason for changing your mind.
If you decide not to continue with the survey, I will not use any information you provide in my
work. In this case, all your information will be destroyed.
This project has received ethics approval from the University of Birmingham’s Humanities and
Social Sciences Ethics Committee. I do not expect participation in the survey to have any positive
or negative effects on you.
Because I will be spending plenty of time with you and other members of the household if you
agree to take part in my research, you cannot be anonymous as people will know I have been
talking to you about my work. However, nobody outside the household, family or compound will
know what you say to me or a research assistant as this information will be treated as confiden-
tial. Your personal information will be stored safely, locked or password-protected. It will only be
accessible to the researcher and other authorised researchers for verification purposes. Other
people helping with this research will not have access to the original data (apart from a research
assistant where they have asked the questions) and will not be able to trace any information back
to you. The information will be treated in accordance with the Data Protection Act 1998. It will
be stored for ten years and then deleted/destroyed after.
I will use the results of the survey in my thesis and any publications and/or workshop or confer-
ence presentations that might arise from it. I will not use your name or address when I use the in-
formation you provide. If you wish, you can receive feedback about the results of the study. You
can specify what kind of feedback you would like to receive, and how, on the consent form.

307
Please feel free to ask for further clarification should anything be unclear. If you agree to partici-
pate in the survey, please read the consent form and sign it.

1.2 Participant Information Sheet for Minors – Household Survey

Hello, my name is Lyn and I am studying at a university in England. I am trying to find out things
about the trees growing in Accra. I want to know what trees are growing here and where they are
growing. I also want to know what people use trees for, what they get from the trees and what
they think about trees (for eg, do people like having trees around? do they think trees are useful
things? would they like to see/have more trees around?).
For my work I am talking to families and other people who live together in the same household
to find out what they think about the trees in Accra, whether they use them, and what for, who
looks after them and how, and what they think should happen to the trees in future. I would like
to talk to you and other members of your family and household for around half an hour. Your
parents/guardian say they are happy for you to talk to me if you want to.
You don’t have to talk to me if you don’t want to. If you agree to talk to me you can change your
mind at any time and not answer any more questions or answer only some of my questions. If
you decide you don’t want to be part of my work anymore, you can say so at any time, even up to
four weeks after we have finished talking. You will not need to tell me why you have changed
your mind. If you decide you no longer want to be part of my work after you have given me in-
formation, I will not use what you have told me in my work.
I will treat everything you tell to me as secret or confidential. That means that I will not use your
name or the place where you live when I write about you in my work or when I talk about what
you tell me to my lecturers and colleagues at university, or to other people at small meetings
and/or big conferences where we talk about people and trees in Accra and other places.
If you want me to, I can send you more information about my work when it is finished. For eg, I
can send you copies of things I write.
If there is anything you have not understood, please ask me to explain it better. If you want to
talk to me I need you to write your name on the form below or tell me that you have understood
what I would like you to do and that you want to be part of my work.

1.3 Participant Information Sheet for Stakeholder Interviews

My name is Lyn Hosek and as part of my PhD degree in African Studies at the University of
Birmingham (UK), I am carrying out research on the trees in the city of Accra. The objective of
the project is to collect information about these trees and to investigate planning and manage-
ment strategies and goals regarding their propagation, use and protection. It further aims to ex-
plore what the inhabitants of Accra think about the role of urban trees in and for their lives and
how they make use of these trees and their services and products.
As part of the research, I am conducting interviews and I am inviting you to participate, as you
were identified as a possible respondent due to being privately/professionally involved in this ar-

308
ea. I am expecting the interview to take approximately 60 minutes. If you agree, the interview will
be audio-taped or video-taped and I might take photos during the interview.
You do not have to take part in this interview. Your participation is completely voluntary. If you
agree to participate, I will ask you to sign a consent form. You can withdraw from the study at
any point during participation and up to four weeks after without giving reasons. In this case, all
your data will be destroyed.
This project has received ethics approval from the University of Birmingham’s Humanities and
Social Sciences Ethics Committee. I do not expect the interview to have any positive or negative
effects on you.
Your data will be treated as confidential, but I cannot guarantee that you will not be identifiable,
as contextual information that I need to provide, e.g. the type of work you do, might make it pos-
sible for some people to find out who you are.
Consequently, you cannot be anonymous but your data will be treated as confidential. Your per-
sonal information will be stored safely, locked or password-protected. It will only be accessible by
the researcher and other authorised researchers for verification purposes. Other people helping
with this research will not have access to the original data, only modified versions that do not link
your responses to your personal information. Your personal information will be treated in ac-
cordance with the Data Protection Act 1998. It will be stored for ten years and deleted/destroyed
after that.
If you wish to participate please indicate your willingness on the consent form, which also allows
you to say if you would like your contribution to be linked to your name. I will present the results
of the interview in my thesis and any publications and/or presentations that might arise from the
research.
If you wish, you can receive feedback about the results of the study. You can specify what kind
of written or other feedback you would like to receive, and how, on the consent form.
Please feel free to ask for further clarification should anything be unclear. If you agree to partici-
pate in the interview, please read the consent form and sign it.

1.4 Participant Information Sheet for Household Case Study

My name is Lyn Hosek and as part of my PhD degree in African Studies at the University of
Birmingham (UK), I am carrying out research on the trees in the city of Accra. The main reason
behind the project is to find out where and what kind of trees are found in Accra; what purpose
they serve; and what their future might be. I also want to know what government and other offi-
cials and ordinary people in Accra think about planting, using and protecting these trees, and who
owns, controls, manages and uses the trees. The objective of the project is to collect information
about these trees and to investigate planning and management strategies and goals. It further
aims to explore what the inhabitants of Accra think about the role of urban trees in and for their
lives and how they make use of urban trees and their services and products I also want to know
whether people like you who live in Accra own trees, what you use them for or get from them
and what you think about the value of trees (for eg, do you like having them around? Do you

309
think they are useful things? Would you like to see/have more around? Who is responsible for
them?).
As part of the research I am doing household case studies and I am inviting your household to
participate in this case study. Your household was identified as a possible participant due to the
answers that you provided in the questionnaire survey you previously participated in. Over the
course of about two to three weeks I would like to spend a total of around fifty hours with the
members of your household. During this time I would like to engage in conversations with you
and conduct interviews which, if you agree, might be audio-taped or video-taped. If you agree, I
might also take some photos. I would also like to observe tree-related activities of household
members about what exactly they do, and how and why. The research might also include rating
answers or producing a timeline of your activities throughout the year.
As before, your household does not have to take part in the case study research. Your participa-
tion is still completely voluntary. If you agree to participate, I will again ask all the members of
your household to sign a consent form. You can withdraw from the study at any point and for up
to four weeks after I finish my research with you and the rest of the household without giving
reasons. If you decide not to continue taking part in the research for whatever reason, I will not
use any of the information you provide for this part of the study in my work. The information
will be destroyed in that case.
This project has received ethics approval from the University of Birmingham’s Humanities and
Social Sciences Ethics Committee.
I need you to be aware of the following issue: The data I hope to collect on your household will
be quite substantial, detailed and private in nature. Your data will be treated as confidential, but I
cannot guarantee that you will not be identifiable, as contextual information that I need to pro-
vide, e.g. the setup of your household or the type of work you do, might make it possible for
some people to find out who you are. Additionally, the time commitment is quite significant.
Therefore, please take your time when deciding if your household wants to participate and feel
free to ask any questions that you have. Due to the nature of the case study, you can only partici-
pate if all household members wish to do so.
Due to the nature of the project, you cannot be anonymous but your data will be treated as con-
fidential. Your personal information will be stored safely, locked or password-protected. It will
only be accessible by the researcher and other authorised researchers for verification purposes.
Other people helping with this research will not have access to the original data, but just to modi-
fied versions that do not link your responses to your personal information. Your personal infor-
mation will be treated in accordance with the Data Protection Act 1998. It will be stored for ten
years and deleted/destroyed after.
I will use the results of the case study in my thesis and any publications and/or workshop or con-
ference presentations that might arise from it. I will not use your name or address when I use the
information you provide. However, if you and the other household members are happy for me
to, I can/will identify you by name in my work. If you wish, you can receive feedback about the
results of the study. You can specify what kind of feedback you would like to receive, and how,
on the consent form.
Please feel free to ask for further clarifications should anything still be unclear. If all household
members agree to participate in the case study, please read the consent form and sign it.

310
1.5 Participant Information Sheet for Minors – Case Study
Hello, my name is Lyn and I am studying at a university in England. I am trying to find out things
about the trees growing in Accra. I want to know what trees are growing here and where they are
growing. I also want to know what people use trees for, what they get from the trees and what
they think about trees (for eg, do people like having trees around? do they think trees are useful
things? would they like to see/have more trees around?).
For my work, I am talking to families to find out what they know and think about the trees in
Accra, whether and how they use them, and what they think should happen to them in future. I
want to talk to you and your family because you have talked to me before and I have some more
questions. I want to spend around a total time of 50 hours with your family over about two to
three weeks doing different things like asking questions and watching what you are doing with
trees and what you can get from them, etc. If you are ok with it, I might audio-record or video-
record some of our talks and take some pictures. Your parents/guardians say it is ok for you to
talk to me if you want to.
You don’t have to talk to me if you don’t want to. If you agree to talk to me you can change your
mind at any time and not answer any more questions or answer only some of my questions. If
you decide you don’t want to be part of my work anymore, you can say so at any time, even up to
four weeks after we have finished talking. You will not need to tell me why you have changed
your mind. If you decide you no longer want to be part of my work after you have given me in-
formation, I will not use what you have told me in my work.
I will treat everything you tell to me as secret or confidential. That means that I will not use your
name or the place where you live when I write about you in my work or when I talk about what
you tell me to my lecturers and colleagues at university, or to other people at small meetings
and/or big conferences where we talk about people and trees in Accra and other places. Howev-
er, if you and the other people in your house or family are happy for me to, I will mention you by
name in my work.
If you want me to, I can send you more information about my work when it is finished. For eg, I
can send you copies of things I write, photographs, voice recordings, etc.
If there is anything you have not understood, please ask me to explain it better. If you want to
talk to me, I need you to write your name on the form below or tell me that you have understood
what I would like you to do and that you want to be part of my work.

311
2 Consent Forms
2.1 Consent Form for Survey Respondents
This information is being collected as part of a research project concerned with urban trees in
the city of Accra, their management and their role in the lives of city inhabitants. The research
is carried out by a PhD researcher from the Department of African Studies and Anthropology
at the University of Birmingham (UK). The information that you supply and that will be col-
lected as part of this research will be entered into a filing system or database. Your personal
information will be stored locked/password protected and will only be accessed by authorised
personnel involved in the project and other researchers for verification purposes. The infor-
mation you provide will be retained by the University of Birmingham and will only be used for
the purpose of research and statistical and audit purposes. By supplying this information, you
are consenting to the University storing your information for the purposes stated above. The
information will be processed by the University of Birmingham in accordance with the provi-
sions of the Data Protection Act 1998.
YES NO
I confirm that I have read and understood the participant information sheet. I
had the chance to ask questions, which were answered satisfactorily.
I understand that my participation is voluntary and that I am free to withdraw
at any point during participation and up to four weeks afterwards without giv-
ing reasons. If I withdraw, my data will be removed from the study and will be
destroyed.
I wish to receive written feedback once the research has been completed
(please specify which type of written feedback you would like).
Summary/E-mail Summary/Post Complete work/E-mail
E-mail:
I understand that my personal data will be processed for the purposes detailed
above, in accordance with the Data Protection Act 1998.
I give permission for _________________________________________
(name of minor) to participate.
Based upon the above, I agree to take part in this survey.
Consent was given orally.

----------------------------------- --------------- -------------------------------------

Name Participant Date Signature Participant

------------------------------------ -------------- -------------------------------------

Name Researcher/Data collector Date Signature Researcher/Data collector26

26 Each consent form also included the contact details of my supervisor and me
312
2.2 Consent Form for Minors – Household Survey

YES NO
I understand what your work is about and what I have to do. I have asked for
explanations of things I didn’t understand and have no more questions.
I understand that I don’t have to be part of the work. I can stop at any time
up to four weeks after answering questions. I understand that I won’t have to
explain why if I decide not to continue as part of your work. If I don’t want
to be part of the work anymore, what I have said will not be used in the work.
I want to find out more about the work when it is done.
Summary/E-mail Summary/Post Complete work/E-mail
E-Mail:

I want to be part of the work.

Consent was given orally.

------------------------------------- --------------- -------------------------------------

Name Participant Date Signature Participant

------------------------------------ -------------- -------------------------------------

Name Researcher/Data collector Date Signature Researcher/Data collector

313
2.3 Consent Form for Stakeholder Interviews
This information is being collected as part of a research project concerned with urban trees in
the city of Accra, their management and their role in the lives of city inhabitants. The research
is carried out by a PhD researcher from the Department of African Studies and Anthropology
at the University of Birmingham (UK). The information that you supply and that will be col-
lected as part of this research will be entered into a filing system or database. Your personal
information will be stored locked/password protected and will only be accessed by authorised
personnel involved in the project and other researchers for verification purposes. The infor-
mation will be retained by the University of Birmingham and will only be used for the purpose
of research and statistical and audit purposes. By supplying this information, you are consent-
ing to the University storing your information for the purposes stated above. The information
will be processed by the University of Birmingham in accordance with the provisions of the
Data Protection Act 1998. No identifiable personal data will be published unless requested or
approved by the participant.

YES NO
I confirm that I have read and understood the participant information sheet. I
had the chance to ask questions, which were answered satisfactorily.
I understand that my participation is voluntary and that I am free to withdraw
at any point during participation and up to four weeks after without giving rea-
sons. If I withdraw, my data will be removed from the study and will be de-
stroyed.
I wish my name/identity to be kept confidential
I agree to the interview being audio-taped
I agree to the interview being video-taped
I agree to being photographed
I wish to receive written feedback once the research has been completed
(please specify which type of written feedback you would like).
Summary/E-Mail Summary/Post Complete work/E-Mail
Photos/E-Mail Photos/Post Audio-recording/E-Mail
Video-recording/E-Mail
E-Mail:
I understand that my personal data will be processed for the purposes detailed
above, in accordance with the Data Protection Act 1998.
I give permission for _________________________________________
(name of minor) to participate.
Based on the above, I agree to take part in this interview.
Consent was given orally.

------------------------------------- --------------- -------------------------------------

Name Participant Date Signature Participant

------------------------------------ -------------- -------------------------------------

Name Researcher Date Signature Researcher

314
2.4 Consent Form for Household Case Study
This information is being collected as part of a research project concerned with urban trees in the
city of Accra, their management and their role in the lives of city inhabitants. The research is car-
ried out by a PhD researcher from the Department of African Studies and Anthropology at the
University of Birmingham (UK). The information that you supply and that will be collected as
part of this research will be entered into a filing system or database. Your personal information
will be stored locked/password protected and will only be accessed by authorised personnel in-
volved in the project and other researchers for verification purposes. The information will be re-
tained by the University of Birmingham and will only be used for the purpose of research and
statistical and audit purposes. By supplying this information, you are consenting to the University
storing your information for the purposes stated above. The information will be processed by the
University of Birmingham in accordance with the provisions of the Data Protection Act 1998.
No identifiable personal data will be published unless otherwise wished by the participant.

YES NO
I confirm that I have read and understood the participant information
sheet. I had the chance to ask questions, which were answered satisfactori-
ly.
I understand that my participation is voluntary and that I am free to with-
draw at any time during participation and up to four weeks after without
giving reasons. If I withdraw, my data will be removed from the study and
will be destroyed.
I wish my name to be kept confidential
I agree to being audio-taped
I agree to being video-taped
I agree to being photographed
I wish to receive written feedback once the research has been completed
(please specify which type of written feedback you would like).
Summary/E-Mail Summary/Post Complete work/E-Mail
Photos/E-Mail Photos/Post Video-recording/E-Mail
Audio-recording/E-Mail
E-Mail:
I understand that my personal data will be processed for the purposes de-
tailed above, in accordance with the Data Protection Act 1998.
I give permission for ______________________________________
(name of minor) to participate.
Based upon the above, I agree to take part in this household case study.
Consent was obtained orally.

------------------------------------- --------------- -------------------------------------

Name Participant Date Signature Participant

------------------------------------ -------------- -------------------------------------

Name Researcher Date Signature Researcherer

315
2.5 Consent Form for Minors – Case Study

YES NO
I understand what your work is about and what I have to do. I have asked
for explanations of things I didn’t understand and have no more questions.
I understand that I don’t have to be part of the work. I can stop at any time
up to four weeks after answering questions. I understand that I won’t have
to explain why if I decide not to continue as part of your work. If I don’t
want to be part of the work anymore, what I have said will not be used in
the work.
I want my name to be a secret
I agree to being audio-recorded
I agree to being video-recorded
I agree to being photographed
I want to find out more about the work when it is done.
Summary/E-Mail Summary/Post Complete work/E-Mail
Photos/E-Mail Photos/Post Audio-recording/E-Mail
Video-recording/E-Mail
E-Mail:
I want to be part of the work
Consent was given orally

------------------------------------- --------------- -------------------------------------

Name Participant Date Signature Participant

------------------------------------ -------------- -------------------------------------

Name Researcher Date Signature Researcher

316
3 Neighbourhood Poverty Ratings
Moderate Very High
Non-Poverty Low Poverty High Poverty
Poverty Poverty
Airport Resi-
Dansoman Accra Central Abeka Ussher Town
dential
Dzorwulu Mataheko Adabraka Abofu Nima
North Dzor- Airport West
Lartebiokorshie Abossey Okai
wulu Residential
Abelenkpe Labone Alajo Akweteyman
Roman Ridge East Legon Alogboshie Chorkor
Achimota Col-
North Ridge Avenor Gbegbeyise
lege
West Ridge Ringway Estate Awudome Kanda
South Industri-
Bubuashie Korle Gonno
al
Asylum Down Cantonments Kotobabi
North Industri-
Dakuman Kpehe
al Area
East Legon Ex-
Kokomlemle Kwashieman
tension
North Kanesh-
James Town Mamobi
ie
Tesano Kaneshie Mamponse
Old Tesano Kissieman Mamprobi
Korle Bu New Town
Korle Dudor North Odorkor
La Osu
La Dadekotopon Russia
Legon Sabon Zongo
Ministerial Area Sempe
Mpoase South La
New Fadama South Odorkor
North Alajo Sukura
South Legon Zoti
Tudu Nyanba Estate
Old Mamprobi
West Abossey
Okai
Lapaz

317
4 Tree Inventory Form
 Check when plot is completed

Plot Information
Plot Stratum: GPS Coordinates Date: Crew:
ID: Lat.
Long.

Percent of Plot measured (%) Plot Tree Cover (%)

Did this Plot have any Trees? (Y/N):_______

Actual Land Use

Actual Land Use: Percent of plot:
Actual Land Use: Percent of plot:
Actual Land Use: Percent of plot:
Actual Land Use: Percent of plot:
Actual Land Use: Percent of plot:
Actual Land Use: Percent of plot:

Ground cover (% of plot covered by)

Bldg. Cement Tar Rock Soil Duff/ Herb/ Main. Unmain. Water
Mulch Ivy Grass Grass

Additional Comments (e.g. about access to plot or plot itself):

318
SURVEY
DATE:
PLOT
ID DBH

TOT LIVE CROWN

TREE LAND HT DBH DBH DBH DBH DBH DBH CROWN
TREE SPECIES TREE CROWN BASE
ID USE DBH 1 2 3 4 5 6 COND
HEIGHT HEIGHT HEIGHT

10

319
 CROWN WIDTH TREE GPS MANAGEMENT
LONG SIDE
TREE CROWN STREET PUBLIC LAT MAINT UTILITY
N-S E-W CLE COORD WALK COMMENTS
ID % MISS TREE TREE COORD Y RECOM CONFL
X CONFL

10

320
5 Household Questionnaire
Section A - Basics
Household ID and description of lo- Interviewer: Date: Language: Comments (e.g. on replace-
cation: ment):

1. ID 2. Household 3. 4. 5. What is this 6. To which 7. What are this 8. How long 9. What is this
House- position (in rela- Sex How person’s na- ethnic group person’s religious has this per- person’s highest
hold tion to head) old is tionality? does this per- believes? son lived in finished educa-
mem- this son belong? Accra? tional qualifica-
ber 1 – Head per- 1 – Ghanaian 1 – No religion tion?
2 – Spouse son? 2 – ECOWAS 1 – Akan 2 – Catholic Based on an-
3 – Child 3 – Other 2 – Ga- 3 – Protestant swer and age: 1 – None
4 – Grandchild African Dangme 4 – Pentecostal/ Has this per- 2 – Kindergarten
5 – Parent 4 – American 3 – Ewe Charismatic son always 3 – Primary
6 – Parent-in-law 5 – Asian 4 – Guan 5 – Other Chris- lived in Ac- 4 – JHS/JSS
7 – Child-in-law 6 – European 5 – Gurma tian cra? 5 – SHS/SSS
8 – Other rela- 7 - Oceanian 6 – Mole- 6 – Muslim 6 – Technical/
tive Dagbani 7 – Traditionalist 1 – Yes Vocational
9 – Non-relative 7 – Grusi 8 – Other (speci- 2 – No 7 – Undergraduate
10 – Househelp 8 – Mande fy) (Bachelor)
11 – Other 9 – Other 9 – Don’t know If no: 8 – Undergraduate
(specify) (specify) a. Where did (HND)
10 – Don’t this person 9 – Postgraduate
know live before? 10 – Don’t know

321
Section B – Progress out of Poverty
Indicator Value Points Score
A. Eight or more 0
B. Seven 4
C. Six 9
D. Five 13
1. How many members does the household
E. Four 14
have?
F. Three 21
G. Two 24
H. One 29
A. No 0
2. Are all household members ages 5 to 17
B. Yes 2
currently in school?
C. No one ages 5 to 17 3
A. No 0
3. Can the male head/ spouse read a phrase/
B. No male head/spouse 2
sentence in English?
C. Yes 5
A. Mud bricks/earth, wood, bamboo, metal 0
4. What is the main construction material
sheet/slate/asbestos, palm leaves/ thatch (grass/raffia), or other
used for the outer wall?
B. Cement/concrete blocks, landcrete, stone, or burnt bricks 5
A. No toilet facility (bush, beach), or other 0
5. What type of toilet facility is usually used B. Pit latrine, bucket/pan 4
by the household? C. Public toilet (e.g. W.C., KVIP, pit pan) 4
D. KVIP, or W.C. 6
A. None, no cooking 0
6. What is the main fuel used by the house- B. Wood, crop residue, sawdust, animal waste, or other 6
hold for cooking? C. Charcoal, or kerosene 13
D. Gas, or electricity 22
7. Does any household member own a A. No 0
working box iron or electric iron? B. Yes 4
8. Does any household member own a A. No 0
working television, video player, B. Only television 2

322
 VCD/DVD/MP3/MP4 player/iPod, or C. Video player, VCD/ DVD/MP3/MP4 player/iPod, or satel- 8
satellite dish? lite dish (regardless of T.V.)
A. None 0
9. How many working mobile phones are B. One 4
owned by members of the household? C. Two 8
D. Three or more 10
A. None 0
10. Does any household member own a
B. Only bicycle 3
working bicycle, motor cycle, or car?
C. Motor cycle or car (regardless of bicycle) 8

Section C – General Questions Public Trees

When answering the following questions, please consider only public trees, such as those along roads or in parks, that are growing with-
in the city of Accra

1. BENEFITS
a. Can you please tell me all the benefits and purposes that you think these trees have? Both for the city and its residents generally and
for you and your household specifically.
A E I M
B F J N
C G K O
D H L P

I would now like you to rank some benefits of public trees. First, according to how important you think they are for the well-being of
Accra’s residents generally. Then according to how important they are for the well-being of your household or some of its members spe-
cifically.
------------------------------------------------------------------------------------------------------------------------------------------------------

323
b. Ranking for Accra’s residents generally

c. Based on this ranking, can you please divide the list into three groups: Benefits that you consider to be of (1) high importance, (2) me-
dium importance, and (3) low importance for the well-being of the city’s residents. You do not need to use all three groups and you can
divide the lists however you want.
-----------------------------------------------------------------------------------------------------------------------------------------------------
d. Ranking for household and members specifically

e. Based on this ranking, can you please also divide the lists into three groups: Benefits that you consider to be of (1) high importance,
(2) medium importance, and (3) low importance for the well-being of your household or some of its members specifically. You do not
need to use all three groups, and you can divide the lists however you want.
-----------------------------------------------------------------------------------------------------------------------------------------------------
f. Are there any public trees in your area of housing that have caught your attention or are especially important for you?
Yes ☐ No ☐ Don’t know ☐
i. If ‘yes’: Can you please tell me what kind of trees and why?

g. Are there any public trees elsewhere in the city that have caught your attention or are especially important for you?
Yes ☐ No ☐ Don’t know ☐
i. If ‘yes’: Can you please tell me what kind of trees and why?

2. MANAGEMENT
a. Would you like to have more public trees in Accra? Yes ☐ No ☐ Don’t know ☐
i. Can you please explain why/why not?

324
 ii. If ‘yes’: Are there any specific trees you would like to see planted? Yes ☐ No ☐ Don’t know ☐
A. If ‘yes’: Which and why?

iii. If ‘yes’: Are there any specific places where you would like them to be planted? Yes ☐ No ☐ Don’t know ☐
A. If ‘yes’: Where and why?

b. Do you think that the public trees are generally in a good condition? Yes ☐ No ☐ Don’t know ☐
i. If ‘no’: Can you please explain what makes you think so?

c. Do you think that the city is generally taking good care of the public trees? Yes ☐ No ☐ Don’t know ☐
i. If ‘no’: Can you please explain what makes you think so?

3. WISHES, THREATS and PROBLEMS

a. Do you see any threats or problems affecting the public trees? Yes ☐ No ☐ Don’t know ☐
i. If ‘yes’: Can you please tell me what threats or problems you see?

b. If you could, would you change anything about the public trees? Yes ☐ No ☐ Don’t know ☐
i. If ‘yes’: Can you please tell me what you would change and why?

4. ADDITIONAL INFORMATION
a. Is there anything else you would like to say about public trees in Accra?

325
Section D – General Questions Public Trees

1. Does your household own land in Accra that has trees on it? Yes ☐ No ☐ Don’t know ☐
a. If ‘no’: Would you like to own land with trees in Accra? Yes ☐ No ☐ Maybe ☐ Don’t know ☐
i. Can you please explain why/why not/why maybe?

b. If ‘yes’: Are any of these trees owned by all household members together? Yes ☐ No ☐ Don’t know ☐
c. If ‘yes’: Does any household member own other land elsewhere in the city with trees on it? Yes ☐ No ☐ Don’t know ☐
IF YES, FILL IN TABLE
A. Who are the owners? (ID/relationship
to owner)
B. How many trees do you own today?
C. Has this number changed over the past?
C1. If ‘yes’: How?

D. Can you name specific types of trees

that you own?

E. Are there any reasons for owning the

specific trees that you do?

326
 F. Has the type of trees you own changed
over the past
F1. If ‘yes’: How?
G. Is anyone looking after the trees?
G1. If yes: Can you please tell me more
about this? I would e.g. like to know who
is responsible for this, what they are doing
and why, when they are taking care of the
trees and how often.
G2. If no: Why is nobody ‘looking after’
the trees?

H. Anything you would like to add about

the management and care of your trees?

2. Is there anything else you would like to say about privately owned trees in Accra, either owned by the household, someone from the
household or someone else?

Section E – Tree Products Household Use

Now I have some questions about the sale of products coming from public and private trees in Accra. Please consider original tree
parts such as fruits, leaves or wood as well as products that have been made from these such as juices, medicine or household items

327
1. In the past 12 months, has any member of your household sold [tree part(s)] or a product derived from them? (fill in first table:
Yes – 1, No – 2, Don’t know – 3)
If ‘yes’: Can you please tell me what kind of trees the [tree product(s)] was/were from? (fill in 2nd table below) (if unknown just tree part)

a. Fruits or b. Seeds c. Flowering d. Leaves e. Wood or f. Bark g. Roots h. Tree liquids

nuts parts woody parts (saps, resins, latex)

Tree part Tree type (if un- 2. How frequently have 3. Did a member of the house- 4. Were any of these pro-
known: write un- you sold hold get some of these from a cessed before you sold
known) [tree part(s)] of [tree tree within the city? them?
type] in the past 12 Yes – 1
months? No – 2 Yes – 1
Don’t know – 3 No – 2
Daily – 1 Don’t know – 3
Weekly – 2 If ‘yes’: Did any of these come
Monthly – 3 from a: If ‘yes’: By:
Less than monthly – 4 public tree – 1 a household member – 1
household owned tree – 2 someone else – 2
another private tree – 3

328
Section F – Tree Product Selling

Now I have some questions about the sale of products coming from public and private trees in Accra. Please consider original tree parts
such as fruits, leaves or wood as well as products that have been made from these such as juices, medicine or household items

1. In the past 12 months, has any member of your household sold [tree part(s)] or a product derived from them? (fill in first table: Yes – 1,
No – 2, Don’t know – 3)
If ‘yes’: Can you please tell me what kind of trees the [tree product(s)] was/were from? (fill in 2nd table below) (if unknown just tree part)
a. Fruits or b. Seeds c. Flowering d. Leaves e. Wood or f. Bark g. Roots h. Tree liquids (saps, resins,
nuts parts woody parts latex)

Tree part Tree type (if 2. How frequently have 3. Did a member of the household get 4. Were any of these pro-
unknown: you sold some of these from a tree within the cessed before you sold them?
write un- [tree part(s)] of [tree city?
known) type] in the past 12 Yes – 1 Yes – 1
months? No – 2 No – 2
Don’t know – 3 Don’t know – 3
Daily – 1
Weekly – 2 If ‘yes’: Did any of these come from a: If ‘yes’: By:
Monthly – 3 public tree – 1 a household member – 1
Less than monthly – 4 household owned tree – 2 someone else – 2
another private tree – 3

329
6 List of Newspaper Articles
No. Title Date Author Newspaper
No name
1 KMA to transform Garden City 10.04.2014 Daily Graphic
(NN)
KMA Launches ‘Greening Kumasi’
2 15.05.2014 Freiku, S. R. The Chronicle
Project
KMA to plant one million trees in me-
3 15.05.2014 NN Daily Graphic
tropolis
Kumasi to become Garden City once
4 22.05.2014 NN Daily Graphic
again
‘Me And My Tree’ Competition Takes Owusu-
5 28.05.2014 The Chronicle
Off In Kumasi Akyaw, R.
6 KMA begins planting one million trees 08.08.2014 Baidoo, F. A. Daily Graphic
Restoring greenery of a pounded city –
7 17.07.2017 NN Daily Graphic
Will KMA succeed this time?
8 KMA Plants 70,000 Trees 19.07.2017 Awuah Jnr., J. Daily Guide
Kumasi observes National Sanitation
9 08.08.2017 Baah, E. Daily Graphic
Day
KMA launches ‘keep Kumasi clean and
10 06.10.2017 NN Daily Graphic
green’
Green Kumasi project progresses –
11 16.08.2017 NN Daily Graphic
7000 seedlings planted
KMA disburses GH₵381,167 to 367
12 19.12.2016 NN The Chronicle
PLWDs
KMA celebrates one year of Rattray
13 13.07.2016 NN Daily Graphic
Park
Mahama, Otumfuo to commission rec-
14 17.06.2015 NN Daily Graphic
reational park in Kumasi
Kumasi Rattray park closed down for
15 23.09.2016 NN Daily Graphic
non-payment of tax
Kumasi: The Garden City without gar-
16 31.07.2013 Asare, G. E. Daily Graphic
dens?
17 The beautiful green danger in Takoradi 02.12.2014 NN Daily Graphic
Prez Mahama inaugurates renovated Duodu, S. and
18 16.11.2015 Daily Graphic
Tamale Market Gyes, Z. K.
19 Tamale Assembly to green Metropolis 16.11.2015 NN Daily Graphic
Why we need open spaces in human Intsiful, G.
20 23.05.2016 Daily Graphic
settlements W. K.
Minister Outdoors Committee To Plan
21 24.05.2013 Adam, F. The Chronicle
Eco-Park
22 Spiritual enclave to be created in 02.10.2013 Enin, E. Daily Graphic

330
 Achimota Forest
Stop planned destruction of Achimota
23 Forest – Nduom tells President Maha- 11.11.2013 Nduom, P. K. Daily Graphic
ma
Mr. President, Stop The Destruction Of
24 11.11.2013 Nduom, P. K. The Chronicle
Achimota Forest!
An open letter to Prez Mahama: Stop
25 15.11.2013 Nduom, P. K. Daily Graphic
the destruction of Achimota Forest
26 Rescuing Achimota Forest Reserve 18.11.2013 NN Daily Graphic
Achimota Forest ecotourism project set
27 03.01.2014 NN Daily Graphic
to begin
Chief Seizes Achimota Forest…School
28 Under Threat, As Lands Commission 06.06.2014 Akli, E. The Chronicle
Fails To Defend Suit
Battle of Achimota…Legal Bullets Ric-
29 09.06.2014 Akli, E. The Chronicle
ochet Across Terrain
Endangered Species Near Extinc-
30 tion…As Depletion Of Achimota For- 11.06.2014 Nonor, D. The Chronicle
est Continues
Chief Seizes Achimota Forest…As For-
31 26.06.2014 Adjei, N. O. The Chronicle
estry Commission Fails To Defend Suit
Fc Seeks Partnership For Achimota Ghanaian
32 21.07.2014 NN
eco-Park Times
Achimota Forest to be turned into rec-
33 07.08.2014 NN Daily Graphic
reational park
34 Accra Eco-Park Project Gets Suitors 07.08.2014 Nonor, D. The Chronicle
Forestry Commission, Police And
35 26.11.2014 NN The Chronicle
Achimota Forest
MPs, Police In Hot Exchange Over FC Attenkah, R.
36 26.11.2014 The Chronicle
Lands and Adam, F.
Forestry Commission, police in tango
37 26.11.2014 NN Daily Graphic
over Achimota Forest
Protecting Achimota For-
38 est…Parliamentary Select C’ttee, team 27.11.2014 NN Daily Graphic
of policemen clash
Forestry Commission, devt partner sign
39 agreement to make Achimota Forest 20.02.2016 NN Daily Graphic
ecopark
Nduom opposes sale of Achimota For-
40 25.02.2016 NN Daily Graphic
est
41 Four sue gov’t over sale of Achimota 27.04.2016 NN Daily Graphic

331
 Forest
Sale of Achimota forest – four go to
42 17.05.2016 NN Daily Graphic
court
Achimota Forest: Major Ecotourism
43 17.05.2016 NN Daily Graphic
destination
Disaster looms if Achimota Forest Eco-
44 tourism Project goes ahead – NDPC 16.06.2016 NN Daily Graphic
boss
45 Must we cut our nose to spite our face? 17.06.2016 NN Daily Graphic
$1.2 billion Accra Eco-Park Project
46 20.08.2016 NN Daily Graphic
takes off
Development of Accra Eco Park now
47 22.08.2016 NN Daily Graphic
on Course
Achimota Forest Not For Sale - Maha-
48 22.08.2016 Larbi, C. Daily Guide
ma
Achimota Forest shouldn’t be turned
Ghanaian
49 into Eco Park – Prof. Frimpong- 13.03.2017 NN
Times
Boateng
Forestry Commission exceeds budget
50 22.12.2013 NN Daily Graphic
by 56 per cent
“Garden & Flower Show” To Boost
51 16.08.2013 Twum, C. The Chronicle
Arts & Crafts
52 Flower and Garden Show begins today 29.08.2013 NN Daily Graphic
Use gardens, flowers as avenue for job
53 31.08.2013 NN Daily Graphic
creation – Second Lady
Stratcomm-Africa Brings Gardening To
54 02.09.2013 Bessey, B. The Chronicle
Life
2nd Garden and Flower Show opens in
55 27.06.2014 NN Daily Graphic
Accra
Ghana Garden And Flower Show Un- Ghanaian
56 02.09.2014 NN
derway In Accra Times
57 Garden show takes off in Accra 05.09.2014 NN Daily Graphic
Breathing New Life into Old Fragments
58 02.09.2015 Boadi, E. The Chronicle
of Wood
Ghana Garden and Flower Awards and
59 17.09.2015 NN Daily Graphic
Clubs launched
60 Garden And Flower Show Ends 22.09.2015 NN The Chronicle
Fourth Ghana Garden and Flower
61 27.07.2016 NN Daily Graphic
Show to promote ecotourism launched
The 4th Ghana Garden and Flower
62 08.09.2016 NN Daily Graphic
Show: Flowers are worth more than

332
 aesthetics
Garden and Flower show opens in Ac-
63 12.09.2016 NN Daily Graphic
cra
Stacomm crowns 4th Garden Show with
64 19.09.2016 NN The Chronicle
awards
65 Enforce regulations on tree cutting 10.01.2015 NN Daily Graphic
Media Tasked To Partner EPA To Save
66 16.06.2014 Anane, E. B. The Chronicle
Environment
67 EPA Marks World Environment Day 06.06.2014 Adam, F. The Chronicle
Amegede, D. Ghanaian
68 EPA creates environmental awareness 12.10.2016
K. Times
69 EPA urges Ghanaians to plant trees 08.06.2014 NN Daily Graphic
70 EPA launches Green City Project 13.06.2014 NN Daily Graphic
EPA Service Personnel Plant Trees
71 18.05.2016 NN The Chronicle
Along Awoshie-Pokuase Road
EPA service personnel partners founda-
72 18.06.2015 NN Daily Graphic
tion to plant trees
ABL, EPA Plant Trees To Save The
73 07.06.2016 Bessey, B. The Chronicle
Environment
GNASBA marks 23rd anniversary with
74 12.12.2016 NN Daily Graphic
tree planting
Philio Delio Foods to undertake tree
75 22.07.2015 Issah, Z. Daily Graphic
planting in 6 northern schools
Accra Premier Lions Club to climax
76 centenary anniversary – 29 Countries 02.05.2017 NN Daily Graphic
expected to attend
77 ABL marks World Environmental Day 06.06.2016 NN Daily Graphic
Help grow more trees in rainy season – Ghanaian
78 20.04.2015 NN
Baffour Oppong Times
Forestry ladies plant trees at Manya Jor- Ghanaian
79 28.09.2016 NN
panya Times
Horticulture can contribute to growth – Ablordeppey,
80 04.09.2013 Daily Graphic
Access Bank S.
Ghana tackles climate change at the
81 23.07.2012 Nonor, D. The Chronicle
beaches
Parks and Gardens to feature in making
82 01.10.2010 NN The Chronicle
Accra a Millennium City
Aburi Botanical Gardens: Botanically
83 19.05.2014 Bentil, N. Daily Graphic
beautiful, structurally ugly
Parks and gardens dept. cries for assis- Abdul- Ghanaian
84 16.06.2015
tance Rahaman, S. Times

333
 Yeboah-Afari,
85 Beyond the euphoria of the new era 06.01.2017 Daily Graphic
A.
86 Pockets of Indiscipline 29.04.2017 NN Daily Guide
From my Rooftop: What next after the
87 23.04.2013 Akordor, K. Daily Graphic
familiarisation tours?
Yeboah-Afari,
88 Where have all the flowers gone? 12.08.2016 Daily Graphic
A.
A Greener, Healthier, More Beautiful
89 03.08.2016 Ogyiri, A. K. The Chronicle
Ghana
AMA rolls out plans to make Accra
90 10.05.2017 NN Daily Graphic
cleanest city
Yeboah-Afari,
91 Accra must sparkle and bloom! 12.05.2017 Daily Graphic
A.
Saving the vegetative cover of an urban-
92 02.05.2017 NN Daily Graphic
ising Accra
93 AMA Cleans Accra 10.05.2017 Tarlue, M. Daily Guide
Gender, Tourism Ministries to facelift
94 22.06.2016 NN Daily Graphic
Efua Sutherland Children’s Park
Line of trees cut to give way to Giffard
95 19.09.2013 Bokpe, S. J. Daily Graphic
Road expansion works
Vandalism of trees in Accra: Becoming
96 22.12.2015 NN Daily Graphic
a norm
Dr. James Orleans-Lindsay’s green
97 28.08.2015 NN Daily Graphic
dream
YEA employs more youth; Targets
98 02.11.2016 Ennin, E. Daily Graphic
100,000 by close of year
Bruce-
99 YEA To Recruit 100,000 Youth 11.11.2016 Daily Guide
Quansah, E.
YEA launches Youth in Greening Gha-
100 30.11.2016 NN Daily Graphic
na Module
Time to halt destruction of our envi-
101 08.03.2016 NN Daily Graphic
ronment
Agyei-
102 Bats: Ghana’s best-kept wildlife secret 11.04.2013 Daily Graphic
Ohemeng, J.
103 Green Ghana project launched 13.07.2015 NN Daily Graphic
Respect Sanitation Work-
104 ers…Zoomlion MD Pleads To The 24.02.2014 NN The Chronicle
Public
Youth urged to support efforts to attain
105 06.07.2016 NN Daily Graphic
SDGs
106 UG to introduce sustainability and envi- 31.03.2017 NN Daily Graphic

334
 ronmental sanitation MA programmes
107 ‘Who we are is what we leave behind’ 16.06.2016 NN Daily Graphic
Boateng, E. Ghanaian
108 MMDAs to develop parks, gardens 16.09.2015
A. Times
109 Ghana fast losing its forest 28.05.2014 NN Daily Graphic
110 Where Are All The Flowers? 01.10.2016 NN Daily Guide
Essence of gardens to human develop- Ghanaian
111 17.08.2015 Boadi, E.
ment Times
Ghanaian
112 Climate Change And Our Survival 15.05.2015 NN
Times
Owusu- Ghanaian
113 Untitled 01.09.2016
Amoah, L. K. Times
NADMO, assembly assist Koforidua Ghanaian
114 14.09.2016 Kodjo, D.
flood victims Times

335
7 Guide for Semi-Structured Stakeholder Interviews
1. Questions about the respondent’s work/job/position/role/relation to urban trees, forests
and/or forestry and, if applicable, general information about the employer/organisation the
respondent is working for

‘Area’ Possible questions Possible follow-ups

- Can you tell me about your professional re-
sponsibilities/your job?
Respondent’s - Can you tell me about your (professional) - Relationship with other
work background/your job ‘history’? stakeholders
- Can you tell me about your organisa-
tion/employer?

2. Questions relating more directly to the urban trees, the urban forest and urban forestry

‘Area’ Possible questions Possible follow-ups

- How to achieve objec-
- Can you tell me about your current and (pos-
tives
sible) future objectives in relation to urban for-
- Time frame
estry/the urban forest?
Planning and - Monitoring
- Have these objectives changed in the past?
Management - Why named factors/
- In your opinion, what factors/events/ pro-
Objectives events/processes
cesses have influenced/are influencing the pro-
- How have they influ-
cess of planning and managing the urban
enced planning and
trees/forest in Accra?
management
- Can you tell me about the trees (species) you
- Where are the trees
are planting/researching/using/ maintain-
from
ing/selling?
Trees - Why certain species
- Can you tell me about the planting and
- Why certain operations
maintenance process/operations (‘work’ direct-
- Planting sites
ly done on/with the trees)?
- Why/how are they
problems/threats/ op-
- Can you tell me about any problems
portunities/strengths
Problems/ with/threats to the urban trees/forest you see?
- How do they impact
Threats - In your opinion, what do you consider to be
job/work
and the main benefits/services coming from the
- Relationship with e.g.
Opportuni- trees/forest?
laws/policies (forestry-
ties/ - If you could simply change one thing, what
related and general city
Strengths would it be (simply = not considering e.g. fi-
planning)
nancial, legal, political aspects)?
- Specific (geographical)
focus area(s)

336
3. Questions relating directly to tangible tree products (to be asked if ‘area’ was not brought up
by respondent before).

‘Area’ Possible questions Possible follow-ups

- why/why not?
- do you/ your organisation consider tangible
- what kind of information
products in planning and management/research
can you give me about spe-
- what is your view/ your organisation’s view
Tangible Tree cific tree products (e.g.
on tangible tree products?
Products amounts sold, percentage
- what kind of tangible tree products (from ur-
from urban trees, changes
ban trees) do you sell/buy/trade or are part of
in volumes traded, quality,
your work/job
etc.)

337
8 List of Trees in Schumacher’s Book
Tree species in Schumacher (1827): Beskrivelse af Guineiske planter: som ere fundne af Danske botanikere,
især af etatsraad Thonning

To confirm and/or update species names provided in the book, I used ‘Plants of the World
Online’ from the Royal Botanic Gardens, Kew27; JSTOR Global Plants database28 as well as ‘The
Plant List’, a collaboration between the Royal Botanic Gardens, Kew; Missouri Botanical Garden
and a number of further contributors29. Where there are doubts, the current species name pro-
vided is marked with a question mark. I included all species which might, even if rarely, present in
a tree habit and also included palms. The species are listed in the order as they appear in the book
and Figure 45 (below table) depicts the locations of some of the mentioned places.

Translations are my own. I did not translate all Danish comments but focused on location de-
scriptions and general information, while I excluded comments on plant descrip-
tions/comparisons and detailed descriptions of use.

Species
Species name
name Page Comments in book (translated)
(current accepted)
(as in book)
Codarium Grows in Accra. Flowers in Sep-
Dialium guineense Willd. 18
nitidum tember and May.
For its fast growth and shade often
Ficus
Ficus umbellata Vahl 25 planted along main roads and in
umbellata
the African towns.
Ficus lutea Ficus lutea Vahl 25 Cultivated.
Along roads between Christians-
Ficus
Ficus laurifolia Lam. 26 borg and Frederiksberg [Frederiks-
ovata
borg]
Here and there. Fruits are eaten by
Ficus calyptrata Ficus calyptrata Vahl 27
the Natives.
Ficus microcarpa Ficus thonningii Blume (?) 28 /
Nauclea africa- Mitragyna inermis (Willd.)
104 Common in the plains.
na Kuntze
Cephalina escu- Sarcocephalus latifolius (Sm.) Here and there under bushes.
105
lenta E.A.Bruce Flowers in May and June.
Ucriana race- Oxyanthus racemosus (Schu- Rare; some places in Aquapim, e.g.
107
mosa mach. & Thonn.) Keay Begusso [Berekuse]
Psychotria triflo- Cremaspora triflora (Thonn.)
108 Grows near Asiama [Afiaman]
ra K.Schum.

27 http://plantsoftheworldonline.org/ (Accessed: 15.06.2021)

28 https://plants.jstor.org/ (Accessed: 15.06.2021)
29 http://www.theplantlist.org/ (Accessed: 15.06.2021)

338
Phallaria spi- Vangueriella spinosa (Schumach.
113 Not common.
nosa & Thonn.) Verdc.
Conocarpus pu- Common near the beach and salty
Conocarpus erectus L. (?) 115
bescens lagoons.
Common in the landscapes near
Cordia guineen-
Cordia guineensis Thonn. 128 the beach; flowers in the rainy sea-
sis
son.
Common; flowers in May. The
Ehretia cymosa Ehretia cymosa Thonn. 129 longest branches are used for an-
chor chains.
Synsepalum dulcificum (Schu-
Bumelia dulcifica 130 Not common in Aquapim.
mach. & Thonn.) Daniell
Celastrus lauci- Maytenus undata (Thunb.)
132 Rare, around Adah [Ada]
folius Blakelock
Carissa dulcis Carissa spinarum L. 146 Fairly common.
Gardenia terni- Gardenia ternifolia Schumach. & In Accra and Adampi [?]; not
147
folia Thonn. common.
In the forests of Fida. According
Gardenia medic-
Gardenia thunbergia Thunb. 148 to Isert, the Natives use it for sev-
inalis
eral illnesses.
Grows in the sandy areas of La and
Asclepias Calotropis procera (Aiton) Dry-
154 Prampram. Flowers in March and
procera and.
April.
Here and there between other
Celtis guineensis Trema orientalis (L.) Blume 160
bushes.
Achras sericea Chrysophyllum albidum G.Don 179 /
Between other bushes on previous-
Ornitrophe
Allophylus africanus P.Beauv. 188 ly cultivated places and near Aqua-
tristachyos
pim.
Noltia Diospyros tricolor (Schumach. &
189 Common near the coast.
tricolor Thonn.) Hiern
Grows on the fields north of
Cupania Christiansborg and Fredensborg
Blighia sapida K.D.Koenig 190
edulis [Ningo]. Flowers in November and
fruits in January.
Clausena anisata (Willd.)
Hook.f. ex Benth. or Clausena Grows here and there, rather
Amyris anisata 191
heptaphylla (Roxb.) Wight & common.
Arn.
Ximenia ameri- Not common, a mile north of
Ximenia americana L. 193
cana Christiansborg. Flowers in June.
Dodonaea re- Common on the shores of the Vol-
Dodonaea viscosa (L.) Jacq. 194
panda ta.
Here and there between the bushes
on the estuary of the Volta, also in
Sophora tomentosa subsp. littoralis
Sophora nitens 201 Poisi [?]. It grows in loose sandy
(Schrad.) Yakovlev
soils and flowers in October and
May.

339
 Most common in the valleys of
Podalyria hema-
Baphia nitida Lodd. 202 Aquapim, even though it is also
toxylon
not common there.
Bauhinia thon-
Bauhinia thonningii Schum. 203 In Aquapim.
ningii
Flowers the whole year. Cultivated
Caesalpinia pul-
Caesalpinia pulcherrima (L.) Sw. 209 in the Danish establishments for its
cherrima
beautiful flowers.
Adenanthera te- Tetrapleura tetraptera (Schum. & Grows in the fertile regions of Aq-
213
traptera Thonn.) Taub. uapim, flowers in May.
Melia angustifo-
Melia azedarach L. 214 Cultivated.
lia
Grows on the fertile mountain
Gomphia flava Schumach. &
Gomphia flava 216 fields and flowers in different sea-
Thonn.
sons.
Limonia mona- Trichilia monadelpha (Thonn.)
217 Grows in Aquapim.
delpha J.J.de Wilde
Erytroxylon Erythroxylum emarginatum
224 /
emarginatus Thonn.
Spondias au- Here and there on fields, not so
Spondias mombin L. 225
rantiaca common.
Psidium longifo-
Psidium guajava L. (?) 229 /
lium
Eugenia corona-
Eugenia coronata Vahl ex DC. 230 Common at the shores.
ta
Lundia mono-
Oncoba monacantha Steud. (?) 231 Rare.
cantha
Chrysobalanus
Chrysobalanus icaco L. 232 /
orbicularis
Capparis tomen-
Capparis tomentosa Lam. 234 Rare, found at Ningo.
tosa
Capparis eryth-
Capparis erythrocarpos Isert 235 Here and there at the shores.
rocarpos
Crataeva guin-
Crateva adansonii DC. 240 Along the Volta, flowers in May.
eensis
Here and there in Aquapim. The
Sterculia verticil- Cola verticillata (Thonn.) Stapf
240 fruit is eaten by the Natives; it has
lata ex A.Chev.
a bitter taste and dyes crimson red.
Deinbollia pin- Deinbollia pinnata (Poir.)
242 Here and there.
nata Schumach. & Thonn.
Smeathmannia pubescens R. Br.
Bülowia illustris 246 /
(?)
Euphorbia
Euphorbia drupifera Thonn. 250 /
drupifera
Uvaria cylindri-
Uvaria chamae P.Beauv. 256 Here and there.
ca
Annona arenar- Grows at Quita [Keta], often be-
Annona senegalensis Pers. 257
ia tween other bushes in loose sandy

340
 ground. Flowers in September, Oc-
tober, November and fruits in De-
cember, January and February.
Annona glauca Schumach. &
Annona glauca 259 Near Ursua [?]; flowers in May.
Thonn.
In the mountains of Aquapim and
Bignonia tuli-
Spathodea campanulata P.Beauv. 273 in Frederiksstad. Flowers in differ-
pifera
ent seasons.
Bignonia glandu- Newbouldia laevis (P.Beauv.)
274 Cultivated here and there.
losa Seem.
Once found in Frederiksberg
Premna quadri- Premna quadrifolia Schumach.
275 [Frederiksborg]. Flowers in Janu-
folia & Thonn.
ary.
Vitex ferruginea Schumach. &
Vitex ferruginea 288 In Aquapim.
Thonn.
Vitex cuneata Vitex doniana Sweet 289 Everywhere.
Grows at the small salty reefs close
to the beach but rarely above 6-8
alen height. At the Volta river and
Avicennia afri- close by areas where the shore side
Avicennia germinans (L.) L. 290
cana wind is not as strong, it grows as
big as the largest oak. Flowers pri-
marily in the rainy seasons but also
at other times.
Adansonia digi- At Quita [Keta], Tubreku [Tog-
Adansonia digitata L. 300
tata bloku], Aquapim.
Bombax pen-
Ceiba pentandra (L.) Gaertn. 301 /
tandrum
Common at the shores of the Vol-
Hibiscus tilliceus Hibiscus tilliaceus L. 313
ta.
Here and there between Asiama
Mimosa glaber- Albizia glaberrima (Schum. & [Afiaman] and Jadofa [Oyarifa].
321
rima Thonn.) Benth. Flowers in April. The wood is used
as firewood.
Mimosa adi- Albizia adianthifolia (Schum.) Grows in Bligusso [Berekuso] and
322
anthifolia W.Wight flowers in April.
Mimosa guineen- Zapoteca portoricensis (Jacq.) Grows in the shrubbery and near
323
sis H.M.Hern. Aquapim.
Mimosa pro- Some bushes of this plant were
Mimosa pigra L. 324
cumbens found near Christiansborg.
Mimosa ad- Acacia nilotica subsp. adstringens
327 In Ningo.
stringens (Schum. & Thonn.) Roberty
Pterocarpus escu- Grows widespread at the shores of
Pterocarpus santalinoides DC. 330
lentus the Volta.
Sommerfeldtia Machaerium lunatum (L.f.) Everywhere at the shores of the
331
obovata Ducke Volta.
Ecastaphyllum Dalbergia ecastaphyllum (L.) Everywhere at the shores of the
332
brownei Taub. Volta.

341
Erythrina latifo- Rarely found wild but cultivated by
Erythrina senegalensis DC. 333
lia the Natives.
Grows here and there on the field,
Robinia thon- Millettia thonningii (Schumach.
349 most commonly, however, near the
ningii & Thonn.) Baker
Volta.
Robinia multi- Millettia irvinei Hutch. & Dal-
350 Not very common. Grows in June.
flora ziel
Robinia argenti- Lonchocarpus sericeus (Poir.)
352 In Ada.
flora Kunth ex
Citrus panicula- Afraegle paniculata (Schumach.
378 /
ta & Thonn.) Engl.
Chrysocoma Vernonia colorata subsp. Colorata
383 /
amara (Willd.) Drake
Acalypha den- Mallotus oppositifolius (Geiseler)
410 Here and there.
tata Müll.Arg.
Jatropha
Jatropha curcas L. 412 Here and there.
curcas
Janipha manihot Manihot esculenta Crantz 414 Cultivated.
Phylanthus an- Flueggea virosa (Roxb. ex Willd.) Common in the plains close to the
415
gulatus Royle sea.
Zanthoxylum Zanthoxylum zanthoxyloides
433 Common.
polygamum (Lam.) Zepern. & Timler (?)
At the Volta, in the Aquapim
Phoenix
Phoenix reclinata Jacq. 437 mountains and in some other plac-
spinosa
es.
Elais
Elaeis guineensis Jacq. 439 Grows wild and cultivated.
guineensis
Borassus flabelli-
Borassus flabellifer L. 443 Here and there.
formis
Hyphaene guin- Hyphaene guineensis Schumach.
445 /
eensis & Thonn.
Ferreola guineen-
Diospyros vera (Lour.) A.Chev. 448 In Quita [Keta].
sis
Schousboea cor- Alchornea cordifolia (Schumach.
449 Here and there.
difolia & Thonn.) Müll.Arg.
Flacourtia edulis Flacourtia flavescens Willd. 450 Here and there.

342
 Figure 45 Location of places mentioned in Schumacher (1827)
Map tiles by Stamen Design, under CC BY 3.0. Data by OpenStreetMap, under ODbL

343
9 Neighbourhood Land Cover Values with Standard Errors
Non- Non-
Canopy Canopy Plantable Plantable
Area plantable plantable
(%) SE (%) SE
(%) SE
Accra 14.0 1.55 40.40 2.19 45.60 2.23
Abeka 5.00 0.97 24.20 1.92 70.80 2.03
Abelenkpe 13.40 1.52 36.80 2.16 49.80 2.24
Abofu 16.60 1.66 34.00 2.12 49.40 2.24
Abossey Okai 3.80 0.86 22.20 1.86 74.00 1.96
Accra Central 6.60 1.11 31.40 2.08 62.00 2.17
Achimota College 53.60 2.23 32.60 2.10 13.80 1.54
Adabraka 9.20 1.29 32.60 2.10 58.20 2.21
Airport Residen-
30.40 2.06 36.00 2.15 33.60 2.11
tial
Airport West Res. 22.40 1.86 39.40 2.19 38.20 2.17
Akweteyman 11.80 1.44 31.60 2.08 56.60 2.22
Alajo 6.20 1.08 33.40 2.11 60.40 2.19
Alogboshie 7.80 1.20 44.60 2.22 47.60 2.23
Asylum Down 12.60 1.48 24.40 1.92 63.00 2.16
Avenor 3.60 0.83 34.80 2.13 61.60 2.18
Awudome 26.20 1.97 37.00 2.16 36.80 2.16
Bubuashie 12.60 1.48 36.60 2.15 50.80 2.24
Cantonments 27.60 2.00 38.80 2.18 33.60 2.11
Chorkor 0.20 0.20 18.60 1.74 81.20 1.75
Dansoman 17.00 1.68 29.80 2.05 53.20 2.23
Darkuman 7.40 1.17 26.60 1.98 66.00 2.12
Dzorwulu 15.00 1.60 41.40 2.20 43.60 2.22
East Legon 14.40 1.57 40.80 2.20 44.80 2.22
East Legon Ext. 20.00 1.79 49.80 2.24 30.20 2.05
Gbegbeyise 2.00 0.68 29.20 2.03 68.80 2.07
James Town 1.20 0.49 41.20 2.20 37.60 2.21
Kanda 21.40 1.83 41.40 2.20 37.20 2.16
Kaneshie 5.00 0.97 33.20 2.11 61.80 2.17
Kisseman 16.20 1.65 37.20 2.16 46.60 2.23
Kokomlemle 3.80 0.86 27.80 2.00 68.40 2.08
Korle Bu 22.40 1.86 53.60 2.23 24.00 1.91
Korle Dudor 2.80 0.74 26.00 1.96 71.20 2.03
Korle Gonno 5.20 0.99 37.00 2.16 57.80 2.21
Kotobabi 4.00 0.88 31.00 2.07 65.00 2.13
Kpehe 4.00 0.88 30.80 2.06 65.20 2.13
Kwashiman 5.60 1.03 35.20 2.14 59.20 2.20

344
La 8.80 1.27 31.60 2.08 59.60 2.19
La Dadekotopon 10.40 1.37 67.20 2.10 22.40 1.86
Labone 19.40 1.77 39.60 2.19 41.00 2.20
Lapaz 5.80 1.05 35.00 2.13 59.20 2.20
Lartebiokorshie 13.00 1.50 33.20 2.11 53.80 2.23
Legon 33.60 2.11 50.80 2.24 15.60 1.62
Mamobi 4.40 0.92 18.60 1.74 77.00 1.88
Mamponse 5.40 1.01 32.20 2.09 62.40 2.17
Mamprobi 8.60 1.25 30.40 2.06 61.00 2.18
Mateheko 7.60 1.19 28.80 2.03 63.60 2.15
Ministerial Area 15.20 1.61 49.00 2.24 35.80 2.14
Mpoase 3.40 0.81 33.80 2.12 62.80 2.16
New Fadama 8.40 1.24 33.80 2.12 57.80 2.21
New Town 3.60 0.83 31.40 2.08 65.00 2.13
Nima 4.40 0.92 17.60 1.70 78.00 1.85
North Alajo 5.80 1.05 38.40 2.18 55.80 2.22
North Dzorwulu 21.00 1.82 40.80 2.20 38.20 2.17
North Industrial 7.80 1.20 31.60 2.08 60.60 2.19
North Kaneshie 15.60 1.62 35.20 2.14 49.20 2.24
North Odorkor 12.40 1.47 37.60 2.17 50.00 2.24
North Ridge 28.40 2.02 35.40 2.14 36.20 2.15
Nyaniba Estate 12.60 1.48 32.00 2.09 55.40 2.22
Old Mamprobi 10.60 1.38 34.20 2.12 55.20 2.22
Old Tesano 11.00 1.40 36.40 2.15 52.60 2.23
Osu 10.20 1.35 36.80 2.16 53.00 2.23
Ringway Estate 30.20 2.05 33.20 2.11 36.60 2.15
Roman Ridge 24.80 1.93 46.40 2.23 28.80 2.03
Russia 6.00 1.06 28.40 2.02 65.60 2.12
Sabon Zongo 5.20 0.99 28.00 2.01 66.80 2.11
Sempe 9.80 1.33 29.00 2.03 61.20 2.18
South Industrial 5.60 1.03 36.40 2.15 58.00 2.21
South Labadi 6.20 1.08 51.80 2.23 42.00 2.21
South Legon 22.00 1.85 57.60 2.21 20.40 1.80
South Odorkor 10.80 1.39 39.20 2.18 50.00 2.24
Sukura 8.60 1.25 37.80 2.17 53.60 2.23
Tesano 21.80 1.85 44.60 2.22 33.60 2.11
Tudu 6.00 1.06 31.80 2.08 62.20 2.17
Ussher Town 2.60 0.71 10.40 1.37 87.00 1.50
West Abossey
9.40 1.31 37.60 2.17 53.00 2.23
Okai
West Ridge 28.60 2.02 40.20 2.19 31.20 2.07
Zoti 17.40 1.70 43.40 2.22 39.20 2.18

345
10 Terms for Interactions between Stakeholders
Interactions
to (not) comply with something to (not) be involved
to (not) have to comply with something to (not) involve someone
to (not) take part in something to (not) need to involve someone
to (not) be part of something to (not) need to get involved
to (not) be supposed to be part of something to (not) warn someone
to (not) work in partnership with someone to (not) make someone aware of something
to (not) partner with someone to (not) interact with someone
to (not) compete with someone to (not) be there for someone
to (not) impose something on someone (e.g.
to (not) do work for someone
charges)
to (not) encourage someone to do something to (not) meet
to (not) talk to someone about something to (not) be able to meet
to (not) direct/refer someone to
to (not) connect with someone
someone/somewhere else
to (not) get interested to (not) motivate someone to do something
to (not) take (legal) actions against
to (not) be interested
someone/something
to (not) make someone interested to (not) approve something for someone else
to (not) feel well-informed by someone to (not) present something to someone
to (not) give/bring something to someone (e.g.
to (not) be well-informed
copies, opportunities, ideas)
to (not) make information available to (not) ask someone to do/give something
to (not) inform someone to (not) need approval from someone
to (not) be informed to (not) be present/active
to (not) supply something to someone to (not) consider who is present/active
to (not) sensitize someone about/for
to (not) acquire something from someone
something
to (not) make an effort to sensitize someone
to (not) have support from someone/anyone
about/for something
to (not) support someone/something to (not) stand up against something/someone
to (not) be able to get support to (not) demonstrate something to someone
to (not) have to seek support to do
to (not) drive someone away/out
something
to (not) force someone to do something to (not) insist that someone does something
to (not) attack someone to (not) ensure that someone does something
to (not) be able to regulate to (not) try that someone does something
to (not) regulate something to (not) expect someone to do something
to (not) make someone do something (e.g.
to (not) collaborate with someone
compensate, pledge, plant)
to (not) cooperate with someone to (not) do something about something
to (not) consult someone to (not) promote something

346
to (not) report to someone to (not) be sponsored by someone
to (not) report someone to (not) get to know things
to (not) share your
to (not) be able to control someone
experience/knowledge/plans with someone
to (not) control someone to (not) take over something
to (not) agree with someone/something to (not) resist to listening to someone else
to (not) assist someone to (not) see value in someone else’s work
to (not) relocate someone to another
to (not) work together
stakeholder's office
to (not) come together to do something to (not) buy into someone else’s plans/ideas
to (not) channel your ideas into someone else’s
to (not) coordinate something
work
to (not) give advice to someone to (not) put pressure onto someone
to (not) trust someone to (not) being pushed by someone
to (not) contract someone to (not) organise something
to (not) accept something to (not) explain something to someone
to (not) supervise someone/something to (not) challenge someone to do something
to (not) be encouraged by someone’s
to (not) guide someone
behaviour/actions
to (not) give conditions on how to do
to (not) stop someone
something
to (not) be able to stop someone/something to (not) merge with someone
to (not) invite someone to do something to (not) correct the attitude of someone
to (not) task someone to do something to (not) conduct a verification
to (not) work closely with someone to (not) do something behind someone’s back
to (not) work closer with someone to (not) assure someone to get something
to (not) finance someone/something to (not) be limited by someone else
to (not) take any side to (not) have your work undone
to (not) receive something/anything (e.g. to (not) require someone to do/have
feedback, training, supplies) something
to (not) get attention to (not) join someone/something
to (not) draw someone's attention to to (not) obtain knowledge from
something someone/somewhere
to (not) interfere to (not) let someone know something
to (not) help someone do something to (not) allow someone to do something
to (not) see how to help someone to (not) be allowed to do something
to (not) restrict someone to do something to (not) do your own thing
to (not) educate someone to (not) tell someone to do something
to (not) need to educate someone to (not) tell someone something
to (not) take care that someone does
to (not) train someone to do something
something
to (not) organise something for someone
to (not) engage
(e.g. training)

347
11 Information on All Household Trees
Trees (T)
HH and Sap- Species Comments
lings (S)
About 80 years old30, branches pruned to
1 T1 Mango about 1,80 m height, DBH 60 cm, 6-7 m
high, crown spread 12 m, holds washing line
Planted about 12 years ago, does not give
1 T2 Avocado fruits, 9-10m high, 2 stems, DBH 8 cm and
9.5 cm, crown spread 6 m,
Planted about 10 years ago, 9-10 m high,
1 T3 Newbouldia laevis
DBH 13 cm
Planted about 10 years ago, 9-10 m high,
1 T4 Newbouldia laevis
DBH 24 cm
Planted about 10 years ago, 9-10 m high,
1 T5 Newbouldia laevis
DBH 8 cm
Planted about 2 years ago, 7 m high, 2 stems
1 T6 Newbouldia laevis with 8 cm and 10 cm DBH, crown spread 3
m
Planted about 2 years ago, 7 m high, DBH 13
1 T7 Newbouldia laevis
cm, crown spread 3 m
Umbrella tree (Terminalia Planted in 1979, DBH 35 cm, 12 m high,
1 T8
catappa) crown spread 11 m
Mango, 4x Soursop, Diali-
1 1S-7S Planted less than a year ago, 0.8-1.2 m high
um guineense, Unknown
12 m high, DBH 40 cm, crown spread 4.5 m,
2 T1 Coconut
holds washing lines
7 m high, DBH 20 cm, crown spread 4 m,
2 T2 Coconut
holds washing lines
2 T3 Coconut 14 m high, DBH 30 cm, crown spread 5 m
7.5 m high, DBH 22 cm, crown spread 5 m,
2 T4 Coconut
shaded by T3 and T5 and T30
2 T5 Coconut 14 m high, DBH 25 cm, crown spread 4.5 m
2 T6 Coconut 7.5 m high, DBH 20 cm, crown spread 4 m
DBH 20 cm, crown spread 5.5 m, coconut
2 T7 Coconut harvester cut some coconuts off this palm
and pruned some fronds
6 m high, DBH 20 cm, crown spread 4 m,
2 T8 Coconut
largely shadowed by surrounding palms
2 T9 Coconut 13 m high, DBH 30 cm, crown spread 5 m
2 T10 Coconut 13 m high, DBH 30 cm, crown spread 5 m
13 m high, DBH 20 cm, crown spread 4 m,
2 T11 Coconut one side of stem black, possibly burnt as lo-
cated next to the HH’s garbage burning place

30 Approximate age at time of data collection in 2017.

348
2 T12 Coconut 5.5 m high, DBH 15 cm, crown spread 4 m
2 T13 Coconut 5 m high, DBH 20 cm, crown spread 5 m
2 T14 Coconut 12 m high, DBH 25 cm, crown spread 5 m
2 T15 Coconut 12 m high, DBH 30 cm, crown spread 4 m
2 T16 Coconut 12 m high, DBH 20 cm, crown spread 5 m
2 T17 Coconut 12 m high, DBH 20 cm, crown spread 4.5 m
2 T18 Coconut 13 m high, DBH 20 cm, crown spread 4 m
2 T19 Coconut 12 m high, DBH 20 cm, crown spread 5 m
2 T20 Coconut 13 m high, DBH 25 cm, crown spread 4.5 m
2 T21 Coconut 13 m high, DBH 25 cm, crown spread 4.5 m
2 T22 Coconut 13 m high, DBH 25 cm, crown spread 4.5 m
2 T23 Coconut 13 m high, DBH 25 cm, crown spread 4.5 m
2 T24 Coconut 6 m high, DBH 20 cm, crown spread 2 m
2 T25 Oil Palm 5 m high
2 T26 Oil Palm 2.5 m high, crown spread 3 m
2 T27 Oil Palm 10 m high, DBH 40 cm, crown spread 5.5 m
2 T28 Papaya 6 m high, 12 cm DBH
4 m high, only one quarter of crown with
leaves, stem almost black in some places, lo-
2 T29 Neem cated close to garbage burning place, crown
spread 3 m in one direction and 1 m in the
other
9 m high, 2 stems with 18 cm and 15 cm
2 T30 Newbouldia laevis
DBH, crown spread 3 m
2 T31 Newbouldia laevis 5 m high, DBH 10 cm
Between 2 and 7 m high, between 5 and 10
2 T32-T41 Newbouldia laevis
cm DBH
2 T42 Newbouldia laevis 7 m high, DBH 10 cm, crown spread 2 m
2 T43 Newbouldia laevis 7 m high, DBH 25 cm, crown spread 2 m
2 T44 Hippocratea sp. 5 m high, DBH 15 cm, crown spread 2 m
2 T45 Hippocratea sp. 4 m high, DBH 10 cm, crown spread 1.5 m
7.5 m high, 2 stems with DBH 20 cm and 15
2 T46 Ficus capensis
cm, crown spread 7 m
Between 60 cm and 2 m high, for sale, several
2 S1-S10 Coconut with broken and dried leaves, completely
shaded by other plants
3 T1 Avocado 7 m high, DBH 30 cm, crown spread 8-9 m
Cut off at 1.5 m height, resprouting, DBH 10
3 T2 Moringa
cm
Cut off at 1.5 m height, resprouting, DBH 10
3 T3 Moringa
cm
4 T1 Coconut 15 m high, DBH 40 cm, crown spread 6 m
4 T2 Coconut 14 m high, DBH 40 cm, crown spread 4 m
4 T3 Oil palm 12 m high, DBH 40 cm, crown spread 7 m
13 m high, DBH 20 cm, crown spread 2.5 m,
4 T4 Coconut
no coconuts
4 T5 Oil palm 12.5 m high, crown spread 8 m

349
 Between 2 and 3.5 m high, between 3 and 10
cm DBH, between 1 and 3 m crown spread,
4 T6-T13 Moringa
apart from T6 and T13 overshadowed by T5,
T6 and T13 noticeably bigger than T7 to T12
4 T14 Coconut 14 m high, DBH 35 cm, crown spread 4 m
4 T15 Coconut 15 m high, DBH 30 cm, crown spread 4 m
13 m high, DBH 30 cm, bottom part of stem
black, possibly from burning as located close
4 T16 Coconut to the HH’s garbage burning area, fronds ap-
pear short and ‘broken’, fronds are yellow
and asymmetrical, crown spread 3 m
4 T17 Coconut 14 m high, DBH 25 cm, crown spread 4 m
About 5 m high, 2 stems with DBH 15 cm
4 T18 Annona squamosa
and 18 cm, crown spread 5-6 m
Dying/almost dead, 11 m high, DBH 70 cm,
4 T19 Mango
crown spread 12 m (almost no leaves left)
4 T20 Coconut 10 m high, DBH 20 cm, crown spread 4 m
4 S1 Lemon 60 cm high, ‘fenced’ with some stones
4 S2 Lemon 25 cm high, ‘fenced’ with some stones
4 S3 Lemon 15 cm high, ‘fenced’ with some stones

350