climatic factors, & adaptation. Little. Partial/full. Full. Degree of stakeholder ...... Arran32, Great Cumbrae, Little Cumbrae and Holy Island) and South Ayrshire ( ...
Performing climate change vulnerability assessments at the local level in Scotland Challenges, analytical trade-offs, and implications for local adaptation
Marta Coutinho Martins Bruno Soares
University of the West of Scotland Thesis submitted in partial fulfilment of the requirements of the award of Doctor of Philosophy
April 2013
This thesis has not been submitted for another comparable academic award.
ii
In loving memory of my grandparents António, Margarida, and João.
iii
| Abstract The complex nature of climate change and its potential impacts on society makes it one of the biggest challenges of our time. Responding to the challenges posed by climate change requires concerted action at various scales to address such impacts, particularly through the development and implementation of adaptation policy. To achieve such efforts however, it is crucial to understand climate change vulnerability which varies across different geographical areas and social groups and, as a result, it is fundamental to understand how changing climatic conditions may impact those natural and social systems. This project contributes to current understanding and knowledge of climate change vulnerability assessments (CCVA). To achieve that, an integrated vulnerability assessment framework is adapted and applied to a case study area – the Ayrshire region – in order to examine how/whether climate change vulnerability can be assessed at the local level in Scotland focusing on two policy areas: farming and flooding. The thesis explores the main challenges encountered whilst performing the CCVA. At a more conceptual level, the thesis explores how the conceptual and analytical elements described in the climate change literature can be pursued in practice. Existing agency at the local level in Scotland to act upon the outcomes of such local vulnerability assessments is also examined. Main findings of the study highlight the key role of scales of analysis and accessibility to data which affect the ability to effectively assess vulnerability as well as the existing mismatch between local agency and the capacity to act at the local level in Scotland. These findings open up to broader issues linked to the ways in which scale and (local) knowledge need to be better understood and addressed in both practice and theory if assessments of climate change vulnerability are to be used as effective governance tools in Scotland and beyond.
iv
| Acronyms and abbreviations
APF Adaptation Policy Framework ASC Adaptation Sub-Committee CAP Common Agriculture Policy CC Community Councils CCC Climate Change Committee CCRA UK Climate Change Risk Assessment CCVA Climate Change Vulnerability Assessment CHES Coupled Human-Environment System CIA Climate Impact Assessment EAC East Ayrshire Council EU European Union FRMA Flood Risk Management (Scotland) Act 2009 GIS Geographical Information System HDI Human Development Index IPCC Intergovernmental Panel on Climate Change LFA Less Favoured Areas NAC North Ayrshire Council NFRA National Flood Risk Assessment NFUS National Farmers Union Scotland SAC South Ayrshire Council SCCAF Scottish Climate Change Adaptation Framework SFPS Single Farm Payment Scheme SRDP Scottish Rural Development Programme UNDP United Nations Development Programme UNFCCC United Nations Framework Convention on Climate Change UKCIP United Kingdom Climate Impacts Programme UKCP09 United Kingdom Climate Projections 2009
v
| Acknowledgements As this journey reaches its end I would like to thank the various people that, one way or the other, helped me pursue and develop this project and to whom I am eternally grateful. This project would have not been possible without the endeavour of both my supervisors Dr Alexandre Gagnon and Dr Ruth Doherty. Alexandre’s support and commitment throughout the PhD project was pivotal for the success of this research and for that I am particularly thankful to him. I would also like to thank Ruth for her help namely reviewing earlier documents and her encouragement throughout the process. I am indebted to the funders of this project the Scottish Alliance for Geoscience, Environment and Society and the University of the West of Scotland. I am also grateful to the many valuable contributors to this research project including the people from the three Ayrshire local authorities, the Ayrshire Joint Planning Unit, the Ayrshire Community Councils, farmers, and national organisations. Although it would be difficult to list all their names here I would like nonetheless to thank all of those involved for their help and knowledge which were indispensable to this project. I am thankful to The James Hutton Institute (former Macaulay Land Use Research Institute) for providing various datasets indispensable for the analysis and a particular thank you to David Donnelly and Professor Campbell for making this possible. Thank also to the Scottish Government for the various datasets supplied, in particular to Paul Gavin for all the help in collating the necessary farming census data. Another particular thank you goes to Christine Cuthbertson from the National Farmers Union Scotland for all her help and expertise regarding farming in the Ayrshire region. On a more personal tone, I would like to thank friends and colleagues from the Scottish Alliance for Geosciences, Environment and Society, the University of the West of Scotland, and all the others I met throughout the PhD. A special thank you goes also to those at Bruno Soares Arquitectos for their support and kindness in the last months of this work. A big thank you to my dearest family for the support, encouragement and comfort they constantly provide me with. Last, but definitely not the least, a very special thank you to Andy for his invaluable help, support, and love throughout this journey.
Thank you all.
vi
| Outline of contents | Abstract ......................................................................................................................iv | Acronyms and abbreviations......................................................................................v | Acknowledgements .................................................................................................. vi | Outline of contents.................................................................................................... 1 | Table of contents ...................................................................................................... 2 | List of figures ............................................................................................................ 7 | List of tables ............................................................................................................. 8 | List of maps .............................................................................................................. 9 Chapter 1 | Introduction .............................................................................................10 Chapter 2 | Climate change adaptation and vulnerability ...........................................31 Chapter 3 | Conceptual framework and methodology ................................................66 Chapter 4 | Scoping stage: the case study region and vulnerable systems................87 Chapter 5 | Assessing the vulnerability of farming to climate change in Ayrshire .....125 Chapter 6 | Assessing the vulnerability to flooding in Ayrshire .................................150 Chapter 7 | Challenges of performing local climate change vulnerability assessments in Scotland and implications for adaptation policy.....................................174 | References ............................................................................................................224 | Appendices............................................................................................................244
1
| Table of contents | Abstract ......................................................................................................................iv | Acronyms and abbreviations......................................................................................v | Acknowledgements .................................................................................................. vi | Outline of contents.................................................................................................... 1 | Table of contents ...................................................................................................... 2 | List of figures ............................................................................................................ 7 | List of tables ............................................................................................................. 8 | List of maps .............................................................................................................. 9
Chapter 1 | Introduction .............................................................................................10 1.1.
Introduction to the thesis............................................................................. 11
1.2.
The problematic of climate change ............................................................. 12
1.3.
The emergence of climate change (adaptation) as a policy issue ............... 13
1.4.
Climate change as a governance challenge................................................ 16
1.5.
Governing climate change adaptation in Scotland ...................................... 17
1.5.1.
The climate change agenda at the local level in Scotland......................... 20
1.6.
Climate change vulnerability assessments in Scotland ............................... 23
1.7.
Aims and contribution of this research project............................................. 26
1.8.
Structure of the thesis ................................................................................. 28
Chapter 2 | Climate change adaptation and vulnerability ...........................................31 2.1.
Introduction................................................................................................. 32
2.2.
Introducing climate change in this study ..................................................... 32
2.3.
Climate change adaptation........................................................................... 34
2.4.
Climate change vulnerability.......................................................................... 37
2.4.1. The concept of climate change vulnerability ................................................ 38 2.4.1.1. Biophysical vulnerability.......................................................................... 39 2.4.1.2. Social vulnerability .................................................................................. 40 2
2.4.1.3. Integrated perspective on vulnerability.................................................... 41 2.5.
Assessing climate change impacts and vulnerability .................................... 42
2.6.
Conceptual elements of climate change vulnerability assessments.............. 44
2.6.1. Who or what is vulnerable? ......................................................................... 45 2.6.2. Who is involved? ......................................................................................... 46 2.6.3. To what is the system vulnerable?............................................................... 48 2.6.4. Why is the system vulnerable? .................................................................... 50 2.6.4.1. Multiple perturbations ............................................................................. 51 2.6.4.2. Causal structures of vulnerability ............................................................ 52 2.6.4.3. Differential vulnerability........................................................................... 53 2.6.5. How to assess vulnerability? ....................................................................... 53 2.6.5.1. Dealing with scale(s) of analysis ............................................................. 54 2.6.5.2. Historic and prospective analysis............................................................ 56 2.6.5.3. Dealing with uncertainty.......................................................................... 59 2.6.6. Current challenges in performing climate change vulnerability assessments60 2.6.7. Examples of vulnerability assessments frameworks .................................... 63 2.7.
Conclusions ................................................................................................. 65
Chapter 3 | Conceptual framework and methodology ................................................66 3.1.
Introduction .................................................................................................. 67
3.2.
Overall orientation and research approach................................................... 67
3.3.
Developing the analytical framework to assess vulnerability......................... 69
3.4.
The CCVA framework used to perform the analysis of vulnerability............... 71
3.4.1. Stage 1: scoping stage ................................................................................ 72 3.4.1.1. Establishing the stakeholders’ process and collecting data..................... 72 3.4.1.2. Selecting the key vulnerable systems ..................................................... 74 3.4.1.3. Selecting the approach and methods for assessing vulnerability ............ 75 3.4.2. Stage 2: Assessing current vulnerability of farming and flooding ................. 78 3.4.2.1. Calculating the indices of exposure to climatic risks................................ 78 3.4.2.2. Calculating the indices of adaptive capacity............................................ 79 3.4.3. Stage 3: Assessing future vulnerability of farming and flooding ................... 81 3.4.3.1. The UKCP09 climate change projections................................................ 81 3
3.4.3.2. Calculating the index of future exposure to climatic risks ........................ 83 3.5.
Conclusions .................................................................................................. 85
Chapter 4 | Scoping stage: the case study region and vulnerable systems................87 4.1.
Introduction .................................................................................................. 88
4.2.
The case study region – the Ayrshire region ................................................. 88
4.3.
The climate of the West of Scotland .............................................................. 91
4.3.1. Climate change projections for Western Scotland ....................................... 93 4.4.
Farming and flooding in Ayrshire ................................................................ 93
4.5.
Agriculture in Scotland and Ayrshire ........................................................... 95
4.5.1. Grassland systems ...................................................................................... 99 4.5.2. Characterising grassland systems in Ayrshire ........................................... 102 4.5.3. Governing agriculture in Scotland.............................................................. 108 4.5.4. Climate change projections and potential impacts on farming ................... 110 4.6.
Flooding in Scotland and Ayrshire............................................................. 113
4.6.1. Causes, types, and impacts of flooding in Scotland ................................... 113 4.6.2. Past flood incidents in Ayrshire.................................................................. 116 4.6.3. Identifying groups vulnerable to flooding in Scotland ................................. 120 4.6.4. Governing flooding in Scotland.................................................................. 121 4.6.5. Climate change projections and potential impacts on flooding................... 123 4.7.
Conclusions .............................................................................................. 124
Chapter 5 | Assessing the vulnerability of farming to climate change in Ayrshire .....125 5.1.
Introduction ................................................................................................ 126
5.2.
Assessing current vulnerability of farming to climate risks in Ayrshire ........ 126
5.2.1.
Exposure to climatic risks ...................................................................... 128
5.2.1.1. Mapping the index of exposure ............................................................. 129 5.2.2. The adaptive capacity of farming communities .......................................... 132 5.2.2.1. Biophysical determinants ...................................................................... 133 5.2.2.2. Socio-economic determinants............................................................... 134 5.2.2.3. Financial determinants.......................................................................... 136 5.2.2.4. Mapping the adaptive capacity of farming communities ........................ 137 4
5.2.3. Mapping current vulnerability of farming to climate change........................ 139 5.3.
Assessing future vulnerability of farming to climate change in Ayrshire..... 141
5.3.1. Mapping the index of future exposure of farming to climate change .......... 142 5.4.
Conclusions .............................................................................................. 149
Chapter 6 | Assessing the vulnerability to flooding in Ayrshire .................................150 6.1.
Introduction ................................................................................................. 151
6.2.
Assessing current vulnerability to flooding in Ayrhsire............................... 151
6.2.1. Exposure to climatic risks .......................................................................... 154 6.2.1.1. Mapping the index of exposure ............................................................. 154 6.2.2. Adaptive capacity to flooding ..................................................................... 159 6.2.2.1. Physical determinants........................................................................... 159 6.2.2.2. Social determinants .............................................................................. 160 6.2.2.3. Economic determinants ........................................................................ 161 6.2.2.4. Mapping the index of adaptive capacity ................................................ 161 6.2.3. Mapping current vulnerability to flooding.................................................... 164 6.3.
Assessing future vulnerability to flooding in Ayrshire................................. 166
6.3.1. Mapping the index of future exposure to climatic conditions ...................... 167 6.4.
Conclusions .............................................................................................. 173
Chapter 7 | Challenges of performing local climate change vulnerability assessments in Scotland and implications for adaptation policy.....................................174 7.1.
Introduction............................................................................................... 175
7.2.
The challenges and analytical trade-offs of performing CCVA in Scotland 177
7.2.1. Stage 1 of the CCVA: the scoping stage.................................................... 178 7.2.2. Stage 2 of the CCVA: assessing current vulnerability ................................ 179 7.2.2.1. Challenges in the analysis of farming.................................................... 179 7.2.2.2. Challenges in the analysis of flooding ................................................... 182 7.2.3. Stage 3 of the CCVA: assessing future vulnerability .................................. 184 7.2.4. Overall challenges encountered throughout the CCVA.............................. 187 7.3.
Pursuing the conceptual elements of CCVA in practice ............................ 190
7.3.1. Addressing the conceptual elements of CCVA in this study....................... 191 5
7.3.2. Reflecting on the challenges of applying the conceptual elements of CCVA to the analysis of vulnerability in the case study region ........................................... 193 7.3.2.1. Issues raised in the analysis of farming vulnerability to climate change 195 7.3.2.2. Issues raised in the analysis of vulnerability to flooding ........................ 196 7.3.2.3. Overall challenges encountered throughout the CCVA ......................... 198 7.4.
Acting upon the outcomes of CCVA.......................................................... 200
7.4.1. Farming and flooding policy and local agency ........................................... 202 7.4.2. The Scottish climate change adaptation agenda and local level agency.... 203 7.4.2.1. The SCCAF’s sectoral plans................................................................. 205 7.5. Contributions of the thesis............................................................................... 207 7.5.1. Advancing knowledge of climate change vulnerability in Ayrshire.............. 207 7.5.2. The ‘local’ in climate change adaptation policy in Scotland and beyond .... 209 7.5.3. Fostering effective mechanisms for local knowledge input into adaptation policy
212
7.5.4. Promoting ‘usable’ science ........................................................................ 214 7.5.5. Conceptualising CCVA in the climate change literature ............................. 216 7.6.
Conclusions .............................................................................................. 217
7.7.
Limitations of the analysis and directions for further research ................... 221
| References ............................................................................................................224 | Appendices............................................................................................................244
6
| List of figures Figure 1.1 - The global governance of climate change ..............................................14 Figure 2.1 - Schematic representation of the effect of an increase in temperature variability, mean, or both on extreme temperatures ................................34 Figure 2.2 - Simplified Climate Impact Assessment model .......................................49 Figure 2.3 - Simplified Climate Change Vulnerability Assessment model ..................49 Figure 2.4 - Pressure and Release model .................................................................52 Figure 2.5 - Illustration of concepts related to scales of analysis................................54 Figure 2.6 - Categories of future characterisations ....................................................57 Figure 3.1 - Analytical components and cross-cutting elements of the APF...............70 Figure 4.1 - Number and size of farming holdings in Ayrshire....................................96 Figure 4.2 - Farm Business Income per type of farm...................................................97 Figure 4.3 - Farm types in Scotland and Ayrshire ......................................................99 Figure 4.4 - Simplified structure of grassland systems.............................................100 Figure 4.5 - Number of holdings with grassland, woodland, crops, and other land in Ayrshire between 2001 and 2011..........................................................105 Figure 4.6 - Number of holdings per main type of livestock in Ayrshire in 2011 .......105 Figure 7.1 - The conceptual elements of climate change vulnerability assessments.194 Figure 7.2 - Main challenges of applying the conceptual elements of CCVA in the analysis of farming in the Ayrshire region..............................................196 Figure 7.3 - Main challenges of applying the conceptual elements of CCVA in the analysis of flooding in the Ayrshire region. ............................................197 Figure 7.4 - The Scottish Climate Change Adaptation Framework sector plans.......205
7
| List of tables Table 1.1 - Main legislation influencing climate change adaptation policy in Scotland.19 Table 1.2 - Overall structure of the thesis. .................................................................30 Table 2.1 - Examples of attributes used to differentiate types of adaptation ..............35 Table 2.2 - Conceptual differences between biophysical, social and integrated perspectives of vulnerability ....................................................................39 Table 2.3 - Differences and commonalities between Climate Impacts Assessment, Climate Change Vulnerability Assessment, and Adaptation Assessment42 Table 2.4 - Current challenges in performing integrated climate change vulnerability assessments...........................................................................................62 Table 2.5 - Examples of vulnerability assessment frameworks. .................................63 Table 3.1 - Main stages, approaches, methods and tools of the CCVA framework. ...72 Table 3.2 - Relation between SRES scenarios and UKCP09 emissions scenarios. ...82 Table 4.1 - Population in the Ayrshire local authorities ..............................................89 Table 4.2 - Climate projections for the West of Scotland ...........................................93 Table 4.3 - Typologies of flooding and mains causes ..............................................114 Table 4.4 - Ranking of past flood incidents in Ayrshire. ...........................................118 Table 7.1 - The analytical stages of the vulnerability assessment framework. .........177 Table 7.2 - Main challenges, compromises, and trade-offs whilst performing the CCVA for farming and flooding in the Ayrshire region. .....................................187 Table 7.3 - Addressing the conceptual elements of CCVA in this study. ..................191
8
| List of maps Map 3.1 - The Met Office 5km grid cells covering the Ayrshire region .......................79 Map 3.2 - The UKCP09 25km grid square covering the Ayrshire region ....................84 Map 4.1 - The Ayrshire region ...................................................................................89 Map 4.2 - Main towns, rivers, and landscape areas in Ayrshire .................................90 Map 4.3 - Main typologies of soil across Ayrshire ....................................................103 Map 4.4 - Land capability for agriculture across Ayrshire.........................................104 Map 4.5 - Less Favoured Areas in Ayrshire.............................................................106 Map 4.6 - Distribution of main types of livestock in Ayrshire ....................................107 Map 4.7 - Past flood incidents and ranking of impacts across Ayrshire.....................119 Map 5.1 - Agricultural parishes in the Ayrshire region..............................................127 Map 5.2 - Index of exposure per agricultural parish in Ayrshire................................132 Map 5.3 - The adaptive capacity of farming communities in Ayrshire.......................138 Map 5.4 - Index of current vulnerability of farming in Ayrshire..................................139 Map 5.5 - Index of future exposure - 2020s using a medium emissions scenario. ...146 Map 5.6 - Index of future exposure - 2020s using a high emissions scenario. .........146 Map 5.7 - Index of future exposure - 2050s using a medium emissions scenario. ...148 Map 5.8 - Index of future exposure - 2050s using a high emissions scenario. .........148 Map 6.1 - Data zones in Ayrshire.............................................................................153 Map 6.2 - The index of exposure to climatic risks in Ayrshire...................................158 Map 6.3 - The index of adaptive capacity across Ayrshire's communities to flooding.163 Map 6.4 - The index of current vulnerability to flooding in Ayrshire. .........................165 Map 6.5 - Index of future exposure - 2020s using a medium emissions scenario ....170 Map 6.6 - Index of future exposure - 2020s using a high emissions scenario ..........170 Map 6.7 - Index of future exposure - 2050s using a medium emissions scenario ....172 Map 6.8 - Index of future exposure - 2050s using a high emissions scenario ..........172
9
1 Chapter 1 | Introduction
1.1.
Introduction to the thesis “This is [...] about nightmares, catastrophes – and dreams. It is also about the everyday, the routines that give our lives continuity and substance. It is about the warming of our planet – a phenomenon which, if it proceeds unchecked, constitutes an existential threat to our civilization” (Giddens, 2011, p. 1).
The complexity of climate change and its impacts on societies represents one of the biggest challenges of our time. The ways in which it will affect natural and social systems is still not completely understood, making it difficult to prepare society for potential changes in climate. A key issue in tackling and dealing with future climate change is to understand vulnerabilities in order to adapt, as far as possible, to such changes. Initiatives and measures to adapt to climate change can be strategically planned in order to ensure that those communities most vulnerable are adequately prepared to deal with changing climatic conditions. However, different geographical areas and social groups are susceptible to suffering in different ways due to the local characteristics that define them (e.g., climatic, biophysical, social, economic, and political). To understand this differential vulnerability it is necessary to perform assessments of how different social and natural systems will be impacted by climate change. Only through this type of examination is it possible to inform and develop adequate adaptation actions and policy to tackle a changing climate. This thesis therefore seeks to interrogate current approaches to the assessment of vulnerability, testing the claims made in the climate change literature by putting them into practice through a case study in one Scottish region. This chapter introduces the overall context and key concepts on which the development of the thesis is based. Sections 1.2, 1.3 and 1.4 describe the overall problematic of climate change, the emerging importance of adaptation to climate change in recent years and its governance as a ‘wicked’ policy problem. To help situate the empirical context of the thesis section 1.5 introduces the current governance model in the UK and Scotland and section 1.6 outline current approaches to the management and governance of climate change. Examples of studies in Scotland looking at climate change impacts and vulnerability are described in section 1.7 before section 1.8 finishes the chapter by explaining the aims and contributions of the study and the overall structure of the thesis.
11
1.2.
The problematic of climate change The latest reports by leading scientists now confirm that the warming of the
climate system is unequivocal and that observed increase in global temperature is very likely to be due to the increase of greenhouse gas (GHG) emissions concentrations in the atmosphere (Parry et al., 2007). The work by scientists such as Jean-Baptiste Fourier in 1824 and later by John Tyndall on atmospheric dynamics greatly contributed to our current understanding of the global warming phenomena. Such scientific efforts unravelled the so called greenhouse effect where GHG act to absorb and emit the energy in the earth’s atmosphere keeping the earth’s surface warmer than without an atmosphere (Le Treut and Somerville, 2007; Bulkeley and Newell, 2010). However, although a naturally occurring phenomenon, the greenhouse effect has been exacerbated since the industrial revolution by human activities (e.g., burning of fossil fuels, clearing of forests) which have significantly increased the amount of GHG released into the atmosphere (Le Treut and Somerville, 2007) leading to an increase in global average temperature with further consequences such as changes in precipitation patterns and other climatic variables (Solomon et al., 2007). Climate change is associated with significant and long lasting changes in the state of the climate by changes in the mean and/or variability of its properties (Solomon et al., 2007; Lavell et al., 2012). These changes tend to fall into three broad temporal groups: long-term changes, inter-annual and decadal variability, and extreme events (Smit et al., 1999; Smit et al., 2000). These typologies of changes in climatic conditions are however inter-dependent since long-term changes encompass climate variability (inter-annual/decadal) which in turn includes extreme events (Smit et al., 2000). Several assessments have shown that warming of the climate system due to anthropogenic causes is already having an impact on many of our biological and physical systems. These impacts however, can be difficult to discern due to other nonclimatic pressures on such systems (Parry et al., 2007). Nonetheless, projections for global climatic changes points towards the continuous rise of temperature accompanied by changes in precipitation patterns, sea level rise and more frequent extremes events such as heat waves and heavy precipitation (IPCC, 2007a). In this context, and despite efforts to cut down GHG emissions, the irreversibility of past GHG emissions makes adaptation to future climate change impacts imperative in a world of continuous climatic change (Parry et al., 2007).
12
1.3.
The emergence of climate change (adaptation) as a policy issue Environmental problems and concerns over the sustainability of humankind and
nature have been in the public and political spheres since the 1960/70s (Carter, 2007a; Giddens, 2011). Originally underpinned by preservation and conservation issues environmental concerns soon expanded into the wider problems associated with global warming and the potential impacts of a changing climate on humans. However, due to the global and transboundary nature of the problematic of climate change it soon became clear there was a need for concerted action across the international community to tackle such a complex problem (Carter, 2007a). The first international effort occurred with the first World Climate Conference in 1979 and throughout the 1980s and early 1990s a number of intergovernmental conferences were held to address scientific and policy issues and call for global action upon climate change (UNFCCC, Undated). In 1988, the World Meteorological Organization and the United Nations Environment Programme established the Intergovernmental Panel on Climate Change (IPCC) with a mandate to evaluate the risks of climate change and provide scientific evidence and support to governments (IPCC, 2010). However, it was only with the creation of the United Nations Framework Convention on Climate Change (UNFCCC) in 1992 that a comprehensive international response on climate change started to take shape (Galarraga et al., 2011). The UNFCCC convention report was signed in 1992 by 154 countries worldwide with the aim of reducing and stabilising GHG emissions at 1990 levels by the year 2000 (Bulkeley and Newell, 2010). The Conference of the Parties (COP) is the UNFCCC main decision making body1 where party governments’ representatives review and discuss the implementation of the convention text as well as other legal instruments to be adopted by the parties (UNFCCC, 1992). Since the coming into force of the convention report in 1994 a number of COP meetings have been held with consequent agreements and mechanisms being adopted to tackle climate change (Figure 1.1).
1
Other UNFCCC bodies include the Secretariat; the Conference of the Parties serving as the Meeting of the Parties to the Kyoto Protocol; subsidiaries bodies; and the Adaptation Fund Board (for more see http://unfccc.int).
13
Figure 1.1 - The global governance of climate change (Based on Bulkeley and Newell, 2010; UNFCCC, Undated).
The UNFCCC convention report established the need to consider and address both mitigation and adaptation to climate change, calling for all parties involved to act upon the responsibility to “Formulate, implement, publish […] regional programmes containing measures to mitigate climate change by addressing anthropogenic emissions by sources and removals by sinks of all greenhouse gases […] and measures to facilitate adequate adaptation to climate change” (UNFCCC, 1992, p. 5). Climate change mitigation is therefore associated with measures and activities aimed at reducing GHG emissions or enhancing the carbon sinks of such gases (e.g., oceans, biomass). Adaptation on the other hand, refers to any adjustments within a particular system (e.g., ecological, social) in response to actual or expected changes in the climate (Smit et al., 1999; Burton et al., 2001; Klein and Smith, 2003; Parry et al., 2007; Füssel, 2007). The capacity to reduce overall impacts and the potential global scale, long-term benefits meant that mitigation measures have traditionally been given more consideration in climate change research and policy than adaptation which was initially 14
perceived as limited in its ability to deal with climate change due to its focus on local scale and short-term measures (Burton et al., 2002; Burton, 2003; Schipper, 2006). For some time it was also argued that a focus on adaptation (rather than mitigation) would reinforce a sense of procrastination and inactivity in terms of tackling the causes of climate change (Thompson et al., 2006; Burton, 2009). Scientific and political interest finally started to shift in the 1990s towards climate change adaptation due to recognition that the climate was already changing and there was a need to adapt to the already unavoidable impacts of climate change (McCarthy et al., 2001; Huq and Reid, 2004; Schipper, 2006). This shift led to an increasing interest in exploring and investigating issues of vulnerability and adaptation (see e.g., Janssen et al., 2006) as well as the development of methodologies and analytical frameworks to support the formulation of climate change adaptation policy. Since then, the UNFCCC has established several programmes to support and fund the uptake of adaptation measures such as the National Adaptation Policies of Action, the Special Climate Change Fund, the Least Developed Countries Fund, and the Kyoto Protocol Adaptation Fund (Huq and Burton, 2003). In addition, numerous organisations and networks have emerged at various levels to address and respond to climate change in recent years. Examples of such organisations include the Climate Action Network, the Global Climate Network, Local Governments for Sustainability, Urban Climate Change Research Network, Climate & Development Knowledge Network, Climate Outreach and Information Network, amongst others. The impacts of a changing climate, which can either be negative or positive, will differ both spatially and temporally and will be experienced differently across social and natural systems. Such diversity of potential impacts requires a range of different adaptation policies and measures at various scales of intervention in order to reduce existing vulnerabilities to climate change. In this context, there has been a move towards understanding adaptation at the local level. Looking at the agricultural sector, for example, Burton and Lim (2005) highlight the importance of developing national [adaptation] policy based on vulnerability assessments performed at the local level in order to ensure the adequacy of such policies and the enhancement of adaptive capacity. Hence, whilst developing adaptation policy is crucial in the overall context of climate change, such policy needs to be informed by assessments of vulnerability at a level of analysis below the national scale due to the differential characteristics and impacts that climatic change may have across different communities and geographical areas. In the UK, such efforts have been mainly led by government at various levels particularly at the local government level (Tompkins et al., 2010).
15
However, understanding local level adaptation requirements is far from simple as climate change is an extremely complex problem where uncertainty at various levels persists. Moreover, the political context within which climate change is embedded also defines and dictates the ways in which policy is developed and implemented. It is therefore crucial to work towards solutions that can be developed and applied in practice as climate change impacts are unavoidable and adaptation solutions will be fundamental to protect those most at risk. In this context, it is imperative to understand two separate but intertwined dimensions regarding the problematic of climate change adaptation:
How does the governance and policy context in which climate change adaptation takes place influence the capacity of those responsible to act upon and implement such adaptation policy? And;
How can we conceptualise and assess climate change vulnerability in order to inform adaptation policy?
The sections below explore these two questions in more detail with particular regard to the Scottish context which forms the empirical focus of this study.
1.4.
Climate change as a governance challenge Despite prevailing uncertainty regarding the timing and impacts of climate
change, the risks and adverse consequences associated with such changes are already clear (Schellnhuber et al., 2006). Efforts to stabilise GHG concentrations in the atmosphere will require drastic reductions to current levels of emissions. Some governments have already expressed serious commitments to achieving this, such as the Scottish Government which has a target for reducing emissions by at least 80% by mid century (The Scottish Parliament, 2009). However, despite mitigation efforts societies will have to adapt to future changes in climate, due to the time lag of the climate system’s response to long-lived GHG once emitted in the atmosphere. Hence there are consequent impacts to which we are already committed (Meadowcroft, 2009). The extremely complex nature of climate change, however, makes it one of the most pressing wicked problems of our time. Initially conceptualised by Rittel and Webber (1973), the term ‘wicked’ is nowadays used to define and capture problems with a set of common characteristics including (Leach and Percy-Smith, 2001):
A multifaceted nature which cannot be resolved by any one level of government and do not fit easily within existing government’s departmental structures (i.e., cross-cutting issues);
Only certain facets of this type of problem can be addressed at the local level; 16
They require interventions the benefits of which will only become apparent in the long term and, as a result, normally fall beyond the scope of typical strategies and plans.
More recently climate change has been described as a “super wicked” by Levin et al. (2010) due to its further exacerbating features. These include the risk of such as running out of time to stop or reverse climate change impacts due to the lengthy lag effects of the climate system’s responses; the lack of central authority regarding responses that need to be addressed to tackle the climate change problem; that those seeking to end the problem are also those causing it since GHG emissions result from activities occurring at various scales across societies; and the fact that even in the face of irrefutable evidence of significant climate change impacts, decision and policymakers tend to disregard this information and base their decisions on short-term horizons (Levin et al., 2010). An important feature of such complex issues “(...) is the need for policy to pay attention to the specificity of local circumstances [...] a centrally determined, ‘one-size-fits-all’ policy regime is unlikely to be effective” (Leach and Percy-Smith, 2001, p. 199). To date efforts to address the problem of climate change have differed between individual countries with some adopting significant measures whilst others continue to ignore the problem (Meadowcroft, 2009). However, as anticipated climate change impacts become clearer in the coming years, the international community will continue to push for action and in this context, “(...) adopting an appropriate response to climate change will be increasingly considered as a normal component of what ‘good governance’ entails” (Meadowcroft, 2009, p. 3). The legislative and policy context also influences the processes through which adaptation policy is developed and implemented. As a result, it is important to understand the governance context influencing responses to climate change. The following sections will describe the governance model adopted in the UK and Scotland and some of its characteristics that will help us to understand the ways in which the wicked problems of climate change adaptation are governed in Scotland.
1.5.
Governing climate change adaptation in Scotland Over the past two decades the concept of governance has emerged in social
sciences (Adger and Jordan, 2009; Pierre and Peters, 2000). Simply put the concept of governance can be understood as the “patterns that emerge from governing activities of social, political, and administrative actors” (Kooiman, 1993, p. 2).
17
In the UK, governance is associated with changes in the role and nature of the state as a result of public sector reforms2 in the 1980s and 1990s (Bevir, 2009). These reforms were largely characterised by the marketization of public services and by the creation of networks and partnerships to improve co-ordination between agencies to enhance institutional arrangements (Bevir, 2009). Two new features characterise the governance model – networks and Multi-Level Governance (MLG) – and influence the ways in which policy is developed and implemented on the ground. Whilst networks can be understood as “(...) a group of interdependent actors and the relationship among them” (Bevir, 2009, p. 137), having the capacity to cross boundaries between different levels of government and across sectors (Newman, 2001; Eckerberg and Joas, 2004; Bevir, 2009); MLG correspond to new forms of interaction between policy actors at various levels concerning “(...) the set of interactions that occur between policy actors across more than one level of government” (Bevir, 2009, p. 134). In an increasingly complex and dynamic world the need to address the wicked nature of climate change, it is imperative to adopt approaches such as these as it allows for different modes of governing, processes, and actors across different scales of intervention to take part in the process of governing complex phenomenon such as climate change ( Betsill and Bulkeley, 2006; Storbjörk, 2007) In Scotland, climate change policy is significantly influenced by these features which not only frame the formulation and implementation of those policies at various levels but also condition the role of different levels of government in their delivery. Although a fairly new effort, the climate change agenda in Scotland is quite an ambitious one. Since the early days of devolution from Westminster, Scotland has been trying to position itself as one of the world leaders in the climate change race (The Scottish Government, 2009). The devolution of powers in 1998 gave Scotland legislative powers over a number of policy areas namely the environment, local government, agriculture, health, and education (Keating, 2005). These new administrative and policy powers led not only to the emergence of a new political system but also to new relationships between Scotland, Westminster, and the European Union (McGarvey and Cairney, 2008). It also led to the emergence of different types of policies ranging from those completely devolved to Scotland to those areas and functions shared with the European Union (Keating, 2005). 2
These public sector reforms were related to particular political ideologies such as the free market neoliberalism of Margaret Thatcher’s conservative administration and later on the “Third way” approach of the New Labour government of Tony Blair (Leach and Percy-Smith, 2001). These reforms marked important shifts in the meaning of government, in the processes of governing, and in the rules and methods by which society is governed (Rhodes, 1997). Such changes in the forms of governing were also accompanied and influenced by other phenomena such as globalization and the emergence of the European (Newman, 2001).
18
As a result, the setting in which climate change policy is developed, and governed, spans across multiple levels and includes a range of different actors from different sectors. For example, operating at a global level the UNFCCC is one of the leading institutions regarding international negotiations on climate change policy (Bulkeley and Newell, 2010). Scotland, as part of the UK, by ratifying the UNFCCC convention report (and other instruments to address climate change e.g., Kyoto Protocol, Cancun Agreements) binds itself to those intergovernmental agreements and targets (The Scottish Government, Undated). Hence, any policy developments within Scotland regarding climate change (e.g., strategic and legislative frameworks) are bound by those global level agreements. Similarly, the European Union also plays an important policy-setting role in Scotland. Since joining the EU in 1973, the UK government (and later the Scottish Government) has been increasingly influenced by European treaties, directives and regulations. In fact, many policy areas in the UK are, to different extents, affected or rooted in a European dimension (Moran, 2005). Climate change policy is no exception as it is influenced by the European Climate Change Programme and its policy on adaptation (Table 1.1). Table 1.1 - Main legislation influencing climate change adaptation policy in Scotland.
Level
Entity
Legislation/policies
International
UNFCCC
UNFCCC Convention text
European
European Commission
White paper ‘Adapting to Climate Change: Towards a European Framework for Action’
UK
UK Government
Climate Change Act
Scotland
Scottish Government
Climate change (Scotland) Act
Moreover, despite being a devolved policy area in Scotland, the climate change agenda needs to have regard to UK legislation including the (UK) Climate Change Act 20083 which sets statutory targets of 80% reduction in GHG emissions by 2050 and provides a framework for concerted action across the UK (The Scottish Government, Undated). Thus, the formulation and implementation of policy in Scotland is influenced by a complex web of legislation and policy at various levels (international, European, and UK). In addition, efforts by the Scottish government to address climate change express the tendency for adopting networks and MLG approaches as the climate 3
The Climate Change Act 2008 also established the Committee on Climate Change (CCC), an independent advisory body to the UK Government on carbon budgeting and climate change adaptation. A further sub-committee on Adaptation (the Adaptation Sub-Committee) was also formed under the CCC banner to provide advice, analysis, and scrutiny of the Government’s adaptation programme and to ensure the UK is effectively prepared for future climate change impacts.
19
change agenda is expected to be implemented and delivered by a range of actors across society including the public sector, local government, private and third sectors, and civil society (The Scottish Government, 2011). This agenda is largely underpinned by the Climate Change (Scotland) Act 2009 (hereafter referred to as “the Act”) which establishes the targets, functions, duties as well as other provisions regarding climate change mitigation and adaptation in Scotland (The Scottish Parliament, 2009a). The Act establishes the legal requirement for an overall reduction in GHG emissions of 42% by 2020 (relative to 1990) and introduces the legal framework for a strategic and coordinated approach to adaptation in Scotland (The Scottish Parliament, 2009a). The Act also states the responsibility of Scottish Ministers to present a programme to the Scottish Parliament setting out their strategic objectives for climate change adaptation and their proposals and policies to achieve those objectives (The Scottish Parliament, 2009a). In the case of climate change adaptation, the Scottish Climate Change Adaptation Framework (SCCAF) is a non-statutory framework and a forerunner to the Adaptation Programme expected to be published in 2012/13. The SCCAF aim is to lead on climate change adaptation across all sectors in Scotland in order to “(...) increase the resilience of Scotland’s communities, and the natural and economic systems on which they depend, to the impacts of climate change” (The Scottish Government, 2009b, p. 3). To achieve that, the SCCAF coordinates Sector Action Plans which address specific issues within key sectors in Scotland (e.g., agriculture, water resource and management, energy, the built environment, health and wellbeing) (The Scottish Government, 2009b).The SCCAF is to be pursued and delivered by a range of actors across different sector areas, partnerships and policy networks (The Scottish Government, 2009e). Another fundamental aspect of the Act refers to the duties on public bodies regarding climate change. These duties require Scottish public bodies4 to contribute to the delivery of GHG emissions targets and climate change adaptation programmes, and to act in a way that they consider most sustainable (The Scottish Government, 2011). The next section looks more closely at the organisation of climate change adaptation policy responses at the local level in Scotland.
1.5.1. The climate change agenda at the local level in Scotland As discussed above, the MLG context of climate change adaptation policy implementation in Scotland is pursued through the SCCAF. The delivery of this 4
In this context public bodies include: Ministers, the Parliament; Local Government; Educational Institutions, Police, The National Health Service amongst other (Schedule 1 of the Freedom of Information, Scotland Act 2002).
20
strategic adaptation framework relies on a range of actors, at various levels of intervention. At the local level, these actors include local communities, local government, the third sector, as well as the private sector (The Scottish Government, 2009b). However, of all these local actors local government is the only public body considered as a major player in the delivery of the climate change adaptation programme (The Scottish Government, 2011). Hence, as a public body and major player5, local government is expected to help deliver and mainstream climate change into their corporate and strategic processes, as well as their actions. Local government’s role in climate change adaptation is also fundamental due to their proximity to local communities and their knowledge of local conditions which “allows adaptation actions to be tailored effectively to localised impacts of climate change” (The Scottish Government, 2009b, p. 27). Local government is responsible for representing local communities, facilitating Community Planning, and innovating and promoting well-being in their local area (McConnell, 2004). This proximity and interaction with local communities provides local government with knowledge and experience of local characteristics and conditions as well as an understanding of existing (and potential future) vulnerabilities of those communities to a changing climate. “Scotland’s communities will often be in the front line in responding to the impacts of climate change and local authorities are ideally placed to lead the community response to climate change. With knowledge of local values, industries, and landscapes, local government allows adaptation actions to be tailored effectively to localised impacts of climate change.” (The Scottish Government, 2009c, p. 27).
Scottish local government can therefore act as an intermediate platform between local communities and central government with regards to understanding of climate change impacts at the local level and potentially feedback such knowledge to higher levels of policy. In addition, local government has certain statutory duties and is responsible for a number of functions ranging from public service provision to implementation of national legislation (McConnell, 2004). The range of services provided by local government can be extremely relevant to help deliver the climate change adaptation agenda in Scotland namely through land use planning, flood management, 5
Major players are those public bodies whose influence or impact on climate change is considered greater than others. These include: public bodies with high impact and influence (e.g., Scottish Government, local authorities), and other public bodies with large estates and large numbers of staff, with large expenditure, or that provide auditing or regulatory functions (The Scottish Government, 2011).
21
environmental and social services, emergency planning, and building control (UK Committee on Climate Change, 2011). However, different policy areas entail different MLG mechanisms and networks for delivering specific agendas in those areas. In the case of climate change, this situation is further exacerbated by its overarching and complex nature which naturally links across many policy areas within local government (e.g., flooding, planning, tourism). This creates a multiplicity of areas of policy intervention at the local level, each of which may involve local government to varying degrees. On the one hand, climate change as a distinct policy area in itself encompasses and links to other policy areas; on the other hand, there is a wide range of policy areas that exist independently of the climate change agenda but influences its outcomes (e.g., flooding management) (cf. McEwen, 2010). Hence, the role and agency of local government sits at the confluence of different policy areas each embedded in its own context of governance, influences, and policy networks. The role of Scottish local government in relation to the climate change adaptation agenda is therefore two-fold. First of all, its proximity to, interaction with and knowledge of local conditions and existing risks across local communities represent a valuable contribution to understanding existing (and future) vulnerabilities to climate change at the local level (cf. McConnell, 2004). This potentially allows a more thorough knowledge and holistic understanding of local vulnerabilities to changing climate conditions of communities in the context of local characteristics e.g., socio-economic conditions, topography, geology. On the other hand, in the context of its statutory duties and responsibilities local government can, to different extents, act upon existing (and future) vulnerabilities to climate change by implementing adaptation policy and measures at the local level. This permits delivering adaptation actions that are more adequate to the scale at which the vulnerabilities and impacts of climate change are felt and therefore allows a closer integration between the scale of the impacts of climate change and the scale of agency at the local level. The role and involvement of local government with local communities is therefore fundamental for an effective public sector response to climate change in Scotland (The Scottish Government, 2011). Another important aspect to consider is the interdependency6 relationship between central and local government in Scotland. Ultimately, local government’s roles and duties are performed according to existing legislation and within its competences and statutory duties. As a result, its autonomy is bounded by central government which 6
This relationship is one of interdependency as on the one hand central government expects local government to deliver public services efficiently and according to national objectives and priorities; tailor service provision to the needs of local communities; to adopt democratic values; and to conduct themselves according to the political, financial and social parameters set by the government. Local government on the other hand, requires from central government the legislative and policy framework for delivering services to their communities; the opportunity to engage in policy consultation; and legal, financial and political autonomy to represent their communities (McConnell, 2004).
22
to a great extent determines how it delivers its services and performs its duties (McConnell, 2004). Nonetheless, local authorities still have considerable discretion regarding how their budgets are allocated although these need to meet the strategic objectives of the Scottish National Performance Framework (The Scottish Government, 2011d; McEwen, 2010). Existing mechanisms and processes such as the Single Outcomes Agreements7 and Community Planning Partnerships8 (CPPs) are in place through which local government is expected to deliver the climate change agenda. Other important mechanisms and tools to compel local government to action on climate change include Scotland’s Climate Change Declaration9 whose signatories aim is to tackle climate change and make a commitment towards action within their local communities (UK Committee on Climate Change, 2011).
1.6.
Climate change vulnerability assessments in Scotland “Vulnerability assessment is a common tool for representing the potential for harm to occur within human and ecological systems of value in response to global climate change. The process of undertaking assessments can contribute to better understanding of community and environmental needs with respect to […] the identification of adaptation actions for vulnerability reduction” (Preston et al., 2011, p.177).
Over the past 40 years the climate in Scotland has been changing in terms of temperature and precipitation patterns which have been leading to drier summers, wetter winters and an increase of heavy rain events (SNIFFER, 2006). Continuous and incremental changes in climatic conditions in the coming decades will only exacerbate potential impacts on Scotland’s society and natural systems. Hence, understanding how the climate is changing and how this may effect and impact Scottish society will be fundamental to developing and implementing effective climate change adaptation (The Scottish Government, 2009b). To achieve that, a range of methodologies and analytical approaches can be pursued to assess the impacts and vulnerability to climate change. The climate change literature distinguishes between different types of climate change 7
Single Outcome Agreements are agreements assumed between Community Planning Partnerships (CPP) and central government setting out how the CPP is planning to work towards national strategic priorities for their local area taking into consideration the scope and purpose of the National Performance Framework (The Scottish Government, 2011d). 8 Community Planning Partnerships are bodies led by local authorities encompassing a range of actors in the public, private, and third sector which work together in order to deliver better outcomes through joinedup services. These tend to include the police, health boards, enterprise networks, transport partnerships, and local communities (The Scottish Government 2011d; McGarvey and Cairney, 2008). 9 All 32 local authorities in Scotland are signatories of the Scottish Climate Change Declaration.
23
assessments according to their focus, the scope and purpose of the analysis as well as the methods and tools utilised to perform such assessments (Carter et al., 2007). Typologies of assessments include for example, climate change impacts, vulnerability assessments, adaptation assessments, and risk-management frameworks (Füssel and Klein, 2006; Carter et al., 2007). Conventional approaches such as the IPCC tend to focus on the quantification of climate change impacts using top-down scenario-driven approaches to assess the impacts of a particular stressor or its effects on a particular sector (UNFCCC, 2004). Alternative methodologies such as the so-called bottom-up approaches focus on the underlying processes, factors, and conditions affecting the vulnerability of a particular system to climate change and, as a result, provide a better understanding of vulnerabilities at the local level (UNFCCC, 2005; Füssel and Klein, 2006). Pursuing both approaches (top-down/bottom-up) as complementary is increasingly perceived as instrumental to improve our understanding of climate change vulnerability, particularly regarding assessments to inform local adaptation policy development (Mastrandrea et al., 2010). In addition, the use of vulnerability assessments as a governance tool to inform and develop climate change adaptation policy needs to account for local characteristics and conditions in order to provide an adequate account of differential vulnerabilities. In the UK, the UK Climate Change Risk Assessment (CCRA) is a 3-year project looking into the potential impacts of climate change. This UK-wide assessment aims to provide information to central government in order to help prioritise the risks presented by climate change against other competing pressures and create the conditions necessary for adaptation policy (DEFRA, 2010). Based on a top-down approach, the CCRA provides outputs for Scotland at a national level identifying key risks, threats, and opportunities from climate change where the government needs to act (DEFRA, 2012). According to the CCRA, the main threats identified for Scotland included changes in soil conditions and reduction in rivers’ flow due to warmer and drier summers; increase of fluvial and coastal flooding due to changes in precipitation patterns; increased risk of diseases and pests affecting agriculture and forestry; and changes in biodiversity. The report also identifies some of the potential benefits of the anticipated changes in climate namely opportunities in terms of crops yields due to longer growing seasons and improvement of land classification as well as increased forestry productivity; and the potential to reduce the number of cold-related deaths (DEFRA, 2012). However, the CCRA does not encompass information on local impacts and vulnerability and, as a result, fails to provide the level of detail that would allow a more thorough understanding of existing climate change vulnerabilities at the local level in Scotland. 24
In Scotland, most academic studies on climate change have been primarily pursued using top-down approaches by focusing on a particular climatic stressor or on the effects of climate change upon a specific element of a system (e.g., a sector). Examples of studies on climate change impacts in Scotland include those focusing on flooding (e.g., Black and Burns, 2002; Prudhomme et al., 2003; Cameron, 2006; Werritty et al., 2007); on crop growth, yields, and soils (e.g., Peiris et al., 1996; Cooper, 1997; Butterworth et al., 2010); on specific fauna and flora communities (e.g., Milne and Hartley, 2001; MacLeod et al., 2005; Trivedi et al., 2008;); and on the tourism sector (e.g., Harrison et al., 1999; Yeoman and McMahon-Beattier, 2006). Since coming into force in 2009, the Climate Change (Scotland) Act has pushed for further actions on climate change which has been pursued in practice through various mechanisms, institutions, and across the public, private and third sectors. This has not only placed duties on public bodies in terms of delivering the climate change agenda but also promoted new initiatives across the board. As a result, other projects and studies of climate change impacts and vulnerability have emerged across Scotland supported by organisations such Adaptation Scotland10 and the Sustainable Scotland Network. Examples of such initiatives include the Climate Ready Clyde 11 and the A9 Project: Climate Change in the Central Highlands12. At the local government level, efforts have also been made towards integrating actions to tackle climate change. A study by the Scotland and Northern Ireland Forum For Environmental Research (SNIFFER) (2005) for example, examined the various activities by Scottish local government with regards to climate change. The study provides valuable insights into how climate change has been addressed at the corporate level, in strategic decision-making, and in service provision across the 32 local authorities in Scotland. It indicates that, despite efforts to take on board the agenda, progress made to date is uneven, as whilst some local authorities have rapidly started taking on board the climate change agenda others were considered yet to develop a concerted response (SNIFFER, 2005). Overall, one of the main barriers identified by local authorities to help further their response and support decision-making and policy was the lack of information and data particularly regarding the “(...) likely climate changes in each region of Scotland and detailed information on how predicted climate changes were likely to interact with local topography and geology” (SNIFFER, 2005, p. 8). The study concludes by stating that “(...) it is clear that further action will be required both by local authorities and other 10
Formerly the Scottish Climate Change Impacts Programme. More information on this project is available at http://www.adaptationscotland.org.uk/4/110/0/Areabased-project--Climate-Ready-Clyde.aspx 12 More information on this project available at http://www.adaptationscotland.org.uk/4/88/0/The-A9Project--Climate-Change-in-the-Central-Highlands.aspx 11
25
agencies to ensure an integrated local response is developed in all areas of the country” (SNIFFER, 2005, p. 27). However, the national climate change agenda in Scotland is underpinned by the outputs from the CCRA which, with its sectoral approach in the analysis of climate change impacts, fails to provide the necessary level of detail at the more local level. Ultimately, it will only be possible to produce a more concerted response to tackle climate change once local level actions on climate change adaptation are based on local assessments13 of vulnerability and impacts that directly affect local communities.
1.7.
Aims and contribution of this research project In the context of climate change adaptation there are two fundamental
dimensions that need to be addressed if adaptation to climate change is to take place adequately and effectively at the local level14. On the one hand is the need to understand the nature of the problem at hand i.e., ‘who’ or ‘what’ is vulnerable to ‘what’ climate change impacts and ‘how’, in order to be able to develop and implement appropriate adaptation measures. To understand such vulnerabilities to climate change different methodologies of analysis can be pursued at different scales. However, given the spatial disparities of climate change impacts across different social and natural systems it is crucial that the assessment of such vulnerabilities is performed at a scale of analysis that provides not only the necessary level of detail to inform the development of relevant adaptation measures but also at a scale where the necessary agency and authority to act upon such outcomes, exists. On the other hand, there is a need to understand the contexts in which climate change policy is developed, implemented, and governed by the multiple actors involved in the policy process. As stated by Betsill and Bulkeley “the governance of climate change is a complex, multilevel process” (2006, p. 154). In Scotland the climate change agenda is embedded in supra-national spheres of influence (UK, European, and international) which establish programmes and targets for action through legislation, regulation, and policy instruments. Similarly, the Scottish Government’s climate change agenda ‘cascades’ down to public bodies, including local government, which are expected to help deliver climate change measures. As a result, it is crucial to understand the various powers, duties and responsibilities of local actors and
13
In this study, local level assessments are understood as those analyses performed below the national level (i.e., sub-national level) such as the regional and local authority levels. 14 These will be further explained and situated in relation to the literature in chapter 2 below.
26
particularly local government and how they can help to deliver the climate change adaptation agenda in Scotland. This project seeks to make a contribution to current understanding and knowledge of climate change vulnerability assessments (CCVA). The original proposal for this research project focused on the assessment of local communities’ vulnerabilities to climate change in order to provide an overall understanding of the main areas and sectors at risk in Scotland. However, it soon became evident that the processes of performing such analyses were significantly undermined by a range of challenges and limitations. As a result, the present study is a meta-analysis of the process of conducting a climate change vulnerability assessment in Scotland at the local level. To achieve that, an integrated vulnerability assessment framework is adapted and applied to a case study area – the Ayrshire region – in order to conduct an assessment of vulnerability to climate change focusing on two particular policy areas: farming and flooding. The thesis therefore seeks to understand how such processes of analysis can be conducted in Scotland at the local level but also to explore the main challenges, barriers, and trade-offs encountered in performing such assessments. It therefore also aims to understand how the conceptual and analytical elements described in the climate change literature as fundamental to the analysis of vulnerability can be pursued in practice in Scotland at the local level. In addition, the thesis also explores existing agency at the local level in Scotland to act upon the outcomes of such local vulnerability assessments. This will be examined in relation to the two policy areas – farming and flooding – utilised to perform the analysis of vulnerability to climate change in the Ayrshire region. The thesis therefore sets out to answer three research questions: 1. What are the challenges and analytical trade-offs of performing climate change vulnerability assessments at the local level in Scotland?
2. How can the conceptual elements of what a climate change vulnerability assessment entail be pursued in the analysis of vulnerability at the local level?
3. What is the agency at the local level to act upon the outcomes of climate change vulnerability assessments for local climate change adaptation policy in Scotland?
The first research question will be explored by performing a CCVA in a ‘real’ case study region – the Ayrshire region. By doing so, it was possible to test and 27
examine how the various conceptual and analytical elements of CCVA (chapter 2) can be addressed and implemented in practice (chapters 4, 5, and 6) as well as the current challenges, analytical trade-offs and compromises made whilst performing the CCVA in Ayrshire (chapter 7). In addition, the answers to the second research question set out in this study were provided by examining the challenges and limitations that emerge from practical engagement with the claims found in the climate change literature (chapter 2) through a meta-analysis of how the conceptual and normative ideals of what a CCVA should encompass were achieved (chapter 7). The third research question, of a more descriptive nature, will describe and explore the existing legislative and policy context in relation to farming and flooding, including the structures of multi-level governance and the policy networks in place to manage and act on climate change adaptation policy at the local level. The agency of local government in relation to the farming and flooding policy areas will be examined to assess its possible contribution to the delivery of climate change adaptation policy at the local level (i.e. how can the outcomes of local CCVA be included in decisionmaking? How can such local outcomes provide input to national policy-making process?).
1.8.
Structure of the thesis The thesis is structured around eight chapters. This chapter has introduced the
literature on climate change governance and policy in Scotland including the Scottish devolution and the Government’s climate change adaptation agenda and the main statutory duties, role and functions of Scottish local government in governing climate change at the local level. The next chapter provides the theoretical and conceptual background to the thesis by expanding on the literature on climate change governance, adaptation and vulnerability theory. It then focuses on the concept of climate change adaptation and reviews the main conceptual perspectives on vulnerability to climate change before assessing the various conceptual and analytical elements normally addressed when performing integrated vulnerability assessments. The chapter ends by presenting some examples of analytical frameworks to perform such analyses. Chapter 3 introduces the overall approach adopted as well as the conceptual framework utilized to conduct this research project. It then introduces the methodology utilized to perform the empirical investigation, therefore establishing the linkages between the literature review chapter (chapter 2) and the analytical chapters (chapters 4, 5, and 6).
28
Chapter 4 then introduces the case study area – the Ayrshire region – in the Southwest of Scotland. It describes its main social and economic characteristics as well as the main climate-related issues in the area and explains how the two analytical themes – farming and flooding – were selected to perform the assessment of vulnerability. The chapter also introduces the context of farming in Scotland and Ayrshire, characterizes farming practices and conditions in Scotland and Ayrshire and outlines relevant legislation and policy. It then characterizes the problematic of flooding in Scotland and Ayrshire and provides a description of the main legislation and policy regarding flooding in Scotland. Chapter 5 cover the analysis of climate change vulnerability regarding farming communities in Ayrshire. The chapter conceptualizes vulnerability in relation to farming practices in the case study region and describes the process of performing the vulnerability assessment of farming in Ayrshire. Chapter 6 examines the vulnerability to flooding in the case study region. The chapter conceptualizes vulnerability to flooding across Ayrshire and describes the process of performing the assessment of current and future vulnerability to flooding in the case study region. Chapter 7 draws together an overall integrated analysis based on the empirical chapters (4, 5, and 6) as well as conceptual chapter (chapter 2). This chapter describes the main challenges, limitations and trade-offs of performing climate change vulnerability assessments in Scotland at the local level, particularly regarding farming and flooding. The chapter then discusses existing barriers and limitations within local government to act upon the outcomes of such assessments. Following from that discussion, the chapter then presents key conceptual and applied/methodological contributions from the thesis which relate back to broader issues of how CCVA can be used as a governance tool in the context of climate change adaptation policy. Finally, a conclusions and further research sections at the end conclude the thesis.
29
Table 1.2 - Overall structure of the thesis.
Chapter 1 - Introduction
Introduction
Introduces the research problematic; Introduces climate change adaptation as a policy issue; Describes the governance of climate change in Scotland at multiple levels; Describes the context of climate change adaptation in Scotland; Presents the aims and structure of the project.
Chapter 2 – Climate change adaptation and vulnerability Literature review
Introduces the concept of climate change adaptation; Explores the literature on climate change vulnerability; Introduces the conceptual elements of integrated vulnerability assessments.
Chapter 3 – Conceptual framework and methodology Conceptual framework and methodology
Describes the overall orientation and research approach; Explains the conceptual framework developed for conducting the analysis; Describes the methodology utilized to perform the climate change vulnerability assessment.
Chapter 4 – Scoping stage: the case study region and vulnerable systems
Empirical investigation
Introduces the case study region and describes the rationale for the analytical themes; Describes the climate and some climate projections for the West of Scotland; Characterises the state of farming and grassland systems in Scotland and Ayrshire; Introduces farming practises and conditions across Scotland and Ayrshire Characterizes the problematic of flooding in Scotland and Ayrshire; Introduces flooding issues and conditions across Scotland and Ayrshire.
Chapter 5 – Assessing the vulnerability of farming to climate change in Ayrshire
Performs the climate change vulnerability assessment of farming in the case study region.
Chapter 6 – Assessing the vulnerability to flooding in Ayrshire
Performs the climate change vulnerability assessment in terms of flooding in the case study region.
Chapter 7 – Challenges of performing climate change vulnerability assessments in Scotland Discussion and conclusions
Describes the challenges and analytical trade-offs of performing CCVA at the local level in Scotland; Examines the challenges of pursuing the conceptual elements of CCVA in practice; Explores the existing capacity to act upon the outcomes of CCVA; Returns to the research questions set out for this research project and describes the main contributions of the thesis; Discusses limitations of the approach adopted, implications of the analysis and suggests directions for further research.
30
2 Chapter 2 | Climate change adaptation and vulnerability
31
2.1.
Introduction Chapter 1 introduced the problematic of climate change and the need not just to
mitigate but also to adapt to unavoidable future impacts of climate change. It was suggested that achieving effective adaptation to climate change requires both the capacity and resources to implement policy and actions, but also the necessary information regarding who or what needs to adapt to climate change (Füssel and Klein, 2006). Hence, it is fundamental to understand (existing or future) vulnerability to climate change. To achieve that, a range of tools have been developed to assess such vulnerabilities. In Scotland, central government initiatives to assess climate change vulnerability have, to date, focused on large scale analysis (such as the UK-wide Climate Change Risk Assessment) which arguably lacks the level of detail necessary to inform adaptation policy at more local levels. This chapter further develops some aspects of the problematic climate introduced in the previous chapter. Section 2.2 starts by further describing the concept of climate change; whilst section 2.3 introduces some of the key issues surrounding adaptation to climate change. Section 2.4 then goes on to discuss the concept of vulnerability and the various conceptual perspectives surrounding this term within the climate change literature. Section 2.5 then looks into the conceptual and analytical features regarded as fundamental when performing climate change vulnerability assessments and the main challenges and limitations associated with this type of analysis. Finally, section 2.6 presents some of the existing frameworks developed to conduct climate change vulnerability assessments15 and section 2.7 concludes the chapter.
2.2.
Introducing climate change in this study Climate is generally described as average weather (normally over a minimum
period of 30 years) and changes to the state of climate is understood as climate change (Le Treut and Somerville, 2007). Climate change is therefore broadly defined as any changes occurring in the state of climate over time which are attributed to human activities as well as natural climate variability (UNFCCC, 1992). It is associated with significant and long lasting changes in the state of the climate by changes in the mean and/or variability of its properties (Solomon et al., 2007; Lavell et al., 2012). These changes can be categorised in three main temporal groups: long-term changes, inter-annual or decadal variability, and extreme weather events. It should be noted however, that these typologies of changes are inter-dependent since long-term 15
Sections 2.3, 2.4, and 2.5 are largely based on work previously published as Bruno Soares et al. (2012).
32
changes encompass variability which in turn includes extreme events (Smit et al., 2009). Long-term changes relate to overall (observed or future) changes in the mean values of climate variables; whilst climate variability regards changes in the frequency, probability, and/or distribution of climatic variables over periods ranging from a few years to decades (i.e., annual or inter-decadal variability) (Smit et al., 1999; Smit et al., 2000). The third typology – extreme weather events – relate to the occurrence of a value above or below a threshold value near the upper or lower ends of the range of observed values in a climatic variable (Field et al., 2012). Hence, climate change should be regarded not only as changes in mean conditions but also in terms of changes in variability and in frequency and intensity of extreme events (Solomon et al., 2007; Lavell et al., 2012). However, understanding changes in climate variability and extremes is difficult due to the interactions between changes in mean and variability (which also vary depending on the climatic variable). For example, Figure 2.1 below illustrates the different effects that increases in both temperature mean and variance have on extreme temperatures. Whilst an increase in temperature mean without a change invariability leads to higher temperatures (figure (a) on top); an increase in the variability without a change in the mean lead to an increase in the probability of both ends of extreme temperatures as well as an increase in the value of the extreme (figure (b) in the middle); and an increase in both the mean and variability will affect the probability of both temperature extremes (hot and cold) as well as fewer cold events and more frequent hot events with more extreme high temperatures (figure (c) at the bottom) (Folland et al., 2001). Other combinations of possible changes in both mean and variability would lead to different effects on temperature extremes. As a result, “(...) it is often uncertain whether the changes are caused by a change in the mean, variance, or both” (Folland, 2001, p. 155). For other variables such as precipitation, changes in the mean can be accompanied by other changes (e.g., frequency of precipitation) which can also affect precipitation extremes and therefore makes it even more complex to understand the various aspects of precipitation extremes (Folland et al., 2001).
33
Figure 2.1 – Schematic representation of the effect of an increase in temperature variability, mean, or both on extreme temperatures (Folland et al., 2001).
However, in this study the different types of possible changes in climate will not be distinguished (e.g., between extreme weather events and long-term changes) in terms of the analysis performed. Hence, climate change will be addressed as long-term changes which encompass climate variability and extremes in those changes.
2.3.
Climate change adaptation The notion of adaptation has long been used in anthropological studies.
However, it was only in the 1990s, with the formation of the UNFCCC, that the concept of climate change adaptation started being considered (Schipper and Burton, 2009; Smit and Wandel, 2006; Burton et al., 2002). Since then a plethora of definitions of what adaptation entails have emerged within the climate change literature (for more see e.g. Smit and Wandel, 2006; Schipper, 2007). One of the most common definitions is the one by the Intergovernmental Panel on Climate Change (IPCC) which defines adaptation as any “adjustments in natural or human systems in response to actual or expected climatic stimuli or their effects, which moderates harm or exploits beneficial 34
opportunities” (Parry et al., 2007, p. 869) (for more adaptation definitions see e.g., Smit et al., 2009). The act of adapting to changing climate is not a new phenomenon as it has been common in human societies to make the best out of climate through available practices and methods (Tompkins et al., 2010; Burton, 2009; Füssel, 2007). However, considering the potential changes in global climatic conditions and the overall uncertainty attached to it brings new challenges to the fore including how we can address and manage such changes and ultimately, adapt our society to future climate change (Preston and Stafford-Smith, 2009; Füssel, 2007). Adaptation to climate change can occur in different systems ranging from changes in socio-environmental processes, people’s perceptions, practices and actions to reduce potential damages and even actions to take advantage of new opportunities that may arise from such changes (Adger et al., 2007). As a result, different types of adaptation can be pursued depending on the vulnerable system, the purpose of the adaptation, and the actors involved (Table 2.1). Table 2.1- Examples of attributes used to differentiate types of adaptation (Adapted from Smit and Pilifosova, 2003; Adger et al.,2007; and Füssel, 2007b).
Attributes Type of system
Examples Coupled human-environment; social; natural;
Purpose
Planned – autonomous
Timing
Anticipatory – reactive
Temporal scope
Short term – long term
Spatial scope
Local – widespread
Form
Technological – legal – institutional – structural - behavioural
Actors
Individuals – community – government
Most types of climate change adaptation only occur in social systems16, such as anticipatory adaptation for example, which occurs before the occurrence of the climate stimulus and/or its impacts in the system, or planned adaptation which normally takes place as a direct result of policy-making and implementation (Parry et al., 2007). Moreover, the capacity of social systems to plan and manage adaptation allows them to pursue goals beyond the mere survival of species such as exploring potential opportunities that may arise from a changing climate (Smithers and Smit, 1997). Adaptation can be prompted by private interests (e.g., individuals, households or organisations) or public interests (e.g., government) (Smit and Pilifosova, 2001; 16
In the context of this study, social systems are understood in their broadest sense including communities, economic sectors, regions, countries.
35
Adger et al., 2007). There is a general acceptance that private and public adaptations largely correspond to autonomous and planned adaptation respectively, due to the type of agency and purpose of these types of adaptations (Smit and Pilifosova, 2001). Adaptation to climate change can therefore be pursued and implemented by different actors, at different scales of intervention, each having different interests, resources, information, and vulnerabilities (Brooks and Adger, 2004). Adger et al (2005) argue that individual adaptation is not however, completely autonomous as individuals are also influenced and constrained by institutional processes (e.g., regulations, property rights, social norms). In fact, adaptation at all scales is influenced by institutional and structural factors some of which can extend beyond nation state boundaries (Adger et al., 2005). For example, the UNFCCC acting at an international level urges governments acting on a national scale to take action on adapting to climate (UNFCCC, 1992). As a response to such calls, governments can pass new legislation and regulations that affect and change the processes and norms which ultimately cascade down to subnational levels (e.g. regional, local, individual). Planned adaptation at the local level (e.g., local authorities, emergency services, local organisations) is therefore bounded by institutional processes and national legislation, regulations and policy and cannot be considered outside the regulatory context in which such actors operate (Adger et al., 2005). Climate change adaptation also occurs in a context of ongoing political, economic, demographic, technological, and cultural changes as well as other transformations associated for example with processes of globalisation and multi-level governance structures (cf. Adger et al., 2005). As such, adaptation tends to be an ongoing, long term process of learning and adjusting, reflecting many factors and multiple pressures not just those that specifically address climate change (Barnett, 2001; Adger et al., 2007; Tompkins et al., 2010). As a result, adaptation may at times be overtaken by other pressing issues such as economic restructuring, migration, and cost effectiveness, potentially leading to situations of “maladaptation” (O’Brien et al., 2006). Even with access to the necessary resources, knowledge, and capacity to implement actions there are factors that can hinder the uptake of adaptation measures including failure to achieve collective decision-making and uncertainty regarding information (Tompkins et al., 2010). In extreme cases, rather than mitigating vulnerability to climate change, ‘maladaptation’ may even exacerbate vulnerability particularly for those most at risk from climate change (Barnett and O’Neill, 2010). Although planned climate change adaptation efforts vary across the globe, they tend to be undertaken by governments, particularly at the local government level, 36
through long-term proactive and planned actions normally stimulated by climate variability and past events (Berrang-Ford et al., 2011; Ford et al., 2011). Focusing on the UK, Tompkins et al. (2010) explore how climate change adaptation is being pursued in practise by a range of actors. According to their study, initiatives to adapt to climate change tend to be led by the public sector at various levels. Major triggers for action were associated with experiencing real or perceived climate change impacts (e.g., drought, heat); legislation, including non-climate related policies at European and national level which influence adaptation action (e.g., EU agricultural policy subsidies); and experiencing both fluvial and coastal flooding. Planned public policy responses to climate change adaptation (which are the focus of this study) should have regard to the protection of those less capable of addressing their vulnerability to climate change, but should also provide information for planning adaptation by other actors and protect important public goods (e.g., public infrastructure, ecosystem services) (Tompkins et al., 2010). According to Füssel (2007b) one of the most critical factors for achieving effective planned adaptation is the need to assess and communicate vulnerability to climate change. As a result, understanding the vulnerability of a system and how climatic change may impact on it is important in order to adequately inform the development of adaptation policy (Füssel and Klein, 2006; Meadowcroft, 2009). Having defined the dimensions of both climate change and planned adaptation the following section examines the concept of vulnerability to climate change and the various conceptual and analytical approaches that currently permeate the climate change literature. It then describes the challenges and limitations associated with performing this type of analysis.
2.4.
Climate change vulnerability “Vulnerability has found its place into the climate change lexicon, with both natural and social scientists eager to measure and assess vulnerability, whether from the perspective of regions, sectors, ecosystems or social groups” (O’Brien et al., 2007, p. 74).
Contributions to vulnerability research from a range of different scientific communities have led to differing understandings and interpretations of what the notion of vulnerability to climate change means and entails. This has also contributed to the development of disparate metrics to assess vulnerability. The sections below describe
37
the main conceptual perspectives of vulnerability in the climate change literature as well as the main typologies of assessment associated with those perspectives.
2.4.1. The concept of climate change vulnerability The linguistic root of the word vulnerability is the state or quality of being vulnerable i.e., able to be wounded, liable to damage or harm (Oxford English Dictionary, 2002). In climate change research the development of this concept has been underpinned by a range of research strands from the hazards literature to human ecology, political economy, political ecology, food security, and development studies (Cutter et al., 2003; Janssen and Ostrom, 2006; Adger, 2006; Janssen et al., 2006; Füssel, 2007b; Mclaughlin and Dietz, 2008). Such multivariate contributions are clearly expressed by the array of vulnerability definitions and conceptualisations which overall contribute to what Hinkel describes as the ‘Babylon confusion’ in vulnerability studies (Hinkel, 2011). This range of understandings ultimately generates problems in developing a consistent definition of what vulnerability to climate change entails and the ways in which it is analysed and operationalised in practice (Preston et al., 2011). Smit et al. (2000) for example define vulnerability as the susceptibility of a system to damage or harm whilst Wisner et al. (2004, p. 11) describe it as “(…) the characteristics of a person or group and their situation that influences their capacity to anticipate, cope with, resist and recover from the impact of a natural hazard.” A third definition by the IPCC conceives vulnerability as “the degree to which a system is susceptible to and unable to cope with, adverse effects of climate change, including climate variability and extremes” (Parry et al., 2007, p. 883) . The first definition by Smit et al. (2000) is quite generalised as it does not specify the subject of analysis or the hazardous event in question. Conversely, the definition by Wisner et al. (2004) explicitly identifies social systems and their characteristics as the subject of analysis whilst recognising natural hazards as the source of harm to the system. The IPCC definition is broader in its scope regarding the subject of analysis but very specific regarding the event affecting the system which is climate change. In this conceptualisation, the subject of vulnerability analysis can differ e.g., an economic sector, a bioregion, or a community. The dissimilarities between the three definitions presented above are associated with distinct perspectives on what vulnerability to climate change involves. In climate change research these conceptual perspectives on vulnerability can be broadly categorised as biophysical, social, and integrated perspectives. These disparate perspectives tend to lead to the application of distinct methodologies to perform the 38
analysis
of
vulnerability
including
climate
impact
assessments,
vulnerability
assessments, and adaptation assessments. Despite conceptual and analytical differences all three perspectives therefore offer different ways of assessing vulnerability to climate change. Table 2.2 below presents some of the main conceptual and analytical differences between the three perspectives on vulnerability to climate change. Table 2.2 – Conceptual differences between biophysical, social and integrated perspectives of vulnerability (Adapted from O’Brien et al., 2007; Eakin and Luers, 2006).
Perspectives on vulnerability Biophysical
Social
Integrated
Focal point of analysis
Biophysical conditions and the hazard
Social systems and social conditions
Both biophysical and social conditions (i.e., the coupled humanenvironment system)
Type of analytical questions
What are the hazards and impacts?
How are people affected? How are they capable to cope with climate change?
How and why do coupled systems change? What is their capacity to adapt?
System of analysis
Sectors, regions, activities, places
Social groups e.g., communities, individuals
Coupled humanenvironment system, ecosystems
Time period of interest
Future climate change
Current climate variability
Current and future climate change
The biophysical view focuses on biophysical conditions and the exposure to hazards of a particular unit of analysis i.e., vulnerability as potential exposure to hazards; whilst the social perspective conceives vulnerability as a pre-existing condition of social systems prior to the hazardous event i.e., vulnerability as a state and/or propriety of a system (Dow, 1992; Brooks, 2003). The integrated perspective is an amalgamation of the two former perspectives and aims to address and integrate both the biophysical and social dimensions of vulnerability in the analysis. The following sections describe in more detail each of these theoretical approaches to the concept of climate change vulnerability.
2.4.1.1. Biophysical vulnerability Vulnerability, as a biophysical condition, addresses the vulnerability and degradation of environmental conditions and extrapolates these in terms of potential impacts on humans (Liverman, 1990). This approach, largely based on the natural hazards tradition, focuses on the distribution of hazardous conditions, human 39
occupancy within hazardous areas, and the degree of loss related to a specific hazardous event (Dow, 1992; Cutter, 1996). In this approach, vulnerability is regarded as an ‘end-point’ (i.e., the outcome of climate change impacts minus adaptation) as its main purpose is generally to provide an understanding of climate change impacts and inform decision-making regarding the costs of adaptation versus the costs of mitigation (O’Brien et al., 2007). The main focus is therefore the source of risk (based on the probability of a hazardous event) which determines the level of vulnerability and issues such as magnitude, duration, and impact of the climatic event normally characterize this type of study. These approaches are also known as risk-hazard approaches (Eakin and Luers, 2006; Turner et al., 2003); outcome-based approaches (O’Brien and Wolf, 2010); or impact-driven studies (Ford et al., 2010). However, although capable of providing an overall understanding of the physical processes generating exposure, this perspective is limited as it excludes the social, economic, political and cultural factors that need to be addressed in the estimation of vulnerability (Ford, 2002; Cardona, 2004).
2.4.1.2. Social vulnerability In the social perspective, vulnerability is conceived as a socially constructed phenomenon resulting from particular social, political, historical and economic processes and structures impacting on individuals/groups which can lead them to vulnerable conditions (Liverman, 1990; Cutter, 1996; Adger, 1999; Brooks, 2003). Contrary to the biophysical view, in this perspective vulnerability is regarded as a property, an inherent condition of a social system (e.g., individuals, communities, social groups). Largely based on political economy and political ecology research traditions (Eakin and Luers, 2006; Füssel, 2007b) this approach emerged as a reaction to conventional modernization theories in geography in the 1960s and 1970s (Hewitt, 1983; Blaikie et al., 1994; Greenberg and Park, 1994). Numerous conspicuous events and industrial accidents during the same period (e.g. The 1970 cyclone in Bangladesh; the 1986 Chernobyl nuclear disaster) (Tierney, 1999; Smith, 2001) also led to an increasing recognition of socio-economic conditions as the main causes of disasters and people’s vulnerability (O’Keefe et al., 1976). Many studies on social vulnerability are therefore related to chronic hazards and disasters in developing countries where, as a result of the hazard, the most vulnerable are affected by e.g. famine, hunger, and drought (Cutter, 1996; Adger, 1999). The focus is drawn to social systems and vulnerability is conceived as having two sides: an external side encompassing the perturbations and risks the system is 40
subjected to, and an internal side which includes the system’s own capacity to cope and respond to hazardous events (Chambers, 2006). As a result, issues such as resilience, sensitivity, resistance, and coping capacity are common in these type of studies (Dow, 1992; Ford, 2002). In this perspective, vulnerability is perceived as the ‘starting-point’ of the analysis where it is considered as a dynamic state resulting from social, environmental, political, and economic processes (O’Brien et al., 2007). This perspective is also known as contextual vulnerability (O’Brien and Wolf, 2010; Ford et al., 2010). However, by over emphasizing the social and political structures and processes generating vulnerability and by neglecting the hazard impact and physical damage, some of the studies using this perspective have only provided a limited understanding of vulnerability (Cardona, 2004).
2.4.1.3. Integrated perspective on vulnerability The limitations posed by the two conventional perspectives and the emergent notion that “any distinction between social and natural systems is arbitrary” (Adger, 2006) led to the appearance of new approaches to the study of climate change vulnerability. This conceptual shift was due to changes in the scientific knowledge of complex systems (e.g., interconnectedness of anthropogenic climate change at various scales; surprise and unpredictability in the day-to-day) and the appearance of new forms of social and political interactions (e.g., more participatory decision-making processes; wider acceptance of environmental issues, human rights). As a result, new ways of conceptualising and defining issues emerged (Gallopin et al., 2001) together with an appreciation that the complexity of problems and situations required better integration of the various dimensions of reality in complex systems. Complex systems such as the human-environment system are perceived as encompassing factors such as multiplicity of scales, irreducible uncertainty, non-linearity, and a range of legitimate perspectives (Gallopin et al., 2001). Integrated approaches to vulnerability aim to address both the biophysical and social dimensions of vulnerability (Füssel and Klein, 2006; Eakin and Luers, 2006; Ford, 2002). As a result, this perspective is broadly perceived as an amalgamation of the other two perspectives of vulnerability. In this view, “nature-society relationships are conceptualized as a mutuality, rather than as a duality” and vulnerability is determined by both biophysical conditions as well as social, political, economic, and institutional processes (O’Brien et al., 2007, p. 76). The process of conceptual integration is pursued by merging concepts from disparate views on vulnerability (Newell et al., 2005). By attempting to blend the two conventional perspectives on vulnerability, this 41
approach is perceived as capable of providing a better and clearer understanding of the multiplicity of processes and dynamics affecting the vulnerability of a system to climate change. This is particularly important in the context of policy-driven assessments aiming to provide measures to inform adaptation policy towards reducing vulnerability to climate change (Füssel and Klein, 2006). The integrated view on vulnerability is regarded as the current paradigm in the analysis of climate change vulnerability and adaptation. It provides a broad conceptual and analytical platform by allowing the integration and application of different conceptual backgrounds as well as a range of methods and tools which have the potential to complement each other and improve the information provided (Mastrandrea et al., 2010). However, the integration of disparate conceptual backgrounds (i.e., biophysical and social perspectives) can be problematic as it requires working with and blending different ways of framing and performing the analysis of vulnerability (cf. Table 2.2 above). This raises a series of analytical and conceptual tensions and challenges identified in the climate change literature and discussed in the following sections.
2.5.
Assessing climate change impacts and vulnerability The different theoretical approaches to vulnerability and adaptation contributed
to the development of different assessment methodologies and tools. Three main types of assessment and the differences and commonalities between them are described in Table 2.3. It should be noted that other types of assessments can also be found in the climate change literature such as risk management assessment (Carter et al., 2007). However, this type of assessment tends to be based on the analysis of risk and not vulnerability and, as a result, was not included in this study. Table 2.3 – Differences and commonalities between Climate Impacts Assessment, Climate Change Vulnerability Assessment, and Adaptation Assessment (based on UNFCCC, 2004; Füssel and Klein, 2006; Carter et al., 2007a).
Climate Impact Assessment (CIA) Main approach
Top-down/scenariodriven
Spatial scale
National to global
Climate Change Vulnerability Assessment (CCVA)
Adaptation Assessment
Bottom-up/vulnerability-driven Local to global
Local to national
42
Key methods and tools
Development of scenarios; downscaling techniques; sectoral impact models
Account for climate variability, nonclimatic factors, & adaptation
Little
Partial/full
Full
Degree of stakeholder involvement
Low (mainly research driven)
Medium (Research/ stakeholder-driven)
High (Stakeholders/ research-driven)
What are the potential biophysical impacts of climate change?
What is the vulnerability to climate change, considering feasible adaptations?
Which are the best adaptation measures for reducing vulnerability to climate change and variability?
Examples of research questions
Vulnerability indicators; past and present climate risks; stakeholder tools; case studies; risk assessment techniques; decision support tools; agent-based methods; narrative methods
Climate impact assessments (CIA), considered as the first type of assessment to emerge in climate change research (Füssel and Klein, 2006), were largely developed in the light of the biophysical perspective on vulnerability. This type of assessment looks at the long-term impacts of climate change on natural and human systems (Füssel and Klein, 2006; Parry et al., 2007; Carter et al., 2007). The IPCC methodology is considered the conventional approach to CIA (UNFCCC, 2004) although other frameworks have been developed and proposed since (e.g., Feenstra et al., 1998). The development of this type of study was driven by the need to understand the potential impacts of climate change and the level of urgency regarding stabilization of greenhouse gases (GHG) and respective mitigation agendas (UNFCCC, 2005). Growing evidence of the unavoidable impacts of climate change allied to slow progress on the reduction of GHG emissions however led to a shift in conceptual focus from a mitigation-based analysis to one focusing primarily on adaptation (Schipper, 2006). Concerns were now over the vulnerability of social systems and the way these may be affected by climate change. In this context, particular attention was drawn to the least developed countries (LDCs) as many have already experienced climate change impacts and related events (e.g., floods, drought) with serious consequences to people’s livelihoods, production systems and the environment (Mertz et al., 2009; Parry et al., 2007). Many LDCs also lack the capacity to adapt and respond to climate change due to general conditions such as fragile economies, low access to technology, generalised poverty, and high levels of inequality amongst populations (UNFCCC, 2011). As a result, an increasing interest emerged in the international community 43
regarding the analysis of vulnerability to develop and implement adaptation policy in those countries (Mertz et al., 2009). This resulted in the appearance of new vulnerability driven methodologies characterized by ‘bottom-up’ approaches more aligned with social and integrated perspectives on vulnerability. In analytical terms, a focus on current climate variability alongside adaptation and non-climatic factors marks the shift from CIA to Climate Change Vulnerability Assessments (CCVA) (Füssel and Klein, 2006). This shift is also associated with new approaches to stakeholder engagement, more sophisticated socioeconomic scenarios, and the consideration of adaptation measures, decisionsupport tools and enhancement of adaptive capacity as ways of reducing vulnerability to climate change (UNFCCC, 2005). Some CCVA termed ‘second generation’ further address relevant non-climatic drivers (e.g. economic, demographic), and the adaptive capacity of the system under analysis (Füssel and Klein, 2006). Adaptation assessment is a further type of methodology which focuses primarily on information requirements from stakeholders for policy development and facilitation of decision-making processes (Füssel and Klein, 2006). However, the distinction between CCVA and adaptation assessment is artificial as they share many analytical components, methods and tools, and both use bottom-up approaches to assess localized vulnerability and adaptation (see Table 2.3 above). They are therefore more attuned to local contexts and more capable of identifying local options and limitations than top-down approaches (UNFCCC, 2004). These different typologies of assessment are not mutually exclusive or competitive as they can complement each other and be integrated in the same assessment exercise (Ribot et al., 2009; Mastrandrea et al., 2010). The integrated perspective on vulnerability aims for that integration through the combination and articulation, whenever possible, of different methods and tools in the analysis. The sections below explore the conceptual and analytical elements that according to the climate change literature should be addressed when performing climate change vulnerability assessments.
2.6.
Conceptual elements of climate change vulnerability assessments As described above, the concept of climate change vulnerability is underpinned
by numerous theoretical contributions which have led to the development of different understandings of what vulnerability to climate change represents as well as a range of methodological approaches to assess such vulnerability. This multiplicity of contributions helped to frame and shape different conceptualizations of vulnerability 44
and consequently helped define the various conceptual and analytical elements regarded as fundamental when performing analysis of climate change vulnerability. These elements can therefore be regarded as normative ideals describing what should be addressed when conducting a climate change vulnerability assessment. These elements will be explored and described by responding to five questions: who or what is vulnerable to climate change? To what is the system of analysis vulnerable? Why is that system vulnerable? How to assess that vulnerability? And who is involved in the assessment? The section finishes by outlining some of the remaining challenges and limitations identified in the climate change literature with regards to performing climate change vulnerability assessments.
2.6.1. Who or what is vulnerable? The core element of any CCVA is the subject of analysis i.e. the vulnerable system under investigation in the vulnerability assessment. It can be a community, a region, an economic sector, a single specie or an entire ecosystem (Preston et al., 2011). In the biophysical view of vulnerability the onus of the analysis tends to be on the hazardous event (e.g., landslide, flooding), the biophysical factors generating exposure to such events, or the impacts on the exposure unit. Conversely, in the social vulnerability perspective the subject of analysis tends to be social systems (e.g., a community, a social group) and their vulnerability to climate change which tends to be associated with its (in)capacity to adapt or recover from climatic-related events. In integrated approaches, the coupled human-environment system17 (CHES) is generally recognised as the main subject of analysis (RASSP, 2001; Gallopin, 2006; Young et al., 2006; Renaud et al., 2010). This type of system is broadly conceptualised as encompassing both the environmental and social components and the interactions between them (Gallopin et al., 1989 in Gallopin et al., 2001). In CHES, “people and nature interact reciprocally and form complex feedback loops” (Liu et al., 2007, p. 1513) making it a synergetic and dynamic system that evolves through time and space. In this context, it is crucial to identify the attribute(s) of concern within the coupled system which will be under scrutiny when performing the CCVA (e.g., human lives, environmental degradation, income and cultural identity of a community) (Luers et al., 2003; Füssel, 2007b). This allows the identification of those linkages most important to the analysis of vulnerability.
17
Also termed coupled human and natural system (Liu et al., 2007); social-environment system (Eakin and Luers, 2006); and social-ecological system (Gallopin, 2006; Adger, 2006; Renaud et al., 2010).
45
Another important aspect regarding the CHES is the notion of place. This idea, re-introduced by Cutter’s model of ‘hazard-of-place’ focuses the assessment on a particular geographical area (Cutter, 1996). The central idea is that the CHES is intrinsically related to a particular area and therefore establishing that delimitation helps to identify the dimensions and scales of interaction and influence relevant to the coupled system under analysis. Moreover, a particular place is associated with a specific geographical context where institutional processes, communication, and decision and policy-making occur (Wilbanks, 2002). This loop of interaction between analysis of vulnerability and policy making can ultimately facilitate the development and implementation of adaptation policy on the ground. Many existing CCVA frameworks consider the CHES as the subject of analysis (Carter et al., 2007) and adopt the notion of place-based assessment (Cutter, 1996; Klein and Nicholls, 1999; Turner et al., 2003; Polsky et al., 2003; Ford and Smit, 2004; Lim et al., 2004). However, a very limited understanding of how human and environmental systems are coupled and the ways in which they interact remains problematic in vulnerability analysis due to the lack of empirical studies (Birkmann and Wisner, 2006; Liu et al., 2007). Liu et al. (2007) uncover some of the complexities and challenges in the analysis of CHES using six disparate case studies across the world. Their study reveals that the complexity of this type of system is still far from being understood and that there is a pressing need for more empirical analysis and more comprehensive portfolios to enhance knowledge regarding coupled systems (Liu et al., 2007).
2.6.2. Who is involved? In bottom-up approaches such as in CCVA there is a general understanding that local actors hold information and possess local knowledge regarding past (climatic) events. Their experiences and knowledge are perceived as valuable contributions to understand the context-specific conditions and existing capacity to adapt to such events (Ford, 2002; Swart et al., 2004; Carter et al., 2007). Moreover, engaging local actors in a CCVA is understood as a valuable process as it allows a more legitimate and transparent production of information, salient policy (particularly regarding adaptation policy) and greater credibility regarding the outcomes of the analysis (Conde and Lonsdale, 2005; Schröter et al., 2005; Meinke et al., 2006; Eakin and Luers, 2006). The framework by Polsky et al. (2003) for example, highlights the importance of involving stakeholders from the outset of the project arguing that ultimately they will be the ones responsible to act upon the outcomes of the CCVA.
46
A stakeholder is understood as ‘a person or an organisation that has a legitimate interest in a project or entity, or would be affected by a particular action or policy’ (Parry et al., 2007). Common stakeholder categories include the scientific community, policy makers, the public, the media, NGOs, and industry (Keskitalo, 2004). The type of engagement will vary depending on the role stakeholders are expected to play (e.g. informants, reviewers) (Carney et al., 2009) which in turn will determine their level of involvement (e.g., participation by consultation, selfmobilization) (Conde and Lonsdale, 2005) at particular stages of the assessment (e.g. at the beginning/end of the project) (Welp et al., 2006). An array of participatory methods and techniques exist to facilitate engagement at various levels such as focus groups, workshops, mapping techniques, development of scenarios, etc (see e.g., Elliot et al., 2005). An alternative approach to ‘local’ and ‘everyday’ knowledge integration in the scientific process is the so called the ‘co-production of knowledge’ through which the interaction of scientists, policy-makers and the public occurs. Such models however require not only a considerable commitment to stakeholder involvement and participation in the research process, but also interdisciplinarity (i.e., capacity and effort by scientists from disparate backgrounds to work together), and efforts towards producing knowledge that is usable (Lemos and Morehouse, 2005). Overall, a number of factors can potentially jeopardise the implementation of this type of model including lack of resources, political constraints, and the level of compatibility between scientists and stakeholders’ needs and expectations (Lemos and Morehouse, 2005). Moreover, critiques of public and stakeholder participation have been noted by Few et al. (2006), including of methods of engagement and whether they represent meaningful inclusion in the assessment process, and more conceptual and operational difficulties in achieving a broad public/stakeholder engagement i.e., who is able to participate (Few et al., 2006). Although recognised as a fundamental part of the vulnerability assessment process, involvement and engagement between those performing the analysis and those utilising the outputs of those analyses can be challenging. Not only in terms of who, when, and how to involve in order to achieve a significant and representative assessment of vulnerability, but also potential challenges when attempting to pursue co-production of knowledge between researchers and stakeholders.
47
2.6.3. To what is the system vulnerable? The assessment of vulnerability generally includes elements of exposure, sensitivity and adaptive capacity regarded as the key components or determinants of vulnerability (Carter et al., 2007; Smit and Wandel, 2006). According to the IPCC, exposure is the “nature and degree to which a system is exposed to significant climatic variations” (McCarthy et al., 2001, p. 987). In this context, the system is considered vulnerable when exposed to particular climate stimulus or conditions (such as precipitation and temperature) that may affect and impact the CHES (Smit et al., 2009). Sensitivity is generally conceived as “the degree to which a system is affected, either adversely or beneficially, by climate variability or change” (Parry et al., 2007, p. 881). Adaptive capacity18 is generally regarded as the “ability of a system to adjust to climate change (including climate variability and extremes) to moderate potential damages, to take advantage of opportunities, or to cope with the consequences” (Parry et al., 2007, p. 869). Perhaps more important than establishing definitional differences is to understand how these elements relate to each other in the various conceptualisations of vulnerability. In the biophysical view the analytical focus is on exposure to climate change and the sensitivity of the subject of analysis to that exposure. Vulnerability is therefore conceptualised and analysed based on these two components and adaptive capacity is normally not accounted for in the analysis. In the social perspective, vulnerability is conceptualised as a pre-existing condition of the system under analysis (Gallopin, 2006). Looking at the vulnerability of farmers in Mexico and Argentina, Eakin et al. (2008) analyse their vulnerability to climatic events by determining the differential sensitivities and capacity to adapt across the various farm households. Exposure is therefore regarded as an external element as vulnerability is conceptualised as a characteristic (and not as a predisposition to damage) of farm households as the result of susceptibilities and/or lack of capacity to adapt (Cardona, 2004). In integrated approaches vulnerability is perceived as a property of the CHES encompassing the dynamics between its social and natural components. As a result, in this approach exposure to climate change (which acts as the stimulus affecting the CHES) is addressed as an internal component of the vulnerability of the coupled system together with the system’s sensitivity to that stimulus and its capacity to adapt to the impacts of such changes in the system (Gallopin, 2006). The analysis of the key components of vulnerability under different typologies of assessment also differs. In the CIA model vulnerability is determined by the 18
Different meanings and understandings of adaptive capacity can be found in the literature (see e.g., McCarthy et al., 2001; Burton et al., 2002; Brooks, 2003; Turner et al., 2003; Brooks and Adger, 2004; Gallopin, 2006; Adger et al., 2007). However, it is beyond the scope of this study to discuss the various interpretations of adaptive capacity in the literature.
48
relationship between sensitivity and exposure of the subject of analysis to climate change (Figure 2.2); whilst in the CCVA model (Figure 2.3), other factors are accounted for in the analysis of vulnerability including climate variability, non-climatic factors and drivers, and the CHES’ capacity to adapt.
Climate change
Exposure
Sensitivity
Climate variability & change Vulnerability to climate change
Exposure
Non-climatic factors & drivers
Sensitivity
Adaptive capacity
Climate change impacts on the CHES Climate change impacts on the system Figure 2.2 - Simplified Climate Impact Assessment model (Based on Turner et al., 2003).
Vulnerability to climate change Figure 2.3 – Simplified Climate Change Vulnerability Assessment model (Based on Füssel and Klein, 2006).
Accounting for adaptive capacity in a CCVA is fundamental as it translates the existing and/or potential capacity to perform adjustments in response to climate change. Whilst adaptation is the occurrence of adjustments in a system in response to climate change, adaptive capacity is the (existing and/or potential) capacity of a system to perform those adjustments or, as Smit and Wandel (2006) posit “(…) adaptations are manifestations of adaptive capacity, and they represent ways of reducing vulnerability” (Smit and Wandel, 2006, p. 286). As vulnerability is generally understood as a function of climate change impacts (exposure and sensitivity) and adaptive capacity, the system’s capacity to adapt is therefore a fundamental determinant of the system’s vulnerability as it represents its ability to deal with potential impacts of climate change (Gallopin, 2006; Smit and Pilifosova, 2003; Brooks and Adger, 2004). In this context, vulnerability and adaptive capacity tend to be perceived as negatively correlated as the higher the adaptive capacity the lower the system’s vulnerability and vice-versa (Smit and Pilifosova, 2003; Brooks and Adger, 2004; Füssel and Klein, 2006). Similarly to vulnerability, adaptive capacity is also understood as a dynamic and differential concept as conditions and risks on the ground vary in time and space and across social groups (Blaikie et al., 1994; Kates, 2000; Smit and Wandel, 2006; Adger et al., 2007). Due to its scale-dependent nature, variations in adaptive capacity occur when looking at different scales (e.g., community, region, country) and the interdependency 49
between these scales of action also affect the capacity to adapt in situ (O’Brien et al., 2004; Smit and Wandel, 2006). The importance of adaptive capacity in the context of climate change is the fact that adaptation occurs in dynamic contexts characterised by social, economic, political, technological and biophysical conditions which are particular to a system (Burton et al., 2001). It is this complex combination of conditions and resources that define the system’s capacity to adapt and which are generally termed as the drivers, components or determinants of AC (Burton et al., 2001; Smit and Pilifosova, 2001; Brooks and Adger, 2004; Smit and Wandel, 2006). These determinants normally include aspects such as economic wealth, information and skills, technology, institutions, social capital, natural resources, and equity (Burton et al., 2001; Smit et al., 2001; Yohe and Tol, 2002; Brooks and Adger, 2004; Smit and Wandel, 2006). Many VA frameworks lack the analytical detail required to understand the system’s adaptive capacity and as a result adaptive capacity frameworks have been developed in the context of particular studies to overcome that limitation. Examples of such adaptive capacity frameworks include the work by Wall and Marzall (2006), Marlin et al. (2007), and Walker et al. (2007). The three key components of vulnerability – exposure, sensitivity and adaptive capacity – are addressed, although to different extents, in most CCVA frameworks. However, the sensitivity of a system to climate variability and change can be difficult to assess particularly in cases where the impacts of climatic conditions on the system are not known or well understood. To overcome such limitation, some authors have developed analytical frameworks where vulnerability is understood as a function of the exposure to climatic risks and the system’s capacity to deal with that exposure (cf. O’Brien et al., 2004; Ford and Smit, 2004; Smit and Pilifosova, 2003). In addition, the relationship between the components regarding their relative importance and influence on the vulnerability of the coupled system is also generally unknown (Hinkel, 2011). To overcome this limitation, some studies involve stakeholders or use expert knowledge to determine the relative importance of each of the key components in the analysis of vulnerability (see e.g., Preston et al., 2009). Another common limitation in CCVA frameworks is the lack of understanding regarding the relationship and dynamics between key components and their relative importance in generating/reducing vulnerability.
2.6.4. Why is the system vulnerable? The vulnerable system i.e., the CHES is subject to a range of factors, conditions, and perturbations that influence its vulnerability and/or capacity to adapt. 50
Other important aspects are the causal structures of vulnerability i.e., the causal chain of structures and factors that lead certain groups to be more vulnerable to climate change than others. Finally, vulnerability is regarded as differential as it varies across different groups in the CHES. These will be described below.
2.6.4.1. Multiple perturbations The vulnerable system i.e., the CHES, is constantly changing due to the numerous pressures and factors influencing it (e.g., political and economic systems, demographic changes) and as it changes, so does its vulnerability to climate change (RASSP, 2001). A perturbation is an influence that may affect (negatively or positively) some attribute of the coupled system and can vary in nature, magnitude, timing, and how the vulnerable system will be exposed and respond to it (Turner et al., 2003; Schröter et al., 2005). The term perturbation is interchangeably used in the literature with terms such as stresses, hazards, and events (Smit et al., 2009). However, some distinguish between these concepts e.g., hazards which can be either perturbations (discrete hazards) or stressors (continuous hazards) (Füssel, 2007; Turner et al., 2003). The notion of hazard is normally related to an external perturbation i.e., external to the subject of analysis (Füssel, 2006, Cardona, 2004). Studies using a hazard/impact approach (i.e., biophysical perspective) tend to focus on a single perturbation and its multiple effects on the vulnerable system (Kasperson and Kasperson, 2001). Conversely, in CCVA the notion of multiple perturbations (largely based on food security studies) is utilised where multiple causes can lead to various critical outcomes (Polsky et al., 2003; Schröter et al., 2005). In broad terms, one can distinguish between internal and external perturbations (Gallopin, 2006) where the former encompass those occurring within and between the human and environmental components of the system (e.g., unsuitable land management practices in a farming community) (Füssel, 2007); and the latter include the exogenous factors and conditions exerting pressure and influencing the coupled system (e.g. national policy on local land management practices). It is therefore fundamental to understand the structures, forces, and dynamics affecting the coupled system (exogenous perturbations) as well as the interactions within and between its elements (endogenous perturbations) (Gallopin et al., 2001). In addition, other threats can be posed by factors other than climate change (e.g., economic crisis; unemployment) or exacerbated by the confluence of multiple perturbations subjecting the coupled system to different exposures and vulnerabilities (O’Brien et al., 2004). To identify those perturbations relevant to the system under analysis Kasperson and 51
Kasperson (2001) suggest using past experiences of vulnerable situations and working backwards towards identifying the perturbations that led to those instances. Another approach focuses on particular unwanted outcomes and works backwards through the causal structures in order to identify the stresses most likely to generate those unwanted outcomes (RASSP, 2001). However, many vulnerability studies focus on external perturbations and, as a result, there is a lack of understanding regarding the role of internal perturbations and interactions within the coupled system (RASSP, 2001; Gallopin, 2006).
2.6.4.2. Causal structures of vulnerability The notion of causality in vulnerability studies is largely underpinned by political economy and political ecology research traditions which emphasise the role of socioeconomic and political factors in generating people’s vulnerabilities. Some authors have been fundamental in terms of understanding causality in vulnerability analysis (see e.g., Sen, 1981; Watts and Bohle, 1993). In a disaster risk context, the Pressureand-Release (PAR) model by Blaikie et al. (1994) (updated later by Wisner et al., 2004) has also contributed greatly to the understanding of causal structures of vulnerability by attempting to explain the processes and mechanisms leading to people’s vulnerability. In this model vulnerability is generated through three interconnected levels: from the root causes of vulnerability, to dynamic pressures, finally leading to unsafe conditions (Figure 2.4).
Figure 2.4 - Pressure and Release model (Adapted from Blaikie et al., 1994).
In their conceptualisation, risk is formulated by two concomitant factors: the occurrence of hazards and vulnerable people. The onus of the analysis is therefore on the underlying causes of vulnerability and the capacity of social groups to anticipate, cope, and recover from the occurrence of natural hazards (Blaikie et al., 1994). Both the PAR and CCVA models offer ways of analysing vulnerability despite using different conceptualisations and formulations, focusing on different hazards and 52
looking at different subjects of analysis. Nonetheless, the underlying notion of multiple and intersecting processes and structures generating vulnerability from the PAR model has permeated into the CCVA model. By analysing the non-climatic factors and drivers influencing the system, the CCVA model allows the examination of the multiple causes leading to vulnerability (Ribot, 1995). However, despite efforts to address the causal structures of vulnerability in climate change research, establishing the empirical relations between underlying processes and impacts on the subject of analysis can be challenging (Eakin and Luers, 2006; Young et al., 2006).
2.6.4.3. Differential vulnerability Vulnerability is perceived as a differential concept as it is unequally distributed across the coupled system (Downing and Patwardhan, 2004; Smit and Wandel, 2006; Adger et al., 2007). The pursuit of better understanding of the nuances of vulnerability was promoted by feminist studies where vulnerability (often linked to poverty issues) is exacerbated by cultural and social norms which place women in more vulnerable positions (e.g., dependency for subsistence, reduced educational opportunities) (Cannon, 2002; Fordham, 2004). Other dimensions commonly associated with conditions of vulnerability are poverty (Chambers, 2006), class (Susman et al., 1983), and also age, religion and ethnicity (Blaikie et al., 1994; Morrow, 1999). Factors such as health and access to education and services (e.g., transport, housing) have also been identified as contributing towards people’s adaptive capacity therefore potentially reducing their vulnerability (SNIFFER, 2009). As a result, accounting for differential adaptive capacity in CCVA is important as it varies amongst groups and can be limited by particular barriers (e.g., political, institutional) or even access to resources (Polsky et al., 2003; Adger et al., 2007). These manifestations of vulnerability are not however mutually exclusive as they can coexist and therefore express different nuances of vulnerability (Morrow, 1999). For example, Pradhan et al. (2007) found that in the flooding events of 1993 in Nepal fatality was highest among young children and women and also the poor living close to flooding waterways in poor quality housing.
2.6.5. How to assess vulnerability? When performing a CCVA, certain analytical processes need to be addressed in order to perform the assessment of vulnerability. These include selecting the scale of analysis, performing both historical and prospective analysis, and dealing with uncertainty. These will be explained below.
53
2.6.5.1. Dealing with scale(s) of analysis There is an array of terminology associated with the word scale such as geographic scale, operational scale, and cartographic scale (Fekete et al., 2010). In this context, scale is understood as the “spatial, temporal, quantitative, or analytical dimensions used by scientists to measure and study objects and processes” (Gibson et al., 2000, p. 218). Issues of scale have become central to broader discussions in human and political geography (cf. Marston, 2000) and such contributions can be extrapolated to the climate change context. For example, a current critique relates to the notion of local scale being associated to a particular territorial space and nested in a vertical hierarchical scale that runs from the local to the regional, national, international (Bulkeley, 2005; Marston et al., 2005). Such critique defends the need to recognise scale as politically and socially constructed within a configuration of various networks and actors where the process of involving specific actors and networks is part of the politics of scaling. Such take on the politics of scale focus on the processes through which particular scales and relations come to be constructed and used to govern certain aspects rather than the more conventional use of delimitations of scale as mere discrete units (Bulkeley, 2005). The concept of scale is intrinsically linked to terms such as levels, extent, and resolution (Evans et al., 2002; Gibson et al., 2000). According to Fekete et al. (2010) scale in vulnerability assessment corresponds to the “vertical axis along which any objects of interest are ranked” (Fekete et al., 2010, p. 731) (Figure 2.5). A level is a particular horizontal dimension of that scale whilst a unit is an entity (e.g., administrative boundaries) which are of the same category within a level (Fekete et al., 2010).
Figure 2.5 – Illustration of concepts related to scales of analysis (From Fekete et al., 2010).
The analysis of vulnerability poses numerous challenges regarding scales of analysis. Firstly, the human and environmental components of the coupled system tend 54
to operate at different spatial, temporal and even operational scales (Cash et al., 2006; cf. Marston, 2000). These scale mismatches can occur whenever the scale at which the components operate do not align properly (Cash et al., 2006; Cumming et al., 2006). This is further exacerbated by the fact that different disciplines tend to use different scales to represent and analyse particular phenomena (Cash and Moser, 2000). In social sciences for example, scale can be defined by applying organisational or functional levels (i.e., city, neighbourhood, household) (Stephen and Downing, 2001); whilst in natural sciences it can relate to particular biophysical conditions (e.g., bioregions) (Cash and Moser, 2000). Secondly, the choice of analytical and methodological approaches (e.g., topdown versus bottom-up approaches) can lead to fundamentally different interpretations of the same phenomenon. For example, CCVA based on the biophysical perspective and using a CIA approach tend to focus on larger scales (e.g., national/global) whilst those employing the social perspective using a CCVA approach normally apply smaller scales of analysis (e.g., local/regional) (Füssel and Klein, 2006). Thirdly, it is fundamental that the scale utilised is congruent with the purpose of the analysis regarding both the process of analysis (in terms of its saliency, legitimacy, and credibility) but also regarding the scales of agency particularly in the context of assessments driven towards the formulation of adaptation policy. This requires a certain match between the scale at which the analysis is performed with institutional and managerial powers to act upon the outcomes of such analysis (Cash and Moser, 2000). In addition, focussing the analysis solely at the local level can miss out certain interactions and dynamics across scales and, as the scale of analysis changes so do the variables explaining a particular phenomenon (Evans et al., 2002). Looking at the vulnerability of Norway to climate impacts, O’Brien et al. (2004b) ascertain the different outcomes arising from analysis of vulnerability at the national level compared to the regional and local level. By doing so, the authors unravel the disparate nuances of vulnerability that emerge when using different scales in the analysis of Norway’s vulnerability to climate change. Hence, determining the scales of analysis to be used in the CCVA frames the analysis (and consequently the outcomes) which, in certain cases, can raise issues of equity, justice and differential vulnerability (Schneider et al., 2007). A fourth aspect regarding the scale at which the assessment of vulnerability is to be performed is linked to the importance of understanding the multi-scalar dynamics between the level at which the CHES operates and other levels of interaction in order to understand the dynamics surrounding the coupled system and the plurality of 55
legitimate perspectives (Gallopin et al., 2001). The notion of hierarchy is commonly used to address cross-scalar dynamics and relates to the idea of grouping together objects along an analytical scale (Gibson et al., 2000; Fekete et al., 2010). Hierarchy theory can be used for the scaling of both biophysical and social data as it allows the consideration of the various levels of spatial and temporal resolution from both dimensions of the coupled system (Evans et al., 2002; Wilbanks, 2002). By addressing the levels above and below the level of reference (i.e. the level at which the coupled system operates) this theory allows integration and analysis of data at multiple levels (Evans et al., 2002). Alternative solutions for dealing with scale in CCVA include converging the assessment to a single intermediate scale of analysis (e.g., regional) or adopting a multi-scale or meta-scale synthesis using information at multiple scales and seeking a greater level of understanding from all the information provided (Wilbanks, 2002). Many tools have been developed to help deal with the analysis of scales, such as geographical information systems, statistical downscaling, and nested economic models. However, these are often limited by the absence of integrating frameworks, lack of multiple-scale data, and integrated modelling capacity (Wilbanks, 2002; Eakin and Luers, 2006). The framework by Turner et al. (2003) for example, addresses the cross-scalar dynamics influencing coupled systems by applying a conceptual scale that includes the local, regional, and global levels. However, methodological and operational challenges are expected when assessing the myriad interactions between scales and the numerous linkages between the system’s components (Eakin and Luers, 2006). Hence, the integration of information from disparate scales of analysis as well as the limited understanding of cross-scale dynamics and interactions can pose serious challenges to the analysis of CHES’ vulnerability to climate change (Cash and Moser, 2000). Issues of scale are identified in the literature as a major challenge in CCVA. Not only due to the different conceptual and operational scales at which the human and environmental dimensions of the coupled system tend to operate but also due to the multi-level nature of vulnerability phenomena, the methodological aspects and analytical challenges of performing analysis of vulnerability, as well as the need to ideally match the scale at which the assessment is performed with the scale of agency to act upon the outcomes of the CCVA.
2.6.5.2. Historic and prospective analysis Another important dimension in terms of how to assess vulnerability is related to performing both historical and prospective analysis (Polsky et al., 2003; Schröter et al., 56
2005). In CCVA, the starting point of the analysis is to understand the current vulnerability of the vulnerable system to climate change (i.e., is the system presently affected by climate change? What are the impacts of climate change on the system?). The analysis of the system’s current vulnerability to climatic conditions allows us to understand the main issues and conditions affecting it e.g. current exposure to particular climate conditions, existing adaptation measures already in place (Lim et al., 2004). In certain cases however, the impacts of climate change may still be unknown or not fully understood within a particular system and, as a result, only prospective analysis can be performed (i.e., future vulnerability to climate change). A range of research methods and tools can be utilised to collate information on current vulnerability such as interviews, workshops and focus groups, mapping, and developing vulnerability profiles (Downing and Patwardhan, 2004). Also common in vulnerability studies is the use of indicators/indices (see e.g., Cutter et al., 2000; Yohe and Tol, 2002; Adger et al., 2004; Benkert and Malone, 2005; and Eriksen and Kelly, 2007). Ideally, indicators should be able to summarize or simplify important properties of the coupled system, as well as quantify, measure, and communicate important phenomena relevant to the analysis of vulnerability (Moss et al., 2001; Brenkert and Malone, 2005). However, by trying to encapsulate a very complex reality, indicators can easily fail to accurately represent a particular reality so caution is required when using indicators to assess vulnerability (cf. Vincent, 2007). Understanding the system’s current vulnerability helps to identify and establish the main issues and priorities in terms of analysis (Ford et al., 2006) as well as providing “(...) a roadmap from known territory into uncertain futures” (Dessai and van de Sluijs, 2007, p. 42).
Figure 2.6 - Categories of future characterisations (adapted from Carter et al., 2007b; Zurek and Henrichs, 2007).
57
In CCVA the analysis of future vulnerability is normally pursued through the use of projections and scenarios of the future. A projection is perceived as “(...) any description of the future and the pathway leading to it” (Carter, 2007, p. 3) and can be deterministic or probabilistic (Dessai et al., 2009). Whilst projections are based modelderived estimates, scenarios are a “(...) consistent and plausible picture of a possible future” (Carter et al., 2007, p. 145) which can represent a range of alternative futures outlining how the world might look under different conditions (Lorenzoni et al., 2000; Carter, 2007b). Some of the methods used for characterising future climate change conditions are shown in Figure 2.6. There are two broad categories of scenarios that can be used to characterise the future in climate change research: climate scenarios which provide information on potential future climate change; and socio-economic scenarios which provide possible trajectories of future social and economic conditions and change (Lorenzoni et al., 2000; Carter, 2007b). In this context, socio-economic scenarios are used to characterise not only the multiple drivers of anthropogenic greenhouse gas emissions in climate scenarios (e.g., IPCC Special Report on Emission Scenarios) but also the vulnerability, sensitivity and adaptive capacity of social systems to that change (e.g., United Kingdom Climate Impacts Programme, 2001) (Dahlström and Salmons, 2005; Carter, 2007b). The advantages of linking socio-economic scenarios with climate change projections are recognised as future climate change will be experienced in different social and economic conditions than current ones. Hence, using socioeconomic scenarios (combined with climate scenarios) can account for the system’s particular conditions and characteristics and therefore develop ideas of possible socioeconomic futures (Lorenzoni et al., 2000). Studies combining climate projections with socio-economic scenarios include for example Lorenzoni (2000), Holman and Loveland (2001), and Wade et al. (2006). Scenario-based approaches can be extremely helpful when considering uncertainty in future pathways (Zurek and Henrichs, 2007) as well as a powerful tool to communicate that uncertainty and explore potential futures with stakeholders (Conde and Lonsdale, 2005). However, to achieve such benefits it is essential to ensure that a range of conditions are in place namely engaging with those stakeholders with a stake in the outcomes of the CCVA, and ensuring that the socioeconomic scenarios are adapted and modified in order to suit the needs of the assessment (Berkhout et al., 2002). Hence, the existing difficulties in accounting for socio-economic variants, the methods utilised to downscale socio-economic information at the sub-national level, and the uncertainty surrounding prospective analysis adds to the overall complexity in 58
the analysis of vulnerability particularly at the local level (Berkhout et al., 2002). Moreover, the uncertainty surrounding prospective analysis particularly regarding complex systems such as the CHES adds to the overall complexity in the analysis of vulnerability (see section below).
2.6.5.3. Dealing with uncertainty Climate change uncertainty can be understood as an “expression of the degree to which a value (e.g. the future state of the climate system) is unknown” (Parry et al., 2007, p. 882). Climate change uncertainty encompasses the natural internal variability of the physical system and its sensitivity to future climate change as well as the uncertainty surrounding the social dimensions of the CHES e.g., its capacity to adapt to climate change and the success of a particular adaptation strategy (Mearns, 2010). There is also uncertainty associated with our limited knowledge of the climate system and the ways in which uncertainty is modelled in studies (Hallegatte, 2009; Foley, 2010). Uncertainty related to our understanding of the climate system includes, for example, trajectories of future GHG emissions and the climate’s sensitivity to future changes (Hallegatte, 2009; Foley, 2010); whilst uncertainty in models’ parameters and data includes unknown errors in model structures and data (Schneider and KuntzDuriseti, 2002; Foley, 2010) and uncertainty in modelling the impacts of climate change and their distribution in the coupled system (Dessai et al., 2009). The problem of uncertainty is further exacerbated by spatial and temporal mismatches i.e., climate change caused by global changes but with local impacts (Schneider and KuntzDuriseti, 2002) and by non-linear processes (i.e., unexpected events, surprises) and self-organisation processes within the coupled system (Gallopin et al., 2001). The conventional approach to climate change uncertainty is underpinned by an evidence-based policy making paradigm which defends the position that uncertainty “needs to be characterised, reduced, managed and communicated” (Dessai and van de Sluijs, 2007, p. 24). Such an approach tends to make use of sophisticated and integrated models and techniques to reduce uncertainty and inform policy and strategies (Morgan et al., 2009; Dessai et al., 2009). An alternative approach places less importance in reducing uncertainty in itself and argues for new and more flexible ways of characterizing uncertainty in policysetting contexts (Mearns, 2010). Lempert et al. (2004) for example suggest an ‘assessrisk-of-policy’ framework to support robust decision-making against currently known uncertainties (Mearns, 2010) whilst Hallegatte (2009) defends the need for more flexible decision making frameworks and proposes different types of strategies to achieve this (e.g., no-regret strategies, reversible strategies). In the face of a certain 59
degree of irreducible uncertainty there is an underlying argument for more iterative, strategic and flexible mechanisms for the implementation and monitoring of adaptation measures and strategies against new knowledge to allow the necessary flexibility to adapt to new situations and knowledge (Lim et al., 2004; Mearns, 2010). The framework by Lim et al. (2004) defends this approach and suggests the need to continuously monitor and evaluate the implementation of the adaptation strategy (following the CCVA) in order to assess success against its objectives and if necessary review and correct the strategy accordingly. Another crucial aspect regarding climate change uncertainty is how to convey this uncertainty to stakeholders and the general public. The responsibility of science to provide information on the rate and magnitude of change and potential damaging events presents serious issues regarding the use of language (Manning, 2003). The IPCC advises full consideration of the various sources of uncertainty in climate change research and the careful use of language to convey such findings since different recipients will have different understandings of findings (IPCC, 2005). Although climate change uncertainty has been receiving growing attention by scholars much of it focuses on the uncertainties surrounding the climate system. Fewer efforts have been made regarding uncertainties in the analysis of climate change vulnerability and impacts of actions taken (Dessai et al., 2007). Looking at seven national adaptation strategies, Biesbroek et al. (2010) review the main factors conditioning the development and implementation of adaptation policy and existing knowledge gaps in the formulation of this type of policy. Based on their study the recommendations stressed the need to further develop effective ways of assessing, managing and communicating uncertainty in a policy context.
2.6.6. Current challenges in performing climate change vulnerability assessments The previous sections outlined the main conceptual and analytical elements in the climate change literature regarding the assessment of vulnerability. This section aims to draw together some of the remaining challenges and prevailing limitations in performing integrated CCVA. The majority of these challenges relate to the lack of praxis which perpetuates our limited understanding of a number of critical issues regarding the assessment of climate change vulnerability. This deficiency is further exacerbated when the aim of CCVA is to provide meaningful and useful information to decision-makers towards developing adaptation policy (Patt and Dessai, 2005). Overall, the need to continue developing empirical studies as well as applying and testing CCVA frameworks is associated with two underlying propositions. 60
The first idea relates to the need to continue exploring and developing new ways of integrating disparate conceptualizations of vulnerability together with the multiplicity of methods to assess vulnerability. As the integrated perspective on vulnerability attempts to amalgamate concepts, ideas and methodologies from both the biophysical and the social perspectives the need to explore new bridges between conceptual perspectives is critical. This effort needs to be developed by recognizing the disparate conceptualizations, framings and interpretations of vulnerability in order to avoid misunderstandings across disciplines (O’Brien et al., 2007), and therefore work towards identifying common points of contact between different worldviews and framings of vulnerability (Newell et al., 2005). This requires a commitment to explore and develop a shared language towards a common conceptual framework (Newell et al., 2005). This blending process needs to be continuously applied and tested in order to “(...) demonstrate the value of integrative research, and [...] promote the growth of trust and understanding across disciplinary boundaries.” (Newell et al., 2005, p.306). Moreover, CCVA frameworks need to be applied to real case studies so as to better understand the practical and operational implications of employing integrated approaches on the ground. The second issue relates to those prevailing challenges and limitations regarding the study and application of the various conceptual elements of CCVA. For example, the interactions and dynamics between the components of the coupled system and the ways in which these relationships influence the CHES’ overall vulnerability to climate change are still poorly understood. This includes not only the interactions between the two components of the system – human and environmental – but also how internal perturbations affect and interact within the coupled system. Moreover, processes of causality between the underlying causes of vulnerability and the impacts on the coupled system also need to be further explored as these remain largely speculative in this type of study. Another example of such limitations is the need to continue exploring and developing new ways of analysing and integrating uncertainty in CCVA as there are clear limitations in many of the assessments being performed (Schneider and Kuntz-Duriseti, 2002). It is also fundamental to improve the ways in which that uncertainty is managed and addressed in the development and implementation of policy to reduce vulnerability. Table 2.4 below lists some of the critical challenges and limitations that prevail regarding the conceptual and analytical elements in CCVA.
61
Table 2.4 - Current challenges in performing integrated climate change vulnerability assessments (Bruno Soares et al., 2012).
Conceptual elements/
Challenges and limitations
normative ideals of CCVA
to the analysis
The CHES
Limited understanding of how human and environmental systems are coupled and the ways in which the two components of the coupled system interact and influence each other.
Who is involved?
Engaging stakeholders
Challenges surrounding co-production of knowledge between researchers and stakeholders; conceptual and operational challenges in achieving a broad public/stakeholder engagement that is significant and representative to the analysis.
Vulnerable to what?
Key components of vulnerability
Limited knowledge regarding the relationship and dynamics between key components; lack of clarity in majority of CCVA frameworks on how to analyse these components.
Multiple perturbations
Limited understanding on the influence of internal perturbations in the CHES i.e., interaction between the human-environment dimensions and how these are affected by internal perturbations.
Causal structures of vulnerability
Empirical challenge regarding the difficulty in establishing the relations between underlying causal structures and processes and impacts on the CHES.
Scales of analysis
Methodological and operational challenges including difficulty in articulating different spatial and temporal scales operating in the CHES; limited understanding of cross-scale dynamics and how to address such dynamics in the analysis.
What or who is vulnerable?
Why is it vulnerable?
How to assess vulnerability?
Historical and prospective analysis
The uncertainty surrounding prospective analysis adds to the complexity in the analysis of climate change vulnerability.
Dealing with uncertainty
Lack of practical application and empirical studies exploring new ways of dealing with and managing uncertainty particularly in policy-setting contexts.
These challenges and limitations can only be overcome through more empirical studies to further explore and test the integration and analysis of the various conceptual elements in CCVA. Moreover, without this empirical knowledge it is extremely difficult to produce relevant information, particularly in policy-setting contexts. Only through praxis can we further our understanding and knowledge regarding the vulnerability of coupled systems to climate change. The previous sections described the main normative and conceptual elements regarded as important when conducting a CCVA. These include defining the coupled system as the main system of analysis; addressing the key components of vulnerability; defining the scales of analysis; engaging stakeholders, and dealing with uncertainty. 62
The following section presents some examples of conceptual and analytical frameworks developed in the literature to assess vulnerability and how the various conceptual and analytical elements are addressed in these frameworks.
2.6.7. Examples of vulnerability assessments frameworks Several conceptual and analytical frameworks to assess climate change vulnerability exist in the literature although some have been developed in different theoretical contexts to that of climate change research (e.g., sustainability, global environmental change, environmental hazards). Nonetheless, these frameworks are important contributions to the field, as they have helped to shape and inform our conceptual and analytical understanding of CCVA in climate change. Table 2.5 below shows some examples of frameworks to assess climate change vulnerability against the various conceptual elements reviewed in the sections above.
Dealing with uncertainty
Historical & prospective
Engaging stakeholders
Differential vulnerability
Causal structures
Scales of analysis
Multiple perturbations
Key components
Climate Change Vulnerability Assessment (CCVA) frameworks
The CHES
Table 2.5 - Examples of vulnerability assessment frameworks.
Cutter (1996) ‘Hazard of place model’ Turner et al. (2003) ‘A framework for vulnerability analysis in sustainability science’ Polsky et al. (2003) ‘Assessing vulnerabilities to the effects of global change: an eight-step approach’ Ford and Smit (2004) ‘Assessing the vulnerability of Arctic communities to climate change’ Lim et al. (2004) ‘Adaptation policy frameworks for climate change’ Hjerpe and Wilk (2010) ‘Baltic Climate Vulnerability Assessment Framework’ Addressed in the CCVA
Partially addressed
Depends on approach
Undefined; N/A
63
The CHES and the key components of vulnerability are the analytical and conceptual elements most common to the frameworks listed in Table 2.5. The only exception regarding the analysis of the CHES is the framework by Lim et al. (2004). This framework allows the use of different analytical and conceptual approaches (e.g., risk/hazard based; vulnerability based) and, as a result, the unit of analysis can differ (e.g., CHES, a community, a bioregion) depending on the analysis. All frameworks in Table 2.5 above address the key components of vulnerability except for Cutter (1996) since this framework does not account for adaptive capacity in the analysis of vulnerability. The causal structures of vulnerability are the least regarded element in the various frameworks, being only considered in the framework developed by Turner et al. (2003). Applying this framework to three case studies, Turner et al. (2003b) unravel the complexity in capturing and analysing such processes and dynamics in empirical studies. This complexity in capturing structural processes and linking them to impacts on the coupled system may be the reason why this conceptual element is not addressed in the majority of CCVA frameworks. The purpose and context in which the various CCVA frameworks were developed also influenced the conceptual elements that are addressed by each of them. For example, the frameworks by Polsky et al. (2003) and Lim et al. (2004) aim to provide general guidance on the process of assessing vulnerability and as a result, are more comprehensive in nature and therefore encompass a larger number of the conceptual elements. Conversely, the frameworks by Ford and Smit (2004) and Hjerpe and Wilk (2010) were developed for assessing the vulnerability of specific communities (Arctic and Baltic communities, respectively). As a result, both frameworks address particular conceptual elements which reflect the specific purpose for which each of these frameworks were developed in the first place. Of all the CCVA frameworks listed above, the Adaptation Policy Framework (APF) by the United Nations Development Programme (UNDP) (Lim et al., 2004) is the most comprehensive regarding the various analytical and conceptual elements discussed in the previous sections. Due to its non-prescriptive nature, this framework allows the selection of the most appropriate approach and methods to perform CCVA given the particularities of the study being performed (Downing and Patwardhan, 2004). The CCVA frameworks presented above propose (to different extents) ways of analysing vulnerability based on the integrated perspective on vulnerability (cf. Table 2.5). However, the complexity in addressing the various conceptual elements in integrated assessments is clear as none of these frameworks fully considers all elements deemed desirable to perform the analysis of vulnerability. 64
2.7.
Conclusions Climate change is understood as any significant and long lasting changes in
climate regarding changes in the mean and/or variability of its properties (Solomon et al., 2007). Adaptation to climate change can take multiple forms, occur at different governance levels and policy areas, and be pursued by a range of different actors. Adaptation actions can be categorised according to their spatial scope, timing, purpose, and form. The uptake of adaptation measures to tackle climate change has increased considerably over the past decade (see e.g., Ford et al., 2011). However, in order to pursue and implement climate change adaptation on the ground it is fundamental to understand vulnerability to climate change. This chapter has discussed how, in climate change research, vulnerability theory is underpinned by a plurality of epistemological and research traditions which has led to a multiplicity of perspectives and conceptualisations of what vulnerability to climate change entails. The current paradigm on vulnerability is the integrated perspective which aims to address both the human and environmental components of coupled systems, where vulnerability is defined by both the exposure to climate change, the susceptibility of the coupled system to suffer (or benefit) from such exposure and its capacity to adapt to those changes. This perspective also defends the need to account for and address a range of conceptual and analytical elements when performing CCVA. These include addressing the key components of vulnerability; engaging with stakeholders in the analysis of vulnerability; analysing the causal structures of vulnerability; accounting for differential vulnerability; addressing the scales of analysis; and performing historic and prospective analysis of vulnerability. The majority of CCVA frameworks in climate change research already address and apply (to different extents) the integrated perspective of vulnerability in their analytical frames. However, the literature on climate change identifies some of the existing and prevailing challenges and limitations of addressing and accounting for these elements when conducting CCVA. This study aims to explore how these conceptual and analytical elements of the integrated perspective on vulnerability can be applied in practice to a case study in Scotland. By doing so, it further aims to investigate the challenges and analytical trade-offs of performing CCVA in Scotland at the local level. In doing so it also aims to further explore the challenges and limitations to realising effective CCVA.
65
3 Chapter 3 | Conceptual framework and methodology
66
3.1.
Introduction Chapters 1 and 2 introduced the overall research problematic being examined
in this study and current understandings of how vulnerability can be assessed to inform climate change adaptation efforts. This chapter describes the overall approach adopted in the research project as well as the conceptual framework developed and the methodology utilized to perform the empirical investigation. Section 3.2 elaborates on the epistemological position adopted to conduct the research and describes the mixed methods approach and strategy of enquiry selected to perform the analysis of vulnerability. The climate change vulnerability assessment framework adopted and the methods used to perform the empirical investigation are then explained in sections 3.3 and 3.4. Section 3.5 concludes the chapter.
3.2.
Overall orientation and research approach The interdisciplinary nature of this research project involves the analysis of
phenomena related to both the social and natural dimensions of a complex system and the interactions between them i.e., the human-environment coupled system. In order to reach an understanding of these various dimensions and analytical components interdisciplinary studies require the collection of information from disparate sources as well as a range of methods and tools to perform the analysis. The epistemological approach adopted in this research project is based on the pragmatic tradition (Robson, 2011). The pragmatist school of thought includes, amongst others, the philosophical work of C.S. Peirce, John Dewey, and more recently Richard Rorty (Cherryholmes, 1992; Creswell, 2009). Although differences exist between different branches of pragmatism, knowledge is generally understood as emerging out of actions and consequences rather than prior conditions as in other philosophical traditions whose focus lie on antecedent phenomena and reporting on past experiences (Cherryholmes, 1992). This epistemological approach is therefore “real-world” and is rooted in practice, focussing on applications and solutions to problems (Creswell, 2009; Robson, 2011). Pragmatism allows for a pluralistic and eclectic approach to both quantitative and qualitative research (Johnson and Onwuegbuzie, 2004; Robson, 2011) as the focus is on the research problematic rather than specific methods of research. As a result it allows the use of a range of available approaches to understand the problem in question (Creswell, 2009). There are however, criticisms of this approach including the so called ‘incompatibility thesis’ which views the use of both qualitative and quantitative research 67
as incompatible given the different nature of the phenomena being studied (Robson, 2011). However, although questions over methodology and integration remain (see e.g., Bryman, 2007) defenders of this approach suggest that pragmatism can still offer a solution to conventional dualisms in research methodology (Howe, 1988; Morgan, 2007). For example, Morgan (2007) considers the separation between theory and research (i.e., inductive versus deductive) as very limited since the “(...) actual process of moving between theory and data never operates in only one direction” (Morgan, 2007, p. 72). As an alternative, the pragmatic approach can adopt a more flexible position that moves back and forward between inductive and deductive approaches. The pragmatic researcher also believes in the existence of different worldviews and assumptions of what constitutes reality (Creswell, 2009). Pragmatism endorses fallibilism where research conclusions are rarely perceived as perfect, absolute or certain, and current truth and knowledge are viewed as changeable over time (Robson, 2011). In light of the epistemological approach selected, the research problematic, and the objectives of this study, a mixed methods approach was adopted for this project. Studies using a mixed methods approach can include qualitative and quantitative approaches in either the types of research questions; the type of data utilised (e.g., numerical and textual); the type of data collection (e.g., interviews and survey); and/or the type of data analysis (e.g., thematic and statistical) (Tashakkori and Creswell, 2007; Gray, 2009). By applying this type of approach “(…) the researcher seeks to elaborate on or expand the findings of one method with another method” (Creswell, 2009, p. 14) therefore allowing for complementarities between the various methods of enquiry in a sequential way. It also permits expanding knowledge from one method to another, therefore allowing a more holistic understanding of the research problematic through the triangulation and convergence of different data (Bryman, 1992; Creswell, 2009; Gray, 2009). The strategy of enquiry selected was the case study as it allows the researcher to investigate the “how” and “why” of a particular contemporary event within a real-life context (Gray, 2009; Yin, 2009) and, as a result renders a more thorough and complete account of social processes and issues (Yin, 2009; Robson, 2011). The case study approach is also compatible with a mixed methods strategy as it is normally based on multiple sources of evidence (i.e., a mix of qualitative and quantitative data) (Gray, 2009). Ultimately, adopting a case study approach allows the project to test the use of a vulnerability assessment framework by applying it to perform the analysis of vulnerability in a specific geographical area in Scotland. This will develop 68
understanding of how climate change vulnerability assessments can be performed in Scotland at the local level as well as the existing capacity of local government to act on those outcomes. The case study area is the Ayrshire and Arran region (hereafter referred as ‘Ayrshire’ or ‘Ayrshire region’) in Scotland. The decision to study this particular area was based on existing contacts within local authorities in the area and the relative proximity to the research institution. The contrast and range of social and economic characteristics as well as the physical and natural attributes of the region were also conditioning factors for the decision to study this geographical area. The Ayrshire region is a single case study with embedded units of analysis (Gray, 2009) each unit corresponding to a different policy area. These policy areas were selected based on a preliminary analysis of the case study area and the rationale underpinning their selection is described in chapter 4. The sections below explain in more detail the overall conceptual framework adopted in this study to respond to the three research questions set out in chapter 1.
3.3.
Developing the analytical framework to assess vulnerability This section describes the conceptual framework which underpinned the
climate change vulnerability assessment (CCVA) framework used to conduct the analysis of vulnerability in this research project. The CCVA framework developed to perform the analysis was based on the Adaptation Policy Framework (APF) of the United Nations Development Programme (UNDP) (Lim et al., 2004). Although originally developed to provide guidance to least developing countries on national adaptation policy formulation, the APF, due to its nonprescriptive nature, allows for the selection of different analytical approaches and methods to perform the assessment of vulnerability given the particularities of the study being performed (Downing and Patwardhan, 2004). In addition, when compared to other frameworks the APF is more comprehensive regarding the various conceptual and analytical elements that should be addressed (Bruno Soares et al., 2012). By permitting the use of different conceptual and methodological approaches the APF allows for the degree of flexibility needed in this study for accommodating different conceptual framings and the analysis of different vulnerable phenomena (cf. Preston et al., 2011). It also permits the use and integration of a range of methods of enquiry which is aligned with the mixed methods approach adopted in this study. This flexibility is particularly relevant in the context of this research project as the vulnerable systems under analysis were selected in conjunction with stakeholders in the first stage of the 69
vulnerability assessment and, as a result, required different analytical framings and consequently different methods and data to perform the analysis of vulnerability. The APF framework is constituted of five analytical components and two crosscutting elements (Figure 3.1).
Stakeholders & Adaptive capacity Continuing the adaptation process Formulating an adaptation strategy Assessing future climate risks Assessing current vulnerability Scoping & designing adaptation project
Figure 3.1 - Analytical components and cross-cutting elements of the APF (Lim et al., 2004).
The starting point, ‘scoping and designing the adaptation project’, sets out the objectives, activities, and expected outputs for the project. It is also at this stage that the analytical approach and methods are decided, stakeholders identified, and existing information on vulnerability and adaptation reviewed (Lim and Malone, 2004).
The second component, ‘assessing current vulnerability’, analyses the current climatic risks of the vulnerable system and also the factors and conditions affecting its vulnerability. To achieve this it is necessary to identify and establish “(...) who is vulnerable, to what, in what way and where” (Downing and Patwardhan, 2004, p. 73). A characterisation of the socio-economic conditions and the analysis of existing adaptation are also addressed at this stage (Downing and Patwardhan, 2004). The outcomes from this component provide an understanding of the dynamics between the system of analysis and climate hence helping “(...) to provide a roadmap from known territory into uncertain futures” (Dessai and van de Sluijs, 2007, p. 42).
The third component, ‘assessing future climate risks’, builds up from component 2 and assesses future climate risks based on climate change projections, future 70
trends in socio-economic conditions, and potential adaptation barriers and opportunities (Lim and Malone, 2004).
Component four, ‘formulating an adaptation strategy’, brings together the three previous components to inform the development of an adaptation strategy (Niang-Diop and Borsch, 2004).
The fifth and final component, ‘continuing the adaptation process’, sets out how to implement and support the adaptation strategy developed (Perez and Yohe, 2004).
Engaging stakeholders is regarded as a cross-cutting element in the APF as involving stakeholders from the outset of the project is considered instrumental in the analysis and development of the adaptation strategy (Conde and Lonsdale, 2004). Adaptive capacity is the other cross-cutting element of the APF and its analysis is ongoing and parallel to the other analytical components (Figure 3.1 above). However, although the APF suggests ways of assessing adaptive capacity under each component (Brooks and Adger, 2004) it does not provide an analytical framework per se. As a result, in this study adaptive capacity was assessed as one of the components of vulnerability (see below). Only the three first components of the APF and the two cross-cutting elements were addressed in this study as it focuses on the challenges and barriers in performing a CCVA and not the development and implementation of an adaptation strategy. The section below describes the adapted CCVA framework used to perform the analysis of vulnerability to climate change in the Ayrshire region.
3.4.
The CCVA framework used to perform the analysis of vulnerability The CCVA framework was based on the APF and consisted of three analytical
stages: the scoping stage, assessing current vulnerability, and assessing future vulnerability. These stages of the CCVA framework are similar to the three first components of the APF. The starting point of the analysis – the scoping stage – aimed to characterise the case study area and understand its main problems and issues, specifically those most related and susceptible to climate. To assess current vulnerability the main climatic events currently affecting the region together with an analysis of the existing capacity to adapt were characterised. The final stage – assessing future vulnerability – used the outcomes of stage 2 as a starting point to assess the potential future impacts of climate variability and change in the region. Table 3.1 below illustrates the main 71
stages of the CCVA framework adopted to perform the analysis of vulnerability as well as the overall approach and the methods and tools utilised to conduct the assessment.
Table 3.1 - Main stages, approaches, methods and tools of the CCVA framework.
Stage 1
Stage 2
Stage 3
Scoping stage
Assessing current vulnerability
Assessing future vulnerability
Aim
Select the methodological approach, identify stakeholders, and characterise the vulnerable systems within the case study region
Characterise the main climatic risks affecting the vulnerability of the systems as well as the sensitivity of those systems to such climatic risks; characterise the capacity to adapt to climate change within those systems of analysis
Characterise future vulnerability of the vulnerable systems to climate change in Ayrshire
Main approach
Bottom-up
Bottom-up
Top-down
Methods and tools
Semi-structured interviews; survey; documentary analysis; statistical data; GIS layers
Semi-structured interviews; farmers’ survey; documentary analysis; statistical data; GIS layers
Climate change projections
All three stages of the CCVA are explained in more detailed below as well as the methods utilised and data collated to perform the analysis of vulnerability.
3.4.1. Stage 1: scoping stage The main aim of this stage of analysis was to characterise the Ayrshire region and identify the main issues and problems influencing it in order to be able to prioritise and select those vulnerable systems most relevant within the context of the case study region. To achieve that, a number of tasks had to be performed namely establishing the analytical approach and methods, selecting the key vulnerable systems to be studied in the analysis of vulnerability, and defining the stakeholder process (Ebi et al., 2004). These are further described below.
3.4.1.1. Establishing the stakeholders’ process and collecting data The adoption of bottom-up approaches is common in CCVA as the involvement of stakeholders is considered important where their knowledge and past experiences can provide an important contribution to understanding vulnerability (see chapter 2). Establishing the stakeholder process involved defining the processes through which stakeholders’ contribution to the analysis was to be achieved (Ebi et al., 2004). 72
Two phases of stakeholders’ involvement were required for this study. The initial phase of stakeholder engagement was pursued during this stage whilst the second phase was performed during the assessment of current vulnerability (stage 2 of the CCVA) (see below). The first round of stakeholders’ involvement aimed at understanding the case study area as a whole. To achieve that, it was crucial to include a wide range of viewpoints and knowledge across the case study region in order to better understand the existing problems and issues within the region. As a result, various organisations were contacted using both interviews and survey to collect data. At this stage of analysis, various datasets were also collected including policy and corporate documents, statistical data, and geo-referenced data. These are described below. A total of 16 semi-structured interviews (19 interviewees) were conducted between 2009 and 2010 with officers across different departments in the three local authorities in Ayrshire (planning policy; sustainability-related departments; economic development; community planning; and emergency services) as well as other relevant organisations in the region such as the Ayrshire Joint Planning Unit, Scottish Natural Heritage,
Forestry
Commission,
Scottish
Agricultural
College,
and
Scottish
Environmental Protection Agency (see Appendix 1 for a schedule of interviews). The interview questions included common themes across a range of issues, from wider subjects such as corporate approach to climate change to particular issues related to specific departments/organisations (see Appendix 2 for a matrix of the questions asked in interviews). The interviews were transcribed from MP3 format to Microsoft Office Word using F4 software. The text from the interviews was then coded and analysed using NVivo software19 which was used to explore the main issues and problems within the case study region. During this stage, a postal survey was also sent to all community councils20 across Ayrshire. This survey covered a range of issues regarding the community council, its environment and climate-related issues (see Appendix 3 for a copy of this survey). The survey was sent to all community councils in October 2009 and a second batch was re-sent to those community councils that did not respond in February 2010. The total response rate was 54% (38 out of 70 surveys sent) 21. Statistical data was also collected from the main data providers in Scotland (see Appendix 4). The data retrieved covered a range of indicators including population,
19
NVivo software is a qualitative data analysis software. A community council is a non-statutory organization run by volunteers whose role is to act as a channel for the views of local communities. There are approximately 70 CCs in the Ayrshire region. 21 Excluding responses considered as void (i.e., missing information regarding the Community Council). 20
73
employment, and education, and were utilised as baseline information for the socioeconomic characterisation of the Ayrshire region. A number of policy documents were collected and analysed in order to better understand the governance context in Ayrshire with particular attention to those areas in local government most closely related to climate change such as policy planning, corporate planning, and sustainability. These included the Ayrshire Joint Structure Plan; the Ayrshire Local Biodiversity Action Plan; Ayrshire and Arran Woodland Strategy; Single Outcome Agreements for all three local authorities in Ayrshire; Local policy plans (and new Development Plans where ready); Sustainability Strategies; Scottish Climate Change Declaration reports; other documents and material (e.g., books and tapes on the history of Ayrshire). Geographical Information System (GIS) shapefiles were also collected to provide geo-referenced information (both spatial and alphanumerical) on the case study area. This data was utilised as background information in the analysis of the Ayrshire region and included information on cartographic background data and administrative boundaries (see Appendix 4 for a list of all GIS shapefiles utilised).
3.4.1.2. Selecting the key vulnerable systems Prioritising the key vulnerable system(s) implies establishing “who” or “what” is vulnerable within the area of study i.e. what or who is going to be the focus of the analysis of vulnerability. Key vulnerable systems can be prioritised based on the nature of the vulnerability such as economic vulnerability (e.g., is the system closely linked to the economy?) or physical vulnerability (e.g., is the system physically vulnerable?), and the importance of such systems within the case study area (Ebi et al., 2004). Reviewing existing data, information, and local knowledge on the case study region was therefore essential to understand the region as a whole. The collection and analysis of such information and data led to the production of a ‘Feedback Report’ which aimed to synthesise the preliminary analysis of the case study region. The report was sent to all stakeholders22 interviewed for their feedback on findings (see Appendix 5 for a copy of this report). The report summarised the main characteristics of the region regarding main economic sectors, socio-economic conditions (e.g., population, education, employment, household composition), as well as deprivation issues across the region. The report also presented the main outcomes from the postal survey sent to community councils and past experiences by those communities regarding weatherrelated conditions and events (e.g., cold temperatures, flooding). Based on this analysis, a report was prepared suggesting three vulnerable systems for the analysis of 22
Stakeholders interviewed from all three local authorities as well as officers from the Ayrshire Joint Planning Unit.
74
vulnerability to climate change in Ayrshire: farming, flooding, and tourism. These systems were selected due to their susceptibility and vulnerability to climate change and relevant to the future development of the region. The majority of the stakeholders who responded to the feedback report agreed with the systems proposed in the stakeholders’ report. The rationale for selecting these topics is further explained in chapter 4.
3.4.1.3. Selecting the approach and methods for assessing vulnerability In analytical terms, the assessment of vulnerability was performed based on the definition provided by the Intergovernmental Panel on Climate Change (IPCC) where vulnerability is understood as a function of exposure, sensitivity, and the adaptive capacity of a particular vulnerable system to climate change (Bruno Soares et al., 2012; Downing and Patwardhan, 2004):
Vulnerability = f (Exposure, Sensitivity, Adaptive capacity)
In this view, the vulnerability of a given system will largely depend on its exposure and sensitivity to particular climatic risks (which together represent the potential impacts on the system) and the capacity to adapt to the impacts of that exposure and sensitivity. A composite index method was adopted to perform the analysis of vulnerability and calculate the exposure, sensitivity, and adaptive capacity of each of the systems under analysis. The method used to develop the composite indices in this study was largely based on the work by O’Brien et al. (2004) where a vulnerability assessment of Indian states to climate change and other global stressors at the regional level was performed based on composite indices. For more information on this method see the technical memorandum of their study in Leichenko et al. (2004). Composite indices are useful as they provide relative measures (rather than absolute) and, as a result, allow the comparison between analytical features (Lindley et al., 2011). All composite indices developed in this study were:
Calculated based on the aggregation (average value) of different indicators (e.g., climatic variables, determinants of adaptive capacity;
Indicators were selected based on their relevance to the analysis of farming and flooding and the existence and accessibility of data to perform the analysis;
75
Normalisation was required prior to any data aggregation as the data/indicators used in the analysis had different measurement units (cf. OECD, 2008).
All indicators were normalised to the range of values in the dataset based on the UNDP’s Human Development Index23 (UNDP HDI) by applying the following formula (Klugman, 2011):
Variable value = (actual value – minimum value) * 100 / (maximum value – minimum value)
In certain cases, the indicators’ values were reversed by using [100 – variable value] in order to ensure that higher values always indicate higher vulnerability (Leichenko et al., 2004);
Equal weights were attributed to all indicators (used to calculate the determinants of adaptive capacity and indices of exposure and adaptive capacity) therefore giving each the same relative importance in contributing (or not) to the index of vulnerability;
Assigning different weights to indicators was also considered as different indicators could be regarded as contributing more or less to vulnerability. Three main methods can be used to allocate weights to indicators: expert judgment (e.g., Brooks et al., 2005); statistical methods such as principal component analysis (Cutter et al., 2003) and equal weights (O’Brien et al., 2004; Leichenko et al., 2004). However, in this study it was decided to attribute equal weight scores to all indicators therefore giving each the same relative importance in contributing to vulnerability. Although this method does not necessarily reflect in an accurate way the relative importance of each of the determinants, it was difficult to apply other methods such as expert judgement and statistical methods (cf. Lindley et al., 2011) as the number of variables considered under each index did not justify using statistical methods such as principal component analysis;
The indices were classified as quantiles (i.e., 5 classes) as it allows the creation of classes that have the same number of features
(cf.
Leichenko et al., 2004). In this study, each of the five classes used represented 20% of the number of geographical features utilised to 23
The UNDP HDI is a composite measure of human development. Minimum and maximum values are set in order to normalise indicators to have an identical range between 0 and 1 (see Klugman, 2011; and OECD, 2008 for a broader discussion on various normalisation methods).
76
perform the analysis, allowing comparison across the region and between those maps using the same geographical features (e.g., agricultural parishes, data zones)(cf. Leichenko et al., 2004);
Labels were then attributed in ArcGIS so that each quantile (each of the 5 classes) corresponded to highest, high, medium, low, or lowest exposure/adaptive capacity/or vulnerability. For example, the maps with the index of vulnerability for farming and flooding show five quantiles: Highest vulnerability, high, medium, low, and lowest vulnerability. The quantile labelled “highest vulnerability” correspond to 20% of the features (either agricultural parishes in the case of farming or data zones for flooding) which scored the highest values in terms of vulnerability and, as a result, are the most vulnerable agricultural parishes/data zones in Ayrshire with regard to farming/flooding. Conversely, the quantile labelled “lowest vulnerability” correspond to 20% of the features which scored the lowest values in terms of vulnerability and, as a result, are the least vulnerable agricultural parishes/data zones in Ayrshire with regard to farming/flooding.
All indices represented in maps (i.e., current exposure, adaptive capacity, vulnerability; and future exposure) are represented according to these quantiles (classes);
The normalisation and classification methods selected (i.e., UNDP HDI formula and quantiles) allowed us to perform and display a comparative analysis between the different features under analysis (e.g. in the analysis of farming vulnerability this allowed to compare each of the agricultural parishes in relation to the other parishes in the region).
The methods selected to collect and analyse data to perform the CCVA were chosen based on the overall advantages for the analysis as well as practical considerations such as data availability, cost and length of time required to obtain and analyse data, research skills, and computational requirements (Ebi et al., 2004). As a result, a balanced judgement regarding methods and data was carried out based on the mixed methods’ approach, the information required to perform the analysis of vulnerability, and the analytical trade-offs imposed by existing constraints (e.g., inaccessibility or inexistence of data; time and financial requirements). The different methods utilised throughout the CCVA will be explained in the sections below.
77
3.4.2. Stage 2: Assessing current vulnerability of farming and flooding The main aim of this stage of analysis was to provide an overall understanding of the current vulnerability of the systems under analysis to current climatic conditions. To achieve that, the analysis of exposure, sensitivity, and adaptive capacity was performed for farming and flooding. The analysis of exposure to current climate conditions aimed at understanding the main climatic variables affecting both farming and flooding in Ayrshire and the existing sensitivity to such climatic variables. The analysis of adaptive capacity focused on existing conditions and factors influencing the capacity to adapt across the region in relation to the vulnerability of farming and flooding. In order to further knowledge of the case study region regarding farming and flooding, six more interviews were conducted (2 interviews on farming and 4 on flooding) with local experts (see Appendix 1 for a schedule of interviews and appendices 6 and 7 for a list of the questions used in the interviews on farming and flooding, respectively). The information obtained in these interviews was utilized to inform the analysis of vulnerability. Overall current vulnerability was examined based on the analysis of composite indices for exposure, and adaptive capacity for both farming and flooding. These are further described and explained in the sections below.
3.4.2.1. Calculating the indices of exposure to climatic risks The analysis of exposure to climatic risks aimed at understanding overall change over the past years and how these differed across different areas within the region. This was based on analysis of time series’ data on observed climatic variables from the UK Met Office national dataset24. This data were retrieved in grid format of 5km cells covering a total of 137 grid points across Ayrshire (Map 3.1).
24
UK Met Office datasets can be viewed and retrieved at: http://www.metoffice.gov.uk/climatechange/science/monitoring/ukcp09/download/access_gd/index.html
78
Map 3.1 - The Met Office 5km grid cells covering the Ayrshire region (The Met Office, Undated b).
The climatic variables most relevant to the analysis of farming and flooding were analysed in terms of trend analysis for each of the grid points across Ayrshire to assess the degree of change over the years. The trend analysis was performed using the nonparametric Mann-Kendall test. The selection of this method was based on its ability to analyze datasets that do not comply with the assumption of normality (cf. Afzal et al. 2011). To achieve that, the ‘zyp’ package of the R Project for Statistical Computing 25 was utilized to carry out the analysis. The climatic variables analyzed to develop the indices of exposure to climatic risks differed for farming and flooding as these systems are influenced and impacted by different climatic risks. These variables will be further explained in chapters 5 and 6.
3.4.2.2. Calculating the indices of adaptive capacity As described in section 3.3 above, the APF framework does not provide a specific framework to assess adaptive capacity. As a result, in this study the analysis of adaptive capacity was performed based on the assessment of the main determinants
25
For more on this software please see http://www.r-project.org/.
79
that influence the capacity to adapt. In general terms, the determinants of adaptive capacity tend to be associated with the existence of, and accessibility to, certain conditions and resources within groups/individuals such as economic wealth, information and skills, technology, infrastructure, governance, social capital, and natural resources (Brooks and Adger, 2004; Adger et al., 2007). Analytical frameworks to assess adaptive capacity exist in the climate change literature (see e.g., Wall and Marzall, 2006; Marlin et al., 2007; Walker, 2007). A common feature amongst these frameworks is the analysis of existing and available resources (i.e. the determinants of adaptive capacity) according to normative categories (e.g. social resources, natural resources, economic resources, etc.) depending on the subject matter. These frameworks also tend to use indicators to perform the analysis of adaptive capacity although in some cases (see see e.g. Marlin et al., 2007) the analysis is largely based on qualitative data. For this study however, it was decided not to adopt any of the existing frameworks since the adaptive capacity and its determinants will differ for the analysis of farming and flooding and so will the determinants of adaptive capacity. Instead, the analysis was based on current theoretical and practical understandings regarding the determinants of adaptive capacity for each of the analytical themes under scrutiny. Information obtained from the interviews conducted with experts also helped to establish the main determinants of adaptive capacity for both farming and flooding. To assess the adaptive capacity of farming communities a range of methods were utilised to collate data. Statistical data from the Agricultural Census in Scotland was requested from the Scottish Government which was provided in aggregated form per agricultural parish. Other reports and literature on Scottish and UK agriculture were also examined including: economic reports on Scottish agriculture; reports on Farms Business Incomes in Scotland; Scotland’s soils; Less Favoured Areas; and Land Capability for Scotland. GIS shapefiles were also obtained including Agricultural Parish Boundaries, Land Cover Scotland 1988, National Soils Map, and Land Capability Map, all provided by the James Hutton Institute (former Macaulay Institute) (for a complete list of the data utilised see Appendix 4). A survey was also prepared and sent to farmers in Ayrshire. The survey included general questions regarding farming practices and weather/climate-related events in the region (see Appendix 8 for a copy of this survey). To assess the adaptive capacity of local communities to flooding statistical data was retrieved from the Scottish Neighbourhood Statistics. The adaptive capacity index was developed using existing data for each of the 480 data zones26 in Ayrshire (The Scottish Government, Undated). This was the scale of analysis selected to perform the 26
Data zones are the smallest areas of statistical geography in Scotland and each zone contains between 500 and 1,000 household of similar social characteristics.
80
assessment because it allows a more detailed analysis of adaptive capacity particularly regarding urban areas which is where the majority of past flooding events were reported in Ayrshire. GIS shapefiles were also collected for the analysis of adaptive capacity to flooding including Scottish Flood Defences Asset Database, SEPA’s Indicative Flood Risk Map for Ayrshire, and data zone boundaries. All GIS shapefiles were requested from Scottish Government and national organisations (see Appendix 4).
3.4.3. Stage 3: Assessing future vulnerability of farming and flooding The analysis of future vulnerability to climate change is generally based on assumptions about how future biophysical and socio-economic conditions will evolve in time (Carter et al., 2007). There is a range of methods to characterize future climate (see chapter 2) although the use of projections and scenarios is commonly utilized in the analysis of future climate change vulnerability. When assessing future vulnerability, it is possible to utilise both climate change projections and socio-economic scenarios in order to allow for a general characterisation on how social and economic conditions may evolve in the future. Such scenarios are normally developed based on existing projections (e.g., demographic, economic) and/or based on stakeholders’ input on how they perceive socio-economic conditions may evolve within a particular area (see chapter 2). However, in this study socio-economic scenarios were not included in the analysis of future vulnerability. This was due to two main factors: first of all, existing projections for Ayrshire were only available regarding employment and demographic conditions at the local authority level which would not provide any analytical detail at lower levels (e.g., data zones); secondly, the alternative approach to develop socio-economic scenarios would have been to involve local stakeholders in developing alternative socio-economic scenarios for the Ayrshire region. However, due to time and financial limitations to this project that option was not considered feasible. As a result, the analysis of future vulnerability was based on the analysis of future climate projections focusing on those climate risks that, over time, may potentially influence the vulnerability of the farming sector and the risk of flooding in Ayrshire.
3.4.3.1. The UKCP09 climate change projections For this study, projections of future climate change were utilised in order to understand the potential changes in climatic conditions and its impacts and effects in the two vulnerable systems under analysis – farming and flooding. The climate change 81
projections used were the UKCP09 whose baseline climate data is taken for 19611990. Both the climate change projections and baseline climate are simulations performed by the UK Hadley Centre HadCM3 global climate model (Murphy et al., 2010). The UKCP09 climate projections are based on a large ensemble of model projections and hence are probabilistic projections which provide the relative probability of different climate change outcomes (Murphy et al., 2010). Through a complex modelling approach, uncertainties were included in these probabilistic projections related to the natural variability of the climate system; the incomplete understanding of climate processes, as well as the limited and imperfect representation of such processes in climate models, the uncertainty attached to future GHG emissions, and model responses to these (see Murphy et al., 2010 for more on how these uncertainties are dealt with in the projections). The UKCP09 provides probabilistic climate change projections for a range of climatic variables, time periods, and emission scenarios. Climatic variables include temperature and precipitation over 30-year periods (e.g., the 2020s period covering the years between 2010 and 2039) over which changes in the climatic variables are averaged. These projections are presented with a probability level attached to them based on the strength of evidence associated to the results (e.g., 90% likelihood that the temperature at a particular location will be 3.5°C warmer than temperatures in the 1961–1990 baseline period) (Murphy et al., 2010). The emission scenarios used in the UKCP09 are those developed by the IPCC’s Special Report on Emissions Scenarios (SRES) where four major scenarios 27 include different pathways of future demographic, social, environmental technological, and economic change (Nakicenovic and Swart, 2000). Each SRES scenario was therefore developed based on a particular storyline (i.e., a consistent narrative of how the future may evolve) (Carter et al., 2007; Jones and Mearns, 2004). The UKCP09 climate change projections can be examined based on three different emissions scenarios which are equivalent to three of the SRES scenarios storylines (Table 3.2). Table 3.2 - Relation between SRES scenarios and UKCP09 emissions scenarios.
SRES scenario storylines
A1
27
Characterized by very rapid economic growth, global population peaking in midcentury, and then declining, and rapid introduction of new, efficient technologies.
UKCP09 scenarios
A1F1
Technology based on intensive fossil fuel consumption
High
A1B
Technology is balanced across sources
Medium
To see the other SRES scenarios see Nakicenovic and Swart (2000).
82
B1
A convergent world, having the population growth of the A1 storyline. Economic structures change rapidly toward a service and information economy, clean and resourceefficient technologies are introduced, with emphases on social and environmental sustainability.
Low
However, the latest studies and literature on climate change points to a continued rise in global GHG emissions and current emissions being nearer the upper end of the IPCC scenarios (i.e., the A1F1 storyline) (Betts et al., 2011; New et al., 2011). Moreover, the current state of the UNFCCC process and efforts to reduce GHG at the global level “(…) make the chances of keeping below 2ºC extremely slim, with 3ºC much more likely, and a real possibility of 4ºC” (New et al., 2011; p. 15). As a result, in this study the UKCP09 scenarios used for the analysis of future vulnerability were the medium and high emissions scenarios as these were thought to better represent current trends on GHG emissions and the latest debates in the climate change literature.
3.4.3.2. Calculating the index of future exposure to climatic risks As described above, the vulnerability approach adopted in this study is defined as a function of exposure, sensitivity, and adaptive capacity of the vulnerable system to climate change. Based on the outcomes of the previous stage of analysis – assessing current vulnerability – it was possible to understand the main climatic conditions affecting farming and flooding and use that information to inform the analysis of future exposure to climate change. Similarly to the analysis of current vulnerability, a composite index method was also adopted to perform the analysis of future exposure to climatic risks in Ayrshire (see section 3.4.1.3. above). To develop the indices of future exposure to climatic risks, projections were retrieved from the UKCP09 user interface28. The climatic projections utilised were the UK probabilistic projections of climate change over land, considering only future climate change29, over 25km grid squares (see Map 3.2).
28
The UKCP09 user interface can be accessed at http://ukclimateprojectionsui.defra.gov.uk/ui/start/start.php 29 The UKCP09 climate change projections provide two types of projections: Future climate change which describes the difference between the absolute future climate and the (modelled) 1961-1990 baseline climatology; and future absolute climate values where projections are provided as absolute values.
83
Map 3.2 - The UKCP09 25km grid square covering the Ayrshire region (DEFRA, Undated).
Data for each of the climatic variables was retrieved as raw data and selected at the 90% probability level. This level of probability was selected as it refers to the top of the range of the magnitude of projected change compared to the baseline period30. The time periods considered for the analysis of future vulnerability were the 2020s (2010 to 2039) and the 2050s (2040 to 2069). The decision to focus on the 2020s (short-term) period of analysis was based on the responses provided by some of the stakeholders interviewed where they were asked what would be the ideal lifespan of the information provided by this type of assessment to help inform local policy31. The responses given emphasised the importance of obtaining information regarding the next 10-20 years rather than long-term periods of analysis.
30
“For example, if a projected temperature change of +4.5°C is associated with the 90% at a particular location in the 2080s for the UKCP09 medium emission scenario, this should be interpreted as it is projected that there is a 90% likelihood that temperatures at that location will be equal to or less than 4.5°C warmer than temperatures in the 1961–1990 baseline period” (DEFRA, Undated). 31 Local policy in terms of planning, sustainable development and economic development departments in the three local authorities interviewed.
84
The analysis of climatic projections for the 2050s was also performed in order to provide a contrasting view of how the future exposure to climatic risks may evolve in the Ayrshire region. The raw data was exported as GIS shapefile where the 13 square grids corresponding to the Ayrshire region were selected (see Map 3.2 above). The values corresponding to the climatic variables per square grid were then analysed in Excel in order to calculate an index of future exposure to climatic risks. As in the analysis of current vulnerability, this was a composite index based on the aggregation of average values of the climatic variables developed to construct the index (see methods in section 3.4.1.3. above). The variables were normalised, classified, and labelled as quantiles and, as a result, allowed comparison between different maps (i.e., indices of future exposure for different time periods) (see methods in section 3.4.1.3. above). In the analysis of flooding, an interpolation method – Kriging method – was utilised to perform the analysis of the index of future exposure. This method was thought to be useful as it generates an estimated continuous surface from a scattered set of points and, as a result, aids the visualisation of the indices of future exposure (rather than showing those values per grid square as in Map 3.2 above) (Oliver and Webster, 1990). This method was not used in the analysis of farming since it was possible to average the values up to the agricultural parish level (see chapter 5, section 5.3.1.). To perform the interpolation, the values of the indices of future exposure which were calculated for each of the 25km grid cells were transferred to 5km grid cells across the Ayrshire region (see Map 3.1 above). As a result, each of the 5km grid cells was given the value of the 25km grid cell within which they fell. Based on the grid cell points the interpolation method was performed using ArcGIS and the results were classified and displayed as quantiles. Chapters 5 and 6 present the analysis of current and future vulnerability to climate change in the Ayrshire region regarding the two policy areas – farming and flooding, respectively. Applying the CCVA framework to the case study region in the manner described above provided an effective means of responding to the first of the three research questions set out for this study (see chapter 1). These practical challenges and limitations will be examined in more depth in chapter 7.
3.5.
Conclusions This chapter has developed the conceptual framework through which the
empirical investigation will be conducted in chapters 5 and 6. Overall the research project is positioned within a pragmatic approach to research underpinned by the use of mixed methods to perform the analysis. The empirical investigation will be pursued 85
using a case study – the Ayrshire region – to explore how climate change vulnerability assessments can be performed at the local level in Scotland. In order to perform the assessment of vulnerability, two vulnerable systems were identified following a preliminary analysis (scoping stage) of the region: farming and flooding. The vulnerability of these subjects to climate change was investigated through a range of research methods, tools, and available data. However, the aim of this study was not just to understand vulnerability within the region but to develop an analysis of the process that would allow the research questions motivating the study to be addressed. The following chapter describes the first stage of analysis – the scoping stage – by introducing the case study region. The chapter also describes the two vulnerable systems under analysis: farming and flooding in Ayrshire. Chapters 5 and 6 then examine stages 2 and 3 of the CCVA in relation to farming and flooding, respectively.
86
4 Chapter 4 | Scoping stage: the case study region and vulnerable systems
87
4.1.
Introduction The previous chapter described the overall approach, methodology, and
methods that were utilised to conduct the empirical investigation in this research project. This chapter introduces the first stage of the empirical investigation, describing the first stage of the climate change vulnerability assessment (CCVA): the scoping stage. This stage aims to introduce the case study area – the Ayrshire region – and characterize the vulnerable systems under analysis in this study: farming and flooding. Section 4.2 describes the main characteristics of the case study are – the Ayrshire region. Section 4.3 presents an overview of the main characteristics of Western Scotland’s climate and also some climate change projections for this part of Scotland. Section 4.4 then focuses on the two vulnerable systems under analysis in this study and presents the rationale behind their selection. Section 4.5 looks at farming in Scotland and Ayrshire focusing on grassland systems as this is the main type of farming system in the case study area. The section then describes the governing context of agriculture in Scotland and finishes by introducing some of the potential impacts of climate change on farming. Section 4.6 then introduces the problematic of flooding in Scotland and Ayrshire focusing on past flood events in the case study region and the main typologies of flooding in the area. The section then provides an overview of the governance context of flooding in Scotland and finishes by describing the potential impacts of future climate change on flooding. Section 4.7 concludes this chapter.
4.2.
The case study region – the Ayrshire region The case study area is the Ayrshire region located in the Southwest of Scotland
on the shores of the Firth of Clyde and adjoined by the councils of Inverclyde, Renfrewshire, East Renfrewshire, South Lanarkshire, and Dumfries and Galloway. It comprises three local authorities: East Ayrshire, North Ayrshire (including the Isles of Arran32, Great Cumbrae, Little Cumbrae and Holy Island) and South Ayrshire (including the Isle of Ailsa Craig33) (Ayrshire Joint Planning Steering Group, 2007) (Map 4.1).
32 33
The Isle of Arran is included in the numbers referring to North Ayrshire except when otherwise stated. The Islands of Little Cumbrae, Holy Island, and Ailsa Craig are barely inhabited.
88
Map 4.1 - The Ayrshire region (Ordnance Survey data © Crown copyright and database right 2011).
Ayrshire has a population of approximately 368,000 people, a total area of 3,400 km2, and 240 km of coastline (Ayrshire Joint Planning Steering Group, 2007) (Table 4.1). Table 4.1 - Population in the Ayrshire local authorities (Mid-2008 Population Estimates Scotland, GRO).
Local authority
Population
Total area (Sq km)
East Ayrshire
119,920
1,262
North Ayrshire
135,920
885
South Ayrshire
111,670
1,222
In Ayrshire the main towns are Irvine, Ayr, and Kilmarnock in the North, South, and East Ayrshire local authoritiy areas respectively. These towns contain the majority of the population, economic activities and services as well as key accessibility links within the region and to the rest of Scotland (Ayrshire Joint Planning Steering Group,
89
2007). Other important towns include Largs, Ardrossan, Saltcoats, Stevenston, Kilwinning, Cumnock, Troon and Prestwick (Map 4.1). Ayrshire is predominantly a lowland region although it has areas of higher elevation such as the peaks in Arran and the Loch Doon mountains. The coastline is quite varied with raised beaches characterising most of the coast south of Ayr and the north part of Arran whilst in the north, between Ayr and Stevenston, is lowland coast with areas of sandy beaches (Land Use Consultants, 1998). Ayrshire has six main rivers - Garnock, Irvine, Ayr, Doon, Girvan, and Stinchar (Map 4.2).
Map 4.2 - Main towns, rivers, and landscape areas in Ayrshire (Ayrshire Joint Planning Unit; Ordnance Survey data © Crown copyright and database right 2011).
The Ayrshire region includes a range of areas of important biodiversity and natural value with numerous areas designated as biological and geological Sites of Special Scientific Interest (SSSI) as well as Special Protection Areas and Special Areas of Conservation, the last two are European designations aimed at birds and
90
habitats protection, respectively. The isle of Arran itself is almost all designated as SSSI due to its uplands and its rich bird collection. Throughout the 18th and 19th centuries, Ayrshire was marked by the development and expansion of its agricultural and industrial base (mainly textiles, mining and iron industries) which led to the emergence of new towns and villages across the region (Land Use Consultants, 1998). The 20th century was marked by the decline of heavy industry leaving behind a landscape of increasingly incongruous settlements once related to the coal and iron industry. Nowadays, many of the villages across Ayrshire (particularly in East Ayrshire) are economically weak due to closures in the mining industry over the past decades. Farming also became more specialised throughout the 20th century particularly the production of dairy, sheep farming, and potatoes which also resulted in more land being given over to grazing (Land Use Consultants, 1998). As in most of the UK, increasing competitiveness due to globalization along with changes in technology has led to profound changes in local economies and their employment base. Ayrshire was no exception to the generalised shift from a manufacturing and industrial employment base to the service sector, with the decreasing number of jobs in these sectors being accompanied by an increase in the tertiary sector, which now accounts for approximately 80% of all jobs in Ayrshire (Ayrshire Joint Planning Steering Group, 2007). In the services’ sector, the highest percentages of employment in 2008 were in public administration, education and health (approximately 32%), distribution, hotels and restaurants (approximately 25%), and banking, finance, and insurance (approximately 10%) (Nomis, 2008).
4.3.
The climate of the West of Scotland As in the rest of the UK, the Scottish climate is considered as a cold and wet
temperate climate where rainfall is more or less uniform throughout the year and with no significant extremes of temperature during the year (The Met Office, 2007). The West of Scotland tends to be milder than eastern areas of Scotland due to the strong maritime influence with prevailing winds blowing from the sea as well as the warm Gulf Stream which also influences the climate. The West of Scotland is one of the most exposed areas in terms of winds which are associated with the passage of deep depressions formed in the Atlantic close to or across the UK (The Met Office, Undated). Some of the main characteristics Western Scotland climate are summarised below (The Met Office, Undated):
91
Temperature – Annual mean temperature between 1971 and 2000 ranged between 7 and 10ºC along the Ayrshire coastline with lower temperatures inland and at higher elevations. Annual mean daily maximum temperature ranged between 11 and 13 ºC along the coast and between 8.5 and 12 ºC inland and at higher grounds. Annual mean daily minimum temperatures varied between 4 to 8 ºC along the coastline and between 0.5 to 5 ºC inland and at higher altitudes. February tended to be the coldest month whilst July and August tend to be the warmest.
Rainfall – Average annual rainfall between 1971 and 2000 was less than 1300mm along the Ayrshire coast whilst inland and at higher elevations average increased to values between 1300 and 17000mm. There were approximately 45 days with rainfall totals above 1mm (‘wet days’) during autumn and winter months, rising to approximately 55 days at higher altitudes; whilst during the spring and summer months this number decreased to approximately 35 wet days along the coastline and 40 days at higher latitudes.
Frost – Average number of days of air frost34 between 1971 and 200 was approximately 40 days whilst on higher ground this average increases up to 80 days a year. In terms of ground frost days these range between 60 to 140 days per year depending on location (similarly to air frost distribution).
Snowfall – Average number of days of snowfall between 1971 and 200 was around 20 days along the Ayrshire coast and between 45 to 55 days at higher altitudes. The number of days of snow lying on the ground35 was more significant during winter months with an average of 7 days at lower altitudes and 15 days on higher ground. Snowfall normally occurs between November and April (with brief falls in October and May in the uplands).
34
Air frost corresponds to the total number of days when the air minimum temperature is below 0°C (The Met Office, Undated). 35 Days of lying snow are counted if the ground is more than 50% covered by snow at 09h00 (The Met Office, Undated).
92
These are some of the main characteristics of climate in the West of Scotland and Ayrshire. The section below will look into future trends with regard to climate change projections for the West of Scotland.
4.3.1. Climate change projections for Western Scotland According to the UKCP09 climate projections, the West of Scotland is expected to change in terms of its climate, including (Table 4.2): Table 4.2 - Climate projections for the West of Scotland (DEFRA, Undated)
Western Scotland 2020s Medium emission scenario 36
2050s Medium emission scenario
2080s Medium emission scenario
Annual mean temperature
+1.3°C (+0.7°C to + 1.9°C)
+2.2°C (+1.3°C to + 3.3°C)
+3.0°C (+1.9°C to + 4.5°C)
Summer mean temperature
+1.4°C (+0.6°C to + 2.3°C)
+2.4°C (+1.1°C to + 3.8°C)
+3.5°C (+1.8°C to + 5.4°C)
Winter mean temperature
+1.2°C (+0.5°C to + 2.0°C)
+2°C (+1.0°C to + 3.0°C)
+2.6°C (+1.4°C to + 4.0°C)
Annual mean precipitation
0% (-5% to +5%)
-1% (-7% to +5%)
-1% (-9% to +7%)
Summer mean precipitation
-6% (-17% to +7%)
-13% (-27% to +1%)
-16% (-33% to +1%)
Winter mean precipitation
+7% (-1% to +16%)
+15% (+5% to +29%)
+21% (+6% to +42%)
Overall, the climate projections for the West of Scotland point towards a general increase in temperature across the three time-periods presented in the table above and this increase is expected to be more accentuated during summer than winter months. Climate projections for precipitation indicate a general overall decrease with a potentially significant decrease in summer precipitation whilst winter precipitation is projected to increase in the coming decades.
4.4.
Farming and flooding in Ayrshire As described in chapter 3 a preliminary analysis of the case study region was
performed covering the main socio-economic characteristics of Ayrshire. Based on this analysis, a report was prepared and sent to the stakeholders involved suggesting three analytical themes for the analysis of vulnerability to climate change in Ayrshire: 36
Values in bold represent the mid-point of the probability range whilst the values in brackets represent the range from the lowest to highest value (10 and 90%, respectively) for each 30-year time period.
93
farming, flooding, and tourism. These themes were regarded as those most susceptible and vulnerable to climate change and relevant to the future development of the region. The majority of the stakeholders who responded to the feedback report (see Appendix 5) agreed with the analytical themes proposed in the stakeholders’ report. The rationale for selecting these themes is explained below. Farming was considered as the backbone of Ayrshire and although this sector only provides a small number of jobs, the land used for farming-related activities covers almost 70% of the total area of the region. In addition, many of the rural (and most remote) communities in Ayrshire still depend on farming and agriculture for their survival. As this sector is highly dependent on weather conditions any changes in the climate may severely affect the region not only economically (e.g., impacts on production) but also environmentally (e.g., potential impacts on biodiversity) and socially (e.g., impacts on communities relying on farming). Flooding was also identified as a recurrent issue in Ayrshire with various incidents occurring in all three Ayrshire local authority areas (although slightly more prominent in North Ayrshire). Although many of these flooding incidents have been linked to maintenance issues (e.g., blocked culverts and drainage) others however, have been caused by the lack of capacity of the river system to cope with the amount of rainfall in a short period of time. Linkages between the river system and the coast were also highlighted in Ayrshire where flooding can occur in certain areas when the “right” weather conditions occur i.e., low lying rivers and the occurrence of storm surge and consequently tidal flooding which can then force the water coming from the rivers inland. Coastal flooding and storm surge have also been highlighted in certain areas of Ayrshire with impacts on infrastructure (e.g., roads, railway, and promenades). However, although sea level is expected to increase across Scotland over the coming decades, these changes are less significant than in other parts of the UK (UK Committee on Climate Change, 2011). Tourism was identified as an emergent economic sector in Ayrshire and its potential for creating employment and investment opportunities is recognised in various development strategies across the region (Ayrshire Joint Planning Steering Group, 2007; NAC, 2010; SAC, 2007; EAC, 2009). The majority of the tourism infrastructure is located in North and South Ayrshire, much being located along the coastline and related to golf, beaches and water activities. The islands of Arran and the Cumbraes are also important tourist destinations within the region attracting a considerable number of tourists every year (Interview with G. Thompson, 2010). Many of the tourism activities in Ayrshire are closely related to, and dependent on, weather and climate
94
(e.g., beaches and golf courses) and as a result, climate change may impact (positively and/or negatively) the development of this sector in the region. All three analytical themes were thought to either represent currently vulnerable groups in Ayrshire (communities and/or infrastructure affected by flooding) or sectors socio-economically sensitive to changes in climatic conditions (farming practices and tourism activities) (cf. UK Committee on Climate Change, 2011). Although data exists on a range of tourism-related indicators for each of the local authorities the data available only covered broad indicators (e.g., tourist numbers, expenditure by tourist activity). In addition, available datasets only covered the year 2008/09 making it difficult to understand and establish the main issues and climatic conditions affecting tourism beyond a snapshot of the situation, and making comparisons with other years was impossible. Hence, the analysis of this sector’s vulnerability to climate change would have required a serious investment in qualitative research in order to understand the linkages between tourism and climate, which due to time and financial constraints was not possible. However, the overall lack of available data regarding tourism led to the rejection of this theme in the analysis of vulnerability. Therefore the analysis of climate change vulnerability was only performed for farming and flooding. The following sections will therefore introduce the two analytical themes selected to perform the CCVA in Ayrshire: farming and flooding. These themes will be examined in relation to Scotland and Ayrshire based on the information gathered (i.e., interviews, survey, documentary analysis, statistical data, GIS shapefiles) to perform the analysis of vulnerability. The assessment of current and future vulnerability to farming and flooding in Ayrshire will then be examined in chapters 5 and 6 respectively.
4.5.
Agriculture in Scotland and Ayrshire Agriculture can be defined as “(...) as the science or business of cultivating the
soil, growing crops and rearing livestock” whilst farming is understood as the skill of practising agriculture37 (UK Agriculture, Undated). In Scotland, the agricultural sector has been declining over recent decades and it currently accounts for a small proportion of economic activity and employment (0.9% of Gross Value Added and 2.65% of employment in 2009, respectively) (Scottish Government, 2011b; DEFRA, 2010). Nonetheless, agriculture remains an important sector not only because it represents approximately 80% of the total land area in Scotland (including common grazing) but also because a significant part of overall rural employment depends on it (The Scottish Government, 2012). In Scotland, where a considerable percentage of the population 37
The terms agriculture and farming will be used interchangeably throughout this study.
95
lives in remote and rural areas, primary sectors such as agriculture and forestry are important, accounting for 17% of employment in 2010 (The Scottish Government, 2010). Agriculture is also associated with other sectors such as distribution and retailing as well as rural tourism which contribute to the wider Scottish economy. In addition, there has also been a tendency towards farm diversification where alternative enterprises are created using farm resources to provide additional income (e.g., tourism, recreation) (The Scottish Government, 2010d; 2011b). The farming sector in Ayrshire has been changing in similar ways to the rest of Scotland. Changes over the past 15-20 years have led to the consolidation of smaller farms into bigger ones in order to improve their efficiency and cope with economic pressures (Interview with C. Cuthbertson38, 2010). In 2010, there were 2,818 holdings across the region with an average of 85 Hectares per holding. Although significantly lower than the national average, a closer inspection shows a mixture of farm sizes in Ayrshire with smaller holdings (farms up to 5 hectares), lower ground with dairy and beef farms (between 100 and 250 hectares), and hill land (which can go up to 700 hectares) (Interview with C. Cuthbertson, 2010; The Scottish Government, 2011c).
Number of holdings (Hundreds)
Figure 4.1 illustrates this variety of farm sizes across Ayrshire. 6 5 4 3 2 1 0 0-
