URBAN WETLAND ECOLOGY AND FLOODS IN KUMASI, GHANA

Benjamin Betey Campion

Institut für Geographie an der Universität Bremen

May, 2012

ii

URBAN WETLAND ECOLOGY AND FLOODS IN KUMASI, GHANA

Benjamin Betey Campion

Institut für Geographie an der Universität Bremen

May, 2012

Submitted for the degree of Doctor of Natural Sciences (Dr. rer. nat.) Supervisors: 1. Prof. Dr. Jörg-Friedhelm Venzke

2. Prof. Dr. Michael Flitner

iii

DECLARATION

I, Benjamin Betey Campion, hereby declare that this thesis has been independently written by me and is original.

Where information has been derived from other sources, I confirm that this has been indicated in the thesis.

Bremen, …27. 09.2012.…………………….…….. …………. Date Benjamin Betey Campion

iv

ACKNOWLEDGEMENT My soul gives thanks to the Lord my God for taking me this far. May the Name of God be blessed forever!

I am very grateful to Prof. Dr. Jörg-Friedhelm Venzke for the bold initiative he took and accepted me as his student. Finding such a father figure from the internet is a million dollar lottery ticket. Prof. Venzke, your friendship, encouragement, sense of humour, humility and simplicity have been my inspiration and support in writing this thesis. I love your open door approach to supervision. I am also very grateful to Prof. Dr. Michael Flitner for accepting to be the second supervisor. Thank you for the invaluable suggestions and criticisms in shaping me into an up-and-coming scientist.

I also received tremendous moral and scientific support from Bruni Hans, Birte Habel, PD. Dr. Karin Steinecke, Dr. Steffen Schwantz and Dr. Marco Langer. I love this small family that nurtured me. You will always be cherished for everything you did to make my stay in Uni-Bremen, UFT, Abeitsgruppe Klima- und Vegetationsgeographie, and above all, Room 0060 a wonderful one. I am also grateful and indebted to my colleagues in the Faculty of Renewable Natural Resources, KNUST, for their support. Prof. Steve Amisah, thank you for advising me to get a PhD.

For the second time, the Katholischer Akademischer Ausländer-Dienst (KAAD) deemed me fit for their highly competitive scholarship. I am highly indebted to you and appreciate your contribution in shaping my life this far. Marko, Simone and the many others at the KAAD office, thank you and God will richly bless you for the trust and confidence you had in me. I look forward to a future of sharing the task of developing humanity with you. Thank you.

To my inspirational friends: Dr. Callistus Mahama, Dr. Nana Ato Arthur, Seth Mensah Abobi, Dr. Julius Fobil, Cephas Kwesi Zuh, Isaac Sackey, Kwabena Asubonteng, and Dr. David Yegbey, thank you.

I will never forget the continuous and diverse support of the Campions: Kofi, Jatuat, Teni, Jakper, Yaw, Yenukwah, Victor, Dumbian, Lardi, Tisob, Sanjaung, Kibirsiar and Cynthia. You are the pillar behind this work. Thank you and let us uphold the family spirit. I could not have made it without a partner. A friend for life, my hope in times of despair, my all: Elma.

“You placed gold on my finger You brought love like I've never known You gave life to our children And to me, a reason to go on. …… But most of all, you're my best friend” (Don Williams).

Palmaak and Iloam, you are my motivation and reason for studying. Thank you for always accepting me after the long periods away from home.

v

DEDICATION To my dad who said, ‘Ben, go to school’; and to my mum who made it all possible.

John Limuub Campion & Mary Konjit Lambon

vi

LIST OF ACRONYMS As Arsenic Cd Cadmium CF Contamination Factor Cr Chromium Cu Copper DEMC District Environmental Management Committee DO Dissolved Oxygen EF Enrichment Factor EPA Environmental Protection Agency FRNR Faculty of Renewable Natural Resources Hg Mercury HSD Hydrological Services Department IFRC International Federation of Red Cross and Red Crescent Societies Igeo Geo-accumulation Index IPCC Intergovernmental Panel on Climate Change ITCZ Intertropical Convergence Zone KMA Kumasi Metropolitan Assembly KNUST Kwame Nkrumah University of Science and Technology MDGs Millennium Development Goals MDS non-Metric Multidimensional Scaling NADMO National Disaster Management Organisation NEPAD New Partnership for African Development Ni Nickel P Phosphorus Pb Lead PCA Principal Components Analysis PLI Pollution Load Index PRS Poverty Reduction Strategy TCPD Town and Country Planning Department TDS Total Dissolved Solids Zn Zinc

vii

TABLE OF CONTENTS DECLARATION ...................................................................................................................iii

ACKNOWLEDGEMENT .................................................................................................... iv

DEDICATION ....................................................................................................................... v

LIST OF ACRONYMS ........................................................................................................ vi

TABLE OF CONTENTS......................................................................................................vii

LIST OF FIGURES .............................................................................................................. xi

LIST OF TABLES ...............................................................................................................xiv

LIST OF PHOTOGRAPHS ................................................................................................. xv

APPENDICES .....................................................................................................................xvi

LIST OF PUBLICATIONS FROM THIS THESIS ........................................................... xvii

ABSTRACT....................................................................................................................... xviii

ZUSAMMENFASSUNG ....................................................................................................xix

1 RESEARCH THEME AND THEORETICAL FRAMEWORK ....................................... 1

1.1 INTRODUCTION....................................................................................................... 1

1.2 RESEARCH BACKGROUND AND JUSTIFICATION ........................................... 2

1.2.1 Urban Environmental Issues ................................................................................ 2

1.2.2 Research Justification .......................................................................................... 4

1.3 RESEARCH GOAL .................................................................................................... 7

1.4 RESEARCH OBJECTIVES ....................................................................................... 7

1.5 RESEARCH QUESTIONS......................................................................................... 7

1.6 CONCEPTUAL AND THEORETICAL FRAMEWORKS ....................................... 8

1.7 HYPOTHESES ......................................................................................................... 12

2 WETLANDS, URBANISATION AND FLOOD NEXUS ............................................. 14

2.1 INTRODUCTION..................................................................................................... 14

2.2 WETLANDS ............................................................................................................. 14

2.2.1 Morphology and Hydrodynamics of Wetlands .................................................. 17

viii

2.2.2 Distribution, Values and Degradation of Wetlands ........................................... 18

2.3 URBANISATION ..................................................................................................... 21

2.3.1 Concept and History of Urbanisation ................................................................ 21

2.3.2 Modern Trends in Urbanisation ......................................................................... 23

2.3.3 Urbanisation Trends in sub-Saharan Africa....................................................... 25

2.4 URBANISATION AND THE ENVIRONMENT .................................................... 26

2.5 WETLANDS AND URBANISATION .................................................................... 28

2.5.1 Urban Wetlands Vegetation ............................................................................... 31

2.5.2 Wetlands, Urbanisation and Agriculture............................................................ 32

2.5.3 Wetlands, Urbanisation and Floods ................................................................... 35

2.6 URBAN POVERTY AND WETLANDS................................................................. 37

2.6.1 Urban Poverty in Africa ..................................................................................... 37

2.6.2 Urban Poverty and the Environmental Kuznets Curve ...................................... 38

2.7 CLIMATE CHANGE, URBANISATION AND WETLANDS............................... 40

2.8 REMOTE SENSING AND URBAN LANDSCAPE PLANNING AND MANAGEMENT................................................................................................................. 42

3 POLITICAL ECOLOGY OF WETLANDS IN KUMASI.............................................. 45

3.1 INTRODUCTION..................................................................................................... 45

3.2 METHODOLOGY .................................................................................................... 46

3.2.1 Physical and Socio-economic Survey of Wetland Communities of Kumasi..... 46

3.2.2 Data Analysis ..................................................................................................... 48

3.3 RESULTS.................................................................................................................. 48

3.3.1 The Physical Characteristics of Kumasi ............................................................ 48

3.3.2 Socio-economic Characteristics of Wetland Communities in Kumasi .............. 52

3.3.3 Community Efforts at Wetland Management in Kumasi .................................. 60

3.4 DISCUSSION ........................................................................................................... 63

4 FLOODS AND ENVIRONMENTAL MANAGEMENT IN KUMASI, GHANA ........ 67

ix

4.1 INTRODUCTION..................................................................................................... 67

4.2 ENVIRONMENTAL MANAGEMENT AND THE NATIONAL ENVIRONMENTAL ACTION PLAN ............................................................................... 69

4.2.1 The Environmental Protection Agency.............................................................. 69

4.2.2 Water Resources Commission ........................................................................... 70

4.2.3 Lands Commission............................................................................................. 72

4.2.4 Local Government and Environmental Management ........................................ 73

4.3 WETLANDS, FLOOD PREVENTION AND MANAGEMENT ............................ 74

4.3.1 Stakeholders’ Capacity for Managing Wetlands and Floods ............................. 74

4.4 DISCUSSION ........................................................................................................... 79

5 WETLAND ECOLOGY AND VEGETATION GEOGRAPHY IN THE KUMASI METROPOLIS ........................................................................................................................ 85

5.1 INTRODUCTION..................................................................................................... 85

5.2 METHODOLOGY .................................................................................................... 86

5.2.1 Reconnaissance Survey of Wetlands in Kumasi and Selection of Study Sites .. 86

5.2.2 Ecological Survey of Wetlands in Kumasi ........................................................ 87

5.2.3 Data Analysis ..................................................................................................... 91

5.3 RESULTS.................................................................................................................. 91

5.3.1 Anthropogenic Activities in the Wetland Areas of Kumasi .............................. 91

5.3.2 Wetland Edaphic Characteristics, Heavy Metals and Stream Physicochemical Characteristics .................................................................................................................. 94

5.3.3 Wetland Vegetation of Kumasi........................................................................ 107

5.4 DISCUSSION ......................................................................................................... 123

6 CLIMATE, LANDUSE CHANGES AND FLOODING IN THE KUMASI METROPOLIS ...................................................................................................................... 128

6.1 INTRODUCTION................................................................................................... 128

6.2 METHODOLOGY .................................................................................................. 129

6.2.1 Rainfall and Flood Data Collection and Analysis............................................ 129

x

6.2.2 Assessment of Urbanisation of Kumasi ........................................................... 129

6.3 RESULTS................................................................................................................ 130

6.3.1 Rainfall Patterns and Trends in Kumasi .......................................................... 130

6.3.2 Floods in Kumasi ............................................................................................. 139

6.3.3 Landcover Change in Kumasi.......................................................................... 146

6.4 DISCUSSION ......................................................................................................... 150

7 GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS ........... 154

7.1 GENERAL DISCUSSION...................................................................................... 154

7.2 CONCLUSIONS ..................................................................................................... 157

7.3 RECOMMENDATIONS ........................................................................................ 159

8 REFERENCES .............................................................................................................. 162

9 APPENDICES ............................................................................................................... 179

xi

LIST OF FIGURES Figure 1.1. Conceptual framework for the investigation of urbanisation, socio-economic activities and incidence of floods in Kumasi ........................................................................... 10

Figure 1.2. A composite model of the effects of human interference on a wetland ecosystem in an urban setting .................................................................................................................... 11

Figure 2.1. An illustration of the continuum between the terrestrial, wetland and aquatic systems ..................................................................................................................................... 15

Figure 2.2. Distribution of wetlands in the world .................................................................... 20

Figure 2.3. Major vegetable farming areas in the Kumasi Metropolis .................................... 35

Figure 3.1. Map of Ghana with an insert showing limits of the Kumasi Metropolitan Assembly and the study sites ................................................................................................... 49

Figure 3.2. Population growth trajectory of the city of Kumasi .............................................. 50

Figure 3.3. The occupation of respondents from flood prone suburbs of Kumasi .................. 54

Figure 3.4. Residential types and homeownership of respondents in the different suburbs of Kumasi ..................................................................................................................................... 55

Figure 3.5. Respondents’ sources of land for construction of houses in the different suburbs of Kumasi ................................................................................................................................. 57

Figure 3.6. Reasons for choice of wetland area for construction of houses in the flood prone suburbs of Kumasi ................................................................................................................... 58

Figure 3.7. Factors or instruments that will motivate residents of the various flood prone communities to relocate ........................................................................................................... 59

Figure 3.8. The predominant uses of wetlands in the various suburbs of Kumasi .................. 61

Figure 3.9. Perception of wetland protection in Kumasi ......................................................... 61

Figure 3.10. Reasons for the continual destruction of wetlands in the various suburbs of Kumasi ..................................................................................................................................... 62

Figure 3.11. A proposed model to describe the decision to move or stay in the flood prone suburbs of Kumasi ................................................................................................................... 65

Figure 4.1. Assessment of stakeholders’ responsibility for the management of floods and wetlands in Kumasi .................................................................................................................. 74

Figure 4.2. Spatial distribution of houses marked for demolition and attitude towards demolition of houses in waterways by the Kumasi Metropolitan Assembly........................... 77

Figure 4.3. A conceptual model for management of floods and wetlands within the Kumasi Metropolis ............................................................................................................................... 83

Figure 5.1. A schematic diagram of a sample study site along a stream X showing the transects and quadrat layout ..................................................................................................... 87

xii

Figure 5.2. Landcover and Landuse activities within 100 m from the stream channel at the different study sites in Kumasi ................................................................................................ 93

Figure 5.3. Soil map of the Kumasi showing study sites . ....................................................... 95

Figure 5.4. A cluster diagram of the correlation matrix of the various environmental variables measured in the different suburbs of Kumasi ........................................................................ 100

Figure 5.5. A principal components analysis plot of the measured environmental parameters ................................................................................................................................................ 101

Figure 5.6. Heavy metals enrichment in the stream sediments of the different suburbs of Kumasi in the rainy season .................................................................................................... 102

Figure 5.7. Heavy metals enrichment in the stream sediments of the different suburbs of Kumasi in the dry season ....................................................................................................... 103

Figure 5.8. Heavy metals accumulation in the stream sediments of the different suburbs of Kumasi in the rainy season .................................................................................................... 104

Figure 5.9. Heavy metals accumulation in the stream sediments of the different suburbs of Kumasi in the dry season ....................................................................................................... 105

Figure 5.10. Seasonal differences in pollution by metals in the stream sediments of the different suburbs of Kumasi................................................................................................... 106

Figure 5.11. Number of species identified at the various wetland sites in Kumasi ............... 116

Figure 5.12. Species richness, evenness and Shannon diversity at the study sites in Kumasi................................................................................................................................................ 116

Figure 5.13. Species diversity indices with respect to distance from the water channel at the different sampling sites in Kumasi ........................................................................................ 117

Figure 5.14. Analysis of similarities of the community composition among wetland sites in Kumasi according to (a) presence or absence of surface water and (b) distance from the water channel ................................................................................................................................... 119

Figure 5.15. A hierarchical cluster of the community composition among wetland sites in Kumasi according to (a) presence or absence of surface water and (b) distance from the water channel ................................................................................................................................... 121

Figure 5.16. A non-metric multidimensional scaling plot of the community composition among wetland sites in Kumasi using the presence or absence of surface water as a factor . 122

Figure 5.17. A non-metric multidimensional scaling plot of the wetland sites in Kumasi using the measured environmental parameters................................................................................ 122

Figure 5.18. A principal components analysis plot of the sampled wetland sites in Kumasi superimposed with vector plots (circle and lines) of the measured environmental parameters................................................................................................................................................ 123

Figure 6.1. Average monthly rainfall pattern of Kumasi from 1961 to 2010 ........................ 131

Figure 6.2. Historical annual rainfall pattern of Kumasi from 1961 to 2010 ........................ 131

xiii

Figure 6.3. Trends in rainfall of Kumasi for the months of January from 1960 to 2010 ....... 133

Figure 6.4. Trends in rainfall of Kumasi for the months of February from 1960 to 2010 ..... 133

Figure 6.5. Trends in rainfall of Kumasi for the months of March from 1960 to 2010 ......... 134

Figure 6.6. Trends in rainfall of Kumasi for the months of April from 1960 to 2010........... 134

Figure 6.7. Trends in rainfall of Kumasi for the months of May from 1960 to 2010 ............ 135

Figure 6.8. Trends in rainfall of Kumasi for the months of June from 1960 to 2010 ............ 135

Figure 6.9. Trends in rainfall of Kumasi for the months of July from 1960 to 2010 ............ 136

Figure 6.10. Trends in rainfall of Kumasi for the months of August from 1960 to 2010 ..... 136

Figure 6.11. Trends in rainfall of Kumasi for the months of September from 1960 to 2010 137

Figure 6.12. Trends in rainfall of Kumasi for the months of October from 1960 to 2010 .... 137

Figure 6.13. Trends in rainfall of Kumasi for the months of November from 1960 to 2010 138

Figure 6.14. Trends in rainfall of Kumasi for the months of December from 1960 to 2010 . 138

Figure 6.15. Trends in total annual rainfall of Kumasi from 1960 to 2010 ........................... 139

Figure 6.16. Respondents’ flood experience in the various suburbs of Kumasi.................... 140

Figure 6.17. Relationship between different rainfall events and total monthly rainfall for Kumasi from 1961 to 2010 .................................................................................................... 141

Figure 6.18. Trends in heavy rainfall events in Kumasi from 1961 to 2010 ......................... 142

Figure 6.19. Recession time of water at the various flood prone suburbs of Kumasi ........... 144

Figure 6.20. Adaptations of residents to floods in the various suburbs of Kumasi ............... 144

Figure 6.21. Flood response model of new and old residents of flood prone suburbs of Kumasi ................................................................................................................................... 145

Figure 6.22. Landcover changes within the jurisdiction of the Kumasi Metropolitan Assembly between 1986 and 2007.......................................................................................................... 147

xiv

LIST OF TABLES Table 2.1. Likely effects of urbanisation on wetland hydrology and geomorphology ............ 18

Table 2.2. Anthropogenic effects on urban wetlands............................................................... 29

Table 2.3. Likely effects of urbanisation on wetland ecology ................................................. 29

Table 3.1. Stakeholder list and the respective data collection tools used in the survey of wetlands in Kumasi, Ghana ..................................................................................................... 47

Table 4.1. Ghana’s international commitments to the environment ........................................ 68

Table 4.2. Stakeholder capacity assessment on wetlands and flood prevention and management in Kumasi............................................................................................................ 76

Table 5.1. Classes of the geoaccumulation index used to define sediment quality ................. 89

Table 5.2. The Braun-Blanquet scale and the description of the values used for sampling of wetland species in Kumasi ....................................................................................................... 90

Table 5.3. Edaphic characteristics of wetland sites sampled in the Kumasi Metropolis ......... 98

Table 5.4. Physicochemical characteristics of water running through the wetlands in Kumasi.................................................................................................................................................. 98

Table 5.5. Species identified in the different wetlands of Kumasi and their importance values................................................................................................................................................ 109

Table 6.1. Descriptive statistics of monthly rainfall pattern of Kumasi from 1961 to 2010 . 132

Table 6.2. Rainfall events and reported floods in some suburbs of Kumasi in 2009 ............ 140

xv

LIST OF PHOTOGRAPHS Photograph 1. Some uses of wetlands in the Kumasi Metropolis............................................ 97

Photograph 2. State of some of the wetlands in the Kumasi Metropolis ............................... 108

Photograph 3. Some high importance value wetland species in the Kumasi Metropolis ...... 114

Photograph 4. More high importance value wetland species in the Kumasi Metropolis ...... 115

Photograph 5. House being built on stilts in a flood prone suburb in the Kumasi Metropolis................................................................................................................................................ 148

Photograph 6. Raised walkways in the compound of a flood prone house in the Kumasi Metropolis .............................................................................................................................. 148

Photograph 7. Embankment built around a flood prone house in the Kumasi Metropolis.... 149

Photograph 8. Embankment built around the main door to an apartment in a flood prone suburb of the Kumasi Metropolis .......................................................................................... 149

xvi

APPENDICES Appendix 1. Survey instrument used for the wetland associated communities in Kumasi ... 179

Appendix 2. Survey instrument used for wetland associated NGOs and government agencies in Kumasi ............................................................................................................................... 185

Appendix 3. Field data sheet used for wetland vegetation and ecological studies in Kumasi................................................................................................................................................ 186

Appendix 4. Study sites and anthropogenic activities within 100 m from the water channel................................................................................................................................................ 187

Appendix 5. A Wilcoxon Signed Ranks Test comparison of the seasonal differences in physicochemical parameters of the wetlands in Kumasi ....................................................... 187

Appendix 6. Metal concentration (mg/kg) in sediments of the different study sites in Kumasi................................................................................................................................................ 189

Appendix 7. Pearson correlation matrix of heavy metals and other measured environmental variables (significant relationships shown in bold) ............................................................... 190

Appendix 8. Principal Component Analysis of environmental variables and heavy metals assessed in the wetlands of Kumasi ....................................................................................... 191

Appendix 9. Comparison of means of heavy metal enrichment factors for rainy and dry season in Kumasi ................................................................................................................... 192

Appendix 10. Comparison of means of heavy metal geoaccumulation indices for rainy and dry season in Kumasi ............................................................................................................. 193

Appendix 11. Comparison of means of Pollution Load Indices for the different sites in the rainy and dry season in Kumasi ............................................................................................. 194

xvii

LIST OF PUBLICATIONS FROM THIS THESIS The following manuscripts have been prepared with material from this thesis and submitted

for publication:

1. Campion, B. B. & Venzke, J. -F. (2011). Spatial patterns and determinants of wetland

vegetation distribution in the Kumasi Metropolis, Ghana. Wetlands Ecology and

Management, 19(5): 423-431

2. Campion, B. B. & Venzke, J. -F. Rainfall variability, floods and adaptations of the

urban poor to flooding in Kumasi, Ghana. This manuscript has been accepted for publication

in Natural Hazards published by Springer.

3. Campion, B. B. & Venzke, J. -F. Beyond defining wetlands to feeling wet: the

political ecology of wetlands in Kumasi, Ghana. This manuscript is under review by

Wetlands Ecology and Management published by Springer.

In all these papers, I developed the concept of the study, collected data and wrote the manuscript

with scientific advice from Venzke, J. –F.

xviii

ABSTRACT Urbanisation and climate change are two of the greatest environmental challenges the world

is experiencing. Wetlands bear the heaviest effects of urbanisation in Kumasi, Ghana. Also,

current forecasts show that climate change will very much affect sub-Saharan Africa and may

cause flooding in many areas. Already, Kumasi has been battling with a growing incidence

and scale of floods, mostly, at the suburbs closer to wetland areas. This research therefore

sought to investigate the interlocked issues of urbanisation, climate change, wetland ecology

and degradation and flooding in Kumasi. The vegetation, edaphic characteristics, hydrology

and physicochemical conditions of the wetlands were analysed. Various stakeholders were

interviewed on wetlands management, floods and adaptations to living with floods. Rainfall

data from 1961 to 2010 was analysed for trends. The landuse change from vegetated to built

environment was also analysed. Whilst there were some plant species common to all the

wetlands, the species composition for each wetland was unique. In each wetland, the level of soil

water saturation was found to influence species composition. The percentage organic carbon of

the wetland soils was the strongest factor that explained the observed spatial heterogeneity of

wetland vegetation in Kumasi. All the wetlands were found to be polluted by all the heavy

metals investigated. Furthermore, the built-up and bare areas within the Kumasi Metropolitan

Assembly (KMA) have been growing at a rate of 1.9 % per annum between 1986 and 2007.

Conversely, the results indicated no significant change in rainfall of Kumasi for the flood

prone months of June and July and, therefore, the floods were not due to climate change. Two

types of floods were, however, identified in the suburbs studied: flash floods due to heavy

rainfall events and “effluent stream floods” caused by a rise in the water table during the peak

rainy seasons. Adaptations to floods by residents included dredging of streams, erection of embankments around houses or apartments, moving property to higher grounds, building

networks of raised walkways and building houses on stilts. People continued to live in the flood

prone areas because, other forms of capital developed after living in these suburbs for a while,

far outweighed the push factors to relocate. Based on these findings, it was recommended that

the KMA should use the presence of effluent floods and the typical wetland vegetation

identified by this research as defining factors to delineate the wetland boundaries.

Considering the limited financial resources available to the KMA for an engineering-biased

urban drainage system, the approach to managing floods in Kumasi should be a combination

of the reconstruction of the natural wetlands and community specific social interventions.

xix

ZUSAMMENFASSUNG Verstädterung und Klimaänderungen stellen zwei der großen Herausforderungen für die

globale Umwelt dar. Urbanisierungsflächen stehen demzufolge unter hohem Nutzungsdruck;

dies gilt besonders für Feuchtgebiete. Eine hohe Verstädterungsrate, eine langsam vonstatten

gehende Raumplanung sowie die Unfähigkeit und Ungleichheit in der Bereitstellung

grundlegender sanitärer Anlagen führen in Ghana zur Bildung städtischer Ballungsräume mit

unterschiedlicher Umweltgüte. Armenviertel (slums) und informelle Siedlungen entwickeln

sich dabei meist in und um Überschwemmungsgebiete. Zudem zeigen aktuelle Vorhersagen,

dass der Klimawandel insbesondere den subsaharischen Raum Afrikas u. a. mit zunehmenden

Überschwemmungsereignissen betreffen wird. Bereits über einen langen Zeitraum mussten

sich die Behörden in Kumasi mit einer zunehmenden Anzahl an

Überschwemmungsereignissen in Armenviertel und informellen Siedlungen

auseinandersetzen.

Die vorliegende Arbeit stellt die notwendigen Informationen bereit, die einen ganzheitlichen

Lösungsansatz zu der komplexen Problematik aus Verstädterung, Verlust und Degradation

von natürlichen Feuchtgebieten sowie Überschwemmungsereignissen in Kumasi bieten.

Folgende Zielvorgaben stehen dabei im Fokus:

� Beschreibung und Charakteristik städtischer Feuchtgebiete in Kumasi, Ghana,

� Bestimmung von Überflutungsparameter sowie Bewältigungsstrategien gegenüber

Überflutungen in Kumasi,

� Überprüfung der Anwendbarkeit der Umwelt-Kuznets-Kurve (Environmental Kuznets

Curve [EKC]) als Instrument der Planung für überflutungsgefährdete Stadtteile in

Kumasi.

Um diese Ziele zu erreichen, wurden zunächst sämtliche Feuchtgebiete des Großraumes

Kumasi registriert. Neben Angaben zu den Vegetations-, Boden- und hydrologischen

Verhältnissen wurden die physisch-chemischen Zustände der Feuchtgebiete analysiert. Die

Informationen zu Feuchtgebietsmanagement, Überflutungen und Anpassungsstrategien an die

Lebensbedingungen wurden durch Befragungen verschiedener Interessensvertreter

gewonnen. Zudem wurden die Niederschlagsereignisse der vergangenen 50 Jahre

xx

ausgewertet. Darüber hinaus wurde auch der Effekt des Landnutzungswandels von

vegetationsbedeckten zu bebauten Flächen untersucht.

Alle untersuchten Feuchtgebiete gehören zu einem Flusssystem. Während einige der

Pflanzenarten in jedem Überflutungsbereich anzutreffen sind, besitzt jedes Feuchtgebiet

trotzdem eine spezifische Artenzusammensetzung. In jedem Feuchtgebiet folgt die Verteilung

der Vegetationsgesellschaften einem hydrologischen Gradienten; die Bodenwassersättigung

ist der dominante Faktor für die entsprechenden Artenzusammensetzungen. Der prozentuale

Anteil an organischem Kohlenstoff bildet eine weitere wesentliche Komponente zur

räumlichen Heterogenität von Vegetationsgesellschaften in Feuchtgebieten. Während

saisonale Unterschiede die chemischen Zustände des Wassers prägen (pH, DO), gibt es

hingegen keine saisonalen Veränderungen im Vegetationsbestand der Feuchtgebiete. Die

Artenvielfalt in den Feuchtgebieten wird nicht durch die benachbarte Landnutzung

beeinflusst; sondern vielmehr durch die hydrologischen und edaphischen Verhältnisse

innerhalb der Feuchtgebiete selbst. Aufgrund der Verhältnisse nahe der KNUST [Kwame

Nkrumah University of Science and Technology]-Polizeistation werden hier die höchsten

Raten an Kohlenstoffbindung erreicht. In allen untersuchten Feuchtgebieten kann eine

Kontamination mit Schwermetallen festgestellt werden. Wahrscheinlich stammt der Anteil an

Hg, Ni, Zn und Pb aus derselben Quelle, während die Immissionen von Cu und Cr andere

Quellen haben.

Überflutungen in Kumasi sind nicht als Folge veränderter Niederschlagmuster in den

Regenmonaten Juni und Juli zu verstehen. Es gibt zwei Typen von Überschwemmungen, die

einen Einfluss auf die untersuchten Feuchtgebietsgesellschaften ausüben: plötzlich

auftretende Überflutungen aufgrund starker Niederschlagsereignisse sowie Folgen eines

steigenden Grundwasserspiegels, der zur Bildung von Flussläufen beiträgt. Die bebaute und

vegetationsfreie Fläche innerhalb der Großstadtregion von Kumasi (Kumasi Metropolitan

Assembly [KMA]) ist im Zeitraum von 1986 bis 2007 um 1,9 % pro Jahr angestiegen. Dieser

Nutzungswandel führt aufgrund eines reduzierten Infiltrationsvermögen, kürzeren

Ablaufzeiten und folglich höheren Abflussmaxima zu höherem Oberflächenabfluss. Dies ist

die Hauptursache für plötzlich auftretende Überflutungen in Kumasi. Hingegen dauert die

Überflutungszeit an der Oberfläche als Folge ansteigenden Grundwasserpegels während der

Hauptregenzeit länger an.

xxi

Die Einwohner haben sich an die Überflutungssituation weitgehend angepasst. Die

Anpassungsstrategien sind der Bau von Drainagegräben und der Bau von Dämmen um

Gebäude, Verlagerung von Besitz in höher gelegene Bereiche, Aufbau eines höher liegenden

Fußwegenetzes sowie Hausbau auf Pfählen. Wegen dieser Anpassungsmöglichkeiten ist die

ortsansässige Bevölkerung nur bereit fortzuziehen, wenn alternative Wohnmöglichkeiten

angeboten werden. Entsprechende Entscheidungen in den überflutungsgefährdeten Vororten

von Kumasi folgen dabei nicht immer den Vorgaben der EKC. Auch bei steigendem

Einkommen wandert die ortsansässige Bevölkerung nicht ab, auch nicht unter staatlichem

Druck. Die anderen Formen der Lebenssicherung, die sich nach einer gewissen

Aufenthaltszeit in diesen Vororten entwickelt haben, überwiegen die Gründe, die

Feuchtgebiete wegen der Überflutungsgefahr zu verlassen.

Die Mehrheit der für diese Studie befragten Interessensvertreter ist der Auffassung, dass die

Feuchtgebiete sehr wichtig sind und daher geschützt werden sollten. Die wichtigsten

Verantwortlichen für das Feuchtgebiets- und Überschwemmungsmanagement sind die

lokalen Clanchefs sowie die KMA. Zu den wichtigsten Aufgabenbereichen gehören die

Verbesserung von Landnutzungsplänen, Abbruch flutgefährdeter Gebäude und die Errichtung

öffentlicher Parks.

Aus den Ergebnissen geht hervor, dass die KMA das Auftreten von Grundwasserbedingten

Überflutungsereignisse sowie von charakteristischer Feuchtgebietsvegetation, die in dieser

Studie erfasst worden ist, als bestimmende Faktoren zur Abgrenzung von Feuchtgebieten

herangezogen werden sollten. Diese Methodik wurde besonders in den Feuchtgebieten nahe

der KNUST-Polizeistation, in den FRNR [Faculty Renewable Natural Resources]-

Bauernhöfen, in Atonsu, Family Chapel, Dakwadwom, Spetinpom und Golden Tulip Hotel

exemplarisch angewendet. Nach Festlegung der Grenzen der Überschwemmungsgebiete

sollten zunächst alle dort vorhandenen Bauten zerstört werden. In den mit neuen Namen

gekennzeichneten, sich mit Vegetation überziehenden Feuchtgebieten werden sich die

natürlichen Ökosystemfunktionen regenerieren. Dazu sollte die KMA die Zusammenarbeit

von Asantehene (König von Ashanti) anstreben, um mögliche Widerstände gegenüber den

Planungsmaßnahmen einzuschränken.

Zukünftig sollten stadtökologische Fragestellungen in den Blickpunkt wissenschaftlicher

Studien mehr berücksichtigt werden. Dazu gehören die mögliche Nutzung von Vegetation zur

xxii

Behandlung von städtischem Abwasser sowie die Bedeutung von Feuchtgebietsvegetation

zum Hochwassermanagement und zur Kohlenstoffbindung. Zudem sollten diese

wissenschaftlichen Untersuchungen die Grundlage für die Bewirtschaftungsmaßnahmen in

abgegrenzten Feuchtgebietsregionen darstellen, um die Biodiversität zu erhalten sowie den

Erholungswert dieser Räume zu steigern.

Um den direkten Eintrag von Schadstoffen in die Gewässer zu verringern, sollte die KMA die

Unterbringung von Abfallsammelbecken in Feuchtgebieten einstellen. Wo es für notwendig

erscheint, sollte eine Mauer um eine derartige Anlage eingerichtet werden, um Überläufe der

Abwässer zu verhindern.

Die Feuchtgebiete um die KNUST-Polizeistation sollte aufgrund ihres ökologischen Wertes

für die Erhaltung von Torfvorkommen geschützt werden. Dieser Standort könnte aufgrund

seines Potenzials als „Kohlenstoffsenke“ im Rahmen des CO2-Handels zur Abschwächung

des Klimawandels als mögliche Einnahmequelle für die KMA fungieren. Beispielsweise böte

die Errichtung einer Gartenstadt eine praktische Umsetzung des von den Vereinten Nationen

verabschiedeten Gesetzes zur „Reduzierung von Emission aus Abholzung und Wald-

Degradierung“ (REDD+). Dies sollte der zukünftige Weg von Kumasi sein, um den

Forderungen der Konvention für globale Städtepartnerschaften für Biodiversität

nachzukommen.

Aufgrund der begrenzten finanziellen Möglichkeiten der KMA zur Errichtung eines technisch

adäquaten städtischen Drainagesystems zur Bewältigung des Überschwemmungsproblems

wären eine Kombination aus Umgestaltungen der Feuchtgebiete, wie schon oben

beschrieben, sowie sozialpolitische Maßnahmen notwendig. Ein derartiges

Überschwemmungsmanagement sollte in jedem Vorort dazu dienen, Schwerpunkte bei der

Reduzierung von Überschwemmungsflächen und -häufigkeiten zu setzen. In diesem

Zusammenhang müssen KMA und die „National Disaster Management Organisation“

erkennen, dass die Einbindung der lokalen Clanchefs, Abgeordneter und relevanter NGO’s

wichtig sind.

1

1 RESEARCH THEME AND THEORETICAL FRAMEWORK

1.1 INTRODUCTION Ghana became a signatory to the Ramsar Convention in 1988. However, more than two

decades later, wetlands are still considered as ‘waste lands’, ‘flood zones’ or ‘mosquito

breeding grounds’ and as such, dredged to facilitate drainage of the water, reclaimed for other

uses, or simply designated as dumping sites for all types of refuse (Ministry of Lands and

Forestry, 1999a). Even with various efforts to protect or improve the wetlands according to

the dictates of the Convention, recent developments in human geography and socioeconomics

continue to be the major threats to wetlands in Ghana.

Urbanisation is one of the greatest environmental changes the developing world is

uszndergoing. The urban land area is therefore under severe pressure for various uses and

wetlands bear the heaviest of them all (Ramsar Convention Secretariat, 2007). Even with

technological progress in city infrastructure, a growing population puts pressure on

biodiversity (Eppinka et al., 2004). Therefore, to safeguard and possibly enhance the

livelihood of many urban and peri-urban communities who subsist on wetlands, it is

imperative that the benefits of the natural wetland ecosystems, including their values for

subsistence economies, are recognised when planning and implementing development

projects (Silvius et al., 2000). This requires identification and delineation these urban wetland

‘ecosystem’ boundaries for their management and sustainability to reduce their conflicting

and competing uses for housing, infrastructure and agriculture.

The problems that climate change, urbanisation and floods present are becoming impossible

to differentiate. The scale, frequency and cost of urban floods are on the rise and are

sometimes attributed to climate change without the requisite support from research. Global

climate change impacts natural vegetation, affects ecological and physiological processes,

alters growing season length, biomass production, competition and leads to shifts in species

ranges and possible extinctions (Levy et al., 2004; Intergovernmental Panel on Climate

Change [IPCC], 2007). Climate change mitigation has therefore been the motivation for

efforts to further understand carbon budgets and carbon accounting at local scales to

contribute to regional and global carbon budget information (Risbey et al., 2007; Tschakert et

al., 2008). The need for information on carbon pathways in major urban areas in Ghana can

2

therefore not be overemphasised. Terrestrial ecosystems represent a large store of carbon,

approximately, three times that of the atmosphere (Levy et al., 2004). Also, metre for metre,

wetlands lock up more carbon than any other terrestrial ecosystems in active exchange with

the atmosphere. For this reason, as interest in climate change mitigation grows, interest in

wetlands intensifies in parallel (Ramsar Convention Secretariat, 2006; Maltby & Barker,

2009). With such growing interest, wetland ecosystems in Kumasi have the potential to serve

effectively as the basis for mitigating and adapting to climate change.

1.2 RESEARCH BACKGROUND AND JUSTIFICATION

1.2.1 Urban Environmental Issues Urbanisation is a threat to many natural habitats and species because species richness,

evenness, and density on farms, lawns or gardens in urban areas are directly controlled by

man (Niemelä, 1999). These human-dictated urban vegetation communities form the template

for other functional groups of species (Taylor, 1993; Barrat-Segretain & Amoros, 1996).

Urbanisation may also lead to increased patch fragmentation (natural, semi-natural or

artificial) and changes in structure (biomass, density, stratification, etc.) and floristic

composition and spatial diversity (Ehrenfeld, 2000; Altobelli et al., 2007; Grimm et al.,

2008). Like most sub-Saharan countries, urbanisation has been a dominant demographic

trend in Ghana. By the year 2020, urban population in Ghana will more than double that at

2000 whilst rural population would have increased by only 3 % (Ghana Statistical Service,

2002a). Given that urban landuse and footprint will continue to expand, the prognosis for

maintaining diversity and function of biological communities and their associated ecosystem

services within and near Ghanaian cities seems dire. Unlike the natural habitats, the

biodiversity of urban habitats is poorly documented in Ghanaian cities, and thus baseline

information is scarce and the possibilities of applying ecological knowledge in urban

planning are limited (Niemelä, 1999).

Factors such as rates of population growth, urbanisation, inappropriate macroeconomic,

social, and technological policies and choices have been considered as some of the causes

and implications of habitat change in Africa (Kidane-Mariam, 2003; Satterthwaite, 2003). For

example, in Ghana, almost all the non-agricultural economic activities take place in the urban

areas. Therefore, peri-urban and rural landowners are usually excited about the potential

opportunities that will be available to them when the urban economy and space merge with

the peri-urban. This leads to an increase in demand for land for infrastructure, industry,

3

service and housing, thereby, causing a drastic landuse change in these transition zones. The

land values rise dramatically and much income is accrued from the sale of lands as they are

put to more productive and profitable uses. Due to the high economic incentives and

developers’ need for lands at already populated areas, wetlands are also sold, although, at

cheaper prices (Payne, 1997; Brook & Dávila, 2000). Urbanisation and economic expansion

then outpace environmental controls and planning (Grimm et al., 2008). With the high rates

of urbanisation, slow pace of urban spatial planning, inability and inequalities in the

provision of basic sanitation services in Ghana, the cities differentiate into spatial clusters of

different environmental quality. The worst zones are the clusters of informal communities or

slums associated with wetlands (World Bank, 2007). The situation is further compounded by

the deliberate filling and dumping of solid and liquid wastes into these wetlands and along

waterways (Water Resources Commission, 2008). In most towns and cities, skip bins are

peculiarly placed at wetlands leading to the gradual conversion of the area into rubbish

dumps (Fobil & Atuguba, 2004).

The two major Ghanaian cities, Kumasi and Accra have been battling with a growing

frequency and scale of floods. These floods are mostly associated with informal and slum

communities in wetland areas. The nature and growth of these cities however make it

difficult to identify the real cause of the floods. Whilst the multitude of physical factors

described above cannot be overlooked, changing climatic conditions may also be a likely

cause of floods.

Slum population in many Sub-Saharan cities accounts for more than 70 % of the urban

population and are dependent on natural resources associated with the already pressured

wetlands (Rietbergen et al., 2002). Poverty alleviation measures for these slum communities

(e.g. industrialisation, agriculture, flood control and protection works etc.) are also drivers of

ecosystem and environmental degradation (World Meteorological Organisation, 2006;

Satterthwaite, 2003). These communities are also the most vulnerable, both physically and

socially, to environmental hazards like the annual floods and tend to be the least influential

economically and politically. Wetland degradation is therefore likely to continue to increase

in slum areas because of the increasing incidence of urban poverty (Baharoglu & Kessides,

2000; Baker & Schuler, 2004).

4

The above vulnerability situation may be widespread in Ghana because the country has no

urban policy. National responses to urbanisation have so far been very piecemeal and not

sufficiently integrated and holistic. According to the United Nations Human Settlements

Programme (UN-HABITAT), the challenges and priorities facing Ghana among others, are

urban planning and management, urban environment and development and vulnerability

reduction (UN-HABITAT, 2008a). Following these rising urban environmental problems, the

Ministry of Local Government and Rural Development announced that it was in the process

of developing a national urban policy (GNA, 2010b). This is supposed to functionally

integrate social, economic, institutional, cultural, environmental and spatial aspects of urban

development. However, the ecological justification of such a policy, if formulated, is likely to

be weak because of the substantial dearth of information on urban ecosystems in Ghana.

1.2.2 Research Justification Global interest in wetlands, climate change and urbanisation has been on the increase in

recent years. Notable recent efforts are the United Nations Conference on Environment and

Development (UNCED) and its aftermath Conventions on Biodiversity and Climate Change,

the International Peat Society Commission on Land use Planning and Environment, IUCN

Wetlands Programme, Ramsar Convention, UN World Charter for Nature, UNESCO Man

and the Biosphere Reserve Programme, UNESCO Convention for the Protection of World

Cultural and Natural Heritage, World Wide Fund for Nature (WWF), the United Nations

Environment Programme’s Cities as Sustainable Ecosystems initiative and the International

Human Dimensions Programme on Global Environmental Change.

The Ramsar Convention Secretariat (2008) has called on all Parties to review the state of

their urban and peri-urban wetlands and to put in place schemes for restoration and

rehabilitation; formulate landuse planning and management to minimize any future impacts

on urban wetlands; and become involved in the Convention on Biological Diversity’s Global

Partnership on Cities and Biodiversity. Another output of the Conference is the call for

preparation of guidelines for managing urban wetlands, in accordance with an ecosystem

approach, taking into account issues such as climate change, ecosystem services and

livelihoods. The following year, the UN chose ‘Planning our Urban Future’ as its theme for

the 2009 World Habitat Day Celebration. The World Disasters Report 2010 also focused on

urban risk impacts of changing precipitation patterns (International Federation of Red Cross

and Red Crescent Societies [IFRC], 2010).

5

These international concerns are relevant to Ghana. Being a member of the international

community and signatory to several treaties and conventions, the Government of Ghana

recognises the importance of wetlands (Ministry of Lands and Forestry, 1999a). Ghana’s

wetland commitments are incorporated into an extensive poverty reduction strategy (PRS)

process that integrates environmental issues with development and poverty reduction

(Griebenow & Kishore, 2009). Principal commitments in the PRS are the Millennium

Development Goals (MDGs), the Kyoto Protocol, the New Partnership for African

Development (NEPAD) and the Convention on Biological Diversity (National Development

Planning Commission, 2005). The peer to peer review by NEPAD of the PRS process

recommended that Ghana should ensure effective environmental sustainability in programme

and policy designs of ministries, departments and agencies of government and improve their

capacity to practice environmental sustainability (African Peer Review Mechanism

Secretariat, 2005). These recommendations have not been implemented by the Kumasi

Metropolitan Assembly (KMA) and all the other assemblies.

In assessing Ghana’s efforts at attaining environmental sustainability (i.e. MDG 7), it is

critical to note that, some of the most cost-intensive interventions have so far not been

addressed. These include developing systems for environmental monitoring, landuse

management, management of watersheds and freshwater ecosystems and biodiversity

conservation among others (Sachs et al., 2004). Projections by the National Development

Planning Commission (2005) and National Development Planning Commission & UNDP

(2010) show that Ghana may not meet the targets of the MDG 7 and therefore set out three

critical areas to be considered:

� institutional resources for monitoring and evaluation, regulation and enforcement and

environmental management;

� resources (human and financial) development for education and training; and

� specific investment in institutional capacity and policy reforms.

It also proposed the development of a landuse master plan that demarcates all lands in

dispute; initiation of measures to stem land degradation; minimization of the impact of

climate change/variability; and promotion of human centred biodiversity conservation

initiatives.

Environmental research and information to meet these targets have been lacking, not only in

Ghana but in developing countries as a whole (Sánchez-Rodríguez et al., 2005). With the

6

issue of climate change rapidly gaining the right political attention and momentum, scientists

need to take advantage of this trend of events to identify possible local climate impacts. It is

now known that climate change will very much affect sub-Saharan Africa (Niasse et al.,

2004; Millennium Ecosystem Assessment, 2005; Intergovernmental Panel on Climate

Change [IPCC], 2007; Satterthwaite, 2008), especially, its water resources (Reinelt et al.,

1998; Choi et al., 2003; United Nations Department of Economic and Social Affairs, 2005;

Jenerettea et al., 2006; Obuobie et al., 2006) and may cause flooding in many areas (Fedeski

& Gwilliam, 2007; IPCC, 2007; International Recovery Platform, 2009; McClean, 2010).

This information on sub-Saharan African countries is generated from running climate models.

However, no studies have been conducted on specific local climate change effects and

adaptations by the vulnerable urban wetland communities in Kumasi, Ghana.

Relevant research conducted in Kumasi and its environs have been on pollution and peri-

urban agriculture (Brook & Davila, 2000; Keraita et al., 2002; Cornish & Kielen, 2003;

Keraita et al., 2003; Cofie et al., 2005), livelihood changes in peri-urban environments

(Holland et al., 1996; Danso et al., 2002; Quansah et al., 2005; Danso et al., 2006; Drechsel,

2006; McGregor et al., 2006), urban planning (Mabogunje, 1990), land and land tenure

(Blake & Kasanga, 1997; Botchie et al., 2007). There is certainly a shortfall in depth and

direction of research that would provide a basis for climate mitigation in the urban

environment in Kumasi, Ghana. Also, the GPRS II does not provide room for GIS-based

ecological and poverty maps and climate change scenarios to offer practical information and

innovation required to better inform development policy. There is no information to use to

develop instruments for the vulnerable to cope with the impacts of climate change by

themselves. There is no information on the effects of climate change on urban wetland

vegetation dynamics, wetland hydrology and carbon sequestration in the light of the

increasing pollution, urbanisation, agriculture and for carbon budgeting. In order to

understand the link between urbanisation and the incidence of flooding, there is the need to

know the spatial extent and characteristics of wetlands, land cover changes and fragmentation

as an information input for better spatial planning in Kumasi. This research is therefore an

attempt to provide the needed information for a holistic solution to the interlocked issues of

climate change, urbanisation, wetlands degradation and flooding in Kumasi.

7

1.3 RESEARCH GOAL This research therefore sought to contribute to knowledge on the state of wetlands (ecology,

carbon stocks, and drivers of change), incidence of flooding and stakeholder needs for

management of wetlands in Kumasi. It will also be an initial attempt to develop the

information base for indicators of wetland biotic integrity in the Kumasi Metropolitan area as

influenced by urbanisation.

1.4 RESEARCH OBJECTIVES � To describe and characterise urban wetlands in Kumasi

� To identify the determinants of floods and coping strategies to flooding in Kumasi

� To test the applicability of the Environmental Kuznets Curve in the decision making

of members of flood prone communities in Kumasi

1.5 RESEARCH QUESTIONS In order to achieve these objectives, this research answered the following questions:

1. What are the characteristics of wetlands in Kumasi and in which ways are they

changing spatially and temporally?

a. What are the physical and chemical characteristics of wetlands in Kumasi how are

they related to the adjacent socio-economic activities?

b. What are the ecological drivers of change in these urban wetland ecosystems?

c. What are the wetland vegetation species and the spatial and temporal differences

in distribution of these species in Kumasi?

d. What is the carbon sequestration potential of these urban wetland ecosystems?

2. What are the effects of climate and landcover-landuse change on the incidence

of floods in Kumasi?

a. Is there any relationship between the incidence of floods and the climatic

(precipitation) characteristics of Kumasi?

b. What are the landuse-landcover changes in Kumasi and how do they affect the

incidence of floods in Kumasi?

3. How does the economy of wetland communities influence their environmental

concerns?

a. What are the socio-economic activities associated with wetland communities in

Kumasi?

8

b. Does the trend of wetland degradation in Kumasi follow the Environmental

Kuznets Curve?

i. How far to left of the continuum of the Environmental Kuznets Curve are

members of these wetland communities in Kumasi?

ii. What do these communities need to reach the peak or flatten the Curve?

4. How are wetland associated communities and authorities managing the

associated wetlands and floods?

a. What are the existing wetland management practices and adaptations to life by

wetland associated communities to life in such environments?

b. What are the opportunities for implementing the Ghana Wetlands Management

Strategy in Kumasi?

1.6 CONCEPTUAL AND THEORETICAL FRAMEWORKS This research is based on the following conceptual framework represented in Figure 1.1. The

total amount of wetlands in Kumasi is considered as a non-renewable stock. This wetland

stock is being reduced through the influence of urbanisation from the conversion of wetland

areas into built environment. Also, these wetland areas, by virtue of their nature, do not

attract high prices and in most places are not even sold. They are therefore invaded by poor

people and converted into informal or slum settlements and the remaining wetland areas are

associated with intense economic activities of these wetland inhabitants. The dense built up

environment and clearing of vegetation in the wetlands therefore lead to increased frequency

and scale of flooding of these suburbs. Storm drains are then built as a flood mitigation

strategy but these have not been successful in solving the problem of floods.

This research assumes that, there are two pathways available to the KMA for the

management of the wetlands and flooding: (i) continue with the business-as-usual

engineering solution or (ii) put in place some wetland management interventions based on

activities and programmes that will be investigated in this study (research themes [RT]).

The latter will lead to an improved stock and state of wetlands in Kumasi offering positive

environmental services including the reduction in floods and or its effects.

The business-as-usual and engineering solution will result in a vicious cycle of

environmental ills. This is because, with increasing urbanisation, the city will frequently have

to expand the storm drains to cope with the increasing runoff. Meanwhile, degradation and

9

loss of the wetland areas continue. The consequence of this is an increase in the frequency

and intensity of floods and loss of environmental quality.

The above conceptual framework is based on the assumption that, urban processes and

natural wetland ecosystems can co-exist. This is because, the perception of wetlands as a

limit to urban expansion and economic growth could be changed to a rather active role in

reducing urban poverty, achieving higher living standards and increasing human development

levels in a sustainable manner (Costantini & Monni, 2008). However, to achieve this, the

very complex links between urbanisation, wetlands, floods and human activities have to be

understood. The theoretical basis of this research, therefore, piggybacks an adapted model of

McDonnell & Pickett (1990) and Ehrenfeld (2000) on urbanisation and ecological

phenomena (Figure 1.2). This research envisages that, there are interactions between the

features of urban areas (column A) and ecological phenomena (column B). Manifestations of

these interactions in Kumasi are shown in column C. Ehrenfeld (2000) proposed some

components and indicator variables for assessment of such urban wetlands (column D).

In this model, the urban environment is divided into four entities: 1. human physical

constructs, 2. the natural areas, 3. humans with their accompanying socio-economic activities

and 4. climate. The human dwellings, roads, factories, offices etc. make up the physical

abiotic developments; the urban greens, parks, wetlands etc. form the natural areas; the

human influence is characterised by various socio-economic activities such as trading,

carpentry, agriculture etc.; and all these interact and are influenced by the local climate.

These form the drivers of change (A) of the environment. Urbanisation provides an

opportunity for all these factors to closely interact (B) to produce the current wetlands of

Kumasi, C (results). These wetlands have reduced ecological value because of encroachment

and activities of the informal and slum communities. The research themes (D) are derived

from phenomena taking place in rows 1, 2, 3 and 4. That is, this research seeks to investigate

how the changes in the phenomena represented by the rows 1, 2, 3 and 4 of column D lead to

the manifestations in column C in Kumasi.

10

RT

Fi

gure

1.1

. C

once

ptua

l fr

amew

ork

for

the

inve

stig

atio

n of

urb

anis

atio

n, s

ocio

-eco

nom

ic a

ctiv

ities

and

inci

denc

e of

flo

ods

in K

umas

i(R

T= r

esea

rch

them

es o

r ar

eas

bein

g in

vest

igat

ed)

outc

ome

decis

ion

curr

ent

stat

e de

cisio

n ou

tcom

e

11

Figu

re 1

.2.

A c

ompo

site

mod

el o

f th

e ef

fect

s of

hum

an i

nter

fere

nce

on a

wet

land

eco

syst

em i

n an

urb

an s

ettin

g (a

dapt

ed f

rom

McD

onne

ll &

Pic

kett,

199

0; E

hren

feld

, 20

00)

A

D

C

B

3 So

cio-e

cono

mic

fact

ors

Resu

lt

1 2 4

Popu

latio

n an

d co

mm

unity

effe

cts

Man

agem

ent

and

capi

tal

appo

rtion

men

t

Stru

ctur

al fe

atur

es

of

urba

n ar

eas

Wet

land

area

s

Dive

rs o

f Cha

nge

Phys

ical

and

chem

ical

envir

onm

ent

Biot

ic an

d en

viron

men

tal

effe

cts

of u

rban

isatio

n

Clim

atic

effe

cts

Rese

arch

Cur

rent

sta

te o

f ur

ban

wet

land

Ecos

yste

ms

Ecol

ogy

(soi

l, ve

geta

tion,

hy

drol

ogy,

phy

sicoc

hem

ical

ch

arac

teris

tics)

Urb

anisa

tion

(spa

tial

and

tem

pora

l ch

ange

s in

phys

ical

struc

ture

s, po

llutio

n)

Socio

-eco

nom

ic co

nditio

ns

(floo

ds,

wet

land

man

agem

ent,

anth

ropo

geni

c ac

tiviti

es)

? ?

Red

uced

va

lue

and

func

tion

of

wet

land

s

Floo

ds,

Slum

s

?

12

1.7 HYPOTHESES The following hypotheses have been formulated to guide this study.

I. According to the Federal Interagency Committee for Wetlands Delineation (1989), Keddy

(2000) and Maltby & Barker (2009), wetland vegetation composition is dependent on the

hydrological characteristics of the site. That is, water level fluctuations and depth of

inundation during the year are important in the development of plant associations (Reinelt et

al., 1998; Sassa et al., 2010). In an urban setting, however, the vegetation is often

manipulated for aesthetics and design, biological conservation, and for reduction in some

forms of pollutants (Botkin & Beverage, 1997; Owen, 1999). The urban vegetation is

therefore often a mix of biological deserts and areas that are rich in biodiversity. Therefore,

the structure, floristic composition and spatial distribution of urban wetland vegetation may

rather be the result of deliberate human interventions to satisfy human interests or adjacent

anthropogenic economic activities (Altobelli et al., 2007). It can therefore be argued that, in

urban wetland environments, the anthropogenic influences or interests rather than the

hydrology determine the vegetation.

Ho: Wetland vegetation composition in the Kumasi Metropolis is not determined by the

hydrological characteristics of the wetland.

II. Nature is the original source of all heavy metals. The presence or absence of a heavy metal

in any environment therefore depends on the site characteristics. Recent developments in less

developed countries have led to a wave of rapid urbanisation (UNFPA, 2007). Even though

these urban areas currently cover less than 3 % of the total land area of the earth, they are

places of intense economic activities (Millennium Ecosystem Assessment, 2000). The direct

effect of urbanisation and the associated intense activities is pollution of ecosystems

(Sánchez-Rodríguez et al., 2005). For this reason, while heavy metals may be ubiquitous in

urban environments due to natural processes, the increased human activities due to

urbanisation and population growth could make a significant contribution beyond their natural

concentrations (Sekabira et al., 2010). Spatial variations in heavy metals in urban

environments could therefore be a reflection of the local anthropogenic activities.

Ho: Concentration of heavy metals in the wetland sediments of Kumasi is not greater than the

natural sediment concentration.

13

III. Flooding has claimed more lives than any other single natural catastrophe (IPCC, 2007).

In the decade 1986-1995, flooding accounted for 31 % of the global economic losses from

natural catastrophes and 55 % of the casualties (Petrov et al., 2005). Also, recent analyses by

the IPCC (2007) show that sub-Saharan Africa will experience the worst effects of climate

change. On the other hand, when urbanisation occurs, some or all of the functions and values

of wetlands are affected by direct activities such as dredging, filling, draining or outlet

modification, while others are affected by secondary activities, such as increased impervious

surfaces, leading to increased quantity of inflow water (Mumphrey & Whalen, 1977; Reinelt

& Horner, 1991; Brody et al., 2007). Roy (2009) therefore concludes that the causality and

vulnerability of urban areas to floods are clearly linked to the very nature of the urbanisation

process. The floods at the various suburbs may therefore be outcomes of the urbanisation

process rather than climate change. From the above arguments, it is therefore hypothesized

that:

Ho: The flooding in the suburbs of Kumasi is not due to climate change.

IV. Studies show that an inverted U-shaped relationship exists between environmental

degradation and per capita income (Magnani, 2001; Dinda, 2004; Costantini & Monni, 2008;

Wagner, 2008). Panayotou (1993) therefore adduces from this relationship that, economic

growth could be a powerful way for improving environmental quality in developing countries.

That is, efforts at improving the economy of developing countries will lead to improved

environmental quality. However, will there be an automatic improvement in environmental

quality as income levels increase or it requires conscious institutional and policy reforms? If

such a relationship exists, does the current environmental quality in the flood prone suburbs of

Kumasi correspond to the economic status of its residents? Are the wetland slum communities

an indicator of a stage of economic development of Kumasi? If such an inverted U-shaped

relationship is true, why are rich countries or economies still degrading their wetlands? It is

therefore hypothesized that:

Ho: The relationship between income and wetland degradation in the various suburbs of

Kumasi will not follow an inverted-U-shaped relationship.

14

2 WETLANDS, URBANISATION AND FLOOD NEXUS

2.1 INTRODUCTION Extensive literature on wetlands, urban wetlands, tropical wetlands, urbanisation, urbanisation

in developing countries, urban ecology, urban poverty, flooding, climate change and carbon

budgets was thoroughly reviewed. From this review, research gaps pertaining to wetlands,

urbanisation and floods in Kumasi were identified. Also, the appropriate research methods

were developed to ensure that methods are standard scientific procedure and comparable to

similar research conducted elsewhere. Literature used in this review were generally peer

reviewed documents from the following databases and resources: CAB Direct, African

Journals Online (AJOL), ELDIS, ScienceDirect, Elsevier, Kwame Nkrumah University of

Science and Technology (KNUST) and University of Bremen library resources and

Government of Ghana reports and policy documents. The search was conducted using

wetlands, urban, flooding, urban planning, urban poverty, ecosystem change, climate change,

urban ecology, Kumasi, Ghana etc. as search words.

The search results showed that on the global level, significant research has been published in

these thematic areas. Much of these publications are, however, focused on developed

countries. There is relatively very little research conducted in these thematic areas involving

Kumasi, Ghana or sub-Saharan Africa. This review is, therefore, very much based on the

specific literature obtained and others which may not be that specific to the study area or

region, but relevant to the assessment of the research and knowledge gaps on wetlands,

urbanisation, poverty, climate change and flooding in Kumasi, Ghana.

2.2 WETLANDS Man has had a very long association with wetlands at different places. The concept of

wetlands therefore has different meanings or sentiments because of the distinctive

relationships with different interest groups in different parts of the world (Maltby & Barker,

2009). The different meanings may be related to the culture or history of a people or

socioeconomic uses to which it has been put etc. Man, in changing from the wanderer to a

settler, has been associated with water bodies. This man-water interaction has been beneficial

and hazardous to both ends of the continuum and the meanings of wetlands sometimes reflect

these experiences. However, the term wetland is a recent vocabulary compared to man’s

association with it (Keddy, 2000). Wetlands are usually associated with floods, swamps,

wastelands, mosquito-infested mud holes etc. It is therefore worthy to note that some of the

15

earlier delineations of wetlands in the USA were for their possible conversion to damp sites

(Federal Register, 1980).

According to Haslam (2003), ‘wetland’ is originally a North American term and is generally

referred to, interchangeably, as bogs, fens, marsh, swamps, sloughs, morasses, moor, mires,

flood meadows etc. in non-ecological literature on the other side of the Atlantic. Wetlands

have a combination of aquatic and terrestrial conditions that produce complex ecosystems that

do not fit completely into aquatic or terrestrial ecology (Figure 2.1). As transitional zones or

ecotones, they have characteristics and species typical of the communities they divide (Mitsch

& Gosselink, 1993). Therefore, the characteristics of wetlands are determined largely by

hydrologic processes which may exhibit daily, seasonal or longer-term fluctuations, in

relation to regional climate and geographic location of the site (Halls, 1997).

Figure 2.1. An illustration of the continuum between the terrestrial, wetland and aquatic systems (source: Mitsch & Gosselink, 1993)

From the illustration in Figure 2.1, wetlands are ecological areas which may be marshes,

peatlands, floodplains, rivers and lakes, and coastal areas such as salt marshes, mangroves,

and seagrass beds, but also coral reefs as well as human-made wetlands such as waste-water

treatment ponds and reservoirs. Because of this broad spectrum of ecosystems and the long

16

historical list of interchangeable words used to describe wetlands in different locations by

different cultures, the definition of wetlands has been of much controversy. For example, in

the 2005 Millennium Ecosystem Assessment report, ‘wetlands’ and ‘inland waters’ are used

interchangeably. Also, “wetlands and mangroves” and “lakes, rivers, wetlands, and shallow

groundwater aquifers” are mentioned as habitat types (Millennium Ecosystem Assessment,

2005). It is not surprising that such a renowned publication by a team of experts will have

problems defining wetlands. Most authors therefore prefer to mention the various wetland

ecosystem types rather than attempt a definition.

One of the earliest formal definitions is that of Ramsar as “areas of marsh, fen, peatland or

water, whether natural or artificial, permanent or temporary, with water that is static or

flowing, fresh, brackish or salt, including areas of marine water the depth of which at low tide

does not exceed six metres” (Ramsar Convention Secretariat, 2006). This definition embraces

almost all land area. Large tracts of land are sometimes temporary flooded due to

anthropogenic activities or changes to the landscape. Also, changes in rainfall patterns

sometimes create large flood zones that may only be temporal. In places where demand for

land is very high, e.g. urban areas, such a broad definition will only exacerbate landuse

conflicts. A less “land grabbing” definition is that by US Army Corps of Engineers and the

US Environmental Protection Agency which considered wetlands as “areas that are inundated

or saturated by surface or ground water at a frequency and duration sufficient to support, and

that under normal circumstances do support, a prevalence of vegetation typically adapted for

life in saturated soil conditions” (Federal Interagency Committee for Wetlands Delineation,

1989). This definition does not include artificial wetlands because under normal

circumstances, they may hitherto have been a different land type and therefore, not supported

wetland vegetation. It is also worthy to note that this definition was formulated as a guideline

for selecting places to be designated as landfill sites in the USA and therefore did not include

coastal zones as wetlands.

Keddy (2000) emphasizes on inundation and defines wetlands as product of a cause-effect

relationship. According to Keddy, ‘a wetland is an ecosystem that arises when inundation by

water produces soils dominated by anaerobic processes and forces the biota, particularly

rooted plants to exhibit adaptations to tolerate flooding’. The use of ‘forces’ in the definition

undermines the fact that wetlands are a natural part of the environment with plants and

animals adapted to such an environment. Therefore, there is no forcing of biota to adapt. The

description of wetlands forcing biota makes wetlands look like a recent artificial creation such

17

that the biota that are ‘caught up’ in such a new environment are forced to adapt. With these

diverse definitions and shortfalls, some wetland interest groups (International Mire

Conservation Group and International Peat Society with support from the Dutch Ministry of

Foreign Affairs, IUCN, Alterra, Wetlands International and Environment Canada)

commissioned a study that revised the definition as follows: “a wetland is an area (of marsh,

fen, peatland or water, whether natural or artificial, permanent or temporal) that is inundated

(up to a depth of 6 m at low tide) or saturated by water (surface/ground, static/flowing,

fresh/brackish) at a frequency and for a duration sufficient to support a prevalence of

vegetation typically adapted for life in saturated soil conditions” (Joosten & Clarke, 2002).

This all-encompassing definition could be considered the most sufficient for delineation,

conservation and management of wetlands. According to Maltby & Barker (2009), the

definitions of wetlands communicate the importance of substrate, vegetation and water in the

identification, development and persistence of wetlands. It is therefore often argued that a

robust definition that can provide effective guidance towards wetlands identification and

delineation must include these three elements. For the purpose of this study therefore, the

definition of Joosten & Clarke (2002) will be adopted. The Government of Ghana however,

uses the Ramsar definition.

2.2.1 Morphology and Hydrodynamics of Wetlands Wetland morphology is as diverse as the landscape in which it is situated. The source and

state of the water, topography and geology all determine the morphology of the wetland. The

prevailing climatic conditions, wetland morphology and hydrology determine the wetland

ecosystem attributes: species composition of the plant community, primary productivity,

organic deposition and flux and nutrient cycling (Halls, 1997; Ot’ahel’ová et al., 2007). In an

urban setting, wetlands are further impacted by anthropogenic activities leading to

modifications in the vegetation, hydrological and morphological conditions as shown in Table

2.1. Upstream activities affect the watershed hydrology through changes in the average runoff

supply and changes in the frequency and severity of extreme events, including flooding and

drought. However, these have been difficult to assess due to complexities in the processes at

work as well as a non-uniform and, in many parts of the world, deteriorating monitoring

network (Millennium Ecosystem Assessment, 2005). In Ghana, gauges are rare on rivers and

data on runoff is therefore difficult to come by.

18

Table 2.1. Likely effects of urbanisation on wetland hydrology and geomorphology (Ehrenfeld, 2000)

Hydrology:

� Decreased surface storage of stormwater results in increased surface runoff (Decreased surface water input to wetland)

� Increased stormwater discharge relative to baseflow discharge results in increased erosive force within stream channels, which results in increased sediment inputs to recipient coastal systems

� Changes occur in water quality (increased turbidity, increased nutrients, metals, organic pollutants, decreased O2, etc.)

� Culverts, outfalls, etc., replace low-order streams; this results in more variable baseflow and low-flow conditions

� Decreased groundwater recharge results in decreased groundwater flow, which reduces baseflow and may eliminate dry-season streamflow

� Increased flood frequency and magnitude result in more scour of wetland surface, physical disturbance of vegetation

� Increase in range of flow rates (low flows are diminished; high flows are augmented) may deprive wetlands of water during dry weather

� Greater regulation of flows decreases magnitude of spring flush

Geomorphology:

� Decreased sinuosity of wetland/upland edge reduces amount of ecotone habitat

� Decreased sinuosity of stream and river channels results in increased velocity of stream water discharge to receiving wetlands

� Alterations in shape of slopes (e.g., convexity) affects water gathering or water-disseminating properties

� Increased cross-sectional area of stream channels (due to erosional effects of increased flood peak flow) increases erosion along banks

2.2.2 Distribution, Values and Degradation of Wetlands The variations in the definitions of wetlands make it a near impossibility to know the total

area and distribution of the world’s wetlands (Barbier, 1994; Finlayson et al., 2000). It is,

however, generally estimated to be about 6 % of the earth surface (Haslam, 2003; Ramsar

Convention Secretariat, 2006; World Meteorological Organisation, 2006; Maltby & Barker

19

2009). Wetlands are very fragmented and represented in all parts of the world (Figure 2.2.)

varying in flora and fauna with diverse ecological functions and anthropogenic uses.

Whilst the definition of wetlands is in controversy, its values and functions are undisputed.

When properly measured, the total economic value of a wetland's ecological functions,

services and resources may exceed the economic gains of converting the area to alternative

uses (Phillips, 1998; Turner et al., 2000). Challenges, however, arise in applying economic

tools such as cost benefit analysis to assess wetlands because of the non-availability of data,

especially, in developing countries (Phillips, 1998). In particular, economic analysis of the

environmental functions of tropical wetlands is underdeveloped. However, the

physical/hydrological and environmental services provided by wetlands around the world

have been well documented and can easily be assessed using various scientific methods. The

hydrological services of wetlands include its prevention of storm damage, flood and water

control (Barbier, 1994; Shutes et al., 1999; Haslam, 2003; World Meteorological

Organisation, 2006), coastal protection and aquifer recharge (Zedler & Leach, 1998), waste

management and improvement of water quality (Biney, 1990; Barbier, 1994; Green & Upton,

1994; Zedler & Leach, 1998; Shutes, 2001; Haslam, 2003; Ramsar Convention Secretariat,

2006; Cooper, 2009) and climate change mitigation (Joosten & Clarke, 2002; IPCC, 2007;

Ramsar Convention Secretariat, 2007). Also, lower soil horizons of wetlands support sulphate

reduction to sulphides and provide the environment for metal-sulphide precipitation by

trapping heavy metals in sediments (Maltby & Barker, 2009).

Wetlands are habitats of enormous biological productivity and high conservation value,

especially, for migratory birds. They typically have high species diversity and are habitats for

various stages in the life cycle of species (Barbier, 1994; Zedler & Leach, 1998; Haslam,

2003; Botkin & Beverage, 1997). Apart from benefit to animals, the socioeconomic

development of many cities can be traced to their association with wetlands through the

provision of potable water (Biney 1990; Barbier, 1994; Millennium Ecosystem Assessment,

2005; Ramsar Convention Secretariat, 2006), expression of cultural values, capture fisheries,

aquaculture and extraction of agricultural and horticultural products (Biney, 1990; Mitsch &

Gosselink, 1993; Barbier, 1994; Millennium Ecosystem Assessment, 2005). The use of

wetlands for recreation and tourism is being heavily exploited in England, the Northern

Territory of Australia, and Kashmir, northern Germany (Zedler & Leach, 1998; Millennium

Ecosystem Assessment, 2005).

20

Fi

gure

2.2

. Dis

trib

utio

n of

wet

land

s in

the

world

(M

illen

nium

Eco

syst

em A

sses

smen

t, 20

05)

21

There are increasing anthropogenic demands for water and other natural resources and

services provided by wetlands and these impact adversely on wetlands (Ramsar Convention

Secretariat, 2007). The capacity of wetlands to continue to provide these resources and

services is therefore declining. A major cause of this decline is the increasing abstraction of

water, especially, for agriculture (Barbier, 1994; Joosten & Clarke, 2002; Millennium

Ecosystem Assessment, 2005; Ramsar Convention Secretariat, 2007), port expansion,

industrialisation and urbanisation (World Meteorological Organisation, 2006). Many surface

and groundwater-dependent wetland systems and their catchments are already water-stressed,

and demand for water is projected to continue to increase (Millennium Ecosystem

Assessment, 2005). For example, access to freshwater is declining for 1-2 billion people

worldwide and this, in turn, negatively affects food production, human health, and economic

development, and has a potential to increase societal conflict (Ramsar Convention Secretariat,

2008).

Following the varied uses of wetlands described above and its consequent degradation, the

allocation of wetland resources requires careful analysis of their values, in particular, their

indirect use values of environmental regulatory functions. The failure to take these values into

account is often a major factor behind decisions that lead to overexploitation or excessive

degradation of wetland systems (Barbier, 1994). The Government of Ghana in celebrating the

World Wetlands Day, 2010, reiterated its commitment to review policies or amend the

relevant legislations that impede the protection of the country's wetlands to ensure long-term

ecological viability (GNA, 2010a). It is unfortunate however, that, this review, if it ever

happens, will not have a strong backing of empirical data, information and literature from the

Ghanaian context.

2.3 URBANISATION

2.3.1 Concept and History of Urbanisation The term urban areas are settlements or localities defined as "urban" by national statistical

agencies. The cut-off points for identifying urban areas or populations therefore vary from

country to country (Millennium Ecosystem Assessment, 2005). The most common ways of

classifying urban settlements are by population, economic conditions, and location. The

minimum population density criterion commonly ranges between 400 and 1,000 persons/km2

(Qadeer, 2004; Pacione, 2009). When population size is used, other economic or social

indices are described. The minimum population size ranges between 1,000 and 5,000

residents; and maximum agricultural employment is usually in the vicinity of 50–75 %

22

(UNFPA, 2007). In each case, however, cut-off points outside these ranges can easily be

found. For example, in Sweden, the urban population threshold is 200 inhabitants; 10,000 in

Benin, Switzerland and Burkina Faso; 20,000 in Nigeria and 30,000 in Japan (Pacione, 2009).

A settlement with a minimum of 5,000 persons is used to describe urban settings in Ghana

(Ghana Statistical Service, 2002a). Although country-specific definitions will remain central

to defining and assessing urban centres for many years to come, the basis for a more uniform

definition is emerging from work using remote sensing and geographical information systems

(Fedeski & Gwilliam, 2007). According to Pacione (2009), whatever the reasons are for

urbanisation, it is likely to be a thing of the past in the near future. This is because, with

globalisation and advances in telecommunication technologies, time becomes instantaneous

and the urban frame will be useless and spatial congregation becomes unnecessary. Man will

return to more ‘rural’ and isolated settlement types but still perform all urban functions.

Following these arguments, the definition of an urban setting is expected to become more

controversial in the future.

Also following recent spatial development trends, urban and suburban landuse form a

continuum as much urban land area is occupied by dense-to-moderate residential housing.

Thus, there are no clear physical boundaries between urban, suburban lands and rural lands, or

between the urban cores of the larger cities and suburbia (Ehrenfeld, 2000). However, to

differentiate economically, urban residents are less likely to work in agriculture and more

likely to work in industry or services. For example, in India, more than 75 % of adult male

population in a town has to be engaged in non-agricultural work to be considered urban

(Pacione, 2009).

As seen above, there is no international agreement on the defining characteristics of urban,

nor are there any scientifically accepted criteria by which to identify urban areas and

populations. Moreover, some urban researchers believe that the distinction between rural and

urban is becoming less relevant (Cohen, 2004) and that the boundary definitions are

inevitably somewhat arbitrary. Even though there are differences in what is urban, the concept

is well studied among sociologists and geographers. Statistically, urbanisation reflects an

increasing proportion of the population living in settlements defined as urban, primarily

through net rural to urban migration. The level of urbanisation is the percentage of the total

population living in towns and cities while the rate of urbanisation is the rate at which it

grows (Millennium Ecosystem Assessment, 2005; UNFPA, 2007).

23

2.3.2 Modern Trends in Urbanisation In the olden days, man was an agricultural being and most of the daily activities involved

procurement of food. With time, technological advances changed the face of agriculture

(Kaplan et al., 2009; Pacione, 2009). Higher productivity in agriculture resulted in food

surplus sufficient to feed non-farmers. This led to division of labour such that, other than

farming, people could become full time craftsmen, traders, soldiers, religious leaders,

politicians etc. Increased geographical mobility and diffusions of goods, technologies and

ideas resulted in concentration of populations. Rather than the earlier focus on agriculture,

trade hubs, political centres and massive public projects to improve human wellbeing started

to become important (Jacobs et al., 1997; Bhattacharya, 2002; Millennium Ecosystem

Assessment, 2005; Smith, 2005; Kaplan et al., 2009). The larger human settlements grew as

cultural centres of literacy, arts, religion and science. This then led to establishment of

political hubs of imperial power and later, economic domination of cities over rural areas

(Harvey, 1985a, b). Then came the period between 1750 and 1950 when most regions in

North America and Europe experienced industrialisation resulting in a wave of urbanisation to

produce the urban industrial societies. Since then, urbanisation and urban growth have

continued to be major demographic trends in all parts of the world (Millennium Ecosystem

Assessment, 2005). The increased urbanisation resulted in increase from only two cities

containing 10 million people in 1950 to seven by 1975 (World Bank, 1991).

Recent developments in less developed countries have led to a second wave of rapid

urbanisation where the number of urbanites will grow from 309 million in 1950 to 3.9 billion

in 2030 (UNFPA, 2007) with Africa being the most urbanising (Bhattacharya, 2002; Kaplan

et al., 2009). According to the Millennium Ecosystem Assessment (2005), decolonisation and

the development of independent nation states had heavy influences on urbanisation in Africa

and much of Asia as controls on the rights of the inhabitants to live in or move to urban

centres were dismantled, and in part because the building of the institutions for independent

governments increased urban employment. The UNFPA (2007) projects the urban population

of Africa and Asia to double between 2000 and 2030 and Grimm et al. (2008) estimates that

more than 95 % of the net increase in the global population will be in cities of the developing

world. Meanwhile, the urban population of the developed world is expected to grow relatively

little: from 870 million to 1.01 billion within the same time frame. At the global level

therefore, all future population growth will thus be in towns and cities. Already, in 2008 the

proportion of the world’s human population residing in urban areas reached 50 % as a result

of rapid urban growth in developing countries (United Nations, 2008; UN-HABITAT,

24

2008b). Urban places will therefore be of fundamental importance for the distribution of

population; in the organisation of economic production, distribution and exchange; in the

structuring of social reproduction and cultural life; and in the allocation and exercise of power

(Pacione, 2009).

It is also argued that the main driver of current urbanisation is economic growth. In general,

the most urbanised nations are those with the highest per capita incomes, and the nations with

the largest increases in urbanisation are those with the largest economic growth (World Bank,

1991; Millennium Ecosystem Assessment, 2005). The current rates of urbanisation in

developing countries could therefore have global economic implications. It can shape

prospects for global economic growth, poverty alleviation, population stabilisation and

environmental sustainability. Cities are already the locus of nearly all major economic, social,

demographic and environmental transformations. They cover only about 2 % of the earth’s

surface, yet consume 75 % of all resources and produce 75 % of all waste (UNFPA, 2007).

Therefore, improving urban planning to make cities more sustainable and increase the quality

of urban life will require integrated urban planning and management and cooperation among

responsible stakeholders (Berkowitz et al., 2003; United Nations Department of Economic

and Social Affairs, 2005). Such stakeholders, however, have different and often conflicting

priorities or insufficient resources to carry out their mandate. This is particularly so in

developing countries which do not have the financial and economic resources to meet the ever

increasing challenges of their uncontrollable urban expansion.

The growth of cities throughout the world presents new challenges to meet societal needs,

which are vulnerable to both climate and population changes. For example, urban water

withdrawal often extends far beyond city boundaries, as municipalities seek to secure new

water sources from the hinterlands (Jenerettea et al., 2006). Cities consume less than 10 % of

the total water used globally (Millennium Ecosystem Assessment, 2005). However, it is the

concentrated nature of that demand which poses a heavy burden on the limited local water

supplies. This is both in terms of volume of demand and pollution from wastewater discharge.

The high costs of collecting, treating and distributing clean drinking water, as well as the costs

of wastewater collection and treatment, are a major burden on public budgets and are beyond

the capacity of municipal or national governments in most developing countries like Ghana

(Millennium Ecosystem Assessment, 2005; Grimm et al., 2008). Any effort at improving

wetlands will therefore be a step in the right direction to reduce the immediate and future

water burden of these growing cities.

25

2.3.3 Urbanisation Trends in sub-Saharan Africa Urban population growth in sub-Saharan Africa is close to 5 % per annum and is the highest

growth rate in the world (Kaplan et al., 2009). Contrary to current urbanisation trends of

developed countries, the urbanisation of Sub-Saharan African countries can best be described

as low density urban sprawl. Urban sprawl, loosely defined as dispersed and inefficient urban

growth (Hasse & Lathrop, 2003; Mundia & Aniya, 2006) has undesirable effects. According

to Brueckner & Fansler (1983) and The World Bank (2000), the phenomenon of urban sprawl

is a symptom of an economic system gone awry. This dramatic expansion of settlements at

the expense of open spaces and natural resources has sparked intense interest and debate on

urban sprawl (Mumphrey & Whalen, 1977; Brueckner & Fansler, 1983; Mitsch & Gosselink,

2000; Hasse & Lathrop, 2003).

Cities in Africa are not serving as engines of growth and structural transformation. Instead,

they are part of the cause and a major symptom of the economic and social crises that have

evolved the continent (The World Bank, 2000). The predominance of agricultural and mineral

trade and the lack of manufacturing have made them more dependent on rural than on urban

economic activities and more susceptible to changes in commodity market prices (Kidane-

Mariam, 2003; Millennium Ecosystem Assessment, 2005). Furthermore, the gains in human

well-being are not distributed evenly among individuals or social groups or among countries

or the ecosystems of the world. Sub-Saharan African urbanisation is one mostly engineered

by deprivation of rural areas (The World Bank, 2000; 2007). In fact, a vast majority of rural-

urban migrants do so to escape the extreme deprivation of amenities. Also, according to

Kaplan et al. (2009), the combination of agrarian economy and high family sizes cause rural

landlessness and poverty making people seek opportunities in the city. There is therefore a net

increase in city dwellings; a lot of which are uncompleted and without access roads,

electricity, water and telephone facilities (The World Bank, 1991; Berkowitz et al., 2003;

Millennium Ecosystem Assessment, 2005; The World Bank, 2007). In these urban centres,

the gap between the advantaged and the disadvantaged is increasing (Baharoglu & Kessides,

2000; White et al., 2001; National Development Planning Commission, 2005; Ravallion et

al., 2007a, b).

According to the records, 7.8 % of the population of Ghana lived in urban centres in 1921. By

1960, urban population had risen to 23.1 % and further to 32 % in 1984 and 43.8 % in 2000

(Ghana Statistical Service, 2002a) and estimated to exceed 50 % in 2008 (UN-HABITAT,

2008). The rapid urbanisation puts enormous pressure on the utilisation of natural resources

26

and the provision of infrastructure and amenities like water supply, sanitation, waste

management, transportation, and housing. The adverse impact of rapid urbanisation has also

exacerbated the incidence of unplanned changes in landuse in the larger cities, especially

Accra and Kumasi (Brook & Dávila, 2000; The World Bank, 2007). Poor residents of these

cities tend to concentrate in the outskirts, in non-functional government structures and land

and along urban wetlands as informal settlements and slums.

2.4 URBANISATION AND THE ENVIRONMENT A significant literature exist on ecosystems within urban areas (The World Bank, 1991;

Zedler & Leach, 1998; Shutes et al., 1999; Rees, 2003; Janssen et al., 2005; Ravallion et al.,

2007a, b); ecology of urban areas, treating urban systems as ecosystems (Botkin & Beverage,

1997; Ehrenfeld, 2000; Keddy, 2000; Capello & Faggian, 2002; Millennium Ecosystem

Assessment, 2005; Berkowitz et al., 2003; Janssen et al., 2005; Secretariat of the Convention

on Biological Diversity, 2005; Ramsar Convention Secretariat, 2007), urban planning

(Mabogunje, 1990; Davey, 1993; Pauleit & Duhme, 2000; Abbaspour & Gharagozlou, 2005;

Obeng-Odoom, 2007), and impacts of urban areas on global environmental change e.g. heat

island effect (Berkowitz et al., 2003; Pongrácz et al., 2006; 2010; Huang et al., 2009;

Takebayashi & Moriyama, 2009), and pollution (Berkowitz et al., 2003; Hidalgo et al., 2008).

The diversity and trends of these studies reflect the themes of interest and concern of the

scientific community.

Urban areas have major influences on the environment in terms of concentration of

consumption and waste discharge, concentrated industrialisation and intensive use of energy and

resources. The relationship between urban economic growth and development and the ecological

effects of this growth is often insidious and manifest when they are almost irrepressible. Cities

have therefore been perceived as parasites on the environment and spitting out wastes to the

detriment of “natural” ecosystems (UNU-IAS Urban Ecosystems Management Group, 2004).

With the development and rise of the modern environmental movements in the 1960s and

1970s, it became fashionable to consider everything to do with cities as bad and everything to

do with wilderness as good. Cities were considered polluted, dirty, artificial, and lack wildlife; therefore, urban environments are bad. While the wilderness is unpolluted and full of wildlife

therefore, it is good (Botkin & Beverage, 1997). Even in recent times, Rees (2003) described

urban areas as the human equivalent of the livestock feedlot: a spatially limited area

characterized by a large population of humans living at a high density and supported by

biophysical processes mostly occurring somewhere else.

27

Cities may not be an ecological disaster, because they generate almost 80 % of the global

economic output (Pauleit & Duhme, 2000). According to the United Nations Department of

Economic and Social Affairs (2005), cities with high population densities use substantially

less land, energy and water per person than low-density cities or suburban or rural areas with

large and widely spaced individual houses. Urban agglomerations tend to have larger populations

and more likely to be the site of large facilities and higher-level administrative functions and

create comparatively densely settled areas involved in the manufacturing and service industry

(Berkowitz et al., 2003; Millennium Ecosystem Assessment, 2005; United Nations Department

of Economic and Social Affairs, 2005). More land is thereby freed for nature. Urbanisation is

therefore not in itself inherently bad for ecosystems.

In contrast, most rural areas are made up of large tracts of land that have been converted to

agricultural monocultures and individual houses widely spaced out in the landscape that do not

support wildlife. Dense urban settlements may therefore be less environmentally burdensome

than rural and suburban sprawl (Brueckner & Fansler, 1983; Capello & Faggian, 2002;

Millennium Ecosystem Assessment, 2005). Also, urban centres facilitate human access to and

management of ecosystem services through, for example, the scale and proximity economies

of piped water systems. Botkin & Beverage (1997) put it very well in their ‘Cities as

Environments’ publication that, urban implosion has dual benefits: improving the lives of most

people and helping to conserve rural land and wilderness. The environmental sustainability of a

city should therefore be in terms of the future health of the environment within the city. That

is, the environment in which the city’s inhabitants dwell is influenced, for better or worse, by

the city’s own growth and change (William, 1990; Fedeski & Gwilliam, 2007).

Urban settlements are agglomerations of people and their activities. Human activities in urban

environments therefore constitute a natural part of urban ecosystem dynamics. Collins et al.

(2000) described a number of conditions that shaped the urban ecosystem. First, compared with

rural ecosystems, urban vegetation ecosystems are highly patchy and the spatial patch structure

is characterized by a high point-to-point variation and degree of isolation between patches.

Second, disturbances such as fire and flooding are suppressed in urban areas, and human-induced

disturbances are more prevalent. Third, because of the higher temperatures in urban systems,

in temperate climates there are longer vegetation growth periods. Fourth, ecological

successions are altered, suppressed or truncated in urban green areas. The uniqueness of urban

vegetation ecosystems can therefore not be overemphasised. The urban spatial diversity and

opportunities are often harnessed in landscaping and aesthetics. Therefore, as people increasingly

28

live in cities, the cities act as both human ecosystem habitats and drivers of local ecosystem

change. The challenge therefore should be how to foster urban systems that contribute to human

well-being and reduce ecosystem service burdens at all scales (Millennium Ecosystem

Assessment, 2005).

Cities are increasingly being recognised as ecosystems by themselves (Botkin & Beverage,

1997; Ehrenfeld, 2000; Berkowitz et al., 2003). The ecosystem approach is used in the analysis

and implementation of the objectives of the Convention on Biological Diversity (CBD). The

principles amount to a strategy for the integrated or holistic management of urban resources

through modern scientific adaptive management practices (Secretariat of the Convention on

Biological Diversity, 2005). Following this Convention, the Global Partnership on Cities and

Biodiversity was formed to support cities in the sustainable management of their biodiversity; to

assist cities to implement practices that support national, regional and international strategies,

plans, and agendas on biodiversity, and to learn from existing initiatives. The Curitiba

Declaration on Cities and Biodiversity adopted in 2007 reaffirms the resolve of mayors of

cities to improving the lives of urban residents to meet the objectives of the Convention on

Biodiversity (Secretariat of the Convention on Biological Diversity, 2007). This has come about

because, historically planners and decision-makers tended to neglect ecosystem services and

relations between ecosystems and human well-being, at least until a local or international crisis

has forced such concerns onto the policy agenda. In order to grow sustainable cities, we need a

renewed emphasis on the positive aspects of urban environments. It is possible for those who are

interested in wildlands and those who are interested in urban environments to both achieve

their goals. This can be achieved by combining environmental scientific knowledge with

spatial planning.

2.5 WETLANDS AND URBANISATION The spread of urbanisation leads to wetland ecosystem fragmentation and exploitation

(Ramsar Convention Secretariat, 2007). Landuse changes and over-use of water leads to

wetland loss and degradation in many parts of the world and at rates faster than other ecosystems

(Millennium Ecosystem Assessment, 2005). This is jeopardising the current and future provision

of their services for human well-being. As the wetland area increasingly gets urbanised, the

environment and in particular, flora and fauna are affected (Tables 2.2 and 2.3). Looking at

these pressures on wetlands, society cannot afford a hands-off approach to management.

Rather, wetland managers must clearly state the goals of management towards maintaining

functions of the wetlands for a sustainable urban environment.

29

Table 2.2. Anthropogenic effects on urban wetlands (Zedler & Leach, 1998) Component Problem Hydrology altered hydroperiod due to dams, flood control channels, excess runoff,

decreased water quality due to contaminants and eutrophication; Habitat loss, fragmentation and degradation; Pests feral animals and increased predation on native animals; invasions of

aggressive plants and negative impacts on native vegetation; Infrastructure Maintenance or dredging for shipping canals, expansion of marinas, filling

for highway expansion, replacement or installation of utilities; Visitors pressure for recreation (e.g., bird-watching, canoeing, bike trails) and

vandalism;

Table 2.3. Likely effects of urbanisation on wetland ecology (Ehrenfeld, 2000) Organism Effect Flora: large numbers of exotic species present; large and numerous sources for

continuous re-invasion of exotics; large amounts of land with recently disturbed soils suitable for weedy,

invasive species; depauperate species pool; restricted pool of pollinators and fruit dispersers; chemical changes and physical impediments to growth associated with the

presence of trash; small remnant patches of habitat not connected to other natural vegetation; human-enhanced dispersal of some species; trampling along wetland edges and periodically unflooded areas; Fauna: species with small home ranges, high reproductive rates, high dispersal rates

favoured; large predators virtually non-existent; “edge” species favoured over forest-interior species; absence of upland habitat adjacent to wetlands; absence of wetland/upland ecotones; human presence disruptive of normal behaviours.

Because of these disturbances, wetlands have taxonomic groups with the highest proportion of

threatened species (Millennium Ecosystem Assessment, 2005). According to Williams

(1990), tropical wetland vegetation species tend to have narrow ecological requirements, with

habitats that are accordingly highly fragile. This is even worst considering the rate of

urbanisation and limited capacities of environmental agencies to monitor and control pollution

in the developing countries. The driving force behind the establishment of the Ramsar

Convention was the continuing destruction of wetlands and the impact of this destruction on

populations of waterbirds (Ramsar Convention Secretariat, 2007). Yet, more than four decades

30

later, the “degradation and loss of wetlands is continuing more rapidly than for other ecosystems”

(Millennium Ecosystem Assessment, 2005). The declining condition of inland waters is putting

the services derived from these ecosystems at risk. The increase in pollution and degradation

of wetlands has reduced the capacity of these wetlands to filter and assimilate waste. It is

established that inland water ecosystems are in worse condition overall than any other broad

ecosystem type (Millennium Ecosystem Assessment, 2005; Ramsar Convention Secretariat,

2007).

Urban land has competing uses for man. The qualities that make a location good for a city

may often also make it an important location for biological conservation. For example, river

mouths, confluences, riparian areas, deltas are good sites for cities because of the access to

transportation, leisure and water for all purposes. However, these locations are modified and

polluted after settlement with the ecosystem cleaning services taking place mostly beyond

urban boundaries. The water requirements of aquatic ecosystems adjacent the expanding

human settlement and freshwater demands increase pressure on the ecosystem. This leads to

changes in flow regime (Reinelt et al., 1998), transport of sediments and chemical pollutants,

modification of habitat, and disruption of migration routes of aquatic biota (Millennium

Ecosystem Assessment, 2005). Landuse change also affects evapotranspiration, infiltration

rates, and runoff quantity and timing (Owen, 1999; Berkowitz et al., 2003; Fedeski &

Gwilliam, 2007; Roy, 2009). Changes of particular importance for human well-being are the

contrasting reductions in the overall quality and quantity of available runoff versus

concentrated peaks of runoff causing flooding. Also, the urban rainwater with absorbed

pollutants from the air falls on road surfaces, parking lots, vehicles, corrosive railings, brick,

and copper roofs thereby resuspending pollutants in the runoff. Wetland plants attenuate the

stormwater flow, bind sediment, and directly accumulate metals (Shutes et al., 1999; Shutes,

2001; Fritioff, 2005). Vegetation also influences urban environmental conditions and energy

fluxes by selective reflection and absorption of solar radiation and may also influence air

quality and human health (Small, 2001). Information on the abundance, quality and spatial

distribution of urban vegetation is therefore very vital for designing systems for the provision

of ecosystem services, park and natural area management and urban planning.

Urbanisation is a relatively permanent landuse change. In general, the first order hydrologic

impacts associated with urbanisation result from increasing impervious area (Choi et al.,

2003; Reinelt et al., 1998; Brody et al., 2007). The hydrologic changes are associated with

changes from major classes of landcover and landuse e.g. from forest to urban system (as the

31

case is in Kumasi). This leads to decreased infiltration and an increase in runoff with the

associated increase in pollution loads, decrease in temporal reliability of environmental systems

and disturbance of wetland biota (Reinelt et al., 1998; Zedler & Leach, 1998). Reduced

infiltration can also lead to longer lifespan of pools with stagnant water, thus providing

increased breeding opportunities for mosquitoes (Millennium Ecosystem Assessment, 2005;

Ramsar Convention Secretariat, 2007). Urban wetlands are also the destination for

contaminants from roads, gutters, storm drains, and occasional wastewater spills.

Whilst the conservation of ecosystems services is undisputedly good for mankind, the urban

dweller rarely plays a significant part in conserving it. There is continued lack of

understanding and appreciation of the complex processes involved because, many of these

services take place at some distance from the urban consumers (Millennium Ecosystem

Assessment, 2005). Also, the effect of the urban dweller’s actions and inactions and the

control and management of these ecosystems go beyond the city boundaries. It is unfortunate

however, that, the people most adversely affected by the loss of ecosystem services tend to be

the least influential economically and politically (such as the urban poor, future generations,

and residents living far from where the decisions are being made) (Millennium Ecosystem

Assessment, 2005).

2.5.1 Urban Wetlands Vegetation Plant species found in wetlands are considered as wetland vegetation. Because migration of

oxygen in waterlogged soils and into root zones is slow and proceeds at an unacceptable or

intolerable rate for most organisms, these areas are often populated by plants adapted to poor

oxygen conditions in the root zone (Lyon, 1993; Haslam, 2003). The major factors believed to

influence wetland biodiversity include hydrology, geomorphology, substratum, chemistry,

adjacent terrestrial system, grazing, anthropogenic influences and individual plant tolerance or

response to the stresses of flood and drought (Harper et al., 1995; Owen, 1999; Casanova &

Brock, 2000; Fortney et al., 2004; Maltby & Barker, 2009; Miller & Fujii, 2010). Conversely,

wetland species play a key role in the water quality, structure and functioning of stream

ecosystems (Barrat-Segretain et al., 1998; Shutes, 2001; World Meteorological Organisation,

2006; Sassa et al., 2010). Streamside vegetation also controls the supplies of allochthonous

and autochthonous organic matter, filters the nutrients and pollutants, helps stabilise banks

and influences the water temperature and light through shading (Zedler & Leach, 1998;

Shutes, 2001; Baarta et al., 2010).

32

Urban wetlands often have unusual physical features and conditions which are the result of

human interventions. These include disturbed soil profiles e.g. through agricultural activities,

mounds of soil from old ditch-digging or channel-widening activities, heterogeneous

materials e.g. from waste dumping or stream deposition, hard concrete surfaces etc. (Reinelt

et al., 1998; Ehrenfeld, 2000; Millennium Ecosystem Assessment, 2005). These forms of

disturbances create spatial and temporal heterogeneity in wetland ecosystems and act as

important mechanisms in maintaining species richness (Pickett et al., 1989; Chaneton &

Facelli, 1991; Barrat-Segretain et al., 1998; Fortney et al., 2004; Ot’ahel’ová et al., 2007;

Grimm et al., 2008). Biological diversity in an urban setting could therefore be a direct

reflection of the spatial diversity provided by the urban landscape.

Apart from the edaphic and hydrologic conditions, landuse is also said to affect the

vegetation. In a study on the watersheds of Wisconsin lakes, U.S.A., agricultural development

explained more of the variation in macrophyte community species richness and abundance

than did urban landuse and that agriculture and urban land uses differ as causal agents (Sassa et

al., 2010). That is, urbanisation and agricultural developments create different perturbation

gradients that appear to affect aquatic plant species differently. With these diverse ranges of

influencing factors and consequences, attempts at wetland restoration have therefore been a

great challenge for many wetland ecologists because of the different and sometimes combined

and synergistic effects of causal agents of degradation. For example, Baarta et al. (2010)

attempted to predict macrophyte development in response to restoration measures. Modelling

the status quo showed that completely isolated water bodies are over grown with macrophytes

resulting in an intensified terrestrialisation process. In contrast, complete reconnection to the

main channel caused a loss of macrophyte abundance and species in most aquatic habitats.

2.5.2 Wetlands, Urbanisation and Agriculture Urban agriculture is defined as an enterprise located within or on the fringes of a town, a city or a

metropolis, which grows or raises, processes and distributes a diversity of food and non-food

products, (re-)using largely human and material resources, products and services found in and

around that urban area, and in turn supplying human and material resources, products and

services largely to that same urban area (Mougeot, 1999; FAO, 2007). According to

Egziabher et al. (1994), urban farming is a basic urban function as ancient as cities

themselves. Asia is leading the way in the sector, with highly complex and efficient systems

for the production and marketing of agricultural products.

33

The idea of urban agriculture got some boost as city planners envisaged an idea of a

sustainable city and tried to achieve it through spatial planning that included agriculture and

wild areas. For example, Sir Ebenezer Howard (1850-1928) in “Garden Cities of Tomorrow”

describes a utopian city in which man lives harmoniously with the rest of nature. His ‘park town’

or ‘garden city’ concepts describe a city that is surrounded and interconnected by greenbelts,

forming a system of countryside and urban landscapes. His idea was to combat urban

overcrowding by the establishment of garden cities of about 30,000 inhabitants, self-contained

in terms of employment, food production, recreation etc. (Mumford, 1961). With such a

concept, variety of habitats ranging from fairly natural ones to highly modified ones are

therefore maintained (Niemelä, 1999). In developing countries’ literature however, urban

agriculture or green areas are more of an evolving economic activity than a result of spatial

planning (Armar-Klemesu & Maxwell, 1999; Egziabher et al., 2004; Faithful & Finlayson,

2005; Erenstein et al., 2006; Foeken, 2006; Foeken & Owuor, 2008; Magigi, 2008) and

governments are introducing institutional and other policy changes that recognize or tolerate,

manage, and promote the activity.

Urban agriculture may be practiced for food (Egziabher et al., 1994; Armar-Klemesu &

Maxwell, 1999; Mougeot, 1999), as a livelihood and asset strategy (Armar-Klemesu & Maxwell,

1999; Erenstein et al., 2006; Foeken & Owuor, 2008), income (Foeken, 2006) and has the

potential to significantly contribute to the city’s sanitation needs (Lydecker & Drechsel,

2010). It is also known that, an increasing proportion of the target group for poverty

alleviation in developing countries is now found in urban areas (Adams et al., 2004; National

Development Planning Commission, 2005; Ravallion et al., 2007a, b). Many of these urban

poor find employment or improve the quality of their daily diet through agriculture in the

open urban spaces. In times of economic crisis, urban agriculture sustains livelihood of most

people and assists with resolving or coping with diverse development challenges.

Urban agriculture is spurred by a complex web of factors still to be understood, not the least

of which are urban poverty and food insecurity (Mougeot, 1999; Adams et al., 2004;

Ravallion et al., 2007b). Even though Pacione (2009) considers the urban populace to be less

employed in agriculture, the 1988 census ranked urban agriculture as Dar es Salaam’s second

largest employer, after small traders and labourers. Urban agriculture occupied 11 % of the

population aged 10 or more (Egziabher et al., 1994). However, few papers from literature

overtly condemn urban agriculture under any form. All near opposition papers are concerned

with the issue of urban planning (Magigi, 2008), public health (Afrane et al., 2004; Matthys et

34

al., 2006; Obuobie et al., 2006; Klinkenberg et al., 2008) and the environment (Adam et al.,

1999; Keraita et al., 2003; Faithful & Finlayson, 2005; Erenstein et al., 2006; Mireri et al.,

2007).

According to Magigi (2008), urban agriculture is a big economic activity in Africa because of

the unguided land development, poor policy and legislative enforcement, out-dated policy

inherited from colonial period and inadequate community involvement in spatial landuse

planning processes. This assertion may be true because urban land values are usually very high

and such lands cannot be used for agriculture if they are not originally part of the urban planning

scheme. In Ghana, urban landuse plans do not include allocation of land for agriculture.

Therefore, lands used for urban agriculture belongs to either government institutions or private

developers or are open areas designated as wetlands along streams (Adam et al., 1999; Brook

& Dávila, 2000; Oboubie et al., 2006). Due to the hilly landscape of Kumasi, most streams

run through inland valleys unsuitable for construction and are therefore prime sites for

cultivation of vegetables, sugar cane and taro (Figure 2.3).

Historically, agriculture has been the main cause of wetland loss (Egziabher et al., 1994;

FAO, 1998; Eppinka et al., 2004; Millennium Ecosystem Assessment, 2005; Ramsar

Convention Secretariat, 2006). This is through direct pollution by fertilizers, water abstraction

for irrigation, drainage of excessively waterlogged places or conversion to paddy farms, and

mining of peatlands, to mention a few. Whilst wetland conversion to agriculture is a common

phenomenon, the permanent wetland loss from agriculture to urban infrastructure or housing

through the urbanisation process has not been given the needed attention (Mathias & Moyle,

1992; Thibault & Zippererb, 1994). These shifts are very much evident in peri-urban areas of

most developing country cities. However, the livelihood of most peri-urban communities

depends to a large extent on the productivity of the natural systems, in particular wetlands

(Brook & Dávila, 2000; Keraita et al., 2003). Therefore, to safeguard and possibly enhance

livelihoods of many urban and peri-urban communities who subsist on wetlands, it is

imperative that the benefits of the natural wetland ecosystems be recognised when planning

and implementing development projects (Silvius et al., 2000; Grimm et al., 2008).

35

Figure 2.3. Major vegetable farming areas in the Kumasi Metropolis (Oboubie et al., 2006)

2.5.3 Wetlands, Urbanisation and Floods The urban development landscape in most developing countries ranges from rich

neighbourhoods with infrastructure to poor slum and or informal settlements. These poor city

neighbourhoods are more prone to flooding and other natural disasters than wealthy

neighbourhoods. These unfavourable initial conditions and disasters then serve as both drivers

and consequences of development patterns (Millennium Ecosystem Assessment, 2005;

Adelekan, 2010). Without adequate spatial planning, governance, regulation, and public

spending, increasing numbers of the urban population are living in wetlands and other

ecologically sensitive areas. This development pattern is making some suburbs of a large

number of African cities vulnerable to natural disasters (Adelekan, 2010). It is predicted that

future flood damages will depend greatly on settlement patterns, landuse decisions, quality of

flood forecasting, warning and response systems (ActionAid International, 2006; Millennium

Ecosystem Assessment, 2005; Adelekan, 2010).

Flooding is a natural phenomenon (ActionAid International, 2006). Very often in nature,

watersheds get flooded. This natural phenomenon has however become problematic because

man is increasingly drawing nearer to these natural systems. The watershed flood frequency

36

and scale has therefore been intensified by human activities. With improved life styles,

industrialisation and urbanisation, watersheds are constantly being inundated with expensive

capital goods and infrastructure. The costs of damage due to floods have therefore continually

risen (United Nations Inter-Agency Secretariat of the International Strategy for Disaster

Reduction, 2005; World Meteorological Organisation, 2006; McClean, 2010). Flood research

has, therefore, intensified and different methods have been applied in urban flooding in

different places with different price tags.

For example, in 1951, a flood along the main stem of the Kansas River was estimated to have

affected mostly urban areas because about 90 % (approximately US $479,000,000) of the

damage experienced occurred in urban areas. This great loss was used to justify the

construction of a stair-step series of lakes in the Kansas River Basin to act as storm water

collection sinks (Koilmorgen, 1953). Since then, the flood control plans for the Kansas River

Basin alone have been so extended and elaborated that they bear a price tag of billions of

dollars. More shattering than the rising costs are the rising number of acres of land that is to

be flooded by building more and more dams. These flood control mechanisms contravene

both the spirit and purpose of resource conservation. Another good intention with a wrong

approach was the building of a series of levées and dredging of channels as a control method

of the December 1955 flooding of Santa Cruz. This only resulted in extreme siltation and loss

of control of the river. Another flood of Santa Cruz in January 1982 proved that the solution

to flooding in urban areas needs a more integrated approach than artificial embankments

(Griggs & Paris, 1982). In practice therefore, floods can never be eliminated entirely.

However, the vulnerability to floods can be mitigated by integration of physical developments

with behaviour and actions especially by residents in flood prone areas (United Nations Inter-

Agency Secretariat of the International Strategy for Disaster Reduction, 2005; APFM, 2006;

Egyir, 2008; Akwei, 2010; McClean, 2010).

The scale and frequency of floods has been on the increase in some suburbs of Kumasi,

Ghana. For instance, residents of Aboabo and Atonsu all in the Aboabo Basin are frequent

victims of floods (Egyir, 2008; Akwei, 2010). However, most of the wetland research

conducted in Kumasi has been on vegetable farming and water quality in river basins of

Kumasi (Keraita et al., 2003; Obuobie et al., 2006; Keraita et al., 2008a, b; Lydecker, &

Drechsel, 2010). Egyir (2008) and Akwei (2010) worked on physical flood control and coping

strategies to flooding in the Aboabo Basin. The watershed hydrology and causes of annual

floods have not been determined by any research but public opinion attributes the floods to

37

obstructed water ways and climate change (Akwei, 2010; Ghana News Agency, 2010). How

much and which of the annual floods is attributable to each of these factors is not known.

According to Bhaskar & Singh (1988), Odemerho (1993), and the United Nations Inter-

Agency Secretariat of the International Strategy for Disaster Reduction (2005), the problems

of floods in third World cities are compounded by the rapid rate of urban expansion into

urban fringe areas often unaccompanied by the development of storm drainage systems.

The effectiveness of wetlands for flood abatement may vary depending on the size of the

wetland, type and condition of vegetation (Ot’ahel’ová et al., 2007), slope, and location of the

wetland in the flood path and the saturation of wetland soils before flooding (Bhaskar &

Singh, 1988). Urbanisation destroys these qualities of wetlands and aggravates flooding

because large parts of the ground are covered with roofs, roads and pavements and building

drains. As a result, even quite moderate storms produce high overland flows quickly filling up

river channels (Griggs & Paris, 1982; United Nations Inter-Agency Secretariat of the

International Strategy for Disaster Reduction, 2005; McClean, 2010). Therefore with

increased climate and landscape variability, most floods are getting out of hand and cannot be

predicted accurately.

Early flood control and protection approaches have been based on structural solutions like

embankments, bypass channels, dams, reservoirs, storm drains etc. This approach resulted in

changes in flow regimes, separation of rivers from flood plains and in most cases increased

flood regimes (Koilmorgen, 1953; Griggs & Paris, 1982). There is now a shift from flood

control to flood management (Integrated Flood Management) with the recognition that floods

are a natural river phenomenon (World Meteorological Organisation, 2006; Adelekan, 2010).

Flood management is more holistic and carried out through a public policy framework. This

integrated flood management approach is, however, not developed in Ghana. Food

management is reactionary through the National Disaster Management Organisation

(NADMO) in collaboration with the District Assemblies. In most places, mitigation measures

depend on resources available to the District Assembly and political expediency.

2.6 URBAN POVERTY AND WETLANDS

2.6.1 Urban Poverty in Africa The post-independence experience of African countries in environmental management has

been characterized by a series of overlapping crises. Most countries are experiencing very

rapid urbanisation, environmental degradation, and increasing poverty (Kidane-Mariam,

38

2003) and a shift or relocation of poverty to the urban areas (IPCC, 2007). These urban poor

are usually easy to locate in the urban landscape. According to the UN-HABITAT (1996),

lack of access to land by these poor, facilitates the development and expansion of slum and

squatter communities within urban areas. This situation is not only typical of Africa. For

example, at end of World War II, cities housed less than a third of the poor in the United

States and this had risen to more than half by the end of the century and these poor people are

spatially concentrated in extremely poor urban neighbourhoods (Kaplan et al., 2009). This

means a relocation of population, poverty and environmental pressures to urban areas. The

challenge now is for a concerted and synergised approach to dealing with this urban poverty-

environmental nexus.

The link between poverty and environment is complex and goes in both directions. The poor

depend heavily on a number of environmental services for their livelihoods and are very

vulnerable to the loss or degradation of environmental conditions (Adams et al., 2004). At the

same time, poverty and unequal access to resources often compel people to use available

natural and environmental resources unsustainably. For these reasons, Griebenow & Kishore

(2009) suggested that strategies to address poverty needed to address environmental problems

as well, and conversely strategies to improve the environment needed to address poverty.

However, in these already financially depraved slum or informal settlements, the lack of

appropriate institutional structures drives ecological and environmental trends towards the

negative. The condition of the environment and human settlements in the urban areas of

developing countries has therefore drawn the attention of the academic community,

international development agencies, regional intergovernmental organisations, nations, and

civic societies at various geographical scales and is well captured in Goal 7 of the MDGs.

Almost all African countries have therefore embarked upon the preparation and

implementation of broad and specific policies and strategies of environmental management

towards meeting this goal and to eliminate urban slums and reduce poverty (African Peer

Review Mechanism Secretariat, 2005; National Development Planning Commission, 2005).

The complex linkages between poverty and environment are therefore supposed to be

mainstreamed in each country’s poverty reduction strategy (Griebenow & Kishore, 2009).

2.6.2 Urban Poverty and the Environmental Kuznets Curve The cause-effect relationship between environmental degradation, poverty and disasters is

complex and has been the subject of many analyses (Beckerman, 1992; Kidane-Mariam,

2003; Stern, 2004; Sánchez-Rodríguez et al., 2005; Constantini & Monni, 2008; International

39

Recovery Platform, 2009). It is widely believed that poverty and other economic conditions

determine the priority placed on environmental issues (Killeen & Khan, 2002; Rietbergen et

al., 2002; Esty et al., 2005; Gunter et al., 2005; Rahman, 2006). That poor people directly

extract environmental resources thereby destroying habitats. However, many poor people live

closely with natural resources and rely heavily on low-input production systems gives. They

therefore have the motivation and advantage in the development of new environmentally

friendly products and services that improves their lives (Rietbergen et al., 2002).

Ecologists wary of the effects of increasing wealth on the environment, arguing that,

economic growth and conservation are incompatible goals (Czech, 2003). Economists, on the

other hand, expect economic growth to be a cure for global environmental challenges; that

wealthier countries have the luxury of investing more to conserve and protect ecosystems

(Cropper & Griffiths, 1994; Adams et al., 2004), fundamental aspects of human well-being,

and the quality of life (Mills & Waite, 2009). According to Costantini & Monni (2008) and

the Secretariat of the Convention on Biological Diversity (2007) environmental resources can

be sustainably harnessed to play an active role in reducing poverty, achieving higher living

standards and increasing human development levels. The difficulty, however, is to define that

optimum or ideal use of the environment which would be sustainable and at the same time,

achieve the economic and developmental goals envisaged by policy makers to reduce poverty.

Environmental degradation tends to increase as structure of the economy changes from rural

to urban or agricultural to industrial, but it starts to fall with another structural change from

energy intensive industry to services and knowledge based technology-intensive industry

(Munasinghe, 1999; Magnani, 2001; Panayotou, 2003; Dinda, 2004; Costantini & Monni,

2008; Wagner, 2008). At the latter stage environmental degradation starts to decrease because

people tend to make preventive environmental expenditures, donations to environmental

organisations or choose less environmentally damaging products. Thus, rich people value the

clean environment and make conscious efforts to preserve it. This systematic relationship

between income and change in environmental quality forms an inverted-U-shaped curve

called the Environmental Kuznets Curve (EKC) (Dinda, 2004).

The turning points of EKCs vary for different pollutants, environmental indicators, variables

used in the model, country, and the policy environment (Magnani, 2000, 2001; Stern, 2004).

For most of the pollution indicators, the estimated turning point lies within the income range

of 3,137 (Panayotou, 1993) to US$ 908,178 (Stern and Common, 2001). No studies have been

published on EKC and wetlands to know the estimated turning point. The closest study is that

40

of McPherson & Nieswiadomy (2005) on threatened species with the turning point between

10,000 and US $15,000 (US $1995 purchasing power parity) per capita income.

Even though the EKC has shown that income–environmental degradation relationship exists,

there is still no consensus on the trajectory (Panayotou, 2003; Barbier, 2004; Dinda, 2004;

Wagner, 2008). For example, Munasinghe (1999) suggests that, developing countries could

take alternative routes that do not reach high levels of environmental degradation by

identifying and addressing the economic imperfections that might exacerbate growth-related

environmental harm. That is, higher levels of development could be attained at lower

environmental cost. Also, the relationship between income and environmental care depends

on the distribution of benefits and costs of polluting activities (Magnani, 2001). The driving

force for flattening of the EKC may therefore not depend on incomes but distribution of costs

and benefits of pollution and the policy interventions that are put in place to protect the poor

and weak from bearing the burden of pollution.

As human populations increase, the marginal per capita value of wetlands first increases

because of increased wetland scarcity per capita. Thus a higher human population implies that

there is less wetland per capita available and therefore, on the average, fewer but more

precious wetlands. But the marginal value of wetlands increases with human development

only to a point because wetland functions begin to be lost through the urbanisation process

(Mitsch & Gosselink, 2000). A planned urbanisation programme and good policy

environment might decrease encroachment on wetlands and other natural areas thereby

securing environmental quality. Therefore Panayotou (1997) and Eppinka et al. (2004) are of

the view that, rather than economic growth, a more secured property rights, better

enforcement and effective environmental regulations will help ‘flatten’ the EKC. Property

rights will reduce the externalities associated with the use of wetlands as environmental

goods.

2.7 CLIMATE CHANGE, URBANISATION AND WETLANDS A major non-climate attribute affecting the ecosystem and its vulnerability to climate is the

increasing urbanisation of the developing world. Close to half of the world’s population now

lives in urban areas. Increasing urbanisation means that the impacts of climate change on

human settlements, if they occur, will increasingly affect urban populations (IPCC, 2007;

Adelekan, 2010). A significant part of the activities leading to the achievement of Goal 7 of

the MDGs (environmental sustainability) will therefore be played out in the urban

environment (Secretariat of the Convention on Biological Diversity, 2007). The hazards to

41

which some cities are exposed are likely to become either more severe or more frequent, and

so the increased risk should be anticipated, if not predicted (Fedeski & Gwilliam, 2007; IPCC,

2007). According to Niasse et al. (2004), the physical and socio-economic characteristics of

West Africa predispose it to be among the most vulnerable regions to climate change

worldwide.

A changing climate can modify all elements of the water cycle: change both the frequency

and intensity of precipitation, evapotranspiration, soil moisture, groundwater recharge,

snowmelt and therefore runoff (Fedeski & Gwilliam, 2007). It is therefore believed that the

most widespread potential impact of climate change on human settlements is flooding

(Fedeski & Gwilliam, 2007; IPCC, 2007; International Recovery Platform, 2009; McClean,

2010). Riverine and coastal settlements are believed to be particularly at risk, but urban

flooding could be a problem anywhere storm drains, water supply, and waste management

systems are not designed with enough system capacity or sophistication to avoid being

overwhelmed (IPCC, 2007). Also, there is likely to be increased cost of losses due to flooding

because of the abundance of more precious and capital goods and services in urban areas

(McClean, 2010). Urbanisation itself explains much of the increase in runoff relative to

precipitation in settled areas and contributes to flood situations (IPCC, 2007). It is estimated

that, by the year 2050, more than 70 % (and 90 % by the 2080s) of people in settlements that

potentially would be flooded are likely to be located in a few regions: west Africa, east

Africa, the southern Mediterranean, south Asia, and southeast Asia (Niasse et al., 2004;

Millennium Ecosystem Assessment, 2005; IPCC, 2007).

Due to climate change and urbanisation, large and growing numbers of people are at high risk

of adverse ecosystem changes, a worsening trend of human suffering and economic losses

from natural disasters (Nelson et al., 2009). There is also increasing demand for ecosystem

services because of increasing consumption per person, multiplied by a growing human

population (Millennium Ecosystem Assessment, 2005). Effects of climate change will be felt

most directly through changes in the distribution and availability of water, thereby increasing

pressures on the health of wetlands. Restoring wetlands and maintaining hydrological cycles

is of utmost importance for addressing climate change, flood mitigation, water supply, food

provision and biodiversity conservation (Ramsar Convention Secretariat, 2007; IPCC, 2007).

According to the Ramsar Convention Secretariat (2008), the management of these wetlands

should be supported by indigenous and traditional knowledge, recognition of cultural

42

identities associated with wetlands, stewardship promoted by economic incentives, and

diversification of the support base for livelihoods.

Wetlands are often biodiversity hotspots and deliver a wide range of ecosystem services

relating to climate change mitigation (Berkowitz et al., 2003; Janssen et al., 2005; Nelson et

al., 2009). Many types of wetlands play important roles in sequestering and storing carbon

(Joosten & Clarke, 2002). However, this is not yet fully recognized in national and

international climate change response strategies, processes, and mechanisms (Ramsar

Convention Secretariat, 2007). Because of the high amounts of carbon stored, wetlands are

particularly vulnerable to climate change impacts and human disturbances of wetland systems

can cause huge emissions of stored carbon. Therefore, recognising the potentially serious

implications of climate change on wetlands, Contracting Parties to the Ramsar Convention

have been called to manage their wetlands in such a way as to increase their resilience to

climate change and extreme climatic events and to ensure that climate change responses such

as revegetation, forest management, afforestation and reforestation and their implementation

do not lead to serious damage to the ecological character of wetlands (Ramsar Convention

Secretariat, 2007).

Despite these emerging international developments on climate change and urbanisation, cities

have the power to establish local environmental policies and regulations, such as landscaping

guidelines and pollution/emissions standards to mitigate the effects of climate change. They

can also integrate biodiversity considerations into other local policies, landuse planning, and

the development and operation of infrastructure, such as transportation and water

management systems (Secretariat of the Convention on Biological Diversity, 2007). Even

with these opportunities, urban areas are still understudied in the analysis of global

environmental change, with a majority of research placing emphasis on the contributions of

emissions of greenhouse gases and the heat island effect to global climate change (Hidalgo et

al., 2008; Takebayashi & Moriyama, 2009). Much less attention has been devoted to the study

of the impacts of global environmental change on urban areas and the people who live in

them. Particularly critical are the conditions in poor countries (Sánchez-Rodríguez et al.,

2005).

2.8 REMOTE SENSING AND URBAN LANDSCAPE PLANNING AND MANAGEMENT The effects of the gradual global dominance of urban settlements extend well beyond the

urban ecosystem. Therefore, as natural areas shrink and fragment through the urbanisation

43

process, humanity needs to conserve biological diversity and ecological integrity of urban

environments for a sustainable future. Meeting the challenge of conserving regional

ecological integrity in urban and urbanizing landscapes will depend on effective urban

management and spatial planning. However, the fragmented nature, isolated and frequently

disturbed conditions of natural areas exacerbate the difficulties inherent in spatial planning and

urban landuse (Mazzotti & Morgenstern, 1997). This is because the distribution of land, water and

forest resources on the earth’s surface is rarely equitable. Therefore the processes of human

ecosystem differentiation, resource allocation and settlement have frequently had spatial

dimensions that are usually characterised by patterns of territoriality and spatial heterogeneity

on many scales (Berkowitz et al., 2003). In managing these spatially heterogeneous and

scarce resources, there is the need for frequent inventory for monitoring and evaluation to

inform policy and reduce conflicts. Large-scale ground-based monitoring of city space,

forests, agricultural lands, wetlands, etc. are however, time-consuming, labour-intensive and

are typically constrained by poor accessibility (Lyon, 1993; Smith, 1993; Vincent & Saatchi,

1999).

In recent times, urban landcover provided by satellite and airborne sensors is an important

complement to in situ measurements of physical, environmental and socioeconomic variables

(Small, 2001; Small & Lu, 2006). Satellite-based earth observation offers the possibility to

provide synoptic measurements of land cover over very large geographic areas and over long

periods of time (Brinson, 1993; Lyon, 1993; Xu et al., 2000; Harvey et al., 2001; Ozesmi &

Bauer, 2002; DeKeyser et al., 2003; Mundia & Aniya, 2006; Small & Lu, 2006; Islam et al.,

2008). The development and application of these technologies has been progressing steadily.

Remote sensing and geographic information system (GIS) technologies have been applied in

wetland studies (Thenkabail & Nolte, 1996; Lunetta et al., 1999; Töyrä et al., 1999;

Thenkabail et al., 2000; Harvey & Hill, 2001; Ozesmi & Bauer, 2002; Choi, 2003), urban

areas (Smith, 1993; Xu et al., 2000; Pongrácz et al., 2006; Altobelli et al., 2007), and spatial

planning (Pauleit & Duhme, 2000; Abbaspour & Gharagozlou, 2005).

The application of remotely sensed observations to studies of the urban wetland vegetation

has, however, been rather limited. This is because data of specific spectral and spatial

resolutions for the discrimination of vegetation types are expensive and hard to come by. For

these reasons, very few remote sensing studies have been carried out in tropical wetland

vegetation in Africa specifically Ghana (Thenkabail & Nolte, 1996; Thenkabail et al., 2000;

Mundia & Aniya, 2006). Whilst Landsat data may have its shortcomings for studies of the

44

built urban environment, it may be sufficient to detect significant spatial and temporal

variations in urban vegetation (Small, 2001). This research therefore forms one of the early

studies on urban wetland vegetation changes in Kumasi, Ghana through the application of

remote sensing technologies.

Up-to-date landuse inventory is mandatory when making plans for solving problems

associated with the rapidly growing areas to make them more sustainable and increase the

quality of urban life (United Nations Department of Economic and Social Affairs, 2005). To

meet these urban spatial planning needs, the Buffer Zone Policy of Ghana advocates for a

buffer of all natural landscapes and waterways in urban areas and integration of natural areas

into urban planning (Water Resources Commission, 2008). This policy is belated and most of

the buffer areas have already been encroached upon. To curb encroachments on natural areas, the

participatory approach as has been applied in city planning in Brazil could be used in Ghana

(United Nations Department of Economic and Social Affairs, 2005). In doing this however, it

is recommended that stakeholders give full recognition to the significance of wetlands in

spatial planning so that the values they represent can properly inform landuse and investment

priority-setting and the adoption of necessary safeguards (The Secretariat of the Ramsar

Convention, 2007). The decentralised administration in Ghana which devolved spatial

planning to the district assemblies offers an opportunity to involve all stakeholders in the

planning process.

The establishment of KUMINFO in the KNUST and the Remote Sensing Applications Unit

(now called CERGIS) in the University of Ghana were some of Ghana’s earliest formal

attempts to propagate remote sensing and GIS applications. Also, Ghana's development

agenda, driven by the Growth and Poverty Reduction Strategy (GPRS II) which is informed by

our commitments to Millennium Development Goals (MDGs), New Partnership for African

Development (NEPAD) and above all, the underlying obligations set out in the Constitution

of the Republic of Ghana, recognises the planning and management of resources for

development. However, the land tenure system and the system of land delivery in Ghanaian

cities have significant implications for spatial development and urban sprawl (UN-HABITAT,

2008a). This research will therefore provide the ecological knowledge to supplement such

efforts at improving the urban environment in Kumasi, Ghana.

45

3 POLITICAL ECOLOGY OF WETLANDS IN KUMASI

3.1 INTRODUCTION In all the regions of the world where the absolute number of poor has increased, a majority of

them are in urban areas (Baharoglu & Kessides, 2000; Devas, 2005). In Ghana, the data on

poverty shows that, urban poverty as a proportion of total poverty is also increasing (White et

al., 2001). Using the expenditure approach, the ratio of urban to rural poverty in the Ghana

1991/92 survey years was 1:35 (Ravallion et al., 2007a). Urban poverty to some extent

reflects active rural–urban migration due to the uneven distribution of facilities and

opportunities in developing countries like Ghana (Baharoglu & Kessides, 2000).

Apart from government acquired lands, land is privately owned in Ghana. The official city

layouts are therefore prepared is consultation with landowners. In such layouts, wetlands and

natural areas or parks are delineated and cannot be built upon. Most of these layouts are not

serviced, such that, the distance to amenities influences the rent on the land and plays the

initial role of spatial resource allocator since it influences location choices of potential land

buyers (Capello & Faggian, 2002). This leads to residential areas reflecting spatial differences

in wealth of residents, types of houses and availability and density of infrastructure. The poor

immigrants tend to settle in the areas that are not part of the formal layout of the city and erect

temporal structures.

A poverty profile of the city will therefore distinguish these poor from the non-poor, identify

their characteristics, facilitate the process of identifying the location of the poor and highlight

groups that may be particularly vulnerable (Baharoglu & Kessides, 2000; Baker & Schuler,

2004). Data is, however, often difficult to obtain in developing countries for poverty profiling.

Whilst data is not easily available, the physical location and characteristics of the poor are

undeniably obvious in Ghanaian urban housing landscape. The wetlands of Kumasi are prime

areas for these informal settlements.

This chapter describes the physical and socio-economic characteristics of Kumasi,

specifically, inhabitants of these informal settlements along the wetlands areas. The political

ecology in terms of land acquisition and demolitions will be discussed. The aim is to identify

plausible development pathways and the choices and actions for (re)shaping these informal

settlements for a sustainable future.

46

3.2 METHODOLOGY

3.2.1 Physical and Socio-economic Survey of Wetland Communities of Kumasi This involved the use of primary and secondary data. The secondary data are mainly from

published works on Kumasi and its environs. The primary data were collected during field

visits through the use of surveys, interviews, observations and photographs at the different

study suburbs. All governmental and non-governmental organisations that are relevant to this

research or work directly or indirectly with the wetlands were identified. Table 3.1 is a list of

stakeholder groups identified and employed in this research. These included: the Kumasi

Metropolitan Assembly (Environment and Sanitation Unit), Friends of Rivers and Water

Bodies, sub-chiefs, Town and Country Planning Department (TCPD), Hydrological Services

Department (HSD) and the National Disaster Management Organisation (NADMO). Informal

visits were first made to the identified organisations and the researcher and research ideas

introduced and relevant personnel suitable for the research identified. Official introductory

letters were then sent to these identified personnel and where necessary, the heads of

departments to introduce the research. The specific involvement of the organisations or

personnel in this research was specified and sought. From the list of stakeholders, various data

collection strategies as shown in the table were employed to solicit the needed information.

3.2.1.1 Wetland communities and the associated anthropogenic activities

With the aid of maps, satellite imagery and the reconnaissance survey, houses, social and

economic activities within 100 m from the stream channels based on the 100 m buffer range

set by the Ministry of Lands and Natural Resources (Water Resources Commission, 2008)

were selected. From this, households were randomly selected and a structured questionnaire

administered to one adult member of the household (Appendix 1). Afterwards, respondents

who had interest and knowledge about the area were selected for a focus group discussion to

probe further, summarise and draw conclusions from the information collected from the

questionnaire survey and interviews. In all, 128 respondents answered the structured

questionnaires.

All anthropogenic activities within the 100 m buffer for each wetland were identified and

recorded. Stakeholders whose activities have direct impact on the wetland e.g. modifications

in water the channel, water quality, wetland vegetation etc. were identified and interviewed.

47

Table 3.1. Stakeholder list and the respective data collection tools used in the survey of wetlands in Kumasi, Ghana

Stakeholder group Stakeholder units # of respondents Data collection

method

Wetland associated communities

Dakwadwom 20 Structured questionnaire and

focus group discussions

Ahensan 21

Aboabo 21

Kwadaso Estates 20

Atonsu 20

Spetinpom 26

Wetland Associated Businesses

Agriculture 5 Interviews

Car washing bays 5

Carpenters 5

Car fitting shops 5

Churches 5

Traditional Authorities

Traditional Elder of Dakwadwom

1 Interviews

Traditional Elder of Ahensan

1

Traditional Elder of Aboabo

1

Governmental and non-Governmental organisations

KMA, HSD, TCPD, NADMO, Friends of Rivers & Waterbodies

5 Interviews

3.2.1.2 Traditional leaders, governmental and non-governmental agencies Each suburb of Kumasi usually has a chief or traditional leader/elder that oversees to

traditional issues in his/her jurisdiction. The traditional leaders of Aboabo, Dakwadwom and

Ahensan were identified and interviewed on traditional wetland management, land tenure,

landuse planning and future of the wetland. Interview sessions were also held with the listed

governmental and non-governmental organisations. Information on research, laws and

policies, and activities on wetlands in Kumasi was sought.

48

3.2.2 Data Analysis Various univariate analyses were used to assess the socio-economic forces that drive wetland

ecosystem and to indicate the degree to which socioeconomic, demographic and

environmental variables contribute to wetland alteration in Kumasi. The results are presented

in charts, graphs and tables.

3.3 RESULTS

3.3.1 The Physical Characteristics of Kumasi

3.3.1.1 Location, history, demography and administration of Kumasi Kumasi is the second largest city in Ghana and the capital of the Ashanti Region. It lies within

latitudes 6°38' and 6°45' north and longitudes 1°41' and 1°32' west on a landscape between

250 and 350 m above sea level. Kumasi used to be known as the Garden City of Africa. This

was because, during the colonial era, planning interventions in Kumasi were in the form of

establishing green belts and recreational parks within the layouts of residential areas. Due to

the undulating topography and numerous streams that transect Kumasi, the green belts and

recreational parks were invariably situated along the major stream channels. The green belts

idea was borrowed from various ‘garden city’ plans popular in England. They were seen as

both sanitary measures and as a way of separating residential areas especially the European,

from non-European zones. The provision of a green belt of about 300 m from buildings

surrounding residential areas was designed to protect Europeans residing in these areas from

mosquito borne diseases. After independence, these green belts were designated as

recreational space and named ‘Nature Reserves’ or ‘Parks’ on zoning maps (Schmidt, 2005;

KMA, 2006). Today, this garden city status is long lost and these green belts are largely either

non-existent, their borders slowly being infringed upon for urban infrastructure and housing,

subsistence agriculture or are used as dumping grounds (Brook & Dávila, 2000). The suburbs

e.g. Dakwadwom, Ahensan, Aboabo and Atonsu situated in or close to these ‘vacant’ land

areas are frequently under floods and therefore the subject of this research.

The Kumasi District as delineated in Figure 3.1 is metropolitan because it has a population

over 250,000 (Ghana Statistical Service, 2005). The Kumasi Metropolitan Assembly (KMA)

is made up of the following Sub-Metropolitan areas: Asawase, Asokwa, Bantama, Kwadaso,

49

Figure 3.1. Map of Ghana with an insert showing limits of the Kumasi Metropolitan Assembly and the study sites [sites are labelled in letters: Atonsu Sawmill (A), High school junction (B), FRNR farms (C), KNUST Police Station (D), Family Chapel (E), Spetinpom (F), Golden Tulip Hotel (G), Kaase Guinness (H), Bekwai roundabout (J) and Dakwadwom (K)]

50

Manhyia, Nhyiaeso, Oforikrom, Old Tafo, Suame and Subin. There is the Metropolitan

Planning Board which integrates the development planning and management of the

Metropolis; monitors and evaluates development planning and management; integrates the

development planning proposals from the sub-metropolitan district councils and advise the

authority thereon. The KMA is bounded by the Bosomtwe, Ejisu-Juaben, Kwabre East,

Afigya-Kwabre, Atwima-Kwanwoma and Atwima-Nwabiagya Districts.

The urban population of the Ashanti Region (which is 51.3 %) is only second to the Greater

Accra Region (87.7 %). The Kumasi Metropolis accounts for about a third of the region’s

population. The population of Kumasi grew slowly from about 3,000 in 1901 to about

180,000 in 1960 (Figure 3.2). Since then, there has been a relatively rapid growth of Kumasi

exceeding a million in 2000 with a current growth rate of 5.9 % and a density of 5,319

individuals/km2 (Ghana Statistical Service, 2002). Between 1984 and 2000, the population

had increased over 200 %. Due to its fast physical and demographic growth, Kumasi now

extends beyond the administrative boundaries of the KMA into the neighbouring districts

(KMA, 2006).

year

1900 1920 1940 1960 1980 2000

popu

latio

n

0.0

2.0e+5

4.0e+5

6.0e+5

8.0e+5

1.0e+6

1.2e+6

1.4e+6

Figure 3.2. Population growth trajectory of the city of Kumasi (Ghana Statistical Service, 2005)

51

3.3.1.2 Climate of Kumasi The climate of Kumasi is tropical, of the wet sub-equatorial type, and strongly influenced by

the tropical continental and tropical maritime air masses. The migration of the Inter-Tropical

Convergence Zone (ITCZ) determines the seasons. Rainfall is weakly bi-modal. The major

rainfall period is between May and June and the minor rainfall period is between September

and October. The dry north-east trade winds (Harmattan) reach Kumasi around mid-

November to early December. Relative humidity on a normal day, in Kumasi, ranges from a

maximum of 100 % at 0900 GMT to 60 % at 1500 GMT. On very dry days it ranges from 40

to 84 % and 85 to 100 % on very wet days. The average minimum and maximum

temperatures in Kumasi are about 21.5 °C and 30.7 °C respectively.

3.3.1.3 Hydrology of Kumasi The Metropolis is located within the Pra basin which is divided into five sub-basins by a

series of ridges running generally in a north-south direction. These sub-basins are Kwadaso,

Subin, Aboabo, Sisai and the Wiwi. The Suatem River flows through North Suntreso and

joins the Kwadaso River at Dakwadwom to form Daban River which flows to Kaase. The

Subin River flows from the Kumasi Zoo also to Kaase. The Wiwi River flows from Nsenie

and merges into the Sisai River at Atonsu. The Sisai, Subin and Daban Rivers finally join the

Oda River, a tributary of the Pra near Asago, south of Kumasi. In the north-western sector of

the Metropolis are several small tributaries of the Owabi River. The Metropolis is therefore

drained by a relatively dense network of rivers whose natural drainage runs generally from

north to south, exhibiting dendritic patterns.

However, estate developments, agricultural encroachment and indiscriminate waste disposal

in the rivers have impacted negatively on the drainage systems and have consequently brought

some of these water bodies to the brink of extinction and increased the incidence and spread

of floods and floodwaters. Some portions of these rivers have also been converted to storm

drains e.g. the portion of the Aboabo River from the Airport roundabout to the High School

Junction. For those that are still flowing, pollution is a serious problem.

3.3.1.4 Landuse and vegetation of Kumasi The Kumasi Metropolis falls within the moist semi-deciduous South-East Ecological Zone.

About 80 % of the total land coverage of the Metropolitan area is planned (KMA, 2006). The

major land uses that make up the Metropolis are residential, industrial, commercial,

educational, civic and culture, transportation and open spaces. Among these land uses, the

predominant ones are residential, commercial, educational and industrial. The residential

52

areas take up to 44 % of the total land area in Kumasi. About 29 % is taken up by the different

land uses such as educational, civic and culture, transportation and open spaces.

Patches of vegetation reserves within the Metropolis house the Kumasi Zoological Gardens

and the KNUST Botanical Gardens. Others are the Otumfuo Children’s Park at Dakwadwom,

Kumasi Children’s Park at Amakom and the Kumasi Golf Course at Nhyiaeso. Aside these,

there are other patches of vegetation cover scattered over peri-urban areas of the Metropolis

e.g. Atasomanso, Breman and Kagyase and Owabi. Predominant local trees are Ceiba,

Triplochlon and Celtis. Due to urban sprawl, most trees have been cut to make way for

buildings, roads and other infrastructure. All lands adjacent to the main rivers have been

declared by TCPD as green belts. These green belts are meant to serve as natural flood plains

of the river but encroachment has significantly reduced their flood storage capacity. Common

vegetation in these wetland areas are shrubs, grasses and food crops and vegetables like

plantain, sugar cane, palm trees, okro, carrot, cabbage etc. associated with urban agriculture.

3.3.1.5 Relief, geology and soils of Kumasi The Metropolis is undulating and lies within the plateau of the south-west physical region

which ranges from 250-300 m above sea levels. The area is underlain with precambium rock.

There are six detailed soil types in the metropolis but the predominant soil type of the

metropolis is forest ochrosol which dominates the entire region. The soil within the basins is

silty clay loam which is moderately slow to slowly permeable resulting in fairly high runoff

rates. This rich soil type makes it possible for foodstuff to be grown in the Metropolis and its

periphery.

3.3.2 Socio-economic Characteristics of Wetland Communities in Kumasi The results of various interviews, focus group discussions with various stakeholders and the

responses of 128 residents of various wetland associated communities are represented in this

section. The specific suburbs being studied are Aboabo, Spetinpom, Atonsu, Ahensan,

Dakwadwom and Kwadaso Estates. Large portions of these suburbs are informal or slum

settlements with haphazardly arranged houses built from wood or block. There is little or no

drainage network. Most houses directly release their domestic waste water through small

channels to the open which finds its way into surrounding streams. Most of the lands used for

these houses are designated as “natural areas” according to the landuse plan of the city.

However, the land owners sell out small portions of these natural areas for various uses

leading to a gradual habitation of such zones. Alleys are very narrow or virtually non-existent

because of encroachments and erection of illegal structures along the alleys.

53

These communities also have high population densities with very little social infrastructure.

Local entrepreneurs take advantage of the dearth in services to provide substandard facilities

to these already economically depraved community. There are several private primary schools

and churches mostly housed in wooden structures. Drug and chemical stores, kiosks and

artisanal shops abound in these suburbs. Also, in these wetland areas, there are no culverts or

bridges linking communities on either side of the water channel. Wooden foot bridges are

constructed across the streams and rivers through local efforts. There is no thoroughfare in

most areas so vehicular movement in such communities is limited.

Figure 3.3 shows the occupational distribution by gender and location of the respondents

considered in this study. Although this study did not seek to investigate gender distribution in

these suburbs, majority of respondents (62 %) were male. Respondents who were porters or

involved in food processing were all female whilst trading, farming, artisanal works and

formal sector employment had both sexes involved. Trading (35 %) and artisanal work (28 %)

form the major occupations of these respondents. The traders are involved in petty trading like

running kiosks, petty and buying and selling of various second-hand goods in the city.

Majority of these traders either do not own the businesses or have ownership with a capital

base less than $100. Most of the petty traders live in Aboabo. The artisans include masons,

plumbers, carpenters, shoemakers, auto mechanics, tailors and hairdressers. Most of these

artisans live in Kwadaso Estates, Dakwadwom and Spetinpom. The porters were only in

Aboabo, Ahensan and Spetinpom and were all immigrants who have been in the metropolis

for less than 10 years.

Various types of houses were identified within 100 m from the water channel (Figure 3.4).

These were permanent block structures in the form of multi-family residential facilities

commonly called compound houses, single family self-contained structures (detached or

semi-detached), L-shaped detached or single room structures and wooden structures. The

wooden structures were found in Aboabo, Ahensan, Atonsu and Dakwadwom. Occupants

were all less than 10 years in these facilities. When the economic conditions of these residents

improve, most of them transform the wooden structures to more permanent block structures.

The predominant housing types for respondents from Aboabo are compound, wooden or L-

shaped detached houses. The houses in Aboabo unlike those of Kwadaso Estates are mostly

multifamily permanent compound structures with relatively large numbers of residents even at

the household level. Those in Kwadaso Estates are mostly single family accommodations.

54

Figure 3.3. The occupation of respondents from flood prone suburbs of Kumasi

55

Figure 3.4. Residential types and homeownership of respondents in the different

suburbs of Kumasi

Majority of the respondents (62 %) either own the houses or are part of the extended family

and do not pay rent in the houses they lived in. The largest proportion of home owners is

those in wooden and permanent single room structures in Ahensan and Atonsu. Most of the

multifamily residential facilities like the compound, detached and semi-detached houses have

tenants who are completely separate households. There is also a relatively high tenant

turnover in these areas. Most people who did not know of the annual floods usually moved

out after one or two flood experiences. The relationship between length of stay and prevalence

of floods was assessed using the Pearson Correlation. There was no relationship between

number of years of stay in an area and presence or absence of floods (r = -0.135). Access to

land for construction of houses by wetland residents is very varied in the different suburbs

(Figure 3.5). The commonest forms of access to land for building a house were family

inheritance, direct purchase from the chief or other landowner or squattering. Respondents in

Aboabo were either squatters or living in rented houses. Most of the respondents from Atonsu

and Dakwadwom inherited the land or the houses they lived in.

56

Various reasons have been ascribed for use of wetland areas for building houses (Figure 3.6).

The commonest reasons were that wetland areas are cheap, close to other parts of the city and

are “no man’s land”. These are usually areas marked out in the landuse plans as natural areas

and not to be developed. However, the landowners later bypass the city authorities and sell

such areas at very low prices. The city authorities do not also actively protect or use such

areas therefore creating the impression that it is a no man’s land. For these reasons, the

wetland areas in Kumasi quickly develop into informal settlements harbouring squatters and

slum communities.

After staying in these environments for a while some residents relocate. Respondents have

different motivations to move out of their current home (Figure 3.7). Some respondents in

Aboabo and Ahensan love living in those suburbs and are not making any efforts to move out.

In all the suburbs, respondents over 75 % of the time, completely agree to move out if given

money or an alternative shelter elsewhere. Forcefully moving people out of these suburbs is

not popular. For all the suburbs, the flood water level is not a first priority to cause them to

move out.

57

Figu

re 3

.5. R

espo

nden

ts’ s

ourc

es o

f lan

d fo

r co

nstr

uctio

n of

hou

ses

in th

e di

ffere

nt s

ubur

bs o

f Kum

asi

0102030405060708090100

family inheritance

from chief

from another person

squatter

rental accommodation

family inheritance

from chief

from another person

squatter

rental accommodation

family inheritance

from chief

from another person

squatter

rental accommodation

family inheritance

from chief

from another person

squatter

rental accommodation

family inheritance

from chief

from another person

squatter

rental accommodation

family inheritance

from chief

from another person

squatter

rental accommodation

Aboa

boAh

ensa

nSp

etin

pom

Aton

suDa

kwad

wom

Kwad

aso

Esta

tes

Proportion of respondents (%)

Subu

rb a

nd s

ourc

es o

f lan

d fo

r con

stru

ctio

n of

hou

ses

58

Figu

re 3

.6. R

easo

ns fo

r cho

ice

of w

etla

nd a

rea

for

cons

truc

tion

of h

ouse

s in

the

flood

pro

ne s

ubur

bs o

f Kum

asi

0102030405060708090100

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

it is cheapuse stream water

economic activities"no mans's land"

close to amenitiesclose to familyin/close to city

Aboa

boAh

ensa

nSp

etin

pom

Aton

suDa

kwad

wom

Kwad

aso

Esta

tes

Proportion of respondents in agreement with reason (%)

Subu

rbs

and

reas

ons

for c

hoice

of s

ubur

bs

59

Figu

re 3

.7. F

acto

rs o

r in

stru

men

ts t

hat w

ill m

otiv

ate

resi

dent

s of

the

vario

us f

lood

pro

ne c

omm

uniti

es t

o re

loca

te

0102030405060708090100

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

do nothing. I love it heregiven money to move

new land/plot elsewherewhen I make some money

when the floods get too muchwhen forced

I am already preparing to moveif provided an alternative shelter

Aboa

boAh

ensa

nSp

etin

pom

Aton

suDa

kwad

wom

Kwad

aso

Esta

tes

proportion of respondents in agreement with reason (%)

subu

rbs

and

reas

ons

to m

ove

out

60

3.3.3 Community Efforts at Wetland Management in Kumasi Wetlands within the Kumasi Metropolis have been used for various purposes (Figure 3.8).

The commonest wetland uses are urban agriculture, grazing grounds, housing and as waste

disposal sites. In Aboabo, Atonsu, Dakwadwom and Kwadaso Estates, waste disposal was

commonest wetland use. The wetlands in Ahensan and Spetinpom were mainly used as

grazing grounds and for housing respectively. The only wetland area with significant

recreational use was that at Dakwadwom. Here, Friends of Rivers and Waterbodies had an

active educational programme on wetlands through the establishment of the Otumfuo

Children’s Park.

Even though these wetlands are not used sustainably, the associated communities believed

that they should be protected (Figure 3.9). Majority of respondents (59 %) were in favour of

protecting the wetland area. Only a few respondents (3 %) were indifferent to their protection.

There are various local laws, customs and beliefs that help in the protection and management

of wetlands. These are however, not being adhered to and respondents believe several factors

are responsible for this (Figure 3.10). Respondents in Aboabo and Ahensan did not believe

that political influence contributed to communities not managing the wetlands. Rather, it is

the lack of political will by stakeholders to protect the wetlands. In Spetinpom and Kwadaso

Estates, respondents were of the view that corruption and the lack of political will to enforce

the laws were the two important factors debilitating wetland management in Kumasi. In all

the suburbs, poverty was also viewed as a very strong factor (more than 50 % yes) accounting

for wetlands not being protected.

61

Figure 3.8. The predominant uses of wetlands in the various suburbs of Kumasi

Figure 3.9. Perception of wetland protection in Kumasi

62

Figu

re 3

.10.

Rea

sons

for t

he c

ontin

ual

dest

ruct

ion

of w

etla

nds

in t

he v

ario

us s

ubur

bs o

f Kum

asi

0102030405060708090100

political influence

changes in government

corruptionpoverty

lack of political will

cost of demolitionloss of cultural values

political influence

changes in governmentcorruption

poverty

lack of political willcost of demolition

loss of cultural values

political influencechanges in government

corruption

povertylack of political will

cost of demolition

loss of cultural valuespolitical influence

changes in government

corruptionpoverty

lack of political will

cost of demolitionloss of cultural values

political influence

changes in governmentcorruption

poverty

lack of political willcost of demolition

loss of cultural values

political influencechanges in government

corruption

povertylack of political will

cost of demolition

loss of cultural values

Aboa

boAh

ensa

nSp

etin

pom

Aton

suDa

kwad

wom

Kwad

aso

Esta

tes

Proportion of respondents in agreement with reason (%)

Subu

rbs

and

reas

ons

for c

ontin

ual d

estr

uctio

n of

wet

land

s

63

3.4 DISCUSSION Similar to the situation in most developing countries, urbanisation of Kumasi is characterised

by very high population growth and construction, merging up the city with the peri-urban

regions. This fast urban growth is however not matched with growth in infrastructure and

services, not even in the planned suburbs. The resource poor suburbs also harbour mostly

rural-urban migrants and the poorest of the population. In recent times, Kumasi has received

large influxes of rural youths from mostly northern Ghana to find menial jobs. A majority of

these low skilled migrants end up working in the informal sector as porters, petty traders,

seasonal labourers (Yeboah & Appiah-Yeboah, 2009). The porters and farmers identified in

this study are likely to be people of this origin. After a little training, most of these immigrants

pick up artisanal jobs like carpentry, masonry, barbers/hairdressers, tailoring/dressmaking etc.

Their economic conditions usually determine their choice of urban space and or housing.

The suburbs considered in this tudy have large immigrant populations and thus, not only

provide important basket of social networks for first time immigrants, but also as political

hotspots. For these reasons, residents from such suburbs have been able to vehemently resist

all efforts by the KMA to demolish their illegal structures. The relatively high population

density and communal spirit of these residents increase their collective bargaining power. In

these suburbs, the only ‘free’ areas are often the wetland areas. The immigrants therefore find

these areas convenient for erecting temporal dormitories with active assistance from old

residents. It is however, surprising that most people will continue to live in such structures or

their improved forms in the same locations after several years of living there and an improved

economic standing.

It is important to state that, majority of the wetland associated inhabitants are natives. These

natives continue to live in such neighbourhoods due to the strong historical, social and

economic ties developed over the years. For an immigrant also, early social networks and the

cost of accommodation might be determinants of their choice of such locations. It is not

surprising that these immigrants first live in makeshift shelters made of wood and may later

move to more permanent single room block structures. Aboabo, Ahensan and Dakwadwom

might be suburbs providing these initial favourable conditions to attract the high immigrant

squatter population. Ownership of such structures might be through a form of transfer of title

or a rental arrangement with the new tenant whilst the economically better off tenant moves

out. Due to the high the demand for these neighbourhoods by immigrants, landowners and

chiefs also then take advantage of these illegalities to lease out or sell the plots to these

64

tenants giving them title to the land. For any given environmental challenge, it is expected

that there is a tipping point at which an individual finds that particular environment unsuitable

for habitation. These suburbs however, are among the densest in Kumasi. The pull factors

motivating people to move in and to continue staying there are stronger than the dissuading

environmental pressures in these neighbourhoods.

According to the theory of the EKC, increased incomes or improved economic conditions

would lead to improvement in the environment or in this case, a desire for better housing

environment away from the annual floods. The findings of this study do not support this

theory either. Rather, improved economic conditions led to improved housing facilities in the

same environment. For example, wooden structures were converted to more permanent block

structures. This is still true even when flood water levels are increasing. A model is proposed

here to describe the relationship between the EKC and the decision to move or stay in these

flood prone suburbs in Kumasi (Figure 3.11). It shows that the initial trajectory of the EKC

curve realistically captures decision-making by poor resident in a flood prone suburb. In

beginning, the victim is only interested in making ends meet. However, at the critical income

threshold, the person starts to invest in improving his/her living space. The trajectory

therefore does not lead to a condition of zero environmental degradation, as would be

expected, but rather a tolerable level through participation in communal efforts like building

of foot bridges, network of walkways etc. In another dimension, this new poor resident

usually has almost 100 % tolerance to the floods. Such a person will only move when forced.

With increasing income, different motivations and needs come to the fore. Consequently,

flood tolerance never drops to 0 % in these communities. There are a few who are ready to

move out ‘when the water level increases’. But such a subjective target is never met and these

people continue to live there. The decision to move or stay in these suburbs is therefore an

abrupt one independent of the environmental and economic conditions.

Homeownership in Ghana is a cultural and societal requirement or an expectation as one

grows older; as having your house is a sign of success. This is aggravated by a national

housing deficit and a very poor mortgage system. Most people therefore acquire the land and

build their houses over several years. Land prices are very high in urban areas and the prices

reflect the quality of the potential houses and services and infrastructure availability in the

neighbourhood. Poor people therefore go for the cheapest available lands that are usually

wetland areas and therefore not part of the official city layout scheme. It is these cheap

65

Figure 3.11. A proposed model to describe the decision to move or stay in the flood prone suburbs of Kumasi

66

unapproved houses built in wetland areas that bear the full brunt of the annual floods. From

the list of pressures or initiatives in the model above, these flood prone inhabitants will rather be

moved more by external pressure than their own volition. The will to continue to stay or the

decreased personal initiative to move out of the wetland may be aggravated by the acquisition of

property rights through the purchase of such lands. According to Lant (1994), the core of the

conflict in wetland destruction is a question of property rights. However, does the owner of

the wetland have the right to alter that wetland for his or her own purposes? Does the public

have the right to the ecological functions and values of a private wetland? Do individual rights

acquired through the purchase of wetlands override national and local government laws? Does

the owner of a wetland have the right to alter it for private purposes at the public expense?

How can the externalities from such alterations be internalised? When can such an individual

right be revoked for the public good?

In demarcating the wetlands as natural areas, it is unfortunate that the KMA (through the

TCPD) does not make any effort at assigning and enforcing property rights to such areas. If

the city wishes to acquire property rights to such areas, it will have to pay a compensation (buy

the rights) to the landowner. By delineating these wetland areas as ‘Nature Reserves’ or ‘Parks’,

the TCPD gives de facto property rights to the public. This will imply that, the landowner has no

right to sell such lands. The land tenure system in Ghana, however, does not recognise such

de facto ownership. The proposals by Panayotou (1997) and Eppinka et al (2004) for more

secured property rights will not work in the Ghanaian context where wetlands delineated for

public good or environmental protection is not paid for. Most of the lands are privately owned

and therefore the city (public) does not automatically obtain rights to the wetlands because it

has been delineated as such. Whilst it may be expensive for the government to pay compensation

to designate these wetland areas as public lands, it is also widely known that government has

not been good managing existing public lands. Therefore, distribution of benefits and costs of

maintaining wetlands as advocated by Magnani (2001) may be a better option to management of

such spaces. The Otumfuo Children’s Park in Dakwadwom is an example which could be

replicated in other wetland areas. Names, uses and attributes should be assigned to the

wetlands and benefits and costs accrued be shared between the landowners and city

authorities for management of the wetland area. Hereafter, better enforcement and effective

environmental regulations will then be needed to maintain the wetland as such.

67

4 FLOODS AND ENVIRONMENTAL MANAGEMENT IN KUMASI, GHANA

4.1 INTRODUCTION Ghana is very much in tune with global environmental concerns. Over the years, it has signed

several conventions and protocols (see Table 4.1) in an effort to strengthen international

cooperation and linkages, and make the environment a better place for its citizenry. Also,

Ghana’s Constitution, Chapter six, Article 41 (k) enjoins the citizens to protect and safeguard

the environment (Government of Ghana, 1992). Apart from the articles in the Constitution,

there are specific laws and policies on the environment. Some of those that are of relevance to

wetlands include the Beaches Obstructions Act (1897), Rivers Act (1903), Town and Country

Planning Act (1945), Land Planning and Soil Conservation Act (1953), Volta River

Development Act (1961), Land Registry Act (1962), Lands Miscellaneous Provision Act

(1963), Ghana Water and Sewerage Corporation Act 310 (1965), Abandoned Property

(Disposal) Act (1974), Irrigation Development Authority Act (1977), Mercury Act (1989),

Small Scale Gold Mining Act (1989), Control and Prevention of Bush Fires Act (1990),

Towns Act (1992), Environmental Protection Agency Act 490 (1994), Lands Commission Act

483 (1994), National Development Planning Commission Act 479 (1994), Local Government

Act 462 (1994), Water Resources Commission Act 522 (1996), Environmental Assessment

Regulations LI 1652 (1999), Wetlands Management (RAMSAR sites) Regulations (1999),

Fisheries Act (2002) and Ghana Meteorological Agency Act (2004).

These laws and regulations have historically been applied to various contexts and use of

wetlands. However, the Environmental Protection Agency Act 490 (1994), Wetlands

Management Regulations (1999) and the Buffer Zone Policy are currently the most important

and of direct use in the protection and management of wetlands in Ghana. These laws and

policies were enacted based on the current Constitution of Ghana and have various institutions

mandated to regulate, direct and implement them. Ghana may be very sufficient in terms of

laws and policies. However, the human resources and institutional framework under which

the laws and policies are being implemented are equally important. This chapter therefore

assesses the laws and policies that apply to the environment in Ghana. Particular emphasis

will be placed on wetland management and floods and the stakeholders and institutional

framework under which they operate.

68

Table 4.1. Ghana’s international commitments to the environment

Convention/Protocol Date of Accession/ Ratification

� Convention on the Continental Shelf, Geneva, 1958 29 April, 1958

� Convention on Fishing and Conservation of the Living Resources of the Seas, Geneva, 1958

29 April,1958

� African Convention on the Conservation of Nature and Natural Resources, 1969

17 May, 1969

� Convention on International Trade in Endangered Species of Wild Fauna and Flora, Washington, 1973

14 November, 1975/ 12 February, 1976

� United Nations Convention on the Law of the Sea, Montego Bay, 1982.

10 December, 1982/ 7 June, 1983

� Convention on the Conservation of Migratory Species of Wild Animals, Bonn, 1979

19 January, 1988

� Convention on Wetlands of International Importance Especially as Waterfowl Habitat, Ramsar, 1971

22 February, 1988

� Convention for Cooperation in the Protection and Development of the Marine and Coastal Environment of the West and Central African Region, Abidjan,1981

20 July, 1989

� Protocol Concerning Cooperation in Combating Pollution in Cases of Emergency, Abidjan, 1981

20 July, 1989

� Vienna Convention for the Protection of the Ozone Layer, Vienna, 1985

24 July, 1989

� Montreal Protocol on Substances that Deplete the Ozone Layer, Montreal, 1987

24 July, 1989

� London Amendment to the Montreal Protocol on Substances that Deplete the Ozone Layer, London, 1990

24 July, 1992

� Convention on Biological Diversity, Rio de Janeiro, 1992 29 August, 1994

� United Nations Framework Convention on Climate Change, New York, 1992

6 September, 1994

� United Nations Convention to Combat Desertification in those Countries Experiencing Serious Drought and / or Desertification, Particularly in Africa, Paris, 1994

27 December, 1996

� Copenhagen Amendment to the Montreal Amendment on Substances that Deplete the Ozone Layer, Copenhagen, 1992

30 September, 2000

� Stockholm Convention on Persistent Organic Pollutants 30 May, 2003

� Montreal Amendment to the Montreal Protocol 8 August, 2005

� Beijing Amendment to the Montreal Protocol 8 August, 2005

69

4.2 ENVIRONMENTAL MANAGEMENT AND THE NATIONAL ENVIRONMENTAL ACTION PLAN

Traditionally, Ghana’s environmental challenges have been the depletion of her resources

through deforestation, desertification and soil degradation (Okley, 2004). Recently,

urbanisation and industrialisation have also come to the fore (UN-HABITAT, 2009). In

March 1988, the Government of Ghana initiated a major effort to manage these environmental

challenges. The exercise culminated in the preparation of a strategy and a National

Environmental Action Plan (NEAP), to define a set of policy actions, related investments, and

institutional strengthening activities to make Ghana's development strategy more

environmentally sustainable and to improve the surroundings, living conditions and the

quality of life for all generations of Ghanaians. The NEAP provided a framework for

interventions deemed necessary to safeguard the environment. For areas related to wetlands,

the policy actions included the establishment of the proposed Water Resources Commission;

adoption of the proposed Water Law; adoption of policy framework for extraction of water

for the different uses; integrated planning of river basins and the control of waste discharge

into water bodies; and establish watershed protection areas. Institutions and detailed working

documents to these objectives have all been produced.

The NEAP also includes a National Environmental Policy to provide the broad framework for

the implementation of the Action Plan. The policy aims at ensuring sound management of

resources and the environment and to avoid any exploitation of these resources in a manner

that might cause irreparable damage to the environment. The policy identified among others,

the following issues to be dealt with: population pressure, land ownership and tenure, landuse

planning, agricultural land use, water resources assessment, problems of water management,

water pollution, watershed protection, urbanisation and urban degradation.

4.2.1 The Environmental Protection Agency The Agency started as the Environmental Protection Council (EPC) in 1974 and was

responsible for all activities and efforts aimed at protecting and improving the quality of the

environment. The EPC functioned as an advisory body until 30th December 1994 when it was

transformed into an Agency by the Environmental Protection Agency (EPA) Act 490. By this

Act, the EPA became a corporate body mandated to implement the NEAP with powers to sue

and be sued. The EPA by law is therefore the leading public body for protecting and

improving the environment in Ghana. The objectives of the EPA, among others of interest to

this research, are to:

70

� mainstream the environment into the development process at the national, regional,

district and community levels;

� ensure that the implementation of environmental policy and planning are integrated

and consistent with the country’s desire for effective, long-term maintenance of

environmental quality; and

� guide development to prevent, reduce, and as far as possible, eliminate pollution and

actions that lower the quality of life.

It performs these with the following guiding principles, inter alios, partnerships, pollution

prevention and control, ecosystem management, environmental justice, environmental

education, compliance and enforcement. The EPA has power to issue environmental permits

and pollution abatement notices in order to control waste discharges and emissions and to

prevent or reduce noise pollution. It is supposed to provide standards and guidelines in

relation to air, water and other forms of environmental pollution. It also has authority to

ensure that developers of commercial entities comply with environmental impact assessments.

4.2.2 Water Resources Commission The major laws that guide the regulation and management of water resources in Ghana are

contained in the Water Resources Commission Act (Act 522 of 1996) and the Water Use

Regulations (L.I. 1692 of 2001). This Legislative Instrument categorises and prioritises water

uses in Ghana and spells out procedure for acquisition of water use permits. Water, like all

other natural resources in Ghana, falls within the provisions of Article 269 of Ghana's

Constitution, and is therefore vested in the President on behalf of and in trust for the people of

Ghana. There is no private ownership of water in Ghana, but that the President, or anyone so

authorised by the President, may grant rights for water use. The Water Resources Commission

was set up and given the powers by the President for the regulation and management of water

resources and for the coordination of policies in relation to them (Government of Ghana,

1996). The Commission is the focal point in fostering coordination and collaboration among

the various actors involved in the water resources sector.

The Commission’s management of water resources is underpinned by the National Water

Policy (Ministry of Water Resources, Works and Housing, 2007). The Policy recognises the

perceived abundance of water resources in Ghana. However, the production and utilisation of

water resources for consumptive and non-consumptive uses is not at optimal level and

therefore calls for an efficient and effective management of available water resources. The

Policy is based on principles of Integrated Water Resources Management (IWRM). In

71

particular, it encourages the sustainable exploitation, utilisation and management of water

resources to ensure full socio-economic benefits to the present and future generations, while

maintaining biodiversity and the quality of the environment.

Wetland ecosystems in Ghana constitute about 10 % of the country’s total land surface and

are grouped as marine/coastal, inland and man-made. Recognising the scale of distribution

and importance of wetlands, the Ministry of Lands and Forestry (1999b) integrated wetland

issues into the National Landuse Policy. The Policy recognises wetlands as environmental

conservation areas and precludes the following practices: physical draining of wetland water,

draining of streams and water courses feeding the wetlands, human settlements and their

related infrastructural developments in wetlands, disposal of solid waste and effluents in

wetlands and mining in wetlands. The Government of Ghana sees its role in wetlands

management as best performed through partnership and co-operation with local people,

governmental and non-governmental organisations, and the private sector.

Ghana became a signatory to the Ramsar Convention in 1988. A major obligation of

signatories to the Convention is the implementation of the principle of ‘wise use’ of their

wetland resources. Therefore, in consultation with stake-holders, the “Managing Ghana’s

Wetlands: A National Wetlands Conservation Strategy” document was prepared to promote

the participation of the local communities and other stakeholders in the sound management

and sustainable utilisation of Ghana’s wetlands and their resources (Ministry of Lands and

Forestry, 1999a). The document provides a national definition and distribution of wetlands, a

management strategy for wetlands and a policy framework for incorporating wetlands

management into the day-to-day activities of Government, organisations, traditional

authorities, communities and individuals. Accordingly, the Government of Ghana has overall

responsibility for perpetuating wetland areas, and also to administer a range of social,

economic and environmental programmes which impact on wetland management throughout

the country whilst local communities directly manage the "wise use" of wetland resources in

their localities.

Also, there is the Buffer Zone Policy for the protection, regeneration and maintenance of the

native or established vegetation in riparian buffer zones in Ghana (Water Resources

Commission, 2008). This policy is a further attempt at managing Ghana’s river basins in

accordance with the IWRM approach and to harmonize traditional and existing public

institutional standards on buffer zones in Ghana. The Policy adopts a 3-tier buffer zone as the

norm for effective erosion control and halting of water quality degradation caused by

72

nutrients, animal or human wastes, sediments and runoff. It however, gives provision for

variation of this optimum buffer “design” from the prescribed values to accommodate local

requirements. The buffer zones are also to be permanently reserved by legislative instruments

with the consent of local authorities and incorporated into the local landuse plans. Local

authorities must take steps to minimize the risk of conversion of buffer zones to uses

incompatible with sustainability of their functions and exclude exploitation or occupation

inimical to the purposes of designation of the areas as buffer zones. Government is also to

ensure that all designated riparian buffer zones along rivers, streams, and around lakes,

reservoirs or other surface water bodies are adequately vegetated and sustainably managed to

restore and maintain their ecological integrity. This is expected to provide socio-economic

benefits to local communities and in fulfilment of Ghana’s overall water, environment and

landuse policies, the MDGs and the Growth and Poverty Reduction Strategy (National

Development Planning Commission, 2005).

4.2.3 Lands Commission Ghana has basically two types of land ownership: state and private. State lands are lands

compulsorily acquired by the government through the invocation of the appropriate

legislation, vested in the President and held in trust by the State for the entire people of

Ghana. On the other hand, private lands in most parts of the country have communal, family

or individual ownership. Stool or skin lands are a feature of communal land ownership in

most Akan traditional groups in southern Ghana and in most traditional groups in northern

Ghana respectively. However, scattered all over Ghana are a number of traditional groups,

which do not recognize a stool or skin as symbolising private communal land ownership. Also

sandwiched between the public and private lands are vested lands, which are a form of split

ownership between the state and the traditional owners. Fundamentally, land ownership is

based on absolute allodial or permanent title from which all other lesser titles to, interests in,

or right over land derive. Normally, the allodial title is vested in a stool, skin, clan, family,

and in some cases, individuals.

Because of this diversity of land tenure regimes, the Lands Commission was established

under article 258 of the 1992 Constitution and came into being through the Lands

Commission Act (Act 483 and 767 of 1994 and 2008 respectively). The Commission is an

integration of the various land agencies for the efficient management of public lands and other

lands, advise and facilitation of good land delivery system in the country. The Commission’s

work is being guided by the National Land Policy. The National Land Policy, District

73

Assemblies Act 462 and the National Development Planning Systems Act 480 enjoin District

Assemblies in conjunction with land owners and the Lands Commission to prepare planning

schemes for all land uses to facilitate dispositions of land for development. In making these

planning schemes, the policy states that wetlands are environmental conservation areas and

the following uses considered incompatible with their ecosystem maintenance and natural

productivity are strictly prohibited: physical draining of wetland waters, damming of streams

and water courses feeding the wetlands, human settlements and their related infrastructural

development in wetlands, disposal of solid waste and effluents in wetlands and mining in

wetlands.

4.2.4 Local Government and Environmental Management For the purposes of local government, Ghana is divided into districts (Government of Ghana,

1993). The Local Government Act (Act 462 of 1993) makes the District Assembly the highest

political authority in any district with deliberative, legislative and executive powers. Among

others, the district assembly is responsible for the development, improvement and

management of human settlements and the environment in the district. The assemblies are

empowered to enact the appropriate bye-laws for planning and management of resources in

their geographical area. In accordance with the NEAP, the responsible committee for

environmental issues at the local level is the District Environmental Management Committee

(DEMC) of the district assembly. The number and composition of this committee varies from

district to district. In general, the DEMCs are made up of about 15 members. These include

the regional programme officer of the EPA, five assembly members (one as chair), at least

two of whom should be women, two representatives of environmental NGOs, officers from

decentralised institutions like the Department of Parks and Gardens, TCPD, Department of

Agriculture, Forest Services Division, District Health Directorate, the District Education

Service, Ghana Water Company or Community Water and Sanitation Division and National

Council on Women and Development. The EPA's Regional Programme Officers and their

staff are expected to provide support and assist with training of DEMCs and to act as a

conduit for dissemination of information from EPA to the districts. The DEMC enforces

environmental standards and regulations set by the EPA and other national regulatory

agencies within its jurisdiction.

74

4.3 WETLANDS, FLOOD PREVENTION AND MANAGEMENT

4.3.1 Stakeholders’ Capacity for Managing Wetlands and Floods The following stakeholders were identified to be involved in wetlands and or flood

management in Kumasi: KMA, HSD, TCPD, NADMO, Assemblyman, Chiefs and NGOs

(Friends of Rivers and Waterbodies).

There were varied perceptions about the responsibilities of stakeholders for the management

of wetlands (Figure 4.1). The commonest stakeholders mentioned were NADMO and the

KMA, especially, in Atonsu and Dakwadwom. Also, at all the study suburbs, chiefs and land

owners were rated high in the management of floods and wetlands. In Ahensan, respondents

completely agreed that the chiefs and landowners were responsible for the wetlands. In

Spetinpom, all the institutions listed as being responsible for managing wetlands and floods

were ranked relatively high by community members.

Apart from the Assemblyman, the other stakeholders were given the opportunity to assess

each other on wetlands protection and management and flood prevention and management.

The outcome of these assessments is presented in Table 4.2. Among the different

stakeholders, NADMO and the KMA were considered the most financially resourced to

manage floods and wetlands. NADMO however lacked the technical expertise at local levels.

The chiefs also lacked the technical and coordinating capacity in wetlands protection and

management and flood prevention even though they were considered very much respected

and could sometimes mobilise the financial resources.

Figure 4.1. Assessment of stakeholders’ responsibility for the management of floods and wetlands in Kumasi

75

4.3.1.1 Kumasi Metropolitan Assembly (KMA)

According to the Local Government Act (Act 462 of 1993), the KMA is the local authority of

the metropolis. This study revealed that decentralised structures for environmental

management in Kumasi are well in place. Flooding issues had been incorporated in the

Medium Term Development Plans of the Assembly (spanning 2006-2009). According to the

Assembly, the floods are caused by residential dwellings and other structures built in

waterways, dumping of refuse in gutters and drains and the inability of existing culverts to

receive large volumes of water whenever there is a heavy downpour. Some mitigating

strategies such as drain constructions or enlargement of existing ones, desilting or dredging of

streams, clearing of waterways (demolition of obstructing structures) and public education

were being carried out. Several houses along waterways have therefore been marked for

demolition. This was an on-going exercise in all the flood prone suburbs like Aboabo,

Ahensan, Anloga, Asafo, Asokwa, Atonsu, Breman, Daban and Oforikrom. According to the

KMA, plans were also in place to curb further building on waterways through strict

enforcement of the building regulations.

Respondents from the communities had varying perceptions about the demolition exercises

carried out by the KMA (Figure 4.2). Almost 50 % of the respondents in Aboabo and

Ahensan had houses in their neighbourhood marked for demolition and virtually none in

Dakwadwom. On the contrary, whilst there is a general support by respondents for demolition

in the other suburbs, no respondent in Aboabo supported it. Most support for demolition of

houses on waterways came from Atonsu.

4.3.1.2 Town and Country Planning Department (TCPD) The TCPD is a government agency responsible for planning and management of the orderly

development of human settlements; providing planning services to public authorities and

private developers and provision of layout plans to guide orderly development in Ghana. It

formulates goals and standards relating to the use and development of land in areas of rapid

urban growth. This Department is present in the Metropolis as one of the decentralised

government institutions. In performing these functions, the TCPD plays a key role in

delineating and protecting wetlands and waterbodies and the prevention of floods in built up

areas.

Apart from Spetinpom and Kwadaso Estates where the TCPD is recognised as responsible for

floods and wetland management, all the other suburbs virtually did not know the TCPD.

76

4.3.1.3 National Disaster Management Organisation (NADMO) NADMO is responsible for the coordination of the management of disasters and similar

emergencies. It ensures that the country is prepared to prevent disasters and manage them well

when they occur. It does this through training, education and awareness creation, coordinating

the activities of various stakeholders in the management of disasters and rehabilitates persons

affected by disasters. NADMO is very well decentralised to regional and district offices. Both

Regional and Metropolitan offices of NADMO are found within Kumasi.

Among all the institutions, NADMO is the most popular with the respondents as being

responsible for floods and wetlands management. This is especially so in Atonsu and

Dakwadwom where respondents virtually assigned all floods and wetlands management

responsibilities to NADMO.

Table 4.2. Stakeholder capacity assessment on wetlands and flood prevention and management in Kumasi

Stakeholder Capacity Rating (increasing from left to right) Comment

1 2 3 4

KMA

Personnel X The KMA knows what to do and how to do it. What is not available are the finances and logistics to carry out their plans. The KMA has other priorities and only reacts in times of flood.

Technical/logistics X Financing X Coordination X

NADMO

Personnel X NADMO has a budget that is usually increased in times of emergency. The organisation is however very politicised and lacks the qualified technical personnel.

Technical/logistics X Financing X Coordination X

TCPD

Personnel X This organisation is well staffed in Kumasi. They however do not coordinate with other stakeholders in their landuse planning mandate.

Technical/logistics X Financing X Coordination X

HSD

Personnel X This is very much an obscure organisation in function and visibility in the KMA. They are relatively under-resourced and much marginalised.

Technical/logistics X Financing X Coordination X

Chiefs

Personnel X They are very important in landuse decisions in Kumasi. Sale of land is a very big source of finance for this institution. Their decisions lack technical backing from the various experts.

Technical/logistics X Financing X Coordination X

NGO (Friends of Rivers and

Waterbodies)

Personnel X Has the personnel and capacity to coordinate wetland management issues. They are well known and resented in some of the suburbs of the KMA

Technical/logistics X Financing X Coordination X

Total and general assessment 7 3 4 10

Generally, the Metropolis has the technical capacity but does not coordinate its flood and wetland management efforts with the little logistical and financial resources available.

77

Figure 4.2. Spatial distribution of houses marked for demolition and attitude towards demolition of houses in waterways by the Kumasi Metropolitan Assembly

78

4.3.1.4 Hydrological Services Department (HSD) The HSD is responsible for assessing the country’s water resources (quantity, quality and

distribution in time and space), the potential for water related development, and the ability to

supply actual and foreseeable demand. The HSD also assesses the environmental, economic

and social impacts of existing and proposed water resources management practices, projects

and the adoption of sound policies and strategies. Above all, they assess the impacts of other

non-water sector activities such as urbanisation, waste management, forest harvesting etc. on

water resources thereby providing security for people and property against water–related

hazards such as floods, storm surges and droughts.

Among respondents, this is the most unknown stakeholder for the management of floods and

wetlands. The HSD and its functions are not known in Aboabo, Atonsu and Dakwadwom.

Like the other institutions, it was only in Spetinpom that the HSD was known and considered,

more than 50 % of the time, responsible for floods and wetlands management.

4.3.1.5 Chiefs and Assemblymen Chiefs or the chieftaincy institution is a traditional leadership arrangement which is

guaranteed in the Constitution of Ghana. The Constitution also recognizes customary law as a

source of law in Ghana and explicitly states that the management of stool lands are to be in

accordance with the relevant customary laws and usage. Majority of the land of the

Metropolis is customary or stool lands and customary law vests the powers in the appropriate

stool (chief) to manage such lands on behalf of and in trust for their people.

The assemblymen on the other hand, are a product of the local government system

representing decentralised decision making bodies at the community level. Officially, they

represent the community in the KMA. In most communities, the assemblymen work very

closely with the chiefs as community leaders to coordinate community development activities

and relief efforts in emergency situations.

Apart from Atonsu and Dakwadwom, the roles of chiefs in wetland and flood management

are very much recognised by all the stakeholders. Respondents in Ahensan and Spetinpom,

almost 100 % of the time, think chiefs are responsible for the management wetlands and

floods. The Assemblyman was rated higher than the chief in solving flood issues and the

provision of relief. The rating of an assemblyman in any community depends on the

dynamism and efforts the person puts in solving community problems. The assemblyman and

the chief will usually be the first line of contact for person(s) or institution(s) coming to the

79

community or bringing assistance in times of flood. Similarly they collaborate and mobilise

communal labour for various flood interventions.

4.3.1.6 Non-governmental organisations (Friends of Rivers and Waterbodies) Several non-governmental organisations (NGO) have interest in water and watershed

management in Ghana. One local NGO that has risen to prominence in the Kumasi Metropolis

is the Friends of Rivers and Waterbodies. The Friends of Rivers and Waterbodies is an

amalgam of volunteer scientists, businessmen, economists, opinion leaders and other

professionals as well as students. Volunteers are mobilised and empowered to execute tasks in

consonance with traditional and scientific values, to provide water and protect waterbodies

from pollution, siltation and abuse. A playground in Dakwadwom called Otumfuo Children’s

Park, was established by the Friends of Rivers and Waterbodies on one side of the wetland.

The NGO has also undertaken several wetland related activities in the community.

Respondents from Dakwadwom therefore rated NGOs high for management of floods and

wetlands. All the other suburbs rated the NGOs relatively low.

4.4 DISCUSSION In spite of the long list of institutions and environmental laws, Ghana’s urban environment is

one of the worst in the sub region. The urban environmental issues described above are

serious and damning to Kumasi and Ghana as a whole. According to the FAO (1985) few

sewage treatment plants existed in Ghana. Since this publication, the situation has rather

become worst because of the increase in urbanisation and little accompanying infrastructure

development (Brook & Dávila, 2000; Keraita et al., 2003; Fobil & Atuguba, 2004; UN-

HABITAT, 2009). Therefore, most of the raw domestic effluents are discharged directly into

water bodies. Information on the extent of pollution caused by domestic wastes on water

bodies exists only for the main rivers and reservoirs that serve as drinking water sources for

communities. This is well captured in a recent United Nations Human Settlements Programme

report that, Ghana’s environmental challenges and priorities are urban planning and

management, housing/urban development and the environment (UN-HABITAT, 2008b).

These environmental issues are the result of conflicts between human and environmental

interests due to urbanisation that has gone haywire. Also, the environmental framework based

on which one can evaluate these issues will be difficult to define because Ghana has no urban

policy. This discussion will therefore be compared to best practices elsewhere and an attempt

made at relating it to local conditions.

80

Appending signatures to conventions and treaties seem not to be a problem for Ghana

irrespective of the obligations and commitments required. This is because the translation of

the obligations of conventions into local laws and policies for implementation and

enforcement has not been followed through to the letter. The divide between policy and

implementation is evident in the contrast between the number of national and international

obligations and the unsanitary conditions and the increasing degradation of wetlands. For

example, Ghana ratified the Ramsar Convention in 1988, demonstrating her willingness and

commitment to reverse wetland loss and degradation. However, it took more than a decade

afterwards to fulfil the requirements of the Convention by recognising wetlands in the

National Land Policy (Ministry of Lands and Forestry, 1999b) and to develop a strategy for

management and sustainable utilisation of wetlands and their resources (Ministry of Lands

and Forestry, 1999a). Now, Ghana has to also fulfil the operational objectives of the Ramsar

Strategic Plan 2009-2015 which among others, required all parties to complete their national

inventories and have a web-based database; put in place a national wetland policy and

management plans founded on sound scientific research and identify priority sites for

restoration. Since becoming a member of this Convention, research on wetlands that could

help meet this operational objective has barely been conducted. Most wetland research has

been on the Volta and Densu Rivers because of their hydroelectric use and provision of

drinking water to Accra respectively.

Even though the environmental governance system is supposed to be decentralised according

to the Constitution and Local Government Acts, the extent of decentralisation is not the same

across the country. Whilst the older and richer districts have the various decentralised

government agencies, most of new or poor districts are still under the oversight of the regional

level agencies. There are, therefore, differences in availability of expertise and even

representation on the DEMC. For example, the EPA, TCPD, Forestry Service Division and

the Land Commission are not present in most districts in Ghana. These differences give multi-

dimensionality to the already complex problems associated with the urban environment. Also

prominent in most districts that are fortunate to have these agencies, e.g. the KMA, is the

conflict of interests among these agencies. For example, Forestry Service Division, EPA,

Department of Agriculture, District Directorate of Health Services, have overlapping

functions and sometimes try to outmanoeuvre each other into approving or disapproving some

environmental issues. This has been aggravated by high staff turnover, frequent changes in the

institutions' mandates, the complexity of the institutional framework and weak coordination.

81

The local government system practised in Ghana is very much in line with international best

practices of participatory governance and bottom-up approaches. Also, according to all the

laws and policies, environmental action in Ghana should be bottom-up. In the spirit of the

Constitution and policies of Ghana, the use of the term ‘Government’ at the local level on

environmental issues refer to the DEMC. This is because environmental protection and

management is a constitutional mandate of every citizen. This mandate is exercised through

local level environmental management and decision making through the DEMC. This is

expected to create ownership and empowerment for the sustainability of environmental

programmes by taking cognisance of local needs and aspirations (Okley, 2004). It is however,

unfortunate that the DEMCs are either non-existent or non-functional in most assemblies. At

the KMA, the EPA has overshadowed this Committee.

Unless land that has been acquired and protected for the public good, the system of land

ownership in Ghana does not augur well for the environment. Urban land value is always on

the ascendency. For example, wetland areas gain value due to urbanisation and improved

construction engineering feats. Like all other jurisdictions, there is market failure in the

environmental services of these wetlands. Wetland uses for residential and commercial

developments therefore often out-compete their use for environmental services as a landuse

priority. Urban wetland areas are therefore very popular sites for fuel stations in Kumasi and

other urban areas of Ghana. Rather than protecting, the EPA’s regulations for siting such

businesses promote the use and destruction of urban wetland areas. Also, most of the

institutions, especially, the DEMC, are non-functional when it comes to enforcing and

implementing the planning schemes for these urban areas.

The rate of urbanisation of Kumasi has overwhelmed the authorities. Settlements are

constructed with layouts from landowners before the TCPD develops or approve the chief’s

layout for the suburbs. In these instances, the wetlands are usually long destroyed by the time

these state institutions start to work. This is because the constitutional powers of governments

to protect the public welfare and the political acceptance of existing environmental

regulations are generally difficult to enforce (Lant, 1994). Albeit different geographical,

economic and social context, Sims & Schuetz (2009) showed that local regulations and

bylaws slowed wetland conversion to residential uses in Massachusetts. In the case of Kumasi

and Ghana at large, a strong DEMC and TCPD working in collaboration with the Chiefs and

other landowners could be a panacea to the wetland loss and environmental management.

82

From the described conditions of Kumasi, the future environment will largely consist of

human-dominated ecosystems, managed to varying degrees, and become harder and harder to

maintain in the poor suburbs. The victims of this looming crisis are both the state and the

persons at these wetland sites. Already plenty resources are spent yearly in supporting flood

victims in Kumasi. Also, the capacity rating of flood and wetland stakeholders has

implications on their level of responsiveness to the affected communities. With the limited

resources, there is the need to strengthen stakeholder collaboration and coordination.

Importantly, though NADMO has been providing relief items and materials to flood victims,

it is usually for immediate and short time relief. It is rather more important for these

stakeholders to focus on the reduction of socioeconomic vulnerability of community members

and the promotion of the culture of preparedness of both the authorities and community

members. This improves our collective resilience to floods.

From the information gathered, a general model for management of floods and wetlands

within the Kumasi Metropolis has been proposed in Figure 4.3. Whilst an integrated and

collaborative approach between stakeholders is advocated, the model takes cognisance of the

strengths and weaknesses of stakeholders in addressing the various flood management

elements. All these elements have to be treated with equal importance. What differ are the

scale and the dimensions of the issues. Based on the findings of this research, different levels

of emphases are proposed for the different dimensions or issues. The proposed order in

decreasing significance of the issues is: social, technical, institutional, environmental and

legal. With the limited financial resources available to the KMA, flood management within

the metropolis should focus on the social and technical elements to achieve better impacts.

The current level of economic development of Ghana is also a contributory factor to

ineffective implementation or non-enforcement of environmental laws. Most obviously, law

enforcement has financial implications and therefore will depend on the wealth of the country.

Developed countries may be more capable and willing to comply with their environmental

obligations than poorer countries because implementation places relatively less burden on the

national cake. On the other hand, if the cost of implementation is very low, or if

implementation produces a net economic gain, a country’s level of economic well-being may

not be very critical for its success. This is because, the decisive factor is the relationship

between the cost of implementation and the capacity of a country’s economy to absorb them.

83

Fi

gure

4.3

. A

con

cept

ual

mod

el f

or m

anag

emen

t of

flo

ods

and

wetla

nds

with

in t

he K

umas

i M

etro

polis

(th

e th

ickn

ess

of t

he l

ine

is a

re

flect

ion

of th

e im

port

ance

of t

he is

sue

or in

terv

entio

n or

inst

itutio

n)

Dim

ensio

ns

of

flood

man

agem

ent

Tech

nical

Le

gal

Stak

ehol

ders

Floo

d M

anag

emen

t El

emen

ts

Dim

ensi

ons

of

flood

man

agem

ent

Land

use

plan

ning

Re

stora

tion

or

prot

ectio

n of

ur

ban

gree

ns o

r wet

lands

Floo

d pr

epar

edne

ss

and

adap

tatio

n str

ateg

ies

Stre

ngth

en

educ

atio

n an

d law

enf

orce

men

t

Stor

m

wat

er

drain

age

Solid

w

aste

m

anag

emen

t

Insti

tutio

nal

Envir

onm

enta

l

Tech

nical

NG

Os

Chie

fs

TCP

HSD

N

AD

MO

KM

A

Socia

l

84

Apart from a nation’s economic situation, national environmental consciousness and

institutions are also very necessary for the protection and management of wetlands. Ghana

needs the birth of a green party. In many developed countries, rising environmental

consciousness has given birth to green parties (or the vice versa) that are revolutionising the

political landscape with green politics by putting environmental issues to the fore in national

policy debates. The challenge now lies as to whether it is the economy or the people that give

rise to green politics. If it is the economy, then Ghana needs to reach a certain level of

economic independence for businesses to go green and still survive the competition. The

Ghanaian economy and businesses are not yet there. If it is the people, there is the need for a

critical mass of people to have a level of dissatisfaction with the environment. That also,

Ghanaians are not yet there. The option left is for the developed countries to dictate the pace

towards a better urban environment through strings tied to bilateral and multilateral assistance

or introduction and enforcement of environment related trade barriers and conditionalities.

85

5 WETLAND ECOLOGY AND VEGETATION GEOGRAPHY IN THE KUMASI METROPOLIS

5.1 INTRODUCTION Human settlements, like other systems, are under constant change involving growth, decline,

aesthetic changes etc. depending on the age of the settlement, demographic trends, planning

and resources available. Urbanisation is currently the biggest change most developing country

settlements are undergoing (Bhattacharya, 2002). It is both a driver and outcome of economic,

political, cultural, social and physical processes. The ecological consequences of the

increasing built environment are pollution and landscape changes resulting in increased range

of spatial diversity from ‘biological deserts’ with paved or gravelled surfaces or manicured

monospecific lawns to biological hotspots like urban wetlands and abandoned industrial sites

and rail lines (Niemelä, 1999). This is because urbanisation and the associated environmental

stress lead to fragmentation of natural vegetation and changes in their structure and floristic

composition (Altobelli et al., 2007). Urbanisation also alters local climate and air quality

through increased concentrations of oxidants (O3, SO2), nutrients and chemicals (nitrates,

cations especially heavy metals), decreased net radiation and average wind speed, increased

cloudiness and precipitation and temperature (“heat island” effect). These changes imply that

urban sites cannot be directly compared with non-urban sites, any more than sites in different

climatic zones (Ehrenfeld, 2000). Therefore the wetlands of Kumasi were considered as

unique domains separate and different from the adjoining peri-urban or non-urban wetlands or

urban wetlands elsewhere in the country.

Most urban ecological research has focused on contributions of greenhouse gases and the heat

island effect (Hidalgo et al., 2008; Takebayashi & Moriyama, 2009). Urban wetlands on the

other hand, deliver a wide range of ecosystem services relating to climate change mitigation

and or adaptation, water catchments, buffer against flooding, soil purification, water and air or

act as carbon sinks and contribute to human well-being (Joosten & Clarke, 2002; Berkowitz et

al., 2003; Janssen et al., 2005; Nelson et al., 2009). Despite their ecological and

socioeconomic importance, urban wetlands have been understudied in the analysis of global

environmental change and flooding. Urban wetland areas are usually characterised by

vegetation patches of different sizes and nature, depending on the landuse options developed

by planners during the urbanisation process or patches left out in the landuse planning. The

vegetation is often affected by the hydroperiod or hydrologic changes and availability of

nutrients. Also, toxic materials in the runoff can interfere with biological processes of wetland

86

plants, resulting in impaired growth, mortality, and changes in plant communities. For

example, direct storm sewer inputs affect plant community structure and vegetation dynamics

(Ehrenfeld, 2000). Plants, as individuals or as communities, could therefore be valuable

indicators of water regime and other environmental conditions (Haslam, 2003).

Reinelt et al. (1998) in a study on the impacts of urbanisation on palustrine wetlands found

that, changes in plant community composition may describe impact of pollution. In other

studies, it was found that contrary to that of Reinelt et al. (1998), vegetation types of different

floristic composition or vegetation cover or density are rather predetermined by

geomorphology, soil conditions, and water regime (Federal Interagency Committee for

Wetlands Delineation, 1989; Barrat-Segretain & Amoros, 1996; Coroi et al., 2004;

Hejcmanovā-Nežerková & Hejcman, 2006; Sorrell et al., 2007). On a large scale, factors such

as soil and climatic conditions may be the determining factors that control the distribution and

abundance of vegetation. However, on a small scale, while the soil and climatic conditions

may not be spatially varied, the vegetation may be very varied. These spatial differences in

species distribution are pronounced in the wetlands of Kumasi. This chapter therefore seeks to

identify the underlying factors that determine the wetland vegetation type and distribution in

Kumasi. Knowledge of the characteristic vegetation is useful in delineating wetland

boundaries and reduces incidence of conflicts associated with landuse planning and wetland

encroachments. This will be an input towards restoring Kumasi to its Garden City status.

5.2 METHODOLOGY

5.2.1 Reconnaissance Survey of Wetlands in Kumasi and Selection of Study Sites The wetlands in Kumasi were identified and delineated using the definition of Joosten &

Clarke (2002). Following a detailed study of drainage maps of Kumasi and use of aerial

photography, physical inspection and ground truthing of these drainage channels was done

using a handheld GPS. Detailed descriptions of landcover and landuse types were recorded

and photographs of salient features taken. The data collected was then be used to validate the

information gathered from the digitised drainage maps and aerial photography.

Based on the information from this survey, some of the major wetland sites of ecological,

landscape, economic and social interests were chosen for the study. Wetlands of ecological

interests included, but not limited to, areas with at least 50 m of wetland vegetation on both

sides of the water channel. Areas with economic and social activities directly affecting or

bordering the wetland (e.g. tie-dye industries, car wash, agriculture, settlement etc.) were

87

selected as sites of economic interests. These sites were selected for detailed socio-economic

studies as described in earlier chapters.

5.2.2 Ecological Survey of Wetlands in Kumasi The various wetlands of interest for this study were selected as described in section 3.2 above.

For each wetland area, e.g. along a stream X, three permanent transects A, B and C were

marked out not less than 50 m apart based on the ecological, landscape and surrounding

economic and social activities (Figure 5.1). The GPS coordinates of these transects were

noted to aid future identification of these sites if markers are removed. Within the permanent

transects detailed ecological studies were carried out at various sections of the transect (Coroi

et al., 2004; Barrat-Segretain & Amoros, 1996).

Figure 5.1. A schematic diagram of a sample study site along a stream X showing the transects and quadrat layout

5.2.2.1 Soil sampling A graduated soil corer of diameter 5.7 cm was used to take soil core samples to a depth of

30 cm (Driessen et al., 2001, Maltby & Barker, 2009) and within 10 m from the water

channel. Each sediment core was carefully studied and described. The sample was then put

into labelled sterile polythene bags and stored in ice chests for further laboratory analysis.

Also, as a proxy for testing soil consistency, the corer was pushed into the soil by hand at

these locations to estimate the resistance to penetration by measuring the total depth of

88

insertion. In the laboratory, a portion of the soil sample was taken and the texture determined

using the Bouyoucos hydrometer method of mechanical analysis (Beverwijk, 1967). The

textural classes were based on the US Department of Agriculture soil particle size ranges.

The total organic matter was estimated from these ice cooled samples using the loss on

ignition method. Each sample was first treated with excess of 10 % HCl to remove any

carbonates that might be present. The water was first removed by air drying in an oven for 24

hours at 70 °C. The dried sample was ground with an agate mortar and sieved through a

1000 µm sieve to homogenise it. A known weight was then placed into a preheated furnace at

500 °C for 2 hours. After ignition, 2-3 drops of (NH4)NO3 was added and the samples put

back into the furnace for 10 minutes in order to remove any unburned organic carbon. The

total organic carbon in each sample was the weight lost on ignition.

5.2.2.2 Stream sediments and physicochemical characteristics Along each permanent transect the presence or absence of water at the surface of each of the

sampling points was noted monthly. The total duration of surface water during the year was

used to assess the effects of wetland hydroperiod on vegetation dynamics.

In each stream, temperature (bottom), turbidity, conductivity, total dissolved solids, dissolved

oxygen and pH of the water were measured using a Hanna multiparameter water quality

portable meter (HI 9828) during the peak rainy season and dry season. Surface sediment

samples were also taken at a depth of 0-15 cm (U. S. Environmental Protection Agency,

1985) and quickly packed in air tight polythene bags to be analysed for various heavy and

trace metals. In the laboratory, samples were air dried in a vacuum oven at 70 ºC for 3-4 days

until constant weight and ground in an agate mortar for homogenisation. Approximately 0.2 g

of dried sample was extracted overnight with 2 ml concentrated nitric acid (HNO3) and

evaporated until near dryness and no nitrous vapours were released. After cooling, 5 ml of an

acid mixture made of concentrated acids HF:HClO4:HNO3 at 6:2:3 ratio was added. The

digestion continued until no coloured vapours were released. The atomic absorption

spectrophotometer (Varian SpectrAA-200FS) was used for the determination concentrations

of Lead (Pb) Mercury (Hg), Copper (Cu), Zinc (Zn), Cadmium (Cd), Chromium (Cr), Nickel

(Ni) and Arsenic (As) and photometric determination used for Phosphorus (P).

Assessment of contamination in the sediments This was done using three parameters: enrichment factor (EF), geo-accumulation index (Igeo),

and pollution load index (PLI).

89

a. Determination of heavy metal enrichment factor (EF)

Iron (Fe) was chosen as the element for normalisation of the enrichment because natural

sources vastly dominate its input in the environment (Schiff & Weisberg, 1999; Ghrefat &

Yusuf, 2006; Sekabira et al., 2010). To obtain the natural Fe concentration in the wetland

sediments of Kumasi, three sediment samples were taken from different locations in the

KNUST Botanical Gardens and pooled. The KNUST Botanical Gardens was chosen because

of its remoteness from known or suspected anthropogenic metal sources. World geochemical

background values in average shale as reported by Turekian & Wedepohl (1961) were used

for all the metals. The EF for each metal was computed as the relative abundance of species in

the sediment to that found in the Earth’s crust (Simex & Helz, 1981) with the formula:

EF = ��� (������)

��� (� ��ℎ′� �����)� Where X is concentration of the metal studied.

Six categories of enrichment were recognised: <1 background concentration, 1-2 depletion to

minimal enrichment, 2–5 moderate enrichment, 5–20 significant enrichment, 20–40 very high

enrichment and >40 extremely high enrichment (Duzgoren-Aydin et al., 2006; Harikumar et

al., 2010).

b. Determination of index of geoaccumulation (Igeo)

The Igeo values were calculated for the different metals using the formula of Müller (1969)

expressed as: Igeo = log2(Cn/1.5*Bn).

Where Cn is the measured concentration of element n in the sediment sample and Bn is the

geochemical background for the element n. The factor 1.5 is introduced to include possible

variation of the background values that are due to lithogenic variations. Müller (1969)

proposed seven sediment quality classes associated with the geoaccumulation index as given

in Table 5.1.

Table 5.1. Classes of the geoaccumulation index used to define sediment quality Igeo value Igeo Class Quality of sediment ≤ 0 0 Unpolluted 0-1 1 From unpolluted to moderately polluted 1-2 2 Moderately polluted 2-3 3 From moderately to strongly polluted 3-4 4 Strongly polluted 4-5 5 From strongly to extremely polluted ≥ 6 6 Extremely polluted

90

c. Determination of pollution load index (PLI)

The PLI is an indicator of site quality. For each metal, the contamination factor (CF) was first

determined using the formula, CF = metal concentration in sediments/background values of

the metal. The extent of pollution by trace metals was assessed using the PLI formula

developed by Tomlinson et al. (1980): PLI = (CF1 * CF2 *.....* CFn)1/n

Where CF = contamination factor and n= number of metals. PLI value >1 is polluted whereas

PLI value < 1 indicates no pollution.

5.2.2.3 Vegetation sampling To fulfil the requirements of the Ramsar Convention Secretariat (2005), the systematic

sampling approach described by Orloci (1978) was used. At each selected wetland site,

starting from the edge of the permanent water channel, 1 m x 1 m quadrats were marked along

each transect. Sampling was done at 10 m intervals. All herbaceous plants and shrubs in the

quadrats were identified to species level with their respective relative abundance and cover

recorded (see Appendix 2) using the Braun-Blanquet scale described in Table 5.2. Cover was

determined from estimates of vertical plant shoot-area projection as a percentage of quadrat

area (Wikum & Shanholtzer, 1978).

The plants were identified with the assistance of a technician and the use of vegetation

manuals by Akobundu & Agyakwa (1998), Cook (1996), Haflinger & Scholz (1980a, b),

Haflinger et al. (1982) and Terry (1983). Plants that could not be identified on the field were

collected and later identified in the laboratory. Around each wetland area, the various

anthropogenic activities up to 100 m from the water channel were noted.

Table 5.2. The Braun-Blanquet scale and the description of the values used for sampling of wetland species in Kumasi

Braun-Blanquet scale

Range of cover (%) Relative abundance Midpoint of over

range (%) 5 75-100 very common, usually single spp.

dominant 87.5

4 50-75 62.5

3 25-50 common, dominance is shared by 2 to 3 other species. 37.5

2 5-25 frequent, a species not sharing dominance but remains significant to the composition of the site

15.0

1 <5; numerous individuals occasional occurrence 2.5

+ <5; few individuals

rare, individuals infrequently seen, low count and crown cover 0.1

91

Sampling was done in June-July 2010 (peak rainy season) and December, 2010-January 2011

(peak dry season). The presence or absence of water on the surface at each sampling site was

noted. Variations in species composition were estimated using the importance value index for

all study sites using the following formulae:

i. density= ������ �� ������� ����! �!��"�#

ii. frequency= ������ �� �"�$� �% &'��' ������� � ������*�$!" %����� �� �"�$� �!��"�#

iii. dominance= *�$!" ��+�� �� ������� � ���! �!��"�#

iv. relative density= ,�%��$- �� ������� �*�$!" #�%��$- �� !"" ������� × 100

v. relative frequency= ���.��%�- +!"�� ��� ������� �*�$!" �� !"" ���.��%�- +!"��� ��� !"" ������� × 100

vi. Relative dominance= ,���%!%�� ��� ������� �*�$!" #���%!%�� ��� !"" ������� × 100

Therefore, importance value index = relative density + relative dominance + relative frequency.

5.2.3 Data Analysis Differences among sampling sites along each stream and sampling periods were found by

means of analysis of similarities (ANOSIM) test at α= 0.05. The various indices of diversity

(Shannon, species evenness and species richness) were used to assess the plant community of

each wetland. The non-metric multidimensional scaling (MDS), cluster analysis and principal

components analysis (PCA) were used to explain species richness/distribution patterns and

environmental parameters that influence/determine these patterns. The BIO-ENV procedure

was used to examine the extent to which the soil texture, physicochemical water parameters

and the total organic matter are related or explain (using the Spearman rank correlation

coefficient ρ) the distribution and type of vegetation in each wetland. The Pearson correlation

analysis was used to analyse and establish associations between heavy metals of wetland

sediments and other variables after the data was transformed and normalised. A PCA was

carried out on the correlation matrix to reduce the dimensionality of the problem and enable

an interpretation of the relationships between variables and sampling sites.

5.3 RESULTS

5.3.1 Anthropogenic Activities in the Wetland Areas of Kumasi The commonest anthropogenic activities identified within 100 m from either side of the

channel are houses, schools, churches, fuel/gas filling stations, farms, car washing bays and

repair shops (Figure 5.2, Photograph 1 and Appendix 3). Activities with the highest land

92

cover are housing and urban agriculture (vegetable farms). The wetlands along the Sisai River

from Family Chapel through Ahensan to Atonsu Sawmill (Sites A, B and E) have the highest

anthropogenic activities associated with them. The predominant anthropogenic activities at

these sites are houses, carpentry shops, car wash and repairs centres and waste disposal sites.

The car repair centres are usually a cluster of wooden shelters serving as garages which do

general repairs, engine oil change, vehicle body building and spraying. A lot of wastes from

these activities are washed directly into the water channel. Natural vegetation cover in these

sites is as low as 30 %.

All the sites have some vegetable farms associated with them. These are commonly small

plots of about 0.25 ha with carrots, cabbage, onions, green pepper and okro. The water from

the channel is used to irrigate the fields to provide an all year round cultivation. Fertilizers,

weedicides, insecticides and pesticides are used to increase yields and quality. The farming

activity at the wetland stretching from the Bekwai to Santasi roundabouts is on a relatively

large scale. Crops cultivated here include maize, oil palm, pawpaw, banana and plantain. Only

about 20 % of the 100 m buffer area is natural wetland vegetation. Houses, infrastructure and

public gardens or space occupy over 75 % of the total wetland area.

93

Figu

re 5

.2. L

andc

over

and

Lan

duse

act

iviti

es w

ithin

100

m fr

om t

he s

trea

m c

hann

el a

t the

diff

eren

t st

udy

site

s in

Kum

asi

94

5.3.2 Wetland Edaphic Characteristics, Heavy Metals and Stream Physicochemical Characteristics

The soils in Kumasi developed from two main primary materials (Obeng, 1971). The western and

north-eastern to south-eastern parts of the city have soils that developed from Cape Coast granites

to form the Kumasi-Asuansi/Nta-Ofin/Bomso-Asuansi compound associations (Figure 5.3). The

wetlands found around the KNUST Police Station (D), Faculty of Renewable Natural Resources

farms (C), High School junction (B), Kaase Guinness (H), the Family Chapel (E), Spetinpom (F)

and Bekwai roundabout (J) belong to this soil category. The soils stretching from the central to

south-western Kumasi developed from the lower Birimian phyllites, greywackes, schists and

gneisses primary materials. The soil groups from these primary materials form the Bekwai-

Nzima/Akumadan-Bekwai/Oda complex associations. The wetlands around Dakwadwom (K),

Family Chapel (E), Golden Tulip Hotel (G) and Atonsu Sawmill (A) belong to this soil category.

The site specific site edaphic conditions are described in Table 5.3. The upper layers of the soils

were generally dark brown and deeper layers were yellowish-reddish brown. Mottles were very

common in wetlands like Atonsu and Spetinpom wetlands. The texture types were mainly loamy

from forest ochrosols and lithosols. All the sites were waterlogged from a minimum 6 months to

being permanently waterlogged. The wetland at the KNUST Police Station had the highest organic

matter content of 34.5% and the whole length of the corer was easily inserted into the soil. At the

Atonsu, Spetinpom, Golden Tulip Hotel and Kaase Guinness wetlands, the corer could hardly

penetrate up to 20 cm and these sites also had comparatively higher proportion of sand and low

organic carbon.

There were spatial and temporal variations in physicochemical parameters measured (Table 5.4).

During the peak rainy season, all the study sites were weakly acidic with pH between 5.4 at Atonsu

to 6.4 at the Golden Tulip Hotel wetlands. The TDS and conductivity were the most varied in the

selected wetland sites. The conductivity of water was directly related to the TDS (conductivity ≈ 2

TDS) for all sites. The water at Atonsu and Kaase Guinness had relatively high TDS of 411 and

423 mg/l and salinity of 0.43 and 0.52 ‰ respectively. The water at the KNUST Police Station also

recorded the least salinity. The channel bottom water temperature, dissolved oxygen were less

varied between sites.

95

Figure 5.3. Soil map of the Kumasi showing study sites [sites are labelled in letters: Atonsu Sawmill (A), High school junction (B), FRNR farms (C), KNUST Police Station (D), Family Chapel (E), Spetinpom (F), Golden Tulip Hotel (G), Kaase Guinness (H), Bekwai roundabout (J) and Dakwadwom (K)] (Source: Obeng, 1971).

These observations were not different for some parameters for the dry season. The conductivity was

also approximately twice the TDS for all the wetlands. For this reason, only the TDS was used in

the subsequent analysis to eliminate the compounding effects of using both parameters. During the

dry season however, pH was generally alkaline with a range from an almost neutral 6.88 at the

KNUST Police Station to 10.6 at site the High School Junction. The biggest seasonal differences

were in dissolved oxygen concentration. For example, dissolved oxygen at sites Atonsu, High

96

School Junction, KNUST Police Station, Kaase Guinness, Bekwai roundabout and Dakwadwom

declined from 4.20, 5.50, 3.30, 4.90, 4.50 and 4.30 mg/l in the rainy season to 0.45, 2.05, 0.62, 0.16,

1.48 and 0.75 ml respectively in the dry season. The non-parametric Wilcoxon Test was used to

assess if there are any seasonal differences in the physicochemical parameters at the various

wetland sites (Appendix 4). The null hypothesis of no seasonal differences was rejected for pH and

dissolved oxygen (α=0.05, p-values = 0.005 and 0.009 respectively).

The concentrations of metals in the different wetland sites investigated are presented in Appendix 5.

Heavy metal concentrations were generally highest in Atonsu Sawmill, Family Chapel and the High

School Junction. The KNUST Police Station had relatively lowest concentrations in all heavy

metals except total phosphorus. The association between these metals and some of the measured

environmental parameters was assessed using the Pearson correlation coefficient (Figure 5.4 and

Appendix 6). The test showed a significantly positive correlation between Hg, Ni, Zn, pH and Pb.

Other clusters of significant correlations are clay, Cu and Cr; dissolved oxygen and salinity. The

percentage of silt in the soil was negatively correlated to pH and Cr; dissolved oxygen and salinity.

To reduce the dimensionality of variables that may be of significance in this study, a PCA was

carried out on the data (Figure 5.5 and Appendix 7). The first two principal components accounted

for 52.9 % of the variation. The major contributors to the first principal component axis are Pb, Zn,

Ni, Hg and pH and those to the second principal component axis are temperature, salinity, dissolved

oxygen and dissolved solids.

The extent of sediment contamination by these heavy metals was further evaluated using the EF,

Igeo and the PLI. All these indices were found to be significantly different between the seasons

(Appendices 8, 9 and 10). The FRNR farms and Spetinpom sediments that did not have significant

enrichment in any heavy metal in the rainy season whilst the KNUST Police Station had very high

enrichment in only phosphorus (Figure 5.6). The Family Chapel, Atonsu Sawmill and High School

Junction sediments had significant enrichment in all metals except cadmium. Sediments from

Family Chapel had very high enrichment of P and Pb and that from Atonsu Sawmill also has very

high enrichment of Pb. Heavy metal enrichment is higher for all sites in the dry season than the

rainy season (Figure 5.7). Very high enrichment was recorded at various study sites for P, Cu, As

and Pb.

According to the Igeo for both seasons (Figures 5.8 and 5.9), most heavy metals were accumulated in

sediments of Family Chapel, Atonsu Sawmill and High School Junction. In the rainy season (Figure

5.8), only Pb, Cu, P and Cr caused pollution class 4 (i.e. sites were strongly polluted by these

metals). In the dry season (Figure 5.9), these sites were extremely polluted by these metals. Extreme

97

pollution was reported of P and (or) Pb in the sediments of the wetlands at Atonsu Sawmill, High

School Junction, KNUST Police Station, Family Chapel and Dakwadwom.

The PLI was used for an overall assessment of the extent of pollution of each site from the

cumulative contributions of all the substances studied. By this criterion, all the sites considered in

this study are deemed polluted (Figure 5.10). Pollution levels in the dry season for all sites were

higher than that of the rainy season. The wetland sites in order of increasing levels of pollution are

FRNR farms, KNUST Police Station, Spetinpom, Dakwadwom, Golden Tulip Hotel, Bekwai

Roundabout, Kaase-Guinness, Family Chapel, Atonsu Saw Mill and the High School Junction.

Wetland being used as rubbish dump Cattle grazing in a wetland

House in a wetland Vegetable farm in a wetland

Photograph 1. Some uses of wetlands in the Kumasi Metropolis (Photograph taken by Author in 2011)

98

Tabl

e 5.

3. E

daph

ic c

hara

cter

istic

s of

wet

land

site

s sa

mpl

ed in

the

Kum

asi

Met

ropo

lis

Loc

atio

n of

w

etla

nd

Des

igna

ted

rese

arch

co

de

Des

crip

tion

of d

omin

ant s

oil f

orm

at 1

0 m

from

wat

er

chan

nel

Dep

th o

f in

sert

ion

of th

e co

rer

at 1

0 m

fr

om c

hann

el

Soil

text

ure

Org

anic

m

atte

r (%

) Sa

nd

(%)

Silt

(%)

Cla

y (%

)

Des

crip

tion

Ato

nsu

Saw

mill

A

Dee

p, b

row

n, m

ottle

d so

il un

derla

in b

y gl

eyed

cla

y. T

op

soil

is da

rk a

nd lo

ose.

Per

man

ently

wat

erlo

gged

28

19

40

41

C

lay

loam

19

.85

Hig

h sc

hool

ju

nctio

n B

Sh

allo

w s

oil u

nder

lain

by

hard

pan

(fer

ricre

te).

The

uppe

r lo

ose

soil

laye

r is

dark

bro

wn

and

low

er h

ard

laye

r is

brow

n w

ith r

ed b

ands

and

gle

yed

patc

hes.

Wat

erlo

gged

fo

r abo

ut 8

mon

ths

10

62

12

26

Sand

y cl

ay

loam

3.

84

FRN

R fa

rms

C

Shal

low

soi

l und

erla

in b

y ha

rdpa

n (f

erric

rete

). So

il is

mad

e up

of d

ark

and

red

porti

ons

with

gle

yed

patc

hes.

W

ater

logg

ed fo

r abo

ut 6

mon

ths

16

38

52

10

Loam

6.

09

KN

UST

pol

ice

stat

ion

D

Dee

p so

ft la

yer o

f org

anic

mat

ter,

loos

e da

rk b

row

n, li

ght

wei

ght a

nd p

erm

anen

tly w

ater

logg

ed w

ith d

eepe

r por

tions

co

ntai

ning

gle

yed

clay

.

> 50

21

46

33

C

lay

loam

34

.50

Fam

ily C

hape

l E

Dee

p, g

rey,

mot

tled

soils

und

erla

in b

y gl

eyed

cla

y,

subj

ect t

o pe

rman

ent w

ater

logg

ed c

ondi

tions

. U

pper

laye

r is

loos

e an

d da

rk b

row

n.

22

46

38

16

Loam

5.

65

Spet

inpo

m

F Sh

allo

w g

rey

soils

und

erla

in b

y ha

rdpa

n fe

rric

rete

. M

ottli

ng i

s co

mm

on in

upp

er p

arts

exp

osed

to a

ir by

root

s of

pla

nts.

Wat

erlo

gged

for

abo

ut 6

mon

ths

10

42

30

28

Cla

y lo

am

14.0

3

Gol

den

Tulip

H

otel

G

R

ed w

ell d

rain

ed s

hallo

w s

oils

unde

rlain

by

hard

pan

ferr

icre

te w

ith c

lear

evi

denc

e of

per

iodi

c w

etne

ss a

nd

late

ral f

low

of w

ater

in th

e so

il pr

ofile

. Wat

erlo

gged

for

abou

t 6 m

onth

s

15

50

32

18

Loam

6.

23

Kaa

se G

uinn

ess

H

Dee

p, g

rey,

mot

tled

soils

und

erla

in b

y gl

eyed

cla

y,

subj

ect t

o pe

rman

ent w

ater

logg

ed c

ondi

tions

. W

ater

logg

ed fo

r abo

ut 7

mon

ths

17

48

30

22

Loam

7.

87

Bek

wai

ro

unda

bout

J

Shal

low

loos

e so

ils u

nder

lain

by

hard

pan

ferr

icre

te w

ith

clea

r evi

denc

e of

freq

uent

wat

erlo

ggin

g an

d la

tera

l flo

w

of w

ater

in th

e so

il pr

ofile

. Wat

erlo

gged

for

abo

ut 8

m

onth

s

20

47

32

21

Loam

5.

24

Dak

wad

wom

K

D

eep,

gre

y, m

ottle

d so

ils u

nder

lain

by

gley

ed c

lay,

su

bjec

t to

perm

anen

t wat

erlo

gged

con

ditio

ns.

Perm

anen

tly w

ater

logg

ed

39

33

31

36

Cla

y lo

am

6.17

Tabl

e 5.

4. P

hysi

coch

emic

al c

hara

cter

istic

s of

wat

er r

unni

ng t

hrou

gh t

he w

etla

nds

in K

umas

i

99

Loc

atio

n of

w

etla

nd

Des

igna

ted

rese

arch

co

de

Rai

ny s

easo

n (J

une)

D

ry s

easo

n (J

anua

ry)

pH

Tem

p (°

C)

Salin

ity

(ppt

) D

O

(mg/

l) T

DS

(mg/

l) C

ondu

ctiv

ity

(µS/

cm)

pH

Tem

p (°

C)

Salin

ity

(ppt

) D

O

(mg/

l) T

DS

(mg/

l) C

ondu

ctiv

ity

(µS/

cm)

Ato

nsu

Saw

mill

A

5.4

28.8

4 0.

43

4.2

411

822

8.6

30.3

1 0.

33

0.45

34

0 67

9

Hig

h Sc

hool

Ju

nctio

n B

6.

3 26

.14

0.10

5.

2 10

3 20

5 10

.6

26.9

8 0.

11

2.05

11

5 23

0

FRN

R fa

rms

C

6.3

25.6

8 0.

11

5.6

120

239

7.05

24

.36

0.08

3.

94

86

172

KN

UST

Pol

ice

Stat

ion

D

5.8

26.3

0 0.

10

3.2

110

220

6.88

26

.87

0.15

0.

62

164

340

Fam

ily C

hape

l E

5.9

27.8

8 0.

24

3.3

252

505

7.06

27

.36

0.25

4.

54

257

512

Spet

inpo

m

F 5.

9 27

.88

0.24

3.

3 25

2 50

5 7.

3 26

.88

0.13

3.

6 14

4 28

8

Gol

den

Tulip

Hot

el

G

6.4

27.0

4 0.

46

4.0

211

430

7.06

28

.02

0.25

0.

82

258

516

Kaa

se-G

uinn

ess

H

6.2

26.5

1 0.

52

4.9

423

805

8.32

31

.77

0.52

0.

16

537

1074

Bek

wai

roun

dabo

ut

J 6.

1 27

.12

0.26

4.

5 32

0 63

0 7.

47

38.1

3 0.

29

1.48

29

7 39

4

Dak

wad

wom

K

6.

0 26

.13

0.29

4.

3 30

5 61

0 7.

67

27.1

4 0.

29

0.75

29

8 59

7

100

Figu

re 5

.4.

A c

lust

er d

iagr

am o

f th

e co

rrel

atio

n m

atrix

of

the

vario

us e

nviro

nmen

tal

varia

bles

mea

sure

d in

the

diff

eren

t su

burb

s of

K

umas

i

101

Figu

re 5

.5.

A p

rinci

pal

com

pone

nts

anal

ysis

plo

t of

the

mea

sure

d en

viro

nmen

tal p

aram

eter

s (T

he v

ecto

r le

ngth

ref

lect

sth

e im

port

ance

of

the

varia

ble’

s co

ntrib

utio

n to

the

two

prin

cipa

l co

mpo

nent

s in

rel

atio

n to

all

poss

ible

axe

s an

d if

ave

ctor

touc

hes

the

circ

le, t

hen

none

of t

hat

varia

ble’

s ot

her

coef

ficie

nts

in t

he E

igen

vec

tors

will

diff

er f

rom

0)

102

Figu

re 5

.6. H

eavy

met

als

enric

hmen

t in

the

stre

am s

edim

ents

of t

he d

iffer

ent

subu

rbs

of K

umas

i in

the

rain

y se

ason

103

Fi

gure

5.7

. Hea

vy m

etal

s en

richm

ent

in th

e st

ream

sed

imen

ts o

f the

diff

eren

t su

burb

s of

Kum

asi i

n th

e dr

y se

ason

104

Figu

re 5

.8. H

eavy

met

als

accu

mul

atio

n in

the

stre

am s

edim

ents

of t

he d

iffer

ent

subu

rbs

of K

umas

i in

the

rain

y se

ason

105

Figu

re 5

.9. H

eavy

met

als

accu

mul

atio

n in

the

stre

am s

edim

ents

of t

he d

iffer

ent

subu

rbs

of K

umas

i in

the

dry

seas

on

106

Figu

re 5

.10.

Sea

sona

l diff

eren

ces

in p

ollu

tion

by m

etal

s in

the

stre

am s

edim

ents

of t

he d

iffer

ent

subu

rbs

of K

umas

i

107

5.3.3 Wetland Vegetation of Kumasi

5.3.3.1 Vegetation abundance and diversity in wetlands of Kumasi There were no seasonal differences (dry season and rainy season) in vegetation, so the data

were pooled and analysed according to the wetland areas sampled. In all, 112 species were

identified (Table 5.5). All the identified species are native vegetation of the forest region of

Ghana except Limnocharis flava and Ceratophyllum demersum which are exotic and invasive.

To assess the species’ individual contributions to the total wetland vegetation, the importance

value index was used. Using this index, species which have high relative dominance,

frequencies and densities in the various wetland areas have high importance value (see

Photographs 2, 3 and 4). The top five species with the highest importance values for each

wetland are marked in bold in Table 5.5. Some of the important species were Coix lacryma-

jobi (found at KNUST Police Station and Spetinpom), Paspalum vaginatum (Atonsu

Sawmill), Thalia geniculata (Atonsu Sawmill, High School Junction, FRNR farms, Family

Chapel and Kaase Guinness), Thelypteris palustris (KNUST Police Station) and Typha

australis (Atonsu Sawmill and Spetinpom). Most of these important species were found in

almost all the wetlands sampled, albeit with relatively low importance values at some sites.

Some species like Achyranthes aspera, Ipomoea carnea, and I. aquatica were found only at

one site each but with very high importance values. Paspalum vaginatum, Brachiaria deflexa,

Chromolaena odorata, Sporobolus pyramidalis and the Cyperaceae were ubiquitous and

formed the dominant species in at least 2 or more wetlands as manifested by their relatively

high total importance values. Species like Thelypteris palustris, Coix lacryma-jobi, Typha

australis, Lemna minor and Thalia geniculata each contributed relatively very high to the

total importance values but were present in few wetland areas whilst Bracharia lata,

Chromolaena odorata, Paspalum vaginatum and the Cyperaceae had relatively lower

importance values but contributed to total importance in all the wetlands sampled.

Different number of species were identified in the different wetlands sampled (Figure 5.11).

The FRNR farms had the highest number of species (86) and Spetinpom wetland with the

least number of species was. Comparing the various wetlands, the Shannon diversity and

species richness were very much varied than the species evenness. Overall, the Dakwadwom,

Golden Tulip Hotel and Bekwai roundabout wetlands had highest species richness and

Shannon diversity while Spetinpom was assessed to be poor in these indices (Figure 5.12).

Dakwadwom had the highest species richness and Shannon diversity of 6.09 and 3.53

respectively.

108

The longitudinal variability in species distribution from the water channel was also

investigated for each wetland area. In general, species richness and Shannon diversity

increased with increasing distance from the water channel (Figure 5.13). Sampling sites

between 30 and 40 m from the water channel were generally highest in species richness and

Shannon diversity. Except at Kaase Guinness, species evenness did not change much with

increasing distance from the water channel. This site has been channelised and there were no

plants growing in the water channel.

River channel at Family Chapel silted with vegetation and rubbish

Photograph 2. State of some of the wetlands in the Kumasi Metropolis (Photograph taken by Author in 2011)

109

Tabl

e 5.

5. S

peci

es i

dent

ified

in

the

diffe

rent

wet

land

s of

Kum

asi a

nd t

heir

impo

rtan

ce v

alue

s (T

op 5

impo

rtan

ce v

alue

spe

cies

for

eac

h si

te a

re s

hown

in b

old)

Iden

tifie

d sp

ecie

s

Sam

plin

g si

tes

Tota

l

impo

rta

nce

va

lue

Ato

nsu

Saw

mill

H

igh

Scho

ol

Junc

tion

FRN

R fa

rms

KN

UST

Pol

ice

Stat

ion

Fam

ily

Chap

el

Spet

in-

pom

Go

lden

Tul

ip

Hot

el

Kaa

se

Guin

ness

Bekw

ai

roun

d-

abou

t

Dak

wa-

dwom

Ach

yran

thes

asp

era

16.3

9.

9 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 26

.2

Aerv

a la

nata

0.

0 5.

9 5.

4 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 11

.3

Aesc

hyno

men

e cr

assi

caul

is

0.0

0.0

2.0

2.2

0.0

0.0

0.0

0.0

2.1

3.5

9.8

Aden

ia lo

bata

0.

0 2.

9 4.

2 0.

0 4.

8 0.

0 2.

2 2.

8 0.

0 2.

9 19

.7

Ager

atum

con

yzoi

des

0.0

1.8

3.7

1.4

0.0

0.0

2.6

0.0

4.0

2.7

16.2

Alch

orne

a co

rdifo

lia

0.0

5.6

5.2

0.0

0.0

0.0

0.0

0.0

2.0

0.0

12.7

A

lcho

rnea

laxi

flora

0.

0 4.

3 6.

5 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 10

.8

Alte

rnan

ther

a se

ssili

s 0.

0 1.

0 0.

0 0.

0 5.

5 6.

8 0.

0 0.

0 0.

0 1.

3 14

.6

Amar

anth

us h

ybri

dus

2.3

2.1

2.0

1.3

5.8

1.6

4.7

0.0

3.0

1.4

24.2

A

mar

anth

us

spin

osus

1.

9 7.

6 2.

0 2.

5 8.

9 5.

8 4.

3 0.

0 3.

1 1.

0 37

.2

Andr

opog

on g

ayan

us

2.7

3.2

2.0

2.3

3.0

6.2

3.2

3.0

1.3

2.5

29.4

Aspi

lia a

frica

na

0.0

0.0

2.8

1.2

0.0

0.0

4.6

0.0

4.8

5.8

19.2

Asys

tasi

a gi

gant

ica

4.3

4.6

5.5

1.8

0.0

0.0

2.7

0.0

5.0

2.4

26.3

Ax

onop

us

com

pres

sus

3.9

3.5

0.0

0.0

6.0

0.0

0.0

0.0

0.0

0.0

13.3

Bam

busa

vul

gari

s 0.

0 0.

0 3.

8 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 3.

8

Bide

ns p

ilosa

0.

0 1.

7 4.

4 1.

8 4.

4 0.

0 2.

8 3.

7 5.

6 3.

2 27

.6

Boer

havi

a di

ffusa

2.

5 3.

7 1.

9 0.

0 4.

6 0.

0 3.

7 3.

8 4.

8 3.

1 28

.0

Bra

chia

ria

lata

5.

4 5.

0 6.

6 7.

4 5.

4 5.

5 3.

5 3.

7 2.

7 9.

3 54

.5

Bra

chia

ria

defle

xa

18.2

0.

0 5.

2 7.

4 0.

0 0.

0 5.

9 6.

4 6.

5 11

.8

61.4

C

alop

ogon

ium

m

ucun

oide

s 6.

0 4.

5 3.

6 1.

2 4.

5 0.

0 6.

5 7.

0 5.

5 9.

2 48

.0

Cas

sia

mim

osoi

des

3.3

1.6

3.8

2.0

0.0

0.0

3.2

4.6

2.0

3.0

23.6

Cas

sia

occi

dent

alis

6.

5 5.

0 4.

2 0.

0 0.

0 0.

0 3.

0 1.

6 0.

0 0.

0 20

.3

110

Cas

sia

siam

ea

5.7

3.9

0.0

0.0

4.6

0.0

6.2

0.0

2.8

0.0

23.2

C

entr

osem

a pu

besc

ens

3.9

3.8

2.7

8.2

0.0

9.0

0.0

5.9

5.0

4.6

43.2

Cer

atop

hyllu

m

dem

ersu

m

1.0

0.0

0.0

3.8

0.0

0.0

0.0

0.0

0.0

0.0

4.9

Chr

omol

aena

od

orat

a 0.

0 6.

9 7.

1 0.

0 6.

8 9.

0 12

.2

10.9

13

.7

9.0

75.5

Cle

ome

rutid

ospe

rma

3.3

2.0

2.3

2.5

0.0

0.0

4.0

0.0

4.4

2.8

21.3

C

oix

lacr

yma-

jobi

0.

0 0.

0 0.

0 13

.6

15.3

24

.1

0.0

0.0

0.0

0.0

53.0

Col

ocas

ia es

cule

nta

6.

1 2.

8 0.

0 0.

0 4.

0 5.

1 1.

9 0.

0 2.

4 1.

9 24

.2

Com

bret

um h

ispi

dum

0.

0 0.

0 3.

8 0.

0 0.

0 1.

4 0.

0 0.

0 0.

0 0.

0 5.

3 C

omm

elin

a be

ngha

lens

is

0.0

4.6

3.9

4.0

3.7

0.0

4.0

5.6

6.7

5.1

37.6

Com

mel

ina

erec

ta

0.0

0.0

0.0

0.0

0.0

0.0

2.7

0.0

2.3

13.5

18

.5

Com

mel

ina

sene

gale

nsis

0.

0 0.

0 5.

3 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 5.

3

Cuc

umis

sativ

us

0.0

2.8

2.3

0.0

2.2

0.0

2.0

1.2

0.9

0.9

12.3

Cym

bopo

gon

citr

atus

0.

0 0.

0 2.

2 1.

0 0.

0 0.

0 1.

7 0.

0 2.

1 1.

7 8.

7

Cyn

odon

dac

tylo

n 11

.6

2.2

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

13.8

Cyp

erus

alte

rnifo

lius

3.1

2.0

2.7

6.3

0.0

3.6

4.3

4.2

5.6

4.1

36.0

C

yper

us c

yper

oide

s 2.

1 2.

9 3.

0 0.

0 0.

0 0.

0 6.

6 5.

9 6.

8 4.

6 31

.7

Cyp

erus

esc

ulen

tus

3.2

3.1

5.4

7.9

8.2

3.0

7.8

6.0

4.6

5.5

54.7

C

yper

us

long

ibra

ctea

tus

2.1

1.6

1.7

4.1

4.9

5.0

7.8

6.8

4.2

1.8

40.0

Cyp

erus

rotu

ndus

3.

9 2.

8 4.

1 4.

5 4.

4 0.

0 5.

6 5.

3 2.

0 5.

8 38

.2

Cyp

erus

stri

gosu

s 5.

0 0.

0 1.

7 4.

7 0.

0 0.

0 4.

6 4.

8 3.

4 5.

5 29

.8

Des

mod

ium

ad

scen

dens

0.

0 8.

0 3.

1 0.

0 0.

0 0.

0 3.

1 1.

8 4.

0 2.

2 22

.2

Echi

noch

loa

colo

num

2.5

1.2

0 0

0 0

0 0

1.8

0 5.

5 Ec

hino

chlo

a ob

tusi

flora

5.

9 3.

1 2.

2 9.

2 0.

0 0.

0 5.

4 3.

2 4.

5 0.

0 33

.6

Eclip

ta a

lba

0.0

0.0

2.7

8.4

0.0

0.0

3.0

2.0

2.4

2.1

20.6

E

leus

ine

indi

ca

6.3

5.7

2.4

4.7

10.5

0.

0 4.

1 7.

1 3.

6 3.

1 47

.5

111

Euph

orbi

a he

tero

phyl

la

0.0

0.0

2.7

1.3

5.1

8.8

4.5

4.7

5.5

5.7

38.1

Fleu

rya

aest

uans

3.

4 2.

7 3.

1 1.

4 4.

4 0.

0 2.

3 4.

3 3.

9 3.

1 28

.7

Glin

us o

ppos

itifo

lius

0.0

0.0

2.0

3.3

0.0

0.0

0.0

0.0

0.0

0.0

5.3

Gom

phre

na

celo

sioi

des

3.4

3.6

1.0

1.1

0.0

0.0

3.7

0.0

2.1

0.0

15.0

Hel

iotr

opiu

m

indi

cum

6.

1 9.

3 0.

8 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 16

.2

Hyd

role

a gl

abra

0.

0 0.

0 0.

0 7.

7 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 7.

7

Hyp

arrh

enia

rufa

3.

6 3.

3 2.

4 2.

5 2.

7 4.

1 3.

1 4.

2 3.

1 2.

3 31

.2

Impe

rata

cyl

indr

ica

3.1

3.7

3.4

7.0

6.0

3.7

2.2

2.8

2.2

6.2

40.4

Ip

omoe

a aq

uatic

a 0.

0 0.

0 0.

0 14

.0

0.0

0.0

0.0

0.0

0.0

0.0

14.0

Ip

omoe

a as

arifo

lia

5.4

4.6

0.0

0.0

8.0

9.8

0.0

7.0

0.0

2.4

37.2

Ipom

oea

cair

ica

0.0

1.0

2.9

0.0

0.0

0.0

5.1

3.8

4.3

5.4

22.4

Ip

omoe

a ca

rnea

0.

0 13

.7

2.6

0.0

0.0

0.0

0.0

0.0

0.0

0.0

16.2

Ipom

oea

invo

lucr

ata

0.0

2.3

2.7

5.3

5.8

6.7

7.0

4.6

4.0

3.6

41.7

Ip

omoe

a m

auri

tiana

0.

0 3.

6 0.

9 1.

2 2.

3 0.

0 2.

3 2.

4 7.

3 4.

0 24

.0

Ipom

oea

subl

obat

a 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 4.

9 5.

1 5.

3 3.

3 18

.7

Just

icia

flav

a 0.

0 0.

0 3.

3 1.

4 3.

9 0.

0 3.

3 2.

5 3.

0 3.

1 20

.5

Lem

na m

inor

16

.9

0.0

0.0

7.6

4.4

0.0

0.0

0.0

0.0

0.0

28.9

Le

ptoc

hloa

ca

erul

esce

ns

3.5

2.8

3.7

3.5

0.0

0.0

4.4

5.4

3.5

5.0

31.8

Lim

noch

aris

flav

a 1.

7 3.

0 1.

8 6.

3 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 12

.7

Ludw

igia

aby

ssin

ica

3.0

1.9

2.9

3.2

4.6

6.3

4.7

5.0

3.1

3.5

38.0

Ludw

igia

dec

urre

ns

0.0

0.0

0.0

15.9

0.

0 6.

0 0.

0 0.

0 0.

0 5.

3 27

.2

Luffa

cyl

indr

ica

2.3

0.0

1.9

0.0

0.0

3.7

2.5

4.9

3.9

2.4

21.7

M

alva

stru

m

coro

man

delia

num

0.

0 3.

5 4.

3 2.

9 0.

0 0.

0 5.

0 4.

4 3.

7 3.

6 27

.4

Mel

anth

era

scan

dens

0.

0 0.

0 3.

1 5.

1 0.

0 5.

5 4.

7 3.

9 5.

2 2.

8 30

.3

Mill

ettia

zenk

eri

1.2

1.4

3.1

2.9

2.7

5.0

5.3

2.9

5.3

2.0

31.7

M

imos

a pi

gra

0.0

0.0

6.9

1.0

0.0

0.0

0.0

0.0

0.0

5.0

12.8

Mom

ordi

ca c

hara

ntia

3.

1 3.

1 2.

9 0.

7 2.

3 4.

4 4.

5 6.

8 2.

4 3.

6 33

.7

112

Nep

tuni

a ol

erac

ea

1.7

2.3

3.1

0.0

3.3

3.3

2.7

3.1

2.2

0.0

21.6

Nym

phae

a lo

tus

0.0

0.0

0.0

3.4

0.0

0.0

0.0

0.0

0.0

0.0

3.4

Oci

mum

bas

ilicu

m

2.8

4.2

2.3

0.0

2.3

5.5

2.4

3.2

1.9

2.3

26.9

O

lden

land

ia

cory

mbo

sa

0.0

0.0

3.6

4.7

0.0

0.0

0.0

1.5

1.8

3.2

15.0

Oxa

lis c

orni

cula

ta

0.0

0.0

2.7

0.0

2.8

2.5

3.8

4.9

2.9

3.4

23.1

Palis

ota

hirs

uta

0.0

0.0

2.9

0.0

0.0

4.0

0.0

3.3

2.7

0.0

12.8

Pa

nicu

m

dich

otom

iflor

um

0.0

0.0

5.8

2.6

8.6

7.9

4.1

6.2

5.7

6.2

47.1

Pani

cum

max

imum

3.

6 4.

8 3.

3 3.

7 8.

9 9.

2 5.

7 6.

1 2.

5 5.

4 53

.2

Pasp

alum

vag

inat

um

14.9

7.

9 6.

9 10

.3

14.9

14

.7

3.8

5.0

4.9

5.9

89.2

Pent

odon

pen

tand

rus

0.0

0.0

0.0

6.6

0.0

0.0

0.0

0.0

0.0

0.0

6.6

Phyl

lant

hus a

mar

us

3.9

6.0

3.4

5.2

4.9

0.0

4.8

5.4

3.7

2.7

39.9

Phys

alis

ang

ulat

a 2.

2 4.

0 3.

9 6.

4 4.

2 5.

1 5.

3 5.

0 4.

4 3.

7 44

.2

Pith

ecel

lobi

um d

ulce

0.

0 4.

6 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 4.

6 Po

lygo

num

la

nige

rum

4.

4 6.

7 0.

0 0.

0 0.

0 5.

4 0.

0 0.

0 0.

0 0.

0 16

.4

Poly

gonu

m

salic

ifoliu

m

0.0

0.0

0.0

4.6

0.0

0.0

0.0

0.0

0.0

0.0

4.6

Rhy

ncho

spor

a co

rym

bosa

6.

6 0.

0 3.

9 0.

0 0.

0 9.

0 7.

3 13

.2

8.4

6.7

55.1

Rici

nus c

omm

unis

4.

1 5.

0 2.

9 0.

0 6.

0 0.

0 0.

0 0.

0 0.

0 0.

0 18

.0

Rot

tboe

llia

coch

inch

inen

sis

0.0

0.0

9.6

0.0

4.5

8.2

11.5

6.

1 9.

4 9.

3 58

.5

Schr

anki

a le

ptoc

arpu

s 0.

0 0.

0 3.

2 0.

0 0.

0 0.

0 0.

0 0.

0 3.

4 6.

8 13

.4

Scir

pus c

uben

sis

1.2

1.8

3.2

0 0

0 0

0 0

1 7.

2 Se

tari

a ba

rbat

a 0.

0 0.

0 2.

6 0.

0 3.

7 5.

9 4.

4 2.

7 6.

8 2.

6 28

.7

Seta

ria

palli

de-fu

sca

5.2

0.0

2.3

0.0

3.8

4.3

2.7

2.7

2.2

2.0

25.4

Sida

acu

ta

2.0

2.9

2.2

3.0

2.8

3.5

3.6

3.6

2.3

1.0

26.8

Sida

gar

ckea

na

2.9

3.2

2.4

0.0

5.1

3.1

4.0

4.1

2.4

2.8

30.0

Sola

num

nig

rum

3.

3 3.

4 1.

4 0.

0 2.

8 0.

0 0.

0 1.

5 3.

4 0.

0 15

.8

Sola

num

torv

um

0.0

4.3

0.0

0.0

3.1

0.0

0.0

0.0

0.0

0.0

7.3

113

Sorg

hum

ar

undi

nace

um

4.6

0.0

3.1

0.0

5.7

5.9

5.3

5.9

6.4

2.5

39.4

Spig

elia

ant

helm

ia

0.0

2.2

1.8

0.0

1.4

3.0

3.4

1.7

2.2

1.9

17.6

Sp

orob

olus

py

ram

idal

is

15.0

4.

9 4.

7 8.

0 5.

3 9.

9 3.

8 3.

3 6.

3 6.

6 67

.7

Syne

drel

la n

odifl

ora

1.1

0.0

2.0

1.0

3.9

4.9

4.4

4.0

3.2

4.1

28.7

Talin

um tr

iang

ular

e 2.

3 1.

3 2.

1 1.

3 4.

0 4.

1 1.

3 4.

3 2.

2 2.

4 25

.3

Thal

ia g

enic

ulat

a 20

.4

16.3

20

.9

0.0

20.6

0.

0 0.

0 19

.1

0.0

0.0

97.2

Th

elyp

teri

s pal

ustr

is

0.0

0.0

2.5

27.4

0.

0 0.

0 0.

0 0.

0 0.

0 0.

0 29

.9

Tria

nthe

ma

port

ulac

astr

um

2.2

2.0

1.4

0.0

2.6

0.0

1.8

0.0

1.2

2.6

13.8

Typh

a au

stra

lis

0.0

13.0

0.

0 0.

0 0.

0 24

.0

0.0

0.0

0.0

0.0

37.0

Ure

na lo

bata

3.

7 4.

7 0.

0 0.

0 2.

2 5.

0 3.

6 2.

3 2.

6 1.

8 26

.0

Vern

onia

cine

rea

0.0

0.9

1.3

1.0

2.9

0.0

0.0

0.0

2.2

1.7

9.9

Vern

onia

colo

rata

0.

0 0.

0 0.

0 0.

0 0.

0 1.

8 0.

8 0.

0 1.

3 2.

7 6.

6 To

tal n

umbe

r of

sp

ecie

s 63

73

86

62

57

46

70

64

78

76

11

2

114

Thelypteris palustris Ipomoea carnea

Luffa cylindrica Achyranthes aspera

Thalia geniculata Coix lacryma-jobi

Photograph 3. Some high importance value wetland species in the Kumasi Metropolis (Photographs taken by Author in 2011)

115

Centrocema pubescens Typha australis

Urena lobata Cyperus alternifolius

Ipomoea aquatica Lemna minor

Photograph 4. More high importance value wetland species in the Kumasi Metropolis (Photographs taken by Author in 2011)

116

Figure 5.11. Number of species identified at the various wetland sites in Kumasi

Figure 5.12. Species richness, evenness and Shannon diversity at the study sitesin Kumasi

117

Figu

re 5

.13.

Spe

cies

div

ersi

ty in

dice

s wi

th r

espe

ct to

dis

tanc

e fr

om t

he w

ater

cha

nnel

at t

he d

iffer

ent

sam

plin

g si

tes

in K

umas

i

118

5.3.3.2 Determinants of wetland vegetation distribution in Kumasi An analysis of similarities (ANOSIM) was carried out to assess the differences or similarities

in species within and between wetlands. There were significant differences in species

composition among the wetland sites with regards to the presence or absence of surface water

at the sampling sites (Figure 5.14a). Also, species changed with increasing distance from the

water channel (Figure 5.14b). The sites could therefore be separated hydrologically and the

distribution of the species also varied according to distance from the water channel (Global

R= 0.24 and 0.37 respectively, at 1 %).

The species found from the water channel up to 10 m away were significantly different from

those found from 30 m and beyond. Typical species in these two categories were Ipomoea

aquatica, Thalia geniculata, Nymphaea lotus, Typha australis, Ipomoea asarifolia, I.

aquatica, I. carnea, Alchornea cordifolia, Thelypteris palustris, Coix lacryma-jobi, Ludwigia

decurrens, Centrocema pubescens, Limnocharis flava, Hydrolea glabra in the water channel

up to 10 m and Cynodon dactylon, Desmodium adscendens, Eleusine indica, Aspilia africana,

Vernonia cinerea, Hyparrhenia rufa, Synedrella nodiflora, Trianthema portulacastrum,

Setaria pallide-fusca, Malvastrum coromandelianum, Ipomoea sublobata, Euphorbia

heterophylla, Oxalis corniculata, Paspalum vaginatum, Eleusine indica, Sporobolus

pyramidalis, Brachiaria deflexa, B. lata, Cyperus longibracteatus, C. esculentus C.

alternifolius from 30 m and beyond. All the species found within the 10 m distance from the

water channel were also found in sampling sites that had water on the surface irrespective of

the distance. Species found about 20 m from the water channel were not significantly different

from those beyond this distance. Cyperus spp were present in various clusters in the different

sampling sites and independent of distance from the water channel or the presence or absence

of surface water.

Having found the sites to be significantly different from each other, a hierarchical cluster

analysis was used to group the sites at the different levels of similarity using the vegetation

data. The clustering was done according to the two environmental factors: presence or

absence of surface water and distance from the main water channel. The sites showed 5

distinct clusters at 28 % similarity (Figure 5.15a and b). Inspecting the two subfigures, the

sites showed a high tendency to cluster according to the presence or absence of surface water

(Figure 5.15a open triangles separate from closed triangles). The clusters were, however, not

distinct when the distance from the water channel was used as a factor for distinguishing the

119

Figure 5.14. Analysis of similarities of the community composition among wetland sites in Kumasi according to (a) presence or absence of surface water and (b) distance from the water channel

120

sampling sites (Figure 5.15b distances symbols show no consistent order). In an attempt to

find underlying reasons for the observed clustering, multidimensional scaling (MDS) was

used. The resultant MDS plot with a stress of 0.17 is shown in Figure 5.16. Apart from site

H2 which in both the hierarchical cluster and MDS plot looked like an outlier, all the other

sites stick together similar to the clusters above according to the hydrological conditions of

presence or absence of surface water. The sites were also further distinguished by their

measured environmental parameters using the MDS (Figure 5.17) and PCA (Figure 5.18). The

MDS shows that the wetlands at Family Chapel, Golden Tulip Hotel, Bekwai roundabout and

Dakwadwom are quiet similar to each other than to the others. From the PCA plots, wetlands

at Family Chapel, Golden Tulip Hotel, Bekwai roundabout and Dakwadwom were different

from the other the wetland sites in the measured environmental parameters. A total variability

of 68.1% is attributed to the PC1 and PC2 (45.7 and 22.4 % respectively). Vectors with the

strongest contributions to the first two axes mentioned above are the percentage organic

carbon, sand, soil consistency, pH, temperature, TDS and salinity.

The assumption from the objective is that the suite of measured environmental parameters are

responsible for structuring the spatial distribution of vegetation or the vice versa such that, an

ordination based on the data would always group sites the same way. In order to match biotic

to environmental patterns, the vectors with the strongest contributions to the first two axes in

the PCA are considered in the BIO-ENV procedure. The combination of variables that best

explained the distribution of the vegetation is percentage organic carbon, sand, temperature

and salinity (in decreasing order of magnitude). Looking at the measures of these important

variables in the various sites, the wetland at the KNUST Police Station has the highest amount

of organic carbon, the lowest salinity, and comparatively low values for percentage sand and

temperature. A further hierarchical step by step analysis of these best results identified

percentage of organic carbon in the soil to be the strongest abiotic factor influencing

vegetation distribution in the studied wetlands.

121

Figure 5.15. A hierarchical cluster of the community composition among wetland sites in Kumasi according to (a) presence or absence of surface water and (b) distance from the water channel (all samples standardised by totals and square root transformed)

122

Figure 5.16. A non-metric multidimensional scaling plot of the community composition among wetland sites in Kumasi using the presence or absence of surface water as a factor (All samples standardised by totals and square root transformed)

Figure 5.17. A non-metric multidimensional scaling plot of the wetland sites in Kumasi using the measured environmental parameters

123

5.4 DISCUSSION A review of literature on soils of Kumasi shows that they are generally forest ochrosols and

lithosols (Obeng, 1971; Adjei-Gyapong & Asiamah, 2000). These soils may belong to the

nitisols and acrisols in the World Reference Base for Soil Resources (Driessen et al., 2001).

Although a detailed soil analysis was not undertaken to enable wetland specific soil

classification, Adjei-Gyapong & Asiamah (2000) states that soils bordering most of the rivers

and network of streams in Ghana are either recent alluvial deposits or presently influenced by

Figure 5.18. A principal components analysis plot of the sampled wetland sites in Kumasi superimposed with vector plots (circle and lines) of the measured environmental parameters. The vector length reflects the importance of the variable’s contribution to the two principal components in relation to all possible axes and if a vector touches the circle, then none of that variable’s other coefficients in the Eigenvectors will differ from 0

124

the floods of these drainage channels and therefore classified as alluvial soils. These alluvial

soils may accumulate autochthonous materials from dying vegetation during the dry seasons

when there is reduced flow and erosive power leading to deposition of organic material at

some wetlands. According to the Driessen et al. (2001), histosols have soil material with more

than 20 % organic matter by weight. Following this definition, soils of Atonsu Sawmill and

KNUST Police Station may be considered histosols. The relatively higher percentage of

organic carbon in wetlands of Atonsu Sawmill and KNUST Police Station could be due to the

low levels pH, salinity/conductivity and dissolved oxygen which all favoured net

deposition/accumulation of organic material.

The pH and dissolved oxygen were also significantly different between the seasons. These

conditions may have caused a decrease in biotic activity in the sites and decreased

decomposition of organic matter. An initial condition of a silty-clay soil might have facilitated

the anoxic conditions leading to the accumulation of organic material. The major

anthropogenic activity in these study sites that could have affected the soil conditions is

agriculture. However, the samples were taken from points that were not under cultivation, at

least not in the recent past. The depth of penetration of the corer was closely related to the soil

textural conditions. It is therefore difficult to isolate soil influence on plant species

distribution and hence accumulation of organic material. Root penetrability and anchorage are

dependent on soil consistency and very important for plant growth (Kim, 2006). The direct

relationship between the TDS and conductivity may be due to the type and amount of

inorganic and ionized substances in the water. The quantities of these substances may be such

that their contribution to conductivity and TDS for all the sites is similar. This direct

relationship also exists with salinity. The range of values for the salinity also shows that all

the study sites are freshwater environments.

By all indices used, Kumasi is said to be polluted by the heavy metals assessed in this study.

This pollution also varies with the seasons. The seasonal differences in sediment

contamination by heavy metals could also be related to the variations in dissolved oxygen and

pH. This is because, heavy metals concentration in sediments depends not only on their inputs

but also upon the ability of the environment to retain the metals. Some factors that determine

heavy metal retention are the textural characteristics, organic matter contents, mineralogical

composition and properties of the adsorbed compounds and depositional environment of the

sediments (Chakravarty & Patgiri, 2009). The seasonal variation in dissolved oxygen and pH

could therefore explain the positive correlation between Hg, Ni, Zn, Pb with pH and Cu and

125

Cr with dissolved oxygen and the respective strong seasonal differences in concentrations of

these metals. These physicochemical conditions might be necessary for the high retention or

adsorption of these metals in the sediments. The significantly positive correlation between

these metals also demonstrates possible co-contamination from similar sources: auto

mechanic shops (car spray and mechanic shops), car wash centres, pesticides and herbicides

from vegetable farms, wood preservatives from waste wood and saw dust from the carpentry

shops, waste from tie-dye industries and vehicle emissions.

If these metals presumably had the same sources, then differences in Igeo, EF and CF values

may be due to the difference in the magnitude of input for each metal in the sediment and/or

difference in the removal or retention rates of each metal from the sediment. This may be

particularly significant due to soil differences and the rainy season when erosion washes away

precipitated and adsorbed metals. Also, the changes in physicochemical conditions could lead

to destabilisation of the adsorbed complexes leading to the measured lower concentrations in

the rainy seasons. In general, when dissolved metals from natural or anthropogenic sources

come in contact with saline water they quickly adsorb to particulate matter and are removed

from the water column to bottom sediments (Schropp & Windom, 1988). This could explain

the high metal concentrations in the sediments at Atonsu Sawmill which has relatively higher

salinity. The Kaase Guinness wetland has been modified to a storm drain and the high

salinity, even though might have increased adsorption, did not reflect in high metal

concentration in the sediment because the concrete bottom of the water channel.

Conversely, no correlations were noted between Cd, P and As with the other heavy metals

suggesting that contamination of these substances in Kumasi might be from different sources

or had a different depositional nature. Cd concentrations were generally low in all sites. This

might be from natural sources such as weathering rather than anthropogenic (Sekabira et al.,

2010). The KNUST Police Station had conditions to potentially contain high concentrations

of heavy metals: fine-grained inorganic particles of clay and natural organic substances which

could have facilitated adsorption, chelation, complex formation or direct precipitation (Gibbs,

1977). Except for P, the low levels of other metals at this site could be due to the absence of

anthropogenic activities that would have introduced such substances. The phosphorus could

be from fertilizers washed from the surrounding agricultural activities and domestic

wastewater.

Trace amounts of heavy metals are always present in fresh waters from terrigenous sources

through weathering of rocks generally at relatively low concentrations. In recent times, due to

126

the combined effects of industrialisation/transport and urbanisation, heavy metal pollution of

aquatic ecosystems is becoming a potential global problem. Ghana is fast developing along

these lines, but like other developing countries, lacks mechanisms and tools to detect and

monitor water quality. Whilst none of these water sources is being directly used for human

consumption, the measured concentrations are so high that it could easily reach humans

through the food chain. For example, these wetlands are used as grazing grounds for livestock

and it will be very necessary to check the concentrations of these metals in the surrounding

vegetation and animals grazing in such wetlands. The Ghana EPA has effluent quality

guidelines for discharges into natural waterbodies. The car wash and repair garages do not

follow these guidelines. Even with the maximum permissible levels for release of effluents

given by the Ghana EPA, heavy metals have the tendency to accumulate in sediments and will

still pollute streams as detected in Kumasi.

Typical of tropical environments, the study sites had very diverse vegetation as expressed by

the Shannon diversity and species richness (Burslem et al., 2005). However, it is difficult to

attribute these levels as a standard for describing wetlands in Kumasi because of the various

forms of disturbances that may be specific to different areas. The inverted U-shaped pattern of

species richness and Shannon diversity might be due to fewer plants adapted to the extremes

and the 30 or 40 m zone providing a more ideal intermediate environment for most species.

The intermediate disturbance hypothesis may hold true for these study sites and thus explain

the pattern of species richness and Shannon diversity (Dial & Roughgarden, 1988).

Even with this inverted U-shaped pattern, species distribution followed specific longitudinal

environmental gradients like flooded conditions (presence or absence of water on the surface).

That is, some of the species are facultative whilst others obligate wetland species. The

presence or absence of water on the surface may have caused the within wetland differences

in species composition and distribution. Since the landscape is not completely flat, it is usual

to find small high ground quadrat plot areas within a wetland that are not flooded and hence

have different species cover. It is worthy to state that, even though no two quadrat plots can

be exactly the same, the edaphic differences may not be so sharp between the quadrat plots to

produce distinct differences in species along the distances that were considered for this study.

Even though this study did not investigate obligate and facultative wetland species, the

association of important value species to flooded sites defines the very unique niches that

characterise such species in the environment. The sites of obligate high importance value

species could be delineated to define the wetland microcosms for each study site.

127

The most popular definitions of wetlands have the key words hydrological conditions and

vegetation adapted to the flooded conditions (Keddy, 2000; Maltby & Barker, 2009). Even

though the cluster analysis shows that the presence or absence of surface water best grouped

species (vegetation distribution), the species found under the different hydrologic conditions

were different for the different wetland sites. What then causes these differences in species

among wetlands in Kumasi? That is, per the definition of wetlands, places with similar

hydrologic conditions in the same area should have similar vegetation. Therefore, other

environmental conditions might be contributing to the species distribution because, the much

used definitions of Keddy (2000) and Maltby & Barker (2009) are not adequate in describing

small scale heterogeneity in the species distribution in Kumasi. The findings of this research

confirm the difficulty to delineate wetlands from a landscape (Mitsch & Gosselink, 1993;

Keddy, 2000; Haslam, 2003; Maltby & Barker, 2009). A wetland is a transition zone between

aquatic and terrestrial environments. The limits and characteristics of this environment may

therefore not be defined by only the hydrological characteristics.

Apart from hydrology, several studies show that wetland geomorphology, substratum,

chemistry, adjacent terrestrial system, grazing, anthropogenic influences and individual plant

tolerance or response to the stresses of flood and drought are significant (Harper et al., 1995;

Barrat-Segretain et al., 1998; Owen, 1999; Casanova & Brock, 2000; Fortney et al., 2004;

Maltby & Barker, 2009; Miller & Fujii, 2010). By virtue of the percentage organic carbon and

other important physiochemical parameters identified by the BIO-ENV procedure, the

wetland at KNUST Police Station can be described as a peatland (Joosten & Clarke, 2002).

The peatland conditions might also have been the defining conditions for the growth and

establishment of Thelypteris palustris not only because of its dominance there but that was the

only site in which this species occurred. Other important species that also occurred in this site

were Coix lacryma-jobi, Hydrolea glabra, Ipomoea aquatica, Ludwigia decurrens,

Centrosema pubescens and Pentodon pentandrus. These species could also be said to be well

adapted to peatland conditions. These wetlands being in urban settings, urbanisation could

also have created perturbation gradients that affect aquatic plant species (Ot’ahel’ová et al.,

2007). The landuse change due to urbanisation could have resulted in alterations in the

hydrology, autochthonous and allochthonous inputs and water chemistry of wetland areas

leading to changes in the local plant communities (Owen, 1999).

128

6 CLIMATE, LANDUSE CHANGES AND FLOODING IN THE KUMASI METROPOLIS

6.1 INTRODUCTION Floods are part of the ecosystem and a natural hydrological feature of most river systems.

They provide benefits to humans through maintenance of ecosystem functioning such as

sediment and nutrient inputs to renew soil fertility in floodplains and floodwaters to fish

spawning and breeding sites (Avakyan & Polyushkin, 1989; Brinson, 1993; Casanova, 2000).

However, this important natural phenomenon causes disasters because human settlements are

getting increasingly closer to the floodplains or river channels such that communities are more

vulnerable and unprepared for these floods. Recent floods in most areas are therefore

preventable or anthropogenically induced and on a spatial and temporal increase (IPCC, 2007;

Millennium Ecosystem Assessment, 2005). On the average, 140 million people are affected

by floods each year, more than all other natural or technological disasters put together and in

addition to human lives, floods cost more than $244 billion damage from 1990 to 1999, the

most expensive of any single class of natural hazard (Millennium Ecosystem Assessment,

2005; IFRC, 2010).

Naturally, climate is dynamic. However, the world has experienced more anthropogenic

induced climate changes over the last 100 years. Observed changes include more frequent and

extreme weather events increasing the risk of flooding especially in urban areas (IPCC, 2007;

IFRC, 2010). The extent to which urban centres are vulnerable to changes in climate is

influenced by a variety of factors. However, this vulnerability could be mediated through

social and economic interventions and the ability of stakeholders and institutions to address

the challenges. Research shows that West Africa is among the most vulnerable regions to

climate change worldwide (Niasse et al., 2004; Millennium Ecosystem Assessment, 2005;

IPCC, 2007). Kumasi has experienced both landuse changes and floods in recent years. The

relationship between these floods and the landuse changes, climate variability or change has

not been investigated. This chapter therefore seeks to identify climate and landuse changes of

Kumasi and assess their influences on the current incidents of floods in some parts of the city.

The adaptations by the vulnerable suburbs to these annual floods are studied.

129

6.2 METHODOLOGY

6.2.1 Rainfall and Flood Data Collection and Analysis Monthly mean rainfall data on Kumasi from 1961 to 2010 were collected from the

Meteorological Services Department, Accra, Ghana. Basic descriptive statistics were used to

evaluate the rainfall data. The Mann–Kendall test (ZMK) was also used to determine the trends

in annual and monthly rainfall and trends in rainy days (Rodrigo et al., 2001; Modarres & da

Silva, 2007; Batisani & Yarnal, 2010). When testing either increasing or decreasing

monotonic trends at a p significance level, the null hypothesis is rejected for absolute values

of Z > Z1-p/2, obtained from standard cumulative distribution tables as 1.96. The standard

significance levels of p = 0.05 is applied. Positive values of ZMK indicate increasing trends

while negative ZMK indicate decreasing trends. The slope of the linear trend (as change per

year) was estimated with the nonparametric Sen’s method Q (Gilbert, 1987). The relationship

between total monthly and annual rainfall and number of rainy days (R≥1 mm/day) and heavy

rainfall events (R≥31 mm/day derived from an estimate of the 95th percentile of the rainfall

events of 1963 [chosen at random among first 3 years of Kumasi’s rainfall data]) was assessed

using correlation. This was to assess if Kumasi’s rainfall is explained more by the total

number of days with rain or by individual heavy rainfall events.

The data on flooding in Kumasi was obtained from the Ashanti Regional Office of the

National Disaster Management Organisation (NADMO). Extra data on the frequency and

duration of floods and extent of damage due to flooding was gathered through focus group

discussions, interviews and questionnaires administered to the affected communities in

Kumasi as described in Section 3.2.2. During the field visits and administration of

questionnaires and interviews, effort was made to identify and understand practices in dealing

with floods. This involved looking out for, noting and recording of behaviours, adaptations,

practices and activities related to flooding and life in slum or informal environments.

6.2.2 Assessment of Urbanisation of Kumasi Urbanisation was assessed by determining the landcover change of Kumasi. Landsat images

of Kumasi (at path 194, row 055) were collected from the landsat website www.landsat.org.

In order to minimise problems related to vegetation phenology, only images acquired during

the dry season were used. These images were captured on 11th January, 1986 and 13th

January, 2007.

130

Various image enhancement and classification techniques were performed to help with the

identification of the landcover types and changes using Leica Geosystems Erdas Imagine 9.2.

A hybrid of supervised and unsupervised classification techniques was performed on the

satellite images. The method was evaluated for accuracy using field verified landmarks with a

GPS. For each study site, the spatio-temporal changes in vegetation cover were estimated.

The final results of the classified image data and change thematic layer were further processed

into maps in ARC-GIS 9.3 environment.

6.3 RESULTS

6.3.1 Rainfall Patterns and Trends in Kumasi

6.3.1.1 Rainfall pattern in Kumasi The annual rainfall pattern of Kumasi is weakly bimodal (Figure 6.1). The first and major

rainy season starts from a relatively dry period in January with an average rainfall of 18.9 mm

and peaks in June with an average of 211.7 mm. The second starts from August with an

average rainfall of 87.7 mm and peaks at an average of 168.9 mm in September and drops to

an average of 23 mm in December. The average annual rainfall of Kumasi was 1372.5 mm

(Figure 6.2). Even though the highest average rainfall occurred in June, the highest monthly

total rainfall ever recorded was 534.5 mm in September, 2007 and the lowest and highest

annual rainfalls were 891.3 and 2345.1 mm in 1982 and 1968 respectively.

The descriptive statistics of the rainfall of Kumasi from 1961 to 2010 is presented in Table

6.1. Within these 50 years, there were 15 January months and 6 December months without

rainfall. All the other months always had rainfall. The months with the highest variability in

rainfall distribution were those between April and October. The rainfall distribution for these

months was also characterised by relatively high coefficients of variation and mostly positive

values for skewness and kurtosis. The rainfall distribution for August and September was

significantly peaked over the normal distribution (kurtosis of 6.98 and 5.50 respectively). All

the monthly rainfall distributions had positive skew values. That is, several extreme rainfall

values lie to the right of the mean. Skewness for August was the highest (2.06).

131

Figure 6.1. Average monthly rainfall pattern of Kumasi from 1961 to 2010

year

1960 1970 1980 1990 2000 2010

Ann

unal

rain

fall

(mm

)

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

Figure 6.2. Historical annual rainfall pattern of Kumasi from 1961 to 2010

0

50

100

150

200

250

JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC

Aver

age

rain

fall

(mm

)

month

1372.5

132

Table 6.1. Descriptive statistics of monthly rainfall pattern of Kumasi from 1961 to 2010

Variable Minimum Maximum Mean Std. Deviation Variance Skewness Kurtosis Mann–

Kendall Z Sen’s

slope Q January 0 111 18.92 25.85 668.05 1.52 1.97 0.00 0.000 February 1 135 55.75 40.03 1602.44 0.43 -0.92 -1.31 -0.546 March 36 310 119.12 53.61 2874.49 1.21 2.31 -2.44 -1.200 April 45 311 146.32 64.54 4165.61 0.80 0.03 0.90 0.631 May 46 307 174.79 63.93 4086.88 0.43 -0.56 -1.15 -0.715 June 41 377 211.73 83.00 6889.14 0.26 -0.66 -0.47 -0.490 July 18 446 146.62 98.14 9630.74 1.10 1.20 -0.74 -0.562 August 7 400 87.70 69.71 4859.20 2.06 6.98 0.49 0.304 September 15 535 168.89 87.10 7586.15 1.80 5.50 0.00 -0.004 October 43 336 156.12 63.55 4038.46 0.77 0.73 -0.50 -0.378 November 3 213 62.38 47.33 2240.15 1.20 1.24 -2.90 -1.145 December 0 116 26.40 26.60 707.48 1.37 1.81 -0.26 -0.025

6.3.1.2 Rainfall trends in Kumasi From the results of the nonparametric Mann-Kendall (Z) and Sen’s slope (Q) methods, it was

not possible to generalise either an increasing or a decreasing trend for the given set of data

(Figures 6.3 to 6.15.). Rainfall for January and September showed no changes (Z=0, Q=0).

Rainfall in April and August were increasing whilst that for February, May, June, July,

October and December was decreasing. In both instances above, the trends were not

significant. Only rainfall for March and November had decreased significantly over the period

(Z= -2.44, Q= -1.20 and Z=-2.90, Q= -1.15 respectively at α=0.05). That is, the average

rainfall for these two months was decreasing at 1.20 and 1.15 mm per year respectively. The

annual total rainfall was also decreasing although not significant (Z= -0.89, Q= -0.59).

133

-20

0

20

40

60

80

100

120

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Jan

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-100

-50

0

50

100

150

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Feb

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Z= 0.00 Q= 0.00

Z= -1.31 Q= -0.55

Figure 6.3. Trends in rainfall of Kumasi for the months of January from 1960 to 2010

Figure 6.4. Trends in rainfall of Kumasi for the months of February from 1960 to 2010

134

-100

-50

0

50

100

150

200

250

300

350

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Mar

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-150

-100

-50

0

50

100

150

200

250

300

350

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Apr

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Z= 0.90 Q= 0.63

Figure 6.5. Trends in rainfall of Kumasi for the months of March from 1960 to 2010

Figure 6.6. Trends in rainfall of Kumasi for the months of April from 1960 to 2010

Z= -2.44* Q= -1.20

135

-150

-100

-50

0

50

100

150

200

250

300

350

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

May

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-200

-100

0

100

200

300

400

500

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Jun

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

0

Z= -0.47 Q= -0.49

Figure 6.7. Trends in rainfall of Kumasi for the months of May from 1960 to 2010

Figure 6.8. Trends in rainfall of Kumasi for the months of June from 1960 to 2010

Z= -1.15 Q= -0.72

136

-200

-100

0

100

200

300

400

500

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Jul

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-100

0

100

200

300

400

500

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Aug

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Figure 6.9. Trends in rainfall of Kumasi for the months of July from 1960 to 2010

Figure 6.10. Trends in rainfall of Kumasi for the months of August from 1960 to 2010

Z= 0.49 Q= 0.30

Z= -0.74 Q= -056

137

-200

-100

0

100

200

300

400

500

600

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Sept

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-150

-100

-50

0

50

100

150

200

250

300

350

400

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Oct

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Z= -0.50 Q= -0.38

Figure 6.11. Trends in rainfall of Kumasi for the months of September from 1960 to 2010

Figure 6.12. Trends in rainfall of Kumasi for the months of October from 1960 to 2010

Z= 0.00 Q= 0.00

138

-100

-50

0

50

100

150

200

250

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Nov

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

-40

-20

0

20

40

60

80

100

120

140

1960 1970 1980 1990 2000 2010

tota

l rai

nfal

l (m

m),

Dec

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Z= -2.90** Q= -1.15

Figure 6.13. Trends in rainfall of Kumasi for the months of November from 1960 to 2010

Figure 6.14. Trends in rainfall of Kumasi for the months of December from 1960 to 2010

Z= -0.26 Q= -0.03

139

6.3.2 Floods in Kumasi

6.3.2.1 Scope and frequency of floods in Kumasi The first rainy peak of the year (May, June and July) has always been associated with floods

in some parts of Kumasi. For the purposes of this study, the definition of flood events by

NADMO as occasions when water from e.g. rainfall events covered temporarily land outside

the limits of the river banks, which is not normally covered, is adopted. Table 6.2 is an extract

of rainfall events that have been reported to have caused floods and documented by the

NADMO Office in Kumasi for June 2009. Except for the flood that occurred on 9th June,

2009, all others occurred during/after heavy rainfall events (R≥31 mm/day). None of the

antecedent rainfall events was heavy. June and July were the months with the most reported

incidence of floods. The flood experience of the residents from the suburbs that are being

studied is shown in Figure 6.16. All respondents in Aboabo and Ahensan had had their homes

flooded before. In the other suburbs, depending on the distance of a house to the water

channel and length of stay of the respondent in the suburb, he/she might have experienced

floods or not. In general, residents who have lived longer than 10 years in the various suburbs

claim the floods are a recent phenomenon.

The relationship between rainy days, heavy rainfall events and the total monthly rainfall was

assessed by correlation. The rainfall of Kumasi was observed to be influenced by both number

-1000

-500

0

500

1000

1500

2000

2500

1960 1970 1980 1990 2000 2010

tota

l ann

ual r

ainf

all (

mm

)

Year

Data

Sen's estimate

95 % conf. min

95 % conf. max

Residual

Z= -0.89 Q= -2.34

Figure 6.15. Trends in total annual rainfall of Kumasi from 1960 to 2010

140

of rainy days and heavy rainfall events (Figure 6.17). However, total monthly rainfall was

found to be explained more by heavy rainfall events than the number of rainy days in the

month (85 % and 72 % of variance in common between heavy rainfall events and rainy days

with total rainfall respectively). A trend analysis of heavy rainfall events is presented in

Figure 6.18. Annual heavy rainfall events are increasing in trend, although not significant (Z=

0.45, Q=0.00). The heavy rainfall trends for the two flood prone months of June and July are

varied. Whilst the heavy rainfall events are decreasing for June (Z= -0.78, Q= 0.00), that for

July is increasing (Z= 0.91, Q= 0.00).

Table 6.2. Rainfall events and reported floods in some suburbs of Kumasi in 2009

Figure 6.16. Respondents’ flood experience in the various suburbs of Kumasi

Flood conditions Antecedent conditions Flood date Amount of

rainfall (mm) Flood location Date of last

rainfall Amount of rainfall (mm)

09/06/2009 27.8 Aboabo, Anloga, Oforikrom, Ahensan,

Kwadaso Estates, Dakwadwom, Sisaso,

Atonsu

08/06/2009 3.1 11/06/2009 60.4 10/06/2009 0.1 15/06/2009 43.3 13/06/2009 19.1 26/06/2009 90.3 25/06/2009 17.3 04/07/2009 53.4 03/07/2009 12.8 09/07/2009 72.7 08/07/2009 27.6

141

Figure 6.17. Relationship between different rainfall events and total monthly rainfall for Kumasi from 1961 to 2010

R2=0.72

R2=0.85

142

-10

-5

0

5

10

15

20

25

1960 1970 1980 1990 2000 2010

tota

l hea

vy r

ainf

all e

vent

s in

the

year

Year

DataSen's estimate95 % conf. min95 % conf. maxResidual

-4

-2

0

2

4

6

8

10

1960 1970 1980 1990 2000 2010heav

y ra

infa

ll ev

ents

in Ju

ne

Year

DataSen's estimate95 % conf. min95 % conf. maxResidual

-3

-2

-1

0

1

2

3

4

5

6

7

1960 1970 1980 1990 2000 2010heav

y ra

infa

ll ev

ents

in Ju

ly

Year

DataSen's estimate95 % conf. min95 % conf. maxResidual

Figure 6.18. Trends in heavy rainfall events in Kumasi from 1961 to 2010

Z= 0.45 Q= 0.00

Z= -0.78 Q= 0.00

Z= -0.13 Q= 0.00

143

6.3.2.2 Adaptations to floods in Kumasi All the suburbs considered have experienced floods at one time or the other. The location of

respondents’ houses influenced their estimation of the number of days the flood waters take to

recede in the suburbs (Figure 6.19). The predominant number of days the people live under flood

conditions was more than 20 days in a year for all the suburbs. For some suburbs like Ahensan,

Atonsu and Dakwadwom, it took more than one month for the water to recede. The estimation

of the loss was usually astronomical and varied depending on the interest and capacity of the

person or institution making the estimate. These losses have however been increasing year

after year.

Several coping strategies have been identified in the various suburbs (Figure 6.20). Each

strategy is used in combination with the others. The most popular strategies are fetching or

pumping out of flood waters and the creation of channels to facilitate the flow of the water.

Respondents who do nothing in these flood situations (about 25% of the time) are usually the

very young and the aged. Respondents of Kwadaso Estates and Spetinpom have relatively

higher educational backgrounds and, almost 75% of the time, more likely to complain to city

authorities.

There are also individual and community adaptations to floods (Figure 6.21). The action a

resident took depended on previous experiences of flood and length of stay within the community. Most new rent paying tenants of these suburbs do not to know that the houses are

flood prone. The first time experience of flood is usually that of shock and annoyance at the

non-disclosure that the house is flood prone. Coping strategies of such persons to the flood are

usually egocentric e.g. victims attempt to retrieve their rent (rent is usually paid 2 to 5 years in

advance); move valuables and or relocate till the conditions improve; seal vents or build

embankments around apartment (Photographs 5,6,7 and 8), complete relocation of property; and

to complain to the city authorities. Most of these individual efforts often result in conflicts

between the person and the landlord or neighbours. Some of the conflict issues were

disagreements on the direction of a new drainage channel created by the individual;

perception of individual not participating in communal efforts; or perception of the individual

being proud. Residents who have lived in these suburbs for a while respond differently to the

floods: they are more accommodating of the landlords and employ more communal adaptations to

the floods e.g. participation in the construction of new drainage channels; (re)dredging of streams;

landlords build embankments around houses; construction of networks of raised walkways to

the various houses or compounds.

144

Figu

re 6

.19.

Rec

essi

on t

ime

of w

ater

at

the

vario

us fl

ood

pron

e su

burb

s of

Kum

asi

Figu

re 6

.20.

Ada

ptat

ions

of

resi

dent

s to

flo

ods

in t

he v

ario

us

subu

rbs

of K

umas

i

Suburb is flooded

individual

� Move residence to new location � Retrieve their rent � Relocate till water subsides � Complain to authorities � Fetch/pump water out of room or

house � Seal water ways and build

embankments around room or house

community

� Things will improve here � God will take care of things. It will

not always be bad � Construct bigger drainage channels

in the neighbourhood � Dredge river/stream channel � Build higher embankments around

house � Do nothing

� Moved out or started to behave like an old resident

� Moved out or started to behave like an old resident

� Construct drainage channels in the neighbourhood

� Dredge river/stream channel � Build embankments around house � Build houses on stilts � Improve on our houses � Do nothing

� Move things to higher grounds � Construct drainage channels in the

neighbourhood � Dredge river/stream channel � Build embankments e.g. with sand

bags or ridges � Fetch/pump water out of room or

house � raised walkways of block and wood � Do nothing

First year (Y)

Next year (Y+1)

Subsequent years (Y+1+2+…n)

New resident Old resident and homeowners

Figure 6.21. Flood response model of new and old residents of flood prone suburbs ofKumasi

conflict

146

In most of these flood prone suburbs, the chiefs and assemblymen harness local labour and

capacity into building walkways, footbridges, drains etc. The public institutions usually

provide some assistance in times of flood. NADMO may provide relief items depending on

the availability of funds and number of people who have been affected. This is also the time

that the KMA and NADMO mostly carry out their public education on floods. The major

flood management strategy of the KMA is the engineering approach. For example, a storm

drain is being constructed on the Aboabo River spanning through Anloga, Asokwa and

Ahensan to Atonsu. In other places, the KMA does desilting, dredging and removal of

structures within the waterways to improve flow of the water.

6.3.3 Landcover Change in Kumasi Landcover change was used as a proxy to assess urbanisation of Kumasi. To do this, it was

assumed that there are two possible landcover types: the land is either covered with vegetation

or built-up and bare areas. The landcover change from vegetated to built-up and bare areas is

shown in Figure 6.22. In 1986, the land area currently under the jurisdiction of the KMA was

about 80 % covered by vegetation. The remaining land area was either built up or bare e.g.

public spaces, construction sites, roads, newly cleared farmlands etc. Land areas adjacent

streams were largely vegetated (Schmidt, 2005; KMA, 2006). The major vegetated

continuums in 1986 were in the eastern and the southern corridors stretching from Asokore

Mampong to Deduako and the north western corridors from Ohwim to Tanoso. Typical built-

up or bare areas were Central Kumasi stretching towards Suame, Airport, Atonsu, Santasi and

Kwadaso. Built-up or bare areas beyond these suburbs were settlements located along the

major routes leading out of Kumasi. By 2007, the vegetated areas within the KMA had

reduced from 80 % to 50 %. The remaining major vegetation patches were at KNUST,

Dakwadwom, Nyankyereniase and the south eastern land area beyond Emena. Vegetation

cover along stream banks had also reduced. That is, the bare and built-up areas had been

growing at about 1.9 % per annum between 1986 and 2007.

147

Figure 6.22. Landcover changes within the jurisdiction of the Kumasi Metropolitan Assembly between 1986 and 2007

148

Photograph 5. House being built on stilts in a flood prone suburb in the Kumasi Metropolis (Photograph taken by Author in 2011)

Photograph 6. Raised walkways in the compound of a flood prone house in the Kumasi Metropolis (Photograph taken by Author in 2011)

149

Photograph 7. Embankment built around a flood prone house in the Kumasi Metropolis (Photograph taken by Author in 2011)

Photograph 8. Embankment built around the main door to an apartment in a flood prone suburb of the Kumasi Metropolis (Photograph taken by Author in 2011)

150

6.4 DISCUSSION The seasonal rainfall cycle of Ghana shows a progressive shift from an all-year-round or two

distinct rainy seasons from the coast to a single rainy season in the north. The results for

Kumasi located around latitude 6°N show two distinct rainy seasons with a lower second

peak. This is because the rainfall in Kumasi is dependent on convection following the

movements of the ITCZ as the dominant mechanism producing rain (Barbé et al., 2002). After

September, the ITCZ is receding and there is less penetration of moisture laden winds to

cause rainfall, hence the weaker second peak. The high variations in rainfall of the rainy

seasons may be due to inconsistencies in quantity and temporal distribution in the rainfall as

shown by the skewness and kurtosis of the rainfall data. Positive skew values mean that the

larger proportions of the rainfall for those months are a result of extreme rainfall events

(Barbé et al., 2002; Modarres & da Silva, 2007; Batisani & Yarnal, 2010). The statistical

results for July, August and September which are the peaks of the rainy season and floods in

Kumasi are therefore not surprising and therefore due to heavy rainfall events as proven by

the correlation analysis. Even though this analysis did not consider time between rainfall

events, the antecedent conditions, as in date of last rainfall event, could have affected the

probability of occurrence and duration of floods.

Apart from January and September, all months experienced a change in rainfall, albeit, most

of them not significant. March and November are the only months with significant decrease in

rainfall at a rate of 1.2 and 1.15 mm/annum respectively. According to Ghana’s Initial

National Communication to the IPCC in 2000, Ghana’s annual rainfall was found to have

decreased by 20% in the past 30 years (Ghana Environmental Protection Agency, 2001). Even

though total rainfall for Kumasi may be decreasing (albeit, not significant), the change is less

than the national figure reported by the EPA. However small it is, this change in can modify

some elements of the water cycle in Kumasi: the frequency and intensity of precipitation,

evapotranspiration, soil moisture, groundwater recharge and runoff (Fedeski & Gwilliam,

2007; Ramsar Convention Secretariat, 2008). Due to the relatively short period of the

available data (50 years), it is difficult to draw conclusions on trends which may only end up

being natural variability in rainfall. Because of this shortfall in the data, these results may

therefore have to be validated by other climate models to draw stronger conclusions on

climate change and rainfall trends in Kumasi.

The rate of growth of bare and built up areas in Kumasi could be a proxy measure of the rate

of urbanisation. The built-up and bare areas within the KMA have been growing at a rate of

151

1.9 % per annum in the past two decades. The total population of Kumasi more than doubled

within this period. The associated anthropogenic removal of vegetation and soil compaction,

e.g. for housing and road constructions resulting from this growth, could decrease infiltration

of rainwater and increase overland flow to cause the recent floods. It is now known that the

most widespread serious potential impact of climate change on human settlements is flooding

(Dazé, 2007; Fedeski & Gwilliam, 2007; IPCC, 2007; Wilby, 2007; International Recovery

Platform, 2009; McClean, 2010). However, from the results obtained, flooding as being

experienced in Kumasi is may not be the result of climate change in June and July but wrong

urban spatial planning and management. The growth of Kumasi is not accompanied by

development of sustainable urban drainage systems and waste management systems with

enough system capacity or sophistication to avoid being overwhelmed by the increased

overland flow. These are aggravated by multiple stresses and low adaptive capacities arising

from endemic poverty and weak institutions (United Nations Inter-Agency Secretariat of the

International Strategy for Disaster Reduction, 2005; Satterthwaite, 2003, 2008).

Even though the floods occurred during or after heavy rainfall events, the relatively long time

it takes for the flood waters to recede may indicate that not all the floods are flash floods or

due to heavy rainfall events. These suburbs are sited on lands, hitherto, considered as

wetlands. This is evidenced by the prevalence and dominance of vegetation adapted to

growing in wetlands (see section 5.3.3). The experiences of these flood prone neighbourhoods

are, therefore, consequences of the spread of physical urban developments into wetlands.

Surface compaction due to urbanisation may increase overland flow and produce flash floods

as described by Griggs & Paris (1982) and the United Nations Inter-Agency Secretariat of the

International Strategy for Disaster Reduction (2005). However, the rising of the water table

during the rainy season (peaking in June) creates a network of effluent streams in large parts

of these suburbs. The heavy rainfall events recorded each year trigger the flow and duration of

these effluent streams. Effluent streams, rather than flash floods, best explain the delay of the

flood waters in drying out from the neighbourhoods. Also, the dense networks of raised

walkways and drainage channels could only be an adaptation to effluent stream flows rather

than flash floods. From the above arguments, these events in those neighbourhoods should not

be considered as floods contrary to the KMA’s definition of floods. This is because, even

though the areas are inundated, the water is very much expected.

It is widely reported that a large and growing number of people are at high risk of adverse

ecosystem changes because of the worsening trend of human suffering and economic losses

152

from floods (ActionAid International, 2006; Fedeski & Gwilliam, 2007; IPCC, 2007; Nelson

et al., 2009). The ecosystem change as a result of the unmanaged urbanisation of Kumasi may

increase the losses to floods. The scale of the risk, extent of vulnerability and losses from

these flood events is much influenced and increased by the poor quality of housing and

infrastructure in Kumasi, the lack of planning and enforcement of wetland management plans

and the low level of preparedness among the city’s population and key emergency services.

The flood prone suburbs have relatively higher human population and densities (Ghana

Statistical Service, 2005) and higher demand for wetland services per person, therefore,

making the human and ecological impacts of flooding more serious.

Also, these flood prone suburbs are predominantly inhabited by the urban poor and new rural-

urban immigrants. Even though there is no spatial poverty profile of the residents of Kumasi,

the physical location and characteristics of the poor are undeniably obvious in the housing

landscape: an agglomeration of dilapidated structures without social infrastructure and

amenities. It is however surprising that estimated economic losses from the floods keep rising

year after year irrespective of the scale of the floods. Because the type and quality of houses

identified in these suburbs are among the worst in the city, these reported increases in losses

may be due to improvement in the economic conditions of these slum residents leading to

increase in acquisition of precious and capital goods and services which may easily be

destroyed by the floods. However, after living in such environments for several years, the

residents adapt very much and losses are supposed to be very minimal except for some off-

season large scale flash floods. Also, there is usually the perception and anticipation of help or

a craving for public sympathy by these victims and therefore the tendency to exaggerate their

losses. Either way it is difficult to independently verify these claims or accurately estimate the

value of properties destroyed by floods in Kumasi.

In recent years, the notion of adaptation has become a significant component of climate

change literature and international negotiations on climate (Niasse et al., 2004; World

Meteorological Organisation, 2006; Intergovernmental Panel on Climate Change, 2007;

Ireland, 2010). However, small scale adaptations by individuals and communities in informal

settlements have not been given attention. What is reported in this study are individual and

communal coping strategies rather than proactive and concerted efforts by city authorities at

managing floods. After several years of stay, residents might have developed strong economic

and social ties in the community and are therefore more reluctant to move out than the new

community entrants. The choice to remain therefore transcends the push effect of annual

153

floods. A key consideration in the adaptation debate should therefore be on the small scale

adaptation efforts and coping strategies of residents of mostly informal settlements who are

most hit by floods. In addition, most of the flood victims have problematic relationships with

the KMA because their houses have either been marked for demolition or they have no

building permits. Yet the KMA is to act with all stakeholders to reduce these risks. In the light

of the increasing urbanisation, the actions of the KMA must be to increase its efforts at not

only re-engineering the city but also dialogue and enforcement of spatial plans and building

regulations to reduce the proliferation of these informal settlements and preserve the

demarcated buffer zones of streams within its jurisdiction.

154

7 GENERAL DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS

7.1 GENERAL DISCUSSION The state of the urban environment in Ghana is undeniably unacceptable. On the one hand,

chiefs, landowners and local government agencies do not coordinate their efforts in spatial

planning, and on the other, urban residents are overwhelmed with their own waste. The result

is that waste is disposed of in any available “free” space, mostly wetlands. Urban wetland

communities therefore bear the full brunt of filth and floods. Cleanliness may be deemed a

virtue. However, this virtue needs to be reinforced and promoted through conscious efforts by

all stakeholders. Whilst residents of Kumasi may be consciously interested in maintaining a

clean environment, it behoves the responsible stakeholders to provide the much needed public

services to enable residents carry out this civic responsibility. Waste disposal in streams and

wetland areas could therefore be a thing of the past if the appropriate spatial planning and

waste management approaches are implemented.

The consequences of urbanisation in developed and developing countries have been very

much described. The KMA does not have to make all the mistakes that have been described or

experienced by other cities elsewhere. There is still a window of opportunity for the KMA to

avert this doom by harnessing the interest of the current King of Ashanti (Asantehene),

Otumfuo Osei Tutu II, in its environmental management agenda. The Asantehene is very

much interested in saving waterbodies. Because of the land tenure system and the authority

the Asantehene wields in Kumasi, efforts to reverse the dire wetland situation could be made

easy. The clearing of waterways through the demolition of houses has been vehemently

resisted in most areas. Political interferences into these activities have sometimes made the

situation more complicated. However, the involvement of the Asantehene in these activities

could provide the much needed impetus that the KMA needs in carrying out their corrective

environmental management and Garden City agenda.

The recent experiences of urban Kumasi make restoration to its former Garden City status not

only desirable, but necessary. More and more people are experiencing floods and the results

of this research show that, total annual rainfall of Kumasi is more from heavy rainfall events

than rainy days. If this trend continues, the city has two options to curb the incidence of flash

floods: (i) either adopt the engineering approach by increasing and expanding the drainage

systems or (ii) improve the conditions of the wetlands to play its natural role of absorbing and

155

slowing down the storm waters. Whilst none of these options will single-handedly solve the

problem, the economics of these options favour the latter. Urban storm drainage systems are

expensive to construct and maintain. Improving wetland areas will not only curb the incidence

of floods but is also an ecologically sound approach for a city that is growing and increasingly

producing more pollutants.

This research has shown that the wetlands of Kumasi are a rich store of plant biodiversity. It

is surprising that, a city that is undergoing such a rapid landcover change still has wetland

patches that have maintained plant diversity comparable to undisturbed environments. That is,

despite the pollution, encroachment and fragmentation of the wetlands, good fragments

worthy of protection still exists. Whilst most of the wetlands have relatively high diversity

indices, the importance value indices show that they are predominantly vegetated by 1-3

species. There seems to be a form of vegetation specialisation based on the soil and water

regime of each wetland. Conservation objectives for these wetlands should therefore be for

their ecological services rather than their store of biodiversity. For example carbon

sequestration should be the conservation objective for the wetlands at the KNUST Police

Station and Atonsu Sawmill; aesthetic values for Golden Tulip Hotel, Bekwai roundabout,

Dakwadwom, Atonsu Sawmill; and FRNR Farms; and flood mitigation objectives for those at

KNUST Police Station, Kaase Guinness. The diverse ecological potentials of these wetlands

should be harnessed for sustainable urban environment of Kumasi. The success of the new

Garden City of Kumasi will depend on how this rich diversity is integrated into the spatial

planning efforts. Whilst the presence of heat island effect has not yet been determined in

Kumasi, the increasing use of air conditioners, vehicles and the reduction in vegetation cover

make it plausible. When these urban green spaces and streams are improved, they could

mitigate the heat island effect and also provide places for relaxation to escape from the

pressures of the city.

Poverty and vulnerability have always been used together in literature on slums and informal

settlements. Even though these suburbs generally house the poor of Kumasi and are exposed

to the annual floods, they cannot necessarily be described as vulnerable. Most of these

residents have accepted the annual flooding as the status quo. Whilst anyone can be

vulnerable to non-seasonal flash flood events, the prevalent form of inundation in these

suburbs is due to effluent streams from the annual rise in the water table. This type of

inundation by its nature and frequency is very much expected. The inhabitants may therefore

be frequently exposed to floods but cannot be described as vulnerable because they expect it

156

and prepare for it. This may explain their reluctance to relocate. Poverty (expressed by their

inability to afford the relatively high rental charges at other suburbs) may be the determinant

in the initial choice to settle in that location. However, these residents soon tap into other

forms of human capital (social, experiential, cultural, spiritual capitals) to reduce their

vulnerability to the floods. These forms of capital need time to mature, improve and increase

in value. Any efforts by the city authorities to relocate these wetland inhabitants should

therefore involve the early identification and removal of factors that provide resilience to

floods. These may, however, be deeply rooted in the social fabric and, therefore, difficult to

remove by the time the inertia ridden city authorities decide to take action.

This study also shows that human environmental behaviour and choices are very fluid and

unpredictable. Different people have different tolerance levels which are related to their

individual value judgements. The Ghanaian society is still very much communal. Even in

urban settings, people easily build relations in their neighbourhood. These bonds are

Ghanaian social values that influence highly the outcomes and decision making processes of

these urban residents. The trajectory of the EKC would apply in very individualistic societies

where persons, or at most the household, take decisions independent of the influence of their

neighbours. Where this independence is lost, the trajectories of flood tolerance or

environmental pollution with increasing wealth deviates from what has been proven by Dinda

(2004), Stern (2004) and Mills & Waite (2009). What is observed is that, after staying in these

communities for a while, income gradually loses its power as a major decision making factor

in residential choices.

Kumasi is growing faster than the inhabitants and planning authorities can cope with. Most

inhabitants are still dependent on agriculture, natural resources and fresh foodstuff. With

increasing urbanisation, these are drawn from increasingly distant environments to satisfy

these inhabitants. The distance from which goods are acquired and to which waste can be sent

is directly related to the spatial developmental stage of the city. Distances and costs to dispose

of wastes are increasing for both the residents and waste management companies. The

distance is also a function of energy (or financial resources). The more the availability of

energy, the further these wastes could be transported or better processed into less harmful

forms. Because of the high costs of managing wastes, streams serve as sources of “free

energy” to transport the waste when it rains. Disposal of waste in wetlands and streams may

therefore be an adaptation by urban residents to reduce this energy function in their lives.

157

These are conscious decisions residents take considering the cost of floods and aesthetic

values of the environment being lost through such waste disposal methods.

Furthermore, the placing of waste containers in wetlands by the KMA could also be a

reinforcing factor to the wrongful use of these wetlands as dump sites. The KMA sees these

wetlands as the remaining islands of free public spaces for these purposes in an already

unplanned and overdeveloped neighbourhood. In most of these poor suburbs, there is a high

default rate of fee payment for collection or disposal of waste. Therefore, domestic waste is

secretly deposited in the immediate surroundings of these containers. These are all washed into

the surrounding streams choking them and increasing the siltation rates. These shallow rivers

therefore easily overflow their banks during the heavy rain events flooding the neighbouring

low lying areas. Therefore, whilst the KMA is working in one dimension to reduce the

incidence of floods by removing structures along the waterways, they should also to be blamed,

in part, for increasing the incidence of floods by siting waste bins in wetland areas.

7.2 CONCLUSIONS This study sought to contribute knowledge on the state of wetlands, incidence of flooding and

stakeholder needs for the management of wetlands in Kumasi, Ghana. Data was collected in

the chosen study sites and suburbs from January, 2010 to December 2011. After various

analyses, the following conclusions were drawn in answer to the research questions and

hypotheses that were set for the study:

� All the wetlands are riverine and associated with Rivers Suatem, Subin, Aboabo, Sisai

and Wiwi which flow through the jurisdiction of the KMA. The size and quality of these

wetlands are being decreased through adjacent anthropogenic activities.

� Whilst there are some species common to all the wetlands, each wetland is unique in

species composition. In each wetland, vegetation distribution followed a hydrological

gradient. That is, the degree of soil water saturation was found to be the main determinant

of species composition for each site (within wetland differences). However, no one

obligate wetland species could be used to characterise all wetlands in Kumasi.

� The percentage organic carbon was the strongest factor that explained the observed

spatial heterogeneity of wetland vegetation in Kumasi (between wetland differences).

Whilst there were seasonal differences in water conditions (pH and dissolved oxygen),

there was no seasonal change in the vegetation. The wetland vegetation species diversity

158

was not found to be influenced by adjacent landuse but by the hydrology and edaphic

conditions of the wetland.

� The hydrology and edaphic conditions at the KNUST Police Station make the wetland

have the highest amount of carbon sequestered among the wetlands investigated. Typical

species at the KNUST Police Station are Thelypteris palustris, Coix lacryma-jobi,

Hydrolea glabra, Ipomoea aquatica, Ludwigia decurrens, Centrosema pubescens and

Pentodon pentandrus.

� All the wetlands were found to be polluted by the heavy metals investigated. The

heavy metals Hg, Ni, Zn, and Pb probably come from the same source whilst Cu and Cr

from another. Major anthropogenic activities that could be the sources of these heavy metals

are the auto mechanic shops (car spray and repair shops), car wash centres, pesticides and

herbicides from vegetable farms, wood preservatives from waste wood and saw dust from

the carpentry shops, wastes from tie-dye industries and vehicle emissions.

� The total annual rainfall is decreasing but not significant. Only rainfall for March and

November is significantly decreasing over the period considered (Z= -2.44, Q= -1.20 and

Z=-2.90, Q= -1.15 respectively at α=0.05). Annual heavy rainfall events are also decreasing

in trend, although not significant. The analyses show that total monthly rainfall of Kumasi is

explained more by heavy rainfall events than by the number of rainy days in the month.

However, there is no significant change in trends of heavy rainfall events for the two

flood prone months (June and July). Floods in Kumasi are therefore not attributable to

changes in climate.

� There are two types of floods that affect the wetland communities studied: flash floods

and ‘effluent stream floods’. The flash floods arise from overland flow due to heavy

rainfall events. The effluent stream floods are due to a rise in the water table during the

peak rainy season of June-July causing a relatively longer surface inundation from

groundwater.

� The built-up and bare areas within the KMA have been growing at a rate of 1.9 % per

annum between 1986 and 2007. This landcover change leads to increase in surface runoff

because of reduced infiltration, decrease in concentration time and time to peak discharge

and consequently increase in peak discharge. This is the major cause of the increased

flash floods in Kumasi.

159

� There was no relationship between number of years of stay in an area and presence or

absence of floods (r = -0.135). The inhabitants are adapted to life in these flood prone

neighbourhoods. Adaptations include: dredging of streams, erection of embankments

around houses, move property to higher grounds, build a network of raised walkways and

building houses on stilts. With these adaptations, residents plan to move out only when

they are provided with an alternative house elsewhere. These are adaptations rather to the

long lasting seasonal effluent floods than to flash floods.

� The behaviour of the flood prone residents does not follow the EKC trajectory. Even

when incomes of these residents increase, they do not move out, not even when they are

forced. The social and experiential capitals developed by staying in these suburbs far

outweigh the financial or economic losses due to the floods. The opportunity cost of

moving out is therefore too high for these flood prone residents to bear. The duration of

stay in a flood prone environment rather than income, is the stronger determinant in the

decision making process of a resident to stay or relocate. Increased income will flatten the

curve but will not bring flood tolerance to 0.

� The stakeholders contacted for this study considered wetlands important and should be

protected. The stakeholders most important for the management of floods and wetlands are

chiefs and the KMA. The responsible authorities should manage wetlands and floods

through the enforcement of landuse plans, demolition of houses on waterways and setting

up public parks. The social, institutional and technical dimensions of flood and wetland

management will achieve better outcomes for all the stakeholders.

7.3 RECOMMENDATIONS Based on the findings of this research, the following actions are recommended:

� The KMA should use the presence of effluent streams and the identified typical

wetland important vegetation species for each wetland as defining factors to delineate the

wetland boundaries. This exercise should particularly target the relatively large stretches

of wetlands at the KNSUT Police Station, FRNR farms, Atonsu, Family Chapel,

Dakwadwom, Spetinpom and Golden Tulip Hotel. After delineating the wetland boundaries,

all structures within the wetlands should be demolished. These recovered wetlands should

be given special names and revegetated to recover their natural system functions. In

doing this, the KMA should liaise with the Asantehene (King of Ashanti) to solicit his

support to reduce opposition.

160

� The wetlands at the KNUST Police Station and Atonsu should be protected for their

peat storage value. These sites could be further assessed for their potential to earn carbon

credits for the city in the on-going climate change mitigation schemes. For example, the

opportunities in the Reducing Emissions from Deforestation and forest Degradation

(REDD+) programme being advocated by the United Nations could be tapped to facilitate

the achievement of the Garden City status. This will also pave the way for Kumasi to

meet the demands of the Millennium Cities Initiative and the Convention on Biological

Diversity’s Global Partnership on Cities and Biodiversity.

� Relevant academic institutions and faculty should develop and actively focus their

research on urban ecology. The research should include the use of vegetation for urban

waste water treatment, wetland vegetation for storm water management and carbon

sequestration. Site specific rainfall input should be studied to assess the wetland

hydrodynamics and its influence on temporal variation in vegetation. Also, scientific

input from this research and others should be the backbone for the management of the

delineated wetland areas to conserve biodiversity and provide ideal environments for

human relaxation.

� The KMA should endeavour not to site waste containers in wetlands to reduce the

direct introduction of waste into stream channels. Where it is necessary to do so, a wall

should be built around such a facility to contain any overflows. Also, waste collection in

these poor neighbourhoods should be free so that residents are encouraged to use the

designated dump sites. Anthropogenic activities around the wetlands should also be

regulated to reduce the direct anthropogenic destruction and pollution of these habitats. In

particular, car wash and repair shops should either be relocated or their waste be managed

so that it is not directly introduced into the wetlands.

� Considering the limited financial resources available to the KMA, an engineering

based urban drainage system is not feasible. The approach to managing floods in Kumasi

should be a combination of the reconstruction of the wetlands as described above and social

interventions. The social flood management elements in each suburb should be identified

and prioritised for action to increase resilience and reduce the scope and incidence of

floods. In this regard, the KMA and NADMO need to recognise and give the needed

support to chiefs and assemblymen of these neighbourhoods and relevant NGOs who are

better placed to achieve these objectives.

161

� The DEMC of the KMA should be strengthened to carry out their objectives and

mandate. The powers vested in this Committee and its apolitical nature will make their

actions more acceptable to the people. When operational, the DEMC should collaborate

with the chiefs and assemblymen of the flood prone suburbs in enforcing community

based flood and wetland management plans.

162

8 REFERENCES Abbaspour, M. & Gharagozlou, A. (2005). Urban planning using environmental modeling and GIS/RS: A Case Study from Tehran. Environmental Quality Management, 14(4): 73-85

ActionAid International (2006). Climate change, urban flooding and the rights of the urban poor in Africa- key findings from six African cities. ActionAid International, Hamlyn House, London N19 5PG, UK

Adam, M. G., Sarfo-Mensah, P. & Warburton, H. (1999). Urban blight and farmers plight: peri -urban agriculture around Kumasi, Ghana. Land, 22: 131-148

Adams, W. M., Aveling, R., Brockington, D., Dickson, B., Elliott, J., Hutton, J., Roe, D., Vira B. & Wolmer, W. (2004). Biodiversity conservation and the eradication of poverty. Science, 306: 1146–1149

Adelekan, I. O. (2010). Vulnerability of poor urban coastal communities to flooding in Lagos, Nigeria. Environment and Urbanization, 22: 433-450

Adjei-Gyapong, T. & Asiamah, R. D. (2000). The interim Ghana soil classification system and its relation with the World Reference Base for Soil Resources. In: FAO (2000). Quatorzième Réunion du Sous-Comité ouest et centre Africain de corrélation des sols pour la mise en valeur des terres. Abomey, Bénin, 9-13 Octobre, 2000

Afrane, Y. A., Klinkenberg, E., Drechsel, P., Owusu-Daaku, K., Garms, R. & Kruppa, T. (2004). Does irrigated urban agriculture influence the transmission of malaria in the city of Kumasi, Ghana? Acta Tropica, 89(2): 125-34

African Peer Review Mechanism Secretariat (2005). Country Review Report of the Republic of Ghana. Midrand, South Africa. http://www.nepad.org

Akobundu, I. O. & Agyakwa, C. W. (1998). Handbook of West African Weeds. 2nd Ed. International Institute of Tropical Agriculture, Oyo Road, P.M.B. 5320, Ibadan, Nigeria

Akwei, E. (2010). Flooding in Aboabo Basin and adaptation: local responses to a global phenomenon. BSc thesis submitted to the Kwame Nkrumah University of Science and Technology, Ghana

Altobelli, A., Bressan, E., Feoli, E., Ganis, P. & Martini, F. (2007). Improving knowledge of urban vegetation by applying GIS technology to existing databases. Applied Vegetation Science, 10: 203-210

Armar-Klemesu, M. & Maxwell, D. (1999). Accra: urban agriculture as an asset strategy, supplementing income and diets. In: Proceedings of International Workshop on Growing Cities Growing Food: Urban Agriculture on the Policy Agenda, La Habana, Cuba, October 1999

Avakyan, A. B. & Polyushkin, A. A. (1989). Flood control experience in the USA. Power Technology and Engineering, 23(1): 53-58

Baarta, I., Gschöpf, C., Blaschke, A. P., Preinera, S. & Heina, T. (2010). Prediction of potential macrophyte development in response to restoration measures in an urban riverine wetland. Aquatic Botany, 93(3): 153-162

163

Baharoglu, D. & Kessides, C. (2000). Urban Poverty. In: Poverty Reduction Strategy Papers - A Sourcebook. The World Bank

Baker, J. & Schuler, N. (2004). Analyzing Urban Poverty: A summary of methods and approaches. World Bank Policy Research Working Paper 3399

Barbé, L. Le, Lebel, T. & Tapsoba, D. (2002). Rainfall variability in West Africa during the Years 1950–90. Journal of Climate, 15: 187-202

Barbier, E. B. (1994). Valuing environmental functions: tropical wetlands. Land Economics, 70(2): 155-173

Barrat-Segretain, M. H. & Amoros, C. (1996). Recovery of riverine vegetation after experimental disturbance: a field test of the patch dynamics concept. Hydrobiologia, 321: 53-68

Barrat-Segretain, M., Bornette, G. & Hering-Vilas-Boas, A. (1998). Comparative abilities of vegetative regeneration among aquatic plants growing in disturbed habitats. Aquatic Botany, 60: 201–211

Batisani, N. & Yarnal, B. (2010). Rainfall variability and trends in semi-arid Botswana: Implications for climate change adaptation policy. Applied Geography, 30: 483–489

Beckerman, W. (1992). Economic growth and the environment: whose growth? Whose environment? World Development, 20: 481–496.

Berkowitz, A. R., Nilon, C. H. & Hollweg, K. S. (Eds) (2003). Understanding Urban Ecosystems: A New Frontier for Science and Education. Sringer-Verlang New York, Inc.

Beverwijk, A. (1967). Particle size analysis of soils by means of the hydrometer method. Sedimentary Geology, 1: 403-406

Bhaskar, N. R. & Singh, V. P. (1988). Planning flood control projects in urban areas. Water Resources Management, 2: 123- 140

Bhattacharya, P. (2002). Urbanisation in developing countries. Economic and Political Weekly, 37(41): 4219-4228

Biney, C. A. (1990). A review of some characteristics of freshwater and coastal ecosystems in Ghana. Hydrobiologia, 208: 45-53

Blake, B. & Kasanga, K. (1997). Kumasi Natural Resources Management Research Project Report. Natural Resources Institute. The University of Greenwich. UK and University of Science and Technology, Kumasi, Ghana

Botchie, G., Akabzaa, T., Gyasi, E. & Sarpong, D. B. (2007). Land tenure, land use and environment in Ghana. ISSER Technical Publication Series 72. Legon, Ghana 78pp

Botkin, D. B. & Beverage, C. E. (1997). Cities as environments. Urban Ecosystems, 1: 3–19

Brinson, M. M. (1993). A hydrogeomorphic classification for wetlands. Wetlands Research Program Technical Report WRP-DE-4. Army Engineer Waterways Experiment Station, Vicksburg, MS

164

Brody, S. D., Highfield, W. E., Ryu H. & Spanel-Weber, L. (2007). Examining the relationship between wetland alteration and watershed flooding in Texas and Florida. Natural Hazards, 40: 413–428

Brook, R. M. & Dávila, J. D. (eds.) (2000). The peri-urban interface: a tale of two cities. School of Agricultural and Forest Sciences, University of Wales and Development Planning Unit, University College London. 251 + vii pp.

Brueckner, J. K. & Fansler, D. A. (1983). The economics of urban sprawl: theory and evidence on the spatial sizes of cities. The Review of Economics and Statistics, 65(3): 479-482

Burslem, D. F. R. P., Pinard, M. A. & Hartley, S. E. (eds)(2005). Biotic Interactions in the Tropics: Their Role in the Maintenance of Species Diversity. Cambridge University Press, 580pp

Capello, R., & Faggian, A. (2002). An economic-ecological model of urban growth and urban externalities: empirical evidence from Italy. Ecological Economics, 40: 181–198

Carter, J. L. & Vincent, H. R. (2001). After site selection and before data analysis: sampling, sorting, and laboratory procedures used in stream benthic macro invertebrate monitoring programs by USA State Agencies. Journal of the North American Benthological Society, 20(4): 658-682

Casanova, M. T. & Brock, M. A. (2000). How do depth, duration and frequency of flooding influence the establishment of wetland plant communities? Plant Ecology, 147: 237–250

Chakravarty, M. & Patgiri, A.D. (2009). Metal pollution assessment in sediments of the Dikrong River, N.E. India. Journal of Human Ecology, 27(1): 63-67

Chaneton, E. J. & Facelli, J. M. (1991). Disturbance effects on plant community diversity: spatial scales and dominance hierarchies. Vegetatio, 93: 143-155

Choi, J., Engel, B. A., Muthukrishnan, S. & Harbor, J. (2003). GIS based long term hydrologic impact evaluation for watershed urbanization. Journal of the American Water Resources Association, 39(3): 623-635

Cofie, O. O., Kranjac-Berisavljevic, G. & Drechsel, P. (2005). The use of human waste for peri-urban agriculture in northern Ghana. Renewable Agriculture and Food Systems, 20(2): 73–80

Cohen, B. (2004). Urban Growth in Developing Countries: A review of current trends and a caution regarding existing forecasts. World Development, 32: 23–51

Cohen, B. (2006). Urbanization in developing countries: current trends, future projections, and key challenges for sustainability. Technology in Society, 28: 63–80

Cook, C. D. K. (1996). Aquatic Plant Book. 2nd, revised edition. Amsterdam: SPB Academic Publishing

Cooper, P. (2009). What can we learn from old wetlands? Lessons that have been learned and some that may have been forgotten over the past 20 years. Desalination, 24: 11-26

Cornish, G. A. & Kielen, N. C. (2003). Wastewater irrigation – hazard or lifeline? Empirical results from Nairobi, Kenya and Kumasi, Ghana. In: Scott C.A. Faruqui, N.I. and L. Raschid-

165

Sally (eds) Wastewater Use in Irrigated Agriculture: Coordinating the Livelihood and Environmental Realities. IWMI, IDRC

Coroi, M., Skeffington, M. S., Giller, P., Smitha, C., Gormally, M. & O’Donovand, G. (2004). Vegetation diversity and stand structure in streamside forests in the south of Ireland. Forest Ecology and Management, 202: 39–57

Costantini, V. & Monni, S. (2008). Environment, human development and economic growth. Ecological Economics, 64: 867 – 880

Cropper, M. & Griffith, C. (1994). The interaction of population and growth and environmental quality. The American Economic Review, 84(2): 250–254

Czech, B. (2003). Technological progress and biodiversity conservation: a dollar spent, a dollar burned. Conservation Biology, 17: 1455–1457

Danso, G., Drechsel, P., Wiafe-Antwi, T. & Gyiele L. (2002). Income of farming systems around Kumasi, Ghana. Urban Agriculture Magazine, 7: 5–6

Danso. G., Drechsel, P., Wiafe-Antwi, T. & Gyiele, L. (2006). Income generated by farming systems in and around Kumasi. In: van Veenhuizen, R. (Ed.). Cities Farming for the Future: Urban Agriculture for Green and Productive cities. RUAF Foundation, IDRC and IIRR. p. 205-207

Davey, K. J. (1993). Elements of Urban Management. World Bank

Dazé, A. (2007). Climate Change and Poverty in Ghana. CARE International. 20pp.

DeKeyser, E. S., Kirby, D. R. & Ell, M. J. (2003). An index of plant community integrity: development of the methodology for assessing prairie wetland plant communities. Ecological Indicators, 3: 119–133

Devas, N. (2005). Metropolitan governance and urban poverty. Public Administration and Development, 25: 351–361

Dial, R. & Roughgarden, J. (1988). Theory of marine communities: the intermediate disturbance hypothesis. Ecology, 79: 1412–1424

Dinda, S. (2004). Environmental Kuznets Curve Hypothesis: A Survey. Ecological Economics, 49: 431– 455

Drechsel, P. (2006). Learning from rapid urban growth and reflection in Kumasi. Case study 25: Ghana. In: Hague, C. et al. (Eds.). Making Planning Work. A guide to approaches and skills. Intermediate Technology Publishing, p. 74-75

Driessen, P., Deckers, J., Spaargaren, O. & Nachtergaele, F. (Eds.) (2001). Lecture notes on the major soils of the world. Natural Resources and the Environment. World soil resources reports series 94. ISBN 925-104637-9

Duzgoren-Aydin, N. S., Wong, C. S. C., Aydin, A., Song, Z., You, M. & Li, X. D. (2006). Heavy metal contamination and distribution in the urban environment of Guangzhou, SE China. Environmental Geochemistry and Health, 28: 375-391

166

Egyir, S. N. (2008). Use of digital elevation models for watershed management in Aboabo, Kumasi. BSc thesis submitted to the Kwame Nkrumah University of Science and Technology, Ghana

Egziabher, A. G., Lee-Smith, D., Maxwell, D. G., Memon, P. A., Mougeot, L. J. A. & Sawio, C. J. (1994). Cities feeding people: an examination of urban agriculture in East Africa. Ottawa, ON, IDRC,. xiv + 146 p.: ill.

Ehrenfeld, G. J. (2000). Evaluating wetlands within an urban context. Urban Ecosystems, 4: 69–85

Eppinka, F. V., van den Bergh, C. J. M. J. & Rietveld, P. (2004). Modelling biodiversity and land use: urban growth, agriculture and nature in a wetland area. Ecological Economics, 51: 201– 216

Erenstein, O., Oswald, A. & Mahaman, M. (2006). Determinants of lowland use close to urban markets along an agro-ecological gradient in West Africa. Agriculture, Ecosystems & Environment, 117(2-3): 205-217

Esty, D. C., Levy, M. A., Srebotnjak, T. & de Sherbinin A. (2005). 2005 Environmental Sustainability Index: Benchmarking National Environmental Stewardship. New Haven, Conn.: Yale Center for Environmental Law & Policy

Faithful, J. & Finlayson, W. (2005). Water quality assessment for sustainable agriculture in the Wet Tropics - a community-assisted approach. Marine Pollution Bulletin, 51(1-4): 99-112

FAO (1985). Review of the state of aquatic pollution of West and Central African inland waters Series. CIFA Occasional Paper-CIFA/OP12, 32pp.

FAO (1998). Wetland characterization and classification for sustainable agricultural development. Food and Agriculture Organization of the United Nations (FAO), Sub-Regional Office for East and Southern Africa (SAFR)

FAO (2007). Profitability and sustainability of urban and peri-urban agriculture. Agricultural Management, Marketing and Finance Occasional Paper 19. ISBN 978-92-5-105881-7. FAO, Rome

Federal Interagency Committee for Wetlands Delineation (1989). Federal Manual for identifying and delineating jurisdictional wetlands. US Government Printing Office, Washington, DC, 76pp.

Federal Register (1980). Guidelines for Specification of Disposal Sites for Dredged or Fill Material". U.S. Government Printing Office, Washington, DC

Fedeski, M. & Gwilliam, J. (2007). Urban sustainability in the presence of flood and geological hazards: The development of a GIS-based vulnerability and risk assessment methodology. Landscape and Urban Planning, 83: 50–61

Finlayson, C. M., Davidson, N. C., Spiers, A. G. & Stevenson, N. J. (1999). Global wetland inventory–current status and future priorities. Marine & Freshwater Research, 50(8): 717–727

Fobil, J. N. & Atuguba, R. A. (2004). Ghana: changing urban environmental ills in slum communities. International Journal of Environmental Policy and Law, 34(4-5): 206-215

167

Foeken, D. (2006). “To Subsidise My Income”: Urban farming in an East African Town. Koninklijke Brill NV, Leiden, The Netherlands. xvi + 224 pp.

Foeken, D. W. J. & Owuor, S O. (2008). Farming as a livelihood source for the urban poor of Nakuru, Kenya. Geoforum, 39(6): 1978-1990

Fortney, R. H. Benedict, M., Gottgens, J. F, Walters, T. L., Leady B. S. & Rentch, J. (2004). Aquatic plant community composition and distribution along an inundation gradient at two ecologically-distinct sites in the Pantanal region of Brazil. Wetlands Ecology and Management, 12: 575–585

Fritioff, Å. (2005). Metal accumulation by plants - evaluation of the use of plants in stormwater treatment. Dissertation for the Degree of Doctor of Philosophy in Plant physiology presented at Stockholm University September 16, 2005. Stockholm University, ISBN 91-7155-111-5 pp 1-56

Ghana Environmental Protection Agency (2001). Initial national communication to the United Nations Framework Convention on Climate Change (UNFCCC). EPA, Accra

Ghana News Agency (2010). Ghana to get radical on preservation of wetlands. Ghanaweb Article 176195. Thursday, 4 February 2010

Ghana Statistical Service (2002a). 2000 Population and Housing Census: Summary Report of Final Results. Accra, Ghana

Ghana Statistical Service (2002b). 2000 Population and Housing Census: Special Report on 20 Largest Localities. Accra, Ghana

Ghana Statistical Service (2005). 2000 Population and Housing Census: Ashanti Region Analysis of District Data and implications for Planning. Accra, Ghana

Ghrefat, H. & Yusuf, N. (2006). Assessing Mn, Fe, Cu, Zn, and Cd pollution in bottom sediments of Wadi Al-Arab Dam, Jordan. Chemosphere, 65: 2114–2121

Gibbs, R. J. (1977). Transport phases of transition metals in the Amazon and Yukon Rivers. Geological Society of America Bulletin, 88: 829–843

Gilbert, R.O. (1987). Statistical methods for environmental pollution monitoring. Van Nostrand Reinhold, New York

Government of Ghana (1992). The Constitution of the Republic of Ghana. Asempa Publishers, Accra

Government of Ghana (1993). Republic of Ghana Act 462 - Local Government Act of 1993. Asempa Publishers, Accra

Government of Ghana (1996). Republic of Ghana Act 522 - Water Resources Commission Act, 1996. Asempa Publishers, Accra

Green, M. B. & Upton, J. (1994). Constructed reed beds: Appropriate technology for small communities. Wetland Systems for Water Pollution Control, 32(3): 339-348

Griebenow, G. & Kishore, S. (2009). Mainstreaming Environment and Climate Change in the Implementation of Poverty Reduction Strategies. The International Bank for Reconstruction and Development/The World Bank, Washington, D.C. 20433, USA

168

Griggs, G. B. & Paris, L. (1982). Flood control failure: San Lorenzo River, California. Environmental Management, 6(5): 407-419

Grimm, N. B., Faeth, S. H., Golubiewski, N. E., Redman, C. L., Wu, J., Bai, X. & Briggs J. M. (2008). Global change and the ecology of cities. Science, 319 (5864): 756-760

Gunter, B. G., Cohen M. J. & Lofgren, H. (2005). Analysing Macro-Poverty Linkages: An Overview. Development Policy Review, 23 (3): 243-265

Haflinger, E. & Scholz, H., (1980a). Grassweeds 1. CIBA-GEIGY Ltd, Basle, Switzerland.

Haflinger, E. & Scholz, H., (1980b). Grassweeds 2. CIBA-GEIGY Ltd, Basle, Switzerland

Haflinger, E., Kuhn, U., Hamet-Ahti, L., Cook, C. D. K., Faden, R. & Speta, F. (1982). Monocot Weeds 3. CIBA-GEIGY Ltd, Basle, Switzerland

Halls, A. J. (ed.) (1997). Wetlands, Biodiversity and the Ramsar Convention: The Role of the Convention on Wetlands in the Conservation and Wise Use of Biodiversity. Ramsar Convention Bureau, Gland, Switzerland

Harikumar, P. S., Prajitha, K. & Silpa, S. (2010). Assessment of heavy metal contamination in the sediments of a river draining into a Ramsar site in the Indian sub-continent. Journal of Advanced Laboratory Research in Biology, 1(2): 157-169

Harper, D. M., Adams, C. & Mavuti, K. (1995). The aquatic plant communities of the Lake Naivasha wetland, Kenya: pattern, dynamics and conservation. Wetlands Ecology and Management, 3(2): 111-123

Harvey, D. (1985a). Consciousness and the urban experience: studies in the history and theory of capitalist urbanization. Johns Hopkins University Press, Baltimore, Md. xix, 293pp.

Harvey, D. (1985b). The urban experience. Baltimore, MD: The John Hopkins University Press

Harvey, K. R. & Hill, G. J. E. (2001). Vegetation mapping of a tropical freshwater swamp in the Northern Territory, Australia: a comparison of aerial photography, Landsat TM and SPOT satellite imagery. International Journal of Remote Sensing, 22(15): 2911–2925

Haslam, S. M. (2003). Understanding Wetlands: Fen, Bog and Marsh. London, Taylor and Francis Group

Hasse, J. E., Lathrop, R. G. (2003). Land resource impact indicators of urban sprawl. Applied Geography, 23: 159–175

Hejcmanovā-Nežerková, P. & Hejcman, M. (2006). A canonical correspondence analysis (CCA) of the vegetation–environment relationships in Sudanese savannah, Senegal. South African Journal of Botany, 72: 256 – 262

Hidalgo, J., Masson, V., Baklanov, A., Pigeon, G. & Gimeno, L. (2008). Advances in urban climate modelling. Annals of the New York Academy of Sciences, 1146: 354-74

Holland, M. D., Kasanga, R. K., Lewcock, C. P. & Warburton, H. J. (1996). Peri-urban baseline studies, Kumasi, Ashanti region, Ghana. Final Technical Report for project R6448. Chatham: NRI

169

Huang, S., Taniguchi, M. Yamano, M. & Wang, C. (2009). Detecting urbanization effects on surface and subsurface thermal environment — a case study of Osaka. Science of the Total Environment, 407(9): 3142-3152

Intergovernmental Panel on Climate Change (IPCC) (2007). Climate Change 2007. Fourth Assessment Report. Cambridge University Press, Cambridge, UK

International Recovery Platform (2009). Environment Issues in Recovery. http://www.gdrc.org/uem/disasters/disenvi/environment-recovery.pdf (accessed 22/06/2010)

Islam, Md. A., Thenkabail, P. S., Kulawardana, R. W., Alankara, R., Gunasinghe, S., Edussriya, C., & Gunawardana, A. (2008). Semi-automated Methods for mapping wetlands using Landsat ETM+ and SRTM data. International Journal of Remote Sensing, 29 (24): 7077 - 7106

Jacobs, G., Macfarlane, R. & Asokan, N. (1997). Comprehensive Theory of Social Development. International. Center for Peace and Development, 2352 Stonehouse Drive, Napa, CA 94558, USA

Janssen, R., Goosena, H., Verhoevenb, M. L., Verhoevenb, J. T. A., Omtzigta, A. Q. A. & Maltby, E. (2005). Decision support for integrated wetland management. Environmental Modelling and Software, 30: 215-229

Jenerettea, G. D., Marussicha, W. A. & Newell, J. P. (2006). Linking ecological footprints with ecosystem valuation in the provisioning of urban freshwater. Ecological Economics, 59: 38 – 47

Joosten, H. & Clarke, D. (2002). Wise Use of Mires and Peatlands- Background and Principles Including a Framework for Decision-making. International Mire Conservation group and International Peat Society. 304pp.

Kaplan, D., Wheeler, J. & Holloway, S. (2009). Urban Geography (2nd Edition). John Wiley & Sons, Inc. NJ, USA

Keddy, P. A. (2000). Wetland Ecology: Principles and Conservation. Cambridge University Press

Keraita, B., Drechsel, P. & Amoah, P. (2003). Influence of urban wastewater on stream water quality and agriculture in and around Kumasi, Ghana. Environment and Urbanization, 15(2): 171-178

Keraita, B., Drechsel, P. & Konradsen, F. (2008a). Using on-farm sedimentation ponds to improve microbial quality of irrigation water in urban vegetable farming in Ghana. Water Science and Technology, 57(4): 519-525.

Keraita, B., Drechsel, P., Huibers, F. & Raschid-Sally, L. (2002). Wastewater use in informal irrigation in Urban and Peri-urban areas of Kumasi, Ghana. Urban Agriculture Magazine, 8: 11-13

Keraita, B., Drechsel, P., Huibers, F. & Raschid-Sally, L. (2002). Wastewater use in informal irrigation in Urban and Peri-urban areas of Kumasi, Ghana. Urban Agriculture Magazine, 8: 11-13

170

Keraita, B., Drechsel, P., Konradsen, F. & Vreugdenhil, R. C. (2008b). Potential of simple filters to improve microbial quality of irrigation water used in urban vegetable farming in Ghana. Journal of Environmental Science and Health, 43: 749–755.

Kidane-Mariam, T. (2003). Environmental and Habitat Management: The Case of Ethiopia and Ghana. Environmental Management, 31(3): 313–327

Killeen, D., Khan, S. R (2002). Poverty and Environment. A contribution to the World Summit on Sustainable Development Process. International Institute for Environment and Development, London.

Kim, J. (2006). A growth model for root systems of virtual plants with soil and moisture control. IEICE Transactions on Information and Systems, E89-D(5): 1743-1750

Klinkenberg, E, McCall, P. J., Wilson, M. D., Amerasinghe F. P., & Donnelly M. J. (2008). Impact of urban agriculture on malaria vectors in Accra, Ghana. Malaria Journal, 7: 1-9

Kumasi Metropolitan Assembly (2006). A Profile of Ashanti Region of Ghana. KMA, Kumasi. 43pp.

Lant, C. L. (1994). The role of property rights in economic research on U.S. wetlands policy. Ecological Economics, 11(1): 27-33

Levy, P. E., Cannell, M. G. R. & Friend A. D. (2004). Modelling the impact of future changes in climate, CO2 concentration and landuse on natural ecosystems and the terrestrial carbon sink. Global Environmental Change, 14: 21–30

Lunetta, R. S., Balogh, M. E. & Merchant, J. W. (1999). Application of multi-temporal Landsat 5 TM imagery for wetland identification. Photogrammetric Engineering and Remote Sensing, 65, 1303-1310

Lydecker, M. & Drechsel, P. (2010). Urban agriculture and sanitation services in Accra, Ghana: the overlooked contribution. International Journal of Agricultural Sustainability, 8(1&2): 94-103

Lyon, J. G. (1993). Practical handbook for wetland identification and delineation. CRC Press, Inc., Boca Raton, Florida

Mabogunje, A. L. (1990). Urban planning and the post-colonial state in Africa: a research overview. African Studies Review, 33(2): 121–203

Magigi, W. (2008). Improving urban land governance with emphasis on Integrating agriculture based livelihoods in Spatial land use planning practise in Tanzania. PhD thesis at Albert-Ludwigs-Universität Freiburg im Breisgau, Germany

Magnani, E. (2000). The Environmental Kuznets Curve, environmental protection policy and income distribution. Ecological Economics, 32: 431–443

Magnani, E. (2001). The Environmental Kuznets Curve: development path or policy result? Environmental Modelling & Software, 16:157–165

Maltby, E. & Barker, T. (eds) (2009). The Wetlands Handbook. Wiley and Blackwell, Oxford

171

Marcotullio, P. J. & Boyle, G (eds) (2003). Defining an Ecosystem Approach to Urban Management and Policy Development. United Nations University Institute of Advanced Studies (UNU/IAS)

Mathias, M. E. & Moyle, P. (1992). Wetland and aquatic habitats. Agriculture, Ecosystems & Environment, 42: 165-176

Matthys, B., Vounatsou, P., Raso, G., Tschannen, A. B., Becket, E.G., Gosoniu, L., Cissé, G., Tanner, M., N’goran, E. K., & Utzinger, J. (2006). Urban farming and malaria risk factors in a medium-sized town in Côte D’Ivoire. The American Journal of Tropical Medicine and Hygiene, 75(6): 1223–1231

Mazzotti, F. J. & Morgenstern, C. S. (1997) A scientific framework for managing urban natural areas. Landscape and Urban Planning, 38: 171-181

McClean, D. (ed.) (2010). World Disasters Report 2010 - focus on urban risk. International Federation of Red Cross and Red Crescent Societies, Geneva, Switzerland. ISBN 978-92-9139-156-1

Mcdonnell, M. J. & Pickett, S. T. A. (1990). Ecosystem structure and function along urban-rural gradients: an unexploited opportunity for ecology. Ecology, 71(4): 1232-1237

McGregor, D., Simon, D. & Thompson, D. (eds) (2006). The Peri-Urban Interface: Approaches to sustainable natural and human resource use. Earthscan/ James & James, xxiv + 336pp.

McPherson, M. A. & Nieswiadomy, L. M., (2005). Environmental Kuznets curve: threatened species and spatial effects. Ecological Economics, 55: 395– 407

Millennium Ecosystem Assessment (2000). Ecosystems and human well-being: current state and trends. Island Press

Millennium Ecosystem Assessment (2005). Ecosystems and human well-being: current state and trends. Island Press

Miller, R. L. & Fujii, R (2010). Plant community, primary productivity, and environmental conditions following wetland re-establishment in the Sacramento-San Joaquin Delta, California. Wetlands Ecology & Management, 18: 1–16

Mills, J. H. & Waite T. A. (2009). Economic prosperity, biodiversity conservation, and the environmental Kuznets curve. Ecological Economics, 68(7): 2087-2095

Ministry of Lands and Forestry (1999a). Managing Ghana's Wetlands: a National Wetlands Conservation Strategy. Accra, Ghana

Ministry of Lands and Forestry (1999b). National Land Policy. Accra, Ghana.

Ministry of Water Resources, Works and Housing (2007). National Water Policy. Accra, Ghana

Mireri, C., Atekyerezab, P., Kyessi, A. & Mushi, N. (2007). Environmental risks of urban agriculture in the Lake Victoria drainage basin: A case of Kisumu municipality, Kenya. Habitat International, 31: 375–386

172

Mitsch, W. J. & Gosselink, J. G. (1993). Wetlands. 2nd ed. Van Nostrand. Reinhold, New York

Mitsch, W. J. Gosselink J.G. (2000). The value of wetlands: importance of scale and landscape setting. Ecological Economics, 35: 25–33

Modarres, R. & da Silva, V. R. (2007). Rainfall trends in arid and semi-arid regions of Iran. Journal of Arid Environments, 70: 344–355

Mougeot, L. J. A. (1999). Urban agriculture: definition, presence, potentials and risks, main policy challenges. In Proceedings of International Workshop on Growing Cities Growing Food: Urban Agriculture on the Policy Agenda, La Habana, Cuba

Müller, G. (1969). Index of geoaccumulation in the sediments of the Rhine River. Geojournal, 2: 108–118

Mumford, L. (1961). The City in History: Its Origins, Its Transformations, and Its Prospects. New York: Harcourt, Brace

Mumphrey, A. J. & Whalen, T. F. (1977). Economic growth, lifestyle preferences, and urban size. Urban Ecology, 2: 259-278

Munasinghe, M. (1999). Is environmental degradation an inevitable consequence of economic growth: tunnelling through the environmental Kuznets curve. Ecological Economics, 29: 89–109

Mundia, C. N. & Aniya, M. (2006). Dynamics of landuse/cover changes and degradation of Nairobi City, Kenya. Land Degradation & Development, 17: 97–108

National Development Planning Commission & UNDP (2010). 2008 Ghana Millennium Development Goals Report. National Development Planning Commission, Accra, Ghana

National Development Planning Commission (2001). Poverty Reduction Strategy (PRS) (2001– 2005). National Development Planning Commission, Accra, Ghana

National Development Planning Commission (2005). Growth and Poverty Reduction Strategy (GPRS II) (2006 – 2009). National Development Planning Commission, Accra, GhanaGhana

Nelson, K. C., Palmer, M. A., Pizzuto, J. E., Moglen, G. E., Angermeier, P. L., Hilderbrand, R. H., Dettinger, M. & Hayhoe, K. (2009). Forecasting the combined effects of urbanization and climate change on stream ecosystems: from impacts to management options. Journal of Applied Ecology, 46(1): 154-163

Niasse, M., Afouda, A. & Amani, A. (eds) (2004). Reducing West Africa's vulnerability to climate on water resources, wetlands and desertification: Elements for a regional strategy for preparedness and adaptation. IUCN: Gland, Switzerland and Cambridge, UK. xviii + 66pp.

Niemelä, J. (1999). Ecology and urban planning. Biodiversity and Conservation, 8(1): 119-131

Obeng, H. B. (1971). Soil Map of Ghana. Soil Research Institute, CSIR, Ghana

Obeng-Odoom, F. (2007). Enhancing urban productivity in Africa. Opticon 1826, 2(1): 1–9

173

Obuobie, E., Keraita, B., Danso, G., Amoah, P., Cofie, O. O., Raschid-Sally L. & Drechsel, P. (2006). Irrigated urban vegetable production in Ghana: Characteristics, benefits and risks. IWMI-RUAF-CPWF, Accra, Ghana: IWMI. 150pp.

Odemerho, F. 0. (1993). flood control failures in a third world city: Benin City, Nigeria -some environmental factors and policy issues. GeoJournal, 29(4): 371-376

Okley, B. L. (2004). Legislation and implementation of international environmental law by African countries: a case study of Ghana. LLM Theses and Essays, University of Georgia. Paper 38

Orloci, L. (1978). Multivariate Analysis in Vegetation Research. Dr W. Junk b.v., Publishers. The Hague. 451pp.

Ot’ahel’ová, H., Valachovič, M. & Hrivnák, R. (2007). The impact of environmental factors on the distribution pattern of aquatic plants along the Danube River corridor (Slovakia). Limnologica, 37: 290–302

Owen, C. R. (1999). Hydrology and history: land use changes and ecological responses in an urban wetland. Wetlands Ecology and Management, 6: 209–219

Ozesmi, S. L. & Bauer, M. E. (2002). Satellite remote sensing of wetlands. Wetlands Ecology and Management, 10: 381 – 402

Pacione, M. (2009). Urban Geography: A Global Perspective (3rd edition). Routledge. 736pp.

Panayotou, T. (1993). Empirical tests and policy analysis of environmental degradation at different stages of economic development. Working Paper WP238, Technology and Employment Programme, Geneva: International Labour Office

Panayotou, T. (1997). Demystifying the Environmental Kuznets Curve: turning a black box into a policy tool. Environment and Development Economics, 2: 465-484

Panayotou, T. (2003). Economic growth and the environment. Paper prepared for and presented at the Spring Seminar of the United Nations Economic Commission for Europe, Geneva, March 3, 2003

Pauleit, S. & Duhme, F. (2000). Assessing the environmental performance of land cover types for urban planning. Landscape and Urban Planning, 52: 1-20

Payne, G. (1997). Urban land tenure and property rights in developing countries: a review. Intermediate Technology, London, UK. 80pp.

Petrov, L. O., Barredo J. I., Lavalle, C., Sagris, V., De Roo, A., Gierk, M. & Genovese, E. (2005). Adaptation strategies for flood mitigation in the urban environment: the case study, Friuli-Venezia Giulia (FVG), Italy. International conference on innovation advances and implementation of flood forecasting technology, 17 to 19 October 2005, Tromsø, Norway. ACTIF/FloodMan/FloodRelief

Phillips, A. (ed) (1998). Economic Values of Protected Areas: Guidelines for Protected Area Managers. Best Practice Protected Area Guidelines Series No. 2. Task Force on Economic Benefits of Protected Areas of the World Commission on Protected Areas (WCPA) of IUCN, in collaboration with the Economics Service Unit of IUCN

174

Pickett, S. T. A., Burch, W. R., Dalton, S. E., Foresman, T. W., Grove, J. M. & Rowntree, R. (1997). A conceptual framework for the study of human ecosystems in urban areas. Urban Ecosystems, 1: 185–199

Pickett, S. T. A., Kolasa, J., Armesto, J. J. &. Collins, S. L. (1989). The ecological concept of disturbance and its expression at various hierarchical levels. Oikos, 54: 129-136

Pongrácz, R., Bartholy, J. & Dezső, Z. (2006). Remotely sensed thermal information applied to urban climate analysis. Advances in Space Research, 37: 2191–2196

Pongrácz, R., Bartholy, J. & Dezső, Z. (2010). Application of remotely sensed thermal information to urban climatology of Central European cities. Physics and Chemistry of the Earth, Parts A/B/C, 35(1-2): 95-99

Qadeer, M. A. (2004). Urbanization by implosion. Habitat International, 28: 1–12

Quansah, C., Drechsel, P. & Bakang, J. A. (2005). Participatory communication strategies to improve natural resources management in peri-urban Kumasi, Ghana. In Mear, A. (Ed.). Food and Water for Millions. Participatory communication for land and water management. IDRC, Ottawa, p.81-88

Rahman, A. (2006). Assessing income-wise household environmental conditions and disease profile in urban areas: Study of an Indian city. GeoJournal, 65: 211–227

Ramsar Convention Secretariat (2005). Wetlands and water: supporting life, sustaining Livelihoods. 9th Meeting of the Conference of the Parties to the Convention on Wetlands (Ramsar, Iran, 1971). Kampala, Uganda, 8-15 November 2005

Ramsar Convention Secretariat (2006). The Ramsar Convention Manual: a guide to the Convention on Wetlands (Ramsar, Iran, 1971), 4th ed. Ramsar Convention Secretariat, Gland, Switzerland

Ramsar Convention Secretariat (2007). Ramsar handbooks for the wise use of wetlands, 3rd edition. Ramsar Convention Secretariat, Gland, Switzerland

Ramsar Convention Secretariat (2008). Healthy Wetlands, Healthy People. 10th Meeting of the Conference of the Contracting Parties to the Convention on Wetlands (Ramsar, Iran, 1971). Changwon, Republic of Korea, 28 October - 4 November, 2008

Ravallion, M, Shaohua, C. & Sangraula, P. (2007a). The Urbanisation of Global Poverty. Background paper to the 2008 World Development Report

Ravallion, M., Chen, S. & Sangraula, P. (2007b). New Evidence on the Urbanization of Global Poverty. World Bank Policy Research Working Paper 4199. Development Research Group, World Bank

Rees, W. E. (2003). Understanding urban ecosystems: an ecological economics perspective. In: Understanding Urban Ecosystems: A New Frontier for Science and Education. A.R. Berkowitz, C.H. Nilon, and K.S. Hollweg (eds.), Springer-Verlag, New York, 115–136

Reinelt, L., Horner, R. & Azous, A. (1998). Impacts of urbanization on palustrine (depressional freshwater) wetlands—research and management in the Puget Sound region. Urban Ecosystems, 2: 219–236

175

Rietbergen, S., Bishop, J. & Mainka S. (2002). Ecosystem Conservation - A neglected tool for poverty reduction. A contribution to the World Summit on Sustainable Development Process. International Institute for Environment and Development, London

Risbey, J. S., Hamza, K. & Marsden, J. S. (2007). Use of climate scenarios to aid in decision analysis for interannual water supply planning. Water Resources Management, 21: 919–932

Rodrigo, F. S., Pozo-Vazquez, D., Esteban-Parra, M. J. & Castro-Diez, Y. (2001). A reconstruction of the winter North Atlantic Oscillation index back to A.D. 1501 using documentary data in southern Spain. Journal of Geophysical Research, 106: 805–814

Roy, M. (2009). Planning for sustainable urbanisation in fast growing cities: mitigation and adaptation issues addressed in Dhaka, Bangladesh. Habitat International, 33: 276–286

Sachs, J. McArthur, J., Schmidt-Traub, G., Bahadur C., Faye M. & Kruk, M. (2004) Millennium Development Goals needs assessments. Country case studies of Bangladesh, Cambodia, Ghana, Tanzania and Uganda. UN Millennium Project

Sánchez-Rodríguez, R., Seto, K. C., Simon, D., Solecki, W. D., Kraas, F. & Laumann, G. (2005). Science Plan: Urbanization and Global Environmental Change. IHDP Report No. 15. International Human Dimensions Programme on Global Environmental Change (IHDP). Bonn, Germany

Sassa, L. L., Bozek, M. A., Hauxwell, J. A., Wagner, K. & Knight, S. (2010). Response of aquatic macrophytes to human land use perturbations in the watersheds of Wisconsin lakes, U.S.A. Aquatic Botany, 93: 1-8

Satterthwaite, D. (2003). The links between poverty and the environment in urban areas of Africa, Asia, and Latin America. Annals of the American Academy of Political and Social Science, 590: 73–92

Satterthwaite, D. (2008). Climate change and urbanization: effects and implications for urban governance. United Nations Expert Group Meeting on Population Distribution, Urbanization, Internal Migration and Development. UN/POP/EGM-URB/2008/16

Schiff, K. C. & Weisberg, S. B. (1999). Iron as a reference element for determining trace metal enrichment in Southern California coast shelf sediments. Marine Environmental Research, 48: 161–176

Schmidt, S. (2005). Cultural influences and the built environment: an examination of Kumasi, Ghana. Journal of Urban Design, 10(3): 353-370

Schropp, S. J. & Windom, H. L. eds. (1988). A guide to the interpretation of metal concentrations in estuarine sediments. Office of Water Policy, Florida Department of Environmental Protection, Tallahassee

Secretariat of the Convention on Biological Diversity (2005). Handbook of the Convention on Biological Diversity Including its Cartagena Protocol on Biosafety, 3rd edition. Montreal, Canada

Secretariat of the Convention on Biological Diversity (2007). Cities and Biodiversity: Achieving the 2010 Biodiversity Target. Mayors’ Meeting on the Contribution of Cities to the Achievement of the 2010 Biodiversity Target. Curitiba, Brazil, 26-28 March 2007. UNEP/CBD/Cities/1/3

176

Sekabira, K., Oryem Origa, H., Basamba, T. A., Mutumba, G. & Kakudidi, E. (2010). Assessment of heavy metal pollution in the urban stream sediments and its tributaries. International Journal of Environmental Science and Technology, 7(3): 435-446

Shutes, R. B. E. (2001). Artificial wetlands and water quality improvement. Environment International, 26: 441 - 447

Shutes, R. B. E., Revitt, D. M., Lagerberg, I. M. & Barraud, V. C. E. (1999). The design of vegetative constructed wetlands for the treatment of highway runoff. Science of the Total Environment, 235: 189-197

Silvius, M. J., Oneka, M. & Verhagen, A. (2000). Wetlands: lifeline for people at the edge. Physics and Chemistry of the Earth (B), 25(7-8): 645-652

Simex, S. A. & Helz, G. R. (1981). Regional geochemistry of trace elements in Checapeake Bay. Environmental Geology, 3: 315-323

Sims, K. R. E. & Schuetz, J. (2009). Environmental regulation and land use change: do local wetlands bylaws slow the conversion of open space to residential uses? Regional Science and Urban Economics, 39(4): 409-421

Small, C. & Lu, J. W. T (2006). Estimation and vicarious validation of urban vegetation abundance by spectral mixture analysis. Remote Sensing of Environment, 100: 441 – 456

Small, C. (2001). Estimation of urban vegetation abundance by spectral mixture analysis. International Journal of Remote Sensing, 22(7): 1305-1334

Smith, M. B. (1993). A GIS-based distributed parameter hydrologic model for urban areas. Hydrological Processes, 7: 45-61

Smith, M. E. (2005). City size in late post-classic Mesoamerica. Journal of Urban History, 31(4): 403-434

Sorrell, B. K., Partridge, T. R., Clarkson, B. R., Jackson, R. J., Chague´-Goff, C., Ekanayake, J., Payne, J., Gerbeaux, P. & Grainger, N. P. J. (2007). Soil and vegetation responses to hydrological manipulation in a partially drained polje fen in New Zealand. Wetlands Ecology and Management, 15: 361–383

Stern, D. I. & Common, M. S. (2001). Is there an environmental Kuznets curve for sulfur? Journal of Environmental Economics and Management, 41: 162–178

Stern, D. I. (2004). The rise and fall of the environmental Kuznets curve. World Development, 32(8): 1419–1439

Takebayashi, H. & Moriyama, M. (2009). Study on the urban heat island mitigation effect achieved by converting to grass-covered parking. Solar Energy, 83(8):1211-1223

Taylor, B. L. (1993). The Influence of Wetland and Watershed Morphological Characteristics on Wetland Hydrology and Relationships to Wetland Vegetation Communities. MSCE thesis, Department of Civil Engineering, University of Washington

Terry, P. J. (1983). Some common crop weeds of West Africa and their control. USAID Regional Food Crop Protection Project

177

The World Bank (1991). Urban Policy and Economic Development; an agenda for the 1990s. A World Bank Policy Paper, World Bank

The World Bank (2000). Entering the 21st Century: World Development Report, 1999-2000, pp. 126-138

The World Bank (2007). Ghana: Country Environmental Analysis. Environment and Natural Resources. Agence Française de Développement (AFTEN), Royal Netherlands Embassy and Department for International Development, Africa Region. Report No. 36985-GH

Thenkabail, P. S. & Nolte, C. (1996). Capabilities of Landsat-5 Thematic Mapper (TM) data in regional mapping and characterization of inland valley agroecosystems in West Africa. International Journal of Remote Sensing, 17(8): 1505-1538

Thenkabail, P. S., Nolte, C. & Lyon, J. G. (2000). Remote sensing and GIS modeling for selection of benchmark research area in the inland valley agroecosystems of West and Central Africa. Photogrammetric Engineering and Remote Sensing, 66(6): 755-768

Thibault, P. A. & Zippererb, W. C. (1994). Temporal changes of wetlands within an urbanizing agricultural landscape. Landscape and Urban Planning, 28: 245-251

Tomlinson, D. L., Wilson, J. G., Harris, C. R., & Jeffrey, D.W. (1980). Problems in the assessment of heavy-metal levels in estuaries and the formation of a pollution index. Helgolander Wissenschaftliche Meeresuntersuchungen, 33: 566–575

Töyrä, J., Pietroniro, A., Martz, L. W. & Prowse, T. D. (1999). A multi-sensor approach to wetland flood monitoring. Hydrological Processes, 16 (8): 1569-1581

Tschakert, P., Huber-Sannwald, E, Ojima, S. D, Raupach, M. R, Schienke, E. (2008). Holistic, adaptive management of the terrestrial carbon cycle at local and regional scales. Global Environmental Change, 18: 128–141

Turekian, K. K. & Wedepohl, K. H. (1961). Distribution of the elements in some major units of the earth's crust. Geological Society of America Bulletin, 72: 175-192

Turner, R. K., van den Bergh, J. C. J. M., Söderqvist, T., Barendregt, A., van der Straaten, J., Maltby, E. & van Ierland, E. C. (2000). Ecological-economic analysis of wetlands: scientific integration for management and policy. Ecological Economics, 35: 7–23

U. S. Environmental Protection Agency (1985). Sediment sampling quality assurance user’s guide. EPA/600/4-85/048. PB85-233542. Environmental Monitoring Systems Laboratory. Las Vegas, NV

UN-HABITAT (1996). An Urbanizing World-Global Report on Human Settlements 1996. Oxford. Oxford University Press. ISBN: 0-19-823346-9

UN-HABITAT (2008a). Country Programme Document 2008 – 2009: Ghana. Regional Office for Africa and the Arab States, Nairobi, Kenya

UN-HABITAT (2008b). State of the Word’s Cities 2008/2009: Harmonious Cities. Earthscan, London

UN-HABITAT (2009). Ghana urban profile. UN-Habitat Regional and Information Office, Nairobi, Kenya

178

United Nations (2008). World Urbanization Prospects: The 2007 Revision (Highlights). United Nations, New York

United Nations Department of Economic and Social Affairs (2005). Changing unsustainable patterns of consumption and production: human settlements and water. Background Paper No.4 DESA/DSD/2005/4

United Nations Inter-Agency Secretariat of the International Strategy for Disaster Reduction (UN/ISDR) (2005). Hyogo Framework for Action 2005-2015: Building the Resilience of Nations and Communities to Disasters. World Conference on Disaster Reduction 18-22 January 2005, Kobe, Hyogo, Japan (A/CONF.206/6)

United Nations Population Fund (UNFPA) (2007). State of World Population 2007. UNFPA.

UNU-IAS Urban Ecosystems Management Group (2004). Cities as Drivers of Sustainable Development. World Urban Forum 2004 Discussion Paper. United Nations University - Institute of Advanced Studies

Vincent, M. A. & Saatchi, S. S. (1999). Comparison of Remote Sensing Techniques for Measuring Carbon Sequestration. Jet Propulsion Laboratory, California Institute of Technology

Wagner, M. (2008). The carbon Kuznets curve: A cloudy picture emitted by bad econometrics? Resource and Energy Economics, 30: 388–408

Water Resources Commission (2008). Final draft buffer zone policy for managing river basins in Ghana. Accra, Ghana

White, H., Killick, T., Kayizzi-Mugerwa, S. & Savane, M. (2001). African Poverty at the Millennium: Causes, Complexities, and Challenges. The International Bank for Reconstruction and Development/The World Bank. Washington, DC

Wikum, W. A. & Shanholtzer, G. F. (1978). Application of the Braun-Blanquet cover-abundance scale for vegetation analysis in land development studies. Environmental Management, 2(4): 323-329

Wilby, R. L. (2007). A Review of Climate Change Impacts on the Built Environment. Built Environment, 33: 31-45

Williams, M. (ed). (1990). Wetlands: a threatened landscape. Basil Blackwell Ltd., Oxford, U.K. 419pp.

World Meteorological Organisation (2006). Environmental aspects of integrated flood management. The Associated Programme on Flood Management (APFM) Technical Document No. 3, Flood Management Policy Series. Geneva, Switzerland

Xu, H., Wang, X. & Xiao, G. (2000). A Remote Sensing and GIS integrated study on urbanization with its impact on arable lands: Fuqing City, Fujian Province, China. Land Degradation & Development, 11: 301-314

Yeboah, M. A. & Appiah-Yeboah, K. (2009). An examination of the cultural and socio-economic profiles of porters in Accra, Ghana. Nordic Journal of African Studies, 18(1): 1–21

Zedler, J. B. & Leach, M. K. (1998). Managing urban wetlands for multiple use: research, restoration, and recreation. Urban Ecosystems, 2: 189–204

179

9 APPENDICES Appendix 1. Survey instrument used for the wetland associated communities in Kumasi

Questionnaire number…………………………Area………………………………...............

Type of house..............................................................Date…………………………………

INTRODUCTION

The researcher is a student of the University of Bremen, Germany working on urban wetland ecology and flooding in Kumasi. The information you provide will be treated as confidential and will be used for academic purposes only. Thank you for your time and cooperation.

1.1 Do you own the house you live in yes [ ] no [ ]

1.2 How did you get access to this land? Family inheritance [ ] Chief [ ] Bought from

another person [ ] Squatter [ ] Not applicable [ ]

1.3 For how long have you been living in the area? Below 5years [ ] 5-10 years [ ]

10-15 years [ ] 15-20 years [ ] more than 20 years [ ]

Please rank the following options according the criterion below:

0= no change/no influence/no contribution/nothing, 1=… 2= … 3=… 4=… in increasing

order to 5=highest increase/most influential/most contribution.

1.4 What do you do for a living?

Job types 0 1 2 3 4 5 1. Trading 2. Farming 3. Artisan (e.g. car fitting/Saw milling/carpentry etc.) 4. Tie-dye manufacturing

5. Oil extracting (palm oil/palm kernel oil etc.)

6. Formal sector job 7. Nothing/unemployed/dependant

Other, please specify 1.5 What are the predominant economic activities in this area?

Activity/use of land at time of moving in 0 1 2 3 4 5 1. Agriculture (vegetables, animals, etc.) 2. Tie-dye industries 3. Hunting 4. Fishing 5. Car washing bays

180

6. Car fitting/spraying shops 7. Carpentry/sawmill 8. Trading

Other (specify) 1.6 What was the previous use of this wetland to the people of this area?

……………...................................................................................................................................

1.7 What is/are the current uses of this stream/wetland to the people of this area?

Uses 0 1 2 3 4 5 1. Agricultural purposes 2. Tie-dye industries 3.Waste disposal 4. Car washing bays 5. Car fitting shops 6. Carpentry 7. Water supply 8. Fishing

9. Rangeland/forage area for cows sheep and goats

10. Recreation/tourism Other, please state

1.8 What are the major landuse activities around the wetland (100 m from water channel)?

Activities 0 1 2 3 4 5 1. Agriculture (vegetables, animals, etc.) 2. Tie-dye industries

3. Waste disposal 4. Car washing bays

5. Car fitting shops

6. Carpentry 7. Rangeland/forage area 8. Recreation/tourism 9. Housing/construction

Other, please specify 1.9 What have been the effects of these activities (1.8 above) on the environment/wetland?

Changes 0 1 2 3 4 5 1. Channel width (size of the river) 2. Water depth 3. Number and type of living organisms (e.g. fish, plants animals)

4. Storm flow (amount of water in the river/channel immediately after rainfall)

5. Suitability of water for domestic/industrial purposes

181

Other, please specify 2.0 What changes have you observed in the wetland vegetation since you first settled?

Changes 0 1 2 3 4 5 1. Plant species diversity 2. Number of woody plants (e.g. trees) 3. Number of herbaceous plants (e.g. grasses) 4. Distribution of plants 5. Density of plants 6. Plant cover 7. Loss of some species 8. Presence of new species

Other, please specify 2.1 Buildings in wetland areas are major problems in Kumasi. Strongly agree [ ] Agree [ ]

Disagree [ ] Not sure [ ]

2.2 What are some of the effects building on wetland areas have on our environment?

Effects 0 1 2 3 4 5 1. Nothing 2. Causes flooding 3. Makes the environment dirty 4. No place to dump wastes anymore 5. Does not make the city beautiful

Other, please specify 2.3 Why do you think people build in wetland areas?

Reasons 0 1 2 3 4 5 1. Wetland areas are cheap 2. Close to amenities 3. Direct use of the water for domestic purposes 4. Close to family 5. Do their economic activities 6. It is “no man’s land”

Other, please specify 2.4 Have you experienced flood in this area? Yes [ ] No [ ]

2.5 If yes, how many times in a year do you experience floods? Once [ ] Twice [ ] Thrice [ ]

more than thrice [ ]

2.6. Which month(s) of the year do you experience the

floods?………………………………………

182

2.7. How long do the floodwaters take to recede? Less than 5 days in a year [ ] less than 10

days in a year [ ] less than 15 days in a year [ ] less than 20 days in a year [ ] less than 25

days in a year [ ] About a month in a year [ ]

2.8. How do you assess the frequency of the floods compared to 10 years ago?

Less frequent [ ] More frequent [ ]

2.9 How do you deal with floods as an individual and community? (mark I or C or both)

Activities 0 1 2 3 4 5 1. Do nothing 2. Fetch/pump water out of flooded areas 3. Move out till water subsides 4. Create channels for easy flow of water 5. Dredge stream 6. Build embankments along stream 7. Complain to authorities

Others, please specify 3.0 Have you ever been educated on the usefulness of a wetland? Yes [ ] no [ ]

If yes, give the name(s) of the organisation(s)

…………………………………………………………………………………………………

3.1 Has there been any attempt by an organisation to move you from this site? yes [ ] no [ ]

If yes, name of organisation……………………………………………………………………

3.2 What will make you move out of this place?

Motivation/reason 0 1 2 3 4 5 1. Nothing, I love it here 2. Given money to move 3. New land/plot somewhere 4. When I make some money 5. When the water gets too much 6. When we are forced 7. I am already preparing to move 8. If I am provided alternative shelter

Other, please specify 4.0 Do you think the wetland should be protected? yes [ ] no [ ]

4.1 How are you and your community protecting this wetland?

Activities 0 1 2 3 4 5 1. We are doing nothing 2. Stop waste disposal on wetland 3. Stop building too close to wetland 4. Create channels for easy flow of water 5. Dredge stream

183

6. Build embankments along stream 7. Stop all economic activities on wetland 8. Complain to authorities

Others, please specify 4.2 Have houses in this neighbourhood ever been marked for demolition because of the

wetland? Yes [ ] No [ ]

If yes, when less than 2 yrs age [ ] 2-5 yrs ago [ ] 5-10 yrs ago [ ]

more than 10 yrs [ ]

4.3 Has any demolition been carried out in this area? Yes [ ] No [ ]

4.4 Do you support the demolition along water ways? Yes [ ] No [ ]

4.5 Who are responsible for the management of the wetlands in Kumasi?

Agencies/institutions 0 1 2 3 4 5 1. KMA 2. Hydrological Services Department 3. Town and Country Planning Department 4. The community and Assemblyman 5. NGOs (e.g. IWMI, Friends of Rivers and

Waterbodies)

6. NADMO 7. The chief/land owners

Others, please specify 4.6 Are there any on-going management practices to protect the wetland? Yes [ ] No [ ]

4.7. What factors affect the implementation of wetland policies and management plans?

Factors 0 1 2 3 4 5 1. Political influence 2. Changes in government 3. Corruption 4. Poverty levels 5. Lack of political will 6. Cost of demolition 7. Cultural values

Other, please specify 4.8 Do you know any of the laws regarding building or sitting of infrastructure in wetland

areas? Yes [ ] No [ ] If yes what is it

Background Information

5.1 Age: below 20yrs [ ] 20-30yrs [ ] 30-40yrs [ ]

40-50yrs [ ] 50-60yrs [ ] above 60yrs [ ]

5.2 Sex: Male [ ] Female [ ]

184

5.3 Education: Primary [ ] Middle School/JHS [ ] Secondary/SHS [ ]

Post-secondary/Tertiary [ ] No formal education [ ] Other [ ]

5.4 Occupation: …………………Household size:..............................................................

185

Appendix 2. Survey instrument used for wetland associated NGOs and government agencies in Kumasi

Date………………………………. INTRODUCTION The researcher is a student of the Kwame Nkrumah University of Science and Technology

working on urban wetlands in Kumasi. The information you provide will be treated as

confidential and will be used only for academic purposes. Thank you for your time and

cooperation.

1. In your opinion; what has been the overall change in the state of the wetland?

2. What are the tasks of inspectors/ officers- in- charge of developments on wetland areas?

3. How often do developers who flout the regulations come to your notice?

4. What punitive measures are there for such offenders?

5. Are these punitive measures effective?

6. How can the effectiveness be improved?

7. Are prospective (building) developers aware of these regulations/rules?

8. How often does your outfit organise educative seminars for people living on wetlands?

9. What medium do you use for such educative forum?

10. Are there any risks/dangers your inspectors/ officers face in visiting development sites in

wetland (e.g. from land guards, mob attack, etc)?

11. Under what circumstances can a building on wetlands be pulled down?

12. Are there maps/demarcations showing wetland areas?

13. What major constraints affect the implementation of wetlands management plan?

14. How do non-economic variables other than per capita income, such as the features of the

political system, and some cultural values, affect the implementation of wetland policies

and management plan?

186

App

endi

x 3.

Fie

ld d

ata

shee

t use

d fo

r wet

land

veg

etat

ion

and

ecol

ogic

al s

tudi

es in

Kum

asi

spec

ies

Dis

tanc

e fr

om w

ater

cha

nnel

Rem

ark/

com

men

t wa

ter

chan

nel

10m

20

m

30m

40

m

≥50m

# co

ver

# co

ver

# co

ver

# co

ver

# co

ver

# co

ver

187

App

endi

x 4.

Stu

dy s

ites

and

anth

ropo

geni

c ac

tiviti

es w

ithin

100

m f

rom

the

wat

er c

hann

el [

min

imum

buf

fer

dist

ance

set

by

Min

istr

y of

La

nds

and

Nat

ural

Res

ourc

es (W

ater

Res

ourc

es C

omm

issi

on, 2

008)

]

Act

ivity

/ la

ndus

e

Ato

nsu

Sawm

ill

A

Hig

h Sc

hool

Ju

nctio

n

B

FRN

R

farm

s

C

KN

UST

Polic

e St

atio

n

D

Fam

ily

Cha

pel

E

Spet

inpo

m

F

Gol

den

Tulip

Hot

el

G

Kaa

se-

Gui

nnes

s H

Bek

wai

roun

dabo

ut

J

Dak

wodj

om

K

# %

la

nd

cove

r #

%

land

co

ver

# %

la

nd

cove

r #

% la

nd

cove

r #

%

land

co

ver

# %

la

nd

cove

r #

%

land

co

ver

# %

la

nd

cove

r #

%

land

co

ver

# %

land

co

ver

Hou

sing

74

34

52

35

1

0 19

6

65

50

45

55

45

20

37

30

152

55

30

20

Farm

ing

2 0

3 5

9 2

15

2 7

10

11

5 12

12

3

15

41

20

3 5

Nat

ural

ve

geta

tion

50

45

98

91

30

30

65

50

0 20

70

Was

te d

ispo

sal

3 10

4

5 0

0 3

0 7

5 2

3 2

0 1

0 2

0 2

0 Sc

hool

s 0

0 1

0 0

0 1

1 1

2 1

1 0

0 2

1 3

0 0

0 C

hurc

h 0

0 5

0 0

0 1

0 7

1 0

1 2

0 4

1 3

0 2

0 M

osqu

e 1

0 0

0 0

0 0

0 0

0 0

1 0

0 0

0 0

0 0

0 C

arpe

ntry

/saw

m

ills

2 5

15

1 0

0 0

0 9

0 0

0 0

0 0

0 0

0 6

0 C

ar w

ashi

ng

bay

2 0

5 0

0 0

0 0

1 0

1 0

2 0

2 0

12

1 0

0 C

ar r

epai

rs

2 0

22

5 0

0 0

0 15

2

0 0

0 0

22

2 12

0

6 2

Roa

d

2 0

5 0

1 0

6 0

6 0

2 0

6 2

3 0

6 2

2 0

Gas

/fuel

fill

ing

stat

ion

0 0

1 0

0 0

0 0

2 0

3 1

2 1

3 1

3 0

0 0

Mar

ket

1

1 2

2 0

0 0

0 0

0 0

0 0

0 0

0 0

0 0

0 K

iosk

s

4 0

6 2

0 0

2 0

0 0

0 2

3 0

12

0 17

0

9 0

Oil

extr

actio

n 0

0 0

0 0

0 0

0 15

0

0 0

0 0

0 0

0 0

6 0

Tie-

Dye

in

dust

ry

0 0

1 0

0 0

0 0

0 0

0 0

0 0

0 0

0 0

0 0

Play

grou

nd

0 0

1 0

0 0

0 0

0 0

0 1

0 0

0 0

0 2

1 3

Fish

pro

cess

ing

1 0

2 0

0 0

0 0

0 0

0 0

0 0

0 0

0 0

22

0 To

tal

94

100

125

100

11

100

47

100

135

100

65

100

74

100

89

100

251

100

89

100

App

endi

x 5.

A W

ilcox

on S

igne

d R

anks

Tes

t co

mpa

rison

of

the

seas

onal

diff

eren

ces

in p

hysi

coch

emic

al p

aram

eter

s of

the

wet

land

s in

K

umas

i

188

N

Mea

n

Ran

k Su

m o

f R

anks

Te

st S

tatis

tics

Z A

sym

p. S

ig. (

2-ta

iled)

ra

iny_

pH -

dry

_pH

N

egat

ive

Rank

s 10

a 5.

50

55.0

0 -2

.803

0.

005

Posi

tive

Rank

s 0b

.00

.00

Ties

0c

Tota

l 10

ra

iny_

Tem

p - d

ry_T

emp

Neg

ativ

e Ra

nks

7d 5.

86

41.0

0 -1

.376

Base

d on

po

sitiv

e ra

nks

0.16

9 Po

sitiv

e Ra

nks

3e 4.

67

14.0

0 Ti

es

0f

To

tal

10

Rai

ny_S

al -

dry

_Sal

N

egat

ive

Rank

s 5g

3.30

16

.50

-.211

Base

d on

ne

gativ

e ra

nks

0.83

3 Po

sitiv

e Ra

nks

3h 6.

50

19.5

0 Ti

es

2i

To

tal

10

Rai

ny_D

O -

dry

_DO

N

egat

ive

Rank

s 1j

2.00

2.

00

-2.5

99

Base

d on

ne

gativ

e ra

nks

0.00

9 Po

sitiv

e Ra

nks

9k 5.

89

53.0

0 Ti

es

0l

To

tal

10

rain

y_TD

S - d

ry_T

DS

Neg

ativ

e Ra

nks

5m

5.40

27

.00

-.051

Base

d on

ne

gativ

e ra

nks

0.95

9 Po

sitiv

e Ra

nks

5n 5.

60

28.0

0 Ti

es

0o

To

tal

10

189

App

endi

x 6.

Met

al c

once

ntra

tion

(mg/

kg)

in s

edim

ents

of t

he d

iffer

ent

stud

y si

tes

in K

umas

i

stud

y si

tes

Ato

nsu

Sawm

ill

Hig

h Sc

hool

Ju

nctio n

FRN

R

farm

s

KN

UST

Po

lice

Stat

ion

Fam

ily

Cha

pel

Spet

inpo m

Gol

den

Tulip

H

otel

Kaa

se

Gui

nnes s

Bek

wai

roun

dabo

u t

Dak

wadw

o m

rain

y se

ason

17

/06/

2010

Ars

enic

138.

6 11

7.5

22.5

17

5.8

44.0

37

.2

54.7

25

.2

45.0

24

.5

Cop

per

652.

5 51

3.2

152.

1 72

.6

621.

3 18

4.2

584.

5 24

5.2

445.

8 12

5.2

Lead

42

5.3

189.

2 35

.8

45.3

45

2.1

49.5

65

.2

75.3

12

2.0

87.5

Zi

nc

842.

6 75

2.5

154.

2 15

2.2

825.

2 15

2.0

242.

2 23

9.0

234.

2 54

1.2

Chr

omiu

m

542.

2 45

2.2

142.

3 14

5.3

1078

.3

142.

8 22

1.5

236.

5 18

9.5

425.

3 C

adm

ium

0.9

0.6

1.0

0.6

1.0

0.6

0.5

0.6

0.6

0.8

Nick

el 44

3.8

479.

2 14

5.2

109.

2 54

2.2

105.

2 17

9.2

102.

7 10

7.2

102.

2 M

ercu

ry

2.1

2.5

0.9

0.7

0.6

0.8

0.8

0.8

0.8

0.8

Phos

phor

us

7785

.2

3022

.2

2230

.5

1889

7.5

1520

2.5

1745

.2

4421

.9

5232

.2

4523

.2

2202

.0

dry

seas

on

14/0

1/20

10

Ars

enic

48.6

18

5.7

25.1

85

.2

200.

1 47

.2

295.

4 27

7.2

189.

1 31

.3

Cop

per

652.

5 62

4.6

214.

2 12

7.4

1052

.1

984.

2 49

7.2

589.

2 38

9.5

323.

2 Le

ad

425.

3 48

9.2

38.5

47

.6

612.

9 84

.9

86.2

40

5.4

342.

4 33

2.1

Zinc

15

42.6

12

52.5

17

7.2

242.

2 10

41.2

42

7.5

327.

6 55

7.2

534.

2 58

2.1

Chr

omiu

m

542.

2 75

2.2

152.

8 32

4.2

532.

9 15

42.9

32

4.9

723.

1 28

7.5

325.

1 C

adm

ium

0.8

1.0

1.3

0.6

0.8

1.0

1.0

0.8

0.9

1.3

Nick

el 84

3.8

779.

2 10

7.6

104.

9 64

2.2

112.

7 12

4.2

148.

2 18

5.2

134.

2 M

ercu

ry

2.5

3.7

0.8

1.0

2.1

0.8

0.8

1.0

0.8

0.8

Phos

phor

us

1778

5.2

1902

2.2

1185

2.3

2252

1.2

1612

5.2

1252

1.3

1689

5.0

1121

4.4

1252

3.2

1754

2.4

190

App

endi

x 7.

Pea

rson

cor

rela

tion

mat

rix o

f he

avy

met

als

and

othe

r m

easu

red

envi

ronm

enta

l var

iabl

es (

sign

ifica

nt r

elat

ions

hips

sho

wn in

bo

ld)

Ars

eni

c C

oppe

r Le

ad

Zinc

C

hrom

ium

C

adm

ium

N

icke

l M

ercu

ry

Phos

phor

us

pH

Tem

pera

ture

Sa

linity

D

isso

lve

d O

2 D

isso

lve

d so

lids

sa

nd

silt

cl

ay

Ars

enic

Cop

per

0.28

7

Lead

0.

358

0.53

8

Zinc

0.

127

0.59

2 0.

852

Chr

omiu

m

0.05

3 0.

743

0.18

8 0.

330

Cad

miu

m

-0.3

98

-0.0

03

-0.1

16

-0.1

27

-0.1

42

Nic

kel

0.12

1 0.

510

0.73

6 0.

939

0.19

6 -0

.155

Mer

cury

0.

141

0.42

6 0.

654

0.86

7 0.

257

-0.1

64

0.95

2

Phos

phor

us

-0.0

66

-0.2

54

0.02

7 0.

255

-0.1

65

-0.3

32

0.33

5 0.

416

pH

0.16

7 0.

243

0.56

7 0.

691

0.33

0 0.

032

0.64

7 0.

775

0.16

5

Tem

pera

ture

0.

438

0.06

8 0.

406

0.20

4 -0

.041

-0

.305

0.

043

-0.0

88

-0.3

05

0.09

0

Sali

nity

0.

441

0.20

3 0.

533

0.29

2 0.

015

-0.3

09

0.05

7 -0

.101

-0

.201

0.

042

0.63

2

Dis

solv

ed

O2

-0.2

22

0.35

2 -0

.092

-0

.067

0.

148

0.38

1 0.

108

0.11

3 -0

.259

-0

.207

-0

.401

-0

.666

Dis

solv

ed

soli

ds

0.43

4 0.

202

0.51

7 0.

279

0.03

2 -0

.325

0.

042

-0.1

15

-0.2

03

0.03

2 0.

628

0.99

9 -0

.672

sand

0.

733

0.15

5 0.

382

0.09

9 -0

.147

0.

179

0.15

6 0.

238

-0.2

23

0.36

2 0.

187

0.07

8 0.

136

0.05

8

silt

-0

.401

-0

.427

-0

.465

-0

.460

-0

.536

-0

.129

-0

.349

-0

.502

-0

.082

-0

.774

-0

.136

-0

.009

0.

046

-0.0

01

-0.5

28

clay

-0

.415

0.

198

0.01

7 0.

270

0.60

7 -0

.131

0.

064

0.07

5 0.

304

0.18

6 0.

015

0.07

3 -0

.289

0.

090

-0.6

77

-0.2

46

191

App

endi

x 8.

Prin

cipa

l C

ompo

nent

Ana

lysi

s of

env

ironm

enta

l va

riabl

es a

nd h

eavy

met

als

asse

ssed

in th

e we

tland

s of

Kum

asi

Eige

nval

ues

PC E

igenv

alue

s %

Var

iatio

n C

um. %

Var

iatio

n 1

5

.52

32.4

32.

4 2

3

.48

20.5

52.

9 3

2

.52

14.8

67.

7 4

1

.93

11.3

79.

1 5

1

.3

7.7

86.

8 Ei

genv

ecto

rs

(Coe

ffici

ents

in

the

linea

r co

mbi

natio

ns o

f var

iables

mak

ing u

p PC

's)

Var

iable

PC

1

PC

2

PC

3

PC

4

PC

5 A

rsen

ic 0.

197

-0

.257

0.31

3

0.05

9

-0.2

14

Cop

per

0.27

3

0.09

8

0.04

8

-0.4

42

0.21

7 Le

ad

0.37

8

-0.0

65

0.

062

0.

028

0.

239

Zinc

0.

387

0.

105

-0

.113

0.06

6

0.23

8 C

hrom

ium

0.

188

0.

143

-0

.178

-0.5

10

-0.2

09

Cad

mium

-0

.100

0.19

7

0.23

6

-0.1

91

-0.0

09

Nick

el 0.

344

0.

194

-0

.031

0.19

1

0.30

9 M

ercu

ry

0.33

7

0.26

1

-0.0

07

0.

231

0.

099

Phos

phor

us

0.05

6

0.15

3

-0.2

89

0.

498

-0

.146

pH

0.

326

0.

151

0.

017

0.

116

-0

.318

Te

mpe

ratu

re

0.15

2

-0.3

73

0.

026

-0

.066

0.

007

Salin

ity

0.18

2

-0.4

50

-0

.093

-0.0

73

0.12

2 D

issol

ved

O2

-0.0

76

0.

355

0.

282

-0

.231

0.

324

Diss

olve

d so

lids

0.17

7

-0.4

52

-0

.105

-0.0

82

0.11

5 sa

nd

0.15

6

-0.0

54

0

.553

0.09

4

-0.2

10

silt

-0.2

90

-0

.124

-0.1

20

0.

122

0.

543

clay

0.07

8

0.08

3

-0.5

39

-0

.233

-0

.224

App

endi

x 9.

Com

paris

on o

f m

eans

of

heav

y m

etal

enr

ichm

ent

fact

ors

for

rain

y an

d dr

y se

ason

in K

umas

i N

orm

ality

Tes

t: Fa

iled

(P <

0.0

50)

Test

exec

utio

n en

ded

by u

ser r

eque

st, R

ank

Sum

Tes

t beg

un

Man

n-W

hitn

ey

Ran

k Su

m T

est

Gro

up

N

Mis

sing

M

edia

n

25

%

7

5 %

EF

rain

y se

ason

90

0

3.24

1 1.

968

6.74

2

EF d

ry s

easo

n 90

0

5.99

4 2.

842

16.0

63

M

ann-

Whi

tney

U S

tatis

tic=

2694

.000

T

= 67

89.0

00

n(s

mal

l)= 9

0 n

(big

)= 9

0 (P

= <

0.00

1)

The

diffe

renc

e in

the

med

ian

valu

es b

etw

een

the t

wo

grou

ps is

gre

ater

than

wou

ld

be e

xpec

ted

by ch

ance

; the

re is

a s

tatis

tical

ly s

igni

fican

t diff

eren

ce (P

= <

0.00

1)

193

App

endi

x 10

. C

ompa

rison

of

m

eans

of

he

avy

met

al

geoa

ccum

ulat

ion

indi

ces

for

rain

y an

d dr

y se

ason

in K

umas

i N

orm

ality

Tes

t: Fa

iled

(P <

0.0

50)

Test

exec

utio

n en

ded

by u

ser r

eque

st, R

ank

Sum

Tes

t beg

un

Man

n-W

hitn

ey

Ran

k Su

m T

est

Gro

up

N

Mis

sing

M

edia

n

25

%

75 %

geoa

ccum

ulat

ion

rain

y se

ason

90

0

1.05

0 0.

330

2.10

7

geoa

ccum

ulat

ion

dry

seas

on

90

0 1.

937

0.86

1 3.

359

M

ann-

Whi

tney

U S

tatis

tic=

2694

.000

T

= 67

89.0

00

n(s

mal

l)= 9

0 n

(big

)= 9

0 (P

= <

0.00

1)

The

diffe

renc

e in

the

med

ian

valu

es b

etw

een

the t

wo

grou

ps is

gre

ater

than

wou

ld

be e

xpec

ted

by ch

ance

; the

re is

a s

tatis

tical

ly s

igni

fican

t diff

eren

ce (P

= <

0.00

1)

194

App

endi

x 11

. C

ompa

rison

of

mea

ns o

f Po

llutio

n Lo

ad I

ndic

es

for

the

diffe

rent

site

s in

the

rain

y an

d dr

y se

ason

in K

umas

i N

orm

ality

Tes

t: Pa

ssed

(P

= 0

.051

) Eq

ual

Var

ianc

e Te

st:

Pass

ed

(P =

0.3

74)

Gro

up N

ame

N

M

issi

ng

Mea

n St

d D

ev

SEM

PLI r

ainy

s

10

0 4.

082

2.19

9 0.

695

PL

I dry

sea

10

0

6.84

1 3.

062

0.96

8

Diff

eren

ce

-2.7

59

t = -2

.314

with

18

degr

ees o

f fre

edom

. (P

= 0.

033)

95

per

cent

conf

iden

ce in

terv

al fo

r diff

eren

ce o

f mea

ns: -

5.26

4 to

-0.2

54

The

diffe

renc

e in

the

mea

n va

lues

of t

he tw

o gr

oups

is g

reat

er th

an w

ould

be

expe

cted

by

chan

ce; t

here

is a

sta

tistic

ally

sig

nific

ant d

iffer

ence

bet

wee

n th

e inp

ut

grou

ps (P

= 0

.033

). Po

wer

of p

erfo

rmed

tes

t with

alp

ha =

0.0

50:

0.49

8

i