UNIVERSITY OF GHANA DEPARTMENT OF GEOGRAPHY AND RESOURCE DEVELOPMENT FLOOD VULNERABILITY AND ADAPTATION IN ACCRA: EXAMINING ASPECTS OF BOTH CLIMATE AND NON-CLIMATE FACTORS ERIC KOFI AFORNORPE 10363823 THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF M.Phil GEOGRAPHY AND RESOURCE DEVELOPMENT DEGREE. MARCH, 2016 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I, Eric Kofi Afornorpe, do hereby declare that, except for references to other people’s work which have been duly cited in this research, the rest of the research is as a result of my own work under the supervision of Dr. Kwadwo Owusu and Mr. Sosthenes Kwadzo Kufogbe of the Department of Geography and Resource Development, University of Ghana, and that neither in whole or in part has this work been presented anywhere for the award of a degree. SIGNATURE: DATE: PRINCIPAL SUPERVISOR: DR. KWADWO OWUSU SIGNATURE:…………………………………. DATE:……………………………………….... CO-SUPERVISOR: MR. SOSTHENES K. KUFOGBE SIGNATURE:…………………………………. DATE: ………………………………………… University of Ghana http://ugspace.ug.edu.gh ii DEDICATION To God be the glory. I gladly and humbly dedicate this work to my family. . University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT “Anyone who has done what Napoleon could not do needs to be uplifted”. The first and foremost recognition and appreciation goes to my supervisors, Dr. Kwadwo Owusu and Mr. Sosthenes Kwadzo Kufogbe for their exceptional guidance. Similar appreciation goes to all lecturers in the Department of Geography and Resource Development in the University of Ghana, and in the Department of Geography and Rural Development (KNUST) for their rigorous training and knowledge they have imparted in me. Next are my sponsors, including PERIPERI-U and BSU-EC for their financial supports. Without them I would not have successfully finished my thesis. To them I say Ayekoo! A lot of appreciation also goes to my Research Assistants and Interpreter including Samuel Dumenya and Jacob Doku Tettey, the community focal persons and respondents including Addrah Pascal. I was able to effectively gather data through their contributions. May your contributions be rewarded. I am equally indebted to the staff of the NADMO (HQ) and communities where data was collected and particularly Ga-Mashie Cantonment, Nima, Agbogbloshie and Korley Dudor, Staffs from AMA and MLGRD, Ghana Health Service, Fire Service, the Hydrological Department, Ghana Meteorological Agency for providing me data. To all my numerous course mates at the Department, I enjoyed, and appreciate our togetherness, advice and all manner of support since we became students in the University of Ghana. University of Ghana http://ugspace.ug.edu.gh iv TABLE OF CONTENTS DECLARATION ..................................................................................................................... iv DEDICATION .......................................................................................................................... ii ACKNOWLEDGEMENT ...................................................................................................... iii TABLE OF CONTENTS ........................................................................................................ iv LIST OF TABLES ................................................................................................................. vii LIST OF FIGURES .............................................................................................................. viii LIST OF BOXES ......................................................................................................................x ABBREVIATIONS ................................................................................................................. xi ABSTRACT ........................................................................................................................... xiv CHAPTER ONE .......................................................................................................................1 GENERAL BACKGROUND OF THE STUDY ................................................................................ 1 1.1 Introduction to the General Background........................................................................................ 1 1.2 Problem Statement ......................................................................................................................... 4 1.3 Research Questions ........................................................................................................................ 6 1.4 Objective of the Study ................................................................................................................... 7 1.5 Propositions to the Study ............................................................................................................... 7 1.6 Justification/Significance of the Study .......................................................................................... 8 1.7 Limitation to the Study .................................................................................................................. 9 1.8 Organization of the Study ............................................................................................................ 10 1.9 Chapter Summary ........................................................................................................................ 11 CHAPTER TWO ....................................................................................................................12 LITERATURE REVIEW..........................................................................................................12 2.1 Introduction.................................................................................................................................. 12 2.2 Background of Flood Disasters.................................................................................................... 12 2.3 Extent of Flood Disaster .............................................................................................................. 14 2.4 Causes of Flood Disasters ............................................................................................................ 17 2.4.1 Non-Climatic Factors and Urban Flooding ..................................................................... 19 2.5 Flood Risks and Vulnerable Areas in Accra ................................................................................ 23 2.6. Adaptation to Flood through Vulnerability Reduction ............................................................... 25 2.6.1 Governments Initiatives to Flood Adaptation .................................................................. 30 2.7 Conceptual Framework of Vulnerability and Adaptation to Disaster .......................................... 34 University of Ghana http://ugspace.ug.edu.gh v 2.8 Chapter Summary ........................................................................................................................ 38 CHAPTER THREE ................................................................................................................40 STUDY AREA AND STUDY DESIGN ........................................................................................... 40 3.1 Introduction.................................................................................................................................. 40 3.2 Study Area Profile ....................................................................................................................... 40 3.2.1 History and Location of Accra......................................................................................... 40 3.2.2 Geology ........................................................................................................................... 42 3.2.3 Climate and Vegetation ................................................................................................... 42 3.2.4 Population and Migration ................................................................................................ 45 3.2.5 Housing ............................................................................................................................ 46 3.2.6 Poverty ............................................................................................................................. 48 3.2.7 Slums, Illegal Settlements and Unauthorized Structures and Dwellers ........................... 49 3.2.8 Disaster Zones ................................................................................................................. 51 3.2.9 Waste Management ......................................................................................................... 52 3.3 Study Design ................................................................................................................................ 53 3.3.1 Types and Sources of Data .............................................................................................. 54 3.3.2 Data Gathering Methods .................................................................................................. 55 3.3.3 Sampling Design .............................................................................................................. 56 3.3.4 Data Analysis ................................................................................................................... 58 3.3.5 GIS Overlay Operation on Flood Zones .......................................................................... 59 3.4 Chapter Summary ............................................................................................................... 61 CHAPTER FOUR ...................................................................................................................62 ANALYSIS OF CLIMATE AND NON-CLIMATE FACTORS TO FLOOD DISASTER ..62 4.1 Introduction.................................................................................................................................. 62 4.2 Background of the Respondents. ................................................................................................. 62 4.3 Knowledge and Perceptions on Disaster ..................................................................................... 63 4.4 Climate Change and Disasters ..................................................................................................... 65 4.5 Causes of Flood Disaster in Accra ............................................................................................... 66 4.6 Changes in Rainfall and other Climatic Conditions .................................................................... 75 4.7 Chapter Summary ........................................................................................................................ 82 CHAPTER FIVE .....................................................................................................................84 SPATIAL DIFFERENCES AND ADAPTATION TO FLOOD DISASTER ..........................84 University of Ghana http://ugspace.ug.edu.gh vi 5.1 Introduction.................................................................................................................................. 84 5.2 Discriminant Analysis of Spatial Variation on Flood Disaster .................................................... 84 5.3 Experience of Flood Disaster in Accra ........................................................................................ 94 5.4. Adaptation Strategies to Flood Disasters. ................................................................................... 96 5.5 Chapter Summary ...................................................................................................................... 100 CHAPTER SIX .....................................................................................................................102 DISCUSSION OF MAJOR FINDINGS .................................................................................102 6.1 Introduction................................................................................................................................ 102 6.2 Flooding in Accra ...................................................................................................................... 102 6.3. Causes of Flood Disasters ......................................................................................................... 103 6.4 Spatial Variation and Severity to Flood Disasters ..................................................................... 105 6.5 Respondents’ Recommended Adaptation Strategies ................................................................. 107 6.5 Chapter Summary ...................................................................................................................... 109 CHAPTER SEVEN ...............................................................................................................111 SUMMARY OF KEY FINDINGS, CONCLUSION AND RECOMMENDATIONS ................... 111 7.0 Introduction................................................................................................................................ 111 7.1 Summary of Key Findings ......................................................................................................... 111 7.2 Conclusion ................................................................................................................................ 114 7.3 Recommendations .................................................................................................................... 115 REFERENCES ......................................................................................................................118 APPENDICES .......................................................................................................................128 APPENDIX A: QUESTIONNAIRES ..................................................................................128 APPENDIX B: AVERAGE MONTHLY AND ANNUAL RAINFALL AND TEMPERATURE DATA FROM 1961-2000 FROM ACCRA AIRPORT STATION ...133 APPENDIX D1: MAP OF SLUM AREAS AND FLOOD DISASTER ...........................136 APPENDIX D2: MAP OF ACCRA METROPOLITAN AREA ......................................137 APPENDIX D3: A MAP OF SLUM AREAS IN THE METROPOLIS ..........................138 APPENDIX D4: A MAP OF COMBINED FLOOD DISTRIBUTION IN ACCRA ......139 APPENDIX D5: A MAP OF COMBINED FLOOD AND SLUMS .................................140 University of Ghana http://ugspace.ug.edu.gh vii LIST OF TABLE Table 2.1 Vulnerability Model ...............................................................................................26 Table 4.1: Background of Respondents ................................................................................63 Table 4.2: Respondents’ Knowledge on Disasters ................................................................64 Table 4.3: Respondents’ Observation of Disasters ..............................................................65 Table 4.4: Disasters with Strongest Relation to Climate Conditions .................................66 Table 4.5 Changes and Variability Observed in the Climate by Respondents .................76 Table 4.6: Observed Changes and Variability on Climate Conditions .............................77 Table 5.2: Knowledge and Perception of Flood Disaster Variations .................................89 Table 5.3 Group Statistics (means and standard deviations) ..............................................90 Table 5.5: Pooled Within-Groups Matrices ..........................................................................91 Table 5.6 Box's M and Log Determinants ............................................................................92 Table 5.7: Summary of Canonical Discriminant Functions (Eigenvalues) ........................93 Table 5.8: Standardized Canonical Discriminant Function Coefficients ..........................94 Table 5.9: People Affected by Flood Disaster in Accra .......................................................95 Table 5.10 Where Respondents Experience Flood Disaster ................................................95 Table 5.11: Response on Efforts to Cope with Flood Disaster ............................................96 Table 5.12: Adaptation Efforts towards Flood Disasters in Accra .....................................97 Table 5.13 Recommended Adaptations from FGD (at Ga Mashie) ...................................99 Table 5.14 Priority Investment ............................................................................................100 University of Ghana http://ugspace.ug.edu.gh viii LIST OF FIGURES Figure 2.2: Adaptation Focus Model .....................................................................................27 Figure 2.3 Conceptual Framework on Flood Disaster, Causes and Adaptation ...............35 Figure 3.1: Map of Accra Metropolitan Area ......................................................................41 Figure 3.7 GIS Overlay Operations of Flood and Communities ........................................60 Figure 4.1: Reasons for Flood Occurrences in Accra ..........................................................67 Figure 4.3: Average Monthly Rainfall over Four Decades in Accra ..................................79 Figure 4.4: Average Annual Rainfall from 1961-2000 in Accra .........................................80 Figure 4.5: Average Monthly Temperature over Four Decades in Accra .........................81 Figure 4.6: Average Annual Temperature from 1960-2000 in Accra ................................82 Figure 5.1: A Map of Slum Areas in the Metropolis ............................................................85 Table 5.1: Listing of Flood Disaster Zones (communities) over Accra ..............................86 Figure 5.2: A Map of combined Flood Distribution in Accra .............................................87 Figure 5.3: A Map of combined Flood and Slums. ..............................................................88 University of Ghana http://ugspace.ug.edu.gh ix LIST OF PLATES Plate 4.1: Deposition of Wastes in Drains at Ga-Mashie .....................................................69 Plate 4.2 Building on Waterways and Chocked Gutters Agbogbloshie .............................70 Plate 4.3 Chocked Gutters in Old Fadama (A and B) .........................................................71 Plate 4.4: Narrow and silted Drain at James Town .............................................................72 Plate 4.5: Waste Collected and left at the Bank of Drain in Ursher Town ........................73 Plate 4.6: Dredging Work on Choked Odaw River .............................................................73 Table 4.5 Changes and Variability Observed in the Climate by Respondents .................76 Table 4.6: Observed Changes and Variability on Climate Conditions .............................77 University of Ghana http://ugspace.ug.edu.gh x LIST OF BOXES Box 4.1: Resident’s Story ........................................................................................................68 Box 4.2: Public Relation Officer’s Story ...............................................................................74 University of Ghana http://ugspace.ug.edu.gh xi ABBREVIATIONS AAP Africa Adaptation Programme AMA Accra Metropolitan Assembly BSU-EC Building Stronger Universities Platform on Environment & Climate CARE Cooperative for Assistance and Relief CDA Classification Discriminant Analysis CRED Centre for Research on the Epidemiology of Disasters DRR Disaster Risk Reduction EM-Dat Emergency Events Database EPA Environment Protection Agency FGD Focus Group Discussion GAMA Greater Accra Metropolitan Assembly GAMADA Ga-Mashie Development Agency GCM Global Circulation Model GHGs Green House Gases GIS Geographic Information System GMet Ghana Meteorological Agency GoG Government of Ghana GPHC Ghana Population and Housing Census GSGDA Ghana Shared Growth Development Agenda GSS Ghana Statistical Service IFRCRCS International Federation of Red Cross and Red Crescent Societies IFRI International Forensic Research Institute IIED International Institute of Environment and Development University of Ghana http://ugspace.ug.edu.gh xii IIPAC Innovative Insurance Products for Adaptation to Climate Change ILGS Institute for Local Government Studies IPCC Intergovernmental Panel on Climate Change KNUST Kwame Nkrumah University of Science and Technology LDCs Less Developing Countries MEST Ministry of Environment Science and Technology MESTI Ministry of Environment Science Technology and Innovation MLGRD Ministry of Local Government and Rural Development MMDAs Metropolitan, Municipal and District Assemblies MWRWH Ministry of Water Resource Works and Housing NADMO National Disaster Management Organization NCAP Netherlands Climate Assistance Programme NCCAS National Climate Change Adaptation Strategy NCCASP Netherland Climate Change Adaptation Study Programme NGOs Non-Governmental Organizations NOAA National Oceanic and Atmospheric Administration OCHA Office of for Coordination of Humanitarian Affairs OFDA Office of Foreign Disaster Assistance OST Office of Science and Technology PERIPERI-U Partners Enhancing Resilience for People Exposed to Risk-University SPSS Statistical Package of Social Science UN United Nations UNCT United Nations Country Team UNDP United Nation Development Programme UNEP United Nation Environment Programme University of Ghana http://ugspace.ug.edu.gh xiii UNESCO United Nation Education, Scientific and Culture Organization UNFCCC United Nation Framework Convention on Climate Change UN-HABITAT United Nations Human Settlements Programme UNISDR United Nation International Strategy for Disaster Reduction US United States of America WFP World Food Programme WMO World Meteorological Organization University of Ghana http://ugspace.ug.edu.gh xiv ABSTRACT Flood is most predominant natural disaster in the world, affecting most cities in Africa especially during rainy seasons. Only an average of 100 floods per decade used to occur. However, since middle of 20th century, flood cases have increased to more than 2,800 per decade. In Sub-Saharan African countries, efforts at managing flood disasters are mainly reactive approaches such as provision of relief to affected people. Reasons are that there is low level of technology for adaptation and there is still high perception that flood is natural and inevitable disaster. This study sought to analyse flood disaster as caused by both natural factors (climate) and human induced factors, including poor drainage systems and waste management making people more vulnerable to floods. The study also examined flood distribution, severity, and the adaptations to such flood disaster. A mixed method research strategy was adopted for data gathering and analysis. Reponses were gathered from 440 respondents using questionnaires, focus group discussions (FGD) and in-depth interviews from household heads and institutional heads. Questionnaires were analysed using SPSS v.20 and results shown in tables, charts and maps. Classification Discriminant Analysis was used to classify flood situation in selected communities. Geographical Information System (GIS) –ArcGIS v.10 was used to map spatial distribution of floods. Flood disaster is a serious challenge in Ghana where it emerges as a leading natural disaster causing damage to lives and properties. Both climatic and human factors were established as contributors to flood disaster in the Metropolis. Nonetheless, the human factors act as catalysts to exacerbate flood occurrence and the extent of damage. Residents’ vulnerability to floods have been linked to human factors including poor housing, inadequate and chocked drainage University of Ghana http://ugspace.ug.edu.gh xv systems, and poor waste management practices among others. Rainfall was identified to be erratic and is becoming more intensive with high run-off. The study also established that flood occurrence and effects in the Metropolis were not evenly distributed. Slum communities and other low residential areas experienced more floods. Efforts by institutions, civil society organization and local communities do not focus on reducing people’s vulnerability to the floods. There is the need for making the city more resilient through efficient planning and building structures such as houses, drainage systems, awareness creation on flood risks, proper waste management systems. Key words: Floods, Climate Change, Vulnerability, Adaptation. University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE GENERAL BACKGROUND OF THE STUDY 1.1 Introduction to the General Background Globally, climate change related disasters have increasingly caused destruction to people’s livelihoods, lives and properties. Munich Re (2002) noted that natural hazards including floods have had extensive negative effects on people’s lives, economic and social infrastructures over the last four decades. Over the past 10 years, the world has suffered an increasing number of natural disasters resulting from changes in climate and non-climate conditions. These disasters include floods that affect more than 2.5 billion people, killing about 500,000 and an estimated economic loss of US$700 billion (Foresight OST, 2013). Askew (1999) stated that floods constitute about one-third of all deaths, one-third of all injuries and one-third of all damages from natural disasters. UN-HABITAT (2010) noted floods to be responsible for half of the world’s disasters constituting 84% of all disaster related deaths. Floods also trigger other problems including spread of diseases such as cholera, diarrhoea, typhoid, scabies, malaria and contaminating water supply sources mostly in urban poor settlements (UN-HABITAT, 2010). This situation has increased global concerns over how to reduce or mitigate people’s vulnerability in exposure to environmental risks especially in cities. Urban areas are seen as crucibles of hazards because of the wide range of risks urban dwellers continue to face (Bull-Kamanga et al, 2003). Although flood disasters are noted to be natural phenomenon, ActionAid (2006) noted that the damages suffered are the consequences of human actions. Bull-Kamanga et al (2003) further stated that the University of Ghana http://ugspace.ug.edu.gh 2 vulnerability of urban populations to disasters is not only natural but rather because of the failed social, economic and political systems, that governs it. The socio-economic structural processes that accelerate the processes of urbanization, population movement and population concentrations increase the vulnerability of urban dwellers (UNEP, 2009). IPCC (2002) provided a definition of vulnerability in relation to climate change as “the extent to which a natural or social system is susceptible to sustain damage from climate change”. This emphasises the significant role of social systems in combating flood disasters in cities. The poor in cities are noted to be the most vulnerable and at risk to natural hazards because of their low resilience (UN, 2009). The increasing urban populations, increasing built up surfaces, poverty and poor city infrastructure planning, exacerbate vulnerability of cities (Bull-Kamanga et al, 2003: Songsore et al., 2009). Adedeji et al. (2012) cited poor infrastructure planning in cities as a restriction to free flow of flood and storm waters, a situation likely to increase flood and its damages. With the rise in world urban populations, the number of urban dwellers at risk to flooding will consequently increase. Global urban population has increased by a factor of five from 0.7 billion in 1950 to 3.9 billion in 2014 (UN, 2014). Further, projections indicate that global urban populations will further increase by 60% by 2050, translating into urban populations of 6.3 million. Adedeji et al (2012) expressed the view that the frequency and intensity of disasters witnessed globally is because of unpredictable climatic conditions manifesting in severe floods, fire, drought, terrorism epidemics and urbanization especially in developing countries. Several cities are becoming hot spots for floods and other disasters, especially in Sub- Saharan Africa, which has also recorded high reported rates of urbanization, urban poverty, vulnerability and extreme rainfall regime over the last few decades (OFDA-CRED, 2010). University of Ghana http://ugspace.ug.edu.gh 3 The period 1997-2008 alone witnessed over 166 urban disasters in 28 African cities (EM- Dat, 2008; cited in Songsore et al, 2011). From 1900 to 2006, floods in cities of Africa killed nearly 20,000 people and affected approximately 40 million people. Total cost of damage was estimated at about US$4 billion (Mulungeta et al., 2007). Floods are among the most devastating natural disasters in Africa, with flash floods among the greatest hazards arising from tropical cyclones and severe storms. Floods contribute extensively to loss of life, damage of property, and promote spread of diseases such as malaria, dengue fever and cholera over coastal communities in the urban Africa (Songsore et al, 2011). According to UNDP (2007), 72% of African urban populations live under slum and other vulnerable conditions predisposing them to flood disasters. The increasing destruction to properties and lives by floods and associated cost prompted the need to put the necessary adaptation strategies in place. Unfortunately, efforts at reducing vulnerabilities associated with flood disaster among slums and other vulnerable locations in over 100 communities in Africa by governments, development partners, non-governmental organizations (NGOs), private and civil society organizations have not yielded much success (Mulungeta et al., 2007). Ghana has suffered devastating floods over the years. There is no specific period of time when floods are most likely to occur, but what is known for sure is that, floods set in with the rains (Agyemang, 2013). The causes of flood disaster in Ghana is due to intense rainfall- run-off, dam burst and tidal waves. The latter occurs mostly along the coastal zones with the first two as a perennial situation occurring all over the country (Karley, 2009). University of Ghana http://ugspace.ug.edu.gh 4 Natural factors are known to trigger flooding in Ghana but human or anthropogenic factors in recent times have aggravated the situation further (Karley, 2009; Songsore et al., 2009; Osei, 2013). Ghana crossed the urban divide in 2010 with more than 50 percent of Ghanaians living in cities, more than 44 percent in the preceding ten years, with the remainder residing in rural communities. The country of over 25 million has rapidly urbanized without adequate infrastructure development. That is to say that the rate of infrastructural development has not been commensurate with the population growth. The situation has led to many towns and cities witnessing uncontrolled developments in the form of slums and illegal settlements with vulnerable conditions that predisposes settlers to flood and other forms of environmental risks. 1.2 Problem Statement Globally, population in cities continues to grow at high rates due in part to the relentless migration of rural populations into urban areas (Price & Vojinovic, 2015). The high populations in urban centres (30% of low and middle-income dwellers), are said to live in informal settlements in overcrowded and deteriorating conditions in Asia and Africa (Dodman et al., 2012). Satterthwaite (2011) found out that such cities lack good infrastructures including poor drainage, water storage, sanitation, roads, healthcare and emergency services system. These conditions increase the vulnerability of such populations to different forms of risks mostly flooding and landslides in Asia and Africa (Hardoy et al, 2001). Ghana like many other developing countries continue to suffer devastating effects of climate change related disasters. Prominent among these disasters is the perennial floods occurring in every rainy seasons of a year. Accra the capital city of Ghana has recorded devastating University of Ghana http://ugspace.ug.edu.gh 5 floods over the years. Factors accounting for the perennial floods in Accra remain both natural and anthropogenic (Songsore et al., 2009; Karley, 2009). Aside the natural factors that contribute to urban floods, the National Co-ordinator of NADMO, cited human factors that exacerbate flood disasters in Accra. He stated “Poor planning resulting from lack of co-ordination, illegal structures as well as undersized hydraulic structures, designing problems, rapid changes in land use patterns due to urbanization, development of settlement in water courses and flood plains were some other causes of floods” (cited in UNEP/OCHA, 2011). This comment was made after October 26th 2011 flooding in the city, which affected over 43,000 people with 17,000 losing their place of residence and 14 losing their lives. There were also large infrastructural damage to roads, waterways and bridges. In addition, over 100 cases of cholera were recorded after the floods (UNEP/OCHA, 2011). Further, NADMO noted that the government of Ghana has spent huge sums of money to the tune of 700 million dollars to restore, rebuild and resettle the flood affected populations between the periods 2007and 2011 (Agyemang, 2013). Agyemang (2013) cited the National Board for Small Scale Business (2011) that the devastations caused by the floods are estimated to cost private and individual firms some 3 billion dollars in Ghana. Historically, Accra has witnessed severe floods and continues to face increasing risks of flooding (Karley, 2009). The incident of flooding in the capital goes far back to the 1930’s caused by the flooding of the Odaw Basin, the creation of the artificial lake on the Volta River and the rains (Karley, 2009). Daily Graphic (2005) also chronicled flood disasters and the severe impacts on the Accra population since 1968 to 2014. Flood disasters over the last decade have shown extensive destruction to life and properties in the capital including the University of Ghana http://ugspace.ug.edu.gh 6 November 1st, 2011 floods in Accra, which destroyed over 43,000 homes, and killing almost 14 people. In a recent report by the UNCT (2015), the June 3rd Flood Disaster brought to a halt all economic activities in the capital city, destroying vehicles, buildings, schools and other infrastructure. During the flood, there was an explosion of a fuel filling station that killed over 150 people and many others left with various degrees of injuries. On his visits to affected flood areas the President of the Republic of Ghana announced a release of 50 million Ghana cedis, partly for relief to victims and partly for clearing of drainage and waterways. The flood disaster has been attributed to the heavy downpour, poor sanitation and poor city planning and enforcement among other human factors. Agyemang (2013) noted that flood forecasting is non-existent in Ghana since monitoring is apparently not available, despite warnings on rainstorms. With authorities facing huge challenges in developing new schemes to plan the city, there is the need to re-examine the human and climatic factors that together cause flooding in Accra. To fill the gap in addressing flood disasters in Accra, additional research is needed to unravel the combined role of natural and human factors that contribute to the vulnerability of urban populations to flood disaster. As such, the current study sought to examine aspects of climatic and non-climatic conditions that increase the vulnerability of city populations to flood disaster. 1.3 Research Questions Subsequent to the above problem, the study posed various questions with focus on people’s knowledge about flood disasters and their distribution, the extent of changes in the climate and non-climate factors. In addition, people’s perception and knowledge on various University of Ghana http://ugspace.ug.edu.gh 7 adaptation strategies adopted were assessed. The study therefore sought to answer the following specific questions; i. What is the nature of flood disaster in the Accra Metropolis? ii. What climatic conditions have a direct bearing on flooding in Accra? iii. What human conditions predispose populations to flood disaster? iv. What is the distribution of flood disasters in Accra? v. What adaptation strategies are adopted to reduce the vulnerability to floods? 1.4 Objective of the Study The general objective of the study was to ascertain climate change and human factors that causes flood disaster, the distribution and the adaptation strategies adopted to reduce people’s vulnerability to flood disaster. To understand the general objective, the following specific objectives were set: i. To assess the nature of flood disaster in Accra. ii. To examine changes in climate conditions such as temperature and rainfall and their direct bearing on flooding in Accra. iii. To investigate the human conditions predisposing populations to severe flood disasters. iv. To examine the level of distribution of flood disaster in Accra v. To assess people’s experience of flood and their adaptation strategies 1.5 Propositions to the Study a) Flood disaster is a predominant disaster in communities in Accra b) There are changes in the climate that have triggered flood disaster in Accra c) Communities with poor drainage, poor housing and poor sanitation experience more flood disaster in Accra d) Adaptation to flood disaster is more of a reactive approach University of Ghana http://ugspace.ug.edu.gh 8 1.6 Justification/Significance of the Study The relevance of research on flood resulting from both changes in climate and human vulnerability is a concern to everybody, individuals, firms, governments, non-governmental organizations and civil society, due to its developmental and environmental repercussions. Flood disaster is among the leading disaster, increasingly destroying homes, properties, infrastructure and lives. Flood disaster also brings many businesses to a standstill. Ghana has a long history of flooding, especially in its major capital Accra. Accra has suffered many devastating flood disasters attributed in recent times to climate change and anthropogenic factors. Many researches have concentrated on climatic factors or anthropogenic causes to floods in Accra. Despite the number of research studies, the situation remains a seasonal affair. Many people have continually blamed central governments and city authorities for not providing the necessary mechanisms for reducing floods in the city. Every year about 85% of NADMO’s resources and efforts are used for addressing issues related to hydro-meteorological disasters in Accra and other parts of the country (MEST, 2012) even though several efforts have been made, from research to policy, to arrest the perennial floods situation. Increasing vulnerability of city dwellers to flood disaster risks has not been properly explored. Large amounts of water flow from rivers upstream enter the city. In addition, there is clear indication of climate change affecting the rainfall regime of Ghana and Accra making it more extreme, more intensive, erratic and unpredictable. Sea level may increase posing a higher flood risk to coastal Accra. The situation in part is exacerbated by the increasing changes in the land use forms in Accra. Such changes include the construction of infrastructure in waterways, the conversion of bare ground to concrete and paved surfaces, inadequate and choked drains. River receptacles among others need better environmentally friendly designs. University of Ghana http://ugspace.ug.edu.gh 9 This study is an attempt by the researcher to unfold some of the findings necessary to address the problems posed by flood disaster as a developmental challenge in Accra. At a policy level, this study will serve as baseline information to the implementation of the national climate policies and strategies on climate change effects. The study is also a sequel to some priority areas and activities provided in the National Climate Change Adaptation Strategy under implementation from 2010-2020, specifically to its priority area four (4): “Adapting to climate change through enhanced research” (MEST, 2011). The study is also to review the main contributions made by researchers on climate, non- climate conditions and flood disaster and linkages between them. This is needed relate the study to concepts and theories developed by other authors in this subject area. This survey would help build on the strengths and the weaknesses of approaches used by other scholars and their recommendations in order to make additional or new recommendations for areas that require further research and implementation. Above all, findings and recommendations arising from this study will be used to deal with similar problems elsewhere in other communities. 1.7 Limitation to the Study The study was confronted with a few challenges arising from data collection, transport, personnel, respondents’ behaviour, language barrier to mention but few. Due to the nature of work involved in data collection, mainly focus group discussions, there was a need for research assistants and an interpreter. The Research Assistants assisted the researcher in reaching out to the respondents, organising FGD meetings, questionnaires and gathering their views. Research Assistants had to be trained for the fieldwork and monitored through the data collection process to attain success on the process and obtain valid data. University of Ghana http://ugspace.ug.edu.gh 10 In addition, the researcher was faced with difficulties in using the local language. To cope with the situation, an Interpreter was used to bridge the language barrier gap. In most cases, the Interpreter had to summarise the views expressed by the respondents. Understanding of terminologies such as climate change, vulnerability and adaptation by some of the respondents demanded further explanation. In most cases, these terms had to be explained through the use of examples to ensure the understanding of respondents. There was also conflict of interest among respondents in identifying the head of the household. On some occasions, the researcher had to visit the same household twice or find alternative household due to disagreement on the head of the household or readiness to respond. This situation did not only incur frustration in the research team but also posed additional cost and time in terms of transport. A lot of data have not been classified into rural urban and rural and community by community in Accra. The challenges were surmounted by successful findings and recommendations. 1.8 Organization of the Study The study has been organised into seven (7) chapters with a summary after every chapter. The first chapter includes the introduction, problem statement, objectives of the study, proposition, significance of the study, and the limitations to the study. Chapter 2 focuses on the literature review on flood disaster, climate and non-climate causes, adaptation to floods and conceptual framework. Chapter three (3) examines the study area profile and study design. Chapter four (4) is devoted to the results and analysis of the data collected including flood disaster and causes. Chapter Five (5) focuses on the mapping and statistical analysis of variations and severity of flood disaster in Accra; and the level of effects and adaptation strategies of the flood disaster. Chapter six (6) discusses major findings, and finally chapter University of Ghana http://ugspace.ug.edu.gh 11 seven (7) includes summary of the study, conclusion and recommendations, references and appendices. 1.9 Chapter Summary This chapter began with a background introduction to the study. This section provided a general background related to flood disaster at a global level, in Africa and Ghana. The background provided the research with a situational context of the problem investigated. The problem statement threw much light on climatic and non-climatic conditions in Accra that together increase people’s vulnerability to flood disaster. This was followed by a section that provided information on the aim and objectives of the study and the proposition that guided the study. The chapter further looked at the significance of the study with details justified the need for additional research to make recommendations to the current problem. Finally, the chapter examined the limitations of the study and details on the chapter organizational structure of the entire study. University of Ghana http://ugspace.ug.edu.gh 12 CHAPTER TWO LITERATURE REVIEW 2.1 Introduction The chapter reviews the existing literature and the contributions by other researchers on the issues, concepts and theories relating to flood disaster, climate change, human induced factors to flooding, and adaptation efforts targeted at mitigating flood disasters. The literature is organized into eight sections, which includes an introduction and outline of the sections of the chapter. The second section examines the background and concepts of flood disaster; section three examines the extent of flood disasters in Africa especially flood records and damage in cities such as Accra; section four is dedicated to flooding caused by climatic and human induced factors. The fifth and sixth sections are focused on flood risk, incidences of flood disaster and adaptation in Accra. Further, section seven dwells on the conceptual frameworks on how climate change and non-climate change conditions interact, their ramifications on flood disasters and possible adaptation strategies. 2.2 Background of Flood Disasters Floods are by far the most hazardous, disastrous, frequent and widespread disaster throughout the world (Dhār and Nandargi, 2001), causing extensive damages to lives and property especially in developing countries where there are still bottle necks in achieving sustainable resilience (Dilly et al, 2007). According to Mulungeta (2007), flood has been one of the most serious threats to city dwellers especially among coastal communities in Africa, where populations continue to rise and there are limited options for sustainable adaptation. University of Ghana http://ugspace.ug.edu.gh 13 Floods usually occur during an overflow of water that submerges land usually dried and not covered by vegetation, irrespective of a cause and effect (Britannica Encyclopaedia, 2013). Descriptions and categorizations of floods vary, based on a combination of a number of factors including the location, sources, causes and impacts (Jha et al., 2012). According to Jha et al (2012) such combinations enable us to characterize floods generally into river (or fluvial) floods, pluvial (or overland) floods, coastal floods, groundwater floods or failures in artificial water systems or dam spillages. Jha et al (2012) added that river floods occur when rivers overflow their banks due to high surges from heavy rainfall upstream that the existing river channel cannot contain, or are due to downstream blockage by soil, wastes or other water bodies. They described overland and groundwater floods as those occurring on the surface of the earth and underground. Bulkeley et al (2009) state that classification of floods as urban and rural is a challenge because definitions are not clear on distinction between urban and rural areas and the similarities in the type of floods that occur over the two geographical spaces. Bulkeley et al (2009) noted that urban floods should be distinguished from rural floods based on the extent of damage and geographical area covered. This is because urban floods occur in smaller geographical areas and such disasters records higher intensity of damage, whereas rural floods happen over considerably large areas with little damage to property and life. Bulkeley et al (2009) could not provide full reasons behind the distinction between the urban and rural floods. Jha et al (2012), contributes to the discourse by stating that the level of infrastructure in rural areas exposed and damaged by floods is low compared to the floods which occur in urban areas. In their literature, Siegel et al (2008) categorize urban flooding into four forms. The first is referred to as localised flood or flash flood. This form of flood occurs more frequently University of Ghana http://ugspace.ug.edu.gh 14 and in most cases covers very small and deplorable areas such as slums with poor drainage, limited drains and blocked culverts with all forms of waste. This flood closely relates to overland or pluvial floods described by Jha et al. The second form of flood occurs in areas along small streams in urban areas, which rise quickly after heavy rains, flowing through small culverts under roads. Although authorities design channels stream very well, changes in the structure of urban areas, including increasing number of settlements along streams coupled with increased rainfall intensity may result in an overflow of the stream when its drainage capacity is exceeded. The third form of flood has to do with bursting of dams and major rivers, which provide water to urban communities, due to poor engineering works, dam water spillage and deposition of sediments in the dam. Further, coastal floods may also occur because of sea level rise and overflow of inland water over immediate delta plains. According to ILGS and IIED (2012), floods that are more frequent in Accra include flash floods and river floods (occurring when there is an overflow of the Weija Dam). Coastal flood also occurs along low-lying coastal communities in Accra (Appeaning, 2011). 2.3 Extent of Flood Disaster Every year, floods affect more than 200 million people with a higher record than the records shown by other climate related disasters (UNISDR, 2011). Floods have become the most frequently occurring and devastating disaster in Africa, causing loss of lives, damage to property and promoting the spread of disease such as malaria, dengue fever and cholera (Baffoe-Bonnie et al., 2006). Mulungeta (2007) highlighted that floods in Africa killed almost 20,000 people and affected nearly 40 million more, and caused damage estimated at about US$4 billion between 1900 and 2006. A number of floods occurred in Ethiopia in May 1968, August 1994, and May 2005, causing damage estimated at US$ 0.9, 3.5, 1.2, University of Ghana http://ugspace.ug.edu.gh 15 and 3.5 million, respectively (OFDA/CRED, 2010). The period 1997-2008 alone witnessed about 166 floods and many other natural urban disasters in 28 African cities (EM-Dat, 2008; cited in Songsore et al, 2011). Flooding has been a serious environmental problem in Ghana causing extensive damage to lives and properties of people living in communities in towns and cities. As early as 1968, heavy floods affected almost every part of the country (Sam, 2009). Narrating the flood antecedents in Ghana, Atuguba and Amuzu (2006) and Sam (2009) indicated that the major rivers including the Pra, Ofin and Ankobra overflowed their banks causing floods in adjoining communities and cities. Transportation (rail and vehicular) was seriously impeded and many commercial activities were halted. Serious flooding has been experienced ever since. Quite recently in the last decade, the years 1999, 2001, 2007, 2008, 2009, 2010 and 2011 experienced quite destructive floods all over the country. In 1999, rainfall induced storms caused coastal floods that resulted in many deaths across coastal parts of Ghana, with Accra very much affected. In June 2001, torrential rains caused widespread flooding in Ghana and particularly, Accra, leaving 11 people dead and over 100,000 homeless. In 2005, Ghana was one of the worst hit countries in western Africa that experienced severe flooding. The Daily Graphic (2005) reported that about 20 human lives were lost across the Upper East, Upper West and Northern regions of the country during the floods of 2005. In 2007, floods affected Northern Ghana; killing 61 people with 25,923 houses affected (damaged, collapsed, or washed away). In addition, 70 feeder roads were destroyed and over 97,000 hectares of farmlands destroyed. The government and other international agencies spent about US$25 million in relief and recovery activities. In 2009, the damage cost was University of Ghana http://ugspace.ug.edu.gh 16 approximately US$5,800,000 and 51,965 people were affected in seven regions (Western, Central, Brong-Ahafo, Volta, Ashanti, Eastern and Greater Accra). Ghana National Disaster Management Organisation (NADMO), estimates that about 350,000 people were severely affected during the 2005 floods and during that time several hundred hectares of crops and farmlands were completely washed away (ActionAid, 2006). The study of ActionAid (2006) projected that many more deaths attributable to flash floods could occur across Ghana, in addition to those resulting from the more usual cause of river flooding. Together with ActionAid (2006), Atuguba and Amuzu (2006), indicated that over 25% of the population of Accra lives on fluvial flood plains or areas identified to be at risk from fluvial flooding. Greater Accra Region has about 50% of its population living on the floodplain of the Densu and tributaries of other rivers (Atuguba and Amuzu, 2006). NADMO (2009) reported that floods had led to the death of twenty people, displaced tens of thousands of households, and destroyed millions of property in communities such as Alajo. The Report of NADMO (2009) also indicated that between 1995 and 2009, more than ten incidents of flooding were recorded in Accra that resulted in human casualty, displacement of households, infrastructure damage, and disruption to economic activities (AMA and UN-HABITAT, 2011). According to Jankowska (2009), the effects of floods cut across all the demographic and socioeconomic groups in the Metropolis varying from delay in economic activities, psychosocial stress, and emotional disturbance among others regardless of where one lives. Jankowska (2009) added that apart from property damage and loss of lives, floods have contributed towards disease outbreak and the pollution of water bodies available for urban consumption. He noted also that many areas in the Accra metropolis have had their drinking waters contaminated and face the risk of a cholera outbreak. University of Ghana http://ugspace.ug.edu.gh 17 2.4 Causes of Flood Disasters Globally, there are increasing changes in the climate, with varying rainfall patterns, wind, temperature, sunshine, etc. Following various scenarios, including that of Global Circulation Model (GCM) and organizations such as IPCC, UNFCCC and WMO, and climate related research on local knowledge and perceptions, much evidence has been obtained across the globe regarding the level of change in the climate, particularly since 1990s (Würtenberger et al, 2011; Selvaraju et al, 2006; IPCC, 2007). IPCC (2007) explained that over the 20th century and until today, there has been a faster rate of change in the climate, resulting in more unpredictable and extreme climate events such rainstorms and excessive heat than previously. In Ghana, annual rainfall is highly variable on inter‐annual and inter‐decadal timescales and long-term trends are difficult to identify (Brown and Crawford, 2009; Meehl et al, 2007; Stanturf et al, 2011). According to Minia (2008), rainfall over Ghana was particularly high in the 1960s, and decreased to particularly low levels in the late 1970s and early 1980s, producing an overall decreasing trend in the period 1960 to 2006, with an average precipitation of 2.3 mm per month (2.4%) per decade. According to NOAA (2007), average global rainfall had increased over last century; however according to Wurtermberger et al (2011), rainfall has decreased in Ghana and most parts of Africa over the last four decades. The decreasing pattern of rainfall in Ghana however, has not correlated directly with increasing floods in Ghana (Karley, 2009), though rainfall is believed to be a natural cause of flooding in most part of the world. Arguments have been made however, that current floods are because of abnormally intensive, unpredictable and destructive rains accompanying tropical cyclones and storms (UNEP, 2012, Ferris, 2012). According to Huong and Pathirana (2013), primary observations made University of Ghana http://ugspace.ug.edu.gh 18 on rainfall is its high intensive pattern in urban communities result from rooftop, flashing and running over impervious urban surfaces with high run-off causing various degrees of floods. Nethler (2012) indicates that the erratic and unpredicted nature of rainfall in modern times is a major contribution to floods in most communities where they occur. Additionally, the unpredicted nature of rainfall makes it difficult for meteorologists to give warning and advice on looming volumes of rainfall in any upcoming season, so that preparation by people in terms of relocation, flood defences and other preparedness that reduces the extent of damage could be made. Wurtermberger et al (2011) noted that during rainy seasons, various rivers and streams are filled and over flow their banks, a situation which has devastated many communities along the River Volta and many other rivers in the country. Similarly, rivers and streams in Accra are always heavily flooded. Such rivers include the River Densu and Odaw River running down to the Gulf of Guinea. Many authors were of the view that the amount of rainfall in Accra is less significant in increasing the volume of rivers and their over flow (Atuguba and Amuzu, 2006; Songsore et al 2011). It was further argued that if floods should be associated with annual rainfall in Accra, which is about 700mm according to its location in a dry coastal savannah and dry coastal vegetation and climatic belt, then positively linking rainfall to increasing river flood is a mere assertion. Separate studies conducted by Adank et al (2011), UN-HABITAT (2011) and Aboagye (2011) revealed that rivers flowing into Accra experience relatively higher rainfall upstream from their main channels and tributaries, whereas Accra is located downstream and in a destination where they finally enter into the Sea (Gulf of Guinea). University of Ghana http://ugspace.ug.edu.gh 19 2.4.1 Non-Climatic Factors and Urban Flooding According to Gallopin (2006), climate- related disasters such as floods are expected to increase due not only to increasing climatic changes but also to also non-climatic factors. Considering climate change and the associated natural disasters, many people viewed it as inevitable and beyond the available technological capacity (Fara, 2001). A study conducted by Cutter et al (2000), however, indicates that, apart from climate change, vulnerabilities generated through human actions have the potential to exacerbate flood disasters. The work of Cutter et al (2000) coincides with Fussel (2009), whose study on climate change vulnerability indicates that communities in developing countries remain more vulnerable to flood disaster due to their deplorable social, economic, political, natural and human capitals, which negatively affect their resilience. According to the IPCC (2007), a combination of natural hazards such as floods with persisting social and economic vulnerable conditions has increased disaster occurrence in poor urban communities, especially in urban areas in Least Developing Countries (LDCs). Vulnerability to climate change may not only be the extent to which a natural or social system is exposed or susceptible to risk of flood hazard but also the capacity of affected populations to recover from their effect (IPCC, 2007). Sanderson (2000) identified rapid urbanization as one of the causes of increasing urban vulnerability to floods and other disasters in urban communities in modern times. “Recent natural disasters show that it is almost always the poorest who are worst affected. The rapidly urbanizing cities of Asia, Africa and Latin America present unprecedented concentrations of poverty, and in so doing mark new levels of vulnerability” (Sanderson, 2000). University of Ghana http://ugspace.ug.edu.gh 20 United Nations, ISDR, Hyogo Framework for Action 2005-2015, indicate that increased population densities and other conditions in cities may worsen the impacts of natural hazards (UN/ISDR, 2009). WHO and United Nations Human Settlements Programme Report (UNHST, 2010) shows that more than half the global population lives in cities. This is projected to reach 70% by 2050, with a growth of nearly 60 million people every year (UNEP, 2011). Despite the deplorable conditions, including poverty in Africa, the continent is fast urbanizing with many more people now living in urban areas (AAP, 2010). Correspondingly, Ghana is one such country in Africa with more than 50% of the population living in urban areas (GPHC, 2010), with a 4.2% annual growth rate. According to EPA and UNDP (2012), the increase in urban populations in Ghana have significant challenges for city authorities. Challenges including urban waste management, vehicular traffic, affordable and adequate housing, sprawling of slums and growing “distressed” spots are some manifestations. Progressively, people in cities continue to face greater risks that expose them to the negative impacts of climate change as a result of a combination of factors including natural hazards (flooding, windstorms, heat spells), high population density, poor infrastructure and low resilience to economic shocks. According to Lino (2012), cities are distinguished by better facilities and better living conditions plus employment, education and business. However, with huge developmental disparities in the rural-urban divide, there is incessant out-migration of poor people from the countryside into the cities. Urban place and space have become more vulnerable and are described by writers as crucibles of hazards. Urban conditions have been described by Red Cross Society as “the Social Geography' of many towns and cities to ‘reflect the vulnerability of different areas to natural hazards’ (Lino, 2012). University of Ghana http://ugspace.ug.edu.gh 21 Rapid urbanization is unfortunately not responding to the increasing demand for housing, employment, waste and water management, drainage systems and proper planning and land use in Ghana (Adank et al, 2011). According to AMA (2011), most slums and other informal communities highly exposed to flood. They are areas where job seeking migrants initially settled with difficulties with accommodation. They resorted to accommodation in the low residential communities and places along the fringes of the city, where rent is less expensive. People occupy buffer zones along drains, which are not officially sold to them. Aboagye (2012a) and Adank et al. (2011) discussed poor drainage systems and lack of maintenance culture as other sources of floods in cities. The UN/1SDR (2011) indicates that most flooding occurs in cities which lack storm water drainage channels to regulate run-off. Lack of well-engineered drainage systems, especially in densely populated areas, water run-off from the roofs of buildings, improper waste management and the extensively paved urban land surfaces all contribute to flooding. Atuguba and Amuzu’s (2006) study in Alajo, recommended the need for the expansion of drainage channels especially from the Odaw River. The subsequent expansion work done on the Odaw River saw a reduction in the incidence of flooding in Alajo community. A study by Anomanyo (2004), shows that lack of and/or proper management of drainage systems in Accra was just one of the bases for flood occurrence in the city. He explained that, bad waste practices have affected the free flow of water in gutters (drains). He stated that, although about 60 to 75% of solid waste generated in the city is collected, the solid waste that remains uncollected often finds its way into open drains, thus obstructing free flow of water causing overflows that result in floods (Anomanyo, 2004; Fobil, 2007 cited in Adank et al, 2011). A few years back, similar studies were conducted in Old Fadama, a typical slum in Accra by ILGSS and IIED (2012), which found that although there is a sturdy improvement in waste practices, attitudinal change remains a major challenge to proper University of Ghana http://ugspace.ug.edu.gh 22 waste management. They indicated that 94% of waste is collected in Old Fadama every day and the rest deposited in hidden corners, gutters or is openly burnt. As a result, waste materials constantly find their ways into open gutters obstructing the smooth flow of water upstream in the Odaw River and other streams, which exit into the Gulf of Guinea. Again, Atuguba and Amuzu (2006) noted that lack of routine maintenance has impeded the efficient functioning of the drainage systems, leading to severe flooding during rainy seasons in Accra. In most cities, land use planning and control are a pre-requisite to any form of development and a first step to building sustainable flood resilience. According to Karley (2009) improper planning in modern cities is characterized by poor land use plans that are manifested in poor zoning of settlements, industrial activities and proper drainage. He indicated that most often certain locations become the receptacle of the negative consequences from poor planning. He also, emphasized the need to manage the urban water system completely to avoid such occurrences as flooding. Studies by Afeku (2005) and many others argued that the master plan for the development of Accra was outmoded to regulate planning and development of Accra and needs updating. Further, he underscored the need to enforce the existing laws properly whilst working hard on attitudinal change as a prime concern to curb the increasing vulnerability of city dwellers. Sam (2009) and Afeku (2005) raised concerns about corrupt and ineffective administration of the system that continues to grant building permits without due diligence. This action results in the haphazard citing of buildings in waterways and in unauthorized locations. Such lands and settlements do not have good access roads, drainage, proper waste disposal spaces and services because of their illegal development. Such areas also become more liable to flooding (Sam, 2009). Recent road projects and other urban developments have University of Ghana http://ugspace.ug.edu.gh 23 made substantial investments in tertiary drains but little or no attention has been paid to primary and secondary drains for the recent increase in volume of run-off. Adank et al (2011) add that institutional frameworks related to storm water drainage are weak. Haphazard urban planning and development, evidenced by the lack of drainage systems in settlements, poor construction of existing drains and corruption in institutions and management system for buildings are some of the identifiable human related causes of flooding. It is culturally and legally unacceptable in Ghana, including Accra to build on spaces that are earmarked for green development or public space but the opposite is the obvious situation (Rain et al, 2011). Rain et al., indicate that housing development on green spaces decreases the natural storage capacity of soil thereby resulting to increased run-off in those areas. In a study that covers urban poor settlements in Accra (Ghana), Kampala (Uganda), Lagos (Nigeria), Maputo (Mozambique), and Nairobi (Kenya), flooding in urban areas can be seen as a consequence of multiple factors including, housing development in floodplains, inadequate waste management and lack of maintenance of storm water drainage channels (Dodman et al, 2013). The combined effect of these factors together with the intensive and erratic rainfall exacerbates flooding problems in the urban communities. Floods are indeed not only a natural environmental problem but also a result of human and anthropogenic factors that have further aggravated the situation (Dodman et al, 2013). 2.5 Flood Risks and Vulnerable Areas in Accra Flood disasters occur in many parts of the country, but are more prevalent in the low income or slum areas. Figure 2.1 shows that where the urban poor live is mostly unplanned and informal in its physical layout making such locations very vulnerable to floods (Atuguba and Amuzu, 2006; Adank et at.2011). Nyarko (2000), in his study used GIS to map out flood University of Ghana http://ugspace.ug.edu.gh 24 hotspots in Ashaley Botwe, Ashaiman, and Legon Farm, Alajo, Odorkor, Accra Newtown, etc. in the AMA of Greater Accra Region. Figure 2:1 Map of Slum Areas and Flood Disaster Source: AMA Medium Term Development Plan, 2010-2013 Communities along the 8km Densu River below the Weija Dam are also vulnerable whenever there is spillage of excess water. The communities that suffer from the spillage include Dansoman area and those along the Lafa Stream, which crosses the Winneba and Motorway extension roads (AMA, 2011). In the low-lying areas of the Korle-Chemu Catchment, flooding is a serious problem, with many houses being inundated by floodwater during and after heavy rains. Similar problems occur in the industrial land cemetery area around the Obetsebi Lamptey Circle, where the interceptor drains and culverts are inadequate and often chocked. University of Ghana http://ugspace.ug.edu.gh 25 There are many other areas including Nima, Dzorwulu, Darkuman, which have experienced localized flooding, caused by inadequate drainage channels and low topography (AMA, 2011). In the Kpeshie Catchment area, drainage in the La Township is inadequate and many waterlogged areas become flooded with light rains. In periods of heavy rains, fence walls collapse and foundations are broken (AMA, 2011). A particular case is the flood that occurred at the end of October 2011. Even though the rains and floods are perennial, heavy rain can occur off-season and cause flooding. Since the preparedness for floods in the off- season is low, the effect of the flood can be quite extensive (UNEP, 2011). Most of the vulnerable areas identified were characterized by limited infrastructure (drainage) and high run-off, including the Lafa and Sakumo II Basin, which are very prone to flooding. Floods occur more extensively in low-lying areas and places with inadequate provision of drains, thus, regulation of human encroachments in the low-lying areas could prevent many annual floods in the AMA (UN-HABITAT, 2011). 2.6. Adaptation to Flood through Vulnerability Reduction Fara (2001) indicates that efforts towards managing floods and other natural disasters include the provision of relief to affected people. Relief in the form of food aid, drugs and clothing mainly provided by governments and NGOs, developed countries and international organizations. However, disasters since late 1990s are known to be due to human induced phenomena (Holloway, 2000), hence adaptation to flood and other disasters have sought to take a human face through initiating actions to build the resilience of people and their communities. According to Birkmann (2006), vulnerability to flood disasters includes exposure to floods as a result of building along flood plains, and low resilient accommodation structures. He added that population which is weak in terms of age, gender, health status, and infirmity suffer flood effects more and are less able to recover. University of Ghana http://ugspace.ug.edu.gh 26 Authors such as Turner et al, (2003); Ellis (2003 and 2013); Broos, (2003) outlined wide range of vulnerability sources which were summarised by Fussel (2007) in Table 2.1. They agreed with other discussions that disasters are not only because of people’s exposure to natural or because of what is termed as biophysical conditions, but also human induced factors, which they term as socio-economic factors. Table 2.1 Vulnerability Model Sphere/ Domain Domain Sphere Anthropogenic conditions Biophysical/Natural Conditions Internal Household income, Topography (e.g., lowlands) Social and economic amenities/services Climate conditions Access to social network/ information Land cover People’s participation in decision making Soil, water, etc. Equity, etc. External Ethnic diversity, migration, Climate change/variability National policies Floods, severe storms International aid Disease outbreak Economic globalization Heat waves International relation/global partnership Sea level change rises Source: Fussel (2007) This distinction is quite comprehensive with Fussel (2007) who indicated that vulnerability as a concept is broad, hence the need to identify specific vulnerable locations, people, elements and other attributes which constitute risk, plus the hazards people have been exposed to and variations in space. Fussel’s vulnerability analysis is summarized in Table 2.1. University of Ghana http://ugspace.ug.edu.gh 27 Metzger et al (2005) states that the term vulnerability is used with reference to particular situations in a particular locations in order to focus adaptation options. According to Smit and Pitifosova (2001), cited in Fussel (2007, 2009), once people’s vulnerability is reduced through well-focussed adaptation strategies, communities and their residents will become resilient, adaptive and robust to hazards. Figure 2.2: Adaptation Focus Model Source: Smit et al (2001) Adaptation strategies targeted towards reducing vulnerability is more sustainable (Gallopin, 2006). IPCC (2000) defines adaptation as adjustments in ecological, social or economic systems, in response to actual or expected climatic stimuli and their effects. In line with this definition, Gallopin (2006) states that adaptation to the effects of climate change should incorporate response to climate change hazards such as flooding and reducing extent of exposure to and building resilience against further incidence through strengthening the necessary infrastructural capacities. ADAPTATION TO WHAT? Climate Related Stimuli Flood over • Time/Space Scales HOW DOES ADAPTATIONS OCCUR? TYPES • Processes • Outcomes People/Comty. WHO AND WHERE? SYSTEM • Definition • Characteristics HOW GOOD IS THE ADAPTATION? •Evaluation • Criteria • Principles NON-CLIMATE Forces and Conditions SYSTEM • Definition • Characteristics University of Ghana http://ugspace.ug.edu.gh 28 Smit et al (2001) developed the general concept of adaptation or adaptive capacity in climate change and climate related impacts extensively. They consider various components including form, scale and process of adaptation as summarized in Figure 2.2. Smit et al deepened the components of adaptation with reference to systems, interest, activities, sector, community or group’s adjustment to something related to shocks, stress or stimulus from climate and climate related hazards. The question “where” and “who” in the adaptation model is relevant since adaptation strategies needed to be targeted towards particular areas and groups of people. With the increasing human induced disasters associated with floods especially in urban communities and dwellers, Smit et al advocated for a greater focus on socio-economic activities and infrastructure and ecosystem management. It is very important to acknowledge the description made by Sanderson (2000) that urban areas are increasingly becoming more prone to disasters especially in regions with endemic poverty. Doing this would promote better adaptation strategies and practices. Jha et al (2012), discussed adaptation options extensively, by looking at the following areas:  Use of probability models to determine vulnerability and potential damages of climate change hazards  Using maps to show communities more affected by disasters in order to determine what strategies are best in adapting to the disasters in those areas  Development of early warning systems  Awareness creation and education  Land use planning and settlement development  Drainage system development One of the views expressed by Jha et al was on clearly mapping the areas that are affected and prone to disaster, an approach they believe is more economical, more realistic and action University of Ghana http://ugspace.ug.edu.gh 29 focused. The capacity to adapt varies considerably among regions, countries, and socioeconomic groups over time. The most vulnerable regions and communities are those that are highly exposed to hazardous climate change effects, and have limited adaptive capacity. Countries with limited economic resources, low levels of technology, poor information and skills, poor infrastructure, unstable or weak institutions, and inequitable empowerment and access to resources have little capacity to adapt and are highly vulnerable (Smit and Pilifosova, 2000). Adaptive measures also differ based on demographic, ecological and socioeconomic characteristics of the people. The ILGS and IIED (2012) noted that at the individual level, residents make prior arrangements for floods, which routinely occur every year. People resort to using blocks, stones and furniture to create higher platforms, stay with family and friends during floods among others as coping mechanisms. During floods or rainstorms, neighbours adopt a number of coping strategies. Many try to draw each other’s attention to the upsurge of flood, especially during heavy downpours of rain. In the process, they help each other to carry property in and out of houses before, during and after floods (Action Aid International, 2006). Some residents undertake collective work to open up drainage channels while others temporarily move to lodgings and public places such as mosques sport stadiums and churches until floodwaters recede. Some also make temporary outlets to ease floodwater movement/flow near their homes (Action Aid International, 2006). Most of these efforts are coping strategies and not sustainable adaptation options responding to the unpredictable nature of rainstorms and the potential increase in floods in recent times. University of Ghana http://ugspace.ug.edu.gh 30 2.6.1 Governments Initiatives to Flood Adaptation Most African governments have national disaster and emergency policies. Uganda has special ministries responsible for disaster preparedness and response. The overall objective of disaster preparedness and management in Uganda, for instance, is to save lives and livelihoods, reduce Uganda’s vulnerability to potential disasters in cooperation with local communities, NGOs, local and international donor organizations among others. They also seek to enhance the country’s ability to contain or minimize the social and economic effects of disasters. In particular, the Uganda strategy aims at mitigating the impact of hazards in order to avoid disasters. This implies good land use planning and avoiding building in hazardous locations (Mulungeta, 2007). In Mozambique, a Master Plan for prevention and mitigation of natural disasters was developed 2006, as part of the 2005–2009 poverty reduction strategy. Notably, the policy highlights the development of flood warning systems and the recognition of effective disaster management as key component of poverty reduction and was a step to achieving the Millennium Development Goals (Mulungeta, 2007). The plan addresses issues at the national level, and does not give special attention to urban areas. Ghana is a signatory to a number of global interventions on climate change and disasters. For example, the United Nations Convention on Climate Change (UNFCCC), which expects national governments to ensure that climate change issues are integrated into national development frameworks and plans. Similarly, Ghana signed for the Hyogo Framework for Action (HFA) 2005-2015 and Sendai Hyogo Framework for Action (HFA) 2005-2015: Building the Resilience of Nations and Communities to Disasters. The frameworks aimed at reducing casualties as well as socio-economic and environmental losses because of disasters, in a sustainable manner (MEST, 2012). University of Ghana http://ugspace.ug.edu.gh 31 Ghana’s initiative towards streamlining climate change adaptation into policies and strategies started little over fifteen (15) years ago under the Netherlands Climate Change Study Assistance Programme (Würtenberger et al, 2011). This support and many others including UNDP Supported Climate Change-Dare Programme led to the formulation of the National Climate Change Adaptation Strategy (2011) and streamlining climate change adaptation into Ghana’s Shared Growth and Development Agenda 2010-2013 and 2014- 2017. The National Climate Change Adaptation Strategy was transformed into a National Climate Change Policy, 2014 to guide the processes leading to streamlining climate into national policies, strategies and programmes. Before a national policy framework is approved as a working document, it is ensured that issues of Climate Change (CC) and Disaster Risk Reductions (DRR) are prominently featured. More importantly, CC/DRR issues have to be accepted as a national priority and a sine qua non to sustainable development. Additional effort made by Ghana include the development of a National Urban Policy with some activities focusing on development of urban drainage, waste collection, and climate change as well as disaster management. One of the challenges that continues to hinder the successful implementation of the policies and programme is lack of proper implementation guidelines, financial constraints and commitments, and lack of proper consultation and monitoring (ILGS and IIED, 2012). In addition, there is very low attention paid to and priority of urban vulnerability and adaptation to the climate change. The extract below indicates urban vulnerability and the need to focus on climate change adaptation for building resilient cities: “Ghana’s urban population is expected to increase from 52% of the total population of 2010 to around 65% in 2030 based on a projected urban rate of around 3% per year. Accra alone constitutes 16.1% of the University of Ghana http://ugspace.ug.edu.gh 32 total population in Ghana with a growth rate of 2.8% in the 2000 (GSS, 2000). The number of people living in slums in Ghanaian cities was estimated at 5 million at a growing rate of 1.8% per annum in 2001. This is pronounced in Accra, Kumasi and Tamale (GSGDA, 2010 – 2013). This puts pressure on infrastructure development. The peri-urban communities, occupied by migrant settlers and low- income urban dwellers have deplorable conditions in respect of infrastructure, services and housing quality. The impact of climate related hazards such as floods are high in view of weak infrastructure due to poor planning and decision, as a result of lack of appropriate building code and design standards and methods” (MEST, 2012). The National Disaster Management Organization Act (Act 517), 1996 created NADMO. The purpose of NADMO is to coordinate emergency and disaster response, and general national disaster preparedness (NADMO Act 517, 1996). Key emergency committees and protective services that operate in Accra are the Ghana Police Service, the Ghana National Fire Service, and the Ghana Red Cross Society. All of these institutions operate under the institutional coordination of the National Disaster Management Organization (NADMO). A number of Sub-committees are under NADMO, addressing issues such as geological hazards including earthquakes, landslides, and coastal erosion; pest and insect infestation; relief and reconstruction; hydro/meteorological disasters; bushfires and lightening; epidemics; and human induced disasters (Atuguba et al., 2006). There are two pressing concerns for NADMO including the capacity to deliver on its mandate and disaster communication. The present ability of the Ghanaian government to respond to anthropogenic and natural disasters is inadequate. Currently, according to studies on disaster preparedness of Accra, disaster response times in Accra are normally as high as 13 to 15 minutes due to bad communication and improper planning of buildings and other structures (Atuguba & Amuzu, 2006). University of Ghana http://ugspace.ug.edu.gh 33 Actions at a policy level for the implementation of the NADMO law include the establishment of NADMO offices at the national, regional and district levels, drawing of disaster management plans at the national, regional and district levels, the establishment of disaster management committees at the national, regional and district levels and plans to coordinate the activities of all collaborating agencies. Atuguba and Amuzu (2006) shared the view that, although there are institutional structures all over the country, the limited resources available have reduced their abilities to implement their mandates fully. NADMO’s activities range from awareness creation, desilting of gutters, provision of relief items, victims’ registration, needs assessment, search and rescue operations. Others carry out demolition exercises on waterways to allow easy flow of run-off in the metropolis (Action Aid International, 2006; NADMO, 2005). The Ghana Red Cross Society has advised communities to reinforce their roofs and build embankments along their walls to strengthen them. Families have been advised to put in place individual emergency and contingency plans during emergencies. In Accra, emergency response teams have continually encouraged communities in flood prone areas to desist from indiscriminate dumping of refuse into gutters and to develop and implement evacuation plans (IFRC,2009). Attempts to solve the perennial flooding have always been a knee-jerk approach. There must be a holistic assessment of the flooding process before, during and after the floods for the appropriate strategies. Such assessments could be very useful in building the capacity and resilience of cities. Studies by Action Aid International (2006) among slum dwellers in six African cities, uncovered very few if any, collective mechanisms towards reducing flood risks, or for managing floods once, they happen in the cities. Local coping strategies and traditional knowledge need to be synergized with scientific knowledge for appropriate interventions University of Ghana http://ugspace.ug.edu.gh 34 (UNFCCC, n.d). These strategies are necessary, as adaptation measures are most effective when developed to incorporate community-based experiences. 2.7 Conceptual Framework of Vulnerability and Adaptation to Disaster Conceptual frameworks are used to show complex interconnections among elements that seek to explain causes and effects of phenomena and integrate several components. This study used a framework (Figure 2.3) to identify climate, non-climate and multiple factors that lead to disasters, and the adaptation resulting strategies used by communities. It also provided a structure for vulnerability assessment for capacity building for climate change impacts management, and adaptation to the disaster (Macchi, 2011 and Fussel, 2007). According to Practical Action (2011) it is emphasized that the relevance of the framework is its contribution to reducing vulnerabilities and strengthening resilience of communities (local community, regional, national and international). The framework indicates that flood disaster occurs as a result of both climate and non- climate conditions. Gallopin (2006), Cutter et al (2000), Fussel (2009), and IPCC (2007) further elaborated this synergy in their research findings. The climate factors are the natural biophysical factors as identified by Fussel (2007). These include precipitation, temperature, winds, and other climatic variables. Accra records very high intensity of rainfall especially in June/July almost every year, which in many cases leads to high run-off and floods. Flooding, especially river flooding is not always due to the occurrence of rainfall in Accra but also rainfall upstream of the rivers/stream flowing through Accra. Non-climatic conditions including economic, social, demographic, technological, and political factors also contribute to flooding. The framework specifically identifies population density in urban communities, encroachment on public habitable and non-habitable areas, housing University of Ghana http://ugspace.ug.edu.gh 35 problems, poor incomes, sanitation, decision making and policy formulation, implementation and regulation as non-climatic factors. Figure 2.3 Conceptual Framework on Flood Disaster, Causes and Adaptation Source: Adapted and Modified from Macchi, 2011 Urban dwellers are supposed to have available capital assets due to opportunities provided in cities; however, increasing population in the cities has reduced the number of available opportunities, and demarcated social classes with the poor mostly living in more deprived areas without adequate housing, drainage, water, good health, etc. Urban poverty has ADAPTATION Policies/Research Individual knowledge Processes Mitigation Adaptation Coping NON-CLIMATIC/ HUMAN FACTORS Population Density Encroachment Housing Income Sanitation Decision-making Policy and Regulation Social protection VULNERABILITY/CATALYTIC CONDITIONS Exposure, Sensitivity Coping Capacity Resilience, Robustness................ Adaptive Capacity CLIMATE CHANGE AND VARIABILITY:  Rainfall (Intensity, Volume, Predictability  Wind (Intensity, Volume, Destruction)  Other climate DISASTERS  Flood  Pest/Disease Illness,  Heat waves  Drought PEOPLE/COM’TY AFFECTED University of Ghana http://ugspace.ug.edu.gh 36 increased beyond rural poverty rate in recent times in Accra (GSS, 2005; GSS, 2008). The poor are indeed simply more vulnerable to physical hazards because of their exposure, sensitivity and low adaptive capacities (UN-HABITAT, 2007). Vulnerability has been related or equated to concepts such as resilience, marginality, susceptibility, adaptability, fragility, and risk’ relating to climate change (Liverman, 1990). Fussel (2007) further added exposure, sensitivity, coping capacity, criticality, and robustness to this list. According to Fussel (2007), the degree of any individual’s vulnerability to impending hazard-including extreme changes in climate and socio- economic conditions may vary depending on the extent of exposure, sensitivity and resilience. Vulnerability to flooding is driven not only by climatic factors but also by many other variables, and had to be developed into a more integrated, policy-driven system as shown in the framework. Climate change vulnerability constitutes the degree to which a system is susceptible to or unable to cope with adverse effects of climate change, including climate variability and extremes (IPCC, 2001). Fussel (2007) describes climate related vulnerability based on the characteristics of the vulnerable system such as the urban poor communities, the type and number of stressors and their root causes, their effects on the system, and the time horizon of the vulnerability. Vulnerability is a function of the character, magnitude and rate of climate variation to which a system is exposed, its sensitivity and its adaptive capacity (IPCC, 2001). It is identified that flood disaster does not occur everywhere except in communities where both climate and human induce factors interplay more extensively. In Accra, such phenomena are common in low residential areas and mainly where there are slums as mapped by AMA and UN-HABITAT (2011). People with low resilient structures, University of Ghana http://ugspace.ug.edu.gh 37 low income capitals and are located close to river banks in Accra are more exposed to fluvial floods. Lastly, the framework indicates coping, adaptation, recovery and mitigation processes regarding climate and non-climate related hazards. Adaptation to climate change is necessary, in addition to mitigation strategies to offset unacceptable impacts of anthropogenic change (IPCC, 2007). Adaptive Capacity is the ability of a system to adjust to climate change, to moderate potential damage, to take advantage of opportunities, or to cope with the consequences (Owaygen, 2010). Mitigation actions are based upon origin and evolution of activities that aim at reducing greenhouse gas emissions directly or indirectly (e.g. by changing behavioural patterns, or by developing and diffusing relevant technologies) through capturing GHGs before they are emitted into the atmosphere or sequestering GHGs existing in the atmosphere by enhancing their sinks (IPCC, 2001). Mitigation usually requires long term and advanced technological course of action. Adapting to flood disaster as per the diagram (Figure 2.3) remains complex in terms of what constitute the adaptation, where, how, when as elaborated by Smit et al (2001) in Figure 2.2. The framework indicates coping within the adaptation process. Individuals cope by temporally relocating to live with relatives and friends when displaced by floods (ILGS and IIED, 2012). In terms of adaptation, some people in Accra develop flood defence walls, sand bags around the water entry points to prevent water from entering their homes. Adaptation is a process that involves policy driven institutions and individual involvement in decision-making. That is why it was very necessary to explore individual knowledge and perception in the field to compliment concepts and findings from other researches and existing policy documents. University of Ghana http://ugspace.ug.edu.gh 38 The framework showed that flood disasters are not only result of intensive rainfall but also exposure of people to vulnerable locations and conditions and other myriad of factors including institutional roles and individual efforts. 2.8 Chapter Summary The chapter reviewed literature beginning with concepts, extent and distribution of flood occurrences. The literature indicates that flood has risen in recent times with increasing damage to property and affecting more than 2.5 billion people, killing about 500,000 and causing economic losses of about US$700 billion. The review proceeded by identifying the causes of flood disaster which includes both climate and non-climate factors. It indicates that changes in the climate is evident by the pattern of rainfall, which is unpredictable, intensive and erratic, co-existing with strong wind and high temperature. Various authors have research focus on urban disasters, vulnerability and adaptation. This chapter has explained that flood disaster has increased in urban communities, which are increasingly characterised by high population density, functions, poor and unorganised and less resilient infrastructure. It has been noted that urban communities are becoming more prone to many disasters including floods in recent times. The literature identified efforts in Ghana and other countries on climate change policies, strategies and disaster management activities. In Ghana for instance, the intensified integration of climate change into national development plans and policies occurred about a decade ago. This led to efforts at formulating the Ghana Climate Change Adaptation Strategy in 2011, and the Climate Change Policy in 2014. The establishment and activities of National Disaster Management Organization was also discussed. The conceptual framework indicated inter-connections among flood disaster, climate factors as natural phenomena and vulnerable conditions arising mainly from human behaviour and University of Ghana http://ugspace.ug.edu.gh 39 influence. The framework for Macchi (2011) was adapted to suit the phenomena in Accra. The framework summarises that flood disasters are not only a result of intensive rainfall but also exposure of people to vulnerable locations and conditions. Adaptation processes were included in the framework and discussed. University of Ghana http://ugspace.ug.edu.gh 40 CHAPTER THREE STUDY AREA AND STUDY DESIGN 3.1 Introduction This chapter addresses two (2) major issues including the profile of the study area and study design. The chapter is divided into four sections. The first provides an introduction and the second focuses on the profile of the study area. The third section explains the study design and fourth sections is a chapter summary. 3.2 Study Area Profile This section examines the geographical and demographic characteristics of the study area. It further unearths socio-economic conditions as situated in the District to help appreciate the potential and challenges of development in the area. In view of this, the chapter specifically looks at the physical background, demographic characteristics, climate conditions, poverty, drainage patterns and the economy of the Accra Metropolitan Area (AMA). The chapter finally provides a detailed methodology adopted for the study. 3.2.1 History and Location of Accra The city of Accra originated from Ga-Mashie, which is made up of James and Ursher Towns (AMA, 2011). It is traditionally believed that the ancestors of the present inhabitants of James Town (Ngleshie) migrated from Nigeria, at the beginning of the 15th century. The Ga-Mashie people – meaning indigenous Ga’s followed them. These people first settled in Accra Central – James Town. New settlements later developed including Nungua, La, Osu (Christianborg), Teshie and Tema (AMA, 2010). At present, Accra Metropolitan Area (AMA) or simply Accra, is found with almost all the tribes or ethnic groupings in the country since its designation as the national capital of University of Ghana http://ugspace.ug.edu.gh 41 Ghana, coupled with rapid migration and urbanization(GAMADA Factsheet, 2008) and it has expanded to the area shown in Figure 3.1. Figure 3.1: Map of Accra Metropolitan Area Source: Field Work, 2013. The Accra Metropolitan Area (AMA) was initially made up of six Area Councils of 58 District Councils that were integrated under the New Local Government System to include Ablekuma, Ashiedu Keteke, Kpeshie, Okaikwei Kpeshie, Osu-Klottey, with semi- autonomy. Like any other MMDAs, the AMA was established by the Local Government Act, 1993 (Act 462) and Legislative Instrument 1615 with Sub-Metropolitan Assemblies including Ablekuma Central, Ablekuma North, Ablekuma South, Ashiedu Keteke, A MAP OF AMA SHOWING STUDY COMMUNITIES University of Ghana http://ugspace.ug.edu.gh 42 Ayawaso Central, Ayawaso East, Ayawaso West, Okaikoi North, Okaikoi South, Osu Klottey (AMA, 2013). It covers an area of 173 square kilometres. The Metropolis is located on Longitude 05°35'N of the Equator and on Latitude 00°06'E of the Greenwich Meridian. Accra Metropolis shares a boundary with Ga South to the North, Ga West to North-West, Ga East to the North, Adenta Municipal to the North East, Ledzo-Kuku-Krowor to the East and the vast Gulf of Guinea to the South (Aboagye, 2012b, AMA, 2010 and 2011). Figure 3.1 shows the map of AMA, which was originally a base map, scanned and digitised. It shows numerous communities in the Metropolis, including that of the communities selected for the study. This involves James and Ursher Towns, Korley Dudor, Agbogbloshie, Nima and Cantonment. 3.2.2 Geology The geology of the AMA consists of Precambrian Dahomeyan Schists, Granodiorites, Granites Gneiss and Amphibolite to late Precambrian Togo Series comprising mainly Quartzite, Phillites, Phylitones and Quartz Breccias. Other formations found are the Palaeozoic Accraian Sediments - Sandstone, Shales and Interbedded Sandstone-Shale with Gypsum Lenses, together with sandy beaches near the mouth of the lagoons that usually provide surface area for erosion by heavy downpours (Quatey-Papafio, 2006). Such rapid erosion has been intersected by the resistant coastline rock of Accra. The coastline is exposed to strong coastal wind action, which result in intense erosion; a phenomena which is a major contributor to coastal flooding (AMA, MLGRD and UNESCO, 2000). 3.2.3 Climate and Vegetation The Accra Metropolitan Area lies in the Coastal Savannah climatic zone with two rainy seasons (Ofori-Sarpong and Annor, 2001). The first rainy season begins in May and ends in University of Ghana http://ugspace.ug.edu.gh 43 mid-July and the second season begins in mid-August and ends in October. The average annual rainfall is about 780mm, which falls primarily during the two rainy seasons (EPA UNDP, 2012). Rains are mostly unpredictable and intensive but with short durations with storms giving rise to local flooding where drainage channels are obstructed (AMA, 2010). Relative humidity is generally high averaging 65% at mid-afternoon to 95% at night. Data available shows that there is a very little variation in temperature throughout the year over the Accra Metropolis (Ghana Meteorological Agency, 2013). The mean monthly temperature ranges from 24.7°c in August (the coolest) to 28°c in March (the hottest) with relatively high annual average of 26.8°c (EPA and UNDP, 2012). According to Rashid- Sally (2011), the high average temperature contributes to the increased heat of Accra especially in the Central Business District where there is frequent temperature inversion. Proximity of the study area to the equator makes daylight hours practically uniform throughout the year. The predominant wind direction in Accra is from the WSW to NNE. Wind speeds normally range between 8 to 16 km/hr. High wind gusts occur with thunderstorm activity, which pass in squalls along the coast. The maximum wind speed recorded in Accra is 107.4 km/hr (58 knots). Strong winds associated with thunderstorm activity often cause damage to property through the removal of roof materials among other damage. The high intensity of strong wind is experienced just before the end of the raining season, especially when accompanied by intensive and unpredicted rainfall that creates problems in Accra (AMA, 2010). Several areas in Accra experience micro climatic effects. Low profile drainage basins with a North-South orientation are not as well ventilated as those oriented East-West. Air is often trapped in pockets and an insulating effect giving rise to a local increase in air temperature of several degrees. This mainly and often occurs in Accra Newtown and Azumah Nelson University of Ghana http://ugspace.ug.edu.gh 44 Sports Complex and its environs. Similarly, some communities immediately along the coast experience sea breeze from WSW with relatively high moisture, which sometimes cause rusting, and erosion to metals, roofs, etc (UN-Habitat, 2011). Vegetation of the Metropolis has been altered in the recent century by climatic and other human factors (Quartey-Papafio, 2006). The Accra Metropolitan Area was believed to have been covered by dense forest with only a remnant trees surviving currently because of both human-environment interactions and changes in climate conditions, which are not favourable to some plant species (Quartey-Papafio, 2006; AMA, 2010). There are three broad vegetation zones in the Metropolitan area, comprising shrub land, grassland and coastal savannah vegetation which appears similar to the Southern shell, Sudan and Guinea Savannah north of the Accra plains. To the west of the Metropolis are shrub lands, which move towards the Aburi hills. The coastal zone comprises of two vegetation types: wetlands and dunes. The wetlands consist of mangroves with estuaries and lagoons. Their protection as coastal wetland is very important to the long-term sustainability of fish and bird which are poached by people.. The dune lands have been formed by a combination of wave action and wind. Most coconuts were planted in this zone in the 1920s but it is estimated that over 80% of those plantations have disappeared because of felling, disease and coastal erosion. The severity of sea erosion along the coast is manifested in the disappearance of long stretches of coconut plantations (AMA, 2010). A number of trees have been introduced including the shrubs such as Neems, mangoes, cassias, avocados; palms and bougainvillea are becoming prominent trees on the landscape of the Metropolis. The Achimota Forest is the only forest reserve in the AMA. The total area covered by the forest has reduced largely due to encroachment, bush fire, sand University of Ghana http://ugspace.ug.edu.gh 45 collection and illegal tree felling. As settlements and other human activities continue to expand, forests are depleted for development (Stow et al, 2012). 3.2.4 Population and Migration The population of Accra is 4.3 million with additional daily influx of 1 million (AMA, 2012) who commute to the City for various socio-economic activities. The population of AMA in general and its sub-metros may see population changes (see Table 3.1 for projected change figures). While the high population provides a large market base for investors, it also poses problems of sanitation, waste management as well as traffic jams especially during the peak hours of the day. During peak hours, Accra hosts about 25% of the national population and has population influx on a daily basis for economic activities aside from the residential dimension captured by the 2000 National Population Census (GSS, 2013; UN-Habitat, 2011). The period between 1960 and 1970 saw rapid industrialisation and expansion in the manufacturing and commercial sectors in some major areas within the Metropolis. This factor contributed to high immigration to Accra, and resulted in high population growth rate (6.32%) in 1960-1970 (UN-Habitat, 2009). The stagnation of the Ghanaian Economy during the 1970’s had an adverse effect on the growth rate of the Accra population as shown in the growth rate (7.51%) of 1970-1984. The decline in agriculture in rural communities in Ghana and industrialisation in urban regions, coupled with the boom in the service sector in the late 1980s and 1990s also propelled immigration to Accra. In addition, the concentration of social infrastructure in the metropolis compared to other regions remains a pull factor attracting people. University of Ghana http://ugspace.ug.edu.gh 46 Nonetheless, rural-urban migration alone accounts for over 35% of the population increase in Accra. Table 3.1: Population Changes of AMA according to Various Sub-Metros Sub-Metro Area (KM2) 2000 2010 2011 2012 2013 Ablekuma Central 11.5 181,541 619554 646195 673981 827950 Ablekuma North 13.6 140,063 483253 504033 525706 295200 Ablekuma South 15.1 211,493 729712 761089 793816 494679 Ashiedu Keteke 4.4 88,717 260174 271361 283030 408060 Ayawaso Central 6.0 155,947 435984 454731 474285 159661 Ayawaso East 5.8 128,641 359643 375107 391237 424173 Ayawaso West 35.2 50,334 140717 146768 153079 447534 La 36.0 133,721 373844 389919 406686 306424 Okaikoi North 21.1 141,085 394433 411394 429083 476533 Okaikoi South 12.0 96,600 270066 281678 293791 392149 Osu Klottey 12.0 96,634 270162 281778 293895 827950 Total 173 1,424,776 4337542 4524053 4718589 5060313 Total population density 173 82.36 250.73 261.51 272.75 292.50 Source: Accra Metropolitan Assembly, 2011 As the population increases in the city, its density over land and pressure on social infrastructure also increases. Population density in 2000 was 82.33 and 292.50 by the end of 2013, with Ashiedu Keteke being the most densely populated sub metro. Accra’s population like that of other urban centres is very youthful, with 56% of the population under the age of 24 years. It will be realised from the age sex ratio that 51% of the population are females and 49% are males. The dominance of females over males is a reflection of the nationwide trend of the 2010 Ghana Population and Housing Census. 3.2.5 Housing It is estimated that annually only 21.4% of housing units are added to total housing units in the Accra Metropolitan Area (AMA, 2010a). This leaves a backlog of 300,000 housing units University of Ghana http://ugspace.ug.edu.gh 47 needed to solve the housing deficit in the AMA. The AMA needs a minimum of 25,000 housing units annually to respond to the huge housing deficit in the AMA (AMA, 2010a). In Accra, the dominant form of housing is private formal and informal housing. Individuals, families, own the private formal houses, cooperate organisations, religious and other non- governmental organizations. They are also rented sometimes out to individuals or families. The private informal buildings are much common in the indigenous communities such as Ga-Mashie. The informal housing structures are predominantly visible in slum communities (UN- HABITAT, 2010). Most of these informal houses are characterised with poor access to essential services such as drainage systems, pipe borne water and electricity, roads and waste management services. About 20 % of all housing structures in Accra are made from temporal structures made up of plywood (GSS, 2010) which hardly stand up to the surface running water. The inability of the AMA to meet the housing needs of the growing population has led to sub-standard structures, insanitary environments, squatter, and slum communities (UN-HABITAT, 2010). People settle on encroached lands, waterways, drainages, open spaces including government and individual lands. On the other hand, there are residential areas with better housing conditions mainly in the high-class residential areas such as Cantonments and Airport Residential Areas. The buildings are solid, with good access to road, sanitation, drainage, good drinking water and fresh air (Quartey-Papafio, 2006). There are under-utilized and vacant lands in high income, low-density residential areas for intensive redevelopment into first class, high quality mixed residential areas for varied purposes including commercial uses. This dichotomy between informal and formal housing areas has perpetuated the level of causes, effects and distribution of disasters in the city. Slum areas unfortunately are more vulnerable and are University of Ghana http://ugspace.ug.edu.gh 48 persistently hit by seasonal floods. Fires also easily occur in the slum communities as result of the wooden structures, poor access and illegal connections to electricity and overcrowding. Malaria, cholera, noise, dust and heat are common phenomena in such informal settlements (Quartey-Papafio, 2006; AMA, 2010). 3.2.6 Poverty In spite of the economic boom (Ghana Human Development Report, 2007), there exists a number of challenges, high unemployment levels and increasing urban poverty across the Metropolis as depicted in Table 3.2. It is contradictory to note that while poverty in Ghana is generally reducing from 39.5% in 1998/99 to 28.5% in 2005/06, that of Accra is increasing from 4.4% in 1998/99 to 10.6% in 2005/06 (MDG, 2010). This is depicted in a number of “poverty pockets” in the Metropolis. Poverty is wide spread mong certain communities in the Metropolis including the La Old Town, Akweteman, Osu Anehor, La Bawaleshie, Ga Mashie, Chorkor, Sabon Zongo, Nima, Ayidiki, Avenor, Gbegbeyese, Mamponse (UMLIS, 2011). Poverty pocket means the segment of the population that verifiably lacks information, power and resources and is usually excluded from development interventions (UN-HABITAT, 2011). It usually includes women and vulnerable groups which always require particular attention. The poor in parts of the city are defined as people who know what they want but cannot realize these desires for lack of or denial of access, or lack of economic means due to sex, age, tribe or social group, location etc. (AMA, 2013). Poverty also relates to social, political, cultural and environmental/physical factors such as discrimination based on gender, ethnicity, caste, age or disability; lack of access to education and training; poor health; lack of representation; lack of empowerment; and vulnerability to shocks and crises University of Ghana http://ugspace.ug.edu.gh 49 (World Bank, 2007). The poor are also vulnerable and socially excluded in several parts of the City especially in low residential areas (Jankowska, 2009). Table 3.2: Poverty and Extreme Poverty Trends in Ghana and Accra Regions Poverty Extreme Poverty 1991/92 1998/99 2005/06 1991/92 1998/99 2005/ 06 Accra - 4.4 10.6 - 1.9 5.4 Greater Accra 26.0 5.2 11.8 13.0 2.4 6.2 National 51.7 39.5 28.5 36.5 26.8 18.2 Source: Ghana Human Development, 2007 Vulnerability is a state of persistent deprivation based on poverty and denial of other basic rights and entitlement leading to the exclusion of certain groups from active participation in economic, political and social life. Social exclusion refers to people who are persistently deprived of their socio-economic rights. They are eventually excluded from the mainstream of activities of decision-making process and lack the ability to compete. The major vulnerable and excluded groups in the metropolis include women, children, and the aged, those who cannot read or write (Jankowska, 2009). 3.2.7 Slums, Illegal Settlements and Unauthorized Structures and Dwellers Urbanization in Accra is at an annual rate of 4.2% and still increasing with increasing rural- urban drift (World Bank, 2007). This phenomenon has resulted in the springing up of slums due to the gross inadequacy in low-income housing for the increasing population (UN- HABITAT, 2011). According to the UN-HABITAT (2011), Ga-Mashie communities have increasingly become a big slum with characteristics described in Table 3.3. About 90% of the populace of the slum dwellers falls within the low to very low-income brackets (AMA, 2010). There are about Twenty-Nine (29) slum communities in Accra, University of Ghana http://ugspace.ug.edu.gh 50 including, Ga Mashie, Chorkor, Gbegbeyese, Mamobi, Sabon Zongo, Nima, Alajo, Ayidiki, Akweteman, Avenor , Abuja, Sodom and Gomorrah (Old Fadama) and Babylon (GAMADA Factsheet, 2008). Table 3.3: Characteristics of Ga-Mashie Community Indicator Characteristics of Ga-Mashie Slum population 100,342 persons Total slum population (disaggregated index Female: 51.8%; Male: 48.2% Land area cover by slum 90.0 hectares Number of dwellings 1794 Population density 1103 Average household size 5.5 person Room occupancy 10.6 person per room Proportion of permanent dwellings (cement block or burnt bricks wall, iron sheet or other permanent roof) 66.6% Proportion of semi-permanent dwellings (mud/pole walls, iron sheet roofs) 32.35% Proportion of temporally dwellings (mud/poles wooden walls 1.1% Availability of period potable water on a compound 27.4% Availability of potable water not on a compound 70.4% Distance to potable water (if not on a compound) Max 50metres Proportion of households depending on sources other than potable water (water vendors, wells) 2.2% Availability of electricity of connections in dwelling 100% Average monthly household income GHc 299.55 (USD 78.83) Source: AMA, 2011 There are a growing number of unauthorised structures throughout the Metropolis. To solve this challenge, the Assembly undertakes demolition exercises to eliminate or reduce these University of Ghana http://ugspace.ug.edu.gh 51 unauthorised structures (AMA, 2010) yet there is still the need to support and invest in the upgrading and transformation of some of these settlements. 3.2.8 Disaster Zones Densu River Catchment and Sakumo Lagoon constitute the largest of all the four coastal basins within the Metropolitan Area (AMA, 2010). Flooding is common along the 8km Densu River below the Weija dam whenever there is overtopping or deliberate release of water over the spillway (Sam, 2009). Flooding is also prevalent in the Dansoman area and along the Lafa stream where it crosses the Winneba and motorway extension roads. Few of the drainage channels in the catchments are constructed. As a result, there is heavy erosion of drainage channels - many of which flow down existing tracks and roads. Access to this area is often cut off and roads become inaccessible during heavy rains (GoG/MWRWH, 2007). Another basin is the Korle - Chemu Catchment which is bounded on the west by a line, running roughly south to north from Gbegbeyese, Dansoman, Kwashieman, and New Achimota; on the north by the Achimota -Legon axis; and to the east by East Legon, the Airport and Ridge (AMA, 2010). The principal streams that drain the catchments are the Odaw River and its tributaries, the Nima, Onyasia, Dakobi and Ado. The principal outlet for water in this catchment is the Korle Lagoon. The Korle-Chemu catchment contains the major urbanised areas of Accra. Many of the drainage channels are poorly developed and maintained. Erosion and siltation of drains is a serious problem. In low-lying areas, flooding is a serious problem, with many houses been inundated by floodwater during and after heavy rains (UN-HABITAT, 2011). University of Ghana http://ugspace.ug.edu.gh 52 In low-lying areas near the Accra Academy in Kaneshie, rapid run-off from Asaredanho overflows into the Bank of Ghana flats because the inlet to the Kaneshie drains is inadequately designed (Karley, 2009). Similar problems occur in the industrial land cemetery area around the Obetsebi Lamptey Circle where the interceptor drain and gullies are inadequate. There are many other areas, like Nima, Dzorwulu, Darkuman and Alajo, which have localised flooding problems caused by inadequate drainage channels and the flat nature of the land terrain. In addition, the Kpeshie Catchment which is bounded on the east by the Military Academy at Teshie and on the north by a line south of Madina and Ajirignano (AMA, 2010; UN- HABITAT, 2011). It covers the eastern part of Accra, Ridge, Cantonments, Osu, Labadi and Burma camp areas. Streams in the catchment generally flow north to south, emptying directly into the principal outlet to the sea at Kpeshie Lagoon or the small Korle Lagoon at Osu. Drainage in the La Township remains inadequate and many waterlogged areas become flooded with light rains. Other areas liable to flooding in the Metropolis are as follows: i. Panbros Salt Ponds ii. Dansoman - Mpoase - South Odorkor corridor iii. Dansoman - Sukura - Chorkor corridor iv. Mataheko - Abossey Okai - Korle Lagoon corridor v. Odaw - Dzorwulu - Awudome - Industrial Areas System vi. Darkuman - North Kaneshie 3.2.9 Waste Management The city generates about two thousand tonnes (2000) of garbage daily, of which the Assembly is able to collect one thousand five hundred tonnes (1,500) daily based on their University of Ghana http://ugspace.ug.edu.gh 53 existing capacity (AMA, 2010). According to AMA (2011) the huge backlog is reflected in choked drains, overflowing garbage heaps, littered pavements, etc. Sources and forms of waste generated in the Metropolis include domestic/households, industrial/commercial, markets, schools and hospitals. Also, 3% of plastic recycling by the informal sector, 80% of metals recycling by the informal sector, organic, paper, plastic, glass, metals, textiles, inert and materials also constitute wastes generated in the Metropolis (AMA, 2011). The collection of solid waste is carried out by barrels and compactor roll on containers, and manual handling from pilling at collection points or by request. Waste collection is a municipal responsibility, which largely is contracted to different private companies (UNEP, 2011). The Assembly also implements Fee-Based Solid Waste Collection System (based on the Polluter Pay Principle) alongside other strategies such as waste-to-energy project and recycling, creation of more dumpsites, mass gutter cleaning exercise to improve the waste management practices in the Metropolis (AMA and UN-HABITAT, 2011). 3.3 Study Design The study made use of mixed method (both quantitative and qualitative) with the method of triangulation of data gathering, analysis and interpretation (Denzin, 2006). The choice of this method was for the purposes of checking or testing consistency or validity of the findings obtained (quantitative); and using one method to clarify, explain, elaborate and justify the other; and fostered the emergence of more research questions (qualitative) (Greene et al., 1997). Different from the concurrent approach was a sequential approach, which was applied to the study during the data gathering process (see 3.3.2). This made it easier for the researcher to collect data using the questionnaire and Focus Group Discussion (FGD). University of Ghana http://ugspace.ug.edu.gh 54 The study also made use of a case study approach where communities were selected and primary data were gathered on knowledge and perceptions regarding changes in climate, effects, and vulnerability and residents’ adaptation strategies. 3.3.1 Types and Sources of Data Two main sources of data including primary and secondary sources were adopted. The secondary data was gathered from books, book-chapters, journals, articles, working papers, newspapers, published and unpublished conference papers. These were pre-existing literature, which serve as secondary sources of information to the study. As part of the secondary data, climate data was gathered from Ghana Meteorological Agency (GMet), specifically from Kotoka International Airport Station. Data included monthly and annual average rainfall and temperature records. The data covered 1961 to 2000, and was used to complement knowledge and perceptions expressed by people on changes experienced in climate conditions. In addition, socio–economic and physical demographic characteristics data was gathered from Accra Metropolitan Assembly and Ga-Mashie Development Agency (GAMADA). This included the profile of AMA and other available documents from the District. The disaster profile of AMA and beyond was also available at the AMA, NADMO and Ghana Hydrological Department. These included the recent disasters especially flood records, extent of damage and communities affected in the City. Primary data based on knowledge and perceptions of changes in climate conditions of the area including rainfall, temperature seasonal variation or fluctuations, flood disasters, vulnerable conditions of the people and adaptation measures were gathered. Concerning the background information of the respondents, data was gathered on age, marital and University of Ghana http://ugspace.ug.edu.gh 55 educational status of the respondents. Others included income levels and occupancy ratios to determine people’s vulnerability due to housing conditions. The perception and experience of flood disaster was interesting, as respondents expressed their views on the flooding situation in the city. At the same time, data on how communities adapt to the disaster situations was also gathered. 3.3.2 Data Gathering Methods The primary data was collected using questionnaires, structured interviews (in-depth interview) and focus group discussions (FGD). The questionnaire captured most of the quantitative data from the field, whilst the FGD and in-depth interviews were used to capture most of the qualitative data from the field. Moreover, the questionnaire was designed mainly for those who could read and write, whilst the FGD and in-depth interviews were designed for those who could not read and write but none-the-less have appreciable knowledge to share on flood vulnerability and adaptation strategies. Wherever possible, the researcher assisted interviewees to provide information required by the questionnaires. The researcher made observation to situations difficult to account for by the respondents on the questionnaires and focus group discussion. . This included the housing conditions, waste management, etc. Again, many participatory methods (FGD, In- depth interviews) were employed as the work sought to understand community-based vulnerability adaptation strategies. To this effect, additional data was gathered sufficient to buttress the quantitative data gathered by questionnaires. Under the FGD, the following methods were used: community historical timeline, seasonal calendar, and community ranking of hazards, while the discussion was going on among the various groups in the communities (see Sub-section 3.3.4). These sub-methods enhanced a better understanding of climate and flood hazard and adaptation strategies and assets of the University of Ghana http://ugspace.ug.edu.gh 56 communities in the study area, changes they perceived, their capacity to cope with the flood, their needs, as well as the role of other institutions in coping and adapting to floods. 3.3.3 Sampling Design The population size of the respective communities was used to generate the total sample. The sampling was done using both probability and non-probability methods of sampling. Using the simple random sampling technique and based on the population sizes, a sample (S) was drawn using the formula [N (N)/N+1]*(a) 2, where N= Total population, a=margin of error, 1=a constant (Basil and Jones, 2009). With a total population size of the selected communities 347, 7799 as indicated in Table 3.4 and margin of error 0.05 and the sample size was determined as follows: S= [N (N)/N+1]*(a) 2 = [347799(347799)/ 347799+1]*(0.05)2 = (120964144401/347800)*(0.0025) = (347798.00)*(0.0025) =434 Stratified or strata probability sampling method was used purposively to select the final sample from the six communities, Nima, Agbogbloshie, James Town, Ursher Town, Korle Dudor and Cantonments. The strata probability method was used as a result of the heterogeneous population distribution among selected communities. In the calculation of sample for a strata sample of the selected communities, Sample for a Strata = Population of Strata * Total Sample Size/Total Population  Nima= 143434*434/347799 = 179  Agbogloshie=19797* 434/347799 = 24  James Town=38209*434/347799 = 47  Korley Dudor =64300*434/347799 = 80 University of Ghana http://ugspace.ug.edu.gh 57  Ursher Town=62133*434/347799 = 77  Cantonments=22104*434/347799 = 27 Table 3.4: Sample Distributions by Community Communities Total Population Sample Nima 143434 179 Agbogbloshie 19,797 24 James Town 38,209 47 Korle-Dudor 64,300 80 Ursher Town 62,133 77 Cantonments 22,104 27 Institutions 6 Total 347799 440 Source: Field Work, 2013. The systematic random sampling method was employed to select the households. The purposive sampling method was used to select household heads for the interview. Purposive sampling method was used because in certain cases, household members had to decide who should be the head. In addition, 6 respondents were purposively selected from the institutions including Ghana Meteorological Agency, Ghana Health Service, AMA, NADMO, Fire Service and Ga Mashie Development Agency (GAMADA). Table 3.6: Sample Distributions of the Communities based on the Data Gathering Tools Selected Communities Questionnaires Focus Group Discussion In-Depth Interview Grand Total Nima 110 24 45 Agbogbloshie 16 8 James Town 21 16 10 Korle Dudor 53 14 13 Ursher Town 36 16 25 Cantonments 22 0 5 Total 242 86 106 434 Source: Field Work, 2013 University of Ghana http://ugspace.ug.edu.gh 58 The Tables 3.4 and 3.5 show distribution of sample by communities and data gathering methods from respective populations. The units of analysis of the study included household heads from the selected communities together with representatives from institutions. 3.3.4 Data Analysis Data gathered through close-ended questionnaire were edited, coded and entered into Statistical Package for the Social Sciences (SPSS) version 20. The open-ended questionnaires were re-grouped, edited, coded and also entered into the SPSS after which outputs were generated in the form of frequency distribution tables, pie chart and graphs. Microsoft Excel 2007 was used to edit tables, charts and lines appropriately. The SPSS includes a statistical tool called Classification Discriminant Analysis (CDA), introduced by Elo and Kyngäs, 2007 to show the differences in two or more situations. In this study, it was used to determine spatial variation of flood disaster under Chapter Five (5), based on people’s knowledge and perception. It was used to compare the flood disaster in the selected communities based on the frequency, fatality and extent of damage of flood disaster. The communities were dependent variables while frequency, fatality and damage indicators were the independent variables. The main assumption of the CDA is that the dependent variables must be different from each other. Because any similarity or correlation between them would not satisfy the result of the CDA. The CDA statistic employs other statistics to enhance the validity and to give further explanations to the results of the CDA. These statistics are briefly explained below and used in chapter 5. i. Group statistical table showed the mean difference between the independent variables. ii. Test of Equality of Group Means further test for differences among the independent variables: Wilk Lambda between 0 and 1, F-test more than 1 and significance less University of Ghana http://ugspace.ug.edu.gh 59 than .05 show higher significant variation or difference among the independent variables. iii. Pooled within Group Matrices statistic explain whether the independent variables are correlated. Once they are correlated, their differences would not be felt hence, their ability to classify the communities in terms of the flood disaster would be weak. iv. Box M and log determinant. It is by this statistic that the differences among the selected communities would be realised. Box M would show that with low/no significant (e.g ≥ 0.05) means that there is less or no difference in the communities. F > 1 would imply there is differences in the communities in terms of the flood disaster. v. Canonical functions using eigenvalues and canonical correlation is a function analysing part of the dependent variables that can or cannot be explained by the independent variables. vi. Standardization was employed to shows which independent variable had highest discriminant score and portray the portions of the dependent variables that can or cannot be explained by the independent variables. 3.3.5 GIS Overlay Operation on Flood Zones Raster maps of the Sub-metros, communities and slums of Accra Metropolitan Area (AMA) were procured. Feature Classes of communities, slums and flood zones were created in ArcCatalog. The individual raster maps were geo-referenced with the following description: i. Feature Class: AMABD new ii. Projected Coordinate System: Ghana Metre Grid iii. Projection: Traverse Mercator University of Ghana http://ugspace.ug.edu.gh 60 iv. Linear Unit: Meter v. Geographic Coordinate System: GCS Leigon vi. Datum: D Leigon vii. Prime Meridian; Greenwich viii. Angular Unit: Degree Figure 3.7 GIS Overlay Operations of Flood and Communities Source: Field Work, 2013. In ArcMap, the editor toolbar was activated and editing was carried out. Digitizing or editing was completed under different layers/feature classes which include Sub-metros, Communities and Slums. Where there were no available map data, GPS Coordinates were picked from the field. The corresponding attribute table of each layer were constructed and edited as well, to include fields of communities affected by flood, number of people affected. Procurement of raster map e.g. communities Procurement of raster map e.g. flood zones Geo-referencing Creation of Feature Class Attribute Table Development for communities, slums and floods Digitizing Symbology/ Map Creation Overlay of vector i.e. Communities, Slums and Floods University of Ghana http://ugspace.ug.edu.gh 61 Additionally, various map outputs were created for display purposes and symbols were applied to generate an overlay of communities, flood and slums in the of AMA. The procedure is summarised in the Figure 3.7. 3.4 Chapter Summary The chapter examined the profile of the study area and the study design. The first section included discussion on history and location of Accra, the geology, climate and vegetation of Accra. Also, the population and migration trends were examined, housing, poverty, slums, illegal settlements and unauthorized structures and dwellers. Further, the section also looked at the disaster zones of Accra and the waste management practices in the study area. The second section discussed the study design. It examined sources of data, data gathering tools and sampling design. Method of data analysis was discussed including the Classification Discriminant Analysis tool for measuring the level of difference in flood disaster in Accra and description of the procedures involved in the GIS overlay operations on flood and community distribution. University of Ghana http://ugspace.ug.edu.gh 62 CHAPTER FOUR ANALYSIS OF CLIMATE AND NON-CLIMATE FACTORS TO FLOOD DISASTER 4.1 Introduction This chapter presents analysis of data gathered from the selected study communities in the Accra Metropolis. The analyses are based on the background of the respondents; the knowledge and perception of respondents to flood disasters and its causes. This was done to relate knowledge and perception of people with the pre-existing concepts and theories of flood disaster. Graphs, charts, box and plates have been combined to give clearer interpretation and understanding of the issues analysed. 4.2 Background of the Respondents. Data was gathered on age and sex in order to explore the social and demographic characteristic of the respondents. Table 4.1 provides a summary of the age and sex distribution of respondents. The respondents (83) were within the ages of 35-44, followed by 73 respondents who were at least 45 years of age (See Table 4.1). The age group (15- 24) had the lowest frequency of 36 respondents. Respondents within the age group of 35- 44, 45 and above constituted the majority because most of them were old and household heads who were born, bread and had lived several years in the community and could bear witness to the long term occurrence of climate change and flood disasters in the city. Most of the respondents within the older age group retired from active service, were resting at home, and were more available to respond to the questionnaires. It was observed that many of the young men and women were not at home at the time of the data collection; some went to school, market and were engaged in other businesses. Most of the elderly people were University of Ghana http://ugspace.ug.edu.gh 63 found in Ga-Mashie, including Ursher Town and James Town because these are indigenous communities where the elderly hardly travel outside their communities. Table 4.1: Background of Respondents Background Communities/Frequencies Age Group Nima James Town Ursher Town Korle Dudor Cantonment Total 15-24 17 0 8 9 2 36 25-34 35 9 2 4 0 50 35-44 28 2 17 27 9 83 45+ 30 10 9 13 11 73 Sub-Total 110 21 36 53 22 242 Sex Male 66 9 14 33 17 139 Female 44 12 22 20 5 103 Sub-Total 110 21 36 53 22 242 Source: Field Work, 2013. Out of the sample population of 242,139 were males and 103 females as indicated in Table 4.1. The females in the study communities constituted the majority as this reflects the national population trend (GSS, 2010). It was discovered that many young men travel to other communities for more opportunities leaving behind the women. This shows that in times of flood, more females would be more exposed as compared to their male counterparts in the Accra Metropolis. Nima however showed predominant male migrant population. This pattern was difficult to justify. It could be linked to the high number of migrants from the northern part of Ghana who are mainly males. Further studies are needed to explore this variation. 4.3 Knowledge and Perceptions on Disaster Flood has been perennial over the last two decades, posing various challenges to the socioeconomic development of Ghana. Exploring knowledge and perception of the respondents about flooding was very important. The study revealed varied views on flood University of Ghana http://ugspace.ug.edu.gh 64 disasters and the implications of those floods. The majority of respondents (88%) were aware of all forms of disaster particularly floods that occur in Accra as shown in Table 4.2. Only a few people (13%) were not aware of any events because of the short duration they have been resident in Accra. This group of people were mostly found in the Nima Township and most had lived less than six (6) months in the Community. Table 4.2: Respondents’ Knowledge on Disasters Awareness Communities, Frequency, Percentage (%) Nima James Ursher Kol.Dudor Cant ‘ment Total Percentage Yes 83 21 36 51 22 213 88 No 17 0 0 0 0 17 7 No Idea 10 0 0 2 0 12 5 Total 110 21 36 53 22 242 100 Source: Field Work (2013) Respondents who indicated they were aware of number disasters indicated this was through their personal or experiences and what they had heard. Amongst the disasters listed, floods were listed (208 / 213) because of their frequency and devastating nature to property in parts of the Accra Metropolis. Discussion with the respondents found that flood occurs almost every year with lives lost and damage to property. Respondents mentioned 1995, 1997, 2010, and 2011 as the years in which they experienced serious flood disasters in Accra. The dominance of flood disaster among the list is not strange because the previous studies by Afedzi (2009) revealed that flood has become a perennial event mostly in the rainy seasons in Ghana. The second dominant disaster was fire (187 of 213 respondents), followed by cholera (158 of 213 respondents), and road accidents (98 of 213 respondents). Others included building collapse, (93 of 213 respondents) among others. Around Agbogbloshie, fires occur almost every day. This was disclosed during the in-depth interview and FGD held in these areas. University of Ghana http://ugspace.ug.edu.gh 65 Respondents stated that fire has been a disrupting event because of the way it destroys people’s property. In a FGD, one respondent shared the view that: “During Harmattan season, we could hear fire events in most communities in Accra” (FDG, March 2013). Table 4.3: Respondents’ Observation of Disasters Disasters Frequency Floods 208/213 Fires 187/213 Cholera Outbreak 158/213 Road accidents 98/213 Others 39/213 Source: Field Work, 2013 The table portray a fair knowledge on disasters in the Metropolis. Further studies are needed on other disasters including fires, cholera outbreak, accidents. It was gathered that cholera outbreak occurs during rainy season after flood inundates many houses especially in the slum areas. 4.4 Climate Change and Disasters Many authors attempt to relate disasters to climate change. Climate has direct are indirect bearing on the nature of occurrence of some disasters hence changes in the climate also affect the level of occurrence of those disasters. In the Table 4.4, 145 of total respondents indicated that flood had high association with changes in climate conditions. A resident in Korley Dudor was of the view: “Flood is our biggest problem........ Whenever it rains flood occurs. Water enters our houses. Some places become muddy. Our gutters are small and chocked” (Focus Group Discussion, 2013) University of Ghana http://ugspace.ug.edu.gh 66 Table 4.4: Disasters with Strongest Relation to Climate Conditions Hazards/Disasters Communities, Frequency Nima James Ursher Kol.Dudor Cantonment Total Floods 56 16 21 33 19 145 Cholera Outbreak/other disease 12 5 10 8 3 38 Fires 15 0 5 4 0 24 Road Accidents 0 0 6 6 Total 83 21 36 51 22 213 Source: Field Work, 2013. Responses on cholera was 38 out of 213. Many of them found it difficult to link up cholera to climate conditions directly. Respondents stated that cholera is common when run-off waters inundate their homes. Interviewing respondents further on cholera and its association with climate conditions, a resident at Agbogbloshie indicated as follows: “Cholera usually occurs when it is a rainy season. Cholera occurs where flood water enters people’s houses and outside. Some of these areas (pointing to other homes) experienced cholera last year............” (In-depth Interview, March 2013). The relationship between climate conditions and fire disaster is quite weak. Only 24 respondents relate fire to climate outbreaks in the city. Meanwhile domestic fire has become an everyday occurrence in mostly slum communities especially in Agbogbloshie and Old Fadama (FGD, 2013). This shows that both cholera and fire outbreak are indirect effects of climate change in Accra Metropolis. 4.5 Causes of Flood Disaster in Accra Flood is an annual phenomenon in the Metropolis affecting properties, houses and businesses of victims in general. Floods result from both climate and non-climate factors. University of Ghana http://ugspace.ug.edu.gh 67 Respondent views were generally explored to identify causes of flood. The results were summarised in Figure 4.1. In figure 4.1, 70 of 145 of the respondents indicated improper waste disposal as a main cause of floods in Accra. Ineffective Settlement planning and implementation (27 responses), drainage problem (24 responses), rainfall (13 responses) and attitudinal behaviour (11 responses) all emerged. Figure 4.1: Reasons for Flood Occurrences in Accra Source: Field Work, 2013. Findings at the focus group discussions augment the result that some people dump rubbish into gutters and the main river at night whether or not there are dustbins. It was explained 0 10 20 30 40 50 60 70 N u m b b er F re q u en cy o f R es p o n se s Communities Causes of Floods in Accra There are no enough drainage in those communities Disposal of sewerage into the gutters Rainfall Settlement planning and implementation Attitudinal behavior University of Ghana http://ugspace.ug.edu.gh 68 that dustbins along the street are not many. Some said that waste collectors could be well paid for the waste collection service but they do not come regularly. At Nima, a retired public servant was interviewed about his experience and observations of waste practices. The resident was named Musa. He was a 68 year old retired Public Servant at the time of the interview. He was sitting in an armchair in front of his gate. It was 11:16 am and he was relaxing and waiting to go to the Mosque at 1:00pm. He originated from Bawku but is now permanently settled in Nima, Accra. He was asked to describe his observations and experiences on waste management in Accra. His experiences are summarised in Box 4.1. Box 4.1: Resident’s Story Source: Fieldwork, 2013 Plate 4.1 to 4.6 were used to illustrate the level of waste management, choked gutters, building on waterways and some few dredging works in Accra. Plate 4.1 specifically shows Life in Accra has changed over his 32-year stay in Nima. Nima was a nice place that attracted every one coming to Accra. The environment was neat and buildings well arranged. Those days my household used to have dustbin at home and every morning my wife, later on my children will sweep the house and rooms, gather the rubbish into the dustbin and carry them to Bola (rubbish dump). We were few in the community and people used to generate small amount of waste. Now population has increased, more waste has been generated and the “Bola” points have been closed down except some small one down there. I no longer allow my children to take bola (rubbish) out. Other people just leave their bola around carelessly and rain carry them away, but for me I can’t complain because some people do not even know where to sleep or what to eat, let alone to provide dust bin themselves. University of Ghana http://ugspace.ug.edu.gh 69 the initial deposition of waste into drains of Accra during dry season. This will accumulate and gradually chock gutter. There are many gutters in the city with this outlook. One would wonder if people are doing this with conscience. I personally met one Taxi Driver dumping in this gutter as soon as he came to park at the station. Plate 4.1: Deposition of Wastes in Drains at Ga-Mashie Source: Field Work, 2013 Passengers left waste into the taxi, alighted and went off. Leaving waste I a car is permitted in Ghana so that people would not dump them on the road. The Taxi Driver collected the waste and dumped them into nearby drain. I asked him to pick it up with all humility and he said to me: Gentleman, mind your business. I cannot carry this rubbish in the car. When we dump it into the gutter here, people come to collect them. People from the Assemblies University of Ghana http://ugspace.ug.edu.gh 70 come for them. We are paying them taxes so they must not give excuses why waste we dump we dump in this gutter should remain there (Resident, 2013) Plate 4.2 Building on Waterways and Chocked Gutters in Agbogbloshie Source: Field Work, 2013 Many plans have been developed in the city to provide effective land use including building, and development of drainage in Accra. The increase in the population and pressure on lands lead people to build in areas earmarked for no habitation. Every year, erection of wooden and illegal structures in communities including, Old Fadama, Agbogbloshie and Ga-Mashie is on the increase. Building near the major streams University of Ghana http://ugspace.ug.edu.gh 71 such as Odaw River, Korley Lagoon is regulated although a few structures were still observed along the river bodies. The general attitude of people in the Accra Metropolis remain a major obstacle to managing urban waste. People’s behaviour and attitudes towards waste disposal was cited to be the third strongest reason for the increasing flood impacts in Accra. It includes non-adherence to building regulations and dumping in open gutters and drains among others. The behavioural aspect cut across all other challenges. Plate 4.3 Chocked Gutters in Old Fadama (A and B) Source: Field Work, 2013 The behaviour of residents influenced the level at which people dump in gutters, building in unauthorised places including building in waterways. Poor behaviour and attitudes is a serious problem because even people who are very educated in both formal and informal A B University of Ghana http://ugspace.ug.edu.gh 72 ways are also part of the problem though efforts are being made to reducing dumping in the city. Plate 4.4: Narrow and silted Drain at James Town Source: Field Work, 2013 The drainage situation forms parts of the reason for the perennial flood disasters in the city. The drainage system are in a deplorable state. Limited drainage was observed in many communities with the available ones been either too narrow, broken or chocked with solid waste. It was observed that drainage is limited and those available either too narrow, broken or chocked with solid waste. University of Ghana http://ugspace.ug.edu.gh 73 Drainage and the waste situation constitute very serious problems in Accra. It was revealed that a much waste goes into the gutters because most of the drains are open and uncovered. Plate 4.5: Waste Collected and left at the Bank of Drain in Ursher Town Source: Field Work, 2013 Plate 4.6: Dredging Work on Choked Odaw River Source: Field Work, 2013 Waste dumbed into the gutters becomes difficult to remove, as it becomes stacked and congested. Worst of it all, waste collectors remove the waste and leave it at the bank of the drains as shown by Plate 4.3. Many respondents, believed that poor land use and planning University of Ghana http://ugspace.ug.edu.gh 74 and poor engineering methods have exacerbated the problem even though they identified rainfall as serious factor to urban flooding. An interview with the Public Relations Officer of AMA on the cause of perennial floods in Ghana. The Officer worked in the Assembly for over 10 years. He was very much conversant with the efforts and challenges of the Assembly. His knowledge was summarised in Box 4.2 Box 4.2: Public Relation Officer’s Story Source: Fieldwork, 2013 Source: Field Work, 2013 Rainfall contributes largely to floods in Accra. There is a high run-off in Accra during rainy season resulting flood. In addition to the responses in Figure 4.1, a resident in James Town who works at Kwame Nkrumah Circle phone shop, Mr. Samuel Armah shared his experience on the pattern of rainfall below: There were number of comprehensive plans for Accra to promote effective land use and development. The recent one was 1993 to 1998. These plans earmarked areas for settlement residential areas, industrial, commercial and administrative areas. Areas not to be inhabited includes the buffer zones along the drains and waterlogs, and recreational areas for evacuation during floods and other dis asters. These spaces were provided so when there is flood in the buffers nobody will lose properties or lives. After some time these areas were encroached, some were sold and distorted. Unstable governance, lack of proper monitoring and enforcement of plans led to improper execution of the comprehensive plans we had for Accra. It is rather unfortunate! Today, there are thousands of people leaving in Alajo, which was marshy and preserved as storage of excess run-off. All buffers along the six drains in the GAMA including Odaw River were inhabited. Development of drainages and adequate measures for enforcement of bye-laws were limited. We should not be surprised why floods inundate these communities and areas at the peak of rainy season. We need to re-engineer the city. University of Ghana http://ugspace.ug.edu.gh 75 Rainfall has become quite different in recent times. In a short, rain could fall heavily and stopp. It could happen like that until many places are flooded here (referring to Kwame Nkrumah Circle). Before one realised, there is flood all over. Where to pass becomes a problem. Water could enter our shops, station and everywhere here (In-depth- Interview, 2013) Section 4.6 examined the changes in rainfall conditions and rainfall intensity in Accra and how these are related to the incidence of flooding. 4.6 Changes in Rainfall and other Climatic Conditions The resident’s awareness about the changes in rainfall and other climatic conditions and their ramification with flood disaster. Results obtained indicated that 70 percent of the respondents were aware of some changes in climatic pattern. Twenty-four percent (24%) of the respondents admitted they had not observed any changes and 6 percent were not certain, perhaps because they could not see marked differences in rainfall. Those who agreed that there were marked changes, did so because of the weather related activities they engaged in. For example, those at James and Ursher Town were predominantly fishermen and so the changes in the sea levels, temperature and wind behaviour provided some clue to changing weather patterns. The basis of the opinions on the changes in the climate was due to the length of time these respondents had been part of the local community, in many cases for at least 30 years. Very few respondents did not observe any change in weather conditions and if they did it was because of their short period of residing in the community. For example, in Cantonments and Nima, residents who did not notice any change in climate conditions were mostly migrants who have not stayed in the community for long. Generally, most respondents across all five (5) communities indicated that there were changes in the climate. University of Ghana http://ugspace.ug.edu.gh 76 Thirty three (33) percent of respondents indicated that there were significant changes in the wind. Usually, a strong wing comes with rainfall in the city. Temperature constituted (32%), rainfall (30%) and others represented by 5 percent. The other elements mentioned included sunshine and heat but with a small percentage. Table 4.5 Changes and Variability Observed in the Climate by Respondents Personal Observation Communities, Frequency Nima James Ursher Kol.Dudor Cantonment Total Total Percentage (%) Yes 59 21 34 46 13 169 70 No 39 0 0 7 9 58 24 No idea 12 0 2 0 0 15 6 Total 110 21 36 53 22 242 100 Source: Field Work (2013) The diverse views expressed by the respondents on weather elements reflected not only the dominance of any particular element at the time of data collection but also related to activities they perform. The extreme events that come with the change in the climate conditions is also a factor. Their choices were largely influenced on how much knowledge they had on one element over the other. The study confirmed that strong winds prevalent at James Town, Korley Dudor and Ursher Town due to their proximity to the Gulf of Guinea coast (sea). The study examined knowledge and perception about the nature of changes in the climatic conditions. The changes were categorised under three major weather elements including rainfall, temperature and wind. In Table 4.6, 16 percent of the respondents said annual University of Ghana http://ugspace.ug.edu.gh 77 rainfall has decreased, 32 percent indicated that rainfall pattern is unpredictable in recent times, 45 percent said rainfall is more intensive and destructive in modern times. Table 4.6: Observed Changes and Variability on Climate Conditions Climate Conditions Frequency Percentage (%) Rainfall Unpredictable 77 32 High Intensity 109 45 Reduction in Annual Quantity 39 16 No much changes 17 7 Sub-Total 242 100 Temperature Generally high 169 70 Colder in wet season and extremely hotter in dried season 36 15 Increasing heat 24 10 No much changes 12 5 Sub-Total 242 100 Wind Increase in intensity/strength 80 33 Increase in magnitude (volume) 19 8 Becoming more destructive 113 47 Did not see much change 30 12 Sub-Total 242 100 Source: Field Work, 2013 Seventy percent of the respondents said temperature has been generally high, 15 percent said temperature is colder in wet season but extremely hot in the dry season. Twenty four (24) respondents accounting for 10% of total respondents admitted that heat in Accra has increased in recent years making living conditions uncomfortable. University of Ghana http://ugspace.ug.edu.gh 78 Respondents suggested that there are changes in the wind pattern. Thirty three (33) percent of total respondents noticed that wind has increased in its intensity and strength over the years. Other respondents (47%) said wind has been more destructive in recent times. Eight percent of agreed to the fact that there has been an increase in the magnitude of wind. All responses provided indication that residents have fair knowledge about climate change, which can help decision in terms of adaptation and mitigation of climate related disaster in the city. Average monthly and annual rainfall and temperature data from Ghana Meteorological Agency (GMet, 2013) shows more scientific and reliable data to determine changes in the climate conditions. This was done to support the knowledge and perceptions of the interview respondents. In the first place, mean monthly rainfall was measured to determine the monthly variation from 1961-2000 year period. It showed a short variation and changes in the climate (over at least 30 years period). The monthly rainfall analysis was also carried out to explain the high run-off and the perennial floods in the June/July. The figure shows that rainfall starts in March/April and peaks in May, June and July, which is the major rainy season. During this time, there is a very high run-off generation leading serious flooding in Accra. In July and August, rainfall is very low and the period that is minor dry season between the major rainy season and the minor rainy season. The second rainfall season starts in September, reaches its peak in October, and is referred to as the minor rainy season. In some years, rain is very intensive and highly unpredictable. December through to March shows the longest period of dryness, where rainfall is scanty with no floods. With the changes and variability associated with climate in recent times, Accra sometimes experiences a short rain in the dry season. University of Ghana http://ugspace.ug.edu.gh 79 In some exceptional cases including October and November, 2011, there were some floods in Accra even though the season was almost into dry season. Usually such flood are very destructive because most people would not be in preparation for it. Figure 4.3: Average Monthly Rainfall over Four Decades in Accra Source: Ghana Meteorological Agency, 2013 In addition to the mothly rainfal pattern over various decades between 1961-2000, mean annual rainfall was also analysed. The trend shows a negative slope. The early and late 1960s, and middle 1990’s show relatively high values. The maximum recession was recorded in the 1980s. The rainfall shows a decline at rate of 10mm per year over the period which may be as a result of climatic changes. The Average Annual Rainfall (ANN) potrays a rise and fall in rainfall occurrence over the four decades in Accra (see Figure 4.4). In addition, the Average Annual Rainfall (Linear y = -1.605x + 76.345 R² = 0.01 0 20 40 60 80 100 120 140 160 180 200 220 1 2 3 4 5 6 7 8 9 10 11 12 R a in fa ll ( m m ) Month Average Monthly Rainfall in Accra Months in 1961- 1970 Months 1971- 1980 Months in 1981- 1990) Months in 1991- 2000 Mean of Months from 1961-2000 University of Ghana http://ugspace.ug.edu.gh 80 ANN) shows the overall decrease in rainfall over Accra. Further characteristic of the rainfall pattern between two-year blocks was compared. Figure 4.4: Average Annual Rainfall from 1961-2000 in Accra Source: Ghana Meteorological Agency (2013). Regarding temperature patterns, monthy and annual average records from 1961-2000 were analysed to have increased over the last year except in the middle of the year where it falls with a positive slope. From Figure 4.5, the average monthly temperature from November through January to April is significantly high with the peak in February. This is due to the dry season or the Harmttan period where there are fewer or no low clouds accompanied by heat through to February. It begins to deline from May to September with the lowest point in July/August. The pattern of temperature over the past 4 decades also remained almost the same, falling across all y = -7.7652x + 967.44 R² = 0.1724 0.0 200.0 400.0 600.0 800.0 1000.0 1200.0 1400.0 1600.0 R a in fa ll ( m m ) Year Average Annual Rainfall (ANN) (1961-2000) in Accra ANN Linear (ANN) University of Ghana http://ugspace.ug.edu.gh 81 decades from 1960 till the present time over Accra. The periods 1980s and 1990s showed levels appreciably above the 1960s and 1970s. Figure 4.5: Average Monthly Temperature over Four Decades in Accra Source: Ghana Meteorological Service Department, 2013 In Figure 4.6, the study analysed temperature per annum from 1961 to 2000. This was carried out to indicate the change in the temperature within the period. In the Figure 4.6, unlike the rainfall trend showed that temperature followed a positive trend. It shows that temperature was low during the 1960s relatively below 27 oC and picked up to about 28.2oC in the early 1970s. The figure later dropped in the 1970s but continued with a rising trend towards the early 1990s. The considerable long term increase in the temperature from late 1970s to 1980s is explained by the long term drought experienced during the period. During that period, there was 25.0 26.0 27.0 28.0 29.0 30.0 31.0 32.0 33.0 34.0 Te m p e ra tu re ( ⁰C ) Month Avearge MonthlyTemperature over Four Decades in Accra Average Max. 1961-1970) Average Max. (1971-1980) Average Max. (1981-1990) Average Max. (1991-2000) Linear (Average Max. 1961-1970)) University of Ghana http://ugspace.ug.edu.gh 82 drastic reduction in the amount of rain which was needed to cool the temperature of the earth. Figure 4.6: Average Annual Temperature from 1960-2000 in Accra Source: Ghana Meteorological Agency (GMet), 2013 The land was then exposed to extensive bush burning and deforestation which aggravated the rising temperature phenomenon. Mean Temperature equally shows a positive trend since 1961 with higher trend from 1980 indicating excess or surplus heat in the atmosphere leading to global warming. 4.7 Chapter Summary The chapter explained the relationship between flood and climate and non-climate factors in Accra. Results were presented under various sections including the background of respondents to determine respondents’ their knowledge and perceptions on flood disasters, climate change and non-climate factor causing the floods. The other sections included the knowledge and perceptions of changes in rainfall and other climatic conditions and the nature of these changes. The result of the study shows that the majority of the respondents y = 0.0374x + 26.493 R² = 0.5839 26.2 26.4 26.6 26.8 27.0 27.2 27.4 27.6 27.8 28.0 28.2 T em p er a tu re (° C ) Year Average Annual Temperature from 1960-2000 in Accra Ann Linear (Ann) University of Ghana http://ugspace.ug.edu.gh 83 were between the ages of 35 and above with more males than females. In relation to the climate change phenomenon, the majority of the respondents demonstrated good knowledge on climatic changes occurring over time. Specifically, temperature was indicated to have increased over the years as did wind which was observed to have increased in intensity and magnitude over the years. However, rainfall has shown a decreasing trend but with higher intensity and run-off in June/July (peak of major season), causing flood disaster. Residents were also assessed on their perceptions on disasters in their localities. In the Accra Metropolis, flooding was noted as the most frequent disaster with very close association with climate change. Specific perceived causes associated with disasters included unpredictable heavy rainfall and run-off, ineffective plan implementation including haphazard building structures, encroachment on low-lying areas, and poor waste management practices. University of Ghana http://ugspace.ug.edu.gh 84 CHAPTER FIVE SPATIAL DIFFERENCES AND ADAPTATION TO FLOOD DISASTER 5.1 Introduction Flood occurrence is a perennial phenomenon in urban communities attributed to changes in climatic conditions and human or anthropogenic factors. This chapter seeks to examine observed spatial differences in flood disaster and adaption strategies adopted. 5.2 Discriminant Analysis of Spatial Variation on Flood Disaster Flood incidence in Accra is not evenly distributed across space. Some communities are more prone to flooding than others as a result of their geographical locational differences. Figure 5.1 shows the map of all the major communities in Accra, highlighting major slums in the metropolis. The UN-HABITAT (2002 p.21) defined slum as “a contiguous settlement where the inhabitants are characterized as having inadequate housing and basic services. A slum is often not recognized and addressed by the public authorities as an integral or equal part of the city”. There are a number of communities in Accra that are classified as slums, with major ones such as Nima, Agbogbloshie, Old Fadama, Korley Dudor among others. Cantonments, Ridge, Kanda, and Asylum Down however are not classified as slums but as residential areas. Figure 5.2 shows the map (enlarged form attached in Annex D) of flood catchment areas in the Accra Metropolis. In-depth discussions held with respondents provided insights into the spatial variations of floods in Accra in terms of frequency of occurrence. Some of the communities noted to be prone to flooding included: Avenor, Gbegbeyese, Korle Dudor/Graphic Roads, Agbogbloshie/Galloway, Darkuman, Kwashieman, Awoshie, Santa University of Ghana http://ugspace.ug.edu.gh 85 Maria, Dansoman, Liberty and Tunga Down, Mukose, Avenor, Atico, Odawna, Adabraka (see list in Table 5.1). Figure 5.1: A Map of Slum Areas in the Metropolis Source: Field Work, 2013. Observation from the field found that almost all settlements in Accra have some segment of slum. Communities such as Cantonment, Ridge, East Legon, and Westland, which are all high residential areas, have some slum areas. These slums came into existence as result of the settlement of the indigenous people who still maintain their old structures. This is typical of the East Legon slums. Another reason is those who used to provide labour for construction works initially going on in those communities. There others who were also University of Ghana http://ugspace.ug.edu.gh 86 selling provisions and food in this communities. There all leaving there temporary initially but now became permanent residents (AMA, 2011). Table 5.1: Listing of Flood Disaster Zones (communities) over Accra Events Locations Flooding in Accra Korle Dudor/Graphic Roads, Agbogbloshie/Galloway, Darkuman, Kwashieman, Awoshie, Santa Maria, Dansoman, Liberty and Tunga Down, Mukose, Avenor, Atico, Odawna, Adabraka, Official Town, West Legon, Alogboshie, Kissema, West Legon, Nii Boi Town, Olengele Kronaa, Apenkwa, New Achimota. Kinaphama Down, Tesano(Adama), Christian Village, Osu Doku, Alajo North and South, Kotobabi North and South, Source: Field Work, 2013 Data on flood disasters in Accra has been gathered through different sources. Records on floods were accessed mostly from NADMO and newspaper reviews. This data was used to create a flood map for the Accra Metropolis (see sub-section 3.3.5 for detailed GIS procedure). The purpose of the map was to determine the spatial distribution and severity of urban floods in the study area. The map of the slum areas and the flood prone zones were overlaid to spatially locate communities with flood incidences. The final map (figure 5.3) (enlarged University of Ghana http://ugspace.ug.edu.gh 87 map in Annex D) reveals that floods are unevenly distributed in AMA with slum areas being the most vulnerable. Figure 5.2: A Map of combined Flood Distribution in Accra Source: Field Work, 2013. Reference is made to areas around Mamobi, Kotobabi, Alajo, Westward to Alajo, Old Tesano. Kaneshie, Kokomlemle, and Abossey Okai, which occupy the central portion of Accra. To the extreme south, along the coast from the west to east, such areas as Korley Gonno, Gbegbeyese, Korley Dudor, Agbogbloshie, Accra Central and James/Ursher Town represented vulnerable communities prone to flooding. Others include Santa Maria, University of Ghana http://ugspace.ug.edu.gh 88 Dansoman, Liberty and Tunga Down, Mukase, Avenor, Atico, Odawna, Adabraka and Official Town. Figure 5.3: A Map of combined Flood and Slums. Source: Field Work, 2013 The spatial distribution of floods in Accra’s slum and low class residential areas, which have dense and packed populations, makes such communities more vulnerable to floods. Notwithstanding this, middle to high class residential areas including Cantonment, Burma Camp, up to Airport Residential and down to Ridge also remain prone to floods (see figure 5.3) (Enlarged size in Annex D). The only difference is that the frequency of flood occurrence, loss of life and property among other negative effects remain highest among slum and low residential areas. University of Ghana http://ugspace.ug.edu.gh 89 The study further sought to assess resident’s knowledge and perceptions on the spatial distribution of flood disasters in Accra. The results of the knowledge and perception of the respondents were analysed using Classification Discriminant Analysis, which was explained in Chapter 3. The flood indicators used included the frequency of occurrence of flood disaster, property damage and fatality. The ratings provided by residents of all the five (5) communities indicated that despite the difficulties of not taking records of flood events, 89 out of 208 respondents indicated that there are community variations in the occurrence and effects of flood events in terms of frequency, damage and fatalities. Notwithstanding, the variations shown in Table 5.2 shows that respondents rated frequency of occurrence as the highest observed variation, followed by property damage and fatality among the indicators (see table 5.2). Table 5.2: Knowledge and Perception of Flood Disaster Variations Independent Variables Nima James Town Ursher Town Korley Dudor Cantonment Total Frequency of Occurrence 11 6 13 19 4 53 Property Damage 7 2 1 12 2 24 Fatality to lives 2 0 0 7 3 12 Total 20 8 14 38 9 89 Source: Field Work, 2013 The study also tested for the extent of difference in community floods using the indicators of frequency of occurrence, property damage and fatality. The three indicators were used as independent variables against all five-study communities as dependent variables. The aim of the analysis was to determine whether these predictor variables would discriminate one University of Ghana http://ugspace.ug.edu.gh 90 community from the other using the Classification Discrimant Analysis with F-test and significance level. Table 5.3 Group Statistics (means and standard deviations) Source: Field Work, 2013. The classification sought to reveal the most important variable(s) by inspecting the group means and standard deviations as provided in sub-section 3.3. According to Table 5.4, there is a significant variation among the independent variable (fatality of disaster and property damage). The variations are shown by Wilks' Lambda values. Inferred from Section 3.3.5 between 0 and 1, values of Lambda are significant to proof the difference between the dependent variables. Table 5.4 shows that frequency of flood disaster is 0.145, Fatality is 0.154 and Property Damage is zero. 127. All three indicators were tested at a 0.000 significance level which means that, the value of P < 0.05 significance test level. This statistics revealed a significant difference in the three indicators (frequency of disaster, fatality of disasters and damage) and the respondent’s community of residence with F = (123.93, 115.74 and 144.52). The Pooled Within-Group Matrices (Table 5.5) was used to establish the inter-correlations. The low correlation indicates that a high difference between the predictors (Frequency, Residential Areas Mean Std. Deviation Total Frequency of the Disaster 3.4944 0.85445 Fatality of the Disaster 3.5393 0.73941 Property Damage 3.4831 1.15908 University of Ghana http://ugspace.ug.edu.gh 91 fatality and property damage) once their predicting behaviour on the dependent variable remain different. Table 5.4 Tests of Equality of Group Means Tests of Equality of Group Means Independent Variables Wilks' Lambda F df1 df2 Sig. Frequency of the Disaster .145 123.932 4 84 .000 Fatality of the Disaster .154 115.740 4 84 .000 Property Damage .127 144.517 4 84 .000 Source: Field Work, 2013. Table 5.5: Pooled Within-Groups Matrices Pooled Within-Groups Matrices Independent Variables Frequency of the Disaster Fatality of the Disaster Property Damage Frequency of the Disaster 1.000 .727 .269 Fatality of the Disaster .727 1.000 .214 Property Damage .269 .214 1.000 Source: Field Work, 2013. The log determinants appear different but with no large variation and Box M is 176.474 with F = 11.615 which are significant at p <0.03 (Table 5.6). This situation is because of the difference between the means of the independent variables. Once the significant figure is University of Ghana http://ugspace.ug.edu.gh 92 less than 0.05 with F value more than one, the conclusion is that disaster occurrence among the study communities differed. Table 5.6 Box's M and Log Determinants Box's M 12.346 F Approx. 1.696 df1 6 df2 11.615 Sig. 0.038 Log Determinants Residential Areas Rank Log Determinant James T 3 -5.432 Ursher F 2 .a Nima 3 -6.531 Korley Dudor 1 .a Cantonment 2 .a Pooled within-groups 3 -7.181 Source: Field Work, 2013 The first function produced the highest Eigenvalues (10.517a) with variation of 96.8% of the total data projected onto a dimension that best separates or discriminates between the communities shown in Table 5.7. The second function accounts for 3.0% whilst the third function accounts for 0.2% indicating lower discriminating ability. These Eigenvalues (Table 5.8) are related to the canonical correlations and describe how much discriminating ability a function possesses. The magnitudes of the Eigenvalues are indicative of the functions' discriminating abilities. The canonical correlation of the University of Ghana http://ugspace.ug.edu.gh 93 predictor variables (frequency of disaster, frequency of fatality and property damage) and the communities were also determined. Discriminating variables were one set of variables and the set of generated grouping variable were considered over a canonical correlation analysis performed. From the analysis, a canonical correlation was arrived at, for example (Table 5.8) a canonical correlation of 0.956 which suggests that the model explains 91.3 percent i.e., (0.956)2of the variation in the dependent variable. About 8.7 percent of flood attributes among the communities was unexplained by the frequency/quantity, fatality and damage of the disasters. Table 5.7: Summary of Canonical Discriminant Functions (Eigenvalues) Eigenvalues Function Eigenvalues % of Variance Cumulative % Canonical Correlation 1 10.517a 96.8 96.8 0.956 2 0.328a 3.0 99.8 3 0.025a 0.2 100.0 Source: Field Work, 2013. The Standardized Canonical Discriminant Function Coefficients was used to calculate the discriminant score for a given case. For the data, frequency/quantity, fatality and damage as the variables were standardized as discriminating variables. For each case, the function scores were calculated using the following equations: Score1=0.312 frequency/quantity+ 0.353 fatality + 0.643 damage Score2=0.617 frequency/quantity + 0.274 fatality - 0.815 damage Each score was standardized to have a mean of zero (0) and standard deviation of one (1). The magnitudes of these coefficients indicate how strongly the discriminating variables University of Ghana http://ugspace.ug.edu.gh 94 affect the score. From the result, it is also realised that the standardized coefficient for damage in the first function is greater in magnitude than the coefficients for the other two variables. Table 5.8: Standardized Canonical Discriminant Function Coefficients Independent Variables Function Disaster Attributes 1 2 3 Frequency of the Disaster 0.312 0.617 -1.306 Fatality of the Disaster 0.353 0.274 1.386 Property Damage 0.643 -0.815 -0.032 Source: Field Work, 2013. Thus, property damage had the greatest impact, with functions of 0.643 damage and -0.815 damage (out of maximum of one or -1), of the three on the first discriminant score. In the second function, property damages -0.815 damage (out of maximum of -1) again the highest impact of discriminating followed by frequency or quantity of the floods. The third function was not included in the score model because its Eigenvalue was not statistically significant. 5.3 Experience of Flood Disaster in Accra The study also collected data on victims of flood and non-victims. The data set assisted in understanding the number of affected people and the extent of coverage. According to Table 5.9, 91 people representing 44% of total respondents have been victims of flood University of Ghana http://ugspace.ug.edu.gh 95 disasters with as many as 119 (56%) stating that they have not personally, or directly been affected by floods. Table 5.9: People Affected by Flood Disaster in Accra Respondents Frequency Percentage Affected 91 44 Never Affected 119 56 Total 208 100 Source: Field Work, 2013. Table 5.10 Where Respondents Experience Flood Disaster Communities Respondents Percentage Nima 42 46 James Town 4 4 Ursher Town 19 21 Korley Dudor 24 26 Cantonment 2 2 Total 91 100 Source: Fieldwork, 2013 The Table 5.10 indicates that majority of respondents (46%) who have suffered from flood disasters live at Nima. Korle Dudor and Ursher Town communities had 26 percent and 21 percent respectively of residents who were victims of flood disasters. James Town and Cantonments had 4 percent and 2 percent of respondents respectively indicating that they have suffered from flood disaster. Discussions with the respondents also showed that most respondents who were affected also lost their properties and belongings. Some specifically identified such lost items as carpets, clothes, books among others. They also indicated some immediate strategies adopted to safeguard their property. Most of them indicated the raising of tables and other platforms University of Ghana http://ugspace.ug.edu.gh 96 above the floor upon which items were placed. According to a Resident of Nima who was severely affected by flood in 2011, she narrated: I had to sleep at my cousin’s residence for some few days while draining the floodwater and clearing my damaged properties from the house...and with the upcoming rainy season, I do not know what am going to do (FGD, 2013). The two victims of flood in Cantonment explained that, even though they have been victims of floods, the incidence occurred rather at their former residence at Accra Newtown but not Cantonment. One person noted that, they moved into Cantonment after his husband got a political appointment. 5.4. Adaptation Strategies to Flood Disasters. Adaptation strategies have always been an important measure to assess community’s ability to respond to an impending hazard or disaster. Aggregating responses from all study communities, majority of respondents (120) indicated that they were aware of certain adaptation strategies to flood hazards. However, 88 respondents could not trace or recall any form of adaptation strategy initiated by governments, NGO’s or civil societies in times of flooding. Table 5.11: Response on Efforts to Cope with Flood Disaster Efforts Communities Responses Nima James Ursher Kol.Dudor Cantonment Total Yes 41 11 22 41 5 120 No 46 7 12 21 2 88 Total 87 18 34 62 7 208 Source: Field Work, 2013 Adaptation efforts varied. Residents mentioned adaptation strategies that are by government, NGOs and individual. Nima, Korley Dudor and Ursher Town had larger University of Ghana http://ugspace.ug.edu.gh 97 number of residents indicating awareness on adaption strategies by government institutions, NGOs, civil society organizations and individuals in the Accra Metropolis. Table 5.12: Adaptation Efforts towards Flood Disasters in Accra Institution Flood Climate Drainage, Waste, Housing Condition Government -Japanese Donation of vehicles to NADMO -Clean-up exercise before major rains -Public sensitization jingles Creating Public Awareness Public awareness NADMO display and educates the public on dos and don’ts through multimedia, maintenance of gutters, provision of reliefs, National Sanitation Day NGO Provision of relief Assist in clean-up exercise Not much Provision of reliefs services Community Assist in clean-up exercise Participating in the Public awareness Programmes Clean-up exercise, education to neighbours, Individual Assist in clean-up exercise Participating in the Public awareness Programmes Fence walls, evacuation Other (Church, Clubs, Associations Assist in clean-up exercise Take part in some of the awareness programme Assist in clean-up exercise, Awareness activities Source: Field Work, 2013. When assessing the specific adaptation strategies for managing flood disasters, varied strategies were adopted at individual, community and institutional levels. Government provides vehicles to institutions managing floods, clean-up exercises prior to rain, public sensitization campaigns among others. Before major rains, NADMO and its disaster University of Ghana http://ugspace.ug.edu.gh 98 management partners undertakes general clean-up exercise in the Metropolis. NADMO partners with institutions such as the Security Services, Zoom Lion and Civil Society Organization, NGOs etc. (see Table 5.12). The private and benevolent organizations also help in the clean-up exercises, especially Zoomlion, corporate societies, market women, etc. NGOs also provide relief services during and after the flood disasters. They help provide temporal housing structures, food, drugs and sensitize communities. For the community at large especially in the flood prone communities, they assist NADMO officials during the clean-up exercises for example; residents at Korley Dudor were organized to assist the officials in the 2012 and 2013 clean- up exercises in and around their community. Churches and other societies sensitize members and communities on flood incidences, although this is not on a regular basis. From all responses, the most common adaptive strategies included the construction of barriers (fence wall), sensitization and desilting of gutters as well as providing relief items. In extreme cases, there are some forms of relocation by victims. These strategies could provide basis for the development of enhanced approaches to flood adaptations. A blend of scientific and local knowledge on flood adaptation strategies will provide a more robust framework to flood management and promote platforms for innovations for city authorities in physical planning. 5.4.1 Adaptation Strategies Recommended by the Respondents The major public stakeholders identified to be managing the urban disasters included NADMO, and AMA. The role of these institutions included intensive education awareness on flood disasters, relocation policies, expanding and constructing new drains, mapping and strategic spatial plans and effective disaster response systems. Private organizations such as CSO, CBO and NGO’s on the other hand, were expected to support projects financially, University of Ghana http://ugspace.ug.edu.gh 99 promote awareness on flood disasters, make technological and innovative strategies to combat urban disasters and engage in effective monitoring of the implementation of supported projects. Table 5.13 Recommended Adaptations from FGD (at Ga Mashie) Institutions Recommendation Ministries, Assemblies (AMA) NADMO Other a) Education and awareness b) Relocation c) Upgrading and rebuilding poor settlement d) Construction of more drainages e) Covering of drainage and cleaning gathers f) Development control g) Proper sanitation management h) Put fines i) Mapping and strategic spatial planes j) Re-zoning and re-alignment k) Effective delivery on disaster emergency CSO,CBO, NGOs a) Promotion of education and awareness b) Motivation to best practices c) Financial support d) Technology and innovation e) Monitoring and evaluation Individual Efforts f) Attitudinal change and discipline g) Proper clean-up h) Cooperation, i) Participation j) Providing primary data in support of decision making Source: Field Work, 2013 Respondents were also requested to recommend some adaptation strategies. From a number of community discussions, key recommendations were recommended as summarised in the Table 5.13. The recommendations made, cut across government institutions, private University of Ghana http://ugspace.ug.edu.gh 100 organizations and individuals. Residents at the individual level are also expected to have attitudinal change towards keeping the environment clean and participate in community clean-up exercises (See Table 5.13) Table 5.14 Priority Investment Strategies Frequencies Planning arrangement 26 Development control and enforcement 40 Education and awareness 79 Citizens participation 42 Others- 21 Not Accounted for 21 Source: Field survey, 2013 The result in Table 5.14 showed that the increasing flooding in Accra is due to increasing human vulnerability as the majority of respondents (79%) called for human centred directives including sensitization programs in addition to development control and law enforcement amongst others in the Metropolis. 5.5 Chapter Summary The Chapter began with an introduction that explained the sub-divisions of the chapter. The next subsection was devoted to the analysis of the spatial variations and severity of the flood disasters in the city. The section revealed that floods are not evenly distributed across the city space but rather in areas with pre-existing flood-favourable conditions. The second section used the Classification Determinant Analysis, to understand the variations in the frequency of flood occurrence, fatality of floods and the damage caused by floods in five study communities. Finally, the chapter examined the adaptation strategies adopted by University of Ghana http://ugspace.ug.edu.gh 101 governments and private organizations as well at the community levels in mitigating flood events and aftermath damage. University of Ghana http://ugspace.ug.edu.gh 102 CHAPTER SIX DISCUSSION OF MAJOR FINDINGS 6.1 Introduction The chapter discusses major findings from the results presented in Chapter 4 and Chapter 5. It focusses on the extent of floods in Accra, its causes, adaptation strategies and recommendations. The chapter is organised into six sections beginning with the introduction; section two on nature and extent of flood in Accra; the third section on causes of flood disaster; fourth section on spatial variation and severity of flood; the fifth section recommends action; and the sixth section is the chapter summary. Where necessary, the results and discussions are placed in the context of the literature reviewed under Chapter 2. 6.2 Flooding in Accra The majority of the respondents (213 of the total respondent representing 88 percent) who were aware of disaster occurrences in Ghana referred to flood as dominant disaster in Accra. Compared with cholera, fires, road accidents and other disasters in Accra, floods are said to be the most frequent, causing area wide damages to the property of individuals and communities especially during the raining seasons. The increased awareness of urban flood disasters is a result of the mass campaign across the mass media and sensitization by civil society organizations. The UNISDR (2011) also identified urban floods as predominant in urban communities worldwide. Floods affected over one hundred people up to the 1940’s; rising to 2800 in the 1990 is which saw over 200 people badly affected. This is also in agreement with Dar and Nandargi (2002) who stated that floods are natural phenomenon aggravated by extreme climate change and hydro-meteorological events. It is noted as the University of Ghana http://ugspace.ug.edu.gh 103 most disastrous, frequent and widespread disaster causing extensive damages to lives and property. 6.3. Causes of Flood Disasters Floods have been linked to several causes. Studies have linked flood events to climatic factors (UNEP, 2012; IPCC, 2011 cited in Gyekye, 2013) even though this study shows that the causes of urban floods are rather multifaceted. Generally, people have agreed to the changes in the climate through their personal experiences, research and media (Wurtermberger, 2011). Respondents expressed their perceptions on the rate at which climate is changing based on the number of years they have stayed in the community and their experiences with the weather and climate conditions. The majority of the respondents (70%) referred to the changes in the climate as intensive rainfall resulting in high run-off and unpredictable weather making it difficult to establish pattern and prepare for its impacts. Another finding was that temperature has increased over time. Responses indicated that all climate conditions are occurring with more extremes including rain and windstorms, flood, heat waves, etc. The changes in the climate based on temperature, supports the position on ISDR (2008) and Stanturf et al. (2011) that global temperature will increase by 6.4% by the end of this century with sea level rising at a rate of 59cm. About 145 respondent representing 68 percent indicated that flooding has strongest relationship with rainfall or rainfall intensity because it is during the rainy season that flooding occurs in Accra. Respondents mentioned improper waste management as serious cause to floods in Accra. This supports the work of Anomanyo, 2004; Fobil, 2007 cited in Adank et al that, although about 60 to 75% of solid waste generated in the city is collected, the solid waste that remains uncollected often finds its way into open drains, thus obstructing free flow of water causing overflows that result in floods. They explained that human University of Ghana http://ugspace.ug.edu.gh 104 induced causes are catalyst to the extent of damage to properties and life lost. Other human induced causes mentioned are housing conditions (27 responses), drainage problem (24 responses), rainfall (13) and attitudinal behaviour (11responses). Poor attitudes include encroachment on low-lying areas or marshy areas, dumping of refuse in gutters and other public spaces. According to Satterthwaite et al. (2007), hundreds of millions of urban dwellers live in poor-quality homes on illegally acquired or sub-divided lands. This reduces the desire of the individuals to invest in more resilient building structures and areas. As a result, many wooden structures are found in these areas, which are easily destroyed by flood (Gyekye, 2013). Most of the slum areas are of weak and temporal structures. The level of peoples’ vulnerability to flood according to Gyekye (2011) also varies due to their socio-economic status, available knowledge on flooding and attitudes of people towards environmental management. The increasing practice of building on watercourses and wetlands, indiscriminate dumping and silting of drains has exacerbated the perennial urban flood in the Accra Metropolis. This result confirms the findings of Aboagye (2012) and Adank et al. (2011) that increased incidence of floods in developing countries is because of the low-lying nature of the land, high rainfall intensity and duration, deposition of sediments in storm drains. Others include dumping of refuse into stream and storm drains, construction of undersized drains and culverts, and building without permit in flood plains mostly because of institutional weaknesses in the Metropolis. Number of respondents linked changes in the climate, especially high intensity and erratic nature of rain to the flood. In addition, many others associated human negligence, such as nature of drainage settlement (planning, building) waste disposal and attitudes to the perennial floods in Accra. It was explained further that human induced factors such as improper waste management, inadequate drainage, and attitudes in terms of excessive University of Ghana http://ugspace.ug.edu.gh 105 encroachment on watercourses as constituting the main factors that influence urban flooding are likely to increase with the increasing level of urban population and poor attitudes. The implication of the result is that, unless concerted effort by institutions is centred on proper urban planning, enforcement of laws, construction of adequate drains and public education, calls for flood disaster mitigation will not yield any fruitful result. 6.4 Spatial Variation and Severity to Flood Disasters The challenges we face is that most of the global environmental problems can find their precedence and causes, directly or indirectly in urban areas because everybody wants to settle in urban areas (UNEP, 2009). Urban areas provide a number of socio-economic opportunities for jobs and income generation, but are simultaneously becoming crucibles of hazards and risks, especially for poorer city dwellers in developing countries. People’s exposure to environmental risk and hazard is because of the physical processes that creates these hazards (for example building and construction, urban planning, infrastructure provision or transportation), and human processes that lead to vulnerabilities (for example, lifestyle choices and consumption). The result of the study shows that even though the people of middle and low-income categories in the Accra Metropolis mostly feel many neighbourhoods in the Accra Metropolis experience floods, the effects. This includes people living low residential areas usually slums, in the streets, etc. People in these communities are more vulnerable to flood disasters as compared to other areas with resilient structures and proper drainage systems in the Metropolis. The results show that flood disaster is more severe in the slum communities such as Avenor, Korley Dudor and Ursher Town. The Classification Discriminant Analysis and Maps reveal that some of the communities have shown uneven flood disaster distribution. The severity University of Ghana http://ugspace.ug.edu.gh 106 of flood incidence in these communities is closely related to the poor environment with poor sanitation, and housing conditions. The findings of Olorunfemi (2011) supported this trend, that most urban poor dwellers are increasingly exposed to hazards due to poor conditions they are subject to (cited in Gyekye, 2013). Findings showed that even though many neighbourhoods in the Accra Metropolis experienced flooding, the people of middle and low-income individuals mostly feel the effect. According to Table 5.9, 91 people representing 44% of total respondents have been victims of flood disasters with 56% stating that they have not personally or directly been affected by floods. The Table 5.10 indicates that majority of respondents (46%) who have suffered from flood disasters live at Nima. Korle Dudor and Ursher Town communities had 26 percent and 21 percent of residents who were victims to flood disasters. James Town and Cantonments had 4 percent and 2 percent of respondents indicating that they have suffered from flood disaster. This agrees with the position of Huntington (2006) that the affected people are mostly those who have low living standards reflected by poor infrastructure and living conditions. Urban areas are not prone to disaster by nature; rather the socio-economic conditions and processes, rapid urbanization, migration increase the risk of urban dwellers to disaster. Migrants, settle in areas that originally are liable to flood with pre-existing weak structural conditions. Some results indicated that urban vulnerabilities are not limited to just low-income residents but middle and high-income groups residing in flood prone areas including waterways or flood plains, and those weak housing structures are more highly affected. In addition, the level of destruction to lives and properties in the low residential areas are more significant. University of Ghana http://ugspace.ug.edu.gh 107 6.5 Respondents’ Recommended Adaptation Strategies Floods usually results in widespread disasters affecting life and properties of victims. To address these problems, governments and private institutions, local communities and individuals must play their roles to minimise flood impacts (see Table 5.13). From the analysis, most of the measures revealed the building of strong resilience in people to floods but centred on relief and aid offering. These measures are temporal and do not holistically address the issues of building resilient cities and communities (UNEP, 2007). Adaptation efforts towards disaster risk management emerged more strongly in 2008 when UNDP started a short project together with the National Disaster Management Organization of Ghana (NADMO) on ‘Enhancing National Strategies for Effective Disaster Risk Reduction in Ghana’ (Würtenberger et al., 2011). This project focused mainly on preparing a Comprehensive Disaster Risk Management and Climate Adaptation Program for Ghana and taking stock of all DRM situations in the country in 2010-2012. Other programmes introduced included the ‘Programme for the Improvement of Capabilities to Cope with Natural Disasters Caused by Climate Change” supported by the Japanese Government. Another one is ‘Raising awareness for climate change project to establish an early warning system for disaster prevention and recovery in Ghana” funded by Vodafone Ghana. They were all aimed at building urban cities capacities towards reducing climate change effects (Würtenberger et al., 2011). In spite of the significant amount of progress made by the aforementioned strategies, the study revealed a number of shortcomings, and most of the efforts did not yield their intended benefits. The programmes have suffered from low reliability of climate models, access to information, participation and financial constrains among others. The initiatives also lacked University of Ghana http://ugspace.ug.edu.gh 108 effective coordination due to inadequate coordinating agencies at the regional and local levels. Local adaptation efforts in the study areas ranged from clean-up exercises, sensitization programmes for communities, donation of foodstuffs and other relief items after floods among others. This finding reaffirms the studies of Atuguba et al. (2009) and Fara (2000) that efforts in Ghana disaster risk management were characterized with the provision of relief items. Disaster response in Ghana has largely been seen as reactive rather than been proactive (Atuguba et al., 2009). More sustainable physical construction work and coordinated effort including planning, development drains or culvert, instituting proper waste management system and reinforcing policies and laws on land use codes are needed. The issue of reducing urban flood vulnerability also faces huge challenge due to rapid urbanization. The increasing immigration of people into the city also exacerbates the urban challenges as the population of Accra continue to exceed access to resources and services. There is the need for a holistic assessment of a shift away from the “gift adaptation” to capacity building and other preventive measures in the form of removing unauthorized structures on flood prone areas, proper waste management, and adequate sensitization, attitudinal change coupled with construction of adequately covered culvert or drains. Human rights concerns should be considered in carrying out these actions. The local people also called for adequate and timely provision of aids, provision of logistics and installation of early warning systems in the prone zones. Strategies and approaches towards the management of flood disasters have to consider the integration of both climate change indicators and urban planning rules into national policies and development programmes towards reducing human vulnerabilities to potential risks. University of Ghana http://ugspace.ug.edu.gh 109 During the FGD’s seventy-nine (79) respondents indicated the need for more education and continuous awareness creation. Such activities will result to a change in perceptions on climate change as a natural and inevitable phenomenon, and the need to strengthen long- term resilience rather than ad-hoc interventions. These changes would reduce human activities that exacerbate flooding in Accra while developing plans to handle the attitudinal changes. The study by ActionAid International (2006) in six African cities and by Jha et al (2012) recommended the need for collective mechanisms towards reducing flood risks, or for managing floods once, they happen in the cities. These strategies are necessary, as adaptation measures are most effective when developed to incorporate community-based experiences. The study revealed that there were few strategies targeted at reducing flood risks or for managing floods once they occurred in the cities. Individuals devise their own means by coping with the flood disasters. Some residents are not ready to leave from the flood disaster prone areas but stay to cope with the situation. There is the need to synergize local knowledge on coping strategies with government policies and scientific knowledge for holistic strategies to adapting and mitigating floods (UNFCCC, n.d). 6.5 Chapter Summary The chapter began with an introduction followed by the nature and extent of floods, the causes, distribution and recommended strategies. It provided discussion on the findings that flooding is the most dominant disaster in Accra. The causes of floods goes beyond rainfall which is more intensive with high run-off, to more human induced conditions including poor waste management, poor attitude of the residents leading to dumping in gutters and building in flood prone areas. The chapter explained the findings that flood occurrences and the effect of flooding varies among communities where slum and low residential areas experience more severe flood disaster that medium to high residential areas. Recommended University of Ghana http://ugspace.ug.edu.gh 110 strategies were discussed. The explanations were linked to the relevant literature reviewed in chapter 2. University of Ghana http://ugspace.ug.edu.gh 111 CHAPTER SEVEN SUMMARY OF KEY FINDINGS, CONCLUSION AND RECOMMENDATIONS 7.0 Introduction Cities are increasingly vulnerable to floods resulting from both climatic and human induced factors. In most African countries including Ghana, efforts at managing flood disasters are mainly reactive approaches due to low technology and high perception that flood is natural and inevitable disaster. The current study aimed at assessing flood disaster resulting from climate change and human vulnerability and adaptation practices of people in the affected communities in Accra to flood disasters. The study was to assess the trend of flood disaster in Accra; to examine changes in climate conditions such as rainfall and temperature and their bearing on flooding in Accra; investigate the human conditions predisposing populations to severe flood disaster; determine the distribution and severity of floods in the communities and finally to ascertain the adaptation strategies to flood disasters. Mixed method research strategy was adopted for data gathering and analysis. Reponses were gathered from 440 respondents using questionnaires, focus group discussions (FGD) and in-depth interview from household heads and institutional heads. Questionnaires were analysed using SPSS v.20 and results shown in tables, charts and maps. Classification Discriminant Analysis was used to classify flood situation from a community to the other. Geographical Information System (GIS) –ArcGIS v.10 was used to map spatial distribution of flood. 7.1 Summary of Key Findings Following analysis and discussion of the key objectives, the study arrived at some findings. The key ones are briefly discussed below. University of Ghana http://ugspace.ug.edu.gh 112 i. It was found that flooding is the most predominant disaster in Accra. Flooding is most frequent and is cumulatively claiming lives and causing damage to property , companies and the public especially during rainy seasons. The increased awareness of urban flood disasters is because of the mass campaign across the city through mass media and engagement by civil society organizations. ii. The study established that factors leading to floods in the Accra Metropolis were multifaceted in nature. This includes poor waste management, inadequate settlement planning and implementation, drainage problems, rainfall, and human behaviour all of which are serious causes of floods in Accra. Increasing urbanisation and population growth has led to the widespread development of slum communities in the Metropolis. There is a growing pressure on urban lands, which forces people to settle on lands that are earmarked for water storage basin and waterways, which are usually prone to flooding. Enforcement of regulations to prevent people building in these areas is weak. Extension of infrastructure and services including waste collection and building of drains in these areas is inadequate. Poor waste management was identified as the most serious challenge. This includes lack of proper waste collection by waste collectors who are paid, dumping of waste in gutters, hidden places and inadequate provision of waste bins at public spaces. The wastes chock gutters making run-off overflow the drains, which leads to flooding. The changing pattern and high intensity and run-off in June/July in Accra has been a major cause of flood disaster. One finding was that rainfall is becoming more intensive within a short duration, which is difficult for fast infiltration. Many part of the city is paved resulting to very high run-off during heavy downpours. A lot more water flows from upstream to University of Ghana http://ugspace.ug.edu.gh 113 Accra through the Odaw River and its tributaries. Drainages are narrow and not maintained to enable easy flow of water from upstream and within Accra. iii. Distribution of flooding in Accra was found to be varied among the communities. Some communities are more susceptible and affected than others. Geographical location of classified communities in Accra has led to the uneven distribution of flood incidences. Mapping flood areas with Geographical Information Systems informed further classifications of the flooding in the city. The Classification Determinant Analysis reveals that the frequency of disaster occurrence, the fatality of disaster and the extent of property damage was significantly dependent on the community of residence. Most of the communities mentioned are low residential areas usually slums. They are characterised by poor and weak housing structures, poor drainage systems and inadequate waste collection services. These communities are highly hit by most of the major flooding in Accra, claiming lives, destroying properties, and halting businesses. iv. Adaptation strategies were identified at different levels of the society from individual, community, civil societies, private organisations and government institutional levels. Across the board, adaptation strategies included awareness creation and sensitisation programs, clean up exercises and evacuation during flood. Awareness creation is a major adaptation efforts made in all communities. A lot of public education and sensitisation have been ongoing to avoid dumping into drains, avoid building in waterways especially with structures that are weak and easily destroyed by run-off. Clean-up exercises by NADMO and other organizations was found to be in place especially just before the major rainy season. These activities are said to have been reducing flooding to an appreciable level. Everyone applauded the clean-up exercise and encouraged it to continue. University of Ghana http://ugspace.ug.edu.gh 114 v. Areas found out to be more important in tackling flooding according to respondents includes planning, building resilient infrastructure, development control and enforcement, educations and awareness, citizens participation. Government through NADMO has spent huge amounts of money on the provision of the relief items. Providing relief items including beds, food and temporal accommodation to flood affected people has been one of the common strategies in adapting to the flood in Accra. There is limited technology and proper early warning systems in place to enable strong flood resilience in Accra. The Town and Country Planning, NADMOs and other institutions must provide the necessary plans, coordinate, and build resilience in Accra. 7.2 Conclusion The Accra Metropolis historically has suffered from devastating floods. Residents are increasingly vulnerable to flood disasters. With flooding incidences unevenly distributed depending on one’s location, the complexities of the urban environment and urbanization and social problems, slum and low residential areas are noted to be the most vulnerable to floods. In spite of the literature supporting the view that flood in Accra is attributable to changes in climatic conditions (rainfall intensity) human factors have intensified the situation. The current study examined the changes in rainfall of Accra Metropolitan Area and the human factors. Data gathered from the Ghana Meteorological Agency confirms the view on changes in climate conditions with rainfall increasing in intensity and run-off generation becoming much more unpredictable over the years. Further, interviews and surveys conducted also revealed that human factors including poor waste management, building in waterways, inadequate drains and poor attitudes of people significantly contributed to the perennial flooding in Accra. University of Ghana http://ugspace.ug.edu.gh 115 Flood frequency of occurrence, fatality of disaster and the extent of property damage were noted to vary among communities. The study recognized the need to sensitize and create awareness on environmental management, reinforce city planning and management, development control and enforcement. The Accra Metropolitan Area remains vulnerable to floods in its major urban challenges. Thus, there is a need to give more urgent and robust measures targeted at reducing vulnerabilities and building a resilient city. 7.3 Recommendations Based on the research findings and suggestions made by city dwellers, the following recommendations have been made for consideration in efforts at reducing and preventing flood disasters and reducing people’s vulnerability in Accra. i. Enhancing public awareness and sensitization in the society is very important. This must be a constant process. People need more understanding and an appreciation of the fact that flooding is not only a natural phenomenon. There are aspects of flooding that are largely caused by human and are avoidable conditions. People must be educated to avoid dumping into gutters even if bins are not adequately provided at public spaces. Education should be done at household level, social media, radio, television, market places, churches and other gathering and NGOs. While doing this, people should be made to fully participate in the discussions that they are comfortable with. School curriculum should incorporate flood disaster, its causes, effects and adaptation. Governments must make this process a top priority in planning and implementation by providing adequate resources to support the awareness process. ii. Land use planning, implementation and enforcement are very necessary in the current flood disaster challenges. With a good land use plans, areas uninhabitable are well-earmarked and proper drainage system can be developed. City authorities University of Ghana http://ugspace.ug.edu.gh 116 and ministries must make sure proper plans are in place and enforced without any fear or favour. The Town and Country Planning Departments and MMDAs must provide and strengthen the necessary actions towards granting development permits, controlling building and construction codes and ensuring that they are strictly followed. This will go a long way to regulate building in waterways and development of slum communities. iii. There must be conscious efforts to desilt chocked gutters, culverts and major drains in the city to ease water flows especially during the rainy seasons. All Assemblies must support households in adopting best waste management practices through provision of dustbins and timely collection of refuse. Most chocked gutters and drains are mostly packed with household and industrial wastes of all kinds. A properly designed scheme to manage city waste generated will potentially add up to efforts to reduce vulnerability to floods. There is a need for proper design of the drainage systems. The hydrological, Geological and Urban Roads Departments all have roles to play in implementing a good drainage system. The capacity of these departments must be built to enable them work. Their works are more capital intensive hence many resources will be needed for their activities. iv. Scientific research, technology and early warning systems must be in place. Further approaches to dealing with flood disaster in urban areas is needed. Urban systems are going through many changes. People are becoming more difficult to handle hence it takes many efforts to enforce the law to guarantee a resilient city. Development planning must consider social dimensions. Where necessary research must find synergies between physical and sociological aspects of development. Appropriate technology must be adopted to reduce run-off by creating water storms. University of Ghana http://ugspace.ug.edu.gh 117 There must be early warning systems to monitor upcoming rain, preparedness in terms of roles and finance, awareness and recovery. University of Ghana http://ugspace.ug.edu.gh 118 REFERENCES Aboagye, D. (2012a). Living with Familiar Hazards: Flood Experiences and Human Vulnerability in Accra, Ghana. Articulo-Journal of Urban Research. Aboagye, D. (2012b) .The Political Ecology of Environmental Hazards in Accra, Ghana. Journal of Environment and Earth Science, 2(10), 157-172. Agyemang, F. (2014) "Perennial Floods in the Accra Metropolis: Dissecting the Causes and Possible Solutions," African Social Science Review 6(1). Accra Metropolitan Assembly (AMA) (2011). “Accra Millennium City: Summary Profile; Policy Initiatives Programs and Projects and Achievements; and Investment Opportunities, AMA, Accra. Accra Metropolitan Assembly (AMA) (2010a). “Medium-Term Development Plan 2010 – 2013: Ghana Shared Growth and Development Agenda” AMA, Accra Accra Metropolitan Assembly (AMA) Official Website http://www.ama.gov.gh/ama/page/5103/ (Accessed 5 May, 2012) Accra Metropolitan Assembly (AMA) (2010b). From Profile to Action: Participatory Slum Upgrading and Prevention (PSUP) Work Plan for Implementing the Action Planning and programme Formulation for Accra Metropolitan Assembly (AMA), Accra Metropolitan Assembly, Ghana Accra Metropolitan Assembly (AMA)/UN-HABITAT, (2011) Participatory Slum Upgrading and Prevention. Millennium City of Accra, Ghana. Slum Situational Analysis Report. Accra Metropolitan Assembly, Ghana Action Aid International (2006). Climate change, urban flooding and the rights of the urban from six African cities. A report by Action Aid. Johannesburg: Action Aid International. ActionAid (2006).Climate Change, Urban Flooding and the Rights of the Urban Poor in Africa Key findings from six African cities. ActionAid International Adank, M., Darteh, B., Moriarty, P., Osei-Tutu, H., Assan, D. and Rooijen, D. van (2011). Towards integrated urban water management in the Greater Accra Metropolitan Area, Current status and strategic directions for the future, SWITCH/RCN Ghana, Accra. University of Ghana http://ugspace.ug.edu.gh 119 Adedeji, O.H., Odufuwa, B.O., and Adebayo, O.H., (2012). Building Capabilities for Flood Disaster and Hazard Preparedness and Risk Reduction in Nigeria: Need for Spatial Planning and Land Management. Journal of Susainable Development in Africa, 14(1) 45-58 Afeku, K. (2005). Urbanization and Flooding in Accra, Ghana. Master’s Thesis, Department of Geography, Miami University Retrieved from http://etd.ohiolink.edu/view.cgi/Afeku %20Kizito.pdf?miami1123271331 (Accessed 12 July, 2012) Anomanyo, E.D. (2004). Integration of Municipal Solid Waste Management in Accra (Ghana): Bioreactor Treatment Technology as an Integral Part of the Management Process. Published M Sc. thesis. LUMES, Sweden Lund University. Appeaning, A.K. (2011). Changing Morphology of Ghana’s Accra Coast. Journal of Coastal Conservation 15 (4) 433-443 Askew, A.J. (1999). Water in the International Decade for Natural Disaster Reduction. In: Leavesley et al (Eds) Destructive Water: Water-caused Natural Disaster, their Abatement and Control. IAHS Publication, 239. Atuguba, R.A. and Amuzu, T.E. (2006). Report on Climate Change and flooding in Alajo, Accra. Accra: The Legal Resources Centre-Ghana (LRC) And Action Aid (International) Ghana. Baffoe-Bonnie, B. Yeboah, F.A, Buabeng S.N, Ofori, A. and Collins A. (2006). Human Health Vulnerability and Public Health Adaptation to Climate Change: Risks and Responses Assessment Report, Government of Ghana Environment Protection AgencyNetherlands Climate Assistance Programme (NCAP) Birkmann, J. and Wisner, B. (2006). Measuring the un-measurable he challenge of vulnerability. Bonn, Germany: UN Institute for Environment and Human Securit (UNU-EHS), Hermann-Ehlers-Str. Encyclopedia Britannica (2013). Flood. Retrieved from http://en.wikipedia.org/wiki/Flood (Accessed, 12 May, 2013) Brown, O. and Crawford, A. (2009). Climate Change and Security in Africa:A Study for the Nordic-African Foreign Ministers Meeting, 2009, International Institute for Sustainable Development 161 Portage Avenue East, 6th Floor, Winnipeg, Manitoba, Canada University of Ghana http://ugspace.ug.edu.gh 120 Bulkeley, H., Schroeder, H., Janda, K., Zhao, J., Armstrong, A., Chu, S.Y. and Ghosh, S. (2009), “Cities and Climate Change: The role of institutions, governance and urban planning”, paper presented at the World Bank 5th Urban Symposium on Climate Change, June, Marseille. Bull-Kamanga, L., Diagne, K., and Lavel, A., (2003). From Everyday Hazards to Disasters: The Accumulation of Risk in Urban Areas. Enviroment and Urbanization 15(1) 193-204. Cutter, S.L., Mitchell, J.T. and Scott, M. (2000). Revealing the vulnerability of people and places: A Case Study of George Town County South America. Annals of the Association of American Geographers, 90(4): 713-737. Daily Graphic (2005). Ghana among the worst hit by Torrential rains, Daily Graphic 28 July 2005 Denzin, N. K. (2006). The elephant in the living room: Or extending the conversation about the politics of evidence. Qualitative Research, 9, 139–160. Dhar ON, Nandargi S. (2001). A comparative flood frequency study of Ganga and Brahmaputra river systems of north India – a brief appraisal. Water Policy 3: 101- 107. Ellis, F. (2003). The Forum for Food Security in Southern Africa. Human vulnerability and food insecurity: policy implications. London: ODI. Retrieved from www.odi.org.uk. (24 January, 2010) Ellis, C.J. (2013). A risk-based model of climate change threat: hazard, exposure and vulnerability in the ecology of lichen epiphytes. Botany, 91: 1-11. Elo, S. and Kyngäs ,H. (2007). The qualitative content analysis process. Journal of Advanced Nursing 62 (1): 107–115. Retrieved from doi: 10.1111/j.1365 2648.2007.04569.x (Accessed17 November, 2012). EPA and UNDP (2012). Exploratory Study on the Assessment of Challenges and Opportunities for Integrating Climate Change into City Level Planning in Metropolitan Areas in Ghana: Case study of Accra and Sekondi-Takoradi Metropolitan Assembly (Unpublished) Fara, K. (2001). How Natural Are 'Natural Disasters'? Vulnerability to Drought of Communal Farmers in Southern Namibia. Risk Management, 3 (3):47-63. Retrieved from http://www.jstor.org/stable/3867913 (Accessed 16 February, 2012) University of Ghana http://ugspace.ug.edu.gh 121 Ferris, E. (2012). Protection and Planned Revocations in the Context of Climate Change Legal and Policy Protection Sries, Brookings – LSE Projec on Internal Displacement, Division of International Protection, Washington DC Foresight OST (2013). Future Flooding. Retrieved from http://www.foresight.gov. up/Previous_Projects/Flood_and_Coastal_Defence/ (26 May, 2013) Füssel, H-M (2007).Vulnerability: a generally applicable conceptual framework for climate change research. Global Environmental Change 17,155–167. Fussel, H.M., 2009. An updated assessment of the risks from climate change based on research published since the IPCC Fourth Assessment Report. Clim. Change 97, 469–482. Gallopin, G.C. (2006). Linkages between vulnerability, resilience, and adaptive capacity. Elsevier Ltd GAMADA Factsheet, (2008). Background Information of Ga Mashie (Old Accra) Development Project. Accra, Ghana. Geneva, 8-13 May, 2011. Geneva, United Nations Strategy for Disaster Reduction, 2011. Retrieved from http://www.preventionweb.net/globalplatform /2011/documents/GP2011-Proceedings.pdf .(Accessed 12 June, 2013) Ghana Human Development Report (2007). Towards a more inclusive society. United Nations Development Program (UNDP), Ghana. http://hdr.undp.org/sites/default/files/nhdr_2007_ghana.pdf (Accessed 23 February, 2016) Ghana Statistical Service (2008).Ghana Demographic and Health Survey. Accra: GSS. Ghana Statistical Service (GSS) (2005). Ghana Living Standards Survey Series Five. Accra: Ghana Statistical Service. Ghana Meteorological Agency (2013). Temperature and Rainfall Data (1961-2000), Accra Airpot Station. (Unpublished). Greene, J. C. and Caracelli, V. J. (1997). Defining and describing the paradigm issues in mixed method evaluation. In J. C. Greene, & V. J. Caracelli (Eds), Advances in mixed-method evaluation: The challenges and benefits of integrating diverse paradigms (pp. 5-18). San Francisco: Jossey-Bass. Gyekye, K.A. (2011) Geomorphic Assessment of Floods within the Urban Environment of Gbawe-Mallam, Accra. Ghana Journal of Geography Vol.3 2011. pp 151-177. University of Ghana http://ugspace.ug.edu.gh 122 Gyekye, A. K. (2013).Environmental Change and Flooding in Accra, Ghana. Sacha Journal of Environmental Studies Volume, 3 (1):65-80 Retrieved from www.sachajournals.com Huong H. T. L and Pathirana, A. (2013). Urbanization and climate change impacts on future urban flooding in Can Tho city, Vietnam Institute of Meteorology, Hydrology and Environment, Hanoi, Vietnam UNESCO-IHE, Institute for Water Education, Westvest 7, Delft, The Netherlands. Huntington, T.G. (2006). Evidence for intensification of the global water cycle: review and Synthesis, Journal of Hydrology 319 (1–4):83–95.Retrieved from doi: 10.106/j.jhydrol.2005.07.003 (Accessed 15 February, 2010). Hardoy, Jorge E. Diana Mitlin and David Satterthwaite (2001), Environmental Problems in an Urbanizing World: Finding Solutions for Cities in Africa, Asia and Latin America, Earthscan Publications, London, 448 pages. IFRC (2009).West and Central Africa: Flood preparedness and response. Emergency appeal N°MDR61003 Final report. International Federation of Red Cross and Red Crescent Societies. IPCC (2001). “Africa” (Chapter 10), in Climate Change 2001: Impacts, adaptation and vulnerability. Cambridge: Cambridge University Press, 487–532. IPCC (2001). Technical summary: climate change 2001: impacts, adaptation, and vulnerability. A Report of Working Group II of the Intergovernmental Panel on Climate Change.Retrieved from http://www.grida.no/climate/ipcc_tar/wg2/ pdf/wg2TARtechsum.pdf. (Accessed 20 February 2012). IPCC (2007). Climate change 2007: impacts, adaptation and vulnerability, Working Group II Contribution to the Intergovernmental Panel on Climate Change Fourth Assessment Report. Cambridge: Cambridge University Press. IPCC (2007). Summary for Policymakers. In: Climate Change 2007: impacts, adaptation and vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on climate change, M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and C.E. Hanson, Eds. Cambridge UK: Cambridge University Press, 7-22. Jankowska, M. (2009). Continuums of Vulnerability in the Slums of Accra, Ghana. San Diego State University, Department of Geography, unpublished. Jha A. K., Bloch, R. and Lamond, J. (2012). Cities and Flooding a Guide to Integrated Urban Flood Risk Management for the 21st. Washington DC: The World Bank. University of Ghana http://ugspace.ug.edu.gh 123 Karley, N. K. (2009). Flooding And Physical Planning in Urban Areas in West Africa: Situational analysis of Accra, Ghana Heriot Watt University, Edinburgh Eh14 4as, Scotland, United Kingdom Lino, I. (2012). Sustainable Urban Peace. (Unpublished Master’s thesis) University for Peace Press, Costa Rica. Liverman, D. (1990), “Vulnerability to global environmental change”, in Kasperson, R.E., Dow, K., Golding, D. and Kasperson, J.X. (Eds), Understanding Global Environmental Change: The Contributions of Risk Analysis and Management, Clark University, Worcester, MA, pp. 27-44. Macchi, M. (2011). Framework for community-based climate vulnerability and capacity assessment in mountain areas. Kathmandu: ICIMOD. Meehl, G.A., Washington, W.M., Ammann, C.M., Arblaster, J.M., Wigley, T.M.L. and Tebaldi, C. (2004). Combinations of natural and anthropogenic forcings in 21st century climate. Journal of Climate 17. Meehl, G. A., Stocker, T. F., Collins, W. D., Friedlingstein, P., Gaye, A. T., Gregory, J. M., Kitoh, R., Knutti, R., Murphy, J. M., Noda, A., Raper, S. C. B., Watterson, I. G., Weaver, A. J. and Zhao, Z.-C. (2007) Global climate projections. Climate Change 2007: The Physical Science Basis. Contribution of Working Group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change Cambridge, UK and New York, NY, USA; Cambridge University Press. Metzger, M.J., Leemans, R., Schro¨ ter, D. (2005). A multidisciplinary multi-scale framework for assessing vulnerabilities to global change. International Journal of Applied Earth Observation and Geoinformation 7, 253–267. Minia, Z. (2008). Climate Scenarios Developed for Climate Change Impact Assessment in Ghana, Report prepared for the Environmental Protection agency under the Netherlands Climate Change Studies Assistance Programme (NCCSAP) Phase 2, Part 1, Accra Ministry of Environment Science and Technology (MEST) (2011). Ghana National Climate Change Adaptation Strategy (Unpublished). Ministry of Environment Science and Technology (MEST) (2012). National Climate Change Policy (Unpublished) Ministry of Local Government and Rural Development (MLGRD) (2012). National Urban Policy and Action Plan (Unpublished) Mulungeta, G., Ayngli, S., Daby, O.P., Gudyanga, F., Lucio, F.and Durrheem, R. (2007). University of Ghana http://ugspace.ug.edu.gh 124 Natural and Human-Induced Hazards Disasters in Sub-Saharan Africa, Science Plan. ICSU Regional Office for Africa. Munich Re Topics (2002): Natural catastrophes in 2002. Review of the Year: 2002 National Disaster Management Organisation Act 1996 (ACT 517) NOAA (2007). “Observing Climate Variability and Change”, Retrieved from http://www.oar.noaa.gov/climate/t_observing.html. (Accessed 24, January, 2013). Nyarko, B.K. (2000). Flood risk zoning of Ghana: Accra Experience. International Archives of Photogrammetry and Remote Sensing, XXXIII (Part B7), 1030-1048. O'Brien, K, Eriksen, S., Schjolen, A. and Nygaard, L. (2004). What's in a word? Conflicting interpretations of vulnerability in climate change research. CICERO Working Paper 04. OFDA-CRED (2010). The OFDA-CRED International Disaster Database 2002. Centre for Research on the Epidemiology of Disasters (CRED). Retrieved from www.em_dat.net/disasters/profiles.php Olorunfemi, F.B. (2011) Managing Flood Disasters under a Changing Climate: Lessons from Nigeria and South Africa. Nigerian Institute of Social and Economic Research (NISER). Discussion Paper No. 1, 201. pp 4-12. Osei, C. (2013). Evaluating Post Flood Disaster Response Strategies in Ashaiman and Agona Swedru. [published online: University of Ghana http://ugspace.ug.edu.gh] Owaygen, M. (2010). Mapping Climate Change Vulnerability in Southeast Asia .IDRC Global Program on Climate Change and Water ESCWA. Price, R.K., and Vojinovic, Z., (2015). Urban Flood Disaster Management, Urban Water Journal, 5(3) 259-276. Quartey-Papafio, J.J. (2006). Changing Livelihoods in Ga Mashie, a Coastal Settlement in Southern Ghana from 1957-2004, Ph.D. Thesis) Department of Geography and Resource Development, University of Ghana, 2006, Legon, Accra. Rashid-Sally, I. (2011).Urban vulnerability and resilience to water mediated climate impacts. Accra: International Water Management Institute. University of Ghana http://ugspace.ug.edu.gh 125 Sam, P. (2009). Flooding in Accra Research Report. Retrieved from http://www.modernghana .com/news/223780/1/flooding-in-accra-research report.html (Accessed 15 June, 2013) Sanderson, D. (2000). Cities, disasters and livelihoods. Risk Management. 2, (4), 49. Retrieved from htt://www.jstor.org/stable/386792. (Accessed 4 January 2013). Satterthwaite D. (2011) How Urban Societies can Adapt to Resource Shortage and Climate Change. Philosophical Transactions of the Royal Society A. Satterthwaite, David (2007), Integrating Adaptation to Climate Change in Decision-making at the Urban/Municipal Level in Low- and Middle-income Nations, (first draft), prepared for the OECD Development Assistance Committee, OECD, Paris, 33. Selvaraju, R., Subbiah, A.R, Baas, S. and Juergens, I. (2006). Livelihood adaptation to climate variability and change in drought-prone areas of Bangladesh. Asian Disaster Preparedness Centre Food and Agriculture Organization of the United Nations. Siegel P.B, Heltberg R, and Steen, L. J. (2008). Climate Change, human vulnerability, and social risk management. Social Development Department, The World Bank Smit, B. and Pilifosova, O. (2001). Adaptation to climate change in the context of sustainable development and equity, Climate Change 2001: impacts, adaptation and vulnerability. Contribution of Working Group II to the Third Assessment report of the Intergovernmental Panel on Climate Change, WMO/UNEP, 877-912. Smit, B., Burton, I., Klein, R. J. T. and Wandel, J. (2000). An anatomy of adaptation to climate change and variability, Climatic Change, 45, 223-451. Smit, B., Wandel, J. (2006). Adaptation, adaptive capacity and vulnerability. Global Environmental Change 16 (3), 282–292. Songsore, J., Nabila, J.S., Yangyuoru, Y., Avle, S., Bosque-Hamilton, E.K.,Amponsah, P.E., and Alhassan, O., (2009), ‘’Integrated Disaster Risk and Environmental Health Monitoring; Greater Accra Metropolitan Area, Ghana’’, in Mark Pelling and Ben Wisner (eds), Disaster Risk Reduction: Cases from Urban Africa, (Earthscan, London), pp.65-85. Songsore, J., Nabila, J.S., Yangyouro, Y., Amoah, E., Bosque-Hamilton, E.K., Etsibah, K.K., Gustafsson, J.-E., and Jacks, G. (2011) Regional Development in Ghana: The Theory and The Reality, (Woeli Publishing Services, New Edition, Accra) University of Ghana http://ugspace.ug.edu.gh 126 Stanturf, J.A, Warren M.L, Charnley, S. Polasky, S.C, Goodrick, S.L, Armah, F and Nyako, Y.A. (2011).Ghana Climate Change Vulnerability and Adaptation Assessment.United States Agency for International Development (USAID). Turner II, B.L., Kasperson, R.E., Matson, P.A., McCarthy, J.J., Corell, R.W., Christensen, L., Eckley, N., Kasperson, J.X., Luers, A., Martello, M.L., Polsky, C., Pulsipher, and A., Schiller, A., (2003). A framework for vulnerability analysis in sustainability science. Proceedings of the National Academy of Sciences of the United States of America 100 (14), 8074–8079. UN (2004). Towards Sustainable Development in Africa: A report on the status of disaster risk management and disaster risk assessment in Africa. UN/ISDR, Africa Development Bank, African Union, New Partnership for Africa’s Development. UN/ISDR (United Nations/International Strategy for Disaster Reduction) (2011). Building the Resilience of Nations and Communities to Disaster. Proceedings of the Conference World Conference on Disaster Reduction, Kobe, Hyogo, Japan, 18th -22nd January 2005. van der Linden and C.E. Hanson, Eds. Cambridge, UK: Cambridge University Press, 7- 22. UN/ISDR (2008). Indicators of Progress: Guidance on measuring the reduction of disaster risks and the implementation of the Hyogo Framework for Action. CH-1211 Geneva 10, Switzerland. UNEP (2002). Assessing human vulnerability due to environmental change: concepts, issues, methods and case studies. UNEP/DEWA/RS.03-5, 2002. UNEP (2007). Cities and urban vulnerability in the context of urban environmental management: concept paper. UNEP. UNEP (2009).UNEP Climate change program. Retrieved from http://www.unep.org/cl imatechange/LinkClick.aspx?fileticket=yQZjN6lCTM%3d&tab d=233&language=en- US (Accessed 2 January, 2013). UNEP /OCHA (2011). Rapid Disaster Waste Management Assessment 26 October Flash Flooding, Central Accra – Ghana. Geneva: UNEP/OCHA Environment Unit. UNFCCC (n.d). Climate Change: Impacts, Vulnerabilities and Adaptation in Developing Countries. United Nations Framework Convention on Climate Change. UN-Habitat (2011).Ghana Housing Profile. United Nations Human Settlements Programme. UNON, publishing services section, Nairobi. ISBN 978-92-1-132416- 7 University of Ghana http://ugspace.ug.edu.gh 127 UN-Habitat (2011). Cities and Climate Change. Global Report on Human Settlements 2011. Earthscan: London. UN-HABITAT (2009). Accra Urban Profile-United Nations Human Settlement Programme,Regional and Technical Corporation Division. UN-HABITAT and AMA (Accra Metropolitan Assembly) (2011). Participatory slum upgrading and prevention Millennium City of Accra, Ghana. UN-HABITAT. UNISDR (United Nations International Strategy for Disaster Reduction) (2009).Global Assessment Report on Disaster Risk Reduction 2009: Risk and Poverty in a Changing Climate, Geneva: UNISDR. http://www.preventionweb.net/english/hyogo/gar/report/index.php?id=9413 (29 May, 2013) UNISDR Global Assessment Report (2011). Revealing risk, redefining development. Geneva, United Nations Strategy for Disaster Reduction, 2011. Retrieved fromhttp://www.preventionweb.net/english/ hyogo/gar/2011/en/home/index.html (29 May, 2013) United Nations (2008). World Urbanization Prospects, the 2007 Revision, Department of Economic and Social Affairs, Population Division, New York: The United Nations United Nations Human Settlements Programme (UN Habitat) (2008) ‘African cities at risk due to sea-level rise’ Retrieved fromhttp://www.preventionweb.net/english /professional/maps/v.php?id=5645 (Accessed 5 April, 2013) United Nations Framework Convention on Climate Change (UNFCCC) (2007): “Climate Change: Impacts, Vulnerabilities and Adaptation in Developing Countries”. Retrieved from www.unfccc.int/resource/docs/publications /impact ts.pdf.3/7/11 (5 January, 2010) World Bank (2009). Assessing vulnerability and adaptive capacity to climate risks: methods for mitigating at local and National levels. Washington D.C: World Bank. World Bank. (2009). Economics of Adaptation to Climate Change-Ghana, Sustainable Development Department, World Bank. Würtenberger, L., Bonzes, I.G, van Tilburg, X. (2011). Initiatives related to climate change in Ghana towards coordinating efforts. Netherlands: Energy Research Centre. Würtenberger, L., Bonzes, I.G, van Tilburg, X. (2011). Initiatives related to climate change in Ghana towards coordinating efforts. Netherlands: Energy Research Centre. University of Ghana http://ugspace.ug.edu.gh 128 APPENDICES APPENDIX A: QUESTIONNAIRES This exercise is in partial fulfilment for the award of Master of Philosophy in Geography and Resource Development. The result of this research is thus for pure academic purpose and your responses will be dealt with utmost confidentiality as per the appropriate research ethic. Thank you. Section A: General Background 1. Age of the respondent......................................................................................... 2. Sex (1) Male [ ] (2) Female [ ] 3. Educational status: (1) No formal education [ ] (2) Basic/JHS [ ] (3) SHS [ ] (4) Tertiary [ ] 4. Marital Status: (1) Married [ ], (2) Single [ ], (3) Divorced/Separated [ ] Section B: Flood Disaster: extent and causes 5. Are you aware of any disaster in Accra (1) Yes [ ] (2) No [ ] (3) Can’t Tell [ ] 6. Could you mention the major disaster you are aware of in Accra? .......................................................................................................................................... .... 7. Identify one major cause of flood disaster .......................................................................................................................................... .... 8. Are aware of the changes in climate of Accra? (1) Yes [ ] (2) No [ ] (3) Can’t Tell [ ] 9. Comment on the table below: the changes you have observed on the corresponding elements. University of Ghana http://ugspace.ug.edu.gh 129 Conditions Changes Observed 1 Changes Observed 2 Changes Observed 3 Rainfall Temperature Wind Humidity Sunshine Atmos. Pressure 10. Which of the following indicators could you associate with the recent floods in Accra? (1) High frequency of flood [ ] (2) High damage to property [ ] (3) High fatality [ ] Section: Spatial Variation in Flood Disaster 11. Could you list some weather elements that have been showing some differences in the recent times? 12. Comment on the table below: List the event, location, date and effects Event Specific Location where disaster occurred Month/ Year Flood Section 4: People’s Experience of Flood 13. Have you been affected by flood before? (1) Yes [ ] (2) No [ ] 14. When was the last time you were affected? (1) This year [ ] (2) No Last Year [ ] (3) Last two years [ ] (4)Every year [ ] 15. How were you affected? (1)Properties destroyed [ ] (2) I lost life [ ] (3)Very disturbed and sick [ ] (4)Other [ ] University of Ghana http://ugspace.ug.edu.gh 130 Section E: Efforts/Policies on Adaptation to Climate Related Hazards/Disasters 16. Do you think a lot of efforts have been done so far? (1) Yes [ ] (2)No [ ] (3)Not aware [ ] 17. Comment on the table below Institutions/Benefactors EFFORTS MADE DATE/YEAR Government NGOs International Organizing Community itself Individuals 18. Are these efforts (question......) enough? (1) Yes [ ] (2) No [ ] (3) Can’t Tell [ ] 19. State the any other efforts you think should be in place .......................................................................................................................................... .......................................................................................................................................... ................ INTERVIEW/FOCUS GROUP DISCUSSION GUIDE FOR THE RESPONDENTS 1. List any disaster your aware of ..................................................................................................................................... ..................................................................................................................................... ............... 2. Indicate the most predominant disaster in Accra ..................................................................................................................................... ........ 3. Could explain how climate conditions and their changes result in flood of Accra? ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ................................ 4. Could you explain how any other factors apart from those from the atmosphere/air result to floods in Accra? University of Ghana http://ugspace.ug.edu.gh 131 ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ........................................ 5. List the communities associated with slums: wooden and unarranged structures, many people could not have a place to sleep, poor waste disposal and poor drainage. ..................................................................................................................................... ..................................................................................................................................... ..................................................................................................................................... ....................... 6. List the communities where flood occur since the last five years 7. What changes have you observed in the weather conditions in the city now as compared with past decades? ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………… 8. Explain how flood has affected people in the communities ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………… 9. Explain what you do before the rainy season ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………… 10. What do you or people do during and after flood in Accra ..................................................................................................................................... ..................................................................................................................................... ............... 11. Apart from your adaptive measures, what are the measures put in place by the government and the stakeholders in curtailing the flood problem in Accra? ……………………………………………………………………………………… ……………………………………………………………………………………… ……………………………………………………………………………………… ……………… 12. Are the measures effective? ……………………………………………………………………………………… ……………………………………………………………………………………… University of Ghana http://ugspace.ug.edu.gh 132 ……………………………………………………………………………………… ……………… 13. Which possible solutions will you recommends reducing if necessary halt the occurrence of those disasters in the Accra Metropolis? ……………………………………………………………………………………………… ……………………………………………………………………………………………… ……………………………………………………………………………………………… ……………… University of Ghana http://ugspace.ug.edu.gh 133 APPENDIX B: AVERAGE MONTHLY AND ANNUAL RAINFALL AND TEMPERATURE DATA FROM 1961-2000 FROM ACCRA AIRPORT STATION Average Monthly and Yearly Rainfall (mm) Years/ Month Jan Feb March April May June July Aug Sept Oct Nov Dec Annual Average Rainfall 1961 51.1 29.0 56.1 85.9 70.6 493.0 83.1 4.3 88.1 11.2 39.4 14.0 1025.8 1962 7.9 16.3 82.5 31.0 138.6 715.0 47.0 22.3 24.1 63.7 26.4 54.9 1229.7 1963 23.6 25.7 35.1 88.4 193.5 248.4 212.3 121.4 190.0 82.5 79.8 48.3 1349.0 1964 16.5 0.0 202.9 208.5 105.7 293.1 19.8 2.0 0.0 0.8 0.0 16.3 865.6 1965 10.2 33.5 5.6 216.7 103.4 367.5 139.7 50.8 28.5 45.2 27.9 24.4 1053.4 1966 8.4 7.9 127.3 136.7 83.6 64.5 62.5 7.1 10.2 65.3 0.0 1.0 574.5 1967 33.8 0.0 77.2 122.2 112.8 319.8 3.8 5.8 93.0 11.4 28.2 3.3 811.3 1968 27.8 40.1 18.3 96.5 79.0 292.9 371.3 85.9 227.1 135.4 33.3 7.4 1415.0 1969 10.4 1.0 95.3 75.2 91.7 222.5 21.3 7.9 21.8 102.1 11.2 6.3 666.7 1970 75.4 83.3 37.6 67.1 257.8 158.0 24.9 7.6 37.6 84.6 40.6 19.1 893.6 Mn(61-70) 26.5 23.7 73.8 112.8 123.7 317.5 98.6 31.5 72.0 60.2 28.7 19.5 988.5 1971 0.0 107.4 59.7 111.0 40.9 277.6 89.1 59.9 64.5 12.2 35.3 10.4 868.0 1972 0.3 37.3 61.0 256.8 40.9 126.3 2.8 2.3 35.3 51.3 22.3 3.6 640.2 1973 0.0 1.8 77.7 56.6 114.8 411.2 78.5 33.0 129.8 51.3 0.0 25.7 980.4 1974 7.9 1.5 78.7 46.7 213.6 345.4 118.1 11.9 97.8 8.4 45.7 22.3 998.0 1975 0.0 71.4 80.5 68.3 138.4 271.3 55.9 6.9 44.7 10.9 64.3 55.6 868.2 1976 0.0 45.7 46.2 132.9 50.2 119.9 12.0 2.3 16.6 100.0 22.8 1.0 549.6 1977 4.0 11.2 24.7 108.0 73.4 55.2 2.5 14.4 22.0 121.3 7.4 12.9 457.0 1978 0.3 11.8 2.8 121.7 260.0 48.2 21.5 4.7 27.7 56.3 0.3 0.0 555.3 1979 0.0 0.0 49.3 39.4 165.2 256.2 52.1 55.9 78.5 160.4 60.4 0.0 917.4 1980 0.0 5.3 44.5 181.1 245.0 167.2 71.1 59.9 98.0 76.0 45.4 7.3 1000.8 Mn(71-80) 1.3 29.3 52.5 112.3 134.2 207.9 50.4 25.1 61.5 64.8 30.4 13.9 783.5 1981 2.0 11.4 39.3 23.4 140.2 144.6 84.0 56.9 98.8 68.0 12.2 0.3 681.1 1982 17.8 18.6 74.4 114.5 170.1 278.0 60.8 2.3 0.3 36.7 0.3 0.0 773.8 1983 0.0 0.0 4.6 33.1 53.8 152.3 1.3 10.5 55.5 2.3 5.4 14.3 333.1 1984 12.8 0.0 61.0 81.0 122.0 96.2 58.7 85.3 99.3 49.0 14.2 25.4 704.9 1985 4.8 6.6 62.0 67.0 212.7 121.6 23.3 20.8 37.5 52.2 64.2 7.9 680.6 1986 0.0 63.7 65.9 19.8 144.1 60.7 33.1 0.8 29.9 83.1 38.5 5.6 545.2 1987 3.8 3.3 21.7 24.6 62.2 16.7 18.7 79.3 275.8 83.2 9.1 41.9 640.3 1988 0.0 9.2 79.6 50.8 241.8 254.9 97.7 12.0 31.4 130.3 53.7 27.5 988.9 1989 0.0 4.1 39.3 132.9 109.6 148.7 74.9 11.0 45.7 86.1 4.4 0.0 656.7 1990 9.6 6.0 1.5 105.9 99.5 127.8 39.6 0.2 24.9 30.7 37.8 85.1 568.6 Mn(81-90) 5.1 12.3 44.9 65.3 135.6 140.2 49.2 27.9 69.9 62.2 24.0 20.8 657.3 1991 13.1 5.5 52.2 174.6 277.7 123.5 263.1 21.1 12.4 64.1 0.7 0.0 1008.0 1992 0.0 0.0 93.2 65.9 171.1 93.0 55.7 8.5 8.2 35.1 26.1 0.2 557.0 1993 20.6 5.8 6.3 96.2 49.0 81.4 4.1 17.2 75.0 37.2 66.7 49.8 509.3 1994 TR 7.1 52.7 6.1 135.2 178.1 11.0 16.6 24.6 90.0 26.5 TR 547.9 1995 0.0 27.5 154.9 89.8 88.5 278.3 273.9 11.5 2.2 39.2 62.7 1.3 1029.8 1996 0.0 50.3 59.0 86.2 246.1 126.3 68.7 38.8 15.0 7.0 15.5 3.7 716.6 1997 2.8 0.0 185.2 269.4 135.7 353.3 37.8 4.6 9.5 112.0 48.7 64.5 1223.5 1998 0.0 8.7 1.5 25.2 178.5 35.9 12.3 0.5 13.4 208.9 14.1 14.6 513.6 1999 19.4 38.6 6.8 47.1 53.1 327.3 61.9 19.4 30.6 26.8 8.4 2.4 641.8 2000 0.7 0.0 59.2 28.2 127.1 116.4 19.9 12.4 8.6 36.8 25.9 77.0 512.2 University of Ghana http://ugspace.ug.edu.gh 134 Average Monthly and Yearly Temperature (°C) Year/ Month Jan Feb March April May June July Aug Sept Oct Nov Dec Annual Average Temperature 1961 27.4 28.0 28.3 27.6 28.1 25.8 24.6 23.9 24.7 26.4 27.2 27.5 26.6 1962 27.7 28.2 27.5 28.2 26.8 24.9 25.2 24.3 25.3 26.2 26.9 26.8 26.5 1963 27.4 27.5 27.8 27.6 27.2 26.4 25.5 25.4 26.2 26.4 27.1 27.4 26.8 1964 27.4 28.4 28.1 27.4 26.9 25.7 24.2 24.1 25.1 25.7 26.6 26.8 26.4 1965 27.1 27.3 28.2 27.3 27.3 25.7 24.4 24.5 25.3 26.4 27.5 27.2 26.5 1966 27.3 28.0 28.4 27.4 27.6 26.5 25.9 25.1 25.5 26.8 27.3 27.8 27.0 1967 27.2 28.3 27.6 27.7 27.2 25.4 24.6 24.4 24.9 25.9 27.0 27.4 26.5 1968 26.7 28.4 27.4 28.4 27.3 26.1 25.3 25.4 25.4 26.5 27.2 27.4 26.8 1969 27.8 28.8 28.3 28.8 28.1 26.3 24.9 25.1 25.6 26.2 27.5 27.8 27.1 1970 27.8 28.6 28.2 28.4 27.2 26.3 25.1 25.0 26.0 26.9 27.2 27.4 27.0 Mn(61- 70) 27.4 28.2 28.0 27.9 27.4 25.9 25.0 24.7 25.4 26.3 27.2 27.4 26.7 1971 27.4 27.2 27.7 27.6 27.8 25.8 25.1 24.6 25.6 26.5 27.3 26.9 26.6 1972 27.7 28.3 28.2 27.6 27.7 27.5 25.5 24.8 25.8 27.2 28.1 28.1 27.2 1973 28.6 29.4 28.8 28.9 28.0 26.4 25.8 25.7 26.0 27.2 27.9 27.4 27.5 1974 27.4 28.5 28.4 28.3 27.2 25.8 25.2 25.3 25.4 26.5 27.6 26.8 26.9 1975 27.5 27.7 27.6 27.7 27.2 25.9 25.2 24.5 25.3 26.6 27.1 27.0 26.6 1976 27.2 27.6 27.8 27.5 27.2 25.7 24.5 24.7 25.6 26.2 27.1 27.5 26.6 1977 27.7 28.4 28.9 28.4 27.9 26.2 25.4 24.9 26.1 26.8 27.8 27.5 27.2 1978 28.2 28.7 28.5 27.7 27.2 25.8 25.0 25.1 25.7 26.7 27.8 28.1 27.0 1979 28.8 29.0 28.5 28.7 27.8 26.2 25.4 25.5 26.0 26.7 27.4 27.6 27.3 1980 28.2 28.6 28.4 28.5 27.0 26.3 25.3 25.1 26.0 26.6 27.3 27.1 27.0 Mn(71- 80) 27.9 28.3 28.3 28.1 27.5 26.2 25.2 25.0 25.8 26.7 27.5 27.4 27.0 1981 27.0 28.7 28.5 28.8 27.3 26.8 25.1 25.0 26.0 27.5 28.1 28.1 27.2 1982 28.2 28.5 28.9 28.9 28.4 27.6 27.1 26.5 26.0 27.2 27.9 28.2 27.8 1983 28.6 29.5 30.0 29.4 28.6 26.1 25.5 25.3 26.1 27.6 28.1 27.7 27.7 1984 28.5 29.0 29.0 28.4 28.5 26.7 26.2 26.4 25.9 27.3 28.1 27.9 27.7 1985 28.0 28.7 28.7 28.3 26.8 26.1 25.5 26.0 26.2 26.3 27.7 27.6 27.2 1986 27.2 28.3 28.1 29.0 28.0 29.5 25.3 25.2 26.1 26.5 27.3 27.7 27.4 1987 28.5 28.8 29.0 29.8 28.6 27.6 27.1 26.4 26.3 27.2 28.5 27.9 28.0 1988 28.4 29.4 28.7 28.9 28.1 26.5 25.5 25.2 26.2 26.8 27.6 27.3 27.4 1989 27.1 28.7 28.5 28.6 27.7 26.7 25.8 25.6 26.1 26.7 28.5 28.4 27.4 1990 28.1 28.6 29.6 28.6 27.6 27.1 25.0 25.4 26.4 27.3 28.3 27.5 27.5 Mn(81- 90) 28.0 28.8 28.9 28.9 28.0 27.1 25.8 25.7 26.1 27.0 28.0 27.8 27.5 1991 28.3 28.7 28.9 27.8 27.5 27.0 25.6 25.1 26.3 26.4 27.8 27.9 27.3 1992 27.9 29.3 29.0 28.7 27.6 26.0 24.8 24.9 25.9 27.1 27.5 28.0 27.2 1993 27.6 28.5 28..5 28.4 28.5 26.9 25.5 25.3 26.4 27.5 27.7 28.0 27.3 1994 27.8 28.8 28.8 29.0 27.9 26.5 25.5 25.3 25.7 26.7 28.1 28.3 27.4 1995 28.5 29.9 28.2 29.0 28.1 26.7 25.9 26.0 26.8 27.3 27.9 28.3 27.7 1996 28.4 28.7 28.8 28.9 27.7 26.5 25.7 25.2 25.8 27.3 28.2 28.1 27.4 1997 28.4 29.1 28.1 27.7 27.3 26.1 25.1 25.3 27.2 27.7 28.1 28.1 27.4 1998 28.5 30.1 30.7 30.5 28.5 27.3 26.3 25.9 26.7 27.5 28.6 28.2 28.2 1999 28.1 28.5 28.9 28.7 28.5 27.0 25.9 25.7 25.8 26.9 28.1 28.6 27.6 2000 28.3 29.0 29.3 28.9 28.2 26.7 25.7 25.5 26.4 27.4 28.0 28.0 27.6 Mn(91- 00) 28.2 29.1 29.0 28.8 28.0 26.7 25.6 25.4 26.3 27.2 28.0 28.2 27.5 University of Ghana http://ugspace.ug.edu.gh 135 APPENDIX C: GROUP MEAN Residential Areas Mean Std. Deviation James T Frequency of the Disaster 1.3750 .51755 Fatality of the Disaster 1.7500 .46291 Property Damage 1.1250 .35355 Ursher F Frequency of the Disaster 2.9286 .26726 Fatality of the Disaster 3.0000 .00000 Property Damage 2.3571 .63332 Nima Frequency of the Disaster 3.4500 .51042 Fatality of the Disaster 3.5000 .51299 Property Damage 3.2000 .41039 Korley Dudor Frequency of the Disaster 4.0000 .00000 Fatality of the Disaster 4.0000 .00000 Property Damage 4.1842 .39286 Cantonment Frequency of the Disaster 4.2222 .44096 Fatality of the Disaster 4.1111 .33333 Property Damage 5.0000 .00000 Total Frequency of the Disaster 3.4944 .85445 Fatality of the Disaster 3.5393 .73941 Property Damage 3.4831 1.15908 University of Ghana http://ugspace.ug.edu.gh 136 APPENDIX D1: MAP OF SLUM AREAS AND FLOOD DISASTER Source: AMA Medium Term Development Plan, 2010-201 University of Ghana http://ugspace.ug.edu.gh 137 APPENDIX D2: MAP OF ACCRA METROPOLITAN AREA Source: Field Work, 2013. University of Ghana http://ugspace.ug.edu.gh 138 APPENDIX D3: A MAP OF SLUM AREAS IN THE METROPOLIS Source: Field Work, 2013. University of Ghana http://ugspace.ug.edu.gh 139 APPENDIX D4: A MAP OF COMBINED FLOOD DISTRIBUTION IN ACCRA Source: Field Work, 2013. University of Ghana http://ugspace.ug.edu.gh 140 APPENDIX D5: A MAP OF COMBINED FLOOD AND SLUMS Source: Field Work, 2013 University of Ghana http://ugspace.ug.edu.gh 141 University of Ghana http://ugspace.ug.edu.gh 1 University of Ghana http://ugspace.ug.edu.gh