Linking Climate Change Adaptation and Mitigation

8 downloads 1354 Views 15MB Size Report
CDM, JI and NAMA are mitigation, and Adaptation Fund and NAPA are ...... building up wealth or the ability to develop and use new technologies by, for instance, ...... were college and final-year high-school students, and had been selected by ...
Linking Climate Change Adaptation and Mitigation - in Agriculture and Forestry

PhD Dissertation Rico Kongsager April 2015

Author

Rico Kongsager

Title

Linking climate change mitigation and adaptation in agriculture and forestry

Supervisors

Dr. Anne Olhoff (Principal supervisor) Head of Programme: Climate Resilient Development UNEP DTU Partnership Prof. Ole Mertz (Co-supervisor) Department of Geography University of Copenhagen, Denmark

Financed by

UNEP DTU Partnership (Ph.D. Grant) Fieldwork financed by UNEP DTU Partnership, Augustinus Foundation, Oticon Fonden Travelgrant, Otto Mønsted’s Foundation, Torben and Alice Frimodt’s Foundation, and OHF and AJ-E Heillmann’s Foundation Support for conference participation by Center for International Forestry Research (CIFOR) and the International Rubber Research and Development Board (IRRDB)

Assessment Committee

Committee Chairman Dr. Ivan Nygaard Senior researcher UNEP DTU Partnership Dr. Daniel Murdiyarso Senior scientist Center for International Forestry Research Dr. Christian Pilegaard Hansen Associate professor Department of Food and Resource Economics, Faculty of Science University of Copenhagen

Front page photo: Maya boy curiously observing the adults at a community meeting (Trio Village, Toledo District, Belize, January 2014).

All photos by author. © Rico Kongsager 2015

Acknowledgements First and most importantly, I owe the greatest thanks to my near family, my wife Christina and our girls Sille and Lucca, for sticking with me during this process, and for distracting me with ‘more important’ things when needed. I would especially like thank you for your willingness to join the fieldwork in Belize – it meant the world to me. Moving to an unfamiliar country was a tough decision, and we all had difficult times during that period, but all three of you handled it fantastically, which I am grateful for. Nevertheless, I hope you are willing to go for it again at another time, because I believe we had the time of our lives (so far). Secondly, I would like to thank colleagues at UNEP DTU Partnership and particularly my Ph.D. fellows. Moreover, I thank my supervisor Anne Olhoff and co-supervisor Ole Mertz for their support, valuable comments and suggestions, and for providing the freedom to let me explore. And Ole, I would also like to thank you for stimulated and encouraged me through my bachelor’s and master’s years, which you were also a great part of. I cannot appreciate that enough. Furthermore, I deeply thank my co-authors, specifically Bruno Locatelli, for your patience and confidence in our ‘project’, and Esteve Corbera for your academic input, but mainly for your hospitality and friendship. For me, you were both a great part of this project. Thirdly, I am indebted to the people of Belize for their kindness and hospitality, as I could not have written this dissertation without your help and patience during the fieldwork. You were always welcoming to me and my family. I would in particularly like to thank Joop (and Martha) Hendrikx for helping with network and practical issues, but chiefly for your friendship and good conversation. Our stay in Belize would not have been so unproblematic without your constant support and enthusiasm. Special thanks also go to Osmany Salas (and family) for great conversations and warm hospitality, and to staff at the Ya’axche Conservation Trust, Maya Mountain Cacao and the Environmental Research Institute at University of Belize for engagement and support. You all were an enormous help – probably more than you know. Finally, this work would have been far less complete without financial support from external sources, namely the Augustinus Foundation, Oticon Fonden Travel Grant, Otto Mønsted’s Foundation, Torben and Alice Frimodt’s Foundation, OHF and AJ-E Heillmann’s Foundation, Center for International Forestry Research, and the International Rubber Research and Development Board. I owe the prolonged fieldwork and my participation in some of the conferences to you.

ii

Supporting activities and courses Presentations at international conferences Date

Conference

Paper presented

Location

Jun 2013

International Rubber Research & Development

4

London, UK

Mar 2014

Global Land Project Open Science Meeting

1

Berlin, Germany

May 2014

Resilience 2014!

2

Montpellier, France

Jul 2014

Carbon, Land and Property

1 and 3

Copenhagen, Denmark

Courses Date

Course

ECTS

Organizer

Location

Nov 2011

Scientific Writing in English (generic course)

2.5

University of Copenhagen

Copenhagen, Denmark

Apr 2012

Conference Presentations in English (generic course)

0

University of Copenhagen

Copenhagen, Denmark

Apr 2012

From Vulnerability to Resilience in Disaster Risk Management

6

The United Nations University Institute for Environment and Human Security

Bonn, Germany

Jun 2012

Political Ecology

8

University of Copenhagen, Forest & Landscape

Copenhagen, Denmark

Jun 2012

Practices, Power, and Knowledge in Participatory Forest Management.

3

Federal Rural University of Rio de Janeiro and Wageningen University

Rio de Janeiro, Brazil

Oct 2012

Forest-Agriculture Frontiers

5

University of Copenhagen, Department of Geography

Copenhagen, Denmark

Mar 2012

Statistical Methods (generic course)

5

University of Copenhagen, Bioscience

Copenhagen, Denmark

May 2013

Multidisciplinary Perspectives on REDD+

10

University of Oslo, Centre for Development and the Environment

Oslo, Norway

May 2014

Philosophy of Science and Research Ethics (generic course)

3

University of Copenhagen, Institute of Food and Resource Economics

Copenhagen, Denmark

Dec 2014

Media school

3

Information (Danish newspaper), Danish Broadcasting Corporation (DR), and Universities Denmark

Copenhagen, Denmark

Workshop and seminar participation Date

Activity

Organizer

Location

May 2012

Adaptation Workshop

Technical University of Denmark

Lyngby, Denmark

Jun 2012

REDD+ Seminar

University of Copenhagen, Forest & Landscape

Copenhagen, Denmark

iii

Jun 2012

REDD+ Workshop

University of Oslo, Centre for Development and the Environment

Oslo, Norway

May 2013

REDD+ Navigating the jungle

Lund University, Centre for Environmental and Climate Research

Lund, Sweden

Dec 2013

Cacao Workshop

Maya Mountain Cacao Ltd.

Punta Gorda, Belize

Oct 2014

REDD+ mini-seminar

University of Copenhagen, Department of Geography

Copenhagen, Denmark

Other presentations of the Ph.D. Date

Audience

Location

Aug 2013

University of Belize – Environmental Research Institute (10 staff members)

Belmopan, Belize

Oct 2013

Caribbean Community Climate Change Centre (~ 40 staff members)

Belmopan, Belize

Feb 2014

Ya´axche Conservation Trust – NGO (12 staff members)

Punta Gorda, Belize

iv

Abstract There is extensive scientific evidence that both adaptation and mitigation are essential to address the problem of climate change. However, there is still limited knowledge about the possibilities for exploiting the interrelationships between these measures in the design and implementation of climate change activities. Against this background, the main objective of this dissertation is to assess whether agriculture and forestry activities in tropical regions can be designed to provide both climate change adaptation and mitigation benefits, and support development locally. Moreover, it is investigated whether this can be done in such a way that the activities complement one another and that any trade-offs can be avoided. The dissertation demonstrates that there is wide scope for adjustments in the design of climate projects. A more holistic approach in the activities can achieve the desired climate effects in terms of both adaptation and mitigation, while at the same time delivering sustainable development benefits locally.

v

Dansk resumé Sammenhænge mellem klimatilpasning og emissionsreduktioner i landbrug og skovbrug Der er omfattende videnskabelig dokumentation for, at både afværgeforanstaltninger og tilpasning er absolut nødvendige for at adressere omfanget og effekterne af klimaændringer. Der er imidlertid stadig begrænset viden omkring mulighederne for at udnytte sammenhænge mellem klimatilpasning og afværgeforanstaltninger i design og implementering af klimaaktiviteter. På denne baggrund er hovedformålet med PhD afhandlingen at vurdere, om skov- og landbrugs aktiviteter i tropiske egne kan konstrueres, så de tilvejebringer både klimatilpasning og afværgeforanstaltninger samt udviklingsfordele lokalt. Desuden undersøges det, om dette kan gøres på sådan en måde, at aktiviteterne supplerer hinanden, og at eventuelle negative indbyrdes påvirkninger kan undgås. Afhandlingen består ud over en synopsis af fire artikler, der belyser ovennævnte problemstillinger empirisk. Den empiriske del indeholder en analyse af 201 skov- og landbrugsprojekter, suppleret med specifikke analyser af skovprojekter og landbrugsjusteringer i Belize og Ghana. Resultaterne illustrerer, at klimaprojekter i dag er meget fokuserede på deres hovedformål (enten klimatilpasning eller afværgeforanstaltninger), og i mindre grad fokuserer på at tilvejebringe andre goder (gensidig opnåelse af tilpasnings- afværgeforanstaltnings- og udviklingsgevinster). Samtidig demonstrerer afhandlingen for de projekter, der undersøges i Belize og Ghana, at projekterne i vid udstrækning med fordel kunne anvende en mere holistisk tilgang især i forhold til den kontekst, de er en del af. Positivt set så viser afhandlingen, at potentialet for at opnå fordele på alle tre områder er til stede, og at mange af aktiviteterne faktisk supplerer hinanden, og i nogle tilfælde ligefrem er synergistiske. Afhandlingen demonstrerer herigennem, at der er et bredere potentiale for gennem justeringer i designet af klimaprojekter, så disse har et mere omfattede idegrundlag, kan opnås ønskede klimaeffekter - både i forhold til tilpasning og afværgeforanstaltninger og også i forhold til at tilvejebringe bæredygtig udvikling lokalt.

vi

Abbreviations CCCCC

Caribbean Community Climate Change Centre

CDM

Clean Development Mechanism

GIS

Geographic Information System

IPCC

Intergovernmental Panel on Climate Change

JI

Joint Implementation

NAPA

National Adaptation Programmes of Action

NGO

Non-Governmental Organization

REDD+

Reducing Emissions from Deforestation and Forest Degradation with sustainable management of forests, conservation of forest carbon stocks and enhancement of forest carbon stocks

UNFCCC

United Nations Framework Convention on Climate Change

WG

Working Group

vii

Table of Contents Acknowledgements ....................................................................................................................................... ii Supporting activities and courses ................................................................................................................. iii Abstract ......................................................................................................................................................... v Dansk resumé ............................................................................................................................................... vi Abbreviations .............................................................................................................................................. vii

PART I: SYNOPSIS .................................................................................................... 1 1

INTRODUCTION ................................................................................................................................... 2 1.1

2

CONCEPTUAL APPROACHES TO LINKING ADAPTATION AND MITIGATION ................. 6 2.1

Background: the adaptation-mitigation dichotomy ....................................................................... 6

2.2

Defining adaptation and mitigation ............................................................................................. 10

2.3

Conceptualizing the linking of adaptation and mitigation ........................................................... 15

2.3.1

Perceptual overlaps and distinctions between adaptation and mitigation................................ 16

2.3.2

Concepts and frameworks ....................................................................................................... 22

2.3.3

Linking adaptation and mitigation in agriculture and forestry ................................................ 30

2.4 3

Objectives, research questions, and structure of the dissertation .................................................. 4

Research gaps .............................................................................................................................. 34

RESEARCH DESIGN........................................................................................................................... 38 3.1

Levels/units and the temporal scale of analysis ........................................................................... 39

3.2

The global study (Paper 1)........................................................................................................... 41

3.2.1 3.3 3.3.1 3.4

Validity and reliability (Paper 1) ............................................................................................. 42 A case study research design (Paper 2, 3 and 4).......................................................................... 43 Case study methodology applied (Paper 2 and 3) ................................................................... 46 Study area, fieldwork methodology and reflections (Paper 2 and 3)........................................... 48

3.4.1

The study area: Belize ............................................................................................................. 48

3.4.2

Why is Belize interesting for this research topic? ................................................................... 50

3.4.3

Timing and fieldwork locations ............................................................................................... 52

3.4.4

Data collection methods .......................................................................................................... 57

3.4.5

Data availability and processing/analysis ................................................................................ 65

3.4.6

Validity and reliability ............................................................................................................. 68

3.4.7

Positioning, power and ethics in fieldwork ............................................................................. 71

3.4.8

Note on excluded survey (Paper 2) .......................................................................................... 74

viii

4

5

SUMMARY OF PAPERS ..................................................................................................................... 78 4.1

Paper 1 ......................................................................................................................................... 78

4.2

Paper 2 ......................................................................................................................................... 79

4.3

Paper 3 ......................................................................................................................................... 80

4.4

Paper 4 ......................................................................................................................................... 81

CONCLUSION ...................................................................................................................................... 82 5.1

Returning to the research objectives ............................................................................................ 82

5.2

Implications for policy................................................................................................................. 85

5.3

Perspectives for future research ................................................................................................... 86

REFERENCES .............................................................................................................................................. 87

PART II: PAPERS .................................................................................................. 102 Paper 1 ....................................................................................................................................................... 103 Paper 2 ....................................................................................................................................................... 139 Paper 3 ....................................................................................................................................................... 178 Paper 4 ....................................................................................................................................................... 215 APPENDICES.............................................................................................................................................. 233 Appendix 1: Interview guide (Paper 2) ..................................................................................................... 234 Appendix 2: Interview questionnaire (Paper 2 - English and Spanish) ..................................................... 240 Appendix 3: Village survey guide (Paper 2) ............................................................................................. 252 Appendix 4: Co-authorship statements...................................................................................................... 257

ix

PART I: SYNOPSIS

Market day (Punta Gorda town, Toledo District, Belize, February 2014).

Market day (San Ignacio, Cayo District, Belize, October 2013).

1

1

INTRODUCTION

The overall aim of this dissertation is to contribute to the empirical knowledge base on the opportunities and challenges of enhancing the linkages between adaptation and mitigation in land-use sectors. The findings should assist funding being invested more effectively, as well as lead institutions and project developers to consider linking adaptation and mitigation in climate-related projects and activities as a trajectory to enhance overall outcomes in combating climate change. Today, there is no longer any doubt about the need for both adaptation and mitigation if climate change is to be abated effectively, and there is agreement that actions are necessary at all spatial scales and that they should address short-term as well as long-term issues. However, the climate debate has in recent decades been focused on a global mitigation agreement with long-term perspectives, with less attention being given to adaptation. This was the starting point for making this Ph.D., and one could ask why climate projects still are one-sided regarding adaptation or mitigation when the doubts about the need for both have gone? The intuitive thought was that it must be feasible and better to make ‘climate’ projects that address both objectives. At least in the cases where interlinkages between adaptation and mitigation exists, because the divide can often be explained by the spatial and temporal differences and sectoral circumstances between the two objectives. However, in the land-use sectors, for instance, which is the field of research in this dissertation, this sharp split and narrow-minded focus appears at first sight remarkable. Others have made the same observation, which some scholars referred to it as ‘the false dichotomy’ framed by scientists and policy-makers (Tompkins and Adger 2005; Bizikova et al. 2007; Biesbroek et al. 2009; Harry and Morad 2013; Duguma et al. 2014a). And several studies had made reflections on a number of different types of linkage between adaptation and mitigation. However, the main way of implementing this linkage ten to twenty years ago was to assess the level of possible substitution between the two objectives, as the need and scope for adaptation was solely an aftermath of our (in)ability to mitigate climate change – a ‘cause and effect’ interaction (Dang et al. 2003). The idea, at that time, was to remove the artificial barrier between mitigation and adaptation in order to make a balance between them in global and national strategies (Wilbanks et al. 2003). Nevertheless, the policy attention towards adaptation started to increase, primarily triggered by the fact that past emissions started to show impacts, particularly in vulnerable developing countries, and because the international climate negotiations about climate change mitigation had made almost no progress (Swart and Raes 2007). These two circumstances might have been the main factors that moved adaptation to ‘the front seat’ next to mitigation, but maybe more interestingly, some scholars turned their research towards the interplay between the ‘driver’ (mitigation) and its ‘co-driver’ (adaptation).1 These scholars investigated the possibilities of complementarities or even synergies and whether their interplay was complicated by trade-offs (e.g. Kane and Shogren 2000; Dang et al. 2003; Klein et al. 2005; Burton et al. 1

These metaphors are used as mitigation still is in the ’driver’ in charge, but adaptation, ‘the co-driver’, now has some

influence on the direction.

2

2007; Klein et al. 2007; Swart and Raes 2007; Ravindranath 2007; Ayers and Huq 2009), and not just adaptation and mitigation as separate measures. These studies, not surprisingly, revealed a complexity of positive, neutral, and negative effects, but now it had been demonstrated and could not be neglected. Later, this interplay has expanded with development issues, as all three have substantial overlaps, and to increase effectiveness and efficiency, there is today a growing quest for creating synergies and suppressing trade-offs between these three objectives (Moser 2012; Harry and Morad 2013; Duguma et al. 2014b). For instance, “it has been suggested that linking adaptation to mitigation in some cases may contribute to repairing the discursive rift between climate policy and sustainable development” (Venema and Rehman 2007 in Ayers and Huq 2009), for which reason this trinity, if successful, has been referred to as ‘triple wins’, multiple ‘wins’, or simply as holistic in approach (Moser 2012; Duguma et al. 2014b; Suckall et al. 2015). It has even been commented “that the separate treatment of mitigation and adaptation in both science and policy has hindered progress against the fundamental sustainable development challenges of climate change” (Ayers and Huq 2009: 754). On the other hand, scholars are also noting that linking adaptation and mitigation into a single portfolio is no panacea and should not be implemented as a ‘forced marriage’, as it can lead to a ballooning of policies and projects that is unmanageable and cost-ineffective (Dang et al. 2003; Klein et al. 2005; Swart and Raes 2007; Wilbanks and Sathaye 2007; Locatelli et al. 2011). However, in some areas, like the land-use sector in the developing world, it is obvious that it should be considered as a central part of the overall guidelines (Duguma et al. 2014a; Suckall et al. 2015). Thus, although the issue has been treated earlier, as a research area it is still in its infancy and suffers from limited conceptual and empirical understandings. The hitherto proposed conceptual frameworks operate at a very superior level, making them difficult to operationalize, and the interpretation of the concepts is not clear (Klein et al. 2007), may can be a reason why the empirical base is very limited (Suckall et al. 2015). Most empirical evidence has narrowed its focus to particular topics, which often makes perfect sense and is understandable for reasons of complexity, but the downside is that important effects (positive and negative) are excluded as they are out of scope. Consequently, there are only a few studies that approach it more holistically, maybe because this approach also has some substantial downsides such as the if you want to say something about the whole, your statements becomes less distinct notion, which is apparently true. Conversely, if research is too specific, it might overlook connections at other scales. Thus, the scientific deficiency in this field of research is that empirical studies that do not only treat separate domains, but also address the overlaps between them. The main exceptions in the analytical literature have been attempts to construct global costbenefit analyses (Huq and Grubb 2007) where adaption and mitigation are purely substitutes, but this is also an approach that has been criticized (Smit et al. 2000; Biesbroek et al. 2009; Warren et al. 2012). Thus, the suggestion is to move beyond a single-impact focus and cost-benefit analyses if studies are to provide better understanding of the possibilities of linking adaptation and mitigation and to avoid trade-offs in projects and policies (Illman et al. 2013; Suckall et al. 2015). This is because little is known about the enabling

3

conditions for design and implementation to move from the current dichotomized approach to the synergy approach (Duguma et al. 2014a). This dissertation addresses this research gap by investigating the possibility of enhancing the empirical understanding of linkages between adaptation and mitigation in the land-use sectors. A stronger evidence base in this research field can ensure that funding is invested effectively, for instance, in ‘no-regret’ options that deliver multiple benefits, whereas the lack of an evidence base of practices to achieve this may significantly limit the capacity of donors to identify, monitor, and evaluate investments (Tompkins et al. 2013; Suckall et al. 2015). The outcome is believed to be a part of the process to overcome sustained conceptual divides between adaptation and mitigation, and it is believed that this will lead to a growing interest in mitigation in the countries that are focused on their adaptation needs, and likewise increase awareness of the benefits that adaptation can bring to mitigation measures. As stated by Biesbroek and colleagues, “it is not the incompatibility of the measures as such that makes an integrated approach to mitigation and adaptation difficult, but the current institutional fabric of directives, treaties and organizations for all sectoral policy domains across all scales of government as a result of the historic dichotomy between mitigation and adaptation which make mainstreaming of climate policy difficult” (Biesbroek et al. 2009: 234).

1.1

Objectives, research questions, and structure of the dissertation

As a response to these global challenges in linking adaptation and mitigation, the overall objective of this dissertation is to explore and assess the scope for and current practice of linking adaptation and mitigation in the agricultural and forestry sectors in the tropics. To attain this overall objective, four specific objectives have been identified: 1. to explore and elaborate on concepts and frameworks, and demonstrate the research gap in the linking of adaptation and mitigation; 2. to examine the extent of the linking of adaptation and mitigation in current agriculture and forestry projects globally; 3. to examine the modalities, potential, and motivations for linking adaptation and mitigation in agriculture and forestry activities; and 4. to suggest adjustments to existing applied instruments to address the linking of adaptation and mitigation to enhance synergies and avoid trade-offs Objective 1 is treated in Section 2 of the synopsis, which provides the overall conceptual approaches to the linking of adaptation and mitigation by highlighting the different frameworks and concepts used in this field of research. This section also includes a scrutiny of the historical background of adaptation and mitigation, as well as a discussion of the definitions of adaptation and mitigation and the perceptual overlaps/distinctions between them. The section furthermore contains an elaboration of the potential for linking adaptation and mitigation in agriculture and forestry, as they are the ones examined in the papers. Finally, research gaps are identified.

4

Objectives 2-4 are examined in the four papers, the first one being based on a review of project documents, and the other three are case studies based on fieldwork in Belize and Ghana. The different methods and techniques chosen for the papers are discussed in Section 3. Section 4 provides summaries and the main findings of the four papers, and Section 5 concludes by returning to the research questions, suggests

policy

implications,

and

provide

perspectives for future research. The four separate papers in Part II address different aspects of the research objectives. They represent a

Figure 1. Visualization of the primary spatial scale and land-use system addressed in the papers.

gradual decrease in scale, as well as different explorations of the adaptation-mitigation link in relation to different land-use intermixtures. Figure 1 illustrates the primary spatial scale of the papers and shows to what extent they address agriculture and/or forestry. Outline of the papers is: 

Paper 1 is a global study that investigates the link between adaptation and mitigation in both agricultural and forestry projects. The main questions are to what extent proposed adaptation and mitigation projects in the agricultural and forestry sectors contribute to each other’s objectives, and does the type of project, sector, and geographical region matter? Secondly, what is the potential for enhancing the linking between adaptation and mitigation in agricultural and forestry projects? Thirdly, what are the modalities and incentives for making this link?



Paper 2 examines linking between adaptation and mitigation in avoided deforestation projects in Belize. The questions examine how adaptation is relevant in carbon sequestration projects in the forestry sector in Belize. And if these projects include adaptation in their design and implementation, and what motivations project developers have for including or excluding adaptation concerns.



Paper 3 scrutinizes barriers in the linking of adaptation and mitigation in maize production in the agricultural sector in southern Belize. The questions asked being how relevant is it to integrate adaptation and mitigation in maize production among farmers in southern Belize, and if so, what are the barriers to doing this?



Paper 4 investigates the mitigation potential of tree crop plantations in Ghana to assess whether these plantations can enter into carbon payment schemes, which could influence adaptive capacity locally. The questions addressed are what is the potential mitigation effect of the main tree crop plantations in Ghana, and do these plantations in the tropics have the ability to partake in forest payment schemes and thereby influence adaptive capacity locally?

5

2

CONCEPTUAL APPROACHES TO LINKING ADAPTATION AND MITIGATION

The climate change literature, especially that concerned with the linking of adaptation and mitigation, forms the basis of the papers in this dissertation. But before turning to the literature on the practice of linking, the background and definitions of adaptation and mitigation will be clarified. This is followed by a conceptualization of the linking of adaptation and mitigation, in which the frameworks in this field of research are analyzed. After this, the linking of adaptation and mitigation in the agriculture and frestry sectors are scrutinized. This is followed by an elaboration of the gaps in research, highlighting especially those this dissertation is trying to fill.

2.1

Background: the adaptation-mitigation dichotomy

The question of dealing with climate change is divided into adaptation, which has been defined as “adjustment in natural or human systems in response to actual or expected climatic stimuli or their effects, which moderates harm or exploits beneficial opportunities”, and mitigation, which is any “anthropogenic intervention to reduce the sources or enhance the sinks of greenhouse gasses” (IPCC 2001). The distinction is also reflected in the working group structure of the IPCC (WGII: Adaptation and WGIII: Mitigation), funding mechanisms (e.g. CDM, JI and NAMA are mitigation, and Adaptation Fund and NAPA are adaptation) and in the scientific sphere, where scholars mostly concentrate on one of these objectives and disregard the other. However, the overall objective of both approaches is the same: to deal with climate change. However, this sharp dichotomy makes less sense today (Biesbroek et al. 2009), as it is widely recognized that adaptation and mitigation are interlinked and that both are necessary for combating climate change (IPCC 2014a). To uncover the explanation for this dichotomy, we have to dig into the history of the climate negotiations. In the 1980s, scientists placed climate change on the global environmental agenda, which also affected the policy agenda. Thus in 1988 the UN General Assembly formally took up climate change for the first time, and in 1992 the UN enacted the UNFCCC in order to facilitate coordination of an international response to anthropogenic climate change (Burton et al. 2007). At that time, science understood climate change to be an environmental pollution problem in the same manner as two other major atmospheric problems: acid rain and the ozone layer depletion, which were both critical transboundary and systemic environmental challenges that had both been handled successfully through international cooperation. Therefore, to solve climate change a policy framework approach was applied in which neutral scientific assessments were conducted (by the IPCC), which were communicated to policy-makers (by the UNFCCC), and were to be transformed subsequently into international emissions targets through a global environmental regime (Burton et al. 2007; Ayers and Huq 2009). At this stage, it was believed that mitigation, i.e. reducing emissions and increasing sinks, was sufficient (Wilbanks et al. 2003), and politicians and scholars who only mentioned adaptation were frowned on, as this was viewed as giving up prematurely (Tol 2005; Moser 2009). Proponents of adaptive strategies were thus ignored as ‘defeatists’, ‘fatalistic’ and linked to ‘do nothing strategies’ (Biesbroek et al. 2009). Some of the

6

latter were referred to as ‘adaptationists’, i.e. those who saw no need to take any action, as they simply trusted in autonomous adaptation and expect future generations to have more income and more sophisticated technologies to enable them to cope with future climate change (Dang et al. 2003; Goklany 2007). Thus, adaptation was treated as a taboo subject, as it was perceived either as representing a failure of mitigation or as a way to weaken mitigation efforts (Pielke et al. 2007). For instance, Al Gore2 forcefully declared his opposition to adaptation in 1992 by stating: “believing that we can adapt to just about anything is ultimately a kind of laziness, an arrogant faith in our ability to react in time to save our skin” (Pielke et al. 2007: 597). Additionally, adaptation was also seen as a disreputable and unfair plan B, as politicians and scholars already knew back then that it would hit developing countries harder, which were held to be irresponsible for causing the problem and as having lower adaptive capacity, whereas the “responsible”3 developed countries had abundant capacity to resist the impacts (Denton and Wilbanks 2014: 1117). As a consequence, consensus was based on the “polluter pays principle” – or the if you broke it, you fix it approach (Garvey 2008: loc. 2041), which also suits public opinion which points at the countries that caused the problem when it comes to resolving it (e.g. Klinsky et al. 2012). The “polluter pays” approach, which today is a fundamental part of climate finance politics, has created a new global relationship between developed and developing countries (Mitchell and Maxwell 2010). Whereas before, in some cases, we were talking about charity and aid, the word now being used in the climate talks is “compensation”. This has been a game-changer, as compensation is less humiliating because it is something you have a right to. However, the notable differences between adaptation and mitigation, especially spatial and temporal scales (Jones et al. 2007; Wilbanks et al. 2007), also played a major role in creating this dichotomy and the focus on mitigation internationally. Mitigation was highly regarded as a global issue to be tackled internationally, whilst adaptation was a local concern to be dealt with by national or local policy-makers. Temporally mitigation had to be dealt with right now, whereas adaptation was considered a problem for future generations, if it were to occur at all (Swart and Raes 2007). This can also be seen in the ultimate objective of the UNFCCC, which is to achieve “stabilization of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system [and] within a time frame sufficient to allow ecosystems to adapt naturally” (UNFCCC 1992: Art. 2). It is also reflected in the IPCC’s third assessment report, which states that “climatic changes today still are relatively small, thus there is little need for adaptation, although there is considerable need for mitigation to avoid more severe future damages. By this logic, it is more prudent to invest the bulk of the resources for climate policy in mitigation, rather than adaptation” (McCarthy et al. 2001).

2

American politician, Vice President of the United States (1993–2001), and climate change and environmental activist.

3

See Garvey (2008: Ch. 3) for elaboration of responsibility in climate change.

7

Pielke (2005) comments that the UNFCCC’s definition of adaptation (also adopted partly by IPCC WGIII) has created a bias against adaptation, as adaptation in this definition only has costs and is viewed as something that should only be employed when mitigation fails to prevent impacts (more on definitions in Section 2.2). One consequence of the UNFCCC’s narrow definition of climate change is that it necessarily subjugated all climate policy to energy policy, as it is widely considered to be the only policy tool that can control the “global machinery”, which fashioned the mitigation focus (ibid.). Nonetheless, a framework for adaptation actions had already been set up within the legal frameworks of the UNFCCC from 1992 (Article 4), which, for instance, recognizes the need for adaptation as well as mitigation by stating that “All Parties ... shall ... formulate, implement, publish and regularly update national and ... regional programmes containing measures to mitigate climate change ... and measures to facilitate adequate adaptation to climate change” (UNFCCC 1992: Art. 4.1b); and further, that countries shall “Cooperate in preparing for adaptation to the impacts of climate change” (Art. 4.1e), and that developed countries shall assist developing countries to meet the costs of adaptation (Art. 4.4). Retrospectively, viewing climate change as an environmental pollution issue that could be solved by international agreements resulted in a mitigation-adaptation dichotomy with a narrow focus on mitigation (e.g. Dang et al. 2003; Burton et al. 2007; Swart and Raes 2007; Biesbroek et al. 2009; Moser 2012). From a precautionary principle4 point of view, which often is an important principle in environmental politics, it was sound of negotiators in the beginning to support mitigation as a primary instrument against the challenge ahead. However, the statements regarding causes and effects in the IPCC reports became stronger, and already in 2001, the IPCC (2001) underscored that significant impacts of climate change were unavoidable, which justified including adaptation strategies before climate change impacts materialized if the precautionary principle was to be followed (Dang et al. 2003). Smith (1997) notes that, “if climate change is inevitable, adaptation to its effects is also inevitable”, and other comment that the need for adaptation to climate variability and extreme events in some regions being contemporary and not just a matter of proactive measures (Smit et al. 2000). A few years later, scholars emphasized that collective adaptation efforts are likely to be needed in most countries (Wilbanks et al. 2003). Therefore, since the Earth Summit (Rio 1992) and COP1 (Berlin 1995), attention to the necessity for adaptation should have increased gradually, as the following negotiations (mainly the COPs) produced very limited results with regard to reducing emissions significantly, and impacts of climate change started to occur (Thompson and Bendik-Keymer 2012: loc. 138). And the attention given to adaptation as an alternative or complementary response strategy to mitigation has increased since the turn of the millennium (Smit et al. 2000; Ayers and Huq 2009; Biesbroek et al. 2009; Moser 2012: Thompson and Bendik-Keymer 2012). Adaptation was translated into policy in 2001 (COP7 in 4

The precautionary principle has been interpreted by many, but the basic idea is to avoid doing actions where the

consequences are uncertain (Gardiner 2006).

8

Marrakech) through the creation of three funds focusing on it, including the 2% CDM-levy financed Adaptation Fund5 and again in 2007 (COP13 in Bali), when adaptation was elevated to be a ‘building block’ alongside mitigation, technological cooperation, and finance (Ayers and Huq 2009). During the same period the production of knowledge on climate change was reshaped from being framed as an environmental problem, with studies not considering adaptation at all (Tol 2005), towards a human-influenced development issue, and it has shifted from being a mono-disciplinary approach with an emphasis on the natural sciences “towards [being] a transdisciplinary research strategy with stronger focus on co-production of scientific knowledge between natural and social scientists, policy makers, and society to support political decisions in the context of sustainable development” (Biesbroek et al. 2009). Nevertheless, only a decade ago the primary focus was still on mitigation, for instance, in the national communications, in which, although adaptation and mitigation were mentioned simultaneously, adaptation was frequently not treated as intensively as mitigation, and government officials and analysts perceived mitigation and adaptation as discrete or pure substitutes of one another (Dang et al. 2003). Further, they (ibid.: S83) asked why it has “taken such a long time for researchers and policy makers alike to pay adequate attention to adaptation?” and why almost all policy-makers and negotiators have “considered mitigation and adaptation as unrelated policy alternatives or even opposing the latter?” In 2007, Burton and colleagues wrote that the underestimation of adaptation largely still persists, and when adaptation is considered it is, unfortunately, “often merely to lament the lack of adaptive capacity in the poorest and most vulnerable communities and countries”, so that “research into the processes and techniques of adaptation still continue to be neglected” (Burton et al., 2007: 373). A few years later, Ayers and Huq (2009: 756) commented that there remains some degree of institutional bias toward mitigation over adaptation despite progress, which is “limiting the scope of action on adaptation to contribute to sustainable development”. Moser (2012) claims that adaptation still follows the history of sometimes disharmonious or at least one-sided discussions of climate change and that it still lags far behind mitigation. She adds that “There is considerable communication yet to be done to challenge and replace old attitudes and perspectives which view mitigation and adaptation as alternatives, rather than as complementary and necessary approaches to managing climate risks” (ibid.: 172). She states that resistance to tackling adaptation is still often based on the belief that talking about adaptation is a form of capitulation (ibid.). Furthermore, discussions about the spatial and temporal scale differences between adaptation and mitigation still take place, but they are less pronounced, as the overlaps are greater than hitherto assumed (see Section 2.3.1 for discussion of overlaps). Nevertheless, the mitigation strategy is mainly a top-down approach that is considered easier to implement in comparison to adaptive strategies, which largely involve a bottom-up approach. And the latter is associated with complicated policy strategies such as participatory approaches, multi-level governance or multi-stakeholder platforms (Biesbroek

5

The others were the Least Developed Countries Fund and the Special Climate Change Fund, both under the auspices of

the Global Environmental Facility (Wilbanks et al. 2003).

9

et al. 2009), which can be instrumental in the continued favoring of mitigation. However, this bottom-up/topdown difference might be a misperception, as mitigation in some cases has to come from the ground up (Kates and Wilbanks 2003). In spite of the increased political attention being given to adaptation, where adaptation now is a prominent aspect (Ayers and Huq 2009), the disharmony between adaptation and mitigation is still apparent in terms of funding, with a dominant role for mitigation. Back in 2007, Pielke and colleagues called adaptation mitigation’s “poor cousin” in the climate policy arena, as historically it had been treated as a marginal policy option by climate change scientists and decision-makers (Pielke et al. 2007: 598). A label that still sticks. In 2010/11 as much as 96% of global climate finance (350 billion USD) was allocated to mitigation activities alone (Buchner et al. 2012) and in 2013 adaptation was still being deprioritized financially, though the figures indicated some improvement, 80% of funding going on mitigation (of which some 15% for REDD), 15% on adaptation, and 5% of funds being distributed to projects with multiple foci (Illman et al. 2013: 33). Besides, the financing mechanisms for adaptation and mitigation are still segregated (Duguma et al. 2014b), and, like a decade ago, the notion remains that money spent on adaptation will reduce the funds available for mitigation and vice versa (Dang et al. 2003). With regard to this factor and to the overall topic of this dissertation, it is naturally very interesting (alarming?) to notice that in 2013 only USD 4 billion or around 1% of climate financing was used for activities with joint mitigation and adaptation objectives (Buchner et al. 2014).

2.2

Defining adaptation and mitigation

In order to analyze the link between adaptation and mitigation, it is necessary to clarify the definitions and substances of mitigation and adaptation. They both represent responses to climate change and are driven by the same problematic, but they have fundamentally dissimilar approaches and essential differences (e.g. Smit et al. 2000; Tol 2005; Klein et al. 2007; Swart and Raes 2007; Wilbanks and Sathaye 2007; Locatelli et al. 2011), which will be elaborated further after discussing the definitions. Climate change novices often become confused over the meanings of adaptation and mitigation, and even replace them, which can appear peculiar to those more versed in matters of climate change. The confusion arises because to mitigate means to abate, moderate or alleviate some expected impacts – at least in engineering and insurance, or in the field of environmental hazards. But in ‘climate change science’ it is considered to be a response in the form of “reducing or stabilizing GHG emissions or levels, in order to mitigate changes in climate” (Smit et al. 2000; 224-5, emphasis added). In the climate change context, mitigation focuses on the causes of climate change (Suckall et al. 2015), and scholars have referred to mitigation as indirect damage prevention (Verheyen 2005), actions taken to reduce the extent of climate change (Tompkins et al. 2013), or actions to avoid the unmanageable (Wilson and Piper 2010). The most commonly cited and applied definition of mitigation runs: “anthropogenic intervention to reduce the sources or enhance the sinks of greenhouse gases”, which comes from the IPCC’s third assessment report (IPCC 2001), and is still used in

10

the fifth assessment report (IPCC 2014a). Mitigation activities can broadly be grouped into five main areas: efficient use of energy (i.e. reducing system waste); use of renewable energies (such as solar, biofuels, wind, ocean thermal exchange); carbon sequestration through enhanced sinks (e.g. reforestation, afforestation); reduced sources of emissions through land use management; and macro-engineered carbon capture and storage (Tompkins et al. 2013). Nevertheless, even though it is the definition related directly to climate change that is used in the papers in this dissertation, mitigation could be viewed in a less reductionistic fashion in the shape of a broader concept. In other words, not narrowly focused on climate change, but also on non-climate-related stressors, such as economic crisis and political conflicts. To adapt, on the other hand, means to modify or alter something to make it more fit for a purpose, while adaptation, a term often used in ecology and social science, “refers to both the process of adapting and the condition of being adapted” (Smit et al. 1999: 203). One climate-related definition of adaptation is adjustments “in ecological-social-economic systems in response to actual or expected climatic stimuli, their effects or impacts” (Smit et al. 2000), but it comes in both narrower and broader versions. This requires its users to specify which version is being used with regard to who (i.e. people, social and economic sectors and activities, managed or unmanaged natural or ecological systems, or practices, processes or structures of systems), what (i.e. climate change, climate variability and/or extreme events) and how adaptation occurs (ibid.). Adaptation is used to describe all activities aimed at preparing for or dealing with the consequences of climate change (Warren et al. 2012), for instance, building dams in flood-prone areas or cutting firebreaks in fire-prone areas. Scholars have referred to adaptation as direct damage prevention (Verheyen 2005) and as managing the unavoidable (Wilson and Piper 2010), as well as actions taken to ameliorate impacts (Tompkins et al. 2013) or actions needed to avoid adverse impacts (Duguma et al. 2014a). Various types of adaptation can be distinguished with regard to timing (anticipatory/proactive/ex-post, spontaneous/concurrent/during, or responsive/reactive/ex-ante), implementers (private and public), awareness (autonomous or planned), intention (intended or unintended), type (technological, behavioral, economic, legal, or institutional) (Smit et al. 2000; Ravindranath 2007; Harry and Morad 2013), or level (incremental, systematic, or transformational) (Rickards and Howden 2012). The most frequently applied definitions are those of the IPCC and UNFCCC, which respectively are that adaptation is “The process of adjustment to actual or expected climate and its effects. In human systems, adaptation seeks to moderate or avoid harm or exploit beneficial opportunities. In some natural systems, human intervention may facilitate adjustment to expected climate and its effects” (IPCC 2014b), and “Adaptation refers to adjustments in ecological, social, or economic systems in response to actual or expected climatic stimuli and their effects or impacts. It refers to changes in processes, practices, and structures to moderate potential damages or to benefit from opportunities associated with climate change” (UNFCCC 2015). However, there is a variety of definitions of adaptation (Dang et al. 2003; Suckall et al. 2014), and discussions concerning meaning and interpretation have been going on since the early 1990s. The IPCC, for instance, states in its assessment report five that they have adjusted the definition compared to earlier IPCC reports to reflect

11

progress in science (IPCC 2014b). The main disagreement in the discussion of adaptation is rooted in how wide a definition to apply, which is not solely a trivial academic debate, as the definition will profoundly influence the available finances and the feasibility of implementation. The main criticism that has been pointed at institutions (i.a. the UNFCCC) that use adaptation in the narrowest sense, by only referring to actions taken in response to climate change resulting from anthropogenic GHG emissions, and thereby excluding the many societies that are already maladapted to current climate conditions (Pielke 2005; Klein et al. 2005). However, this narrow interpretation of climate change is widely accepted, resulting, for instance, in estimates where adaptation actions only represent a cost of human-caused climate change that would be avoided by emissions reduction, thus placing greater emphasis on mitigation, as seen in the past. In contrast to these institutions, the IPCC, like many decision-makers and researchers, does interpret adaptation as “a much broader range of actions that make societies more robust to changes, including, but not limited to, those caused by climate change” (Pielke et al. 2007: 598). The issue is that the UNFCCC’s restricted definition of climate change only focuses on changes in climate that result from human activity (directly or indirectly), which makes sense theoretically, but is very difficult to handle in practice and in the real world of politics and policy (Pielke 2005). Furthermore, this line of thinking will meet little understanding among “those experiencing devastating losses from climate impacts [human or naturally induced] in developing countries” (Pielke et al. 2007). A decade before Pielke, scholars were already suggesting that adaptation and vulnerability should be addressed to climate variability and not solely to long-term climate change (Ribot et al. 1996), and that actions should address current problems of sustainable development in the light of variable and uncertain environments (Smithers and Smit 1997). This points backwards to the ‘no-regret’ term, or measures worth doing, no matter what (IPCC 1995). Furthermore, it was acknowledged that in general higher capacity could prevent ecosystems and societies from dangerous impacts until the limits of adaptation were reached (Klein et al. 2007). Hence, development was recognized as having a significant amount of overlap with adaptation (and vice versa), as reducing vulnerability and increasing adaptive capacity benefits both agendas (Huq and Reid 2007). This advance represented a significant change in perception in handling climate change impacts, because it was no longer merely the physical exposure to some environmental catastrophe that resulted in environmental vulnerability, but something that was deeply linked with the political, economic, and social contexts in which people lived (Ayers and Huq 2009). This was why “effective adaptation to climate change should include core development concerns associated with vulnerability including but not limited to those directly caused by climate change” (ibid: 756). This shift happened in the aftermath of the growing attention being given to political ecology, a research approach to environmental issues and changes that place increased emphasis on the influence of political, economic and social aspects, which differs from apolitical approaches such as cultural ecology or natural science in general. Political ecology is about power relationships and disregards boundaries, contrary to, for instance, land change science, which use boundaries and classes to understand and interpret land changes in a reductionist fashion or as a model of the world (Robbins 2012).

12

Thus, climate change has gradually been accepted as a development issue (Lemos et al. 2007; Reid and Huq 2007) and, during the last decade, the notion of adaptation has evolved in the direction of making societies more robust and flexible in general, which will facilitate robustness to climate change and variability, but also to socio-economic changes, many of which are more rapid than global warming (Tol 2005). Today “there is general acceptance that adaptation involves long-term adjustments within a system […] to better manage external stress” (Suckall et al. 2014), and the overlap with development becomes very natural, as the two have many cross-cutting issues that would contribute to both objectives, for instance, poverty reduction and food security (Mitchell and Maxwell 2010). This involvement with development has made the goals of adaptation less clear-cut, as the actions consist in a wide palette of arrangements, for example, agricultural and land tenure policies, infrastructure, irrigation systems, and livelihood diversification programs (Ayers and Huq 2009). Another significant change is that adaptation actions are not limited to a cost (Goklany 2007; Duguma et al. 2014a), but today are also seen as strategies that “enhance and implement measures to reduce, moderate and take advantage of climate impacts at all scales”. These are therefore aimed at creating diversity; promoting skills, learning and innovation; reducing dependence on ecosystem services; reducing inequality; and ensuring the participation of all stakeholder groups in decision-making processes at multiple levels (Mitchell and Maxwell 2010). In the papers in this dissertation, there is applied the broader definition, namely “actions that make societies more robust to changes, including, but not limited to, those caused by climate change” (Pielke et al. 2007: 598). Consequently, adaptation can be achieved by reducing exposure or sensitivity to climate change, or by enhancing the capacity to adapt (Harry and Morad 2013). In using this definition, a humanist viewpoint at adaptation is adopted, where we adjust “our conception of who we are to appropriately fit the new global context […] instead of attempting to mitigate the impacts of climate change on who we are at present” (Thompson and Bendik-Keymer 2012: loc. 199). Why aim at protecting our contemporary life-form, which in most cases is unsustainable, when we have the opportunity to make changes that reflect the realities and requirements we will unavoidably face in the new world that we are actually already a part of? Thus, we have to adjust who we are and enable ourselves to meet new and projected conditions (ibid.). This view is in contrast to the ‘pollutionist’ and ‘development’ views of adaptation identified by Burton (2009 in Thompson and Bendik-Keymer 2012), where the pollutionist’s stand is that higher pollution levels are acceptable if the vulnerability of those affected is reduced. This means the more adaptation the higher the GHG concentration to be accepted, which is a cynical perspective. Nevertheless, this is the way we have headed in recent decades, with very limited success in decarbonizing the global economy (Thompson and Bendik-Keymer 2012). Some argue that developed countries have chosen this trajectory deliberately, as adaption is cheaper for them than mitigation (Dang et al. 2003; Ayers and Huq 2009). On the other hand, the ‘development’ view, which is held by many, focuses on buffering societies from climate change, but in a mainstream fashion, by implementing adaptation to development planning. This makes adaptation just a form of development, as in building water

13

systems and electricity, which enhances the resilience of communities to any harm, including climate change. This view does not favor preserving current emissions as the pollutionist, but about protecting human capabilities. However, the drawback is the limited attention this gives to fundamental changes in how we live, as the imperatives of development continue unaltered (Thompson and Bendik-Keymer 2012: loc. 195). As seen in this discussion of adaptation, climate change is not simply a science or matter of economics, but very largely also an ethical one, with conflicting world views and moral disagreements about justice, fairness, and equity. Many of the conflicts are not about factuality, at least not anymore for most of us, but disputes about principles and their application (Garvey 2008: loc. 715). “Science can tell us what is going on, but not what we should do about it”, as it mostly depends “on what we value and how we think about our values” (ibid: loc. 512). Furthermore, “The question of what they [governments and other agents] should do is a normative one and thus has ethical dimensions because it generally involves the conflicting interests of different people. The answer rests implicitly or explicitly on ethical judgments” (Kolstad and Urama 2014). The ethical discussions have especially been about the issues of responsibility and the distribution of benefits and costs. Should developed countries bear the total responsibility because of their emissions in the past, or should developing countries mitigate as well,6 or should emission targets and cuts be in line with consumption rather than production (as suggested by Harry and Morad 2013)? The willingness to mitigate is further complicated by the circumstance that the direct benefits of mitigation activities usually fall on others, in other places and/or at other times (Swart and Raes 2007). The IPCC provides six different perspectives on the sources and possible mitigation obligations for world emissions, which all contain ethical questions and demonstrate the complexity of the climate negotiations (Victor and Zhou 2014: Ch. 1.3.2). Beyond the responsibility question, generational justice is also being discussed, which has two dimensions: intragenerational equity, or the responsibility for increasing the well-being of the most disadvantaged in today’s society, and intergenerational equity, or ensuring that the prospects of future generations are not compromised (Fleurbaey and Kartha 2014). The ethical discussions are mainly triggered by the fact that significant temporal and spatial differences exist. Different countries have very dissimilar emissions (in the past, currently, and will have in the future) and climate change damage is not expected to be distributed spatially equally. As a result, some countries will face greater damage than others, and some might even benefit from the changes (Dang et al. 2003). The main unfairness is that those who had limited emissions in the past are those who will be affected the most in the future, which is why they deserve support for adaptation from the emitters who are responsible (Wilbanks and Sathaye 2007; Illman et al. 2013; Duguma et al. 2014a). Conversely, some regions and agents will benefit

6

Growth in global emissions will most likely come from developing and least developed economies (Raupach et al. 2007), as several developing countries have become major emitters of greenhouse gases and with emissions levels rapidly increasing (Illman et al. 2013).

14

from a changing climate (Goklany 2007), for instance, at higher altitudes and latitudes some currently nonarable landscapes may be transformed into potentially productive croplands (Jarvis et al. 2011). Moreover, benefits can be created by climate policy by providing funding opportunities for reducing and storing carbon, or benefitting the biofuel industry or the lithium industry from the increased demand for cleaner fuels and newer generation of batteries respectively (Mitchell and Maxwell 2010). These examples must be mentioned, as they show that climate change is not solely about losing,7 but certainly also has some winners (both historically and in the future), which it is important to acknowledge and to introduce into the ethical discussion, as ‘winners’ might have a moral responsibility for the ‘losers’ (Pielke 2005). Furthermore, ignoring the winner issue is problematic, as it can lessen the chances of policy-makers taking any kind of action (Glantz 1995 in Pielke 2005) and can suppress eventually hidden agendas. These were some of the ethical comments that are important to bear in mind when reading this dissertation. The issue has been elaborated further by numerous writers (e.g. Adger et al. 2006; Garvey 2008; Adger et al. 2009; Thompson and Bendik-Keymer 2012; Hayward 2012; Klein et al. 2014: Ch. 16.7; and publications by Stephen Gardiner, Dale Jamieson, Peter Singer, and Henry Shue).

2.3

Conceptualizing the linking of adaptation and mitigation

In this dissertation, it is mainly the word ‘linking’ that is utilized as the key term, by which is meant to ‘join’ or ‘couple’. In this case it is linking of adaptation and mitigation, thus not treating them by default as separate measures, but as measures that potentially have complementarities and synergies. The term ‘linking’ has been used in a similar way by others (Kane and Shogren 2000; Verchot et al. 2007; Bizikova et al. 2007; Ayers and Huq 2009; Martens et al. 2009; Locatelli et al. 2011; Suzuki 2012). However, a wide range of terms have been applied by the climate change research community, development organizations, policy-makers, NGOs and practitioners to cover the same or similar topics. Some use ‘integration’ (Klein et al. 2003; Wilbanks et al. 2003; Goklany 2007; Swart and Raes 2007; Wilbanks and Sathaye 2007; Ayers and Huq 2009; Jarvis et al. 2011; Thuy et al. 2014), others ‘interaction’ (Rosenzweig and Tubiello 2007; Warren 2011; Warren et al. 2012), and yet others ‘inter-relationships’ (Huq and Grubb 2004; Klein et al. 2007; Neufeldt et al. 2010). Which all are terms also utilized in this dissertation to cover the same as ‘linking’. In the literature referred above, the terms are used inconsistently to cover what here is referred to as linking, and any discussions related to different meanings of them has not been located, and it has not been possible to determine the reasons for different scholars’ selection of term. The most significant difference in the literature relates to definitions of adaptation, but this is not mirrored in the term used. Nevertheless, some use ‘synergies’ deliberate (Dang et al. 2003; Ravindranath 2007; Smith and Olesen 2010; Illman et al. 2013; Rahn et al. 2013; Duguma et al. 2014b), 7

This is the reverse of a statement from the Executive Director of the United Nations Environment Programme, who in

2001 stated, “There are no winners, only losers, in the climate change scenario. Now is time to act collectively and decisively” (cited in Pielke 2005).

15

which only covers one type of linking, but not the topic as a whole, as explained in the Section 2.3.2 on frameworks and concepts. But before going on to this it will highlighted in which areas adaptation and mitigation are assumed to be distinct or to overlap respectively in order to provide an understanding of the potential challenges and/or opportunities involved in linking them. 2.3.1

Perceptual overlaps and distinctions between adaptation and mitigation

Similarities and differences between adaptation and mitigation have been widely discussed in recent decades, especially in the literature advocating or opposing their integration. The main distinctions have been targeted at spatial, temporal, and sectoral issues, but scholars differ when discussing these. Furthermore, the gap seems to have decreased during the period, mainly because limited mitigation action has occurred, leading to calls for an increased emphasis on adaptation. In the past, the perception was that commonalities were minor, for instance, that jurisdictions were different (Dang et al. 2003; Wilbanks et al. 2003; Tol 2005), but other scholars have emphasized that the gap is smaller than perceived (Tompkins and Adger 2003; Pielke 2005; Pielke et al. 2007; Biesbroek et al. 2009; Moser 2012). It must emphasized that we not deny the conceptual differences between the adaptation and mitigation, but in the papers we follow the latter group of scholars when they recommend a ‘less sharp, less categorical rendering of them’ (Lawson 2003: 200; cited in Tompkins and Adger 2003). The dominant focuses or ‘common wisdom’ (Swart and Raes 2007) of the two strategies found in the literature is shown in Table 1, together with some exceptions. To provide a better overview it is divided into different issues, but this breakdown has several overlaps. First, the sharp and simple spatial division that adaptation is local and mitigation is global (Table 1) is less pronounced in the present discussion, as this notion has been challenged. For instance, Swart and Raes (2007) and Moser (2012) mention examples where adaptation actions can have global consequences (e.g. more resistant crops that are grown globally) and mitigation actions that will have local consequences (e.g. less local air pollution by closing coal-fired power plants, or the effect of biofuels on local food security). Furthermore, mitigation has started to be an important local issue with regard to implementation and cooperation, as individuals, communities, cities, states and companies take action without any direct payments or benefits (Swart and Raes 2007), but simply out of moral or altruistic reasons. However, some might do this for economic incentives too, as altruistic behavior is a part of their business model, 8 while others simply see opportunities in the ‘green economy’ that is developing, and yet others see it as a ‘no-regret’ action, as with energy-efficient solutions (ibid.).

8

Some of these entities might refer to mitigation initiatives, but with limited actions included, which has been referred to

as ‘green-washing’ (Laufer 2003).

16

Table 1. The dominant focuses of adaptation and mitigation, including deviations. Sources (in chronological order): 1 Dang et al. 2003; 2 Tompkins and Adger 2003; 3 Wilbanks et al. 2003; 4 Adger et al. 2005; 5 Murdiyarso et al. 2005; 6 Tol 2005; 7 Burton et al. 2007; 8 Goklany 2007; 9 Klein et al. 2007; 10 Pielke 2005 and Pielke et al. 2007; 11 Ravindranath 2007; 12 Swart and Raes 2007; 13 Wilbanks and Sathaye 2007; 14 Wilbanks et al. 2007; 15 Ayers and Huq 2009; 16 Biesbroek et al. 2009; 17 Mitchell and Maxwell 2010; 18 Locatelli et al. 2011; 19 Tanner and Allouche 2011; 20 Banerjee 2012; 21 Matocha et al. 2012; 22 Moser 2012; 23 Warren et al. 2012; 24 Harry and Morad 2013; 25 Illman et al. 2013; 26 Tompkins et al. 2013; 27 Duguma et al.2014b. Issue

Spatial scale in general

Adaptation: dominant focus       

Spatial scale: degree of cooperation required

           

Mitigation: dominant focus

italics indicate deviations from the dominant focus unfair: the ‘victims’ are not always responsible for causing climate change (1  free-riding problem, especially motivated with countries less vulnerable to climate change (1 more a problem for developing countries, but will also affect developed countries (12 until recently, considered to be a problem of developed countries (12 attention to adaptation tends to increase as the scale of attention becomes more fine-   has been treated as an issue for developed countries, which bear the greatest responsibility grained (13 for climate change, and because the mitigation capacity in the South was lower (15 more of a priority for vulnerable developing countries in the South, whose mitigation prioritized by industrialized countries that have the greatest responsibility for climate capacity is lower than in the North, yet whose vulnerability to climate change is acute (15  change (15 adaptation has often been preferred by developing countries that are more vulnerable to priority for all countries (17 climate change impacts and that have lower mitigation capacity given their low level of  emissions compared with developed countries (25  primarily an international issue, as mitigation provides global benefits (18 increasingly relevant for both developed and developing countries (25  increasingly relevant for developed and developing countries (25 despite increasing adaptation challenges, developed countries continue to focus on  despite increasing adaptation challenges, developed countries continue to focus on mitigation, while adaptation is a key priority for more vulnerable developing countries (27 mitigation, while adaptation is a key priority for more vulnerable developing countries (27 national, regional (1  global (1 bottom-up decision-making (2  top-down decision-making (2 successful adaptation is a combination of efforts from the local to the regional, national  primarily a matter of national governments in the context of international negotiations (6 and international scales (4  addresses, first and foremost, the Ministries of Energy and Finance and, to a lesser extent, primarily a matter of local managers of natural resources, and individual households and the Ministries of Transport, Agriculture, Forestry and, perhaps, Environment (6 companies, in the context of a regional economy and society (6  primarily justified by international agreements and the resulting national public policies aimed at local water managers, farmers, health officials, coastal zone managers, tourist (9 suppliers, architects, or energy suppliers; decision-makers on the national level would only  relatively few public or corporate decision-makers have direct responsibility for both be involved tangentially (6 adaptation and mitigation. For example, adaptation might reside in a Ministry of decisions are often made at the sub-national and local levels (9 Environment, while mitigation policy is led by a Trade, Energy or Economic Ministry (9 relatively few public or corporate decision-makers have direct responsibility for both  lower political urgency/legitimacy (although there are exceptions to this, e.g. short-term adaptation and mitigation. For example, adaptation might reside in a Ministry of co-benefits, local air pollution, energy security and jobs enhance urgency (12 Environment, while mitigation policy is led by a Trade, Energy or Economic Ministry (9  concrete mitigation as well as adaptation actions necessarily involve decisions by most adaptation is motivated by the private interests of affected individuals, households individuals at the local level (12 and firms, and by public arrangements of impacted communities and sectors (9  administrative scales: predominantly (inter)nationally orientated (16 poor status at all scales of policy-making (10  decisions must be taken either by major emitters at the national and international levels higher political urgency/legitimacy (although there are exceptions to this, e.g. proactive (20 adaptation with high costs and uncertain effect can have low urgency (12  involving primarily higher (national and international) scales of governance. However, actions necessarily involve decisions by individuals at the local level (12 reality is far more complex and demands … a perspective to appreciate that in most instances, adaptation and mitigation will interact with each other … wherever they are efforts focused mainly on local-scale efforts focused primarily on the relatively near-term implemented (22 (i.e., with immediate effectiveness). However, reality is far more complex and demands

17

Spatial scale: implementat ion and motivation

      

… a perspective to appreciate that in most instances, adaptation and mitigation will interact with each other … wherever they are implemented (22 administrative scale: generally local (27 local and regional priorities and interests, as well as global concerns (9 the common assumption that marginal adaptation is a local issue gone wrong (10 adaptation measures can occur at population, community, personal or productive-system (food, forestry and fisheries) level (11 often incentives are not needed, as the action is strictly required (although there are exceptions to this, e.g. anticipatory actions without immediate benefits may need incentives (12 adaptation is predominantly the responsibility of interest groups or local or regional governments (16 often falls to practitioners where local responsibility is clearer, although it often depends on support from national and global levels (19

 

   

Spatial scale: costs and benefits

     

      

Temporal scale: costs, implementat ion and effect

  

primarily local benefits (1 some options are beneficial in the absence of climate change (‘win-win’ and ‘no-regret’ options) (1 should benefit the victims (2 local costs, often relatively local benefits (3 adaptive measures are short-term investments with short-term solutions to the impact of climate change and natural variability (8 works on the scale of an impacted system, which is regional at best, but mostly local (although some adaptation might result in spill-overs across national boundaries, for example, by changing international commodity prices in agricultural or forest-product markets) (9 mainly benefits those who implement it (egoistic) (although there are exceptions to this, e.g. adaptation of temperate farmers may have global consequences) (12 from an economic perspective, the benefits of adaptation leads to greater local benefits (12 benefits are often more localized (14 adaptation necessitates locally appropriate actions that have context-specific benefits for the countries and regions that implement them (15 benefits: local (16 primarily a local issue, as adaptation mostly provides benefits at the local level (18 generally benefits only those that implement it (although there are exceptions to this, e.g., urban design projects with low energy requirements and low vulnerability can constitute both a mitigation and an adaptation measure (24 reactive adaptation can wait until more concrete evidence of climate impacts is available. Successful proactive adaptation has difficulties justifying itself, as the “baseline” impacts are unknown (1 benefits can more or less be appropriated by those bearing costs (1 costs whenever, benefits may come later or relatively soon after (3

            

    

local and regional priorities and interests, as well as global concerns (9 the common assumption that mitigation is a global issue, requiring global coordination, does not hold any more. With the ongoing failure of many rich nations to reduce emissions, action on mitigation has become increasingly diffuse, as communities, cities, states and companies pursue emissions reductions (10 usually incentives are needed (although there are exceptions to this, such as ‘no-regrets’ options (e.g., energy efficiency) (12 the effectiveness of mitigation policy is dependent on the participation of all countries at the international level. However, to stay below globally agreed targets, in practice GHGs have to be reduced at the local level where the activities take place (16 mitigation measures can take place varying from the individual to the international level (16 at the level of individual households, communities, and firms or of entire sectors and countries (23 global benefits, but varying across regions (1 some options have high local secondary benefits (e.g. reduce local air-pollutants). Some options may even be directly financially viable (1 should be done by the perpetrators (2 local costs, global benefits (3 global (benefits might be realized at the local/regional level) (9 global impact (11 mainly benefits others (altruistic) (although there are exceptions to this, such as cobenefits for short-term local air pollution, energy security, jobs (12 from an economic perspective, the benefits of mitigation are greater at the global scale and external to a local area (12 benefits are more global (14 benefits of mitigation are perceived to be greater at the global scale and external to a local area (15 benefits: mainly external to the region (16 has the capacity to benefit most (although there are exceptions to this, such as where urban design projects with low energy requirements and low vulnerability can constitute both a mitigation and an adaptation measure (24 benefits are generally considered to be global in nature, meaning that a reduction of emissions in one part of the world will benefit all continents (25

setting emissions targets has to be adjusted regularly to take into account new projections (1 benefits to later generations (1 costs now, benefits delayed. Lag in time: benefits to later generations (3 must act earlier despite greater uncertainty (3 mitigation strategies are short-term investments for long-term climate results (8

18

        

 Sectors

   

    

may act later after reducing uncertainty, although anticipatory adaptation may require earlier action (3 adaptation measures would be effective immediately and yield benefits by reducing vulnerability to climate variability. As climate change continues, the benefits of adaptation (i.e., avoided damage) will increase over time (9 the time span between the expenditures of and returns on adaptation is usually much shorter compared to mitigation (9 policies generally have a positive and direct immediate effect on the countries and regions that implement them (11 often the main driver is a short-term benefit due to reducing vulnerability to current climate (although there are exceptions to this, such as preparing for long-term impacts) (12 adaptation reduces vulnerability to immediate and near-term climate risks (15 also long-term non-climate benefits of adaptation (16 short-term effect on the reduction of vulnerability (18 efforts focused mainly on the relatively near-term (i.e., with immediate effectiveness). However, reality is far more complex and demands a long-term, life-cycle and systems perspective to appreciate that, in most instances, adaptation and mitigation will interact with each other for the duration of their implementation (22 the effects of adaptation are commonly observed immediately, reducing the vulnerability of those that are able to adapt (25 very heterogeneous, but with some stress on agriculture (1 adaptation within government will primarily be dealt with by spatial planners and different (non-energy) sectors of the economy (2 very heterogeneous, but more feasible for certain sectors and regions than others (3 adaptation is, by nature, sectoral. While this is true in many respects, several adaptation strategies cut across sectors and include efforts to broadly enhance adaptability, e.g., building up wealth or the ability to develop and use new technologies by, for instance, strengthening systems for education, research and development. Similarly, reductions in vulnerability to malaria would improve the nutritional status of the affected populations and increase their capacity to participate more fully in education or the labor force. This strategy, therefore, would also provide benefits across several sectors. Other strategies and measures that would have significant climate change-related spillover effects in other sectors include reducing hunger while providing co-benefits in terms of public health and reduced pressures on biodiversity (8 mainly urban planning, water, agriculture, forestry, health, coastal zones (although there are exceptions to this, such as renewable energy sources that are vulnerable) (12 may involve more dispersed stakeholders at the local level across sectors (14 a priority in the water and health sectors and in coastal or low-lying areas, and but also relevant in the agricultural and forestry sectors (18 often focusing on engineering, land-use planning and broader developmental approaches to reducing future risks of flooding, water scarcity, or other weather-related risks (21 relevant to a wide spectrum of sectors such as agriculture, tourism and recreation, human health, water management, coastal management, urban planning and nature conservation, to name just a few (25

       



        

benefits of mitigation carried out today will be evidenced in several decades because of the long time greenhouse gases remain in the atmosphere (ancillary benefits such as reduced air pollution are possible in the near term) (9 delay between incurring the costs of mitigation and realizing its benefits (9 long-term impact on GHG damage (11 long-term benefit from avoided climate change (although there are exceptions to this, such as co-benefits in the short term, e.g. in the form of reduced air pollution or, in some ‘noregrets’ cases, in the form of economic benefits and jobs (12 the effects of mitigation are only apparent over longer time scales (15 also short-term, non-climate benefits of mitigation, such as energy savings (16 long-term effect on climate change because of the inertia of the climatic system (18 tend to be viewed as long-term solutions. However, reality is far more complex and demands a long-term, life-cycle, and systems perspective to appreciate that, in most instances, adaptation and mitigation will interact with each other for the duration of their implementation (22 the climate benefits of mitigation activities often take time to observe due to the long time greenhouse gases remain in the atmosphere. Obviously some major co-benefits may accrue in the short-term, such as health benefits from reduced air pollutants (25

focus on emissions from fossil fuels (1 energy planning and the carbon intensity of economic growth, for example, are usually high in the priorities of industry sectors, government and of consumers who are interested in emergy security (2 focus on emissions from energy consumption (3 mainly energy, transport, building and industry (although there are exceptions to this, such as the mitigation options also available in water and land management) (12 energy, transport, industry, and the domestic sectors + renewable energy sources such as biofuels, hydropower, wind energy, and solar energy (12 may involve powerful industrial stakeholders from the energy sector concentrated at higher levels of decision-making (14 a priority in the energy, transportation, industry and waste management sectors, but the agricultural and forestry sectors are also relevant (18 usually emphasizes carbon efficiency in industrial processes, transport, housing, energy generation, etc., as well as, more recently, reforestation and forest conservation (21 has often been a popular strategy in sectors where energy consumption is dominant and/or where impacts on carbon sinks are considerable when looking at environmental impacts. These include, for example, energy, transportation, construction and manufacturing, forestry, waste management and increasingly agriculture. These sectors generally consist of well-organized actors that are used to taking longer term decisions on planning and investments. However, during the 21st century all GHG emitting activities have to be addressed in one way or another in all sectors, taking note that on the global level negative emissions may be a necessity in the long-term to stay within the 2°C limit (25

19

Nevertheless, even that emission-reduction activities have to take place locally (Moser 2012), the effectiveness of climate change is dependent on support at the international level and on the participation of all countries (Biesbroek et al. 2009; Tanner and Allouche 2011). Besides, there has been a shift away from the notion that adaptation is for developing countries and mitigation for developed countries, which has been generated by the increasing emissions in developing countries. Moreover, attention to adaptation has increased in the developed world, which is partly a consequence of the significant damage being caused by extreme weather events worldwide. Nevertheless, on the administrative scale the existing constituencies of adaptation and mitigation only marginally overlap (Tompkins and Adger 2003), as mainstreaming into the different agencies at large has failed, and relatively few public or corporate decision-makers have direct responsibility for both adaptation and mitigation. For example, adaptation is handled by some ministries (like agriculture, forestry, building and infrastructure), while mitigation is located in others (especially energy, trade and finance) (Tol 2005; Klein et al. 2007). Secondly, there is a general attitude regarding the temporal scale of adaptation and mitigation that scholars largely agree on (Table 1). Adaptation (benefits, costs and implementation) is immediate and near-term due to reducing vulnerability to current climate and climate variability, whereas mitigation benefits will affect later generations, although implementation and costs relate to the present. Thus, for mitigation there is a delay between incurring the costs and realizing the benefits, whereas adaptation provides benefits immediately.9 However, this perception has been challenged, for instance, by Moser (2012: 169), who states that “reality is far more complex and demands a long-term, life-cycle, and systems perspective to appreciate that in most instances, adaptation and mitigation—regardless of the level at which they are initiated—will interact with each other for the duration and wherever they are implemented”. Furthermore, she shows that adaptation and mitigation have a greater temporal (and spatial) overlap (Figure 2) than hitherto anticipated in the literature (e.g. Tol 2005). Examples could include countries that adopt a national policy to foster renewable energy, which at that local level will interface with local ecosystems and species that need to adapt to rapid climate change, or the interaction between biofuels and food production (Moser 2012). Biesbroek et al. (2009) comment that the temporal difference is exacerbated and that there is a growing awareness that proactive approaches to both mitigation and adaptation will prove more effective than reactive approaches to climate change. Others exceptions regarding the dominant focus are, for instance, that many mitigation actions would have short-term benefits such as reduced air pollution or, in some ‘no-regrets’ cases, economic benefits and jobs (Klein et al. 2007; Swart and Raes 2007; Biesbroek et al. 2009; Moser 2012; Illman et al. 2013). Alternatively, adaptation actions taken today may also have long-term benefits (Swart and Raes 2007; Biesbroek et al. 2009), as is the case in all large (infrastructural) adaptation measures with a lifetime of 9

This difference creates some difficulties in producing economic assessment of the cost and benefits of the two

approaches, especially the impact of using discount rates.

20

centuries or more (Moser 2012). Lastly, Biesbroek et al. (2009) add that the temporal difference over time will be highly influenced by technological development, innovation and scientific progress, but we do not know if it will increase or decrease this gap.

Figure 2. The temporal and spatial overlap between adaptation and mitigation. a) The common range of spatial and temporal scales of adaptation and mitigation options. If and when there is only a limited overlap, the universe of potential synergies and tradeoffs is relatively constrained. However, b) over the full lifecycle of a policy option, the overlaps are significant and the universe of potential synergies and tradeoffs vast (Moser 2012: 169).

Thirdly, there used to be a sharper division between adaptation and mitigation in the sectors, but the reality today is that both are relevant more or less in all sectors, which is why no exceptions have been emphasized in Table 1. Nevertheless, adaptation is mentioned as being most relevant with regard to the agricultural, forestry, tourism, urban, and coastal zones, as well as to nature conservation, health and water management (Swart and Raes 2007; Matocha et al. 2012; Locatelli et al. 2011; Illman et al. 2013). Hence, there is a wide spectrum of sectors which involve a very dispersed set of stakeholders across several scales (Wilbanks et al. 2007), and where spillover effects occur between sectors (Goklany 2007). Mitigation, on the other hand, is sectorally less complicated, as the focus is on the large emitters, which are energy, transport, industry, waste and housing (e.g. Dang et al. 2003; Tompkins and Adger 2003; Wilbanks et al. 2003, Wilbanks et al. 2007; Illman et al. 2013). Nevertheless, it went from almost being only an energy sector issue to be addressed in all emitting activities in all sectors, as it is judged to be necessary to stay within the 2°C limit (Illman et al. 2013). For instance, more recently there has been a growing interest in the water and land management (agricultural and forestry) sectors as well (Swart and Raes 2007; Locatelli et al. 2011; Matocha et al. 2012; Illman et al. 2013). And renewable energy sources such as biofuels, hydropower, wind energy and solar energy (Swart and Raes 2007) are naturally important substitutes for fossil-fueled energy sources. In most countries renewable energy sources are found in the energy sector, but they are also relevant as direct energy sources in all other sectors (e.g. solar-driven water pumping systems in agriculture).

21

Finally, mitigation and adaptation use very diverse units of measure, which makes it difficult to compare them. Mitigation is easily measured using internationally accepted methods (Biesbroek et al. 2009) and provides hard, quantitative metrics (Ayers and Huq 2009; Illman et al. 2013) that are expressed in CO2 equivalents for individual mitigation actions. These can be compared, as a common currency (Burton et al. 2007), with other mitigation activities, and if the costs of implementing the measures are known, their cost-effectiveness can be determined and compared (Murdiyarso et al. 2005; Klein et al. 2007). In contrast, adaptation has been said to be complex, messy and context-specific, without a standardized measurement methodology, without predefined targets, and difficult to quantify or express in a simple metric, thus impeding cost-benefit comparisons between adaptation efforts, or with mitigation (ibid.; Ayers and Huq 2009; Biesbroek et al. 2009; Illman et al. 2013). For instance, “adaptation options cover such a broad spectrum of costs and benefits that the definition of a common currency is far more ambiguous, if not impossible” (Burton et al. 2007). Furthermore, most of the adaptation measures that are applied are similar to broader sustainable development actions, which is why, for instance, the Adaptation Fund has struggled to come up with appropriate metrics for adaptation. Consequently, most funds accept some actions in the projects that are similar to development activities, without being explicitly designed to target climate change (Illman et al. 2013). This missing common metric is the main reason why no direct comparisons between adaptation and mitigation actions are made in the papers, but concentrate exclusively on the presence and modality of the actions and adjustments. 2.3.2

Concepts and frameworks

The overall idea for this study originates from reading Klein et al. (2007), who present a framework (Figure 3) that can be used to cluster examples according to the type of linkage and ordered according to the entry point and scale of decision-making. The framework distinguishes between four types of inter-relationship between adaptation and mitigation: 

A→M (adaptation affecting mitigation): adaptation actions that have consequences for mitigation. The affect is unintended and/or unnoticed. For instance, efficient community use of natural resources (positive affect) or natural resource management solely to sustain livelihoods (negative affect).



M→A (mitigation affecting adaptation): mitigation actions that have consequences for adaptation. The affect is unintended and/or unnoticed. For instance, land use projects that support local economies and livelihoods (positive affect), or afforestation projects that lead to depleted water resources and other ecosystem effects, with consequences for livelihoods (negative affect).



∫(A, M) (adaptation or mitigation): decisions that include trade-offs or synergies between adaptation and mitigation. Adaptation and mitigation are somehow connected. For instance, public-sector funding and budgetary processes that allocate funding to both adaptation and mitigation, where one is selected over the other.

22



A∩M (adaptation and mitigation): simultaneous consideration of adaptation and mitigation. For instance, water management with allocation of water between hydroelectricity and consumption.

The inter-relationships can be direct, involving the same resource base (e.g., land) or stakeholders, indirect (e.g., effects through public budget allocations) or remote (e.g., shifts in global trade flows and currency exchange rates) (Klein et al. 2007).

Figure 3. Typology of inter-relationships between climate change adaptation and mitigation (Klein et al. 2007:761).

The figure (3) summarizes the interlinkages identified and analyzed in the published literature, but it can also be used as a framework to investigate how examples at the policy level (e.g., portfolios, funding, strategies) are related to the inter-relationship between adaptation and mitigation (Klein et al. 2007: 760-3). In practice until today, the emphasis has largely been on the first two (i.e. M→A and A→M links) and attention to the last two (i.e. ∫(A,M) and A∩M) has been limited, despite their being necessary to progress along the synergy continuum (Duguma et al. 2014a). In the four papers in this dissertation, all four types of linkages are in focus, but the emphasis on the different kinds of interlinkages varies between the papers. In Paper 1 the focus is on M→A and A→M links, as few of the projects focus at A∩M because they are explicitly adaptation or mitigation projects. Furthermore, the ∫(A,M) issues could not be investigated in this paper, as the project documents refrain from describing any trade-offs and only point out a few synergies. Paper 2 focuses on the M→A link but also finds some trade-offs, and furthermore suggests some disregarded synergies (∫(A,M). Paper 3 focuses both on ∫(A,M) and A∩M issues, as the suggested adjustments are dedicated to benefiting both

23

adaptation and mitigation, while at the same time recognizing potential synergies and trade-offs. In Paper 4 there is attention on M→A by investigating how mitigation can affect adaptation. However, other frameworks that concentrate on the linking of adaptation and mitigation have emerged since the IPCC emphasized the importance of considering and establishing enhanced knowledge on the interlinkages and interrelationships between adaptation and mitigation (Klein et al. 2007). Biesbroek et al. (2009) advocate for a river basin approach as an administrative framework for integrating adaptation and mitigation in practice, where adaptation and mitigation measures should be examined from a systems perspective for the length of the measure’s life-cycle and its impacts on natural and human systems. Murdiyarso et al. (2005) draw up an analytical framework emphasizing three key parameters of adaptation and mitigation activities that should be identified to determine the degree to which there are potential synergies or trade-offs between mitigation and adaptation objectives: 1) who will be participating in the activities; 2) where the activities will take place; and 3) which activities will be implemented. This framework inspired the approaches applied in Paper 2 and 3. Others stress the link from adaptation and mitigation to development, as in the ‘climate compatible development’ framework (Figure 4), which is defined as “development that minimizes the harm caused by climate impacts, while maximizing the many human development opportunities presented by a low emissions, more resilient future” (Mitchell and Maxwell 2010). Common to these three frameworks is the limited attention given to the character of the different types of connection between adaptation and mitigation (and development), which will be elaborated on further below.

Figure 4. Climate-compatible development framework, which shows the overlaps between adaptation, mitigation and development (Mitchell and Maxwell 2010).

24

According to Klein et al. (2007: 753), “Adaptation and mitigation may be seen as substitutes in a policy framework at a highly aggregated, international scale: the more mitigation is undertaken, the less adaptation is necessary and vice versa”. However, in practice adaptation and mitigation are measures that occur separately, in complementarity, or in synergy with each other (Figure 5), which is why it is appealing to create multiplewin solutions with a single climate-policy option (ibid.). The separate measures are not in contact but run in parallel, whereas the complementary measures function in a symbiotic fashion whereby the outcome of one supplements or depends on the outcome of the other (ibid.). The synergistic version represents a further advanced interaction of adaptation and mitigation, whereby their combined effect is greater than the sum of their effects if implemented separately (ibid.). The word ‘synergy’ originates from synergia (Latin) or synergos (Greek), which translates into working together (Merriam Webster 2015), and has been defined as combined or co-operative effects that are literally produced by things that operate together (e.g. parts, elements or individuals). In the superadditive or classical conceptual model of synergy (von Eye et al. 1998), this is something that produces more than the things can produce alone (Corning 1998). This is why this model is the one to strive towards in implementing adaptation and mitigation measures, as the primary incentive behind the synergistic approach is to increase effectiveness, minimize costs and ensure continuity of production and/or service provision by minimizing risks (Kane and Shogren 2000; Ayers and Huq 2009; Illman et al. 2013; Duguma et al. 2014a). Nevertheless, two less prominent types of synergies are represented in the literature. The first is the subadditive model, where the aggregated outcome of the intervention is less than the sum of the individual intervention outcomes (ibid.), which can be justified, for instance in projects to avoid significant trade-offs (e.g. in Paper 1 when advocating suboptimal projects where there are room for both adaptation and mitigation in order to make the projects more viable). The second is the non-additive model, where interaction between the interventions is the primary focus, irrespective of their individual effects (von Eye et al.1998), for example, in chemical reactions (Duguma et al. 2014a), a version not covered in the papers.

Figure 5. Schematic illustration of the separate, complementarity, and synergy approaches in adaptation (ADA) and mitigation (MITI) measures. The size of the spheres is a relative indicator of the priorities for the measures with time (Duguma et al. 2014a: 425).

25

Swart and Raes (2007) adopted a less theoretical approach in defining synergies by simply distinguishing between 1) indirect synergy, which is a positive side-effect of an action, for example, climate policies that lead to reduced air pollution that lead to lower negative health impacts and consequently higher resistance to climate stresses. Comparable to M→A and A→M in Klein et al. (2007); and 2) direct synergy, or actions targeted to reduce greenhouse gas emissions and vulnerability simultaneously, which is comparable to A∩M in Klein et al. (ibid.). In this definition, the output sum is less important, as the focus is targeted at how intentional the synergistic effect of the action is. However, the indirect type is not a synergy (cf. Duguma et al. 2014a), but a complementarity approach, as it uses adaptation or mitigation as an entry measure (the major-minor notion), which then provides the other objective with a co-benefit. This is because, according to Duguma and colleagues, the goal in the synergistic approach is “Reducing impacts of climate change by addressing adaptation and mitigation within an integrated framework without prioritizing among the two and giving due attention to system integrity and functionality” (ibid.: 422, emphasis added). Furthermore, the whole is more important than the parts, as is the case in the complementarity approach (ibid.). Climate-smart agriculture (Paper 3) and agroforestry (Paper 4) are highlighted as good examples of synergies, whereas a forest plantation established for sequestering carbon and providing services like micro-climate amelioration and habitat for wildlife (Paper 2) is an example of complementarity (ibid.). There are also good examples of synergies in Paper 1, especially in the projects with a more holistic approach that are less focused on delivering specifically adaptation or mitigation benefits. However, the projects analyzed in this paper mainly have complementary actions, as the major-minor notion is pronounced. This leads to a consideration of project design, which in synergetic projects should take place through a multistakeholder process in order to ensure component integrity and system functionality (as it could be done in Paper 3), whereas the complementarity model often adopts a top-down approach mainly involving climate change professionals, donor agencies and target communities (Duguma et al. 2014a), as seen in the projects in Paper 2. Some other disadvantages in the complementarity model are that it is less cost effective in general (Klein et al. 2005) and that it creates competition between adaptation and mitigation for already limited resources (Tol 2005). The latter will make developing countries prioritize among different measures, which will inevitably make them deprioritize mitigation. However, concerns have been raised about the associated transaction costs and the risks of failure of overambitious and complex projects (Klein et al. 2005; Klein et al. 2007; Swart and Raes 2007), which raises the question of whether it is possible to capitalize the benefits in reality (Suckall et al. 2014). Accordingly, deliberately seeking synergies is not automatically an effective way to address the management of climate change risks (Illman et al. 2013). However, an advantage with the synergistic approach is the possibility to reduce or basically avoid trade-offs, which will potentially occur if interventions are implemented as parts and not as a whole (Duguma et al. 2014a). This conception, however, starts a discussion about the boundaries of the whole, which, particularly in land use-related projects, can be difficult to demarcate. Numerous studies have gone into the details of trade-offs in this sector providing

26

valuable insights (see Table 2 in Section 2.3.3), and scholars have also commented on potential trade-offs in other sectors (Klein et al. 2005; Tol 2005; Huq and Grubb 2007; Swart and Raes 2007; Moser 2012; Suckall et al. 2014). However, evidence-based case studies of trade-offs are rare (Tompkins et al. 2013), which Klein et al. (2007) partly explain by the likely circumstance that in reality trade-offs are insignificant in most cases. The above should not be interpreted to mean that complementarity is of no use, as it can be a stepping-stone to creating synergies as time progresses and scopes are brought closer (Figure 5). However, for synergy to happen there has to be resource relatedness (i.e. overlap in the resources utilized) between the actions to be synergized and resource complementarity. The latter refers to cases where actions that lead to an increase in one resource also increases the return to the other resource. Especially the land use sector has been pointed out as having substantial resource relatedness and resource complementarity between adaptation and mitigation measures, for instance, by addressing issues of food, energy and water supply with practices that are attractive to all land users (Klein et al. 2005; Duguma et al. 2014a), which has been shown possible (see Section 2.3.3). Hence, sectors matter when the aim is synergies, as well as convergence between the temporal and spatial perspectives of the activity. The greater the overlap (spatially, temporally and sectorally) the higher the potential for synergy and the need for attention to possible trade-offs. Duguma et al. (2014a) created a systematic approach to making the move from complementarity to synergy, which requires identification of the right approaches and concepts that enhance multifunctionality to ensure the provisions of simultaneous benefits. They also advocate the landscape approach (e.g. Scherr et al. 2012) to move towards synergy, as it places the emphasis particularly on multifunctionality and interactions among components. Figure 6 exemplifies this approach in a real landscape, where the linkages between adaptation and mitigation are a mixture of different types that diverge from the stylistic representation in Figure 5. Location #1 (Figure 6) indicates separation between adaptation and mitigation, where forest conservation has limited contact with the village (see more in Paper 2). However, the villagers still adapt – just not as a part of the project. Thus, adaptation and mitigation run in parallel with little or no effort to find complementarity or synergy between the two. Actually, trade-offs occur in both ways (e.g. resource exclusion by the projects and illegal extraction by the villagers). Location #2 is also taken from Paper 2 but indicates complementarity, where the projects deliver ecological adaptation benefits by protecting it, and adaptation activities, like fire management and thinning protecting the forest and thereby the carbon. This cannot be characterized as synergy (cf. Klein et al. 2007), at least not of the superadditive kind, as the superadditive type is a symbiotic function whereby the outcome of one supplements or depends on the outcome of the other. Synergies, however, are represented in locations #3a and #3b, which represent climate-smart agriculture (Paper 3) and agroforestry (Paper 4).

27

Figure 6. #1: separated measures, #2: complementarity, #3: synergy (Photo: Jalacte village, Belize, January 2014).

Finally, Jarvis et al. (2011) offer a framework for analyzing synergies and trade-offs between mitigation and adaptation. The latter is divided into short-term risk management and long-term progressive management. It explores three scenarios. The first is ‘transition (win-win-win)’, which is the best-case scenario in which risk management strategies contribute smoothly to progressive adaptation, all the while mitigating climate change without any real trade-offs (Figure 7: left). This is a win-win-win system, which, for instance, can be obtained by adopting some of the suggested adjustments in Paper 3 if payments for the amount of carbon sequestration provided are disbursed. This will decrease risk by offering immediate financial capital relief, increasing carbon storage and ensuring progressive adaption “by creating incentives and opportunities to diversify and further invest in agricultural and non-agricultural income sources” (ibid.: 195). The second is ‘disjointed adaptation (win-win)’, a system whereby risk management does not easily transit into transformational adaptation, but where there are synergies between each of these and mitigation (Figure 7: center). Mitigation actually acts as a bridge in the transitioning. It can, for instance, be in an alley-cropping system as in Paper 3, where climatetolerant legumes provide additional fodder (risk management), biomass sequesters carbon (mitigation) and the landscape is transformed into an improved natural resource base (progressive adaptation) (ibid.: 195). The third is ‘disjointed adaptation (no win-win)’, which is the worst-case scenario, in which there always are tradeoffs, no opportunities for win-win, and no smooth transition from risk management to progressive adaptation (Figure 7: right). This, for example, can be a case of small producers whose land will become unsuitable for agriculture in the future, and who have no clear long-term adaptation strategies. They will have to find an alternative, which in the most cases means migrating to deforest new land elsewhere, which has been a common pattern in southern Belize (Paper 3). In these cases, assistance is necessary to help and encourage the affected parties to adapt sustainable ways (ibid.). In Jarvis et al.’s framework, the temporal perspective is the

28

important factor, as it focuses on what will occur in the future by applying different strategies. Therefore it offers help to explore the potential overlaps between adaptation and mitigation, which is the first step towards producing synergetic measures and reducing trade-offs (cf. Duguma et al. 2014a). This framework is originally intended to be an analytical tool to be applied in the agricultural sector, but the approach can with advantage be used to assess other sectors as well.

Figure 7. From left to right: 1) transition (win-win-win), 2) disjointed adaptation (win-win), and 3) disjointed adaptation (no win-win) (Jarvis et al. 2011: Fig. 3-5).

The framework developed below (Figure 8) is an attempt to merge some of the features from the concepts and frameworks mentioned above. Time and scope develop towards the right. Hence, the ultimate aim is a constellation between adaptation (broadly defined) and mitigation, which is referred to as climate compatible development. In the beginning it is important to analyze the potentials and needs of the two main objectives, which is followed by a second analysis of the overlaps. In these steps it is essential to locate who is involved and who is affected, where activities are taking place or will occur in the future, and which activities have potentials or needs. When this is done, potential overlaps will appear which makes it possible to create activities that are complementary or synergetic (C and S in the figure). Nevertheless, the focus in the first step is to find complementarity and synergy between climate-related measures to ensure that these are incorporated and not overshadowed by the more development-related measures. Some will be separate measures (SM in the figure) on their way to the right, but later find complementary or synergetic effects and finally end at the ultimate aim. Others remain separate measures, as they are detached from the other objectives and have nothing in common whatsoever with them. However, some measures will remain as trade-offs (UT in the figure) throughout the process and will “survive” as trade-offs as they are assessed to be essential or impossible to close down. These last two – unavoidable trade-offs and separate measures – are expected to decrease over time, and can in rare cases be the preferable path. Nevertheless, the impact on development has to be considered carefully in these cases. Related to the papers of this dissertation, then, most projects in Papers 1 and 2 are managed as separate measures, but as highlighted in Paper 1 and suggested in Paper 2, many of these projects do have the potential to select one of the ‘preferable paths’ in the framework, though they often refrain from them due to the lack of incentives provided by the standards applied. Paper 3 tries to establish an agricultural system that creates complementarities and synergies, where the coupling to the broader adaptation can be emphasized later to end as ‘climate compatible development’. Last,

29

Paper 4 starts with mitigation as a separate measure, which has some complementarity with climate adaptation, for instance, less soil erosion compared to non-tree crops. Furthermore, if payments for the carbon sequestered by trees are disbursed, the system will be heading towards ‘climate compatible development’, as the payouts assist the broader adaptation needs.

Figure 8. A suggested framework to move from separate measures of adaptation and mitigation to climate compatible development.

2.3.3

Linking adaptation and mitigation in agriculture and forestry

As identified in Section 2.3.1, adaptation and mitigation are relevant for almost all sectors. But relevance is not the same as complementarity or synergy, because adaptation and mitigation can be separated spatially and/or temporally in a sector, or have trade-offs. However, it has been assessed to be highly relevant in the sectors of agriculture and forestry (see Table 1 in Section 2.3.1), which have been called the obvious candidates for the linking of adaptation and mitigation (Dang et al. 2003; Klein et al. 2007; Swart and Raes 2007; Illman et al. 2013; Duguma et al. 2014a; Suckall et al. 2015). This can be explained by the fact that adaptation and mitigation are highly relevant for almost all agricultural and forestry activities, and that in most cases these activities have implications for both adaptation and mitigation. Furthermore, there is the substantial resource relatedness and resource complementarity between adaptation and mitigation measures in the land-use sectors (Klein et al. 2007). Agriculture and forestry are inextricably linked to both adaptation and mitigation in climate change, as they are highly affected by climate change and, at the same time, contribute significantly to the world’s greenhouse

30

gas emissions. The IPCC estimates that agriculture, forestry and other land-use was responsible for 24% of global anthropogenic emissions of greenhouse gases in 2010 (Smith and Bustamante 2014). Around 10-12% came from agriculture (UNEP 2013; Tubiello et al. 2013) and 10-13% from changes in land use and land cover. The latter is mainly driven by deforestation associated with food production (Houghton et al. 2012), which is expected to increase with growing demands for food (Bajželj et al. 2014). More positively, the mitigation potential of these sectors is estimated to be substantial, and both are important players in closing the gap between the current emissions trend and the politically agreed 2ºC limit. Agriculture and forestry can contribute 32% (17-43%) of what is needed, which is more than other sectors: power (18%), industry (18%), transport (12%), buildings (13%), and waste (5%).10 These reductions are achievable through improved agricultural management practices (Smith et al. 2008; Smith and Olesen 2010) and appropriate forest management (Houghton et al. 2012). In the case of agriculture this could be achieved through climate-smart agriculture (FAO 2010a; World Bank 2011; FAO 2013), climate-smart landscapes (Scherr et al. 2012; ICRAF 2014), organic agriculture mitigating climate change (Kotschi and Müller-Sämann 2004; Jordan et al. 2009), conservation agriculture (Thomas et al. 2007; FAO 2008a; FAO 2008b), ecological intensification (Doré et al. 2011) and sustainable intensification (Garnett et al. 2013; Godfray and Garnett 2014). For forests this could include forest conservation, protected area management, afforestation and reforestation, bioenergy plantations, agroforestry, sustainable forest management and urban forestry (Metz et al. 2007; Ravindranath 2007). In relation to adaptation, the IPCC estimates that agricultural production and food security are already being affected by climatic change (Porter et al. 2014). Land-use systems globally are expected to change in response to future climate change, which will cause significant changes in livelihoods and landscapes (Steenwerth et al. 2014). The demand for food is increasing as a consequence of population growth, and at the same time crop yields are declining, which is putting extra pressure on the fertility of the land and will increase prices for the main staple foods (Hertel et al. 2010; Lobell et al. 2011; Wheeler and von Braun 2013). One meta-analysis of regional climate impacts on agriculture predicts an eight percent decline in overall crop productivity by 2050, which will affect food security significantly (Knox et al. 2012). Smallholders practicing rain-fed farming in tropical regions (which is the research area in all the papers in this dissertation) are particularly exposed to climatic change and low food security (Cline 2007; Jarvis et al. 2011; Knox et al. 2012; Rosenzweig et al. 2014). Furthermore, their vulnerability will increase when food prices rise because they typically buy more food than their agricultural systems produce. Consequently, the pace of climate adaptation measures in the agricultural sector needs to speed up (Porter et al. 2014). Adaptation in relation to forests can, according to Locatelli et al. (2011), be divided into ‘adaptation for forests’, which are the needs and actions that will maintain its functioning status, and ‘forests for people’s adaptation’, which is the role forests play in the adaptation of communities and the broader society. Concerning 10

Using a mean estimate. Figures used in making the estimates are from UNEP (2013: xv).

31

‘adaptation for forests’, studies have shown that measures are needed to adapt forests to future climates, since forest ecosystems can be vulnerable to climate variability and climate change, but studies have also revealed that forest projects rarely consider adaptation (Locatelli et al. 2008; Guariguata et al. 2008; Reyer et al. 2009; Keenan 2015). Regarding ‘forests for people’s adaptation’, studies have estimated that forests contribute significantly to rural livelihoods in many countries (Angelsen et al. 2014) and that forest products provide safety nets for local communities when agricultural crops fail, for instance, as a consequence of climatic events (Fisher et al. 2010). Thus, forests are central to the adaptive strategies of local communities and well-managed ecosystems can help local communities and the wider society to adapt both to current climate variations and to future climate change by providing a wide range of ecosystem services that reduce the vulnerability (Vignola et al. 2009; Turner et al. 2009; Pramova et al. 2012; Locatelli et al. 2015). For instance, hydrological ecosystem services imbedded in forest systems as base flow conservation, storm flow regulation, and erosion control are of the utmost importance in buffering the impacts of climate change on water users (Locatelli et al. 2011). Table 2 provides a variety of examples of synergies and trade-offs between adaptation and mitigation in the agricultural and/or forestry sectors, which all originate from the literature. However, the examples greatly depend on their context, design and implementation, with site selection and management practices being especially essential. What is intended to be a synergy can turn out to have only neutral or even negative impacts on the other objective. Thus, actions have to be tailored to the specific conditions (Fleurbaey and Kartha 2014), as few if any outcomes are universal (Knowler and Bradshaw 2007). For instance, it is very difficult to assess how susceptible community livelihood conditions are to a forest carbon sequestration project if extraction happens informally (or even illegally), thereby affecting carbon sinks and hence mitigation efforts (Duguma et al. 2014a). The examples in the table indicate that the forestry sector delivers more synergies and has more trade-offs when compared to agriculture. It could be that this disparity derives from the fact that forest areas contain significantly more carbon, but at the same time they also compete with alternative land-uses (mainly agriculture) that have significant higher opportunity costs (Börner and Wunder 2012). Table 2. Examples of synergies and trade-offs between adaptation and mitigation in the agricultural and/or forestry sectors.

Synergies / linking between adaptation and mitigation (examples)

Agricultural sector  carbon sequestration in agricultural soils can create an economic commodity for farmers (sequestered carbon) and make the land more valuable by improving soil and water conservation, thus enhancing both the economic and environmental components of adaptive capacity (Klein et al. 2007)  fallow systems transformed to continuously cultivated areas (to maximize production under

Forestry sector  several forest adaptation management strategies, like fire management, using drought-resistant varieties in planted forests, and the promotion of native species, will increase carbon sequestration beyond adaptation (Guariguata et al. 2008)  adaptation projects that reduce the vulnerability of communities in forested areas or in the surrounding region have the potential to avoid deforestation and forest degradation (Locatelli et al. 2011)  integrating adaptation into mitigation projects may increase their resilience to climate variations, the permanence of carbon storage, and their acceptance by local communities, as adaptation responds to local issues (Locatelli et al. 2011)  using drought-resistant varieties of tree species in planted forests to improve tree species’ resilience to water stress while increasing carbon sequestration (Locatelli et al. 2011)

32











 

 





heavier precipitation conditions)  mitigation projects have the potential to facilitate the adaptation of can increase the ability of soils to forests to climate change by reducing anthropogenic pressures on sequester carbon (Rosenzweig forests, enhancing connectivity between forest areas and and Tubiello 2007) conserving biodiversity hotspots (Locatelli et al. 2011) synergies could include practices  mitigation projects can protect the ecosystem services that are to improve soil water-holding relevant for people‘s adaptation, such as water regulating services capacities by adding crop or the provision of forest products used as safety nets (Locatelli et residues and manure to arable al. 2011) soils, which not only affects soil  incentives for forest conservation and increasing forest cover properties and nutrient cycling, would not only avoid greenhouse-gas emissions but would also but also lowers emissions (Smith result in benefits for local climate, water resources and biodiversity and Olesen 2010). (Klein et al. 2007) soil carbon sequestration has  forestry mitigation projects (e.g. forest conservation, afforestation synergies with adaptation, as this and reforestation, biomass energy plantations, agroforestry, urban can result in better plant nutrient forestry) can reduce water evaporation and vulnerability to heat content and increased water stress (Swart and Raes 2007) retention capacity, leading to  mangrove plantations that protect coastal areas from storms and higher yields and greater simultaneously sequester carbon (Ayers and Huq 2008; Locatelli resilience (FAO 2009a; FAO et al. 2011) 2009b)  coastal afforestation which stabilizes shorelines and provides by increasing the amount of soil protection against storms and other extreme events while carbon, fertility and water simultaneously improving carbon sequestration (Ayers and Huq retention are improved, thus 2009) leading to improved yields.  afforestation and reforestation that increase carbon sinks and Estimates show that for every 1 simultaneously reduce water evaporation and lower vulnerability ton of carbon that is sequestered to heat stress (Klein et al. 2007) into degraded soil, maize yield  reforestation prevent flooding and erosion while sequestering increases by 10–20 kg/ha carbon (Swart and Raes 2007) (Suckall et al. 2015)  early warning systems and forest fire management can modeling suggests that investing prevent/limit emissions from fires, but also help adapt to increasing substantially in adapting climate variability and extreme events causing forest fires such as agriculture to climate change in droughts or storms (Swart and Raes 2007) some regions — Asia and North  ecosystem management can contribute to a reduction in America — can result in greenhouse gas emissions through improved land-use and reduced substantial mitigation co-benefits, deforestation on the regional scale. Further, conserving and while the latter may be restoring naturally functioning ecosystems, which is often one of insignificant in Africa (Lobell et the goals of ecosystem management, can significantly contribute al. 2013) to buffering ecological responses to climate extremes such as Conservation agriculture that droughts and wildfires. Moreover, ecosystem management helps increases soil carbon, and build capacity for learning and adaptation at multiple scales. As a simultaneously generates higher result, societies will be better prepared to respond to surprises and yield and reduces labor uncertainties associated with climate change (Mori et al. 2013) requirements (Suckall et al. 2015) positive synergies between agricultural production, on-farm biodiversity and carbon storage in agroforestry (Henry et al. 2009) agroforestry provides a particular example of a set of innovative practices that are designed to enhance productivity in a way that often contributes to climate change mitigation through enhanced carbon sequestration, and that can strengthen the system’s ability to cope with adverse impacts of changing climate conditions (Verchot et al. 2007) shade use and reforestation in coffee can influence both climate vulnerability and carbon stocks (Rahn et al. 2013) agroforestry in coffee production zones and upstream reforestation is expected to reduce the impacts of climate change on coffee production by improving water regulation and soil fertility and reducing landslides and erosion and provide mitigation benefits by enhancing carbon stocks in the landscape (Locatelli et al. 2011) agroforestry contributes to multifunctional landscapes that support mitigation and adaptation and can lead to improvements in livelihoods, whereby provision of fuel wood, timber, fruits and/or fodder is often associated with the co-benefits of improved soil fertility, water infiltration and below- and above-ground carbon sequestration (Jerneck and Olsson 2013; Mbow et al. 2014) agroforestry and ecosystem conservation are key approaches in the integration of climate change adaptation and mitigation objectives, often generating significant co-benefits for local ecosystems and biodiversity. Synergies between climate change adaptation and mitigation actions are particularly likely in projects involving income diversification with tree and forest products, reduction of the susceptibility of land-use systems to extreme weather events, improvement of soil fertility, fire management, windbreaks, and the

33

Trade-offs between adaptation and mitigation (examples)

  











2.4

conservation and restoration of forest and riparian corridors, wetlands and mangroves (Matocha et al. 2012) see also Roshetko et al. 2007 and Schoeneberger et al. 2012 particularly in the distribution of  the types of trees best suited to preventing flooding and protecting scarce resources among rival biodiversity may not be the most effective to sequestrate carbon activities (Steenwerth et al. 2014) (Dang et al., 2003) increasing fertilization and  competition for land by mitigation projects would increase land irrigation requires more energy, rents, and thus commodity prices (Klein et al. 2007) and unless this can be produced  in arid and semi-arid regions, afforestation strongly reduces water from non-fossil sources, yields. This has direct and wide-ranging negative implications for emissions will increase (Klein et adaptation options in several sectors such as agriculture al. 2007) (irrigation), power generation (cooling towers) and ecosystem unavoidable clashes between protection (minimum flow to sustain ecosystems in rivers, water supply for agriculture and wetlands and on river banks) (Klein et al. 2007) other water utilizing sectors (e.g.  bioenergy crops might affect food prices negatively (Klein et al. hydropower) (Klein et al. 2007) 2007) more sustainable practices to  large-scale afforestation and reforestation aiming at carbon sequestrate more carbon can sequestration could reduce runoff and water available off-site affect yields negatively (Zomer et al. 2006) (Steenwerth et al. 2014)  carbon sequestration competing with ecosystem services relevant organic practices to increase soil for adaptation (Locatelli et al. 2011) organic carbon can lead to  conserving forests and cultivating bioenergy crops on a large scale competition for feed for livestock produce conflicts with respect to food supply, food prices or fuel, or even to decreases in (especially in the tropics) and water resource management (Popp average yields, thereby et al. 2012) exacerbating forest conversion to  forest projects can limit local communities’ access to forest agricultural land (Smith et al. resources (Matocha et al. 2012) 2013) fast-growing tree monocultures for mitigation conflict with local tree and forest uses, making livelihoods more vulnerable when trees are planted in water-scarce areas conflicting with local water uses, and in some cases when “climate-smart” agroforestry practices conflict with the need for agricultural intensification to produce increasing amounts of food for a growing population (Matocha et al. 2012) additional trees in shaded coffee systems to enhance carbon can reduce yields and increase pressures from disease (Matocha et al. 2012)

Research gaps

The adaptation-mitigation dichotomy described in Section 2.1 is also reflected in research, as demand-driven research should be policy-relevant, as a consequence of which it has followed the language and the structure of the UNFCCC’s Framework Convention (Burton et al. 2007; Biesbroek et al. 2009). Studies rarely analyze the consequences of their primary domains for the other objective, for example, the impacts of adaptation options on greenhouse-gas emissions. Also, the adaptation and mitigation science communities, which principally are separate and unconnected, tend to use different methods (Klein et al. 2007). The mitigation research community has mainly used a ‘top-down’ approach with a strong focus on technological and economic issues, while the adaptation research community has been more interdisciplinary by nature, tending to conduct ‘bottom-up’ studies with the emphasis on local and place-based analysis (ibid.; Biesbroek et al. 2009). The latter has shared its “research approach with the development studies and disaster risk-reduction communities that minimizes immediate- and short-term impacts of climate trends and shocks in the most vulnerable, primarily developing countries” (Ayers and Huq 2009). These facts have influenced how the two are perceived by policy-makers and the wider public, where adaptation science is assumed to be a soft, uncertain science compared to mitigation science, which has a hard, indisputable reputation (Harry and Morad

34

2013). These differences have created difficulties in finding common ground, which has affected the focus on linking of adaptation and mitigation in the world of science (Biesbroek et al. 2009). The scientific literature on linking has been dominated by theoretical analysis and modeling of the possible synergies and trade-offs on the global scale. To start with, it was modeling studies of the optimal mix between adaptation and mitigation that was interesting (e.g. Kane and Shogren 2000), as the two objectives were seen as substitutes, rather than complementarities or synergies (Dang et al. 2003; Klein et al. 2005; Biesbroek et al. 2009; Moser 2012). However, the optimal mix models proved to be a misguided assumption, as it is impossible to determine ‘the’ optimal mix of mitigation and adaptation options: whatever mix is optimal will depend on local conditions, values, preferences, uncertainties, etc. (Klein et al. 2005). Later the focus shifted to integrated assessment modeling, which provides approximate estimates of the relative costs and benefits (mainly of mitigation) at highly aggregated levels, whereas only a few models include feedbacks from the impacts of climate change (Klein et al. 2007). However, these approaches were met with some criticism, for instance, that many of the adaptation aspects are “inherently difficult to capture in global scale simulations”, and that the studies “fail to account for adaptation limits and the context-specific nature of adaptation”, “often assume that societies are optimally adapted to their climates, whereas in fact many adaptation deficits currently exist”, “tend to underestimate damage costs and adaptation costs, and to overestimate the potential for adaptation”, and “[are] detached from the reality of decision-makers” (Warren et al. 2012). Nevertheless, it was commented that the growing awareness, that society plays a far greater role than previously assumed, has led to “a gradual shift from the fundamental and mono-disciplinary research domain towards a more transdisciplinary research strategy” (Biesbroek et al. 2009). This was an essential change in science in finding socially accountable solutions, which is assessed to be located in the interaction between the theoretical and applied science communities (Klein et al. 2007; Sathaye et al. 2007; Swart and Raes 2007). However, a drawback of this process proved to be the vast amount of research being conducted on climate change, which has often increased the demand for more knowledge, as more remains unexplained for scientists and politicians. This has created a failure to act upon existing knowledge or a ‘mitigation–adaptation paradox’ (Biesbroek et al. 2009). Thus, discussion of the linkages between adaptation and mitigation is just only getting started, and studies are increasingly beginning to view the two as complementary to each other (Illman et al. 2013), even though all too often the debate is conducted in a very generalized manner. Nevertheless, this new paradigm in the linking studies, which truly gained momentum in 2007, was chiefly fueled by the IPCC, who stated that, despite still having a small literature base, linking was a new research field that was growing fast (Klein et al. 2007). Other publications that had an impact on establishing this research area included two special journal issues, particularly Mitigation and Adaptation Strategies for Global Change (Issue 5, 2007) entitled ‘Challenges in Integrating Mitigation and Adaptation as Responses to Climate Change’ (e.g. Goklany 2007; Huq and Grubb 2007; Ravindranath 2007; Halsnæs and Verhagen 2007; Wilbanks and Sathaye 2007), and to a lesser degree Climate Policy (Issue 4, 2007) entitled ‘Integrating climate change actions into local development’ (e.g. Burton

35

et al. 2007; Swart and Raes 2007). The latter focused more on linking climate change and sustainable development than on linking adaptation and mitigation, as in the former. Thus, the research on linking strategies was being broadly recognized in 2007, but it had taken a long time, as calls for this to happen can be traced all the way back to 1991 (Goklany 2007). Since 2007, the amount of research and the resulting evidence base has still remained rather scattered (Illman et al. 2013; Suckall et al. 2015), and as a field of research it is very poorly developed, particularly concerning material at the regional and sectoral levels (Tompkins et al. 2013). The main lack today is empirical examples of local, on-the-ground case studies that can improve our understanding of the synergies and trade-offs between adaptation and mitigation (Burton et al. 2007; Klein et al. 2007; Locatelli et al. 2011; Moser 2012; Illman et al. 2013; Duguma et al. 2014a). However, it is not studies narrowly focusing on climate change that are required so much as the current linking paradigm’s search for ‘triple wins’, which is why development issues have to be taken into consideration as well (Tompkins et al. 2013; Suckall et al. 2015). Adaptation and mitigation are too commonly examined separately, as the historical dichotomy between adaptation and mitigation still persists, and if they are studied together, is it in the absence of development issues (Suckall et al. 2015). Thus, there is a need to move beyond the single-impact focus if studies are to provide a better understanding of the possibilities of mainstreaming multiple ‘wins’ and to avoid trade-offs in projects and policies (Illman et al. 2013; Suckall et al. 2015). However, the potential for and feasibility of multiple ‘wins’ remain rather uncertain (Tompkins et al. 2013), as almost no evidence exists with which to assess them (Suckall et al. 2015). Also, little is known about the enabling conditions for design and implementation to move from the current dichotomized approach to the synergy approach (Duguma et al. 2014a). Furthermore, the current analytical frameworks for evaluating the links between adaptation and mitigation are inadequate, which is why it is difficult to take advantage of synergies or to explicitly evaluate trade-offs (Klein et al. 2007). However, a strong evidence base on linking can ensure that funding is invested effectively, for instance, in ‘no-regret’ options that deliver multiple cobenefits for adaptation, mitigation, and development. The lack of an evidence base for practices that will achieve triple wins will significantly limit the capacity of donors to identify, monitor, and evaluate investments (Tompkins et al. 2013; Suckall et al. 2015). Furthermore, studies of linking may “save money by making more efficient use of scarce resources, e.g. not building separate institutions and processes to support adaptation and mitigation and avoiding conflicting policies” (Suckall et al. 2015: 38). Finally, the need for empirical case studies (as in Papers 2-4) mentioned above can function as showcasing in tackling “any concerns related to potential inefficiencies and to address other remaining challenges in for example project planning” (Illman et al. 2013: 4). For instance, comparative studies of the impacts of carbon projects on local communities and their adaptive capacity are required, as they “are needed to grasp the necessary lessons and develop best practices for mainstreaming adaptation and mitigation” (Locatelli et al.

36

2011: 444), as explored in Paper 2. The identification of barriers to implementing adaptation and mitigation like behavioral change are some other prominent examples of the research needed in this area (Jarvis et al. 2011; Moser 2012), as revealed in Paper 3. Additionally, the need has been raised for potential synergies and overlaps between existing portfolios, as this could help to adjust funding guidelines to allow for synergies and avoid trade-offs (Illman et al. 2013), which is the main idea in Paper 1. These are some of the gaps in knowledge addressed in the papers in this dissertation, but more specific research gaps are laid out in the individual papers.

37

3

RESEARCH DESIGN

This section presents the applied research design and the methods adopted in collecting and handling data to achieve the research objectives. The overall research approach was influenced and determined by my academic background (geography), the conceptual field and research gaps found in the literature (Section 2), and the geographical setting of the research (the case areas). And, selecting the research design, including the steps in data collection and data processing, was guided by the research questions (Section 1), while the data-processing phase especially was appropriate to the questions being addressed (cf. Creswell 2009:3). The researcher must ask which methods are the most applicable with regard to the research questions, as the former are closely related to successfully answering the latter. An incorrect method might give answers, but maybe not the correct ones or the type wanted. Nevertheless, it is also important to understand that it is a dynamic process that often develops during the research. Issues emerge, together with reflections and new considerations, which is why decisions have to be taken during the process to redirect and shape the methodology being applied. There are several reasons for selecting the research designs applied in the papers, and justification of each design is given in the sections. The aim of the sections is not to repeat what has already been written in the papers, but rather to add background, reflections and additional information not found there. It is tried to reduce any overlap with the paper versions to a minimum. However, each description should be able to stand on its own. Table 3 provides a methodological overview of the methods and techniques that have been adopted in the papers. Table 3. Overview of the applied research designs Paper 1

Paper 2

Paper 3

Paper 4

Case study: imbedded (multiple-units of analysis)

Case study: imbedded (single-unit of analysis)

Case study: holistic (multiple-units of analysis)

Research tools: data collection

Mixed method (mainly sequential exploratory). Systematic review of systematic selected project documents Analytical framework developed in Excel

Adjustments developed and used as interview guide. Maps, aerial photos, dictaphone, notebook and pen. Interview schedules

Measurement guidelines, local forester, and measuring tape

Data collection

Project databases

Descriptive data translated to quantitative data using codes Statistical analysis of variance (ANOVA), followed by qualitative examples

Semi-structured interviews, field visits (observation) Qualitative

Field measurements

Data

Analytical framework developed and used as interview guide. Maps, aerial photos, dictaphone, notebook and pen. Interview schedules Semi-structured interviews, field visits (observation) Qualitative

Coding and categorization using NVivo, and assign meaning

Coding and categorization using NVivo, and assign meaning

Calculations and literature analysis

Research design

Data analysis

Quantitative

38

In the following sections, there will elaborated and reflected on the methodological choices made for this dissertation, including presentation of the main fieldwork area. The four papers are grouped into two main sections, as the approach in Paper 1 differs from those used in the case studies (Papers 2-4), which is why this separation was believed to make most sense for the reader. Paper 1, which is a desk study that applied a mixed method, is elaborated on in Section 3.2, where especial emphasis is placed on issues of validity and reliability (Section 3.2.1). Papers 2-4 use a case study approach, which is why this research design is discussed (Section 3.3) before the different issues of the Belizean case studies (Paper 2 and 3) are handled (Sections 3.3.1). Section 3.4 describes the research area and reflect on the fieldwork methodology applied in the Belizean papers. Particular emphasis is placed on the semi-structured interviews that are the core of the Belizean case studies with respect to data collection and analysis. Besides these scientific steps, practicality issues related to the selection of Belize and to conducting fieldwork are included. However, first of all the different levels and units relating to the papers are discussed.

3.1

Levels/units and the temporal scale of analysis

‘Level of analysis’ is a geographical term used to identify the location, size, or scale of a research target, whereas the ‘unit of analysis’ is the major entity (e.g. individuals, groups, villages) being analyzed. The latter is also defined as the case, which relates to how the research questions are defined, and which guides the research design and data collection strategy (Yin 2009; Creswell 2014). The papers address different levels of analysis from the meso-level (i.e. village and district) to the macro-level (i.e. country and international), and different units of analysis from projects to crop varieties (Table 4). The levels/units applied were found to be the most appropriate in targeting the research questions. Table 4. Levels/units of analysis and time perspective applied in the papers Level of analysis Unit of analysis

Temporal scale of analysis

Paper 1 Africa, Asia and Latin America (project locations and surroundings) Project descriptions in major portfolios of climate change that include forestry and/or agricultural activities Subunit: portfolio standards Project lifetime

Paper 2 Belize (project locations and surroundings)

Paper 3 Toledo District, Belize

Paper 4 Agricultural Research Center of Kade, Ghana

Certified mitigation projects. Subunit: buffering villages

Maize cultivation practices Maya villages

Tree crop plantation varieties

Project lifetime (and retrospective)

Snapshot/contemporary (and retrospective and future outlook)

Snapshot/contemporary

In Paper 1, project descriptions in major portfolios of climate change projects in Africa, Asia, and Latin America that included forestry and/or agricultural activities were the unit of analysis. Projects without land management activities (e.g., animal waste management or biogas production) and national projects in capacitybuilding or institutional strengthening without agricultural or forestry activities on the ground were excluded.

39

In Paper 2, mitigation projects in Belize were used as the primary unit of analysis, while buffering villages appear as subunits. The latter should intentionally have played a greater part – see the explanation about the excluded survey in Section 3.4.8. In Paper 3, an ethnic group in a particularly interesting geographical district in Belize was used. It could be argued that the more traditional unit of analysis choice (i.e. the household) could have been chosen, but it was refrained from doing so since several scholars have already conducted extensive household-level research in the district, suggesting that such an approach would not have produced any new insights. In Paper 4, tree crop plantation varieties cultivated in comparable conditions were used as the unit of analysis. Choices made with regard to the time perspective (the temporal scale of analysis) reflect what the data collected with the selected unit of analysis allowed any statements about. In Paper 1, we focused on activities and their outcomes in the lifetime of the project, as the project descriptions provided information exclusively about that timespan. In Paper 2, we concentrated on the project lifetime’s activities and outcomes, as described in the project documents, but we also included retrospective information from interviewees and literature to place the projects and findings in their historical context. Paper 3 is a snapshot of the contemporary situation for making the suggested adjustments in maize cultivation in the Maya villages, and again retrospective information from interviewees and the literature was used to add a historical dimension to the findings and the supposed barriers. In addition, because Paper 3 deals with adapting to the future, including expected climatic changes, it to some degree incorporate a future outlook. Paper 4 provides a snapshot of the current carbon content, but this can claim to be retrospective as well, because we have snapshots from different ages for two of the crops, making it possible to interpolate carbon content backwards.

The latter, Paper 4, was actually the motivating force behind the overall research topic in this dissertation, but it will only be mentioned very briefly here, as method applied is fully explained in the paper. We wanted to investigate the possibility of linking a mitigation activity to adaptation. The paper touches on how tree crop plantations in the tropics, more specifically in Ghana, can take the form of both adaptation and mitigation strategies. Data collection for this paper was conducted as a part of the fieldwork for my master’s thesis. Nevertheless, the paper was written at the beginning of this Ph.D., which is why it features in this dissertation. Three weeks of fieldwork were conducted in January 2011, of which two weeks were used collecting tree measurements at the Agricultural Research Centre in Kade, located 123 km NW of Accra. The last week was used in Accra to conduct qualitative interviews and collect secondary data. The study site was selected because the research station had the tree crop plantations we were searching for. The case is expected to be representative, and lessons learned are assumed to be informative about experiences with the average case (Yin 2009: 48).

40

3.2

The global study (Paper 1)

The research questions for Paper 1 required both quantitative and qualitative answers, which is why we chose a mixed methods approach. Recent literature (Tashakkori and Teddlie 2010; Creswell and Plano Clack 2011) tends to use the term mixed methods, but previous writings have also featured terms like ‘integrating’, ‘synthesis’, or ‘mixed methodology’ (Creswell 2014: 217). Mixed methods research employs a combination of quantitative and qualitative approaches and utilizes the strengths of both to address complex issues where the use of either quantitative or qualitative approaches by themselves is inadequate. The main idea behind the mixed methods approach is to gain insights from the combination of the two approaches in order to provide an expanded understanding of the research issue. This involves the collection of both qualitative (open-ended) and quantitative (closed-ended) data and includes analysis of both (Bryman 2008; Yin 2009; Creswell and Plano Clack 2011; Creswell 2014). We chose to apply a mixed methods approach in Paper 1 because of its strengths in drawing on both quantitative and qualitative data and minimizing the limitations of both approaches. In doing this, we achieved both breadth by having a large sample, and depth by including specific qualitative findings. Furthermore, this method provides a sophisticated, complex approach to research procedures at the practical level and proved a useful strategy to obtain a more complete understanding of the research question at the procedural level (Creswell 2014: 218). For instance, we explained the quantitative results in the paper with reference to qualitative data from concrete examples from particular projects or general trends in how mitigation projects contribute to a particular aspect of adaptation and vice-versa. The challenge with this method was the extensive data collection, the time-intensive nature of analyzing both quantitative and qualitative data, and the requirement that the researcher be familiar with both forms of research (ibid.). In addition, it was a challenge to explain the method in the paper in brief and at the same time make the flow of research activities in the design understandable for the reader. Our approach followed a multi-step sequential procedure (Creswell 2014), which is illustrated in Figure 9. Qualitative information from the literature feeds into creating the analytical instrument (i.e. the adaptation and mitigation aspects) to be applied in exploring the projects. Thus, we used an exploratory approach (cf. ibid.) in the initial phase. The instrument was used to contain qualitative quotes from the project documents that were transformed concurrently into a quantitative dataset (i.e. connected approach, cf. ibid.), which is referred to as a semi-quantitative approach, as the data build on each other. This, the quantitative part of the database, was analyzed initially to search for statistically significant tendencies. Next, the statistical tendencies were exemplified with qualitative data, and general observations on the qualitative data were used to explain the results. However, in some cases we returned to the original project documents to obtain additional information. In this way, the qualitative dataset and background material assisted in interpreting the statistical results by elaborating on some of the quantitative results. Thus, we used an explanatory approach (cf. ibid.) in this phase.

41

Overall, the approach provided a deeper understanding of the issues than if we had solely applied a quantitative or qualitative approach. The emphasis in the study is on the statistical analysis of the quantitative data in step 4 (Figure 9). We test whether linking between adaptation and mitigation is pursued in practice. We did this by testing whether mitigation and adaptation projects contributed explicitly to each other’s objective and whether this contribution was substantiated. Our independent variable is project type (adaptation or mitigation), followed by four intervening variables: 1) portfolio (10 different funds/standards), 2) sector (agriculture, forestry, or mixed), 3) geographical location (region and sub-region), and 4) length of document (contiguous number). The statistical analysis is to test whether these variables have a significant influence on the dependent variables (i.e. the outcomes) because it was assumed that these variables would affect the outcomes of the projects. The test, analyses of variance (abbreviated ANOVA), assessed whether a pattern occurs by chance by comparing groups with regard to an outcome. See Paper 1 for additional details of the method used (e.g. data collection and analysis methods).

Creating the analytical instrument

Data collection: developing database

• Data: Literature review (qual)

• Data: Project documents (qual)

Transformation of dataset • Data: database (qual → quan by using codes)

Data analysis

Explaining results

Discussion

• Data: database (QUAN part)

• Data: database (qual part)

• Data: mix of QUAN and qual findings

Figure 9. Process diagram of the sequential procedure applied in Paper 1. “quan” and “qual ” are quantitative and qualitative, respectively. Capitalization indicates priority.

3.2.1

Validity and reliability (Paper 1)

There are certainly important caveats with the applied method, which may have affected the validity of the results. One issue is that the analytical instrument has been developed solely for this assignment and has not been tested before. Thus, we cannot be sure that it can find the intended content in the project descriptions. When an analytical framework is created, we are limited by our knowledge and our perception of the world we are investigating (Grimen and Ingstad 2007: 285). In this study, we had selected what we were searching for, which may have created a bias in the results. Furthermore, since it would be impossible to find evaluations of that many projects, we had to use project documents with aims which we cannot be sure will be realized, or that merely represent lip service, for example, to achieve certification or political will. A majority of projects

42

use the most likely outcome of a specific activity, but since the outcome of many activities is case-specific, it is uncertain what the outcome will be in the end. Since the main aim of this study was to search for links between adaptation and mitigation, we were not interested in the scale of the project or activities, and therefore the magnitude of the activities included in the aspects was not considered. For instance, the degree of reforestation of a particular area makes a difference in terms of its contribution to mitigation, but not when searching for existing or potential ways to link adaptation and mitigation. Besides, projects that contribute to many aspects of either mitigation or adaptation are not necessarily superior in achieving integration, as a project may contribute to only one aspect (e.g. reducing emissions from deforestation) and thus be an exemplary linking project. Further, negative impacts were not considered, as only a few projects mentioned them, and the majority of projects that described negative impacts mitigated them in the implementation phase. Thus, when negative impacts are not described it is difficult to find trade-offs between adaptation and mitigation in the implementation of activities, and we refrained from doing this, as it would have led to unsubstantiated conjecture. However, it should be noted that most activities that do not have a positive effect on the other objective are expected to have a neutral effect (Klein et al. 2007). To achieve reliability, that is, the consistency of a particular approach across different researchers and different projects (cf. Creswell 2014: 246) of the database, a pilot of the data collection phase was made. This can be regarded as an intercoder agreement, where two or more researchers reach a consensus on the codes used for the same passage in the text to archive a high level of consistency in coding (Creswell 2009: 191). The Achilles heel in this study was especially the subjective judgment in converting the qualitative text in the project documents into a quantitative score. Thus, we picked four projects (one adaptation in agriculture, one adaptation in forest, one mitigation in forest, and one mitigation in agriculture) to be analyzed by two of the authors independently. Results were compared, and disagreements were discussed in detail. Most of the disagreements were minor, and no major disagreements were found. Thus, the analytical tool was refined to avoid overlapping and misinterpretation. Another measure to ensure reliability was to let only one person collect the majority of the data from the project documents and have one of the co-authors make spot checks. On several occasions individual projects were discussed intensively, and in the beginning we had to reanalyze projects because some aspects in the analytical tool were adjusted.

3.3

A case study research design (Paper 2, 3 and 4)

“…a scientific discipline without a large number of thoroughly executed case studies is a discipline without systematic production of exemplars, and a discipline without exemplars is an ineffective one.” T.S. Kuhn cited in Bent Flyvbjerg’s “Five Misunderstandings About Case Study Research” (Flyvbjerg 2006)

43

With that quote and the findings in Section 2 in mind, and because we wanted to explore a contemporary phenomenon within its real-life context using multiple sources of evidence (Flyvbjerg 2006; Yin 2009), we decided to apply a case study approach in these papers. The case study is an empirical enquiry involving both single and multiple cases, which has a distinct advantage over other methods when seeking to explore contemporary events over which the investigator has little or no control, especially when the boundaries between the phenomenon being researched and the context in which it exists are indistinguishable (Yin 2009). Further, it is a strategy in which the researcher explores the depth of a program, event, activity or one or more individuals that are bounded by time and activity. Data, or detailed information from multiple sources, are collected using a variety of procedures over a sustained period of time (Creswell 2014). The case study approach is based on the power of the example, and the value of this method has long been established, as in studies of aspects of livelihoods, or of organizations or institutions, where in-depth research is imperative for understanding a complex social phenomenon within these systems (Stake 1995; Flyvbjerg 2006; Yin 2009; Creswell 2014). In general, “the focus of the case study is on developing a narrative or revealing a phenomenon based on an in-depth, real-time, or retrospective analysis of a case” (Edmonds and Kennedy 2013: 113). Sayer (1992) notes that, “In designing concrete research we have to keep in mind the nature of our objects of interest.” Consequently, against the background of the overall research objectives, and due to the complexity inherent in the Belizean human–environment systems, it was found relevant to adopt a case study approach to my research. And because empirical fieldwork is necessary to understand relationships in their settings. Case study research is most closely associated with anthropological work and is often confused with ethnographic methods, which focus on the use of close-up, detailed observation, but nature of case studies is that they are interdisciplinary (Yin 2009). The approach allows the use of wide-ranging research methods, which was principally important for selecting the approach for these papers. Furthermore, case studies can also be confused with qualitative research (as in Creswell 2014), though evidence can be based on any mixture of quantitative and qualitative data (Stake 1995; Flyvbjerg 2006; Yin 2009). In these studies there have used a broad variety of both qualitative and quantitative methods or techniques, arguing that there is a strength in this methodology. In addition, Edmonds and Kennedy (2013: 113-14) provide an overview of different applied definitions of a case study. Flyvbjerg (2006) argues for the power of the case study and rejects the common view that it cannot provide evidence and reliable information by confuting five general misunderstandings about the approach. One general misunderstanding is that one cannot generalize on the basis of individual cases, which is why the case study cannot contribute to scientific development. This is a misperception, as case studies can enter into the collective process of knowledge accumulation in a given field. Studies without any attempt to generalize can certainly be of value in this process, and they have often helped on the path to scientific innovation. Certainly formal generalizations are essential and effective for scientific development, but they are not the only legitimate method of scientific inquiry (ibid.: 227). Concerning Paper 2 and 3, similar areas, projects and issues

44

as those described in the papers are found across the tropics; however, due to the wide variations in local contexts, the findings are not generalizable or directly transferable to other areas. Furthermore, none of the studies are intended to be critical cases, which are defined as having strategic importance in relation to the general problem, following the principle of “if this is (not) valid for this case, then it applies to all (no) cases” (Flyvbjerg 2006: 230; Yin 2009: 47). Nevertheless, they contribute to the critical research and theoretical advances with regard to linking adaptation and mitigation with analytic generalization, which is different from statistical generalization, where one can generalize to populations or universes (see Yin 2009: 15). An associated misunderstanding is that it is difficult to summarize and develop general propositions and theories on the basis of specific case studies, because the studies are irreducible, and essential contextual information will be lost by being summarized (Flyvbjerg 2006). Flyvbjerg (ibid: 241) agrees with this, especially concerning the case process, but points out that it is less correct with regard to case outcomes. Consequently, the researcher studies other cases and generalizes the findings as a summary of a topic. One example is land-change science, which draws on a broad array of synthetic and meta-analysis techniques to generate global and regional knowledge from local case studies. Furthermore, the demand for the generation of global and regional knowledge will continue to grow in order to support adaptation and mitigation policies that are consistent with both the local realities and the regional and global environmental and economic contexts of land change (Magliocca et al. 2014). Moreover, methods for analyzing and summarizing the findings from the case studies are proposed, because findings in summarized form might tell us new things about geographical trends (Rudel 2008), and internet-based engines as the GLOBE Project can accelerate the synthesis of local studies in land-change science (Ellis 2012). However, the increasing focus on global-scale work has been met with some skepticism (Fischer et al. 2011; Lindenmayer and Likens 2011) because big data can create new challenges like loss of context, intuition and real-world relevance. The focus has shifted from people to pixels, and decision-making has moved from local to higher scales in the search for generalizations rather than specifics (Fischer et al. 2011). However, others believe that the larger scale is favorable, as continental-scale interactions are fundamental (Schimel 2011), and meta-analysis of existing case studies, where the case study descriptions are held together with contextualized location factors in GIS layers, is a fusion of two complimentary approaches. They are therefore not contradistinctions, as this kind of synthesis of case studies adds value to the understanding at a larger scale (Verburg 2014). Nevertheless, the causes and consequences of local land change and their coupling with global and regional systems remain one of the fundamental challenges in land-change science, and challenges and potential biases are inherent in land-change science synthesis approaches (Magliocca et al. 2014). Finally, another misunderstanding is that case studies contain a larger bias toward verification than other approaches, that is, a tendency to confirm the researcher’s preconceived notions. However, case studies have a rigor of their own that is no less strict than the rigor of other research methods (Flyvbjerg 2006: 235; Yin

45

2009: 14). In addition, case study researchers often report that their preconceived views were wrong, which compelled them to revise essential points. Besides, the question of subjectivism and bias toward verification applies to all methods, for example, the element of arbitrary subjectivism in the choice of categories and variables in a quantitative investigation. In addition, mistakes are more likely to be corrected during the study because the case study researcher becomes very close to the subjects being studied (Flyvbjerg 2006: 235). 3.3.1

Case study methodology applied (Paper 2 and 3)

This section describes the cases being investigated in direct response to the research questions for Paper 2 and 3. The study area is described, and the arguments for selecting it as a primary case study area are set out first, followed by clarification regarding the timing and fieldwork locations. This is followed by explanation and comment on the data collection method and data availability and processing, which is succeeded by arguments regarding validity and reliability. Positioning, power and ethics in the fieldwork as conducted are then discussed, and finally, a note on a part that were later excluded from Paper 2 is provided. However, before these sections, some comments will be added regarding the distinctive features of the case study design and approach applied in the two papers (an overview is given in Table 3). The research subject is highly relevant to the choice of the case study approach and the choice and design of the qualitative methods used. Since the linking of adaptation and mitigation is a relative new way of thinking in which little empirical research has been done, the research in these papers adopts an exploratory approach, which is reflected in the methods of data collection and analysis. Examining an issue at an early stage which has not been clearly defined calls for a methodological approach that allows the researcher to retain the holistic and meaningful characteristics of such a contemporary, real-life event (Yin 2009: 32). Exploratory studies can provide significant insights into given situations, and the applied qualitative approach can give some indication of the why, how and when of situations, but it cannot provide how often or how many, which have to be investigated using quantitative approaches (Creswell 2009). Nevertheless, even an exploratory case study should be preceded by statements about what is to be explored (Yin 2009:37), which is why the purpose and the criteria of the exploration are stated in the papers. Beyond being explorative, both papers (mainly Paper 3) have certain elements of an ethnographic approach too, as the behavior of a ethnic group in its setting is the core of the study (cf. Creswell 2009: 195). In selecting the cases, we used information-oriented selected cases (contrary to random selected), where cases are selected on the basis of expectations about their content (cf. Flyvbjerg 2006: 230). The selected cases are intended to be characterized as representative cases with regard to the research questions, which is why it is assumed that the lessons learned from these cases will prove informative for cases with similar circumstances and conditions but without generalizable conclusions (Yin 2009: 48). Others would refer to these cases as being instrumental, which are case studies that provide insights into an issue or specific theme (cf. Creswell in Edmonds and Kennedy 2013: 188). However, there is no certainty that they are not extreme or unique cases

46

(i.e. unusual cases, which can be especially problematic or especially good in a more closely defined sense (Flyvbjerg 2006: 230)), or critical cases (i.e. cases described as having strategic importance in relation to the general problem and cases that permit logical deductions of the type (ibid.: 229-230; Yin 2009: 47)). The design in Paper 2 is imbedded, which is defined as case studies where each case is imbedded in the same system, but where they are invested individualy (Yin 2009: 59). And it is multiple-case, as four different aspects (subunits) are investigated in three projects (the cases). The project developer’s organizations each represent a case with its own context (ibid.: 46), which includes the local history, land uses and environment, and a common context that among things consists in a political framework. Paper 3 is also imbedded, but a single-case design, where information is pooled across the region and information related to individual villages becomes part of a larger main unit of analysis (i.e. the Toledo District). Furthermore, it is a single case because all villages are treated as operating in the same context, which is obviously not totally correct, as sub-contexts are present in each community; this is why we allowed ourselves to zoom in on certain villages in the text (i.e. the subunits of analysis). A multiple-case design (Paper 2) was selected for purposes of comparativeness and with the intention to maximize informational diversity, as well as because it is expected to produce a stronger evidence, thus strengthening and broadening generalizability and conclusions (ibid). However, the design implies a compromise over obtaining an exhaustive understanding of the cases investigated, which was more reachable in the single-case design (Paper 3). The projects in Paper 2 and villages in Paper 3 can be perceived as collective cases, which refer to cases studied together in order to examine a phenomenon, population, or general condition, and they “are chosen because it is believed that understanding them will lead to better understanding, perhaps better theorizing, about a still larger collection of cases” (Stake 2005: 445-6). The approach is qualitative, which, according to Creswell (2009: Ch. 9) is characterized by research that takes place in real settings, relies on the researcher as the instrument for data collection, employs multiple methods for data collection, is inductive, often involves the use of a theoretical lens, is based on participants’ opinions, is emergent (meaning that the initial plan for the research may not be tightly prescribed), is interpretive and is holistic. These are overall labels that can be attached to Papers 2 and 3. Furthermore, in qualitative research the dominant focus is on exploring experiences, perceptions, thoughts, notions, values, actions or feelings, and where the data analysis is designed to investigate subjects’ responses, narratives, or observations in order to identify themes and describe them in a descriptive or narrative manner (Kumar 2005: Ch. 2; Grimen and Ingstad 2007). The approach provides the possibility to understand how a culture acts by investigating its members experience, view and interpret the world (Grimen and Ingstad 2007: 281). For example, sustainable practices in agriculture in developing countries are promoted in many places, but many farmers have not implemented them. Why is this so, and what are the barriers that discourage farmers from employing these practices? Quantitative research, such as surveys, could provide indications, but qualitative research can obtain a more differentiated picture. Furthermore, one disadvantage with quantitative surveys and why that approach was disregarded in these papers is that the researcher must know what to ask for in advance. The qualitative

47

approach, on the other hand, is particularly suited to studying connections in complex social and cultural phenomena and for studying how they function in practice. What values do people hold, and how are values bound into a system? In qualitative research, the focus is on the relationship between human behavior and the various contexts (ibid.: 283), which is why this approach was selected. Furthermore, as the research subject and the study area are complex entities, it would have been difficult to locate or construct quantifiable variables that could be linked directly to the research question, which is why variables only would be proxies with little validity.

3.4

Study area, fieldwork methodology and reflections (Paper 2 and 3)

The history of the country, its extraordinary ethnical diversity, and the beauty and huge variations in nature were some of the things that personally fascinated me about Belize. Beyond these more personal selection criteria and that, in general, Belizeans are open and kind, the academic selection criteria for studying ‘linking adaptation and mitigation in agriculture and forestry’ were fulfilled. In Belize, I found a country that both had adaptation and mitigation on the agenda in the agriculture and forestry sectors, with forest sequestration projects and where deforestation is an issue (mitigation), but also challenges regarding the impact of future climatic change (adaptation). Furthermore, in Belize I found a less studied part of the region. The latter points are elaborated further, but first some general knowledge about the country of Belize provided. Specific details about the areas investigated are described in the papers. 3.4.1

The study area: Belize

Main geographical information Belize is located between 15˚52' and 18˚30' North latitude and 87˚28' and 89˚14' West longitude at the northernmost point and on the Caribbean coast of the Central American isthmus; it borders Mexico, Guatemala and the Caribbean Sea (Figure 10). It is the second smallest country in Central America, measuring 280 km from north to south and 110 km from east to west, with a land area of 22,963 km2. The country is topographically divided into two physiographical regions: 1) lowlands with semi-deciduous forests, savannahs with numerous freshwater wetland areas, and coastal lagoons and mangroves along the coast in the northern part and southern coastal plains, and 2) the Maya Mountains. The Maya Mountains, and partly also the plains of the south, are dominated by submontane and montane broadleaf forest interspersed with pine savannas. The country has 85 types of recognized ecosystems due to the varied topography, geology and strong seasonality of rainfall (Young 2008). The country’s infrastructure has improved tremendously in recent decades, and the main stretches (i.e. the Western, Northern and Southern Highways) are now paved (Berman 2011). In the coming year a third entry point into Belize will be added, when the Southern Highway is extended into Guatemala (Cisco 2015). However, roads outside the main highways are dirt roads in poor condition (Berman 2011). The Belize

48

Electricity Ltd. company has extended the electricity and communication grids: 83% of households are now connected, 25% of the population use the internet, and 76% of households have a mobile phone. The water supply source for 85% of households is public piped water, but for the majority purified bottled water is used for drinking. The figures are relatively uniform across the districts, but significant lower in Toledo District in the south (SIB 2011). Independence, economy, governance, and land tenure Belize was known as British Honduras until 1973, but after 125 years as a British colony, independence was obtained in 1981. A border conflict with Guatemala, which is ongoing, delayed the independence process, and Guatemala refused to recognize the new nation until 1992. Before independence, the region was disputed by the British and Spanish in the 17th and 18th centuries, and further back, it had several Maya city states, which declined at the end of the first millennium A.D. This is documented by the many archeological sites present today. The political system is a unitary parliamentary constitutional monarchy with two chambers (the Senate and the House of Representatives), which is dominated by two parties: the center-right United Democratic Party (UDP) and the center-left People's United Party (PUP) (Berman 2011; CIA 2015). Economically, tourism has become the backbone of the economy, followed by exports of marine products, citrus, cane sugar, bananas and garments. The country has the second highest per capita income in Central America, but income disparity between rich and poor is enormous. Among major concerns are the sizable trade deficit, the heavy foreign debt burden and high unemployment. Another political concern is growing involvement in the Latin American drug trade, which generates high crime rates (CIA 2015). Regarding land tenure, the country is divided into national land (owned by the government, including lease-land), forest reserves (administered by the government), private land and Indian reservations. Companies or individuals own most properties (63%), while another 25% is under lease. Reservation land is only represented in Toledo District among the Mayan communities (24% of households in Toledo) (SIB 2011). Population With a population of only 322,453 inhabitants, Belize’s population density is the lowest in the region (14 per km2). However, due to fertility and immigration population increased by more than 30% from 2000 to 2010. Until recently, most of the country’s population was rural; however, due to increased internal migration, the majority of the population now lives within the coastal zone (Young 2008). Nevertheless, the census of 2010 still showed that 55% of the population lives in rural areas (SIB 2011). For its size, Belize is extraordinarily ethnically diverse. Mestizos (Spanish/Latino), Creoles and the Maya are the three largest groups, with 49%, 21% and 10% of the population respectively, followed by the Garifuna (5%), Mennonites (5%), East Indians (2%), Asians (1%), Caucasians or Whites (1%) and others (SIB 2011). The official language is English, but all the ethnic groups have their own language(s) or dialect(s).

49

Figure 10. Administrative and political map of Belize (source: Land and Survey Department, Belize, 2010).

3.4.2

Why is Belize interesting for this research topic?

“A case is an edited chunk of empirical reality where certain features are marked out, emphasized, and privileged while others recede into the background. As such, a case is not “natural,” but a mental, or analytical, construct aimed at organizing knowledge about reality in a manageable way.” (Lund 2014: 224).

First, the country has a vast forest conservation potential with one of the largest contiguous forest areas in Central America (Emch et al. 2005). The estimates suggest that around 60% of Belize’s land area is covered by forests, but high deforestation rates have been documented (Whitman et al. 1997; Emch et al. 2005; Cherrington et al. 2010; Meerman et al. 2010; FAO 2010b; Cherrington et al. 2012). Belizean forests are facing

50

increasing pressure, including encroachment by swidden agriculture (a theme in Paper 3) and conversion of forest to pasture, citrus groves and large mechanized farms (Chomitz and Gray 1996; Government of Belize 2011). Additional drivers of deforestation and fragmentation include logging and urban growth (Horwich and Lyon 1990). To protect nature, 94 protected areas in 115 management units have been established, and 36% of terrestrial land is now under some form of conservation, which makes the country incredibly rich in biodiversity, including several endangered species (Meerman 2005). Many of the protected areas, including the projects in Paper 2, form part of the greater Mesoamerican Biological Corridor that connects forested areas from southern Mexico to northern Argentina, emphasizing its importance in national as well as regional conservation networks. The primary function of the biological corridor is to provide sufficient habitats (natural and semi-natural) within the landscape to allow for biological connectivity. Secondly, Belize is highly vulnerable to natural hazards and climate change. Its long, low-lying coastal areas are especially vulnerable to more intense and frequent tropical storms and hurricanes, flood damage and rising sea levels. The country has experienced frequent natural disasters of catastrophic proportions: most recently it suffered the impact of a category 1 hurricane (Richard in 2010), and Hurricanes Keith in 2000 and Iris in 2001 caused damage reaching 45% and 25% of GDP respectively (World Bank 2013). The IPCC’s (2013) 2050 projections for Central America predict increased temperatures with subsequent increased evaporation losses, decreased precipitation (minor), shorter rainy seasons and longer dry seasons, increased frequency and intensity of heavy rain events, stronger hurricanes with increasing disaster losses, and a general rise in extreme events like droughts and floods. Moreover, according to the latest national communication to the UNFCCC, the country might experience declines in crop production and forest damage as a result of severe weather events, which will in turn pose challenges for poverty reduction and development efforts, the country’s overriding priorities (Government of Belize 2011). Thirdly, there is a lack of natural science studies from Belize (Young 2008), especially on land-use change. This is partly explained by the fact that natural science research in Belize is mainly conducted by foreigners such as myself, and partly because educational institutions in Belize are primarily teaching institutions where research has a lower priority (ibid.). However, there are studies published in scientific journals on various topics this dissertation can build on, which are categorized below: 

Land-use change estimates (Fairweather and Gray 1994; Di Fiore 2002; Emch et al. 2005; Bray 2010; Cherrington et al. 2010; Meerman et al. 2010). Method: remote sensing.



Roads, land use and deforestation (Chomitz and Gray 1996). Method: a spatial GIS model.



Impact of farming systems and traditional agroforestry on forest degradation (Bernsten and Herdt 1977; Steinberg 1998; Levasseur and Olivier 2000; Emch 2003). Method: qualitative and partly quantitative.

51



Land-use system modeling (Rosenberg and Marcotte 2005). Method: economic land-use system model.



How cultural change has affected agriculture practices (Bernsten and Herdt 1977; Wilk 1997; Steinberg 1999; Steinberg 2002). Method: qualitative and partly quantitative.



Tenure and user rights (Wilk 1997; Steinberg 1998; Wilk 1999; Clark 2000; Steinberg 2002; Emch 2003). Method: qualitative.



Anthropological study of modernization at the household level (Wilk 1997). Method: qualitative.



Assessing drivers of smallholder deforestation (Wyman and Stein 2010; Wyman et al. 2011). Method: remote sensing and quantitative household survey.



Community conservation management (Hartup 1994; Diemont et al. 2011). Method: qualitative and partly quantitative, semi-structured interviews.



Forest management and greenhouse gas offsets (Cutright 1996; Corbera et al. 2007). Method: qualitative.



Farmer labor-exchange networks in swidden agriculture (Wilk 1997; Downey 2010; Couvet 2012). Method: qualitative and quantitative, household survey.

3.4.3

Timing and fieldwork locations

The fieldwork took the form of two separate trips to Belize between January 2013 and March 2014, during which numerous different locations were visited (Figure 11). I selected different fieldwork areas in order to address the research objectives, and I will briefly explain the considerations for selecting the base locations. I started with a reconnaissance trip around Belize (e.g. Belmopan, San Ignacio, Mountain Pine Ridge and Punta Gorda) for two weeks in January 2013, which had two aims. The first was to examine the feasibility of conducting the research by assessing the availability of the requested information, for instance, by checking the willingness and openness of expected interviewees, like government officials and farmers, to participate and answer the specific research questions. This was done as some issues might be controversial or sensitive for some. Staff members at the Environmental Research Institute at the University of Belize were very helpful in this process and provided the first important contacts, and I had the chance to test my ideas and some of the questions. Secondly, to assess the practical and safety issues11 with regard to bringing my family with two small children to the country for seven months, regarding, for instance, accommodation in safe areas. The trip clarified that both issues could be satisfied, and it also provided a good understanding the contemporary farming methods and the specific organization of different settlements. Together with earlier Belizean studies, the information gathered (interviews, informal conversations, observations) was used to select specific case

11

Belize is the sixth most violent country in the world, with an extreme high murder rate per capita, and public and social

life in Belize has been devastated by an epidemic of crack cocaine for decades (Wilk 2001; OSAC 2014).

52

study areas and focal points for the papers, which were later developed into the research tools for the more prolonged period of fieldwork.

Figure 11. Locations visited during fieldwork in Belize. The ‘projects’ are those studied in Paper 2.

The primary fieldwork lasted for almost seven months (205 days) between August 2013 and March 2014 (Table 5). The duration of the fieldwork, with considerable time spent in the areas investigated, was based on the prescriptions of Stake (1994: 242), who argues that qualitative case study research involves “the main

53

researcher spending substantial time, on site, personally in contact with activities and operations of the case, reflecting, revising meanings of what is going on”. There is no doubt that the fieldwork could have been accomplished more quickly, but it would have been at the expense of essential details. The primary means of transportation was a four-wheel drive SUV, which we bought on arrival and sold before returning to Denmark. Renting a car in Belize is extremely expensive, and many of the locations would have been impossible to visit using public transport. Moreover, the car also made it possible to relocate the whole family relatively easy, as it could contain all our belongings and four persons. Nevertheless, during the days I spent in the villages I walked or biked. For logistical reasons the fieldwork was conducted from three different bases (central, northern, and southern Belize), which are described further in the following paragraphs. Field notes were written at the base, which made it possible to take a step back and look at the experiences from the field. Writing up notes during the evenings was important as new connections became apparent, leading to the development of new ideas to be investigated later on. Table 5. Main fieldwork (205 days): spatial and temporal frames, and main aims Spatial frame:

Temporal frame:

Base locations Town (District)

Start and end date (no. of days)

Belmopan (Cayo)

8/8-2013 → 8/11-2013

Purpose (location)

     

Qualitative interviews (Belmopan, Belize City, San Ignacio, Gallon Jug, Succotz, Mountain Pine Ridge, Yalbac, San Antonio, Central Farm, Belize Int. Airport Village surveys (El Progresso, Upper Barton Creek) Other areas visited (Spanish Lookout) Project areas visited (Bull Run) Visited citrus processing plant (Pomona) Presentation of work (Belmopan)

Blue Creek (and Orange Walk Town) (Orange Walk)

8/11-2013 → 29/112013

    

Qualitative interviews (Blue Creek, Orange Walk Town, San Felipe, San Carlos) Village surveys (San Felipe, San Carlos) Other areas visited (Indian Church, Indian Creek, La Milpa, Reinland) Project areas visited (Rio Bravo) Visited chicken slaughterhouse (Tres Leguas) and papaya plantation company (SE of Rio Bravo)

Punta Gorda (Toledo)

30/11-2013 → 26/22014



Qualitative interviews (Punta Gorda, Maya Mountain Research Farm, Eldridgeville, San Jose, Naluum Ca, Jalacte, Crique Sarco, Blue Creek) Village surveys (Indian Creek, Pine Hill) Maya villages visited (Crique Sarco, Trio, Blue Creek, San Jose, Jalacte, Big Falls, Indian Creek, Naluum Ca, San Antonio, San Felipe, Silver Creek, San Miguel, San Pedro Colombia, San Vicente, Santa Elena, Santa Cruz, Golden Stream) Other area visited (Maya Center, Cockscomb Jaguar Basin, Belcampo, Maya Mountain Research Farm, Placencia, Monkey River) Project areas visited (Boden Creek) Visited chocolate-processing facility (Punta Gorda) and banana plantation and packing facility (Logans Bank) Community meetings (Crique Sarco, San Jose, Jalacte) Presentation of work (Punta Gorda)

       Belmopan (Cayo)

26/2-2014 → 28/22014



Qualitative interviews – cross-checking findings and initial conclusions (Belmopan, Central Farm)

For the first three months (August to October 2013) we lived about ten kilometers west of the capital, Belmopan. I chose to start here, as the official Belizean governmental institutions are located here, as are most NGOs. In addition, organizations and people not located in Belmopan visit the capital frequently, and meetings

54

can be arranged. However, day-trips from Belmopan to several places, such as the former capital (Belize City: 1 hour NE of Belmopan), were made regularly to conduct interviews. Furthermore, one of the three projects studied in Paper 2 (Bull Run: 1.5 hour SW of Belmopan) could be reached on day trips from there, where among other things I conducted village surveys in El Progresso and Upper Barton Creek (further comments on village surveys in Section 3.4.8). Spanish Lookout (30 min W of Belmopan), inhabited by the most progressive Mennonites in Belize, was visited on several occasions because the area is very interesting in terms of agriculture. In Spanish Lookout the yield is the highest in the country, and the area is called the food basket of Belize. However, agricultural expansion is causing deforestation of former protected areas, so an interesting dilemma occurs. Additionally, the citrus processing plant in Pomona (1 hour SE of Belmopan) was visited through a private guided tour in order to acquire a better understanding of the production structure of the most important commodity for Belizean GDP. The following month (Nov 2013), we moved to Orange Walk District, where the main base was the Blue Creek Mennonite community. This base was chosen because it is near one of the three projects studied in Paper 2 (Rio Bravo), and because the Mennonite communities in this area are the driving factor for deforestation on the land buffering the protected Rio Bravo Conservation area. The tourist facility, La Milpa, inside the Rio Bravo Conservation area was visited in order to view the protected area from the inside. However, the greatest amount of time was spent in the Spanish-speaking communities buffering Rio Bravo, where I observed livelihoods, especially farming and forest utilization, and where village surveys in San Felipe (Picture 1) and San Carlos were conducted. Lastly, I was given private guided tours at the largest chicken slaughterhouse (in Tres Leguas) and papaya plantation and packing (SE of Rio Bravo) facilities in Belize, to absorb additional information about large-scale farming in the country. The fieldwork in this month was complicated by flooding, which was caused by the highest rainfall in 25 years and made several areas isolated and inaccessible. For the last three months (December 2013 to February 2014), our base was Punta Gorda in Toledo District in southern Belize. This part was mainly devoted to spending time in Maya villages, particularly Crique Sarco, San Jose and Jalacte, and to interview farmers and informants about Maya farming and their livelihoods in general. Much time was also spent in Punta Gorda, the district capital, which has the main organizations working in the villages. Furthermore, Punta Gorda holds the only market days in the region, which attracts people, including potential interviewees, on a weekly basis. One hour from Punta Gorda is the third and last of the projects studied in Paper 2 (Boden Creek), which was visited, with village surveys being conducted in Indian Creek and Pine Hill. The rainfall was still extraordinary for the season, which made some isolated villages inaccessible. For instance, after I had postponed visiting the Mennonite community Pine Hill for two months (Picture 2), I decided to use a mountain bike to make the last part, as the car would probably have become stuck on the muddy road. I already had been pulled out twice in other locations during fieldwork, which was very troublesome.

55

Picture 1. Main Street (San Felipe Village, Orange Walk District, November 2013).

Picture 2. Road to Pine Hill in the dry season (Toledo District, Belize, February 2014).

56

Other interesting fieldwork sites in Toledo were visited through private guided tours to a chocolate-processing facility (Punta Gorda), a banana plantation and packing facility (Logans Bank), the Maya Mountain Research Farm (San Pedro Colombia), where they experiment with multistory permaculture farming, and finally on several cacao trial projects. Furthermore, I was invited to participate in a workshop at Maya Mountain Cacao (one of the two cacao operators in Belize), where the task was to plan the ideal planting pattern of 120 acres of cacao agroforestry, complete with timber trees for shade and diverse intercrops. Before leaving Toledo District, I arranged three community meetings (in Crique Sarco, San Jose and Jalacte), and a meeting with the scientific staff of a local NGO working with forest conservation and agriculture – all in an effort to verify the discoveries from Toledo. Finally, I returned for two days to Belmopan to cross-check my initial findings and conclusions from the entire fieldwork with four key informants I had interviewed at the beginning of the fieldwork. They were carefully chosen, as I assessed them as holding the big picture concerning both the research in question and Belize in general (further comments on this in Section 3.4.6). Information for Paper 2 and 3 was collected continuously and in parallel during the fieldwork. However, the emphasis for the first four months was on Paper 2, and in the last three months on Paper 3. This was done because potential interviewees did not necessarily live in proximity to the areas of investigation. For instance, a key informant who wrote his master’s thesis about conservation in southern Belize ten years ago now lives in northern Belize. Thus, for logistical reasons he was interviewed in November before I moved south. This was the situation with several interviewees, and many were interesting and knowledgeable with regard to the themes of both papers, which is why they were interviewed on both issues. These interviews were separated into two parts, but not with an equal emphasis on both, as the interviewee often had forest/conservation (Paper 2) or agricultural/Maya (Paper 3) as the main area of knowledge. 3.4.4

Data collection methods

To answer the research questions, multiple data-collection approaches and techniques for different specific purposes were used. Even though personal observations and literature provided a lot of background information and topics to explore, it was semi-structured interviews that proved the key to obtaining the necessary knowledge to grasp the complexities and interrelationships being investigated. This was mainly because semi-structured interviews are designed to capture the intangible aspects imbedded in the questions, such as inclusion in conservation projects in a certain cultural and political context. This is why the emphasis is on this part in the following, but first I provide a paragraph on how the interviewees were selected. Additional information, such as interviewees’ affiliations and number of interviewee encounters, can be found in the two papers. Selecting interviewees As mentioned in the previous section, the Environmental Research Institute at the University of Belize was helpful in providing the first important contacts during the first stay. But the second stay also started well

57

concerning contacts. I had rented a house for the first three months outside Belmopan, which by coincidence was the house of the former head of UNDP in Belize, who had formerly taught social sciences at the university and was now working as an independent consultant. He was Dutchman, married to a Belizean (who was working at UNDP), and he had lived in Belize for the majority of his life. For these reasons, he had extensive contextual knowledge and an enormous network with researchers at the University of Belize, government officers and leaders in ministries, independent researchers and consultants, employees and leaders in NGOs, etc., which he gladly shared with me. He also knew many of the potential interviewees’ educational backgrounds and previous engagements, which was very beneficial. Belize is a small country with few educated people, which is why many have a past in other institutions and organizations where I also interviewed people. On occasion some interviewees were willing to inform about their experiences from previous work, thus providing information that was difficult to obtain from the current employees, as they were unwilling to provide that kind of information. I assessed him to be neutral with regard to the research, reliable and independent, and he was also companionable, reflective and articulate, which gave rise to many informal conversations about Belize which clarified many of the issues in the research. However, the conversations with him and his wife are not directly apparent in the papers, as they mainly provided contextual information and contacts. The overall idea in selecting interviewees was to obtain as broad a range of people who were knowledgeable about the issues as possible, while at the same time choosing interviewees from a diverse set of affiliations. This was done to cover the topics as broadly as possible and to obtain information about potential incongruences between the different groups and affiliations. Furthermore, both internal and external interviewees were needed in order to examine whether they had different views on the research issues. They were selected deliberately, meaning that individuals who would best help to understand the research problem and the research questions were selected (Creswell 2014: 246). Potential interviewees were not chosen for their statistical representativeness, but for their knowledge, opinions and accessibility. However, some were selected because they represented a certain organization or group, for example, those who were directly involved in the projects in Paper 2 or persons from organizations working in the Maya villages in Paper 3. Other informants were less obvious and were traced through other people through snowball sampling, that is, respondents being identified by previously interviewed respondents – a beneficial method to use when people are challenging to locate and access (Valentine 2005). Hence, by the end of an interview, the interviewee was asked if they knew anyone else who could contribute to the issues in question. This paid off with additional contacts, but inevitably fewer newcomers emerged as the fieldwork progressed. Using this approach, I achieved contact with almost all the desired interviewees. Some declined or did not answer to start with, but they were later persuaded to participate, though a few (i.e. two international auditing/validation companies referred to in Paper 2) proved impossible to enter into contact with. I stopped collecting interviews when the issues became saturated, that is, when fresh data no longer sparked new insights or revealed new properties

58

(cf. Creswell 2014: 189). However, this was a cumbersome approach to use, especially in the beginning, as one does not know the image or the number of pieces of the puzzle one is collecting. Moreover, the orientation in a case study does change during fieldwork as new topics emerge, which is the flexibility and thus the strength of this approach, but slippage from the original research question can at the same time open up a space for criticism (Yin 2009: 52). Thus, the researcher has to control the situation and keep the research on track, which is why some individuals were interviewed more than ones. New topics emerged, and I had to return to some interviewees to obtain clarifications or ask for their knowledge about the new discoveries. However, persons with several interviews can also indicate that one meeting was simply not enough time-wise. Arrangements for the interviews Interviews were arranged by sending a request by email, in which I introduced the topic, the purpose of the interview and myself. Next the date and location were arranged by email (or phone). In many cases I sent a reminder to them the day before, especially if I had to travel a long way (in some cases 2-3 hours each way on bad forest roads). The interviews with white-collar workers mainly took place in their offices. However, some were conducted in restaurants (over lunch or dinner) or in their private homes. I always came to the location they had selected to make it as convenient for them as possible. The majority of interviews were conducted in privacy and face to face, but I did some on Skype, as the extreme rain in 2013 made several areas inaccessible for long periods, or else because the interviewee was traveling or lived abroad. Interviews with farmers and people working in the field (e.g. field stations) took place in the field, that is, in the forest, on farmland, along the road, in plantations, at a market or in farmer’s houses. These interviews were conducted face to face, and the number of people present during my interviews was kept to a minimum, because I sensed that the interviewee became more reluctant to speak when others were present. Interviews in the field benefitted from being more concrete, directed towards specific and directly observable matters, with farmers, for instance, explaining their fields (e.g. crops, spacing and challenges) and agricultural practices (e.g. the shifting cultivation pattern). It was clear that farmers and foresters liked the interview situation in the field, where the researcher has less control, better than around the house or in an office, and the researcher– interviewee relationship seemed more equal, with the privacy of the field making the interviewee less reluctant to participate. The semi-structured interviews I always started the interviews by explaining that I was a Ph.D. student and presented the main topic, and explained that the findings would result in a Ph.D. dissertation and that I guaranteed anonymity (i.e. no names in the final products of the research without their acceptance). In some cases I had to explain that my research was not intended to result directly in a development project or similar, but that I was hoping that my work would be beneficial for the Belizean Government or donor organizations when it came to targeting their initiatives. I then invited the interviewee to tell about their affiliation, educational background and their work,

59

and how they were related to the topic. This was followed by some clarification regarding definitions and understandings of key terms, especially the terms ‘adaptation’ and ‘mitigation’. Then I started with questions from the interview guides that I judged fitted the interviewee (in accordance with their affiliation, education, location, position in organization and knowledge of the village or topic), meaning that the interviewee was presented with certain parts from the interview guides. For instance, the meteorologist was asked questions about weather patterns and climate change and the forester about the impact of weather and climate on the forest. Thus, the interview guides (see for example the guide used for Paper 2 in Appendix 1) were meant to guide the interview. Regarding Paper 2, the guide contains all the topics I needed answers on, whereas in the case of Paper 3 this was more open because the idea was that a conversation about the suggested adjustments should reveal the potential barriers. The interview guides were partly inspired by the literature, project documents and the initial scoping mission. As well as the interview guides I used other tools like visual aids (maps, satellite images, a seasonal calendar), transect walks and field visits to guide some interviews. For instance, satellite images of the three project areas in Paper 2 were used during the interviews to visualize and identify boundaries, access roads, forest types and buffering villages. The interviews conducted in relation to Paper 2 featured a ranking exercise (Box 1), and some of the comments obtained during this exercise feature in the paper, but the actual results from the exercise was ultimately excluded from the paper. Interviews in general are very useful when the subject or participants cannot be directly observed (Creswell 2014: 191), as with the barriers in Paper 3. Furthermore, the interviewees can provide historical information that is not available elsewhere, which then allows the researcher to control the line of questioning (ibid.), which, for instance, is not possible in a questionnaire. The principle of the semi-structured interview is to have a freely flowing conversation on a selected topic in which the researcher seeks to acquire the trust and confidence of the person interviewed. There should be no correct answers, as the exercise involves exploring both the width and depth of the topic. However, the researcher must steer the conversation with his interview guide and not to let himself be tempted by the many interesting side topics (peripheral issues), or else he will fail to cover essential themes concerning the main subject (Grimen and Ingstad 2007: 294). However, researchers are often, like me, outsiders without absolute knowledge of custom, local conditions, power structures or the personnel history of organizations, thus room must be made for an open discussion in order to bring in new issues that the researcher is unaware of, and to sense what matters the most to the interviewee. The idea with semi-structured interviews is that they should be non-directive, non-standardized, open-ended and directed towards understanding the interviewee’s perspectives, experiences or situations in his or her own words (Spradley 1979; Kvale 1996), and not to overly structure the conversation, as this might prevent surprising insights and connections from emerging that it was not possible to conceive a priori (Kvale 1996; Taylor and Bogdan 1998: 99). The latter also depends on the scope of the particular interview. When doing case studies, data collection is often both about the subject of the case study (the “phenomenon”) and data external to it (the “context”) (Yin 2009). However, sometimes it is hard to distinguish between these two data

60

Box 1. Ranking exercise (Paper 2) The exercise was made with 32 of the interviewees who had extended knowledge of 1) forest projects in Belize, and 2) adaptation and mitigation in forest projects. The duration was around ten minutes, and it was carried out at the end of the interview, where the interviewee had a greater understanding of the topic. The aim of the exercise was to confront interviewees very directly with the research topic by having them decide whether they agreed or disagreed with concrete statements. This was designed to determine whether the interviewees were in favor of linking adaptation and mitigation, neutral or resistant. I applied the Q methodology (van Exel and de Graaf 2005), which forced interviewee to acknowledge their attitudes to the topic in question. Creating the statements was guided by the frame set out in Table 6, where some are in favor of linking, others are neutral or resistant. The interviewee had to insert twelve statement cards into a template (Figure 12) to determine whether they agreed or disagreed with the statement. The fact that nine of the statements are mainly against linking adaptation and mitigation in the projects created dilemmas for the interviewee. If more or less the same number of positive and negative statements were made with regard to the linking, the exercise would be too predictable. Furthermore, the forced methodology using dilemmas also resulted in additional interesting discussions. Table 6. Statements used for the ranking exercise (the numbers solely functioned as statement-ID) In favor of linking

Neutral or resistance to linking

Linking adaptation and mitigation in forest projects in general

8: It could be beneficial with a standard for forest projects that focused on both adaptation and mitigation.

2: I find the simplicity of a project important. Meaning, activities, should be focused on the main aim of the project.

10: Mitigation in some projects, adaptation in others.

Include ecological adaptation in forest mitigation projects

9: Other ecological services, such as biodiversity and watershed protection, are important in forest mitigation projects, and should, therefore, be included proactively in the activities. 11: Carbon sequestration could be increased by including local communities proactively in the project design.

4: Funders and standard verifiers are only interested in mitigation, and think that social and ecological adaptation complicates projects. 1: Biodiversity is an important issue, but not important in forest mitigation projects.

5: Selection of tree species is not important in these mitigation projects since the trees will naturally adapt to the future climate.

6: The design of the project is decided by the funder and the standard applied, not by the project developer.

12: Sustainable development is less important in a mitigation project, as carbon is the main aim.

7: I do not think that the project needs to consider social adaptation that much, as it is a mitigation project.

3: The impact of the project on nearby communities is less important since the project zone is private property.

Include social adaptation in forest mitigation projects

Figure 12. Template used for the ranking exercise: From ‘most disagree’ (1) to ‘most agree’ (6).

61

types because the boundaries between the case study and its context are rather fluid (ibid.). For instance, in the interviews related to Paper 3 it was unavoidable not to have a contextual discussion of land tenure in Toledo District in between questions directly related to the main research topic. Consequently, during the interviews it was difficult to maintain a sharp distinction between the central research topic in question (the linking of adaptation and mitigation) and contextual information. In the interviews, with their open-ended questions, I sometimes allowed the conversation to go down byways, which sometimes proved to be blind alleys, but on other occasions resulted in new discoveries that would have gone unmentioned if I had controlled the conversation too strictly. Secondary sources of information Collecting a range of secondary sources and records, such as reports and books, which are inaccessible (online and hardcopy) from abroad was a very important additional activity during fieldwork. These documents became accessible during interviews or by showing up in official offices (e.g. University of Belize, Statistical Institute of Belize, Belize National Meteorological Service, and Forest/Agricultural Department), NGO offices and field stations. Among the information gathered was national population and housing census data; agricultural statistics of yields and production of crops at the district level; statistics on quantities of livestock at the district level; monthly precipitation and temperature figures for the investigated areas (historical and projections); education statistics; and farm-level information on orange and cacao production. This information was mostly used to estimate rough national and district trends because the confidence in specific data was weak due to collecting and handling procedures. Another treasured piece of information, especially for a geographer, came from GIS layers obtained from two of the interviewees (interviewees 56 and 78 in Paper 2:Online Resource 2) and the project material related to the projects in Paper 2. These layers provided information regarding protected areas, fire risks, hurricane damage, historical hurricane tracks, logging (legal and illegal), settlements, roads, ecosystems and land cover, project polygons, soil types, topography, watersheds, biological corridors and district and national borders. In addition, libraries had valuable information, and the library at the Caribbean Community Climate Change Centre (CCCCC) in Belmopan was especially applicable, as it contained historical documents, reports and books on my research topics. At CCCCC I was allowed to borrow the material in order to make copies. In other cases, I was permitted to take photos of the documents, or simply handed a copy. Books were purchased, but the book market in Belize is limited, which is why books about Belize have to be tracked down, for instance, at tourist locations, where there is a small market for books. Other sources of information and networking Observations in villages, markets, fields and forests on a daily basis were another method of acquiring contextual information, for instance, information about house construction; available food the small local shops; willingness to talk to outsiders; food available close to the houses; access to transportation; encounters

62

with protected animals; and encounters with people with shotguns or extracting timber in a strictly protected forest. As Robbins (2012) comments, even though this kind of information is difficult to imbed in the analysis or in final papers, it is irreplaceable in acquiring an overall understanding. It provides observations like: how the traditional houses make sense in the heat; how to store corn without getting weevils in it; seeing buses with workers from San Felipe coming into Blue Creek every day (which says a great deal about why San Felipe is not a village of farmers any more); or the origin of products at the market (often Mexico, even though similar products were produced locally). Many of these personal observations were supplemented by informal conversations with the individuals present, which facilitated access to valuable information. Other informal conversations in churches, private homes, sports fields, restaurants and bars, preschools and tourist locations added further information. Some speak about these secondary activities and observations during fieldwork as “learning from below” and refer to it as “hanging out” (Wainwright 2011). We as a family participated in social gatherings where information was obtained in an informal way and networking happened. For instance, we participated in a baseball match with the Mennonites in Blue Creek and in celebration of Independence Day in Belmopan, as well as a party for the Swedish ambassador for Central America.12 On my own, I simply met people, for example, by providing lifts or when biking around Punta Gorda. The relatively informal circumstances at these encounters provided an atmosphere of confidence where information was shared in both directions. Furthermore, the informal encounters gave me an opportunity to explain the objectives of my fieldwork, which led to some contacts and showed me who was connected. The latter aspect was important in a small country like Belize, where many of the interviewees know each other, which can affect the fieldwork both positively and negatively. For instance, it had a negative impact in one situation, where too late I became aware that two leading figures I was in contact with were a married couple, and information was running ahead of me, which made a particular NGO increasingly reluctant to share information or to give interviews. I another situation, it had a positive impact in the sense that many of the persons I had already interviewed opened the gate to additional interviews that could have been hard to obtain on my own as another unimportant Ph.D. student from the North. For instance, a contact established through the owner of my children’s preschool gave me access to the CCCCC library mentioned above. Many of these happenings occurred by coincidence, for instance, when my wife got into conversation at a carwash with a Maya women, whose father had worked for the British as a development officer in Crique Sarco (one of the main Maya villages in Paper 3) in the 1960s. I later interviewed him, which would not have happen without this coincidence, as I was not aware that he was still living in the area. Another, almost unbelievable coincidence was that one of my project areas in Paper 2 was formerly owned by a Danishmanaged company, which, given the closeness in nationality, provided remarkable historical information about ownership, land-use and land-fertility. The last coincidence I will describe here occurred because my wife and 12

We were invited because the husband of the owner of our children’s preschool in Belmopan was the Swedish consul.

63

children went to church in Blue Creek (Orange Walk District) while I was doing interviews in the other villages. The Mass on this particular Sunday was in Low German (Plattdeutsch), which is impossible to understand, even for my High German-speaking wife. A young woman offered to translate. Later, at a private dinner, I discovered that she was married to the owner of the land buffering one of the projects in Paper 2, which we visited together, where he provided extraordinary information about, for example, land use, landownership and illegal logging. Without this coincidence, I would probably not have found out who owned the most important piece of land for my research in that area, because land ownership information in Belize is controversial and, therefore, difficult to obtain. In general, the personal observations and information from informal encounters revealed the power structures within society and delivered useful contextual information that could only have been obtained through my long-term presence in the country in the company of my never-ending curiosity. Furthermore, this kind of information was valuable for my general understanding of the areas investigated, and it fed into the discussions with interviewees. Recording and notes All the semi-structured interviews conducted in offices, homes, restaurants, etc. were recorded with a small digital recorder, which I asked permission to use (one person refused). In a few cases, I found that the presence of the recorder prevented the interviewee from speaking freely, and it might have resulted in withholding information. In the field not all interviews were recorded, as walking, rain and noise, among other things, made it troublesome. Notes were taken, however, no matter whether the interview was recorded or not, but if not recording the notes was more thorough. On many days I had a long way to go back to the base, and while I was driving I used the time to voice-record observations, reflections on an interview, or to note that an earlier interviewees had to be seen again to obtain their views on a newly emerged topic. Further, when unexpected information appeared, new questions were merged into the interview guides to cross-check the information. This emergent design, in which research regularly evolves, is characteristic of qualitative research (Creswell 2014: 186), and exercising adaptability and flexibility are among the important strengths of the case study design (Yin 2009: 70-1). Back at the base, I had time in the evening or the following day to reflect on and make an initial analysis of the interviews, which was done simply by listening to the recordings and reading the notes, as well as by discussing issues and findings with the people in my daily social sphere. All the notes were entered into Word documents, one for each paper, which were divided into topics to start with, but where new topics emerged continuously. Some topics later merged together, while others were separated into several new topics. This dynamic structure in these documents functioned as an ongoing discovery, and the process of analysis, in which themes were identified, and concepts and propositions developed (cf. Taylor and Bogdan 1998: 141), was well under way. However, I had to take care in this process not to lose the overview and perspective of all

64

the qualitative data (cf. Kvale 1996: 176), which certainly happened to me momentarily. To prevent this, I had integrated into the Word documents featured descriptions of the components within the broader research questions and the data needed to address each component, which I returned to regularly to keep the data gathering on track. In the first two months of the fieldwork the notes were rather unstructured, but as fieldwork progressed my notes became more focused as I narrowed down the interviewee list and as the questions I asked during the interviews became more accurate. Beside these Word documents, I had other dynamic tools that evolved on a daily basis. One document listed potential interviewees or persons to get hold of with contact information, affiliation and subjects for interview, which were colored-coded according to importance: “have to be contacted”, “potentially” and “long-shot”. Other tools were my schedule and a do-to list of different tasks I needed to carry out during the fieldwork. All the documents (the two case studies, the list of potential interviewees, the schedule, the to-do list) were my ‘fieldwork protocol’, which were specifically used to guide the fieldwork and later on to write this dissertation, including the papers. 3.4.5

Data availability and processing/analysis

This paragraph describes the analytical technique that was used to explore the collected data, but it should be mentioned that the analysis phase in itself was a continuous process, in which the collection and the processing phases are difficult to distinguish from each other. As described in the section above, the fieldwork did feature some degree of analysis, that is, the interpretation of the collected data started in situ while in the field while writing notes and listening to interviews. The distinction between data collection and data analysis is in principle present, but it becomes blurred in practice. Nevertheless, the following will treat the process of making the data manageable and the actual data analysis, because of its importance in the research design. The data analysis followed the steps described by Creswell (2009), as shown in Figure 13. The raw data (i.e. interview recordings) had to be organized and prepared to start with. This process started when returned to Denmark, because I considered it important to have all the interviews done before the actual analytical process began, as I then had the experience of all the interviews in the back of the head while handling the individual interview. For each interview I listened to it, taking notes. The notes were divided into the topics that occurred during the interview, including background and contextual information. Some sentences were transcribed literally if it was assumed that the statement could be used as a quote in the papers. Furthermore, during this process I added my personal interpretations and notes as links to what was said in other interviews, for instance, recurring themes, convergent or conflicting views, and any connection between people’s affiliations and their points of view. This step tended to overlap with the forthcoming steps, as will be described later. On average, it took me two to four hours to boil a one-hour interview down to one to three A4 pages of notes, which made the interviews manageable in the analysis.

65

Figure 13. Data analysis process in qualitative research (Creswell 2009: 185)

Conducting full transcriptions of the interviews was judged unnecessary for several reasons. For instance, the less strict control during the interviews, in order to make the situation less formal and comfortable, produced many fascinating detours that had little importance for the papers but provided essential contextual information. Consequently, full transcriptions of all the conducted interviews would have resulted in around 1000 pages,13 which would have been unmanageable in the analysis, and besides incredibly time-consuming. Beyond the high costs, would it have been difficult to use transcriptions made by external persons or companies, as one had to be present at the interview to make a satisfactory transcription. This was because prior knowledge was taken for granted in the conversation, which made the interviews complex to transcribe and transcriptions difficult to use. Deciding when someone is talking about a particular topic is often a matter of interpretation, as people very rarely use the precise words that researchers have chosen to label our analytical categories with (Mason 1994). Furthermore, location noise and dialects (many Belizeans speak a Creole dialect of English) were other concerns against outsourcing this task. After the initial screening of all the interview notes and my own notes, NVivo (version 10) was used to assist in the qualitative data analysis. NVivo is an analytical tool that allows you to organize, sort and search the information collected in the interviews. Basically, it is an advanced copy and paste program that makes it 13

The interviews lasted for 67.15 hours, which, at 15 pages/hour, is 1,007.25 pages.

66

possible to code imported Word, PDF and audio files. First, coding was conducted, that is, the process of organizing the material into chunks or segments of text and assigning a word or phrase to the segment in order to develop a general sense of it (Creswell 2014: 241). Coding makes extensive material easier to handle and maneuver in, and it also helps develop and refine interpretations of the data later in the process (Taylor and Bogdan 1998). A sentence, for example, that contains information about the inclusion of communities in forest projects, one specific village and adaptation needs in agriculture is coded using these three nodes. When traveling through the gathered data, other information fits the same nodes and is copied into them. This is done in order to identify common narratives and to bring together recurring themes related to the research questions. The search for themes is a technique that has been suggested as a useful way of systematically grasping qualitative data (e.g. Mason 1994; Creswell 2009). The lists of nodes used for the two papers were a combination of emerged and predetermined nodes (Figure 14), which is a common approach (Creswell 2014: 199). For instance, among the predetermined nodes were ‘project information’ and ‘adaptation needs for forests’, whereas ‘additionality’ and ‘tax and forest’ emerged during the fieldwork and data analysis. Figure 14 shows the final structure of the developed hierarchical sorting of nodes, which emerged during the analysis, as patterns began to appear. However, in the beginning only some of the major themes existed, and the final structure is the product of refining, adding, and collapsing or expanding the nodes, which also made me aware of the analysis I could make and what the data were showing.

Figure 14. Overview of nodes (capital letters) and codes extracted from NVivo applied to the analysis of the interviews made for Paper 2.

67

The main coding themes used in Paper 3 are not shown here, but they were primarily the barriers and secondary contextual information. Most key informant interviews have been used in both papers; however, some provide information of a contextual character, while others focus more narrowly on specific topics. Finally, data ordered into themes and sub-themes were interpreted, which means that the researcher draws meanings from the findings of the data analysis (cf. Creswell 2009: 4). During the writing of both papers, for reasons of verification I returned to the data to consider how it could have been biased or disturbed in other ways by the applied collection method (the loop in Figure 13). 3.4.6

Validity and reliability

Validity strategies in qualitative research are procedures used to demonstrate the accuracy of the findings and to convince readers of this accuracy, whereas reliability is consistency in a particular approach across different interviews, researchers and projects (Creswell 2014: 246). The tactic behind constructing validity in a case study is to use multiple sources of evidence, establish a chain of evidence and have material reviewed by informants (cf. Yin 2009: 41), which all are tactics applied in these papers. Reliability is sought in these papers by providing explicit and detailed descriptions, for instance, lists of interviewees (Online Resource 2 and 1 attached to Paper 2 and 3, respectively), interview guides (e.g. Appendix 1) and themes and codes of analysis (Figure 14), which all make the studies repeatable and increase reliability (Yin 2009:45). Nevertheless, it can be difficult to locate exactly the same interviewees, as they have been anonymized to protect them. In order to ensure validity and reliability, the research followed the criteria described below, which appear in chronological order. First of all, by spending prolonged time in the field (in this case seven months), the researcher develops an indepth understanding of the phenomenon under study and can convey details about the site and the people; that is, the more experience the researcher has with the actual settings, the higher is the validity of the findings (Creswell 2014: 202). Secondly, I personally bring some bias to the studies, as my interpretation is shaped by my background (gender, culture, history, socioeconomic origin) (Creswell 2009: 191), but certainly also by my having been trained as geographer and not, for example, as an economist or anthropologist. This sort of bias is imbedded in all studies, and it cannot be detached, but it is important that the reader of these studies reflects on this (more on this in Section 3.4.7). In preparing for the interviews, I drew up the interview guides and selected the interviewees (in the case of Paper 2 this was done together with co-author), which, of course, was intended to be done in such a way that objectivity was obtained, for instance, by selecting interviewees from different standpoints to obtain diverse opinions. Nevertheless, this part may have been biased by the choices I made, like overlooking issues, questions or interviewees. To counteract this bias, at the end of the interview interviewees were encouraged to bring up issues that had not been touched on and to mention potential interviewees (see snowballing: Section 3.4.4) that related to the research in question.

68

Thirdly, the procedure followed during the interviews is critical, as validity depends on the quality and reliability of the respondents’ information. One of the greatest concerns for validity in qualitative research using interviews is the limitation that information is filtered through interviewees (Creswell 2014: 191). Thus, to minimize this issue, questions were framed cautiously in order to lessen biases, such as interview biases or strategic answers (Bateman et al. 2002). Kvale (1996: 34) comments that the interviewer in a semi-structured interview should lead “the subject toward certain themes, but not to certain opinions about these themes”, while Spradley (1979: 58) comments that an interview is a friendly conversation “into which the researcher slowly introduces new elements to assist informants to respond as informants”. For instance, the interviewee may be affected by the designated place of the interview (natural field setting vs. an office or restaurant) or by the presence of other people (including the researcher, which is elaborated on in Section 3.4.7), which is why I had the interviewee select the setting and preferred tête-à-tête. To avoid misunderstandings and confusions in during the interviews, I started by explaining the applied definitions of adaptation and mitigation if it was intended to use these terms in the particular interview. Regarding strategic answers, those who were interviewed might have an agenda, some obvious and disclosed, others hidden from the interviewer. To minimize this issue, responses and affiliations were cross-checked with other interviewees. Overall, answers and the interpretation of them were cross-checked with future interviewees, as well as with information obtained from informants and other respondents, which was done to ensure a high degree of reliability pertaining to the collected information. This procedure is especially important for contradictory information that runs counter to the themes (Creswell 2014: 202). However, after many interviews, it is hard to disregard the fact that not all people are equally articulate and perceptive (e.g. project developers/verifiers vs. village chairpersons in Paper 2), which, if not handled with respect, can affect the inclusion of their viewpoints. Beyond the cross-checks between interviewees, I used member checking (Creswell 2009: 191) to determine the accuracy of the qualitative findings by taking the findings and specific descriptions back to informants. In some cases, where I was in doubt about the interpretation, I returned (in person, by email or Skype) to the interviewee to validate and elaborate on the interpretation. This explains some of the additional encounters with some respondents (see Online Resource 2:Paper 2 and 1:Paper 3). Other interviewees were a part of my daily life at the base locations, and they served as a check throughout the fieldwork by having an ongoing dialog with me regarding the interpretations of the data. Furthermore, regarding Paper 3, community meetings with villagers were held in three villages to check if the most distinct features of the perceived farming system had been understood correctly, and to obtain additional feedback from farmers about feasibility and the barriers to the suggested adjustments. Finally, I applied peer debriefing to increase accuracy, which involves locating people with extensive knowledge to assess, for example, findings and descriptions to see if they resonate with other people than the

69

researcher (Creswell 2014: 202). More precisely, I presented my work (including preliminary findings and descriptions) to about forty staff members at the Caribbean Community Climate Change Centre after the first three months (22 October 2013) and for twelve employees of the Ya’axche Conservation Trust (the largest NGO working with forest conservation and sustainable agriculture in southern Belize) at the end of the fieldwork (25 February 2014). Both events ended with a discussion, which provided useful feedback and insight, but it also increased validity by not eliciting any major concerns from the attendees. Finally, before leaving Belize (26-27 February 2014), I presented and discussed the overall findings and descriptions that I wanted to use in the papers with four key informants. They were selected because of their different backgrounds and affiliations (anthropologist at the University of Belize, head and researcher at CARDI14, head and researcher at IICA15, and CEO in the Agriculture Department), and because they had extensive knowledge concerning the research questions and all-round knowledge about Belize. They all accepted, with minor corrections, the information presented to them and added further input to the interpretation of findings. Fourthly, back home, reliability was acquired by using the same procedure for all interviews, which was why I waited to handle the interviews thoroughly until I had returned to Denmark. The reason for this emerged when I began to listen to some of the first interviews after two months of fieldwork, and noticed that I had overlooked important details and issues, which had emerged when doing the later interviews. Thus, the approach chosen was to handle the first interviews in the same way as the last ones. Whereas, starting analyzing the interviews before the last one was complete could have created a bias, for instance, for the themes that were in my head at that particular period of the fieldwork. Further, my dependence on changing conditions and contexts was recognized more fully with this approach. In writing the papers, validity was constructed by having several sources support the statements put forward and by maintaining chain of evidence throughout the research. The chain of evidence can be kept tight when it is possible to trace the conclusions throughout the process all the way back to the case study questions (Yin 2009: 123). Furthermore, findings were compared with previous research conducted in Belize and reviewed internally by co-author (Paper 2) and another scholar (Paper 3), who both had conducted research in Belize on related issues. These steps, referred to as triangulation, that is, using multiple sources (interviews, observations, document analysis) to build a coherent justification (Creswell 2009: 190; Yin 2009: 114-17), was a central aspect of the case study research design, as it provides accuracy and credibility to the conclusions that can be drawn (ibid.). The process of triangulation also involved visiting fields and markets actually to observe and validate the information given during the interviews, though this kind of triangulation is not popular in academic papers. Nevertheless, this information and other sources feed into the creation of rich, thick descriptions of the case studies and its findings, making them more realistic and adding validity to the findings (cf. Creswell 2009: 191).

14

Caribbean Agricultural Research & Development Institute

15

Inter-American Institute for Cooperation on Agriculture

70

However, the sections on research findings in the papers are based primarily on information from interviewees, whereas other information sources are used in discussing the findings themselves. One may ask how many examples should be gathered from the semi-structured interviews in order to support a claim. If only one interviewee mentioned the issue, I cross-checked the claim against others, but if that failed, I excluded the claim or brought it in emphasising that only one interviewee had mentioned this particular issue. Some information is too important to omit, however, and it could be that the interviewee is the only one who has that particular information. Nevertheless, some off the record information was obtained for Paper 2 (e.g. regarding certain land-tenure arrangements), which could not be verified whatsoever, so unfortunately it had to be omitted. Furthermore, the data were searched for counter-evidence to a claim, that is, attempts were made to find examples that contradicted claims. However, direct contradictions were rare, though different ways of seeing an issue certainly emerged (e.g. land tenure in Toledo in Paper 3). Another validity issue that can arise is external validity (i.e. generalizability), which occurs when conclusions are transferred incorrectly to other persons, settings or time periods (ibid.). This has been a major barrier to doing case studies, as critics typically state that single cases offer a weak basis for generalization (Yin 2009: 43). However, case studies are not intended to be generalized to a larger world, as they deal with analytical generalization (which differs from statistical generalization) in a specific context, where the researcher is striving to generalize a particular set of results to a broader theory. However, replicating the test of the theory in other geographical locations might establish stronger support for it (ibid.; Edmonds and Kennedy 2013: 113). One example is Paper 2, where a multiple-case research design shows similar evidence in three individual cases, making for stronger proof than if we only had investigated one case. However, Paper 3 has some external validity issues by generalizing the results to all Maya villages, which happens because the interviewees often spoke generally about the villages. Consequently, it is pointed out in the paper that the villages are comparable but still diverse. Furthermore, case studies can be collected and analyzed with regard to a particular topic, which can allow researchers to make generalizations related to a topic and/or a larger geographical area (Verburg 2014). However, this is sometimes troublesome, as studies use a huge variety of methods, which can make the results incomparable, though, extended descriptions increase transferability, as the reader has a solid frame for comparison (Creswell 2009: 200; Yin 2009). 3.4.7

Positioning, power and ethics in fieldwork

The role of the researcher has been the subject of discussion by many scholars (e.g. Oliver 2003: 3-25; Kumar 2005; Grimen and Ingstad 2007: 287-91; Briggle and Mitcham 2012: Ch. 2; Creswell 2014: 92-101+187-8), who agree that the researcher has impact on the research, as he or she engages with the people being studied as people and not simply as respondents to research instruments (Bryman 2008: 24-6). Furthermore, as human beings we are always somewhere, we perceive the world from a particular point of view, and we give expression to a certain and necessarily incomplete world view (Shweder 1991: 18-9). Consequently, I, with

71

my personality, age, gender, ethnicity, political convictions, life philosophy and educational background, influenced the approaches and findings (cf. e.g. Creswell 2009: 191), especially as all fieldwork data used in the papers was collected and analyzed by me. There could be a bias through, for example, my sympathy for an underdog group (e.g. smallholders in these papers). Furthermore, the author brings some aims, expectations, hopes and attitudes with him to the field, which influences what he sees and how he sees it, and which together with issues of trust and power influence the product. Therefore, neutrality and objectivity do not exist in social science studies, and value-free research is an illusion (Bryman 2008: 24-6). Hence, in reflecting on my fieldwork, an important question is to consider what positions I (and my interpreters) had while conducting the fieldwork, and how these positions affected my understanding of the study area. Above all, research was conducted in accordance with international ethical standards (e.g. Oliver 2003: 3-25; Kumar 2005; Grimen and Ingstad 2007: 287-91; Briggle and Mitcham 2012: Ch. 2; Creswell 2014: 92101+187-8). For instance, I had no past experience with any of the communities and interviewees that could influence the research, and the fieldwork funding bodies had no influence on the structure or content of the dissertation or the publications. Furthermore, I obtained consent from respondents prior to the interview and explained to them who I was, what their participation would involve, the nature of the research and how the data would be used and later published. Interviewees were also informed that data would be protected from any authority or NGO in Belize, which was important for the responders in the villages, as some questions were of a sensitive nature (especially in Paper 2, which discusses illegal activities like poaching and logging in protected areas). I avoided any harm or undue intrusion to respondents to the best of my abilities, as I had to consider the consequences of the respondents’ participation in the research and ensure that they would not suffer hardship because of their involvement (cf. Kvale 1996). This is also why the specific identities of respondents have been anonymized in the papers, though their names can still be identified in the raw data. All the interviews with people working in multilateral organizations, projects, NGOs, the public sector, farmers’ organizations and academia, were well-off in Belizean terms, so issues of power were minor. However, trust certainly played a role in these interviews, as it emerged that interviewees became more willing to share information during the interview or in subsequent meetings. This spoke in favor of longer interviews, where there was room for conversation on topics outside the research area as well. I also tried to use intermediaries to recommend me or mentioned some of the people I had interviewed in order to break down mistrust. In a country that is visited by few researchers and where limited research is conducted by educational institutions, some people, even the educated, have a hard time understanding what a Ph.D. student from Denmark is doing in Belize. Some thought I was working for Forest Department or an NGO to start with, which could have influenced their agenda and thus the answers they gave. However, additional explanations and handing out my business card were in most cases enough to convince people of my intensions.

72

In all the villages, I obtained permission to talk to the villagers and to visit the fields from the chairperson. The main issues in my research, my intentions in coming to the village and how the data would be handled were explained. Furthermore, villagers were informed that my activities would not be disruptive and that I was not coming with a development project or similar, which was mentioned to avoid raising any expectations. However, I often mentioned that my research would hopefully be noticed by institutions working in Belize, and in that way it might benefit the villages indirectly. The interviews in the villages and on farms had the same issues relating to trust as mentioned in the paragraph above, such as a fear that information would find its way back to public administrative offices like the local tax department. As I did not stay for long periods in the villages, because for logistical reasons I had decided to use base locations, trust had to be acquired by having a friendly attitude. This was simply done by smiling to and greeting anybody, and to engaging in small talk about anything. The topic ‘differences between Denmark and Belize’ was especially effective in generating trust, as workers from, for example, a Forest Department or NGO hardly had extensive knowledge about Denmark. In some conversations, the motives and aims of my research became secondary, as the interviewee wondered more about issues like my personal activities and hopes, a factor also noticed by Taylor and Bogdan (1984: 86). In addition, the lengths of stays in villages is two-sided, as long-term stays, with the possibility of growing into a locally understood role, both limits and enlarges the researcher’s opportunities (Cohen 1984: 222). In some villages (those further away from the major roads) my presence caused some stir, which reminded me that I was white westerner and thus by definition well-off, but through this process, people began to understand that I was not working for a development project or a government agency. After successive visits people recognized me, and my harmlessness was noticed, which increased their willingness to speak with me. However, I was still a stranger, as became evident in, for instance, an incident where I walked into the mountains behind a Maya village to track down some farmers to interview and to investigate their maize fields. Two Maya women and their children were hiding because they were scared of me, but laughed and became helpful when I explained my business for being there. On another occasion, a person in the same village had seen me together with someone (to whom I gave a lift) from the local NGO, who was disliked by some, as the NGO was trying to conserve areas they wanted to farm. He refused to talk to me because of seeing this and spread his untruthful discovery to a few others. The damage was minor, but it made me more careful about who I should be seen with, and I constantly emphasized my independence as a researcher. To indicate independence (and for family reasons), from time to time I brought my wife and children to join me in the villages, which also was a fantastic icebreaker with people. It made me appear harmless, and while I was conducting interviews around the village, the local villagers would often invite my family inside the houses (e.g. to learn to make traditional tortillas); also, whenever I returned to these locations, they instantly asked if my family was around. However, this approach should be used with care for many reasons. For instance, some villagers might feel intimidated by a whole family or the safety of family members in exposed areas.

73

3.4.8

Note on excluded survey (Paper 2)

For Paper 2, a questionnaire survey was conducted which is described below, but it was later excluded from the paper for several reasons, which will be mentioned below. It nonetheless deserves mentioning in the dissertation, as it formed a substantial part of the fieldwork. The questionnaire survey had three objectives: 1) To assess the vulnerability and adaptive capacity of the buffering communities (note: nobody lives permanently inside any of the project areas, which is why buffering communities were approached) 2) To examine how they exploit the forest and land (including agricultural pattern, crops and practices) 3) To examine the impact of the projects on communities and vice versa The structure of the questionnaire was as follows: 1) General information about the household, 2) Agricultural farming, 3) Perceptions of the climatic variability, 4) Use of forest products, 5) Impact of the project, and 6) Economy of the household. The questionnaire was developed and written in English, in collaboration with the co-author, and then translated into Spanish (see household questionnaire in Appendix 2). Before the household survey the villages were visited and the chairperson was interviewed to obtain a general understanding of the village (see village questionnaire in Appendix 3), to explain my mission and expected activities, and to ask permission to conduct fieldwork in the village. Furthermore, a pilot of eight households was made in the first village (El Progresso) to test the questionnaire design for various types of error. A specific concern was to minimize response errors by testing whether the questions were understood as intended, but the pilot also enabled unnecessary questions to be deleted. Sampling was done in the two villages nearest to the project area in the three cluster areas (projects 1, 2 and 3), as these communities were expected to affect the project areas directly and to be impacted by the projects in their turn (Table 7). A so-called two-stage sampling approach was used (Yin 2009). Villages were segregated into sections using a GPS device (Garmin Oregon 650) with aerial photos, and in each section was a random sampling approach applied (e.g. every other house). An equal number of households was sampled in each village section. The sampling strategy was to target the heads (male or female) of household, but when the household head was not present, the spouse and other family members participated, and in the case of missing data follow up visits were scheduled. Despite the fact that English is the official language and that most people understand it, was it proved difficult for some people to answer in English. Three of the six selected villages are mainly Spanish-speaking, and as my Spanish skills are limited, were different approaches used in these villages. English was well understood in Upper Barton Creek, Indian Creek and Pine Hill, so in those cases I was able to do the interviews myself. Furthermore, the conservative Mennonites, as in Upper Barton Creek and Pine Hill, are considered to be very closed communities, but I judged them to be less reluctant to talk to a white male from Europe on his own, as local people, especially officials and women, have difficulties in approaching these communities. In the case of San Carlos, as only a few interviews were needed, I conducted them using a Spanish-English interpreter.

74

Table 7. The six buffer communities in which household interviews were conducted. a measured, as the crow flies, using Google Earth from the village center to the nearest project boundary. b data from SIB 2011. Village Distance to project areaa Ethnicityb Population / householdsb Interviews conducted / % of households

Rio Bravo16 San Carlos

San Felipe

Bull Run El Progresso

10 km

7 km

4 km

Upper Barton Creek 6 km

Mestizo/His panic 95% 138 / 29

Mestizo/Hispa nic 96% 1500 / 332

Mestizo/Hispa nic 94% 482 / 96

5 HH / 17%

48 / 14%

39 HH / 41%

Boden Creek Indian Creek

Pine Hill

0 km

0 km

Mennonites 95% 80 / 12

Maya 94% 722 / 134

Mennonites 100% 206 / 39

9 HH / 75%

24 HH / 18%

7 HH / 18%

In the case of San Felipe and El Progresso most interviews were conducted in Spanish by myself using a Spanish-English interpreter plus three and two enumerators, respectively. Before using the enumerators, I trained them in the overall topic of my research with reference to how they should present the research, how to understand and phrase the questions, the units being applied, how to approach interviewees and the sampling strategy. Furthermore, on the first day in a village, they watched me doing some initial interviews, and throughout the day I observed them during the interview sessions. My abilities in Spanish were good enough to understand most of the conversation, as I knew the questions and the most common answers in Spanish. Hence, I could correct the interpreter if I thought the conversation was being translated insufficiently. The enumerators were outsiders in relation to the village in question, as other researchers at the university had informed me that villagers were often reluctant to answer questions given to them by insiders. The enumerators were college and final-year high-school students, and had been selected by teachers at their institutions because they had the skills and experience. I paid the enumerators and the interpreter a per diem (8 am to 6 pm) of 45 BZD (22.5 USD) and lunch, which was higher than the norm, as 45 BZD was standard pay for an educated person and 30-35 BZD for an uneducated person. The higher salary was explained with the long workdays on Saturdays and Sundays, and simply to make the work attractive. Weekends, especially Sundays, were preferred, as farmers were available on these days, whereas market days were avoided. The interviews started with an introduction to the survey for the interviewees: the topic, who I was, the handling of their answers, and by telling them that the chairperson had approved the survey. We were often confronted with the what’s in it for me question from interviewees, and we had to explain that no project or funds would be forthcoming because of the survey, but that we hoped that it would prove important for government institutions or donor agencies working in communities in Belize in general. This was usually accepted, but 16

Indian Creek/New Hope in Orange Walk District is the closest village to Rio Bravo, but I was not allowed to conduct

interviews there, so San Carlos was added instead. Indian Creek/New Hope it is a very closed and conservative Mennonite community not involved very much with the outside world (Roessingh et al. 2009). Their involvement in the project had previously assessed to be minimal (Salas and Castillo 2013).

75

some declined to participate. The interviews lasted for between twenty and sixty minutes, depending on the number of detours during the interview, and were conducted either in the respondent’s home or by their fields. The interview process allowed contextual information and farmers’ perceptions of specific issues touched on in the survey to be captured, enabling both quantitative and qualitative data to be collected. Answers were written on printed questionnaires, which I double-checked over lunch and after the last interview. Back at the base, the responses were typed into a premade Excel database. In total, 132 households were interviewed in six villages (Table 7). The number of households interviewed in each village depended on village heterogeneity (higher homogeneity = fewer interviews) and the size of the village (fewer inhabitants = a higher percentage of total households interviewed). A ratio of 30% for small populations (under 1000) is suggested as a rule of thumb, or higher if data are intended to cover several variables simultaneously, as in regression models (Angelsen et al. 2011). Smaller samples can be justified when the underlying population is homogeneous (ibid.), as the case in all the villages selected for this research. Data from SIB (2011) show that 94% or more of the population in these villages had the same ethnicity, which according to the literature (e.g. Wilk 1997: Maya communities; Roessingh et al. 2009: Mennonite communities; Salas and Castillo 2013: Mestizo communities) suggests a high level of homogeneity with regard to the information being sought. Excluding the survey Even though the interviews were conducted at the household level, it was the villages on the more general level I was interested in. Consequently, conducting the interviews at the household level may have been excessive, but these interviews did provide additional information (narratives) that was not asked for directly, but that came up during the interviews. However, these narratives (often the same) could also be obtained through the interviews with key informants, where information regarding agriculture, forest use and relationship to the project were also obtainable. Furthermore, the descriptive information (e.g. population, age, religion, tenure, commodities, access to electricity) at village level could be obtained from the decennial censuses, the latest dating from 2010. Hence, the survey only revealed a few additional findings compared to the information from key informants in the villages (mainly chairpersons or other village heads). Another issue was that the knowledge held by the man in the street (i.e. in the households) about the details of projects was typically limited, key informants being better informed, as they had a greater overview of the effects of the project on the village and vice versa. Moreover, doing surveys pose some challenges, one of them being asking of questions retrospectively and the implications this may have for data quality (Bergeron and Pender 1999). Particularly the questions in Part 3 of the questionnaire (Appendix 2), about perceptions of climatic variation over the past twenty years, were challenging, because weather history was difficult for respondents to recall (see also Beckett et al. 2001), or simply because interviewees were too young. Further, the questions in Part 2.2 (in the questionnaire), about

76

land ownership and tenure disputes, caused some problems. Many villagers squat on the land and have ongoing land disputes, which makes them reluctant to speak freely about this issue. Another issue concerns farm size and the amount of livestock owned (Part 2 in the questionnaire), because the figures may cause envy and begging from other villagers, and some respondents accordingly over- or understate the figures for opaque reasons. For the same reasons, and to reduce informants’ reluctance to answer, questions regarding income, remittances and yields were omitted from the questionnaire after the pilot. Respondents in some villages were also reluctant to answer questions regarding forest use (Part 4 in the questionnaire), as hunting and extracting activities are illegal in certain areas. All these issues, namely about land tenure, agriculture and forest use, were adequately covered by asking key informants (chairpersons and other official village heads), who were more keen to answer, as the questions concerned the whole community and not them personally. For instance, they talked in general terms about which villages are making encroachments on to protected areas, but without mentioning people’s names.

77

4

SUMMARY OF PAPERS

Part II of this dissertation consists of a collection of four papers (Table 8). Each paper revolves around the research topic and questions presented in Section 1, and they also draw on and provide specific new insights and perspectives to the conceptual setting presented in Section 2, which has been used to construct the analytical frameworks applied in the papers. Furthermore, the associated literature has been reviewed in order to explore the research objectives and understand the empirical findings. The specific context they are each a part of is described in the papers. Table 8. Overview and status of papers Paper

Title

Author(s)

Status (22 April 2015)

Journal

1

Addressing climate change mitigation and adaptation together: a global assessment of agriculture and forestry projects

Rico Kongsager Bruno Locatelli (CIFOR; CIRAD) Florie Chazarin (CIFOR)

In review. Submitted February 2015

Environmental Management

2

Linking adaptation and mitigation in carbon sequestration projects: evidence from Belize

Rico Kongsager Esteve Corbera (Universitat Autónoma de Barcelona)

World Development

3

Barriers to linking mitigation and adaptation in smallholder farming systems: lessons from maize cultivation among the Maya in southern Belize The carbon sequestration potential of tree crop plantations

Rico Kongsager

In review. Submitted October 2014, resubmitted after revision February 2015 In review. Submitted February 2015

4

Rico Kongsager Jonas Napier (University of Copenhagen) Ole Mertz (University of Copenhagen)

Published 2013

Mitigation and Adaptation Strategies for Global Change Mitigation and Adaptation Strategies for Global Change

All the papers are presented in the form in which they have been published or submitted for publication, hence the differences in style that reflect the requirements of the respective journals, which are peer-reviewed international journals that are multidisciplinary or topic-specific in scope. Below, each of the four papers is briefly summarized, focusing on the point of departure, approach, and main findings. The logic behind the order in which the papers appear in the Part II is based on scale, going from the global to local levels: Paper 1 (global/tropics), Paper 2 (country), Paper 3 (district), and Paper 4 (an agricultural system).

4.1

Paper 1

Kongsager, R., Locatelli, B., and Chazarin, F.: Addressing climate change mitigation and adaptation together: a global assessment of agriculture and forestry projects This paper was triggered by the current gap in the availability of meta-analyses and the need to obtain an overview and status of the situation in linking adaptation and mitigation in the agriculture and forestry sectors. Prior studies had focused on a few projects in particular regions, but we wanted to conduct a more comprehensive study across the tropics to obtain the big picture. It was expected that adaptation and mitigation

78

would be highly separated for historical reasons, as this has influenced the scope and modalities in the funding mechanisms for land-use projects. The two climate change objectives, adaptation and mitigation, have been and to a certain extent still are considered separately in policies and institutions because of their different objectives and characteristics, which was expected to influence the projects. Thus, climate change projects in forestry and agriculture were assessed to investigate if and how they integrate adaptation and mitigation. We analyzed 201 project documents from adaptation funds (e.g., the UNFCCC Adaptation Fund), adaptation plans (e.g., National Adaptation Programmes of Action), mitigation instruments (e.g., the Clean Development Mechanism), and project standards (e.g., Climate Community & Biodiversity (CCB)). We particularly scrutinized whether projects established for one objective contributed explicitly to the other (i.e. whether mitigation projects contributed to adaptation and vice versa). More importantly, we examined 36 aspects to assess whether their activities or expected outcomes allowed for potential contributions to both objectives. It was found that, despite the institutional separation between them, more than a third of the project documents explicitly mention a contribution to the other objective, although only half of those substantiated it by having activities to support these statements. However, more interestingly, most projects showed the potential, at least in some aspects, to contribute to the other objective. Thus, the potential exists for projects to harness synergies and avoid trade-offs between adaptation and mitigation, but most of the current funding mechanisms neglect to provide the incentives for project developers to consider the incorporation of both objectives. Projects are often narrowly focused on their primary aim, which here is adaptation or mitigation. However, some of the mechanisms have a broader focus, which, beyond adaptation and mitigation, also have substantial development benefits. Formerly published studies on the topic are used to reflect on the findings.

4.2

Paper 2

Kongsager, R. and Corbera, E.: Linking adaptation and mitigation in carbon sequestration projects: evidence from Belize The point of departure for this paper is the findings in Paper 1, which provide an overview of the current status of linking adaptation and mitigation in selected projects, but without delivering details beyond the project documents. Therefore, we wanted to dig deeper into some of the projects from that paper, and by zooming in on fewer projects, the intention was to increase the level of detail, which was assessed to be achievable using a qualitative approach. A case study design is used to compare three avoided deforestation projects in Belize. In this paper, a multi-dimensional analytical framework, which is a refined version of the one applied in Paper 1, was developed to explore the extent to which carbon forestry initiatives integrate adaptation concerns. Before this the overall context of carbon sequestration projects in the forestry sector is outlined. The analysis is mainly based on semi-structured interviews with individuals directly or indirectly linked to the projects, as well as government officials or carbon standard managers. We demonstrate that the rhetoric of linking adaptation and mitigation is not easy to retain because the existing mechanisms to sell carbon offsets are weakly enforced, and their mandate does not incorporate adaptation concerns. Furthermore, project developers

79

lack the resources and capacities to deal with both dimensions, and have the perception that there are few incentives – or even disincentives – to pursue a stronger link to adaptation. The mitigation part of the projects is concentrated on avoided deforestation without any additional activities to enhance biomass stocks, the main shortcoming being the questionable additionality in all projects. Therefore, the benefits of the projects are mostly non-carbon benefits, such as biodiversity and watershed protection, but to a lesser extent to climate mitigation – and adaptation. However, avoided deforestation per se is argued to contribute valuable ecological adaptation benefits. We conclude by emphasizing that the integration of adaptation and mitigation in Belize’s carbon offset projects remains a laudable but still elusive goal, but at the same time we specify activities and implementation methods to enhance the adaptation-mitigation link in the attempt to achieve synergies and circumvent negative impacts. The findings are discussed in relation to former studies of forest projects, especially those focused on social benefits, participation, land grapping, and the commodification of ecosystem services.

4.3

Paper 3

Kongsager, R.: Barriers to linking mitigation and adaptation in smallholder farming systems: lessons from maize cultivation among the Maya in southern Belize The point of departure here was the growing interest in the research communities for climate-smart agriculture, which is heavily related to the overall research topic. Interest was also triggered by the limited number of agricultural projects with mitigation aims, which was one of the findings of Paper 1. Thus, the paper explores the opportunities for designing certain agricultural activities that can achieve joint adaptation and mitigation benefits while avoiding substantial trade-offs, and without making radical changes to the current system. A case-study design is used, and the approach is qualitative, with semi-structured interviews and field observations. First, the paper zooms in on the agricultural methods applied by one ethnic group, the Maya Indians, in one particular district in Belize. Secondly, four extensive low-input adjustments to maize production are identified with the aim of obtaining both adaptation and mitigation benefits. Thirdly, the study investigates the barriers to making these adjustments. The paper finds that adjustments to enhance the link between adaptation and mitigation in small-scale maize production in Maya villages in southern Belize are possible, but that several barriers can make the overall climate-smart objective difficult to achieve in practice. The barriers are of a proximate and indirect nature, exist at different spatial scales and involve various levels of governance. It is furthermore shown that these barriers are not homogenous across the villages in the region. For instance, some communities have adopted strategies that will most likely make them less vulnerable to climatic change, as well as to other changes that are likely to happen. These findings indicate that an overall district-level strategy that provides both adaptation and mitigation benefits will need a wide variety of approaches if it is to address the different communities properly. The study also highlights that changes are a necessity at various administrative scales, as well as culturally in the Maya communities, if changes are to be made to a system that is currently poorly adapted and causes deforestation. Discussion of the findings is

80

centered around the research on the adoption of agricultural practices in similar contexts, but it also draws on earlier findings on Maya culture by anthropologists and geographers.

4.4

Paper 4

Kongsager, R., Mertz, O., and Napier, J.: The carbon sequestration potential of tree crop plantations This paper, which, as mentioned earlier, was actually the motivating force behind the overall research topic, deals with the potential of a specific agricultural system to partake in mitigation payment schemes, thereby providing adaptation benefits in the shape of payments for ecosystem services. The point of departure was a consideration of the absence of tree crop plantations in mitigation projects, when it was possible to have forest plantations, established as afforestation and reforestation, as a part of a national carbon budget or payment scheme. Thus, the first question was whether tree crop plantations have the potential to sequestrate carbon, and whether this could be measured with simple methods by locals. This would make it possible for these plantations to partake in larger mitigation schemes with payments involved, as this would reduce transaction costs. The economic benefits were judged to be necessary if farmers in developing countries are to assist in the sequestrating of carbon as a long-term strategy. Furthermore, tree crop plantations can also couple adaptation and mitigation beyond payments, as they provide both wood and non-wood products with the potential to increase the resilience that is often claimed to be necessary in order to adapt to unavoidable climate variability and change. The measurements for this case study were conducted in Ghana, and it was estimated that the carbon sequestration potential of four major plantation crops cultivated in the tropics (cocoa, oil palm, rubber, and orange) was considerable. However, this is only the case if they are established on land with modest carbon content such as degraded forest or agricultural land, and do not cause deforestation of primary forest, as the latter would result in significant carbon losses. The latter would also impact on other ecosystem services such as biodiversity, erosion control and water supply negatively, which will most likely exclude tree crop plantations from partaking in payment schemes such as REDD+. Lastly, the paper demonstrates that simple carbon assessment methods can provide reliable results, which makes it easier for developing countries to partake, as measurement costs can be kept down.

81

5

CONCLUSION

This dissertation has set out to explore the linking of adaptation and mitigation in the agricultural and forestry sectors by applying a range of methods suitable for dealing with the complex nature of the questions under investigation. The methods adopted in the papers operate at multiple scales, use multiple data sources, and represent a mixture of qualitative and quantitative data. This has made it possible to generate four diverse papers, though still under the same umbrella theme of linking adaptation and mitigation in agriculture and forestry in the tropics. The first section returns to the research objectives outlined in Section 1.1, which will be answered by drawing on the findings from Section 2 and the papers. This is followed by suggesting policy implications and perspectives for future research.

5.1

Returning to the research objectives

The linking of adaptation and mitigation is not a new phenomenon, but important research gaps and limited empirical evidence make it difficult to examine how the linking of adaptation and mitigation can move from conceptual thinking to practice. The same, limited material also applies to the conceptual frameworks developed in this field, at least in terms of concepts and frameworks that might assist in analyzing synergies and trade-offs between adaptation and mitigation in projects. This dissertation makes an attempt to enhance the link between adaptation and mitigation by developing a framework (Figure 8) in which climate measures start out separately (adaptation or mitigation), but as time and scope evolve, complementarities and synergies are constructed between the measures that eventually attain the ultimate aim of climate-compatible development. The developed framework acknowledges that some measures will remain separate or constitute trade-offs and never reach that aim. However, separate measures and trade-offs, of which only the last is a real problem, will most likely decrease in number and impact as approaches to establish synergies and complementarities evolve. Another important issue highlighted in this dissertation relates to the history of how adaptation and mitigation have been perceived, which was explored to provide an understanding of how the current dichotomy between adaptation and mitigation originated, which is of importance in realizing where overlaps are feasible, or alternatively considered to be cumbersome. The overlaps have until recently been perceived as limited, mainly due to differences in spatial and temporal scopes, but also because of institutional, administrative and sectoral arrangements. However, scholars have lately stressed that the divide is not substantial enough to segregate adaptation and mitigation as sharply as before. Overlaps are present, spatially and temporally, especially in some sectors, such as land use, where synergies and complementarities from the literature demonstrate that the agriculture and forestry sectors contain enormous potential for linking adaptation and mitigation, but also have crucial trade-offs that have to be dealt with. The latter was is the visited literature found more substantial in the forestry sector than the agricultural sector, whereas something in between, like agroforestry, was

82

overwhelmingly emphasized in the literature as being a positive approach to the linking of adaptation and mitigation. The extent of linking adaptation and mitigation in current agriculture and forestry projects was examined globally. The first discovery was that projects primarily focus on one objective – either adaptation or mitigation. Not one of the 201 project documents analyzed had considerable adaptation and mitigation elements both imbedded in the project design. However, this is not surprising, as projects originate in the adaptation-mitigation dichotomy just mentioned, and are therefore driven by the specific interest of the international donors and national institutions that construct the frames in which these projects are rooted. Project developers simply follow the guidelines and requirements to achieve funding and support – nothing more or less. However, by strictly complying with institutional standards, the projects in the papers were revealed as potentially creating trade-offs between adaptation and mitigation. For instance, giving only limited attention to ecological adaptation can affect the mitigation effects of a forest project if hurricanes or drought eventually strike the area. These effects can be handled with complementary adaptation tools, such as fire management, thinning, or planting activities, and even with synergistic approaches, such as fast-growing climate-tolerant species, which, beyond increasing resilience, would also increase the mitigation effect. Moreover, the analysis showed that many projects have the potential to contribute to the other objective (i.e. mitigation contributing to adaptation, or vice versa). In particular, mitigation projects demonstrated the integration of adaptation aspects, which can partly be explained by the circumstance that adaptation can deliver tangible effects in mitigation projects. Adaptation projects, conversely, have less to achieve by imbedding mitigation. Nevertheless, this is possible by, for instance, paying for carbon uptake, which can be used for adaptation activities, or directly planting trees to sequester carbon (mitigation), which will simultaneously protect coastal shores, crops, or soils (adaptation). Nonetheless, the current institutional setting on land-use projects does not create incentives to search for synergies and/or to avoid trade-offs between adaptation and mitigation. And what might be worse is that many of these projects have an extended duration, up to a century for some of the forest projects, without considering the impacts of climate changes. For instance, the climate impacts on the selected species planted in a reforestation project are often not assessed. The international funds and standards to which the projects are attached can assist the latter in pursuing synergies and harnessing trade-offs between adaptation and mitigation by providing information and technical assistance. Furthermore, assessing the benefits integrating the other objective can be made mandatory. Such an approach would not force a marriage between adaptation and mitigation in the projects, but it can improve the overall efficiency of climate funding by delivering joint benefits at local and global levels, as well as assist avoiding trade-offs, without increasing project complexity and costs significantly.

83

Moving more locally, more exactly to Belize, the dissertation shows similar results. Adaptation is ignored in the three mitigation projects investigated, the effect of which on livelihoods is mainly neutral. In addition, the projects can be contested by neighboring communities because of property claims and food security concerns. It is argued that the structure of the global carbon markets and the provisions of the applied standards exclusively focus on the mitigation aspects and largely ignore the adaptation dimensions in certified forest carbon sequestration projects. Furthermore, project developers have few incentives, but also lack the resources and capacities, to address adaptation in these mitigation projects. However, ecological adaptation benefits are delivered, which are implicitly imbedded in this forest project type (avoided deforestation). Several suggestions are made, which are related directly to these projects in respect of how they can enhance adaptation while simultaneously benefitting mitigation. One such tactic would be to involve the surrounding communities actively in the projects already in the initial phase to allow them to take some ownership of the projects, potentially decreasing encroachments into the protected areas. Another very important issue explored is the weak additionality in the projects, which casts doubt on their actual contribution to mitigation. Hence, climate projects that mostly provide non-carbon benefits, such as biodiversity, watershed protection and limited climate adaptation, are exceedingly problematic. Moving away from the project level and the attached provisions to the agricultural setting of one ethnic group (Maya Indians) in one district (Toledo in southern Belize). This is a case of how adjustments to enhance the integration of adaptation and mitigation will potentially encounter considerable barriers. The suggested activities are likely to provide the desired climate-smart benefits, but different barriers, such as the structure of the current land-tenure system, provide less incentive to adopt these practices. The paper reveals a high degree of complexity for making adjustments, which are not limited to climate-related adjustments, but to changes in the agricultural system in general. The stakeholders have different agendas concerning this region, which are often not straightforward or disclosed. It is emphasized that small-scale farming systems are highly complex and that, if climate-smart measures are to be rolled out, several approaches are needed to account for household and village diversity. One important issue mentioned is the irrational way in which these farmers may take decisions, at least from an economic point of view. For instance, the suggested changes would potentially deliver higher yields and other benefits for the farmer, as well as spare land as a bonus to the global climate, but this is not necessarily adopted, as farmers’ choice may be guided by other drivers such as religion or family conflicts. It is also shown that some agricultural systems in the tropics can actually deliver considerable mitigation benefits, as is proved by estimating the potential carbon uptake in tree crop plantations at one location in Ghana. With these mitigation benefits, the system has the possibility of dual-purpose use, as it mitigates climate change while increasing local incomes and thereby strengthening resilience locally. However, the question is asked whether tree crop plantations will be tolerated in the possible future payment scheme, as this can lead to perverse incentives and have adverse impacts.

84

5.2

Implications for policy

This dissertation documents the need to create incentives at the institutional level in the agriculture and forestry sectors for improving the linking of adaptation and mitigation in the implementation of climate-related projects. The dichotomy between adaptation and mitigation, which is mainly a politically and scientifically constructed one, is found to be prevalent. Thus, even though the focus has changed, and interlinkages between adaptation and mitigation have started to gain attention, this dissertation shows that a lot more has to be done both internationally and at the project level. However, institutions may be somewhat intimidated in linking adaptation and mitigation, as it is far from easy to manage. It can be compared to a hedgehog, as it sticks out (in many directions), moves slowly and only in the dark (lack of frameworks and clear concepts), which is why it can easily be run over (by separate measures or ‘doing nothing’). Nevertheless, by making changes that prioritize this integration, the institutions involved can create the potential to pursue the advantages highlighted in this dissertation. On the other hand, if incentives are not put forward politically and financially, this can create inefficiency, and we may find ourselves following approaches that solve one issue but exacerbate others. However, politicians also have a responsibility for creating systems that are manageable in practice. And linking adaptation and mitigation should only be attempted in cases with spatial, temporal, and sectoral overlaps, and in activities that have the potential to develop complementarity and synergy. Some climate measures have no complementarity or synergy, maybe even trade-offs, but if they are essential, they have to be implemented nevertheless, and disfavoring these measures in the pursuit of linking adaptation and mitigation should be avoided. Moreover, since cost-efficiency is still crucial in making them attractive to stakeholders and funding agencies, then shall all activities in climate measures have to contribute substantially to an objective (adaptation or mitigation) – and they shall consequently not just be far-fetched activities making an unimportant link between adaptation and mitigation. The dissertation also finds that the current carbon market approach, based on projects, standards, certifiers, developers and auditors, is unlikely to deliver both mitigation and adaptation, or at best only tangentially. The dissertation puts forward a number of suggestions for a change of approach by shifting the focus on to policy programs that support fair tenure regimes and promoting sustainable resource management practices in agriculture and forestry, thus effectively and legitimately restricting unsustainable resource use with the support of local communities. Formalizing customary resource rights and usufruct rights, for instance, can form synergies between adaptation and mitigation, as insecure property rights can influence people’s adaptive capacity and can indirectly cause deforestation. Improving national policies can remedy some of the perverse incentives, while policies that incentivize the sustainable management of forest and agriculture can be adopted instead, which in the long run would benefit both locals and the national economy. Furthermore, politicians can establish structures for community-based natural resource management, which has been an effective method for managing resources sustainably in many forest areas. This can be done by

85

empowering local forest-user groups and agrarian communities, instead of selling the land to foreigners that might establish forest projects that exclude the locals, giving rise to conflicts. Thus, the focus on single projects and carbon trading seems to be diverting attention from the environmental, socio-economic and cultural dynamics in which forests and rural people are involved. International and national policy frameworks for climate change mitigation should therefore refrain from providing continuous support for narrowly designed carbon projects and promote more holistic and territorially based approaches focusing on the well-being of rural populations instead. However, for these initiatives to have an impact, it is necessary to mainstream awareness about adaptation needs and mitigation potentials, including their various connections, at the various administrative levels. This can remove barriers and ultimately result in more efficient climate policies both nationally and locally, as well as being an important step away from projects that have limited effect globally.

5.3

Perspectives for future research

There is still scope for future research to further develop concepts and frameworks that are operational and that can investigate and evaluate the different connections that exist between adaptation and mitigation, principally in the areas with the greatest overlaps between the need for adaptation and the potential for mitigation. The potential is there; but improved documentation on where it lies and which tools that can be used to exploit it is needed. No great innovations are needed, as the technologies and knowledge presumably already exist, but consideration of how to combine these technologies and knowledge is required if adaptation and mitigation are to be successfully linked. Cross-disciplinary research would have an important role to play, as knowledge across a vast variety of academic disciplines is needed. More specifically, better knowledge on the government mechanisms that are most effective in fostering the linking between adaptation and mitigation at multiple levels is necessary. It is also crucial to undertake more empirical studies to establish a greater understanding of the many practices and processes involved because their interactions and outcomes are often not predictable, making it difficult to determine the possibilities of synergies and to avoid trade-offs. Furthermore, studies that demonstrate robust criteria and indicators to evaluate multiple gains are needed, together with case studies that show where and how these gains and synergies are to be obtained. Moreover, studies of different payment configurations would also be highly relevant, as the identification of effective methods to distribute payments for ecosystem and other services remains open to research. Without these studies, there is a risk of implementing projects with imbedded inefficiencies and high transaction costs.

(Adger et al. 2005)

86

REFERENCES Adger WN, Arnell NW, Tompkins EL (2005) Successful adaptation to climate change across scales. Glob Environ Chang Policy Dimens 15:77–86. Adger WN, Dessai S, Goulden M, et al. (2009) Are there social limits to adaptation to climate change? Clim Change 93:335–354. Adger WN, Paavola J, Huq S, Mace MJ (2006) Fairness in Adaptation to Climate Change. MIT Press, Cambridge, Massachusetts Angelsen A, Jagger P, Babigumira R, et al. (2014) Environmental Income and Rural Livelihoods: A GlobalComparative Analysis. World Dev. doi: 10.1016/j.worlddev.2014.03.006 Angelsen A, Larsen HO, Lund JF, et al. (2011) Measuring livelihoods and environmental dependence: Methods for research and fieldwork. Earthscan Ayers JM, Huq S (2009) The value of linking mitigation and adaptation: a case study of Bangladesh. Environ Manage 43:753–64. doi: 10.1007/s00267-008-9223-2 Bajželj B, Richards KS, Allwood JM, et al. (2014) Importance of food-demand management for climate mitigation. Nat Clim Chang 4:924–929. doi: 10.1038/nclimate2353 Banerjee SB (2012) A Climate for Change? Critical Reflections on the Durban United Nations Climate Change Conference. Organ Stud 33:1761–1786. doi: 10.1177/0170840612464609 Bateman IJ, Carson RT, Day B, et al. (2002) Economic Valuation with Stated Preference Techniques. A Manual. Edward Elgar Publictaions Beckett M, Da Vanzo J, Sastry N, et al. (2001) The quality of retrospective data: an examination of longterm recall in a developing country. J Hum Resour 36:593–625. Bergeron G, Pender J (1999) Determinants of land use change: evidence from a community study in Honduras. EPTD Discussion paper no. 46. Washington, D.C. Berman J (2011) Moon Belize. Avalon Travel Publishing. 9th Edition. Bernsten RH, Herdt RW (1977) Towards An Understanding of Milpa Agriculture - Belize Case. J Dev Areas 11:373–392. Biesbroek GR, Swart RJ, van der Knaap WGM (2009) The mitigation–adaptation dichotomy and the role of spatial planning. Habitat Int 33:230–237. doi: 10.1016/j.habitatint.2008.10.001 Bizikova L, Robinson J, Cohen S (2007) Linking climate change and sustainable development at the local level. Clim Policy 7:271–7. Bray DB (2010) Forest cover dynamics and forest transitions in Mexico and Central America: towards a “great restoration”? In: Nagendra H, Southworth J (eds) Reforesting Landscapes Link. Pattern Process. Springer, Berlin, pp 85–120 Briggle A, Mitcham C (2012) Ethical concepts and theories. In: Briggle A, Mitcham C (eds) Ethics Sci. An Introd. Cambridge University Press, Cambridge, UK, pp 23–61

87

Bryman A (2008) Social research strategies. Soc. Res. methods. Oxford University Press, pp 3–29 Buchner B, Falconer A, Herve-Mignucci M, Trabacchi C (2012) The Landscape of Climate Finance 2012. Climate Policy Initiative Buchner B, Stadelmann M, Wilkinson J, et al. (2014) The Global Landscape of Climate Finance 2014. Climate Policy Initiative. Burton I (2009) Climate Change and Adaptation Deficit. In: Schipper ELF, Burton I (eds) Adapt. to Clim. Chnage. Earthscan, pp 89–99 Burton I, Bizikova L, Dickinson T, Howard Y (2007) Integrating adaptation into policy: upscaling evidence from local to global. Clim Policy 7:371–376. Börner J, Wunder S (2012) The Scope for Reducing Emissions from Forestry and Agriculture in the Brazilian Amazon. Forests 3:546–572. Cherrington E, Cho PP, Waight I, et al. (2012) Forest Cover and Deforestation in Belize, 2010-2012. 2010– 2012. Cherrington E, Ek E, Cho P, Howell BF (2010) Forest Cover and Deforestation in Belize: 1980-2010. SERVIR, Panama City, Panama Chomitz K, Gray D (1996) Roads, land use, and deforestation: A spatial model applied to belize. World Bank Econ Rev 10:487–512. CIA (2015) The world factbook - Belize. https://www.cia.gov/library/publications/the-worldfactbook/geos/bh.html. Central Intellegence Agency (CIA). Accessed 15 February 2015. Cisco (2015) Southern Highway (Mile 14) - Belize/Guatemala Border Road Project. http://cisco.com.bz/placencia-road-upgrading-2/. Accessed 10 Feb 2015 Clark C (2000) State leasehold and Mayan customary cultivation rights in Belize. Soc Nat Resour 13:15–32. Cline WR (2007) Global Warming and Agriculture: Impact Estimates by Country. Center for Global Development, Washington DC, USA Cohen AP (1984) Participant Observation. In: Ellen RF (ed) Ethnogr. Res. A Guid. to Genral Conduct. Academic Press, London, UK, pp 216–229 Corbera E, Kosoy N, Tuna MM (2007) Equity implications of marketing ecosystem services in protected areas and rural communities: Case studies from Meso-America. Glob Environ Chang Policy Dimens 17:365–380. doi: 10.1016/j.gloenvcha.2006.12.005 Corning PA (1998) “The synergism hypothesis”: On the concept of synergy and its role in the evolution of complex systems. J Soc Evol Syst 21:133–172. doi: 10.1016/S1061-7361(00)80003-X Couvet D (2012) Migrations Between Villages: Incidents or Significant Drivers of Swidden Agriculture Changes? Ecol Soc 17:1133755. Creswell JW (2009) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. Sage Publications. 3th Edition.

88

Creswell JW (2014) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. Sage Publications. 4th Edition. Creswell JW, Plano Clack VL (2011) Designing and Conducting Mixed Methods Research. Sage Publications. 2nd Edition. Cutright NJ (1996) Joint implementation: Biodiversity and greenhouse gas offsets. Environ Manage 20:913– 918. Dang H, Michaelowa A, Tuan DD (2003) Synergy of adaptation and mitigation strategies in the context of sustainable development: the case of Vietnam. Clim Policy 3:S81–S96. doi: 10.1016/j.clipol.2003.10.006 Denton F, Wilbanks TJ (2014) Climate-resilient pathways: adaptation, mitigation, and sustainable development. In: Field CB, Barros VR, Dokken DJ, et al. (eds) Clim. Chang. 2014 Impacts, Adapt. Vulnerability. Part A Glob. Sect. Asp. Contrib. Work. Gr. II to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 1101– 1131 Di Fiore SL (2002) Remote Sensing and Exploratory Data Analysis as Tools to Rapidly Evaluate Forest Cover Change and Set Conservation Priorities along the Belize River, Belize. MA thesis. Columbia University, New York, NY Diemont SAW, Bohn JL, Rayome DD, et al. (2011) Comparisons of Mayan forest management, restoration, and conservation. For Ecol Manage 261:1696–1705. doi: 10.1016/j.foreco.2010.11.006 Doré T, Makowski D, Malézieux E, et al. (2011) Facing up to the paradigm of ecological intensification in agronomy: Revisiting methods, concepts and knowledge. Eur J Agron 34:197–210. doi: 10.1016/j.eja.2011.02.006 Downey SS (2010) Can properties of labor-exchange networks explain the resilience of swidden agriculture. Ecol Soc 15:15. Duguma LA, Minang PA, van Noordwijk M (2014a) Climate change mitigation and adaptation in the land use sector: from complementarity to synergy. Environ Manage 54:420–32. doi: 10.1007/s00267-0140331-x Duguma LA, Wambugu SW, Minang PA, van Noordwijk M (2014b) A systematic analysis of enabling conditions for synergy between climate change mitigation and adaptation measures in developing countries. Environ Sci Policy 42:138–148. doi: 10.1016/j.envsci.2014.06.003 Edmonds AW, Kennedy TD (2013) An Applied Reference Guide to Research Designs: Quantitative, Qualitative, and Mixed Methods. Sage Publications, Inc Ellis E (2012) The GLOBE project: accelerating global synthesis of local studies in land change science. Newsletter of the Global Land Project No. 8. Emch M (2003) The human ecology of Mayan cacao farming in Belize. Hum Ecol 31:111–131.

89

Emch M, Quinn JW, Peterson M, Alexander M (2005) Forest cover change in the Toledo District, Belize from 1975 to 1999: A remote sensing approach. Prof Geogr 57:256–267. Fairweather PN, Gray DA (1994) The Land Use of Belize: 1989/92. Land Information Centre, Ministry of Natural Resources. Belmopan, Belize FAO (2010a) “Climate-Smart” Agriculture: Policies, Practices and Financing for Food Security, Adaptation and Mitigation. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. Rome, Italy FAO (2013) Climate-Smart Agriculture Sourcebook. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. Rome FAO (2008a) Investing in Sustainable Agricultural Intensification: The Role of Conservation Agriculture. A Framework for Action. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. Rome FAO (2008b) Conservation Agriculture: Looking beneath the Surface. Nakuru, Kenya. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. Rome FAO (2009a) Harvesting agriculture’s multiple benefits: Mitigation, adaptation, development and food security. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. FAO (2009b) Food Security and Agricultural Mitigation in Developing Countries: Options for Capturing Synergies. FAO - Food and Agriculture Organization of the United Nations, Rome, Italy. FAO (2010b) Global Forest Resources Assessment 2010, Country Report Belize. Rome Fischer J, Hanspach J, Hartel T (2011) Continental-scale ecology versus landscape-scale case studies. Front Ecol Environ 9:430. Fisher M, Chaudhury M, McCusker B (2010) Do Forests Help Rural Households Adapt to Climate Variability? Evidence from Southern Malawi. World Dev 38:1241–1250. Fleurbaey M, Kartha S (2014) Sustainable Development and Equity. In: Edenhofer O, Pichs-Madruga R, Sokona Y (eds) Clim. Chang. 2014 Mitig. Clim. Chang. Contrib. Work. Gr. III to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, United Kingdom and New York, NY, USA., p 68 Flyvbjerg B (2006) Five misunderstandings about case-study research. Qual Inq 12:219–245. Gardiner SM (2006) A Core Precautionary Principle. J Polit Philos 14:33–60. doi: 10.1111/j.14679760.2006.00237.x Garnett T, Appleby M, Balmford A (2013) Sustainable intensification in agriculture: premises and policies. Science (80- ) 341:33–34. Garvey J (2008) The Ethics of Climate Change: Right and Wrong in a Warming World. Kindle version. Continuum. London, UK

90

Glantz MH (1995) Assessing the impacts of climate: the issue of winners and losers in a global climate change context. In: Zwerver S, van Rompaey RSAR, Kok MTJ, Berk MM (eds) Clim. Chang. Res. Eval. Policy Implic. pp 41–54 Godfray HCJ, Garnett T (2014) Food security and sustainable intensification. Philos Trans R Soc Lond B Biol Sci 369:20120273. doi: 10.1098/rstb.2012.0273 Goklany I (2007) Integrated strategies to reduce vulnerability and advance adaptation, mitigation, and sustainable development. Mitig Adapt Strateg Glob Chang 12:755–786. Government of Belize (2011) Second National Communication (Belize) to the United Nations Framework on Climate Change. Grimen H, Ingstad B (2007) Qualitative research. Chapter 10. In: Laake P, Benestad HB, Olsen BR (eds) Res. Methodol. Med. Biol. Sci. Academic Press, London, UK, pp 281–310 Guariguata MR, Cornelius JP, Locatelli B, et al. (2008) Mitigation needs adaptation: Tropical forestry and climate change. Mitig Adapt Strateg Glob Chang 13:793–808. doi: 10.1007/s11027-007-9141-2 Halsnæs K, Verhagen J (2007) Development based climate change adaptation and mitigation—conceptual issues and lessons learned in studies in developing countries. Mitig Adapt Strateg Glob Chang 12:665– 684. doi: 10.1007/s11027-007-9093-6 Harry S, Morad M (2013) Sustainable development and climate change: Beyond mitigation and adaptation. Local Econ 28:358–368. doi: 10.1177/0269094213476663 Hartup BK (1994) Community Conservation in Belize - Demography, Resource Use, and Attitudes of Participating Landowners. Biol Conserv 69:235–241. Hayward T (2012) Climate change and ethics. Nat Clim Chang 2:843–848. doi: 10.1038/nclimate1615 Henry M, Tittonell P, Manlay RJ, et al. (2009) Biodiversity, carbon stocks and sequestration potential in aboveground biomass in smallholder farming systems of western Kenya. Agric Ecosyst Environ 129:238–252. doi: 10.1016/j.agee.2008.09.006 Hertel TW, Burke MB, Lobell DB (2010) The poverty implications of climate-induced crop yield changes by 2030. Glob Environ Chang 20:577–585. Horwich R, Lyon J (1990) A Belizean rain forest: The community baboon sanctuary. Orang-utan Press Houghton RA, Werf GR van der, DeFries RS, et al. (2012) Chapter G2 Carbon emissions from land use and land-cover change. Biogeosciences Discuss 9:835–878. Huq S, Grubb M (2004) Scientific assessment of the inter-relationships of mitigation and adaptation. IPCC Expert Meet. Geneva Huq S, Grubb M (2007) Preface. Mitig Adapt Strateg Glob Chang 12:645–649. Huq S, Reid H (2007) Community based adaptation: an IIED briefing. International Institute of Environment and Development, London.

91

ICRAF (2014) Climate-Smart Landscapes: Multifunctionality in Practice. World Agroforestry Centre (ICRAF). Nairobi, Kenya Illman J, Halonen M, Rinne P, et al. (2013) Scoping study on financing adaptation-mitigation synergy activities. Nordic Working Papers. Nordic Council of Ministers. IPCC (2001) Climate Change 2001: Synthesis Report. Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Cambridge, UK IPCC (2014a) Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva, Switzerland IPCC (2014b) Annex II: Glossary. In: Barros VR, Field CB, Dokken DJ, et al. (eds) Clim. Chang. 2014 Impacts, Adapt. Vulnerability. Part B Reg. Asp. Contrib. Work. Gr. II to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 1757–1776 IPCC (1995) Second Assessment Climate Change. Intergovernmental Panel on Climate Change. IPCC (2013) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia,. 1535 pp. Jarvis A, Lau C, Cook S, et al. (2011) An Integrated Adaptation and Mitigation Framework for Developing Agricultural Research: Synergies and Trade-Offs. Exp Agric 47:185–203. Jerneck A, Olsson L (2013) More than trees! Understanding the agroforestry adoption gap in subsistence agriculture: Insights from narrative walks in Kenya. J Rural Stud 32:114–125. doi: 10.1016/j.jrurstud.2013.04.004 Jones RN, Dettmann P, Park G, et al. (2007) The relationship between adaptation and mitigation in managing climate change risks: a regional response from North Central Victoria, Australia. Mitig Adapt Strateg Glob Chang 12:685–712. Jordan R, Müller R, Uodes A (2009) High Sequestration, Low Emission, Food Secure Farming. Organic Agriculture – a Guide to Climate Change and Food Security. IFOAM and IFOAM EU Group, IFOAM and IFOAM EU Group 2009, D-Bonn and BE-Bruxelles Kane S, Shogren JF (2000) Linking adaptation and mitigation in climate change policy. Clim Change 45:75– 102. doi: 10.1023/A:1005688900676 Kates R, Wilbanks T (2003) Making the Global Local Responding to Climate Change Concerns from the Ground. Environ Sci Policy Sustain Dev 45:12–23. Keenan RJ (2015) Climate change impacts and adaptation in forest management: a review. Ann For Sci 72:145–167. doi: 10.1007/s13595-014-0446-5 Klein RJT, Huq S, Denton F, et al. (2007) Inter-relationships between adaptation and mitigation. Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth

92

Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK Klein RJT, Midgley GF, Preston BL, et al. (2014) Adaptation opportunities, constraints, and limits. In: Field CB, Barros VR, Dokken DJ, et al. (eds) Clim. Chang. 2014 Impacts, Adapt. Vulnerability. Part A Glob. Sect. Asp. Contrib. Work. Gr. II to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 899–943 Klein RJT, Schipper ELF, Dessai S (2005) Integrating mitigation and adaptation into climate and development policy: three research questions. Environ Sci Policy 8:579–588. doi: 10.1016/j.envsci.2005.06.010 Klein RJT, Schipper LE, Dessai S (2003) Integrating mitigation and adaptation into climate and development policy: three research questions, Tyndall Centre Working Paper No. 40. Klinsky S, Dowlatabadi H, McDaniels T (2012) Comparing public rationales for justice trade-offs in mitigation and adaptation climate policy dilemmas. Glob Environ Chang 22:862–876. doi: 10.1016/j.gloenvcha.2012.05.008 Knowler D, Bradshaw B (2007) Farmers’ adoption of conservation agriculture: A review and synthesis of recent research. Food Policy 32:25–48. doi: 10.1016/j.foodpol.2006.01.003 Knox J, Hess T, Daccache A, Wheeler T (2012) Climate change impacts on crop productivity in Africa and South Asia. Environ Res Lett 7:034032. doi: 10.1088/1748-9326/7/3/034032 Kolstad C, Urama K (2014) Social, Economic, and Ethical Concepts and Methods. In: Edenhofer O, PichsMadruga R, Sokona Y (eds) Clim. Chang. 2014 Mitig. Clim. Chang. Contrib. Work. Gr. III to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, Cambridge, United Kingdom, and New York, NY, USA, p 76 Kotschi J, Müller-Sämann K (2004) The role of organic agriculture in mitigating climate change. Kumar R (2005) Research Methodology: A Step-by-Step Guide for Beginners. Sage Publications. Kvale S (1996) InterViews. An Introduction to Qualitative Research Interviewing. Sage Publications. Laufer S (2003) Social Accountability Corporate and Greenwashing. J Bus Ethics 43:253–261. Lawson T (2003) Institutionalism: on the need to firm up notions of social structure and the human subject. J Econ Issues 175–207. Lemos M, Boyd E, Tompkins E, et al. (2007) Developing adaptation and adapting development. Ecol. Soc. 12:26: Levasseur V, Olivier A (2000) The farming system and traditional agroforestry systems in the Maya community of San Jose, Belize. Agrofor Syst 49:275–288. Lindenmayer DB, Likens GE (2011) Losing the Culture of Ecology. Bull Ecol Soc Am 92:245–246. doi: 10.1890/0012-9623-92.3.245

93

Lobell DB, Baldos ULC, Hertel TW (2013) Climate adaptation as mitigation: the case of agricultural investments. Environ Res Lett 8:015012. doi: 10.1088/1748-9326/8/1/015012 Lobell DB, Schlenker W, Costa-Roberts J (2011) Climate trends and global crop production since 1980. Science 333:616–20. doi: 10.1126/science.1204531 Locatelli B, Catterall C, Imbach P (2015) Tropical reforestation and climate change: beyond carbon. Restor. Ecol. Locatelli B, Evans V, Wardell A, et al. (2011) Forests and Climate Change in Latin America: Linking Adaptation and Mitigation. Forests 2:431–450. doi: 10.3390/f2010431 Locatelli B, Kanninen M, Brockhaus M, et al. (2008) Facing an Uncertain Future: How Forest and People can Adapt to Climate Change. CIFOR: Bogor, Indonesia Lund C (2014) Of What is This a Case?: Analytical Movements in Qualitative Social Science Research. Hum Organ 73:224–34. Magliocca N, Rudel T, Verburg P, et al. (2014) Synthesis in land change science: methodological patterns, challenges, and guidelines. Reg Environ Chang 15:211–226. Martens P, McEvoy D, Chang C (2009) The climate change challenge: linking vulnerability, adaptation, and mitigation. Curr Opin Environ Sustain 1:14–18. doi: 10.1016/j.cosust.2009.07.010 Mason J (1994) Linking qualitative and quantitative data analysis. In: Bryman A, Burgess RG (eds) Anal. Qual. data. Routledge, pp 89–110 Matocha J, Schroth G, Hills T, Hole D (2012) Integrating climate change adaptation and mitigation through agroforestry and ecosystem conservation. Agroforestry-The Futur. Glob. L. Use. Springer, pp 105–126 Mbow C, Van Noordwijk M, Luedeling E, et al. (2014) Agroforestry solutions to address food security and climate change challenges in Africa. Curr Opin Environ Sustain 6:61–67. doi: 10.1016/j.cosust.2013.10.014 McCarthy JJ, Canziani OF, Leary NA (2001) Climate Change 2001: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Third Assessment Report of the Intergovernmental Panel on Climate Change. IPCC: Intergovernmental Panel on Climate Change, Cambridge University Press, Cam. Meerman J (2005) Belize Protected Areas Policy and System Plan: Result 2: Protected Area System Assessment & Analysis. Meerman J, Epting J, Steininger M, Hewson J (2010) Forest Cover and Change in Belize circa 1990 - 2000 2005. Center For Applied Biodiversity Science, Arlington, VA, USA Merriam Webster (2015) Merriam Webster Dictionary. http://www.merriamwebster.com/dictionary/synergy. Accessed 12 April 2015. Metz B, Davidson OR, Bosch PR, et al. (2007) Contribution of Working Group III to the fourth assessment report of the Intergovernmental Panel on Climate Change.

94

Mitchell T, Maxwell S (2010) Defining Climate Compatible Development, CDKN ODI Policy Brief. Climate & Development Knowledge Network Mori AS, Spies TA, Sudmeier-Rieux K, Andrade A (2013) Reframing ecosystem management in the era of climate change: Issues and knowledge from forests. Biol Conserv 165:115–127. doi: 10.1016/j.biocon.2013.05.020 Moser SC (2012) Adaptation, mitigation, and their disharmonious discontents: an essay. Clim Change 111:165–175. doi: 10.1007/s10584-012-0398-4 Moser SC (2009) Good morning, America! The explosive US awakening to the need for adaptation. California Energy Commission, Sacramento. Murdiyarso D, Robledo C, Brown S, et al. (2005) Linkages between mitigation and adaptation in land-use change and forestry activities. In: Robledo C, Kanninen M, Pedroni L (eds) Trop. For. Adapt. to Clim. Chang. search Synerg. Bogor, Indonesia: Center for International Forestry Research (CIFOR), pp 122– 153 Neufeldt H, Jochem E, Hinkel J, et al. (2010) Climate policy and inter-linkages between adaptation and mitigation. In: Hulme H, Neufeldt H (eds) Mak. Clim. Chang. Work Us Eur. Perspect. Adapt. Mitig. Strateg. Cambridge University Press, Cambridge University Press, Cambridge, UK, pp 3–30 Oliver P (2003) Research and ethics. Student’s Guid. to Res. Ethics. Cambridge University Press, Cambridge, UK, pp 3–25 OSAC (2014) The Overseas Security Advisory Council (OSAC) Belize 2013 Crime and Safety Report. https://www.osac.gov/pages/ContentReportDetails.aspx?cid=14034. Accessed 21 Nov 2014 Pielke R (2005) Misdefining “climate change”: consequences for science and action. Environ Sci Policy 8:548–561. Pielke R, Prins G, Rayner S, Sarewitz D (2007) Climate Change 2007: Lifting the taboo on adaptation. Nature 445:597–598. Popp A, Krause M, Dietrich JP, et al. (2012) Additional CO2 emissions from land use change — Forest conservation as a precondition for sustainable production of second generation bioenergy. Ecol Econ 74:64–70. doi: 10.1016/j.ecolecon.2011.11.004 Porter JR, Xie L, Challinor A, et al. (2014) Food Security and Food Production Systems. Clim. Chang. 2014 Impacts, Adapt. Vulnerability. Contrib. Work. Gr. II to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Pramova E, Locatelli B, Djoudi H, Somorin OA (2012) Forests and trees for social adaptation to climate variability and change. Wiley Interdiscip Rev Chang 3:581–596. Rahn E, Läderach P, Baca M, et al. (2013) Climate change adaptation, mitigation and livelihood benefits in coffee production: where are the synergies? Mitig Adapt Strateg Glob Chang 19:1119–1137. doi: 10.1007/s11027-013-9467-x

95

Raupach MR, Marland G, Ciais P, et al. (2007) Global and regional drivers of accelerating CO2 emissions. Proc Natl Acad Sci U S A 104:10288–93. doi: 10.1073/pnas.0700609104 Ravindranath NH (2007) Mitigation and adaptation synergy in forest sector. Mitig Adapt Strateg Glob Chang 12:843–853. doi: 10.1007/s11027-007-9102-9 Reid H, Huq S (2007) Community-based adaptation: A vital approach to the threat climate change poses to the poor. Briefing Paper. IIED. International Institute for Environment and Development (IIED), London,UK Reyer C, Guericke M, Ibisch PL (2009) Climate change mitigation via afforestation, reforestation and deforestation avoidance: and what about adaptation to environmental change? New For 38:15–34. doi: 10.1007/s11056-008-9129-0 Ribot J, Najam A, Watson G (1996) Climate variation, vulnerability and sustainable development in the semi-arid tropics. In: Ribot JC, Magalhães AR, Panagides SS (eds) Clim. Var. Clim. Chang. Soc. Vulnerability Semi-Arid Trop. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 13–54 Rickards L, Howden SM (2012) Transformational adaptation: agriculture and climate change. Crop Pasture Sci 63:240–250. Robbins P (2012) Political Ecology: A Critical Introduction. Kindle version, 2nd Editio. Wiley-Blackwell Roessingh CH, Plasil T, Visser DP (2009) Between Horse & Buggy and Four-Wheel Drive: Change and Diversity among Mennonite Settlements in Belize, Central America. VU University Press Rosenberg DE, Marcotte TP (2005) Land-use system modeling and analysis of shaded cacao production in Belize. Agrofor Syst 64:117–129. Rosenzweig C, Elliott J, Deryng D, et al. (2014) Assessing agricultural risks of climate change in the 21st century in a global gridded crop model intercomparison. Proc Natl Acad Sci U S A 111:3268–73. doi: 10.1073/pnas.1222463110 Rosenzweig C, Tubiello F (2007) Adaptation and mitigation strategies in agriculture: an analysis of potential synergies. Mitig Adapt Strateg Glob Chang 12:855–873. Roshetko J, Lasco R, Angeles M (2007) Smallholder Agroforestry Systems For Carbon Storage. Mitig Adapt Strateg Glob Chang 12:219–242. Rudel TK (2008) Meta-analyses of case studies: A method for studying regional and global environmental change. Glob Environ Chang 18:18–25. Salas O, Castillo (2013) Socio-economic assessment of Programme for Belize’s activities on the primary buffering communities of the Rio Bravo Conservation and Management Area. Sathaye J, Najam A, Cocklin C (2007) Sustainable Development and Mitigation. In Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the

96

Intergovernmental Panel on Climate Change [B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (eds)],. Sayer A (1992) Method in social science. A realist approach. Routlledge. London Scherr SJ, Shames S, Friedman R (2012) From climate-smart agriculture to climate-smart landscapes. Agric Food Secur 1:12. Schimel D (2011) A reply to Fischer et al. Front Ecol Environ 9:430–431. doi: 10.1890/11.WB.022 Schoeneberger M, Bentrup G, de Gooijer H, et al. (2012) Branching out: Agroforestry as a climate change mitigation and adaptation tool for agriculture. J Soil Water Conserv 67:128A–136A. doi: 10.2489/jswc.67.5.128A Shweder RA (1991) Thinking Through Cultures: Expeditions in Cultural Psychology. Harvard University Press. SIB (2011) (Statistical Institute of Belize). Belize National Population and Housing Profile-2010. Belmopan, Belize Smit B, Burton I, Klein RJT, et al. (2000) An anatomy of adaptation to climate change and variability. Clim Change 45:223–251. Smit B, Burton I, Klein RJT, Street R (1999) The Science of Adaptation: A Framework for Assessment. Mitig Adapt Strateg Glob Chang 4:199–213. Smith JB (1997) Setting priorities for adapting to climate change. Glob Environ Chang 7:251–264. doi: 10.1016/S0959-3780(97)00001-0 Smith P, Bustamante M (2014) Agriculture, Forestry and Other Land Use (AFOLU). Clim. Chang. 2014 Mitig. Clim. Chang. Work. Gr. III Contrib. to IPCC 5th Assess. Rep. ‐ Chang. to underlying Sci. Assess. p 181 Smith P, Haberl H, Popp A, et al. (2013) How much land-based greenhouse gas mitigation can be achieved without compromising food security and environmental goals? Glob Chang Biol 19:2285–302. doi: 10.1111/gcb.12160 Smith P, Martino D, Cai Z, et al. (2008) Greenhouse gas mitigation in agriculture. Philos Trans R Soc B Biol Sci 363:789–813. Smith P, Olesen JE (2010) Synergies between the mitigation of, and adaptation to, climate change in agriculture. J Agric Sci 148:543–552. Smithers J, Smit B (1997) Human adaptation to climatic variability and change. Glob Environ Chang 7:129– 146. doi: 10.1016/S0959-3780(97)00003-4 Spradley JP (1979) The Ethnographic Interview. Fort Worth: Harcourt Brace Jovanovich College Publishers. Stake RE (1995) The art of case study research. Thousand Oaks, CA: Sage Publications. Stake RE (2005) Qualitative case studies. In Denzin, NK & Lincoln, YS (Eds.), The Sage handbook of qualitative research (3rd ed., pp. 443-466). Thousand Oaks, CA: Sage Publications.

97

Stake RE (1994) Case Studies. In: Denzin NK, Lincoln YS (eds) Handb. Qual. Res. Thousand Oaks, CA: Sage., pp 236–247 Steenwerth KL, Hodson AK, Bloom AJ, et al. (2014) Climate-smart agriculture global research agenda: scientific basis for action. Agric Food Secur 3:11. doi: 10.1186/2048-7010-3-11 Steinberg MK (1998) Political ecology and cultural change: impacts on swidden-fallow agroforestry practices among the Mopan Maya in southern Belize. Prof Geogr 50:407–417. Steinberg MK (2002) The globalization of a ceremonial tree: The case of cacao (Theobroma cacao) among the Mopan Maya. Econ Bot 56:58–65. Steinberg MK (1999) Maize diversity and cultural change in a Maya agroecological landscape. J Ethnobiol 19:127–139. Suckall N, Stringer LC, Tompkins EL (2015) Presenting Triple-Wins? Assessing Projects That Deliver Adaptation, Mitigation and Development Co-benefits in Rural Sub-Saharan Africa. Ambio 44:34–41. doi: 10.1007/s13280-014-0520-0 Suckall N, Tompkins E, Stringer L (2014) Identifying trade-offs between adaptation, mitigation and development in community responses to climate and socio-economic stresses: Evidence from Zanzibar, Tanzania. Appl Geogr 46:111–121. doi: 10.1016/j.apgeog.2013.11.005 Suzuki R (2012) Linking Adaptation and Mitigation through Community Forestry: Case Studies from Asia. RECOFTC – The Center for People and Forests. RECOFTC – The Center for People and Forests Swart R, Raes F (2007) Making integration of adaptation and mitigation work: Mainstreaming into sustainable development policies? Clim Policy 7:288–303. Tanner T, Allouche J (2011) Towards a new political economy of climate change and development. IDS Bull 42:1–14. doi: 10.1111/j.1759-5436.2011. 00217.x. Tashakkori A, Teddlie C (2010) Handbook of Mixed Methods in Social & Behavioral Research. Sage Publications, 2nd Editio. Sage Publications Taylor SJ, Bogdan R (1984) Introduction to qualitative research methods: the search for meanings (2nd ed.). John Wiley & Sons. 302. Taylor SJ, Bogdan R (1998) In-Depth Interviewing. In: Taylor SJ, Bogdan R (eds) Introd. to Qual. Res. Methods A Guideb. Resour., 3 edition. Wiley, pp 76–105 Thomas R, Pauw E de, Qadir M (2007) Increasing the resilience of dryland agro-ecosystems to climate change. J. Semi- Arid Trop. Agric. Res. 4: Thompson A, Bendik-Keymer J (2012) Ethical Adaptation to Climate Change: Human Virtues of the Future. Kindle version. The MIT Press. Thuy P, Moeliono M, Locatelli B, et al. (2014) Integration of Adaptation and Mitigation in Climate Change and Forest Policies in Indonesia and Vietnam. Forests 5:2016–2036. doi: 10.3390/f5082016

98

Tol RSJ (2005) Adaptation and mitigation: trade-offs in substance and methods. Environ Sci Policy 8:572– 578. doi: 10.1016/j.envsci.2005.06.011 Tompkins EL, Adger NW (2003) Defining response capacity to enhance climate change policy. Tyndall Centre Working Paper No. 39. Tyndall Centre for Climate Change Research and School of Environmental Sciences. University of East Anglia. Norwich, UK. 23. Tompkins EL, Adger NW (2005) Defining response capacity to enhance climate change policy. Environ Sci Policy 8:562–571. Tompkins EL, Mensah A, King L, et al. (2013) An investigation of the evidence of benefits from climate compatible development. Working Paper 124. Leeds and London, UK: Centre for Climate Change Economics and Policy. Tubiello FN, Salvatore M, Rossi S, et al. (2013) The FAOSTAT database of greenhouse gas emissions from agriculture. Environ Res Lett 8:015009. doi: 10.1088/1748-9326/8/1/015009 Turner WR, Oppenheimer M, Wilcove DS (2009) A force to fight global warming. Nature 462:278–279. doi: 10.1038/462278a UNEP (2013) The Emissions Gap Report 2013: A UNEP Synthesis Report. United Nations Environment Programme. UNFCCC (1992) United Nations Framework Convention on Climate Change. UNFCCC (2015) Homepage: http://unfccc.int/focus/adaptation/items/6999.php - Accessed 22 Apirl 2015. Valentine G (2005) Valentine, G. (2005): Tell me about…: using interviews as a research methodology. In: Flowerdew R, David MM (eds) Methods Hum. Geogr. A Guid. students doing a Res. Proj., 2nd Editio. Pearson Education Limited, Malaysia, pp 110–127 Van Exel J, de Graaf G (2005) Q methodology: A sneak preview. 30. Venema HD, Rehman IH (2007) Decentralized renewable energy and the climate change mitigationadaptation nexus. Mitig Adapt Strateg Glob Chang 12:875–900. doi: 10.1007/s11027-007-9104-7 Verburg PH (2014) Resilience Conference (May - 2014) Montpellier, France - The value of place-based research for large scale landscape change modelling. Session 67: The value of place-based research in an era of big data. Verchot L, van Noordwijk M, Kandji S, et al. (2007) Climate change: linking adaptation and mitigation through agroforestry. Mitig Adapt Strateg Glob Chang 12:901–918. Verheyen R (2005) Climate Change Damage and International Law: Prevention, Duties and State Responsibility. Martinus Nijhoff, Netherlands, 418 pp Victor D, Zhou D (2014) Introductory Chapter. In: Edenhofer O, Pichs-Madruga R, Sokona Y (eds) Clim. Chang. 2014 Mitig. Clim. Chang. Contrib. Work. Gr. III to Fifth Assess. Rep. Intergov. Panel Clim. Chang. Cambridge University Press, United Kingdom and New York, NY, USA., p 42

99

Vignola R, Locatelli B, Martinez C, Imbach P (2009) Ecosystem-based adaptation to climate change: what role for policy-makers, society and scientists? Mitig Adapt Strateg Glob Chang 14:691–696. Von Eye A, Schuster C, Rogers WM (1998) Modelling Synergy using Manifest Categorical Variables. Int J Behav Dev 22:537–557. Wainwright J (2011) Decolonizing development: Colonial power and the Maya. Kindle version. John Wiley & Sons Warren R (2011) The role of interactions in a world implementing adaptation and mitigation solutions to climate change. Philos Trans A Math Phys Eng Sci 369:217–41. doi: 10.1098/rsta.2010.0271 Warren R, Benzie M, Arnell N, et al. (2012) Scoping study: Modelling the interaction between mitigation and adaptation for decision making. UEA/Tyndall Centre, Stockholm Environment Institute, Walker Institute, University of Southampton, University of Cambridge. Wheeler T, von Braun J (2013) Climate change impacts on global food security. Science (80- ) 341:508–13. doi: 10.1126/science.1239402 Whitman AA, Brokaw NVL, Hagan JM (1997) Forest damage caused by selection logging of mahogany (Swietenia macrophylla) in northern Belize. For Ecol Manage 92:87–96. Wilbanks TJ, Kane SM, Leiby PN, et al. (2003) Possible responses to global climate change: integrating mitigation and adaptation. Environ Sci Policy Sustain Dev 45:28–38. Wilbanks TJ, Leiby P, Perlack R, et al. (2007) Toward an integrated analysis of mitigation and adaptation: some preliminary findings. Mitig Adapt Strateg Glob Chang 12:713–725. doi: 10.1007/s11027-0079095-4 Wilbanks TJ, Sathaye J (2007) Integrating mitigation and adaptation as responses to climate change: a synthesis. Mitig Adapt Strateg Glob Chang 12:957–962. doi: 10.1007/s11027-007-9108-3 Wilk R (2001) Consuming Morality. J Consum Cult 1:245–260. doi: 10.1177/146954050100100211 Wilk RR (1999) Whose forest? Whose land? Whose ruins? Ethics and conservation. Sci Eng Ethics 5:367– 374. Wilk RR (1997) Household ecology: economic change and domestic life among the Kekchi Maya in Belize. Northern Illinois University Press, DeKalb, Illinois, USA. Wilson E, Piper J (2010) Spatial planning and climate change. Routledge, London, UK World Bank (2011) Climate-Smart Agriculture: Increased Productivity and Food Security, Enhancing Resilience and Reduced Carbon Emissions for Sustainable Development, Opportunities and Challenges for a Converging Agenda: Country Examples. Washington, DC World Bank (2013) Adaptation Fund Project: Belize Marine Conservation and Climate Adaptation Project. Wyman MS, Stein T V (2010) Modeling social and land-use/land-cover change data to assess drivers of smallholder deforestation in Belize. Appl Geogr 30:329–342.

100

Wyman MS, Stein T V, Southworth J, Horwich RH (2011) Does Population Increase Equate to Conservation Success? Forest Fragmentation and Conservation of the Black Howler Monkey. Conserv Soc 9:216– 228. Yin RK (2009) Case study research: Design and methods. Sage Publications. Young CA (2008) Belize’s Ecosystems: Threats and Challenges to Conservation in Belize. Trop Conserv Sci 1:18–33. Zomer R, Trabucco A, van Straaten O, Bossio D (2006) Carbon, Land and Water: A Global Analysis of the Hydrologic Dimensions of Climate Change Mitigation through Afforestation/Reforestation; IWMI Research Report 101. International Water Management Institute, Colombo, Sri Lanka

101

PART II: PAPERS

Maya village (Jalacte, Toledo District, Belize, January 2014).

Land clearing (near San Jose, Toledo District, Belize, December 2013).

102

Paper 1

Addressing climate change mitigation and adaptation together: A global assessment of agriculture and forestry projects

In review at ‘Environmental Management’

103

Addressing climate change mitigation and adaptation together: A global assessment of agriculture and forestry projects Rico Kongsager1, Bruno Locatelli2+3, Florie Chazarin2 1. UNEP DTU Partnership on Energy, Environment and Sustainable Development, UN City, Marmorvej 51, DK-2100 Copenhagen O, Denmark, Phone: +45 61 69 04 68 2. CIFOR, PO Box 1558, 15024 Lima, Peru 3. CIRAD BSEF, Avenue Agropolis, 34398 Montpellier, Cedex 5, France Corresponding author: Rico Kongsager. E-mail address: [email protected] (R. Kongsager) Author Contributions: RK and BL developed methods, RK and FC collected and processed primary data, RK and BL analyzed results and wrote the paper. Abstract: Adaptation and mitigation share the ultimate purpose of reducing climate change impacts. However, they tend to be considered separately in projects and policies because of their different objectives and scales. Agriculture and forestry are related to both adaptation and mitigation: they contribute to greenhouse gas emissions and removals, are vulnerable to climate variations, and form part of adaptive strategies for rural livelihoods. We assessed whether climate change projects in forestry and agriculture integrated adaptation and mitigation, by analyzing 201 projects from adaptation funds (e.g., UNFCCC Adaptation Fund), adaptation plans (e.g., National Adaptation Programmes of Action), mitigation instruments (e.g., Clean Development Mechanism), and project standards (e.g., Climate Community & Biodiversity (CCB)). We analyzed whether projects established for one goal contributed explicitly to the other (i.e., whether mitigation projects contributed to adaptation and vice versa). We also examined whether their activities or expected outcomes allowed for potential contributions to the two goals. Despite the separation between the two goals in international and national institutions, 37% of the project documents explicitly mentioned a contribution to the other objective, although only half of those substantiated it. In addition, most adaptation (90%) and all mitigation projects could potentially contribute at least partially to the other goal. Some adaptation project developers were interested in mitigation for the prospect of carbon funding, whereas mitigation project developers integrated adaptation to achieve greater long-term sustainability or to attain CCB certification. International and national institutions can provide incentives for projects to harness synergies and avoid trade-offs between adaptation and mitigation. Keywords: REDD+; emissions; vulnerability; landscape; ecosystem services; livelihoods

104

1 Introduction As both mitigation and adaptation are needed to respond to climate change, several studies have argued that projects and policies should aim to avoid trade-offs and maximize synergies between the two approaches (Kok and De Coninck 2007; Swart and Raes 2007; Ayers and Huq 2009). Adaptation and mitigation share the ultimate purpose of reducing climate change impacts but have different objectives: mitigation aims to reduce emissions or enhance the sinks of greenhouse gases, while adaptation addresses the effects of climate change on people and ecosystems (Tol 2005). Because of their different objectives and scales, adaptation and mitigation tend to be considered separately in projects and policies (Klein et al. 2005). Agriculture, Forests, and Other Land Use (AFOLU) activities are relevant to both mitigation and adaptation, because they emit or capture greenhouse gases, are vulnerable to climate variations, and provide goods and services that are central to the adaptive strategies of local communities and reduce the vulnerability of local communities and broader society to climate variations (Ravindranath 2007; Guariguata et al. 2008; Locatelli et al. 2011; Matocha et al. 2012). Considering both adaptation and mitigation in climate change initiatives could help avoid trade-offs. For example, a forest plantation project that sequesters carbon for mitigation can reduce water availability for downstream populations and increase their vulnerability to drought. Failure to consider mitigation in adaptation initiatives may lead to adaptation measures that increase greenhouse gas emissions, which is one type of maladaptation, according Barnett and O’Neill (2010) and the fifth assessment report of the IPCC (Porter et al. 2014, p.34): “key maladaptation would be one which increased emissions of greenhouse gases”. Similarly, we could use the term “malmitigation” for an initiative that reduces greenhouse gas emissions but increases vulnerability. Several benefits of addressing adaptation and mitigation jointly in projects and policies have been mentioned in the literature. For example, integrating adaptation into mitigation projects may increase their resilience to climate variations, the permanence of carbon storage, and their acceptance by local communities, as adaptation responds to local issues (Locatelli et al. 2011). Project developers could access alternative sources of adaptation and mitigation funding and, if mitigation projects produce adaptation outcomes, carbon funding could bridge the adaptation funding gap (Matocha et al. 2012). Another advantage would be to improve the cost effectiveness of the overall climate change funding (Ravindranath 2007; Suckall et al. 2015). However, concerns have been raised about the feasibility of pursuing adaptation and mitigation together, the associated transaction costs, and the failure risk of overambitious and complex projects (Klein et al. 2005; Klein et al. 2007; Swart and Raes 2007). In 2012, around one billion US$ was committed globally to activities aiming at both adaptation and mitigation, out of a total of 80 billion US$ of climate finance committed by the International Development Finance Club, a network of national and subregional development banks, but no detail is available on the joint mitigation and adaptation activities (Ecofys-IDFC 2013). According to the OECD’s Creditor Reporting

105

System on aid activities for the global environment (which describes whether adaptation and mitigation are primary or secondary objectives of funded activities), 22% of funding to forestry and agriculture between 2010 and 2012 had both adaptation and mitigation as objectives, either primary or secondary (OECD 2014). As this funding includes broad institutional support, which reports contributions to all available environmental objectives, more analysis is needed to understand the degree of integration of adaptation and mitigation. Several emerging approaches to development (e.g., climate-smart development or climate-compatible development) consider adaptation and mitigation jointly and aim at achieving triple wins for development, adaptation and mitigation (Someshwar 2008). However, there is limited information on the applicability of these approaches in AFOLU activities. Previous studies have showed that forest mitigation projects rarely consider adaptation (Guariguata et al. 2008; Reyer et al. 2009) or have limited effects on local development and the vulnerability of local populations (Caplow et al. 2011). Other studies have assessed how specific AFOLU activities can contribute to mitigation and adaptation, for example, agroforestry (Schoeneberger et al. 2012), and how specific mitigation instruments can contribute to sustainable development, for example, the Clean Development Mechanism (Olsen 2007). However, analyses of how different AFOLU activities can contribute to both mitigation and adaptation are lacking (Suckall et al. 2015), although the IPCC called for more case studies on the relationships between adaptation and mitigation (Klein et al. 2007) and previous studies have called for more research to establish the conditions under which adaptation and mitigation can be effectively integrated in AFOLU activities (Dang 2003; Verchot et al. 2007; Locatelli et al. 2011). Before the integration of adaptation and mitigation is tested in policies and funds, we need to know if the current AFOLU projects, designed with either mitigation or adaptation objectives, contribute or could potentially contribute to the other objective. This paper analyzes the integration of adaptation and mitigation in proposed climate change projects in forestry and agriculture in the tropics. We adopted a semi-quantitative approach for reviewing 201 documents describing adaptation or mitigation projects under different portfolios of national programs, global instruments, funds or certification standards. We hypothesized that the separation of adaptation and mitigation in policies and funding is mirrored at the project level and that there is great potential for increasing this integration, but that incentives to harness synergies between adaptation and mitigation are lacking.

2 Methods and data 2.1 Project selection We searched project descriptions in major portfolios of climate change projects (Table 1). For building our database, we selected only projects with forestry and agricultural activities (projects with both, including agroforestry, were labeled “mixed”) in Africa, Asia, and Latin America, with project documents prepared before January 1, 2013, available online in English, French, or Spanish. We excluded projects without land management activities (e.g., animal waste management or biogas production) and national projects on capacity

106

building or institutional strengthening without agricultural or forestry activities on the ground. Projects of any scale and at any stage of development were included; however, we recorded the type and length of the project description (project design documents, identification note, profiles, or concept notes). Our database included 284 projects from 79 countries (see list of projects in Online Resource 1). Of these, 13 came from two portfolios: 10 projects verified by two standards (CCB plus VCS or PV) and three registered under the CDM and verified by one standard (CCB or PV).

Table 1. List of selected portfolio and acronyms Portfolio

Acronym

National Adaptation Programme of Actions Clean Development Mechanism

NAPA CDM

Goal A: adaptation M: mitigation A M

CarbonFix Climate Community & Biodiversity Plan Vivo Verified Carbon Standard Adaptation Fund Special Climate Change Fund under GEF Least Developed Countries Fund under GEF Global Environment Facility other than SCCF and LDCF: SPA-CBA (Strategic Priority on Adaptation for CommunityBased Adaptation), SFM (Sustainable Forest Management Program)

CF CCB PV VCS AF SCCF LCDF GEF

M M M M A A A Either A or M

Type

National plan International mechanism Project certification standard

International fund

Because the length of project documents was likely to influence the extent to which project objectives were substantiated and because adaptation project documents were significantly shorter than the mitigation project documents (respectively 45 and 100 pages on average), we excluded the 83 projects that had documents with 15 or fewer pages (including all the 70 NAPA short project descriptions extracted from the national adaptation plans). With this new set of 201 projects, there was no significant difference in document length between adaptation and mitigation projects (respectively 86 and 103 pages on average). In addition, this exclusion of short project documents reduced double-counting, as many long LCDF project documents originated from short NAPA project description. The remaining 120 mitigation and 81 adaptation projects were distributed across 68 countries (Fig. 1 and Online Resource 1). A higher proportion of projects aimed at mitigation in Latin America than elsewhere. The sample contained more forestry projects (52%) than agriculture (25%) and mixed projects (23%). The mixed projects were equally split between adaptation and mitigation, whereas agriculture projects were exclusively about adaptation and forestry projects mainly about mitigation (94 out of 105). Projects documents were dated from 2004 to 2012 (68% from 2010 or after).

107

Fig. 1. Distribution of the 201 analyzed projects by country (map); by region, primary project goal and sector (bar charts); by region and portfolio (pie charts)

2.2 Analysis First, we browsed the full text of all project documents to search for statements about the contribution of projects to adaptation and mitigation goals and for statements substantiating this contribution (see examples in Table 2). For adaptation, we considered three goals: social adaptation (i.e. reducing the vulnerability of people to climate variations with measures that are not aimed at adapting agriculture or ecosystems), agricultural adaptation (i.e. vulnerability of agriculture), and ecosystem adaptation (i.e. vulnerability of forests and other nonagricultural ecosystems). For projects explicitly associating adaptation and mitigation, we searched for statements about the project developers’ rationale for doing so.

108

Table 2. Examples of statements in the project documents about explicit and substantiated contributions to the two climate change goals Goal

Examples of explicit contributions

Substantiation

Social adaptation

The project objective is to reduce the vulnerability of communities to climate change. The project will empower subsistence farmers to combat the effects of drought.

Project proponents have assessed the vulnerability of livelihoods to climate variability and climate change. Specific activities, groups or communities are identified as the most vulnerable within the region.

Agricultural adaptation

The project will enhance the climate resilience of the agricultural sector by improving water and land management. The project will introduce agroecological practices that help reducing agricultural vulnerability to climate change.

Project proponents have assessed the vulnerability of agriculture and food production systems to climate variability and climate change and identified key threats.

Ecosystem adaptation

Project activities will reduce fire risk in forests. The project will plant native trees that are resistant to floods. Climate change risks are reduced through planting a mix of species. Protecting continuous forests across altitudinal ranges will assist species in adapting to climate change by allowing migration.

Project proponents have assessed the vulnerability of forests, mangroves or other ecosystems to climate variability and climate change and identified key threats.

Mitigation

Reforestation will lead to the long-term sequestration of atmospheric carbon in biomass and soils. The project aims to halt deforestation and its associated greenhouse gas emissions.

Project proponents have estimated potential carbon sequestration or emission reductions.

We defined a degree of explicitness and substantiation of the contribution of a project to mitigation and the three adaptation goals, by using the following scoring: “0” if not explicit; “1” explicit but not substantiated; “2” weakly substantiated; “3” strongly substantiated. Using analyses of variance, we assessed whether this degree depended on its primary goal (adaptation or mitigation), the sector (forestry, agriculture, or mixed), the portfolio and the location (region and sub-regions), after controlling for the effect of the length of project documents because longer documents were likely to have a more explicit and substantiated contribution. This analysis aimed at highlighting what factors influenced how projects reported their contribution to adaptation and mitigation, for example whether adaptation projects were more likely to report a contribution to mitigation than the contrary. Second, regardless of whether a contribution was explicit and substantiated, we analyzed how project outcomes could potentially contribute to adaptation and mitigation goals. For this, we established a list of project outcomes based on a literature review, and after grouping them, selected 36 outcomes (Table 3). We browsed the full texts of project documents to search for statements about these outcomes. For all projects, each of the 36 outcomes was scored: “0” if not mentioned; “1” if explicitly mentioned; or “2” if mentioned explicitly and substantiated. We also recorded qualitative descriptions of the outcomes for providing examples.

109

Table 3. Goals and outcomes used in the project analysis Goal Social adaptation

Project outcomes that can potentially contribute to the goal 15 outcomes: Livelihood diversificationacde; Incomeacde; Healthf; Energyabe; Tenure and rightsca; Food securityghc; Clean and reliable waterabe; Market accessac; Education and capacityaci; Disaster risk reductionbi; Protective infrastructurebi; Protective ecosystems (against landslides, floods, etc.)bcdi; Resilient infrastructure and housingbei; Strengthened institutions (incl. policies) and social networksaj; Gender and women’s empowermentkl

Agricultural adaptation

6 outcomes: Agricultural water management; Resistant or diversified crops and varieties; Farming practices for increased resilience; Post-harvest management; Resistant farm animals; Trees on farm for crop or animal resilience

Ecosystem adaptation

5 outcomes: Protection against climate-related disturbances (e.g., fires, pests); Reduced human pressures on ecosystems or restoration of degraded ecosystems; Connectivity between ecosystems at the landscape scale; Management of production forests and plantations for increased resistance or resilience; Management of agroforestry systems and multiple use forests for increased resistance or resilience 10 outcomes: Reduced carbon losses in ecosystems; Carbon storage in ecosystems; Carbon storage in products; Carbon storage in soils; Carbon storage in agricultural vegetation; Reduced N2O/CH4 emissions from soils; Reduced N2O/CH4 emissions from vegetation; Reduced N2O/CH4 emissions from animals; Reduced N2O/CH4 emissions from animal and plant waste; Reduced emissions from energy

Mitigation

Sources (primary sources in bold) a

Below et al. 2012:Tables 2-6 Noble et al. 2014:Sec. 14.3 c Pramova et al. 2012:Cases dMatocha et al. 2012:Table 1 e Angelsen et al. 2014 f Smith et al. 2014 g Porter et al. 2014:Sec 7.5 h Easterling et al. 2007:Sec.5.5 iZou and Wei 2010:Table 6 j Adger et al. 2011:Table 1 k Terry 2009:especially 101-10 l Rocheleau & Edmunds 1997 Clements et al. 2011:Table 1.1 Below et al. 2012:Tables 2-6 Cruz et al. 2007:Table 10.8 Rosegrant et al. 2010:72-6 Porter et al. 2014:513-20 Pramova et al. 2012:Case 2 Matocha et al. 2012:Table 1 Anderson and Zerriffi 2012:Table 1 Reyer et al. 2009:27-28 Guariguata et al. 2008:Table 1 Ravindranath 2007:Table 1 Locatelli et al. 2008:Table 1 Easterling et al. 2007:295 Matocha et al. 2012:Table 1 b

Smith and Bustamante 2014:11.3 Börner and Wunder 2012:Table 1 Ravindranath 2007:Table 1 Uprety et al. 2012:Table 1.1 Smith and Olesen 2010 Rosegrant et al. 2010:56-72 Matocha et al. 2012:Table 1 Anderson and Zerriffi 2012:Table 1

2.3 Limitations of the applied method One limitation of the analysis is its use of project design documents rather than evaluation of implemented projects. Few evaluations were available, and many of the projects are not implemented yet. We could not analyze the implementation of projects and the delivery of expected outcomes. As project design documents rarely reported expected negative outcomes, it was difficult to identify trade-offs between adaptation and mitigation, which may occur during project implementation. Our selection of outcomes for assessing the potential contribution to adaptation or mitigation probably influenced the results; for example, the high number of adaptation outcomes and their similarity to sustainable development outcomes have possibly overestimated the number of mitigation projects with a potential

110

contribution to adaptation. Whereas our assessment informs about potentials, some adjustments in project designs may be needed to achieve actual contributions: for example livelihood diversification may not automatically contribute to adaptation and project developers may need to analyze how diversification can reduce people’s vulnerability to climate variations.

3 Results 3.1 Contributions of adaptation projects to mitigation Few adaptation projects considered mitigation explicitly. Only 24 of the 81 adaptation projects (30%) explicitly mentioned a contribution to mitigation (Fig. 2), such as increasing carbon stocks in trees through afforestation, reforestation, and agroforestry or increasing soil carbon through land management. For example, in one adaptation project in Togo (with the identifier P017 in Online Resource 1) within the AF portfolio, mitigation benefits were reported from the sustainable management of degraded forests but it was recognized that this was not the priority of the project.

Fig. 2. Distribution of adaptation or mitigation projects according to their explicit, substantiated, and potential contributions to the other goal

Most adaptation projects could potentially contribute to mitigation. The majority of adaptation projects (73 out of 81 or 90%) reported mitigation-related outcomes, such as increased carbon storage in forests, soils or agricultural vegetation, and reduced carbon losses in forests. For example, an adaptation project in Colombia (P002) aimed at reducing risk and vulnerability to climate change and proposed climate change-resilient agricultural practices (agricultural adaptation) and ecosystem restoration with native trees resistant to flood conditions (ecosystem adaptation) to reduce floods in downstream communities (social adaptation). Several

111

expected outcomes could likely result in increased carbon storage in soils (e.g., soil restoration, decreased erosion, and improved fertility) and trees (e.g., agropastoral systems and reforestation). Even though many adaptation projects could potentially contribute to mitigation, some could definitely not. Among the 57 adaptation projects that did not state an explicit contribution to mitigation, eight (Fig. 2) had no outcome potentially contributing to mitigation and were agriculture projects aimed at improving technology and practices (e.g., crop varieties, water management) or infrastructure (e.g., irrigation canals, coastal dykes). The other 49 (Fig. 2) could mention a contribution to mitigation if there were any interest or benefit in doing so. Most adaptation projects that integrated mitigation did not explain their motivations for doing so. Mitigation was considered as a side effect in most adaptation projects with explicit contribution to mitigation, without further analysis of what opportunities this may represent. For example, project proponents in Eritrea (P014) mentioned that the project “will result in increased carbon sequestration” as an added benefit. One project in Bolivia (P034) saw mitigation as a possible funding source (“creating a future plan for carbon compensation”). In two projects, mitigation was perceived as a way to scale-up adaptation: project developers in Cambodia (P001) and Uruguay (P010) planned to explore whether REDD+ (Reducing Emissions from Deforestation and forest Degradation) could help promote adaptation at a larger scale (stating, in Uruguay, that “adaptation should be the most relevant plan of action” for the country). Two other projects, in Ghana (P071) and Togo (P017), mentioned the cost effectiveness of pursuing adaptation and mitigation jointly. However, the project developers in Togo highlighted possible trade-offs between the two goals and commented that their approach did “not optimize mitigation”. 3.2 Contributions of mitigation projects to adaptation Nearly half of the mitigation projects contributed explicitly to adaptation. Among the 120 mitigation projects, 43% explicitly stated a contribution to adaptation and 31% substantiated it (Fig. 2). Contributions were mainly related to social adaptation (33%) and more frequently to the adaptation of ecosystems (23%) than agriculture (16%), because most mitigation projects were forestry or mixed rather than agricultural. All mitigation projects had an outcome that could potentially contribute to at least one adaptation goal. Almost all (98%) could contribute to social adaptation, especially through education and capacity building, livelihood diversification, and increased incomes. For example, a mitigation project in Zimbabwe (P110) will increase income and diversify nonagricultural activities such as beekeeping as an adaptation measure and an alternative to wood harvesting, a driver of forest degradation. All mitigation projects could potentially contribute to ecosystem adaptation, especially through increased connectivity between forests at the landscape scale, reduced human pressures on biodiversity, management of climatic disturbances such as fires and the plantation of native species or species resistant to climatic stressors such as droughts. For example, a community carbon project in Mozambique (P090) aimed at restoring and protecting natural forests and biodiversity, reforesting for increasing landscape connectivity, harvesting timber sustainably, and creating

112

firebreaks and patrols. More than half of the mitigation projects (55%) could also contribute to agricultural adaptation, particularly through agroforestry, improved water management, and livestock or farming practices that improve resilience. For example, a mitigation project in Indonesia (P087) associated forest conservation with agricultural activities for water conservation, irrigation, improved technologies, and multispecies agroforestry systems. The potential contribution to agricultural adaptation often resulted from agricultural activities and sometimes from other activities, e.g., reforestation of riparian areas by a project in Brazil (P137) was expected to improve water for downstream irrigation. Few mitigation projects explained their motivations for integrating adaptation (only a quarter of the mitigation projects with explicit contribution to adaptation). Some project proponents stated that adaptation was needed because of climate change impacts, but it was unclear why the project would address climate vulnerability rather than other vulnerabilities. It seemed that most projects addressed adaptation because it was required by the CCB Standards to attain “gold” certification (there were much more projects integrating adaptation in the CCB portfolio than in other portfolios). For example, a CCB conservation project in Indonesia (P087) stated that “there is value in adopting a portfolio or mix of strategies that includes mitigation [and] adaptation”, but without further justification, even though the project developers recognized the “expected increasing costs of implementing programs and activities to [adapt to the] impacts of climate change”. Four projects included adaptation measures for ecosystems because of the perceived risk of climate variability or change for the permanence of carbon storage. For example, in a forest conservation project in Paraguay (P086), climate change impacts “may start to be felt towards the end of the project lifetime, potentially affecting carbon stocks”. Adaptation was also integrated into mitigation projects when the impacts of climate variations on agriculture were recognized as a threat to projects, ecosystems, and carbon. Examples include a project in Kenya (P085) where decreased food productivity, if not addressed, would increase pressure on biodiversity, and a REDD+ project in Zimbabwe (P110) where “drought threatens the project activities that focus on agricultural improvements”. 3.3 Projects integrating both adaptation and mitigation A mitigation project in our sample was more likely to integrate the other goal explicitly and substantiate it than an adaptation project, according to the analysis of variance (p