COLLEGE OF HUMANITIES SCHOOL OF SOCIAL SCIENCES HAZARDS AND VULNERABILITY MAPPING FOR ADAPTATION TO CLIMATE RISKS IN SAVANNAH ECOSYSTEM: A CASE STUDY OF THE UPPER EAST REGION, GHANA BY GERALD ALBERT BAERIBAMENG YIRAN (10211304) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF DOCTOR OF PHILOSOPHY GEOGRAPHY AND RESOURCE DEVELOPMENT DEGREE DEPARTMENT OF GEOGRAPHY AND RESOURCE DEVELOPMENT JULY, 2014 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh I Declaration I hereby declare that this Thesis is the result of my own research carried out in the Upper East Region of Ghana. I further declare that this document has not been submitted in part or in whole to any other institution for an award with the exception of references which have been duly acknowledged. Name of Candidate: Gerald Albert Baeribameng Yiran Index Number: 10211304 Signature:……………………………….. Date ………………………. Principal Supervisor: Prof. E.A. Gyasi Signature: …………………… Date: ……………. Department of Geography and Resource Development University of Ghana P.O. BOX LG59 Legon, Accra edwin.gyasi@gmail.com Co-Supervisor: Dr. E.M. Attua Signature …………………… Date ………… Department of Geography and Resource Development University of Ghana P.O. BOX LG59 Legon, Accra emattua@ug.edu.gh Co-Supervisor: Dr. Kwadwo Owusu Signature ………………… Date ……………. Department of Geography and Resource Development University of Ghana P.O. BOX LG59 Legon, Accra kowusu@ug.edu.gh University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh mailto:edwin.gyasi@gmail.com mailto:emattua@ug.edu.gh mailto:kowusu@ug.edu.gh II Dedication This dissertation is dedicated to my family especially lovely daughters, Geraldine Yinborim Yiran and Geralda Masong Yiran. I get the inspiration to work harder when I look at you. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh III Acknowledgement I am extremely grateful to all my supervisors, Prof. E.A. Gyasi, Dr. E.M. Attua and Dr. Kwadwo Owusu for believing in me and giving me the courage to carry on. I cannot use words to express my gratitude but only hope that the output of this work will ginger you up to take on future students with the same zeal and enthusiasm. I most gratefully thank Prof. L.C. Stringer who has worked tirelessly with me throughout my one year stay at University of Leeds. You also brought on board Prof. Andy Challinor to act as a second supervisor. Dear Prof Lindsay, I am highly indebted to you as your great suggestions and quick response to my submissions have led to the timely completion of this work. I will also seize this opportunity to thank Prof. Andy Challinor of University of Leeds for the immense contribution and direction on the work acting as a second Supervisor in Leeds. My appreciation also goes to Dr. Andrew Evans of School of Geography, University of Leeds, for taking time to read and give guidance on the spatial mapping aspects of the thesis. I also Thank Prof. J. Songsore, Dr. Osman Alhasan, Prof. J. Yaro, Dr. Alex Owusu Barima and Dr. J. Kusimi for their support. I express my profound gratitude to the University of Leeds for the opportunity to study there. Finally, I thank the Head and staff of the Department of Geography and Resource Development for their immense support and encouragement during my study. I also extend a hand of gratitude to all the DADU Directors and their extension officers, NADMO officials, Directors of the NGOs, the Chiefs and people of the Region for the support they offered me during the fieldwork. I thank all my field assistance and Denis Asampila, J.B. Ayeridine and Stephen Tobazaa for their special assistance. My appreciation also goes to my lovely wife, Mrs. Gloria-Sheila A. Yiran for her patience and understanding. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh IV Sponsors UNIVERSITY OF GHANA University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh V Table of content Declaration .................................................................................................................................. I Dedication ................................................................................................................................. II Acknowledgement ................................................................................................................... III Table of content ........................................................................................................................ V List of Tables ........................................................................................................................... IX List of Figures ........................................................................................................................... X Abstract .................................................................................................................................. XII Chapter 1 Background of the study ........................................................................................... 1 1.1 1.1 Introduction ........................................................................................................... 1 1.2 Problem statement ....................................................................................................... 6 1.3 Objectives.................................................................................................................... 9 1.4 Structure of the thesis ................................................................................................ 10 Chapter 2 Literature Review .................................................................................................... 11 2.1 Introduction ............................................................................................................... 11 2.2 The concepts of hazards and vulnerability ................................................................ 12 2.2.1 Hazards .............................................................................................................. 12 2.2.2 Vulnerability ...................................................................................................... 13 2.2.3 Adaptation and its relationship with vulnerability to climatic hazards ............. 20 2.2.4 Climate risk ........................................................................................................ 21 2.3 Conceptual framework .............................................................................................. 21 2.3.1 Theoretical overview ......................................................................................... 21 2.3.2 The conceptual framework ................................................................................ 26 Chapter 3 Research Design and Methodology ........................................................................ 31 3.1 Introduction ............................................................................................................... 31 3.2 Methodological review ............................................................................................. 31 3.2.1 Vulnerability Assessments................................................................................. 31 3.2.2 Vulnerability measurement and mapping .......................................................... 33 3.3 The study area ........................................................................................................... 35 3.3.1 Biophysical characteristics of the Region .......................................................... 36 3.3.2 Socio-economic background of the Upper East Region .................................... 42 3.3.2.1 Social and cultural ............................................................................................. 42 3.4 Synthesis ................................................................................................................... 47 3.5 Data collection .......................................................................................................... 47 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh VI 3.5.1 Primary data ....................................................................................................... 47 3.5.2 Secondary data ................................................................................................... 54 3.5.3 Data analysis ...................................................................................................... 55 Chapter 4 Determination of Hazards and Vulnerabilities ........................................................ 56 4.1 Introduction ............................................................................................................... 56 4.2 Hazards...................................................................................................................... 57 4.2.1 Observed climate data ........................................................................................ 57 4.2.2 Exposure of the human-environment system to hazards ................................... 63 4.2.3 Sensitivity of the human-environment system to the hazards ........................... 75 4.2.4 Susceptibility/impacts ........................................................................................ 81 4.3 Summary ................................................................................................................... 86 Chapter 5 Determining adaptive capacities ............................................................................. 87 5.1 Introduction ............................................................................................................... 87 5.2 Livelihoods assets ..................................................................................................... 88 5.2.1 Natural capital .................................................................................................... 88 5.2.2 Financial capital ................................................................................................. 89 5.2.3 Social capital ...................................................................................................... 92 5.2.4 Human capital .................................................................................................... 95 5.2.5 Physical capital .................................................................................................. 96 5.3 Dynamic pressures .................................................................................................. 104 5.4 Transforming structures and processes ................................................................... 106 5.5 Synthesis ................................................................................................................. 109 5.6 Coping strategies ..................................................................................................... 110 5.7 Summary ................................................................................................................. 112 Chapter 6 Development of indicators of vulnerability .......................................................... 114 6.1 Introduction ............................................................................................................. 114 6.2 Development of susceptibility layers ...................................................................... 117 6.2.1 Susceptibility indicator datasets ...................................................................... 117 6.2.2 Susceptibility to drought/high temperatures layers ......................................... 119 6.2.3 Susceptibility to floods/high precipitation layers ............................................ 127 6.2.4 Susceptibility to windstorms............................................................................ 134 6.3 Adaptive capacity indicator datasets ....................................................................... 137 6.3.1 Social capacity ................................................................................................. 137 6.3.2 Resilience ......................................................................................................... 142 6.4 Normalisation and transformation .......................................................................... 149 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh VII 6.5 Weighting and aggregation ..................................................................................... 151 6.6 Conclusion .............................................................................................................. 153 Chapter 7 Vulnerability mapping and analyses ..................................................................... 154 7.1 Introduction ............................................................................................................. 154 7.2 Visualising vulnerability ......................................................................................... 154 7.2.1 Sectoral vulnerability ....................................................................................... 155 7.2.2 Spatial vulnerability ......................................................................................... 162 7.3 Evaluation ............................................................................................................... 168 7.3.1 Robustness tests ............................................................................................... 169 7.3.2 Sensitivity and uncertainty analysis ................................................................. 171 7.4 Analysing vulnerability ........................................................................................... 172 7.4.1 Agricultural sector ........................................................................................... 173 7.4.2 Health sector .................................................................................................... 175 7.4.3 Water, housing and infrastructure sectors ....................................................... 178 7.4.4 Overall vulnerability ........................................................................................ 180 Chapter 8 Adaptation to climatic hazards: Policy implication .............................................. 183 8.1 Introduction ............................................................................................................. 183 8.2 Barriers to adaptation .............................................................................................. 184 8.3 Policy analysis......................................................................................................... 188 8.3.1 FASDEP II and METASIP .............................................................................. 188 8.3.2 Health policy .................................................................................................... 195 8.3.3 Land use and housing policies ......................................................................... 198 8.3.4 Water policy ..................................................................................................... 200 8.3.5 The National Climate Change Policy (NCCP) ................................................ 202 8.3.6 Autonomous adaptation ................................................................................... 204 8.3.7 Critical areas for investment ............................................................................ 205 8.4 Summary ................................................................................................................. 207 Chapter 9 Overall summary and conclusion .......................................................................... 208 9.1 Summary ................................................................................................................. 208 9.1.1 Climate variability and change ........................................................................ 208 9.1.2 Adaptive capacity and vulnerability ................................................................ 209 9.1.3 Conceptual framework ..................................................................................... 213 9.2 Challenges and limitations ...................................................................................... 215 9.3 Conclusion .............................................................................................................. 216 Reference ............................................................................................................................... 219 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh VIII Appendix A1 Household questionnaire ................................................................................. 244 Appendix A2 Institutional questionnaire ............................................................................... 253 Appendix B Plots of SPI values (April –October, 1988-2012) ............................................. 257 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh IX List of Tables Table 3.1 Distribution of questionnaires ................................................................................. 51 Table 3.2 Number of participants and location of focus group discussions ............................ 53 Table 4.1 Month the rainy season starts .................................................................................. 59 Table 4.2 Observed changes in rainfall pattern over the last 30 years .................................... 61 Table 4.3 The occurrence of hazards from questionnaire survey ............................................ 64 Table 4.4 Frequency of droughts in the Region from 1983 – 2013 ......................................... 65 Table 4.5 Duration of drought ................................................................................................. 66 Table 4.6 Characteristics of drought between 1988 and 2012 extracted from 3-month SPI ... 69 Table 4.7 Frequency of floods according to respondents ........................................................ 70 Table 4.8 Consequences of hazards ......................................................................................... 72 Table 4.9 Cause of reduction in source of household food supply (in percentages) ............... 74 Table 4.10 Sensitivity of human-ecological system to hazards ............................................... 76 Table 4.11 Statistics on the impact of floods in the Upper East Region ................................. 76 Table 4.12 Sensitivity of crops to events of droughts and floods in the Upper East Region .. 80 Table 4.13Malaria Burden in the Upper East Region .............................................................. 84 Table 5.1 Employment status .................................................................................................. 90 Table 5.2 Harvest of the major crop in year of good rain ........................................................ 92 Table 5.3 Strategies for recovery from major flood events ..................................................... 94 Table 5.4 Health facilities ........................................................................................................ 99 Table 5.5 Critical Health Staff to patient ratio ........................................................................ 99 Table 5.6 Means of carting produce to the market ................................................................ 103 Table 6.1 Indicators for susceptibility mapping .................................................................... 118 Table 6.2 Ranking of localities of flash flooding .................................................................. 132 Table 6.3 Rating of land cover types for windstorm susceptibility based on tree density .... 136 Table 6.4 Scores of financial institutions in the districts ....................................................... 141 Table 6.5 Ratings of Markets based on functionality and structures ..................................... 146 Table 6.6 Weights of susceptibility indicators and the reasons for the weights .................... 152 Table 7.1 Adaptive capacity indicators for sectoral vulnerability mapping to drought/ high temperatures ........................................................................................................................... 156 Table 7.2 Adaptive capacity indicators for vulnerability of sectors to flood ........................ 158 Table 7.3 Weights for aggregate vulnerability mapping ....................................................... 160 Table 7.4 Mean volatility of different methods ..................................................................... 170 Table 7.5 Mean Volatility of nine scenarios ran with weighted sum technique .................... 172 Table 8.1 Barriers to adaptation to climatic hazards in the study area .................................. 185 Table 8.2 Selected outputs and actions of objectives 1 and 2 ............................................... 190 Table 8.3 Areas for policy consideration for adaptation in the agriculture sector................. 194 Table 8.4 Areas for investments by government and partners to enhance adaptive capacity 206 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh X List of Figures Fig. 1.1 Map of ecological zones of Ghana ............................................................................... 2 Fig. 1.2 Rainfall variability in the Guinea Savannah Ecological Zone of Ghana ..................... 4 Fig. 2.1 Relationship between Hazard and Disaster ................................................................ 12 Fig. 2.2 An integrated conceptual framework for vulnerability assessment and mapping ...... 28 Fig. 3.1 Map of the Study Area ............................................................................................... 37 Fig. 3.2 Map showing selected Towns/Villages for questionnaire interviews ........................ 49 Fig. 4.1 Framework developed to analyse data for hazards and susceptibilities ..................... 56 Fig. 4.2 Monthly Rainfall at four Agro-climatic Stations ....................................................... 58 Fig. 4.3 Maximum average monthly temperatures (°C) from 1991 to 2012 ........................... 62 Fig. 4.4 Area of cultivation of major crops in hectare (1992-2012) ........................................ 77 Fig. 4.5 Maize farms in the Upper East Region ...................................................................... 78 Fig. 4.6 Sources of Household food in the Region .................................................................. 81 Fig. 4.7 The effects of the 2007 floods in the Upper East Region .......................................... 82 Fig. 4.8 Health effects of warming (high temperatures) on the inhabitants ............................ 84 Fig. 4.9 Effects of windstorms and heavy rainfall on buildings in the Region ....................... 85 Fig. 5.1 Aspects of the conceptual framework for analysing adaptive capacity ..................... 87 Fig. 5.2 Responses to drought by respondents ........................................................................ 96 Fig. 5.3 The nature of rural houses .......................................................................................... 97 Fig. 5.4 Houses built in waterlogged areas in Bolgatanga ...................................................... 98 Fig. 5.5 Potholes on the Bolgatanga-Bawku Road ................................................................ 100 Fig. 5.6 Source of water for dry season farming or gardening .............................................. 101 Fig. 5.7 Location of market centres where produce from farms are sold .............................. 104 Fig. 5.8 Proportion of rural and urban population according to years ................................... 105 Fig. 5.9 Some strategies used to adapt to the hazards in the housing sector ......................... 111 Fig. 5.10 Stone bunds and animal droppings on farm ........................................................... 112 Fig. 6.1 Susceptibility and adaptive capacity indicators ........................................................ 115 Fig. 6.2 Susceptibility of agriculture to drought/high temperatures ...................................... 121 Fig. 6.3 Susceptibility of humans to drought/high temperatures ........................................... 124 Fig. 6.4 Susceptibility of water sector to drought/high temperatures .................................... 126 Fig. 6.5 Susceptibility of agriculture to flooding ................................................................... 128 Fig. 6.6 Susceptibility of humans to floods/high precipitation .............................................. 130 Fig. 6.7 Susceptibility of the housing sector to floods/high rainfall ...................................... 133 Fig. 6.8 Susceptibility of the road sector to floods/high precipitation ................................... 135 Fig. 6.9 Susceptibility to windstorms .................................................................................... 136 Fig. 6.10 Maps of Human and Social capitals ....................................................................... 139 Fig. 6.11 Financial Capital ..................................................................................................... 142 Fig. 6.12 Maps of Natural capital and Technology ............................................................... 145 Fig. 6.13 Maps of infrastructural coverage ............................................................................ 147 Fig. 6.14 Coping capacity of households in the districts ....................................................... 149 Fig. 7.1 Aggregation of indicators for vulnerability mapping ............................................... 155 Fig. 7.2 Vulnerability of sectors to drought/high temperatures ............................................. 157 Fig. 7.3 Map of vulnerability to floods/high rainfall according to geographical sectors ...... 159 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh XI Fig. 7.4 Map of Vulnerability to droughts/high temperatures and floods/high rainfall ........ 161 Fig. 7.5 Vulnerability map of the Upper East Region to windstorms ................................... 162 Fig. 7.6 Map showing spatial variation of vulnerabilities of the various sectors .................. 163 Fig. 7.7 Maps showing test for Robustness ........................................................................... 171 University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh XII Abstract The climate is warming, a trend that is projected to increase with increasing frequency and intensity of climate related hazards. Impacts are likely to be greater in developing countries because of dependence on climate sensitive livelihood activities. The nature and impacts of climate hazards require a study that goes beyond investigating a single hazard or livelihood activity to embrace multiple hazards and their impacts on the various livelihood activities. It is against this background that this research was set up to undertake a multi-hazard/multi-sector analysis and mapping of vulnerability and identify policy options to enhance adaptation in the savannah ecosystem, using the Upper East Region as a case study. To achieve this, an integrated framework was developed to guide the research. Assessment of the various components of the framework required varied datasets and different methods of data collection and analysis. The primary data was collected using questionnaires, focus group discussions, photography, as well as personal observation. The secondary data consisting of daily weather and socio-economic and geographic data was collected from relevant institutions, policies and published reports. The weather data was analysed using the Standard Precipitation Index and plotted with R. The questionnaire survey and socio-economic data was analysed using Statistical Package for Social Science and Excel while the focus group discussions were subjected to manual content analysis. The datasets were used to develop indicators of exposure, sensitivity and adaptive capacity and converted into geographic layers for mapping in ArcGIS 9.3. The individual indicators were georeferenced using the district boundaries since most of these datasets were collected at district scale. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh XIII The analyses identified hazards occurring as droughts, floods, high temperatures, heavy rainfall and windstorms. They occur very frequently, sometimes alternately or concurrently. These hazards have occurred more than 9 times on the average from 1983 to 2012 and have caused considerable damage to most sectors, particularly agriculture, health, housing and water sectors. The maps produced reveal where the Region is vulnerable to these climatic hazards. Droughts/high temperatures are shown to be the most devastating hazards. The agriculture sector is the most vulnerable to droughts/high temperatures and was highly vulnerable in 9 districts, while the health and water sectors show high vulnerability in 8 and 6 districts respectively. The Region was resilient to floods/heavy rainfall with 3 and 6 districts recording low vulnerabilities in the agriculture and health sectors respectively. Vulnerability to these hazards calls for adaptation. Analyses suggest that adaptation policies and programmes should target areas where adaptive capacity is low. Options to increase adaptive capacities and reduce vulnerability include irrigation, improving healthcare, education, extension and outreach, livelihood diversification, use of drought/heat tolerant crops and transplanting crops. Government could also consider transforming the Region’s economy by partnering with the private sector to invest in available opportunities such as mining, tanning, revamping the tomato cannery, to create alternative employment, propel development and advance further research in vulnerability studies. The major contributions of this work are the multi-hazard/multi-sector approach to analysis of vulnerability and adaptation to climate risks and an integrated conceptual framework it presents. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 1 Chapter 1 Background of the study 1.1 1.1 Introduction The Inter-Governmental Panel on Climate Change (IPCC) has revealed that “projections of future vegetation distribution under climate change indicate that many biomes could shift substantially, including areas where ecosystems are largely undisturbed by direct human land use” (IPCC, 2014a:7).The degree of the shift is seen to increase with increasing global mean warming without a sudden threshold (Scholze et al., 2006; Pereira et al., 2010; Rehfeldt et al., 2012, all cited in IPCC, 2014a). According to Gonzalez et al. (2010), the observed climate and vegetation projections indicate that one tenth to one half of global terrestrial area may be highly to very highly vulnerable to biome changes. These biome changes are likely to alter ecosystem structure and the provision of ecosystem services, affecting the livelihoods of many people (Gonzalez et al., 2010). One such biome (or ecosystem) likely to experience changes is the savannah ecosystem, which is supporting the livelihoods of many people, particularly in sub- Saharan Africa. The savannah ecosystem is home to a significant number of plant and animal species that provide tourism revenue, food through hunting and gathering, medicinal plants, fuel wood, construction material, cultural, regulating and supporting services (IPCC, 2014a) to both the people living in it, who depend on the system for their survival, and those situated in other areas. The savannah ecosystem of West Africa covers the northern part of Ghana, specifically as the Guinea Ecological Zone and Sudan Savannahs, which extends from about latitudes 7° 30' N to11° N and longitudes 0° 5' E to 2° 30' W (Fig. 1.1). These ecological zones support the livelihood activities of the over 10 million inhabitants who predominantly engage in agriculture (Ghana Statistical Service, 2012a). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 2 Fig. 1.1 Map of ecological zones of Ghana The Guinea Savannah and Sudan ecological zones are considered a food basket as they produce most of the major food crops and livestock in the country (MOFA, 2010). The Guinea zones also contribute more than 54 percent of total runoff in the country into the Volta Lake, which University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 3 is used largely for producing about 80 percent of the country’s electricity needs (Andah et al., 2004; Gao and Margolies, 2009). Another important ecosystem service derived from the ecological zones is water for domestic and industrial use (Yidana et al., 2011). These demands are addressed (though not adequately) from both surface and ground water sources. Socio-economic and livelihood activities in the zones are highly climate sensitive. It is often loosely said that without water, there will be no life on earth. Rainfall is the single source of water in the savannah ecosystem that feeds all other sources and supports life (Barry et al., 2005). The ecosystem thus depends solely on the amount and distribution of rainfall, both spatially and temporally, for its surface and groundwater recharge. The area has one peak rainy season of about 900-1200 mm of rainfall between April and October with the rest of the year being dry (Owusu and Waylen, 2009). Agriculture in Ghana and across the interior (Guinea and Sudan) savannah is largely rainfed. Only about 0.2% of land in the country is irrigated (MOFA, 2010). Rainfall in the zone is highly variable (Fig. 1.2). It often switches from years of high rainfall to years of low rainfall, resulting in flash floods, dry spells and/or droughts. The consequence of these climate events to agriculture is often crop failure and loss of livestock (Thornton et al., 2008). It has been projected that rainfall will decrease by about 13% by 2080 in Ghana’s Savannah region (Minia, 2004, cited in EPA, 2005). Already, there is evidence of an increasing number of dry spells, droughts, flooding and reduction in the length of the rainy season, and these have had adverse implications for agriculture (Owusu and Waylen, 2009). For example, droughts occurred in 1983/84, 1997/98, 2003, 2006/7 and affected food production in the country, leading to food shortages, especially in the north (FAO, 2009; De Pinto et al., 2012). Six floods also occurred University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 4 in the country between 1991 and 2008 and displaced millions of people and destroyed properties and farms (The World Bank Group, 2009; BBC, 2010). Fig. 1.2 Rainfall variability in the Guinea Savannah Ecological Zone of Ghana Source: EPA, 2005 Apart from the high variability in rainfall, temperature is also on the rise and is projected to go up by 1°C – 3°C by 2060 and 1.5°C – 5.2°C by 2090, with greatest impacts in the north (McSweeney et al. (undated), cited in De Pinto et al.. 2012). The rising temperatures bring about high evapotranspiration, ranging between 1450mm/annum and 1968mm/annum, and annual pan evaporation of about 2540 mm (Barry et al., 2005). According to Oguntunde (2004), nearly 80% of rainfall is estimated to be lost to evapotranspiration during the rainy season. These high values reduce water availability in the soil for proper plant growth and surface and groundwater availability for other livelihood activities. Thus, agriculture, (Total) 5 per. moving average (Total) University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 5 vegetation, and water, the main ecosystem services derived from the Savannah Ecological Zones are highly sensitive to temperature because of its influence on soil moisture content and water availability. The magnitude of these expected impacts mean that further research in this area is urgently required. This climate sensitive ecosystem is further reported to be undergoing environmental stress from anthropogenic activities (Barry et al., 2005; Yiran et al., 2012). Studies have shown that the savannah vegetation is degrading as a result of land clearing for farming and other human activities (O’Higgins, 2007; Yiran et al., 2012). Studies have also shown that the soils have been heavily degraded due to exposure to agents of erosion, over-cultivation and the high intensity of heat radiation (see Senayah et al., 2005; Aniah et al., 2013). As a result of the stress experienced associated with both climate change and growing anthropogenic pressures, the economies of the Regions in the savannah zone are becoming increasingly vulnerable (Gyasi et al., 2006; IPCC, 2014a). Research such as that presented in this study is therefore needed in order to better understand the vulnerabilities faced. The savannah ecosystem of Ghana has been subjected to a number of drought and flood events in recent times which may be attributed to climate variability and change (EPA, 2005; Owusu and Waylen, 2009). The frequent exposure and high sensitivity of the Region to hazards and risks make it highly vulnerable to such pressures. Vulnerability is due largely to the low adaptive capacity of the system and particularly its inhabitants, principally as a result of poverty (IPCC, 2014b). All three Regions of northern Ghana, which lie in the Guinea Savannah Ecosystem, are amongst the poorest in the country but the Upper East Region (Fig.1.1) is the most affected, and is an area in which more than 80% of its people are classified as poor (UNDP, 2012). The high poverty in the Region is due to the dependence on agriculture as the University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 6 major economic and livelihood activity which in turn is climate dependent. Together with land degradation, this greatly explains the low crop/livestock production of the Region (MOFA, 2010). It is becoming virtually an annual event that the early part of the rainy/wet season is marked by erratic and insufficient rainfall, whereas the end is characterised by heavy downpours (EPA, 2005). This highly variable rainfall pattern, coupled with high temperatures, results in dry spells, droughts and floods. These not only affect agriculture but also the health of the people, and the ecosystem’s ability to provide certain services such as absorption of floods, medicine, water, construction material, etc. (IPCC, 2014a). Recent crop modelling studies suggest that agriculture in West Africa will be affected disproportionately as a result of climate change (Lobell et al., 2008, cited in Antwi-Agyei et al., 2012). This has serious implications for livelihoods, particularly for people of the Upper East Region who depend almost entirely on what is produced from the farm. The occurrence of these natural hazards (particularly floods and droughts) has also revealed the weaknesses in the disaster preparedness and emergency response systems, further exposing the people, land use and infrastructure to climate risks (The World Bank Group, 2009). 1.2 Problem statement From the fifth assessment report of the IPCC, the savannah ecosystem is more highly sensitive to climate change and climate variability than was previously thought (IPCC, 2014a). According to Boko et al. (2007), “Africa’s food production systems are among the world’s most vulnerable because of extensive reliance on rainfed crop production, high intra- and inter- seasonal climate variability, recurrent droughts and floods that affect both crops and livestock, and persistent poverty that limits the capacity to adapt” (cited in IPCC, 2014b:9). Recent University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 7 studies show that agriculture in the lower latitudes, especially in the Guinea Savannah Ecological Zone of West Africa, will be disproportionately affected by the impacts of climate change and climate related hazards (Lobell et al., 2008; Thornton et al., 2009; Thornton et al., 2011; IPCC, 2014b). Besides impacts on agriculture, some studies suggest that climate change among other factors will have an important effect on water scarcity (Carter and Parker 2009; MacDonald et al., 2009; Notter et al., 2012; Tshimanga and Hughes, 2012). The population at risk of increased water stress in Africa, for the full range of the Special Report on Emissions Scenarios (SRES), is projected to be 75-250 million and 350-600 million people by the 2020s and 2050s respectively (Arnell, 2004, cited in IPCC, 2007). Some studies also point to similar negative impacts on population, health and other systems in Africa (IPCC, 2014b). Despite the negative impacts of climate change pointed out by these and many other global and regional studies, very little is known of vulnerability in the Guinea Savannah Ecological Zone, especially in the Upper East Region of Ghana, where the frequency and severity of climatic events are increasing (NADMO, 2011). The IPCC (2014b) has attributed this dearth of knowledge to lack of technical means in African countries to assess the impacts of climate change and determine effective adaptive strategies. This lack of knowledge could also be due to the complex human-environment interactions resulting in the formulation of several theories and concepts and a lack of appropriate empirical application of these concepts (Acosta-Michlik, 2005). Many studies on vulnerability have often centered on humans or society, rather than integrating human and biophysical aspects, and this has been the case for most vulnerability assessments in Ghana (Codjoe and Owusu, 2011). Another factor that has contributed to the paucity of knowledge is that many studies have concentrated on impacts of single hazard or extreme event (e.g. flood or drought) (e.g. Antwi- University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 8 Agyei et al., 2012; EPA, 2012). Nonetheless, the frequent and alternate occurrence of these events, sometimes in the same rainy season, warrants a study that considers all climatic hazards to enable effective adaptive strategies to be identified and enacted. Furthermore, most studies are done at a macro-spatial (national or higher) level which does not take into account very minute spatial differences and socio-cultural settings that gradate the vulnerabilities of societies and ecosystems (e.g. Davies and Midgley, 2010). It has been stated that, studies at these macro- spatial scales do not allow realistic assessments of human responses and differential vulnerabilities to global environmental change (Acosta-Michlik 2005; Abson et al., 2012). Several studies have also shown that practical initiatives that address and improve societal adaptive capacity, thereby reducing vulnerability, are commonly expected to be evident at the micro-spatial scale (Ford and Smit, 2004; Smit and Wandel, 2006). It is also at this scale that the cultural practices which have allowed societies to survive environmental stresses can be studied, such that they can inform adaptation (O’Brien and Holland, 1992, cited in Smit and Wandel, 2006). As observed by Greiving et al. (2006), spatially oriented risk assessment must first of all be multi-hazard oriented and must go beyond sectoral considerations. A multi-hazard and multi- sector study will allow for full diagnosis of the characteristics of vulnerability and offer insights into ways of adapting to the hazards. This study targets these knowledge gaps by adopting an integrated approach to study and map the vulnerabilities of the coupled human-environment interactions to climatic hazards. To do this, it uses Geographic Information systems (GIS) at the local level. GIS offers an opportunity to integrate natural science and social science data and is therefore appropriate for this study. It is important to emphasise at this level that most of the concepts and terms used here are defined and/or discussed in chapter two. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 9 1.3 Objectives As shown above, the Savannah ecosystem is experiencing different hazards of different magnitudes across space and time and the socio-ecological responses also vary accordingly. This means that vulnerability (measured by the impacts and adaptive capacity) is dynamic (Abson et al., 2012). It is against this background that this study aims to investigate hazard dynamics, and to map vulnerability to multi-hazards at a single snapshot in time, focusing on the savannah ecosystem using the Upper East Region as a case study. This is important in order to understand the vulnerabilities and offer policy options for adaptation in the savannah ecosystem. The aim will be achieved by: 1. Assessing the trends of climatic hazards over a 30 year period (1983-2013). In addressing this objective, answers to the following questions will be sought:  What are the principal hazards occurring in the savannah ecosystem?  What are the dynamics of the hazards over the study period? 2. Identifying adaptive/coping strategies that can be used to manage the impacts of the hazards identified in (1). The questions to be addressed under this objective will be:  Historically, what has been the response to these impacts in the ecological zone over the past 30 years?  What capacity exists to implement these measures? 3. Mapping current hazards and vulnerabilities to climate risks using GIS. This objective will be addressed by considering the following questions  Who and what are vulnerable to these hazards?  Why are these elements/groups vulnerable to the climatic hazards? University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 10 4. Examining current policies and assessing how they can be improved to further support adaptations to the climatic hazards in the savannah ecological zone.  What are the barriers to adaptation to the hazards?  What policies exist or can be formulated for adaptation?  How can the adaptive capacities identified be incorporated into the policies? 1.4 Structure of the thesis This thesis is divided into 9 chapters. Chapter 1 has set out the background to the study. The problem and the objectives have been stated, as well as the research questions to be addressed. Chapter 2 reviews literature on the concepts, theories and conceptual frameworks that have been used to analyse and assess vulnerability and develops a novel framework for application in this study. In Chapter 3, the first part reviews the methodologies that have been used to analyse and map vulnerability, and profiles the study area in relation to hazards, while the second part presents the methods of data collection and analysis. Chapter 4 addresses objective 1 by analysing the data (both observed climate data and field survey) to identify hazards and their nature of occurrence. In Chapter 5, objective 2 was addressed by analysing the field survey and socio-economic data to identify the adaptive capacities present in the Region. Chapter 6 involves developing indicators of susceptibility and adaptive capacity identified in Chapters 4 and 5, and incorporates into geographic layers. Chapter 7 maps and analyses the vulnerabilities, thus addressing objective 3 by combining the layers developed in Chapter 6. Chapter 8 addresses objective 4 by reflecting on the barriers and then analysing the policy implications for adaptation to the hazards. The summary and conclusion are presented in Chapter 9. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 11 Chapter 2 Literature Review 2.1 Introduction The term vulnerability has been used in a number of contexts: medical science, biophysical science and more increasingly in social science (e.g. UNISDR, 2009; Cuevas, 2010; Antwi- Agyei et al., 2012; Abson et al., 2012; Cardona et al. 2012; Ciurean et al., 2013;Garbero and Muttarak, 2013). Although all these disciplines are concerned with vulnerability, their conceptualisations and approaches for assessing it are different (Brooks, 2003). For example, social scientists view vulnerability as representing the set of socio-economic factors that determine people’s ability to cope with stress or change (Allen, 2003). On the other hand, climate scientists often view it in terms of the likelihood of occurrence and impacts of weather and climate related events (Stott et al., 2011; Christidis et al., 2012). In many instances, scientists from these fields and other areas such as development studies and disaster management address similar problems and processes using different languages because of their different backgrounds (Brooks, 2003). As a result, several definitions, concepts/conceptual frameworks and methodologies have been developed for vulnerability assessment and/or analysis (Hinkel, 2011). Several researchers have reviewed the different definitions (e.g. Füssel, 2009; Birkmann, 2006) given the evolving nature of concepts of hazards and vulnerability. This chapter will provide a synthesis of the main points emerging from existing reviews, introducing the concepts of hazards and vulnerability and its components; exposure, sensitivity and adaptive capacity/resilience. It will discuss different ways of systematising vulnerability and present a new conceptual framework that is relevant for the present study and which may be applied to the study of vulnerability of the savannah ecosystem and elsewhere, to climate variability and change. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 12 2.2 The concepts of hazards and vulnerability 2.2.1 Hazards The term hazard is usually used to denote environmental threats like earthquakes, wind, and floods, but hazards exist in all aspects of human life. Some of these may occur naturally while others may come about as a result of human actions. Recognising this, the IPCC defined a hazard as “the potential occurrence of a natural or human-induced physical event that may cause loss of life, injury, or other health impacts, as well as damage to and loss of property, infrastructure, livelihoods, service provision, and environmental resources” (IPCC, 2012:44). Hazards become disasters when physical events occur and actually cause harm to people and property. The United Nations International Strategy for Disaster Reduction (UNISDR) define a disaster as “a serious disruption of the functioning of society, causing widespread human, material or environmental losses which exceed the ability of affected society to cope on its own resources” (UNISDR, 2009:9). That is, a disaster is the outcome of a hazard, measured in human terms (lives lost, people affected, economic losses, environmental losses) mediated by the properties of the social and environmental systems that are exposed to and affected by the hazard (UNISDR, 2009). This can be represented diagrammatically, as in Fig. 2.1. Fig. 2.1 Relationship between Hazard and Disaster (Source: Author’s own construct) As shown in Fig. 2.1, when a hazard becomes a disaster, in general, either the affected people become impoverished and recover slowly or they are able to cope and recover fast, though sometimes a mix of the two can occur. This way of thinking admits that disasters are socially INPUT HAZARD MEDIATOR HUMAN & ENVIRONMENT SYSTEMS OUTPUT DISASTER Poor/ slow recovery Coping/ Fast recovery University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 13 constructed events, that is, they are products of the impact of hazards on people whose vulnerability have been created by social, economic, political and other mediating conditions (Cannon and Müller-Mahn, 2010). This makes some scholars regard vulnerability as a far greater determinant of disaster risk than the existence of hazards themselves (Ward and Shively, 2011; Hewitt, 1983, both cited in Alexander, 2006:2). Therefore, disasters caused by hazards are not only influenced by the magnitude and frequency of the hazard event but are also determined by the vulnerability of the affected society and its natural environment (Cardona et al., 2012; UNISDR, 2009). It is in this light that Helmer and Hilhorst (2006) believe that the core insight that disaster studies can bring to climate-related research is that “vulnerability is critical to discerning the nature of disasters” (Helmer and Hilhorst, 2006:2). 2.2.2 Vulnerability The everyday use of the term vulnerability refers to the inability to withstand the effects of a hostile environment but its scientific use depends on the field and purpose of the study (Ciurean et al., 2013). Historically, vulnerability has its roots in geography and natural hazards research (Füssel, 2006; Janssen et al., 2006) but has evolved to include several research contexts within the social sciences (e.g. Gassebner et al., 2010; Noy and Vu, 2011). From the natural hazards perspective, Blaikie et al. (1994) defined vulnerability as “the characteristics of a person or group in terms of their capacity to anticipate, cope with, resist and recover from the impacts of natural hazard” (cited in Ciurean et al., 2013:7). Vulnerable persons or groups, referred to in this definition, are frequently classified according to their age, class, ethnicity, gender and physical or mental disability. Vulnerability however varies among different groups depending on the nature of the hazard and the socio-cultural context. The focus of the above definition of vulnerability is on the consequences of the natural hazards on society and could be considered University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 14 as more socially focused (see also Malone, 2009; Lei et al., 2014; for more definitions of vulnerability in this field). A more recent contribution to the literature on vulnerability comes from climate change researchers. In climate change research, vulnerability is used as an integrative measure of the probability of occurrence and impacts of the hazard (Cuevas, 2010). Much of the research on vulnerability in climate change is reflected in the numerous report of the IPCC. In the fifth assessment report of the IPCC, vulnerability is defined as “the degree to which a system is susceptible to or unable to cope with, adverse effects of climate change, including climate variability and extremes. Vulnerability is therefore a function of the magnitude and rate of climate variation to which a system is exposed, its sensitivity and its adaptive capacity” (IPCC, 2014c:24). This definition makes the physical causes and their effects an explicit aspect of vulnerability while the social context encompasses the notions of sensitivity and adaptive capacity (IPCC, 2012). Vulnerability can also be interpreted as the residual impact of climate change after adaptation measures or adaptive capacity has been accounted for (Wilby and Miller, 2009). Therefore, vulnerability is dependent on the context and spatial scale as it reflects in the variations in wealth, social equality, availability of food, the status of health and education, infrastructure, access to natural resources and technology, which also govern the ability to adapt (Wilby and Miller, 2009). Though the definitions of vulnerability are contextual, vulnerability can basically be put into two categories: biophysical and social vulnerability (Cutter et al., 2009; Clare and Weninger, 2010; Nelson et al., 2010). This categorization is necessary as it will aid in eliminating or reducing the inconsistencies in the definitions (Cuevas, 2010). But there is a thin line between social and biophysical vulnerabilities: societal vulnerability to a hazard is not only the product University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 15 of the physical event itself but also attributable to the prevailing social and economic system of the affected community (Clare and Weninger, 2010). Thus, depending on the complex interaction between natural events and properties of the afflicted community, natural hazards mayor may not give rise to significant disturbance in the stability of the social system (Clare and Weninger, 2010). Biophysical vulnerability as seen by Jones and Boer (2003) is a measure of indicators such as monetary cost, human mortality, production costs, or ecosystem damage (cited in Cuevas, 2010). When viewed this way, biophysical vulnerability is focused on the traditional risk analysis of a system concerned with the probability of occurrence of a hazard and its ultimate impacts often expressed in terms of the amount of damage experienced by the exposed system (Cuevas, 2010). This is critical, especially in relation to ecosystem vulnerability as it considers the location of the exposed system and the resources available to cope (Cuevas, 2010). Social vulnerability on the other hand explicitly focuses on socio-economic and demographic factors that increase or decrease damage from hazards (Cutter et al., 2009). It is determined by factors such as poverty and inequality, age structure, food entitlements, ethnic composition, housing quality among others (IPCC, 2014b; Turner II, 2010). These factors should not be seen as synonymous or be equated to vulnerability but, they translate climate vagaries to human suffering and loss (Ribot, 2013). As indicated earlier, several definitions of vulnerability exist but this research will adopt the definition given by the IPCC as outlined earlier (IPCC, 2014c). The reason for choosing this definition is that it expresses vulnerability mathematically which can be quantitatively implemented in Geographic Information Systems (GIS). It also captures the factors of both University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 16 biophysical and social vulnerability in its components. The following subsections therefore reflect on these components of vulnerability (i.e. exposure, sensitivity and adaptive capacity). 2.2.2.1 Exposure Exposure is employed to refer to the presence (location) of people, livelihoods, environmental services and resources, infrastructure, or economic, social, or cultural assets in places that could be adversely affected by physical events and which, thereby, are subject to potential future harm, loss, or damage (IPCC, 2012). If the elements are not located in hazard prone areas, then they will not be exposed. But as population increases, there will be increasing diverse demands for land and the gradual decrease in the availability of safer areas mean that humans and human endeavour will unavoidably be located in potentially dangerous places (IPCC, 2014c). Exposure will therefore depend on where populations choose or are forced to live and how they construct their settlements, communities and livelihoods (Brooks, 2003). Burton et al. (1993) take a slightly broader view of exposure, where it is the nature and degree to which a system experiences environmental or socio-political stress whose characteristics include the magnitude, frequency, duration and areal extent of the hazard (cited in Adger, 2006). This means that the intensity of some hazards is greater than others and some occur more often, faster and/or last longer than others. It further shows that areas affected by the hazard also vary depending on these characteristics and other environmental properties of the location where it occurs. Thus, it can be deduced that exposure partly encapsulates the spatial and temporal dimensions of vulnerability and makes it dynamic in both time and space (Abson et al., 2012). 2.2.2.2 Sensitivity Once a system is exposed to a hazard, then it may suffer harm or damage. The degree or amount of damage the exposed system suffers is its sensitivity. According to the IPCC, sensitivity is University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 17 defined as “the degree to which a system is affected, either adversely or beneficially, by climate variability or change. The effect may be direct (e.g., a change in crop yield in response to a change in the mean, range or variability of temperature) or indirect (e.g., damages caused by an increase in the frequency of coastal flooding due to sea-level rise)” (IPCC, 2014c:24). A key phrase in the definitions of sensitivity is “the degree to which”. This phrase is used as a measure of the severity of the climate stimuli or the amount of damage the system will (or will likely) suffer under the influence of climate stimuli. The term ‘affected or ‘modified’ in the definitions often imply negative consequences, thus sensitivity is more often than not used to connote adverse impacts. One reason that explains the focus on adverse impacts is the fact that climate change impact assessments are more concerned with finding ways of minimising the adverse effects of climate stresses on humanity, rather than maximising new opportunities that emerge. Sensitivity of a system to climate change may depend on the innate physiological or biological variables of the system, specific physical or ecological factors and/or may be highly influenced by the presence and extent of other human-related factors (Glick et al., 2011). These characteristics will make the system naturally fight or resist the stimuli before yielding to them if the effects overwhelm the characteristics. This means that some systems can suffer more damage than others, even when exposed to a stimulus of the same magnitude (Simelton et al., 2012). For example, people who live in the tropics are more likely to withstand a heat wave than those in the Polar Regions. Similarly, flood damage will be greater in areas with mud buildings than concrete buildings. Thus, sensitivity also varies with space and time depending on the characteristics of exposure. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 18 2.2.2.3 Adaptive capacity/resilience The “ability of systems, humans, institutions, or other organisms to adjust to potential damages, take advantage of opportunities, or cope with the consequences” is referred to as adaptive capacity (IPCC, 2014c:2). The system is able to adapt to climate risks through its physiological or biological characteristics or through human interventions that enhance its capacity to adapt. Thus, adaptive capacity is shaped by the interaction of environmental, social, cultural, political and economic forces that determine vulnerability through exposures and sensitivities and the way the system’s components internally react to the shocks (Gitz and Meybeck, 2012). It follows from the definitions that adaptive capacity is also dynamic, changing from place to place and over time in reaction to the spatio-temporal changes in exposure and sensitivity (Birkmann, et al., 2013; Cuevas, 2010). Adaptive capacity therefore has two dimensions: the internal reactions to shocks (coping ability or short-term adaptive capacity) and mobilisation of resources to adapt to change (management capacity or long-term adaptive capacity) (Gitz and Meybeck, 2012). The first dimension, coping capacity, is the means by which people or organizations use available resources (both in normal times as well as during crises or adverse conditions) and abilities to face adverse consequences that could lead to a disaster (Gitz and Meybeck, 2012). Coping signifies ‘here and now’ capacity and includes a set of actions available to those at risk (Birkmann et al., 2013). According to Bermann et al. (2012), coping capacity can be transformed into sustainable adaptive capacity by institutions, thus enhancing adaptive capacity to future climate hazards. The second dimension goes beyond coping capacity to embraces possible adaptation measures (Levina and Tirpak, 2006). The determinants of adaptive capacity include economic wealth, technology, information and skills, infrastructure, institutions, and equity (Cuevas, 2010). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 19 Adaptive capacity when viewed as described above applies to only social systems (Levina and Tirpak, 2006). However, its counterpart, resilience, fills the gap in the ecological system. Although the field of origin of the concept is being contested (Manyena, 2006), resilience became an important research concept following Holling’s work on “Resilience and Stability of Ecological Systems”, where he defined resilience as “the amount of disturbance that can be sustained by a system before a change in system control or structure occurs. It could be measured by the magnitude of disturbance that the system can tolerate and still persist” (Holling, 1973, cited in Lei et al., 2014:614). From the time of Holling, the concept of resilience has evolved from primarily concerning the structural balance of a system to concern system functions, including its abilities to self-organise, learn, and regenerate after a disaster (Lei et al., 2014). As shown in the case of adaptive capacity, resilience also has two dimensions: inherent resilience and the adaptive resilience (Zhou et al., 2010). What is clear from these definitions and concepts of adaptive capacity and resilience is that they are meant to aid the system to reduce its vulnerability to hazards. Thus, some scholars consider that vulnerability and resilience are at opposite ends of a spectrum, that is, a less vulnerable system or community to a particular hazard is a more resilient community to that same hazard and vice versa, though there are still some disagreements (Adger, 2006; Renaud et al., 2010).Therefore, building resilience is often seen as tantamount to reducing vulnerability (Gallopin, 2006). According to the Resilience Alliance (2009), resilience is the degree to which a system can build and increase the capacity for learning and adaptation. It is therefore necessary to understand adaptation and its relationship with vulnerability. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 20 2.2.3 Adaptation and its relationship with vulnerability to climatic hazards Consistent in the climate change scholarship is the conceptualisation of vulnerability as a function of the exposure and sensitivity of that system to hazardous conditions and the ability or capacity, resilience of the system to cope with, adapt or recover from the effects of those conditions (Smit and Wandel, 2006). The manifestations of adaptive capacity or resilience as ways of reducing vulnerability are regarded as adaptation. Adaptation, according to IPCC, is the “process of adjustment to actual or expected climate and its effects” (IPCC, 2014c:1). The IPCC differentiates adaptation in terms of human and natural systems. In human systems, adaptation seeks to moderate harm or exploit beneficial opportunities while in natural systems, human intervention may facilitate adjustment to expected climate and its effects. Two types of adaptation are thus proposed (IPCC, 2014c:1): i. Incremental adaptation referring to adaptation actions where the central aim is to maintain the essence and integrity of a system or process at a given scale. ii. Transformational adaptation which considers adaptation that changes the fundamental attributes of a system in response to climate and its effects. Other views of adaptation exist (see Lei et al., 2014; McLaughlin, 2011; UNISDR, 2009) but the general idea that emerges is that adaptations are the adjustments in a system’s behaviour that reduce its vulnerability to the changing climate. As is seen in the IPCC definition and all other definitions, adaptation connotes practical actions or strategies taken by society or a system to ameliorate vulnerability to climate risk. This means that adaptation measures must be designed in such a way as to make social and ecological systems suffer little or no losses to climate related hazards. This may be achieved by understanding the spatio-temporal variations of hazards and vulnerabilities to climate risks. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 21 2.2.4 Climate risk Climate risk, according to Mudelsee (2010), is the probability of adverse effects from extreme values of variables in the climate system. That is, climate risk relates to extreme values of the weather or climate variables such as high precipitation, storms, floods, droughts, etc. Because climate changes, so can the various types of climate risk (floods, storms, etc.). Thus climate risks can be seen as the hazards resulting from climate variability and change. Clark et al. (2000) in identifying the dimensions of vulnerability as comprising exposure, sensitivity and resilience also suggest that sensitivity and exposure reasonably define risk and vulnerability by just adding resilience to risk. Following this view, it can be argued that the relationship between vulnerability and risk is linear. This argument is further strengthened by the risk-hazard model (Eakin and Luers, 2006; Füssel, 2006). However, the relationship between risk, vulnerability and resilience/adaptive capacity has been conceptualised differently in the diverse theories from the different fields of study. 2.3 Conceptual framework Diverse theories and concepts have been used to address the complexity of vulnerability (see Miller et al., 2010) as there is no single theory that has fully dealt with this complex problem. This calls for a review of the theories and concepts. 2.3.1 Theoretical overview The theoretical backgrounds shaping vulnerability research have largely come from hazard studies in the geophysical sciences, geography, human ecology, constructivism, and political ecology/economy (Eakin and Luers, 2006; McLaughlin and Dietz, 2008). These theories can be broadly grouped into four perspectives: biophysical, political ecology/economy, human ecology and constructivist perspectives (McLaughlin and Dietz, 2008). The biophysical University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 22 perspective on vulnerability focuses only on the vulnerability or degradation of biophysical conditions and their direct or indirect impact on the human occupants of a landscape (see Soares et al., 2012). This approach conceives vulnerability as an end-point, that is, the impact of climate change after adaptation has been accounted for (Soares et al., 2012). This is because it generally provides an understanding of climate change impacts and informs decision-making regarding the costs of adaptation versus the costs of mitigation (O’Brien et al., 2007). The main focus is therefore upon the source of risk or hazard, which determines the level of vulnerability and issues such as magnitude, duration, and impact of the climatic event normally characterise this type of study. Studies using the biophysical approaches are also known as risk-hazard approaches or impact-driven studies (Eakin and Luers, 2006; Ford et al., 2010). However, the narrow focus on environmental factors in this approach invariably neglects the social, economic and political factors that shape the exposure to and impacts from environmental threats (Eisenack and Stecker, 2010; Wisner et al., 2004, cited in McLaughlin and Dietz, 2008). It also fails to deal with the role of human agency (both individual and institutional) as well as culture in producing vulnerability, and does not consider the various livelihood and coping strategies used to mitigate or reduce vulnerability (Cutter et al., 2009). It therefore implies that human beings do nothing in the face of disasters. Such a view gives an impression that human agents are merely passive victims of risks of climate change (Kassam et al., 2011). Also, such a perspective lays more emphasis on extreme events but there are social processes that influence differential vulnerability between individuals and social groups (Cutter et al., 2009) and these need to be taken into account. Therefore, analyses of vulnerability will not be complete without taking into consideration, the resilience and coping strategies of humans (Kassam et al., 2011). Recent research has attempted to address these University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 23 shortcomings with the introduction of human ecology, political economy/ecology and constructivist perspectives. Human ecology has been defined as “the complex and varied systems of interaction between man and his living and non-living environment” (Editors, 1972, cited in Kassam et al., 2011:218). Some of the earliest attempts to integrate social factors into the analysis of vulnerability can be found in works on the human ecology of natural hazards (see Cutter et al., 2009; Kassam et al., 2011). The principal contribution of this concept was that it highlighted environmental variation as a causal force controlling social change and vulnerability. However, progress towards integrating the environment with the social has been limited by the persistence of essentialism (McLaughlin and Dietz, 2008). Essentialist theories conceptualise change as an interaction between natural tendencies and secondary forces that impede those tendencies. Essentialism particularly fails in practice because functionalist and developmental theorists have not been able to systematically theorise secondary interfering forces, resulting in descriptions of social change that fail to account for the diversity of actual histories (Bock, 1956, cited in McLaughlin and Dietz, 2008). These theories also fail to explain how populations maintain themselves in an ecosystem with varying cultural practices through adaptation and homeostasis (Clay and Olson, 2008). In trying to fill in the gaps in human ecology, political economists took a much wider view than human ecologists by emphasising the sociopolitical, cultural, and economic factors that together explain differential exposure to hazards, differential impacts, and, most importantly, differential capacities to recuperate from past impacts and/or to cope and adapt to future threats (Eakin and Luers, 2006; Abson et al., 2012). The other major contribution political economists have made to vulnerability analysis is their constant emphasis on the role that inequality and University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 24 differential political and economic power play in increasing the vulnerability of poor and marginalised groups (Pelling, 2001 and Wisner, 2003, cited in McLaughlin and Dietz, 2008). Vulnerability to food insecurity, for instance, is explained through entitlement theory as a set of linked economic and institutional factors (Adger, 2006). Important to the vulnerability discourse of the entitlements theory is its influence on coping strategies and adaptive capacity in general since the resources available to a person or society will be commanded and used in times of adversities. This is because vulnerability in the entitlement framework is the risk that a household’s commodity bundles will fail to buffer them against hazards (Ribot, 2010). Political ecology, inspired by a strong critique of the technocratic focus of traditional natural hazards researches, has been particularly influential and still has some influence on climate change and hazards research (Miller et al., 2010). Political ecology has its roots in structuralist and neo-Marxist thinking, providing a framework which is characterised by analyses of social and economic processes with interacting scales of causation and of social difference (Eakin and Luers, 2006). A political ecology approach aims at contextualising vulnerability at the local scale with any external or local pressures or drivers that may have an influence on, for example, food security. The strength of political ecology is that it offers researchers a way to explain external forces such as the practices of transnational corporations over local activities such as agricultural production (Bryant, 2001; cited in Khan, 2013). It also uses time series datasets to explain how vulnerability increases or decreases over time, and by this, is able to inform on the genealogy of narratives concerning the environment and identify power relationships supported by such narratives (Stott and Sullivan 2000, cited in Adams and Hutton, 2007). This justifies why the theory is essential for vulnerability studies as it focuses on the social relations that shape practice. In its sympathy with the poor and exploited, it addresses the plight of the vulnerable in terms of their abilities and constraints (Watts, 2000:164). Political ecology, University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 25 moving on from a purely structuralist approach to various aspects of vulnerability, has begun to consider carefully both resilience and adaptive capacity as part of the whole schema of differential vulnerability to natural hazards, risk and environmental change. In order to eliminate the barriers of essentialism, political ecologists have drawn upon constructivist theory (Christmann et al., 2014). Social constructivism is based on observation and scientific study about how people learn and create meaning of their social reality (Christmann et al., 2014). As human ecologists and political economists focus on the dynamics of social structure in their explanations of vulnerability, constructivists emphasise the role of human agency and culture, and view vulnerability as asocial construction process where potential threats are collectively assessed and negotiated by members of a society (Christmann and Ibert, 2012). This theory sees vulnerability as a collectively selected entity (that is valuable and to be preserved), delimited and located at the centre of an actor-network (with its inherent social, immaterial, material, and/or spatial structure) at a certain point in time (Christmann et al., 2014). Actors, according to Snow et al. (1986), interpret their experiences in relation to ‘‘frames’’ which provide ‘‘‘schemata of interpretation’ that enable individuals ‘to locate, perceive, identify, and label’ occurrences within their life space and the world at large’’ (cited in McLaughlin and Dietz, 2008:102). This is very important because people or communities identifying certain occurrences within their environments as hazards depend largely on their perception. In turn, people’s perceptions affect whether they adapt or not. Constructivist perspectives have helped fill the lacunae within both the human ecology and political economy perspectives by providing insights such as transforming our understanding of the role played by culture and agency in producing differential vulnerability within society; disaster victims are not merely victims but also survivors and active agents, while vulnerability University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 26 researchers are an integral part of the everyday social interactions that could contribute to, or mitigate, vulnerability (McLaughlin and Dietz, 2008). However, constructivists have been critiqued for failing to adequately theorise the dynamics of social structure which can be traced to the persistence of nominalism within the constructivist tradition (McLaughlin, 2001, cited in McLaughlin and Dietz, 2008). Realising this, constructivists shifted from what they called radical constructivism to moderate constructivism which conceptualises social categories as bounded networks with increased interest in the dynamics of social boundaries (e.g. Lamont and Bail, 2008; O’Flynn, 2014). It is obvious from the discourse provided in this chapter that no single theory is able to adequately address the issue of vulnerability. Thus, there is an increasing cross-fertilisation and convergence of theoretical perspectives. In this regard advocates of political ecology have demonstrated a lead role by attempting to synthesise the conceptualisations of a number of theoretical perspectives (McLaughlin and Dietz, 2008). As stated by McLaughlin and Dietz (2008), “a comprehensive theory of vulnerability must be capable of addressing the interrelated dynamics of social structure, human agency and environment(s)” (McLaughlin and Dietz, 2008:104). As such, this research will be driven by a strong consideration of these factors especially at the local level, where the interplay of these factors are important in determining vulnerability. 2.3.2 The conceptual framework Some of the conceptual frameworks express vulnerability in terms of exposure to a hazard, the sensitivity of the exposed system, the resilience/adaptive capacity of the system and the possible adaptations. For example, Bohle’s (2001, cited in Ciurean et al., 2013) double structure of vulnerability views vulnerability as having an external and internal side. The external side is related to the exposure to risks and shocks while the internal side, also called University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 27 coping, relates to the capacity to anticipate, cope with, resist and recover from the impact of a hazard (Ciurean et al., 2013). The pressure and release model developed by Blaikie et al (1994) also views vulnerability in relation to risk and hazard (cited in Cutter et al., 2009). This framework works at different spatial, functional and temporal scales and takes into account the interaction of multiple perturbations and stresses/stressors (Ciurean et al., 2013). It conceptualises vulnerability at three progressive levels: 1) the root causes (e.g. limited access to power, structures or resources or political ideologies or economic systems); 2) dynamic pressures such as demographic or social changes in time and space (e.g. rapid population growth, rapid urbanisation, lack of local institutions, appropriate skills or training) and 3) unsafe conditions posed by the physical environment (e.g. unprotected buildings and infrastructure, dangerous slopes) or socio-economic context (e.g. lack of local institutions, prevalence of endemic diseases) (Ciurean et al., 2013). Another framework worth mentioning here is the sustainable livelihood framework. The sustainable livelihood framework examines the vulnerability context in which people are living, their livelihoods assets and the transforming structures and processes that generate livelihood strategies leading to livelihood outcomes (Birkmann, 2006). Important here are the five livelihood asset categories and transforming structures and processes that shape poverty which is considered as one of the reasons for low adaptive capacity (IPCC, 2012). For a review of the conceptual frameworks, see Birkmann (2006) and Ciurean et al. (2013).Though the conceptual frameworks enhance our understanding of vulnerability, only some of them result in paradigms of quantitative or qualitative vulnerability assessment (Ciurean et al., 2013). Those that result in quantitative or qualitative vulnerability assessment will be integrated and used for assessing and mapping vulnerability in the savannah ecosystem in the present research. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 28 In this regard, the framework (Fig. 2.2) designed here integrates the double structure of vulnerability, the sustainable livelihood framework, the Pressure and Release (PAR) model and the global environment change framework, while the overarching premise that guides the study is that of the IPCC, which defines vulnerability as a function of exposure, sensitivity and adaptive capacity (IPCC, 2014b). Fig. 2.2 An integrated conceptual framework for vulnerability assessment and mapping Source: Author’s own construct Thus, the essential parts of the other frameworks that help in the identification and quantification of indicators are brought into play under the components of vulnerability defined by IPCC and used to construct the framework. In the framework (Fig. 2.2), the coupled human- environment system is perceived as a container which is receiving, producing and moderating the drivers of the climate system. It is from this human-environment system that the greenhouse Human Environment Climate change and variability Hazards  Livelihood Assets (Human, financial, social, physical and natural)  Transforming structures and processes (Government, private sector, policies, institutions, culture etc.)  Dynamic pressures (Skills, local markets, urbanisation, deforestation, decline in soil fertility, freedom of expression, etc.) S en si ti v it y Susceptibility Adaptive capacity/Resilience Vulnerability Exposure Characteristics Threatened elements Frequency Individuals Magnitude Household Duration Community Ecosystem State, etc. Unsafe condition Dangerous location Unprotected buildings Low income levels Lack of local institutions Lack of preparedness, etc. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 29 gases which cause climate variability and change are emitted (IPCC, 2012). Climate variability and change also manifests itself in the form of hazards to which the human-environment system is exposed. Exposure can be expressed in terms of components (i.e. threatened systems) and its characteristics and is explained by the biophysical, political economy and entitlements theories (Ciurean et al., 2013). Within the debate of social vulnerability, the term exposure also deals with social and institutional features, meaning “processes that increase defenselessness and lead to greater danger, such as exclusion from social networks” (Birkmann, 2006:19). These could be linked to unsafe conditions in the PAR model (Birkmann, 2006). The degree to which the human-environment system (i.e. the exposed system) yields to these hazards by suffering damage, loss of property and entitlements, or even loss of life, is related to the sensitivity of the system and is also explained by the biophysical, political economy and entitlements theories. The combined effect of exposure and sensitivity according the IPCC (2014b) is the potential impact or susceptibility of the system to climate change. However, within the coupled human-environment system are a set of social and environmental assets which humans use in order to mitigate or minimise these susceptibilities and sustain their lives (DFID, 1999, cited in Birkmann, 2006). Thus, borrowing from the sustainable livelihood approach are the five livelihoods assets (human, financial, physical, natural and social) and the transforming structures and processes (levels of government, private sector, culture, laws, policies) that influence livelihood strategies and their outcomes (Birkmann, 2006). Central to the sustainable livelihood approach, which is essential for this work, is that it views people and communities on the basis of their daily needs, acknowledging the various capabilities the poor and marginalised people can offer (de Haan and Zoomers, 2005). University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 30 The daily needs of people and communities guide their activities which may increase or decrease their susceptibility to hazards. The use of these livelihood capitals and transforming structures and processes could result in dynamic pressure (see PAR in Birkmann, 2006) such as demographic and social change over time and space which bears on the coping or adaptive capacity of the human-environment systems. However, the sustainable use of the livelihood assets and structures and processes will lead to positive dynamic pressures which will enhance the adaptive/coping capacity of the system. The human-environment system and its interactions may also be able to adjust itself to withstand or transform to a state that it suffers very little or no damage from the hazards. The residual impact after adaptive capacity has been accounted for when a sensitive system is exposed to a hazard, taken from IPCC’s (2014b) definition, is vulnerability. Thus, with increased coping/adaptive capacity, the vulnerability of the human- environment system will be greatly reduced if not eliminated. Inherent in the framework is that vulnerability is dynamic over space and time (Birkmann, et al., 2013; Cuevas, 2010). The framework (Fig. 2.2) brings together several frameworks and their theoretical underpinnings into a single model and therefore takes advantage of the strengths of these frameworks to quantitatively and qualitatively assess and measure vulnerability. As noted by Ciurean et al. (2013), there is no general model that can satisfy all needs, thus, this framework is by no means assumed to satisfy all needs. Rather, it is an attempt to capture most of the variables that can be used to measure and map vulnerability. Recognising these variables are dynamic, this then paves the way for a methodology to be designed to capture the variables mentioned in the framework for vulnerability analysis and mapping, even though the maps produced are acknowledged to reflect a particular snapshot in time. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 31 Chapter 3 Research Design and Methodology 3.1 Introduction As stated in Chapter 2, there are several methods for assessing and measuring hazards and vulnerability. Each of these methods has its strengths and weaknesses. The objective of this chapter is to take a look at these methods and the study area, and choose those methods that can best be applied to the current investigation in relation to the study area. Thus, the chapter begins with a discussion of the methodological approaches for assessing and mapping hazards and vulnerability and a description of the study area. It then presents the methods that were used to collect and analyse data for the present study. 3.2 Methodological review 3.2.1 Vulnerability Assessments Vulnerability assessments require a step by step approach using tools that are essential to collect credible, significant, and valid information (Cash et al., 2003, cited in Polsky et al., 2003). Some scholars have devised seven steps, and others, eight steps for carrying out hazard and vulnerability assessments (e.g. Polsky et al., 2003). Each of the steps might require different kinds of tools to execute. However, the selection of method(s) depends upon the availability of resources, models, and data. In data scarce areas such as the Upper East Region, it will be more prudent to collect data at the local level as well as use tools that generate data at higher spatial levels and disaggregate to lower levels to supplement. These approaches to vulnerability assessments have been described as two alternative but complementary approaches to impact assessment and can be summarised as (IPCC-TGICA, 2007): 1. A top-down approach involving the interpretation and downscaling of global-scale scenarios to regional level and University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 32 2. A bottom-up approach that builds scenarios by aggregating from the local to regional scales. The first approach relies heavily on the use of scenarios but the resolutions of these scenarios are too coarse for studies of this nature. The advancement in technology and refinement of models has led to the introduction of regional models. However, credible assessment at the community level requires projections of community development to be compatible with ongoing and prospective trends. This means that at the local level, historical data and information about ongoing trends as well as qualitative and other information from local resource managers, decision makers and NGOs, is of great importance, so as to obtain credible scenarios at the local and regional scale (IPCC-TGICA, 2007). Analysis of observed data combined with downscaled climate data will provide information on the hazards prevailing in the area over the past, present and the future. Information about the level of exposure and sensitivity inherent in the coupled human-environment system (Fig. 2.2 in chapter 2) is best collected at the local level as the characteristics are generally system dependent and manifest more at the local level. Similarly, extracting information on adaptive capacity also requires a blend of local and the higher level techniques at different spatial levels. Whilst a detailed description of the tools chosen and data processing procedures for gathering data pertinent to this study are discussed later, the above review suggests that local level assessments such as this should rely more on the bottom-up approach with the downscaled information providing a framework for scenario building. The information gathered at this level is very useful for deriving metrics for vulnerability mapping. University of Ghana http://ugspace.ug.edu.ghUniversity of Ghana http://ugspace.ug.edu.gh 33 3.2.2 Vulnerability measurement and mapping Vulnerability is a complex and difficult concept to measure. It is often more difficult to define criteria for quantifying vulnerability because it is not a directly observable event (Downing et al., 2001). Thus decisions about which systems or geographical locations are more vulnerable than others would most likely find acceptance if based on agreed criteria that are transparent, robust and objective (Adger et al., 2004). In supporting this thesis, Luers et al. (2003) catalogued a number of scenarios that make it difficult to identify which system is more vulnerable without some defined criteria on which to base the comparison (Luers et al., 2003:256). Notwithstanding the complexities as expounded in the literature, the assessment of vulnerability requires a reduction of potentially available data to a set of important indicators and criteria that facilitate an estimation of vulnerability (Damm, 2010). The need to develop systems of indicators of disaster risk and vulnerability at national and sub-national scales that will enable decision-makers to assess the impact of disasters was emphasised in the Hyogo Framework for Action 2005-2015 (UNISDR, 2005) developed after the World Conference on Disaster Reduction. Several definitions of indicators exist in the literature (e.g. Damm, 2010; Moldan and Dahl 2007; Birkmann et al. 2006; Nardo et al. 2005). This research adopts the definition given by Hammond et al. (1995) who defined indicators as “quantifiable constructs that provide information either on matters of wider significance than that which is actually measured, or on a process or trend that otherwise might not be apparent” (cited in Vincent, 2004:6). Indicators are not measured directly but are composed from data (primary and/or secondary; quantitative and/or