Adaptation to Climate Change Part 1

Adaptation to Climate Change.

Mark Pelling.

Acknowledgements.

This book would not have been possible without the inspiration and generous exchange of ideas with colleagues, in particular: Kathleen Dill (Cornell University), Cris High (Open University), David Manuel-Navarrete (King's College London) and Michael Redclift (King's College London). Research underpinning this book was undertaken as part of three grants awarded by the UK Economic and Social Research Council (RES-221-25-0044-A, RES-228-25-0014 and RES-062-23-0367). Without this financial support the work would not have been possible. I would also like to give thanks for the many rich discussions I've enjoyed with PhD and masters students in the Department of Geography at King's College London two PhD graduates, Marco Gra.s.so and Llewellyn Leonard, have their theses referenced. Wider discussions, in particular through the Global Environmental Change and Human Security Programme led by Karen O'Brien at the University of Oslo, have also been instrumental in shaping this work. For patience beyond the call of duty I thank the Routledge editorial and print teams, in particular Andrew Mould. Special recognition is also due to Ulli Huber for creating the time and atmosphere necessary to complete this work and to Lilly Pelling for her word processing and computer management skills. Of course, the real thanks go to all those respondents who have freely given of their time and energies to provide the empirical backbone for this work, many of whom remain at the sharp end of adapting to the consequences of climate change and development failure.

Part I.

Framework and theory.

1.

The adaptation age.

Everyone has the right to life, liberty, and security of person.

(Universal Declaration of Human Rights, Article 3).

Climate change adaptation is an opportunity for social reform, for the questioning of values that drive inequalities in development and our unsustainable relationship with the environment. But this outcome is by no means certain and growing evidence suggests that too often adaptation is imagined as a non-political, technological domain and enacted in a defensive rather than a progressive spirit. Adaptation has been framed in terms of identifying what is to be preserved and what is expendable, rather than what can be reformed or gained. Dominant development discourses put the economy as first to be preserved, above cultural flourishing or ecological health. There is a danger that adaptation policy and practice will be reduced to seeking the preservation of an economic core, rather than allowing it to foster the flourishing of cultural and social as well as economic development, or of improved governance that seeks to incorporate the interests of future generations, non-human ent.i.ties and the marginalised.

The argument put forward in this book suggests that adaptation is a social and political act; one intimately linked to contemporary, and with the possibility for re-shaping future, power relations in society. But it also recognises that different actors perceive contrasting roles for adaptation. That there may be multiple ways of adapting is already recognised in the literature through the range of different scopes and timings for adaptive interventions (for example, Smit et al., 2000; Smit and Wandel, 2006). These are important technical considerations but more emphasis is needed on the underlying socio-political choices that are made through the selection of adaptation pathways. Here we propose three such pathways leading to resilience (maintaining the status quo), transition (incremental change) and transformation (radical change). No one pathway necessarily leads to 'progressive' or more equitable and efficient outcomes than the others. The evaluation of pathways and subsequent outcomes will be a function of context and the viewpoint of individual actors. Opening a.n.a.lysis of how it is that individual adaptive pathways come to dominate or be marginalised is one of the aims of this book, which offers theoretical and empirical exploration.

Recent experience suggests that consensus on a progressive adaptation will not be easy. Our current age of adaptation is the second time in recent history that a global environmental challenge has provided an opportunity to question dominant forms of development. The first, coalescing around the notion of sustainable development, has (to date) manifestly failed. The international roots of the sustainable development agenda lay in a concern that the environmental limits to economic growth were fast approaching. Indeed the combination of mitigation and adaptation agendas represents a reprise of the sustainable development agenda, and climate change a strong signal that existing developments are far from sustainable (Le Blanc, 2009). Underlining the significance of adaptation for sustainable development, Adger et al. (2009a) remind us that climate change adaptation decisions have justice consequences across as well as within generations.

The first mainstream expression of a sustainable development approach was the Brundtland Commission, 1983, which stimulated a search for radical ecological and social alternatives to development (Redclift, 1987). These peaked in public awareness at the UN Conference on Environment and Development (UNCED) in Rio de Janeiro, Brazil, 1992. Here, differences in the prioritising of development and environment between rich and poorer nations and the influence of a strong industry lobby limited reform at the international level. The parallels with current challenges facing international negotiations at the UNFCCC are striking. The policy legacy of UNCED has been a constrained version of sustainable development largely restricted to ecological modernisation and an acceptance of the subst.i.tutability of environmental for economic value (Pelling, 2007a). Where some success has been achieved through this process it is outside of the compromised domain of international politics through the innovations of civil society groups, fair trade companies and concerned individuals, where environmental and social justice goals have been brought into projects for economic development (Adams, 2008). But these initiatives remain fragmented and overwhelmed by the global policy consensus.

Can climate change adaptation reinvigorate these debates and provide an impetus for stronger sustainable development action? Might climate change adaptation be both a reprise of sustainable development and a new opportunity in its own right? The origins of the UNFCCC process lie partly in UNCED where the first Framework Convention on Climate Change agreement was opened for signature. This connection to debates on sustainable development also reminds us that climate change and resultant adaptation are but one expression of an underlying crisis in environmentsociety relationships. The deepest root causes of climate change and the inability of those with power in society (locally and globally) to act lie in the dominant processes and values of the political economy that increasingly concentrate wealth in the hands of a few, with unjust social and environmental externalities as accepted. At this level climate change risk is but one expression of a deeper social malaise in modern society. For the poor, comfortable and rich alike aspiring and acquiring in order to consume have become the rationale for development; a rationale propelled as much by fear of failure as the pleasures of consumption. Can the burgeoning academic and policy interest in adaptation be levers to address these deeper questions of sustainability and justice, as well as adjusting to meet the more proximate risks presented to us by a changing climate?

Here we propose and ill.u.s.trate a framework to help reveal and understand the social, cultural and political pathways through which adaptation to climate change unfolds. Adaptation is conceptualised through three layers of a.n.a.lysis (Chapters 35) which build from a starting point in the notion of resilience to encompa.s.s adaptation as a process of socio-political transition and transformation. Each stage of theoretical a.n.a.lysis brings together work from systems theory with a wider literature including regime a.n.a.lysis, discourse, risk society, human security and the social contract. This reflects the strong influence of systems thinking on adaptation work but also enables the theoretical precision derived from systems thinking for example, on social learning and self-organisation to run throughout the book while bringing to the fore power, which is more ably addressed through other theoretical discourses. These theoretical discussions are then ill.u.s.trated through three case study chapters showing how adaptation can unfold through contested politics in organisations, urban systems and nation states.

Power lies at the heart of this conceptualisation of adaptation. Power asymmetries determine for whom, where and when the impacts of climate change are felt, and the scope for recovery. The power held by an actor in a social system, translated into a stake for upholding the status quo, also plays a great role in shaping an actor's support or resistance towards adaptation or the building of adaptive capacity when this has implications for change in social, economic, cultural or political relations, or in the ways natural a.s.sets are viewed and used. Accepting that adaptation is contested makes interpreting adaptation as progressive hostage to the observer's viewpoint. This requires the imposition of a normative framework to provide a consistent and transparent positionality for a.n.a.lysis. Here we are guided by Rawls' theory of justice that identifies procedural (inclusion in decision-making) and distributional (social and spatial) elements. Rawls (1971, see also Paavola et al., 2006) prioritises human rights over public goods; holds the social contract between citizens and the state in dynamic tension so that it is liable to capture by vested interests at moments of pressure; and argues that society should be governed by principles that protect inclusive governance and seek to enhance the quality of life of the poorest. This final statement is perhaps the most important for making judgements on comparative adaptation pathways.

In seeking to make the social and political elements of adaptation visible three questions run throughout this book and structure its narrative: 1. How is adaptive capacity shaped?

Or, to what extent is adaptive capacity dependent upon existing inst.i.tutional and actor capacity; can it be constructed anew through external influence or through autonomous actions?

2. How is adaptive capacity turned into adaptive action?

Or, what inst.i.tutions and actors are important in mediating this threshold and the wider feedback between action and future capacity?

3. What are the human security outcomes of adaptive actions?

Or, how far do framing inst.i.tutions and individual actors control processes of adaptation and how does this affect the exercise of rights and responsibilities in society, and the social distribution of well being, basic needs, human rights and subsequent adaptive capacity?

The following sections in this introductory chapter establish the scope of the book. First adaptation is defined and the approach taken to make climate change and a.s.sociated adaptations visible explained. Second, to help contextualise this work, some of the main strands in contemporary adaptation theory are presented. Finally the structure of the book is outlined.

Adapting to climate change.

Adaptation in the face of environmental change is nothing new. Individuals and socio-ecological systems have always responded to external pressures. But climate change brings a particular challenge. Uncertainty in the ways through which climate change will be felt set against its speed and scale of impact, combined with the invisibility of causal linkages in everyday life, bring new challenges for the sustainability of socio-ecological systems. It is for this reason that understanding adaptation to climate change is a critical challenge of our time. As the t.i.tle of this book suggests, adaptation is conceived of here as a dynamic phenomenon as a process rather than a status. An individual or business may be well adapted to a particular moment in history, but the dynamism of climate change requires an adaptation that can coevolve with it. Climate change is no longer an external threat to be managed 'out there', but is an intimate element of human history both an outcome and driver of development decisions for individuals, organisations and governments. This requires a closer look at social relations and practices, even values, as sites for adaptation, and suggests that it is necessary, but not sufficient, to control the impacts of climate change through technological innovations like environmental engineering and crop selection.

There are many ways of characterising adaptation, which as an intellectual construct cannot be directly observed. Here a key distinction is made between adaptation that is forward or backward looking. As a backward looking attribute, adaptation is revealed by capacity to cope during moments of stress or shock. For example, well-adapted urban communities have fewer losses to hurricane events. Greater capacity in Cuba's early warning and evacuation systems when compared to the southern states of the USA in large part explain the far lower human losses in Cuba from hurricane events (UNDP, 2004). As a forward looking attribute, adaptation cannot be revealed through impacts (which have not yet happened) and instead is made visible through theoretically identified components a.s.sociated with adaptive capacity. An important gap in our understanding of adaptation comes from the difficulty of being able to follow adaptive processes over time and so verify through observation the contribution of theoretically defined components on adaptive practices.

Despite this caveat, our focus is on forward looking adaptation. It is here that adaptation has the potential to intervene in development policy and practice through progressive risk reduction. To this extent the work is driven by theoretical understandings of what const.i.tutes adaptive capacity. On the ground, however, past experiences that reveal backward looking adaptation can feed in to local understandings of the pressures shaping capacity looking forward. A full discussion of adaptation theory is presented in Chapter 2.

For researchers and policy makers alike the invisibility of forward looking adaptive capacity is compounded by the dynamism of climate change. For specific physical or ecological systems change can be gradual and persistent for example, in sea level rise. For others temporary equilibrium may be violently disrupted when thresholds are breached and systems enter new states for example, the potential reversal of the thermohaline circulation system in the North Atlantic. The impact of such global scale processes is mediated by local socio-ecological and environmental conditions. This has led many to argue that adaptation is a local agenda in contrast to mitigation, which is global. While our concern is with adaptation, we make a case for both agendas to have local and global components and indeed national level action too. High level legal frameworks and voluntary agreements can support local action, but local level action is also a potential driver for higher level policy. Where political will is absent at higher levels, local action has the potential to be decisive in determining capacity and action and influencing higher level policy. This is the case for mitigation and adaptation for investing in zero carbon lifestyles and technology as much as livelihood diversification. Those fundamental social attributes that enable and shape adaptive capacity also influence the potential for local contributions to mitigation (Bulkeley and Betsill, 2003).

Climate change is also a slippery concept to demonstrate empirically. Outside of the imaginary worlds of computer models it is as yet impossible to determine the proportion of any hydrological or meteorological event that is attributable to climate change. O'Brien and Leichenko (2003) were among the first to argue that searching for the incremental risk a.s.sociated with climate change is a lost cause and many years away from resolution. Meanwhile the numbers of people and socio-ecological systems at risk and bearing loss from climate change a.s.sociated events is increasing. Climate change is manifest locally through extreme events and in the heightened variability of precipitation, temperature and wind. We may never understand the precise contribution of anthropocentric climate change to these events and trends but we can be certain that climate change is a decisive contributing factor and that vulnerability exists, demanding action.

The idea of adaptation.

While mitigation was clearly defined in the original United Nations Framework Convention on Climate Change (UNFCCC) negotiated in the Rio Summit, 1992, adaptation was not. Despite this the term was used in the agreement text. Its meaning continues to be debated (Burton, 2004). Arguably it is the slipperiness of the term that has been part of its attraction for discussion in academic and policy circles alike. Here we present an overview of some of the main contributions to the adaptation debate as scholars and policy makers have sought to make sense of the term handed to them from the UNFCCC. The section begins with an a.s.sessment of the influence of the IPCCUNFCCC on scholarly work on climate change adaptation, of the ways in which climate change impacts are evaluated and the geographical distribution of climate change impact risk. From this point an overview of work on social aspects of adaptation is presented around four questions that cross-cut research. This discussion is a prelude to that in Chapter 2, which offers an extended response to the intellectual inheritance and current shape of adaptation to build a conceptual framework.

The IPCCUNFCCC frame.

The IPCC and UNFCCC procedures and agendas have greatly influenced the direction of thinking as well as policy on climate change adaptation. There is a high level of interaction between these inst.i.tutions, with the IPCC feeding into the UNFCCC process, which in turn helps to drive funding and political will for adaptive actions and research. The stated aim of the IPCC is to support national policy on climate change through offering scientific consensus. Founded in 1988, the IPCC has produced a.s.sessment Reports in 1990, 1995, 2001 and 2007. Each in turn has included a greater emphasis on adaptation as evidence has acc.u.mulated. In this way the IPCC has acted as both a stimulus and a resource for research on adaptation to climate change.

The First a.s.sessment Report helped to shape the UNFCCC and drive its ratification at the UN Conference on Environment and Development, Rio, 1992 (Agrawala, 2005), but said relatively little about adaptation. It was in the Second a.s.sessment Report that the socio-economic aspects of climate change were seriously addressed for the first time. The report concluded by sketching out the scope of support needed for adaptation. It argued that efficient adaptation depended upon the availability of financial resources, technology transfer and cultural, educational, managerial, inst.i.tutional, legal and regulatory practices, both domestic and international. The vision was firmly on the potential roles and responsibilities of international actors with limited evidence of local adaptive behaviour. The Third a.s.sessment Report included a greater focus on adaptation strategies and concluded that adaptation was necessary to complement mitigation efforts raising the significance of adaptation in the UNFCCC process and helping to achieve the Nairobi work programme. The Fourth a.s.sessment Report stated that adaptation was necessary to address the impacts of climate change, was clear that this was already occurring and that more extensive adaptation than was being undertaken would be necessary to address future vulnerability to climate change. Hinting at the possibility of a progressive adaptation agenda, the report also connected sustainable development with vulnerability to climate change, and argued that climate change could impede national abilities to follow sustainable development pathways. This report provided the scientific basis for the Bali Action Plan reached by parties at the 13th session of the Conference of the Parties (COP).

While the IPCC process is scientific it reports back to governments and is influenced by their interests and priorities (Grundmann, 2007). It has been described as a boundary object, and a hybrid science-policy project at the interface between science and politics. As a consensus organisation, and one open to intense public scrutiny, it is conservative, careful to follow core science rather than policy or advocacy trends. This has been its strength in terms of scientific credibility for policy makers, but has also made it difficult for some evidence on climate change and human reaction to be included. For example, much local evidence for climate change impacts and experience in adaptation, particularly in Africa, Asia and Latin America and the Caribbean, is gained by local actors and held by civil society actors or published nationally or regionally but not in international peer review journals and so has been difficult to include. The IPCC in this respect offers a conservative, rigorous view of climate change, but should not be seen as a full acccount of existing information or knowledge. The Fourth a.s.sessment Report from Working Group II began to address this by drawing also from the grey literature produced by governments and NGOs. A concern for inclusiveness in scientific representation from all world regions has also led to a quota system and travel funds to support partic.i.p.ation from scientists based in low- and middle-income countries. Even so, this does not mean that governments are equally happy with IPCC findings, with various US governments largely ignoring the IPCC while others (especially in Europe) have endorsed and acted upon it through the UNFCCC and unilaterally (Grundmann, 2007).

The IPCC process has also been constrained by its slow recognition of the full contribution to climate change debates to be made from parallel disciplines or policy areas that may cover very similar ground but not use the language of climate change or publish in climate change a.s.sociated journals. Thus, for example, the considerable academic and policy literatures on disaster risk reduction, social security and food security in developing countries, community-based water management and risk insurance are making only slow impact on the IPCC. Such a sharp focus was perhaps appropriate to managing information on the natural and physical science components of climate change. For Working Group II's remit of vulnerability and adaptation, where impacts and responses often build on past experience but ultimately transcend policy or disciplinary boundaries, this is less helpful at times threatening that the IPCC will reinvent theoretical or methodological lessons that could better be brought in from other specialisms.

For example, much of early conceptual work on adaptation mirrored existing work on coping within the food security and disaster risk disciplines. The Fourth a.s.sessment Report from Working Group II went some way to addressing this concern with the inclusion of cross-sector and indeed cross-report case studies including the consequences and responses to Hurricane Katrina and the vulnerability of mega-deltas (IPCC, 2007). Working Group II also took the lead role between 200911 in organising a Special Report on Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation, to help bring knowledge from disaster risk management into climate change adaptation. At this stage of mapping and understanding vulnerability and adaptation the inclusion of knowledge from cognate areas is important. However, it is also key that the IPCC be seen in context. It is a mechanism for consolidating knowledge for the policy community on climate change. As the IPCC increasingly recognises parallel communities the challenge will be in retaining its core purpose and intellectual focus while embracing ever wider sources of knowledge. For the academic community the challenge is to communicate effectively with the IPCC process without restricting a.n.a.lysis and thought to the priorities of the IPCC.

The costs of adapting.

It is difficult to estimate the future costs of adapting to climate change. From the more restricted world of disaster management we know that the difference between investing in prevention and the costs of a disaster impact can easily exceed a ratio of 7:1 (DFID, 2004a). The costs of adapting to climate change are more far-reaching. Future demands for adaptation are also to be shaped by the actions we take now to mitigate it. One estimate, by Stern (2006), suggests adaptation costs in the order of 5 to 20 times the estimated costs of containing climate change through mitigation. The economic costs of adaptation are also not evenly distributed worldwide. Using data from past natural disaster events shows that richer nations with an acc.u.mulated legacy of physical infrastructure and housing have the most absolute economic exposure. However, these countries also have the a.s.sets to adapt (and arguably have substantial power and duty to mitigate, and in this sense have some control over, their own destiny). Poorer countries have less physical a.s.sets exposed but economies tend to be more dependent upon primary production and ecosystem services. As a proportion of GDP potential economic losses are highest in low and middle income countries. In addition, it is the same countries from Africa, Central, South and Southeast Asia and Central America and the Caribbean that record the highest mortality rates from natural disasters, adding human to economic vulnerability (UNDP, 2004). Past experience and projected risk of human loss through mortality and morbidity are also strongly skewed to poorer countries where income is dependent on primary extraction and where populations are not protected from environmental hazards by safe buildings, infrastructure, health services, and transparent and responsive governance (IFRC, 2010).

Observed data based on losses to past patterns of disaster events is the best guide to current vulnerability and backward looking adaptive capacity, but climate change means past patterns of hazard may not be as useful a guide to the future as had once been a.s.sumed; the so-called problem of non-stationarity (Milly et al., 2008). Figure 1.1 shows an example of forward looking a.s.sessment of relative vulnerability to climate change and extremes under a warming of 5.5 degrees C. It incorporates adaptive capacity as a component of vulnerability. Vulnerability is calculated based on input variables for human resources (dependency ratios and literacy rates), economic capacity (market GDP per capita and income distribution) and environmental capacity (population density, sulphur dioxide emissions, percentage of unmanaged land). The advantage of this approach is that it is not tied to past experiences of extreme events. Despite this, results largely confirm the burden identified above for poorer countries. High levels of vulnerability are a.s.sociated with low and middle income countries in South America, southern Asia and Africa. High vulnerability is also found in China and some of Eastern Europe. But this method also suggests North America and Europe are extremely vulnerable, painting a portrait of widespread vulnerability across the globe where adaptive capacity is overwhelmed by climate change even over the next 40 years (Yohe et al., 2006). This is a compelling case for the need for urgent and deep levels of mitigation alongside the need to support adaptation to reduce vulnerability from current and inescapable future climate variability and extremes.

The IPCC (2007) calculates that for the most exposed countries, such as coastal states in Africa, adaptation costs may be as high as 10 per cent of national GDP. For low-lying small island developing states the relative costs are even higher. Oxfam (2008) estimates that at least US$50bn is needed annually to support adaptation in developing countries. UNDP (2007) identifies an additional need of around US$86bn by 2015 (0.2 per cent of developed country GDP) on top of existing overseas development a.s.sistance budgets from bilateral and multilateral donors. These are large sums, but not unprecedented. The UNDP (2007) equates its total cost estimate to around 10 per cent of the current military expenditure by OECD countries.

The international architecture for support of adaptation is developing as adaptation rises on the political agenda. The UNFCCC provides one management structure for support of low and middle income countries. Bilateral and multilateral agencies, such as the development banks and other UN agencies, also provide financial and technical support. Investment decisions in the corporate private sector also impact on adaptation, including policy decisions from the insurance and reinsurance sectors, and are likely to increase in importance as businesses in middle and high income countries are forced to adapt. The emerging infrastructure is, however, built around existing poles of power nation states and the UN system which is beholden to them, or banking interests, with nation states or private investors at the helm. Can these actors be expected to embrace adaptation as anything other than resilience acts to reinforce the status quo? Indeed should they be encouraged to do so? a.s.serting more radical change in social and political systems needs to come from below through the actions of people at risk building on existing social and political reform movements.

With the costs of climate change increasing and adaptation being increasingly demanded, meeting the funding gap for adaptation in the short term is a key Figure 1.1 Global distribution of vulnerability to climate change. Combined national indices of exposure and sensitivity (Source: Yohe et al., 2006).

challenge. Without additional and earmarked funds for adaptation there is a risk of money being forced from existing overseas development a.s.sistance (ODA) budgets. ODA finance is already being squeezed by increased recent demand for humanitarian and disaster reconstruction funding (White et al., 2004). Agrawala (2005) has estimated that between 1560 per cent of official development a.s.sistance (ODA) flows will be affected by climate change. This trend is a particular tragedy as ODA is a key mechanism for reducing generic vulnerability to disaster risk and climate change impacts as well as achieving broader human security goals. A range of proposals exists for identifying additional funds. Oxfam (2008) proposes that funding be generated from auctioning a fraction of emissions allocations to developed countries under the post-2012 agreement, including proposed new emissions-trading for international aviation and shipping. Other proposals include increasing the share of the Clean Development Mechanism contributing to adaptation and increasing the role played by private capital through venture capital or commercial loans.

Conceptual development.

Since its reintroduction into social scientific and policy debates following the Rio Summit, the interests of different a.n.a.lysis have made adaptation a slippery concept. For some, adaptation's contribution would best be as a tightly defined, technical term (like mitigation in the existing UNFCCC doc.u.mentation) that can add universal clarity to policy formation including at the international level (for example, Schipper and Burton, 2009). Others, who see adaptation not as a technical category but as a research field, tend to have a wider view. Fankhauser (1998) suggests that adaptation can be synonymous with sustainable development. This challenge was noted as early as 1994 by Burton, just two years after the Rio Summit, and the plethora of interpretations has continued to grow as individual disciplines and intellectual communities have invested adaptation with their own worldviews (Kane and Yohe, 2007).

The adaptation to climate change debate is driven by four questions: * What to adapt to?

* Who or what adapts?

* How does adaptation occur?

* What are the limits to adaptation?

None of these questions have easy answers.

What to adapt to?

Climate change itself is agreed to be manifest in at least three interacting and overlapping ways: climate change has come to encompa.s.s long-term trends in mean temperatures and other climatic norms, importantly precipitation, and secondary effects like sea-level rise together with variability about these norms from inter-seasonal to periods of a decade with particular implications for infrastructure planning, agriculture and human health, and extremes in variability that can trigger natural disasters such as floods, hurricanes, fires and so on (IPCC 2007). Furthermore, local studies of adaptation make it increasingly clear that while international and national policy makers may seek a clear measurement of impacts and adaptation a.s.sociated with climate change the incremental costs of mitigating or adapting to climate change, as the Global Environmental Facility puts it (Labbate, 2008) on the ground, any meaningful measurement of adaptation needs to accept climate change is contextualised with the other risks (social, economic and political as well as environmental) that shape and limit human well being and the functioning of socio-ecological systems (Pelling and Wisner, 2009). This is the difference between an economic a.n.a.lysis of the farming sector of a country, and understanding the competing choices that shape adaptive capacity and actions for an individual farmer put in the context of the markets and regulatory regimes within which the farmer operates. Both are useful but partial lenses. The overlapping of seasonal and other climatic cycles with variation in baseline climate change and extremes makes it very difficult for specific events to separate out climate change signals from background weather patterns. Both short-term uncertainty in variability and extremes and long-term trends need to be considered (Adger and Brooks, 2003).

Who or what adapts?

Initial work on a.s.sessing who or what adapts came from the a.s.sessment of regional or national scale agro-economic systems. For example, Krankina et al. (1997) refer to boreal forestry management strategies as a means of a.s.sisting forests adapt. Here the system of interest was ecological and the management system an intervening variable between it and climate change. This kind of work complements well the scale of resolution available from climate modelling and the existing understanding of ecological adaptation within agricultural sciences, but is less suited to explore well the social processes driving and limiting adaptive decision-making. Economic a.s.sessment has also operated well at this scale, seeking to identify the costs (and benefits) of climate change scenarios for agricultural systems and to varying extents factoring in human adaptation. In a review of the economics of climate change literature, Stanton et al. (2008) observe the narrow framing used to approach decision making for climate change policy. Harvey (2010) goes further, arguing that a new macro-economic vision is needed to help move past the internal contradiction of contemporary economics that promotes energy intensive growth and so accelerates climate change with consequent growth inhibiting outcomes. Contemporary incentives push for greater and greater economic growth in an attempt to grow our way out of climate change and its attendant risks. The extraction and concentration of wealth that results increases collective vulnerability while simultaneously accelerating climate change a.s.sociated (and other environments) hazards.

More human-centred a.n.a.lyses have also flourished which seek to identify the human and social characteristics that determine the capacity of communities to face a shock or stress (Adger et al., 2005a). Local viewpoints help to contextualise adaptation within development and explain why people are unable or unwilling to take adaptive action (helping to identify the limits to climate change adaptation). From an a.n.a.lysis of two communities in Puerto Rico, Lopez-Marrero and Yarnal (2010) found that concerns for health conditions, family well being, economic factors and land tenure were given more priority by local actors than adaptation to climate change, despite their exposure to flooding and hurricanes. The results show the importance of addressing adaptation within the context of multiple risks, and of people's general well being.

How does adaptation occur?

The diversity of work examining processes of adaptation has benefited from a number of typologies of adaptive action and their coherent synthesis, see Smit et al. (2000), Smit and Wandel (2006), Burton et al. (2007). Carter et al. (1994) distinguished between autonomous (automatic, spontaneous or pa.s.sive adaptations) that occur as part of the routine of a social system, and planned (strategic or active) adaptations. Smit et al. (2000) also add that adaptations may occur unintentionally as an incidental outcome of other actions further emphasising the importance of contextualising a.s.sessments of adaptive capacity and action. The timing of the adaptation relative to its stimulus has led to additional types. Some draw from the disasters community, which uses a staged model of actions for tracking behaviour before and after disasters. Burton et al. (1993) distinguish adaptations that prevent loss, spread loss, change use or activity, change location or engage in restoration. More generally, adaptations can be reactive, concurrent (especially important for a.n.a.lysis of adaptation to gradual and ongoing changes in climatic norms) and antic.i.p.atory. Adaptive actions can be long- or short-term, and this has come to be a.s.sociated with a distinction between actions aiming for short-term stability (coping) or longer term change (adaptation) (see Chapter 2). Adaptation has also been characterised according to the form of action (technological, behavioural, financial, inst.i.tutional or informational), the actor of interest (individual, collection), the scale of the actor (local, national, international) and social sector (government, civil society, private sector); and the costs and ease of implementation (Smit et al., 2000). Maladaptation is used to describe those acts that, through bad planning or inadvertent consequences, cause either local or distant consequences that outweigh gains (Smit, 1993).

What are the limits to adaptation?

Literature on the limits to adaptation has largely been framed by the concerns of international actors. The challenge so defined is to provide guidance for policy makers on what might be achieved through adaptation to limit or avoid the dangerous impact of climate change as a parallel agenda to mitigation to help achieve a balance in investment between mitigation and adaptation (Hulme, 2009). At first this seems a technical problem of a.s.sessing the economic costs of a range of technical solutions and applying costbenefit a.n.a.lysis to a range of scenarios. This can certainly help. But if adaptation is to move us towards a more sustainable development path then technological investments are only part of the solution. Changes in values and a.s.sociated governance regimes will also need to be on the agenda, or may force themselves on as established inst.i.tutions fail in the face of climatic extremes. Examining these limits is more difficult.

At root this approach to defining the limits to adaptation is contingent upon the levels of risk a.s.sociated with climate change that are socially acceptable (Adger et al., 2009a). Given the unequal social and geographical distribution of costs likely to come from mitigation or adaptation (and acceptable, or 'unavoidable' impacts) led strategies, this is also a political question. The limits to climate change adaptation when framed in this way are cultural, social and political. This may produce some surprising outcomes. Kuhlicke and Kruse (2009), for example, show how local adaptive actions to reduce flood risk along the Elbe river, Germany, rely mainly on antic.i.p.ation and a.s.sumptions about state support, the latter actually being seen to undermine local resilience. In Australia risk of the re-introduction of mosquitoes carrying dengue disease is increasing as a consequence of government advice for households to adapt to increasing drought risk by installing domestic water tanks, the perfect breeding environment for the Ae. Aegypti mosquito (Beebe et al., 2009).

So, how can the limits of adaptation be ascertained? There are big lessons from the past in the failure of sophisticated civilisations from the Greenland Norse to Easter Island and the Maya of Central America. In each case a changing climate interacted with dynamic social pressures to undermine productive systems that overwhelmed adaptive capacity and led towards collapse (Diamond, 2005). Contemporary extractive land management systems are revealing their limits too. Sharer (2006) shows how contemporary industrial agriculture in lowland Guatemala and southern Mexico supports fewer people and generates more local environmental degradation than Mayan farming practices, forcing an intensifying drive to fell more forest for short-term productive gain. More than anything these cases tell us that the risk from environmental change is a product of social amplification the failure to recognise and respond in time to emergent risk rather than an intrinsic quality of the hazard itself. More contemporary evidence of the limits to adapt alongside climatic variability and extremes comes from the failure of coping and past rounds of adaptation made manifest by natural disasters from regional food security crises, to major hurricanes and floods, or local events such as flash floods, water logging and landslides that are local disasters (ISDR, 2009).

Disasters occur when socio-ecological systems coping capacities are overwhelmed (ISDR, 2004). There are four basic pathways for this failure. First, as a result of a lack of resources and the marginality that underpins this. This can force people (the poor and marginalised) to knowingly live or work in places exposed to risk in order to access other benefits such as close proximity to livelihood opportunities. Lack of resources also limits people's ability for self-protection. Second, as a result of a lack of information. Proactive adaptation is constrained when new hazards or vulnerability drivers emerge that are not planned for and may not be recognised until it is too late. This was the case in the 2003 heat-wave that claimed more than 35,000 lives in Europe, with earlier events, notably in Chicago, USA (Klinenberg, 2002), failing to stimulate learning and antic.i.p.atory adaptation in European cities. At the local level social networks can be as important as formal extension and advisory services for learning. Most acutely information fails when early warning is not provided (IFRC, 2005). Third, as a result of inst.i.tutional failures. This is the princ.i.p.al reason for physical infrastructural failure the proximate cause for many events. Inst.i.tutions fail to enable adaptation when those at risk and managing risk are not able to learn critically, but rather are trapped in cycles of marginal improvements of existing behaviour (see Chapter 4); when those at risk and their advocates cannot hold risk managers to account; and when information and resources cannot be used effectively or equitably (Wisner, 2006). Fourth, as a result of the speed of development and application of appropriate technological innovations. In South Asia, in the s.p.a.ce of a generation cell phone technology has enabled mobile phones to spread from being the preserve of the wealthy to a ubiquitous feature of urban and rural life alike with knock-on benefits including providing early warning for disaster risk (Moench, 2007). These accounts indicate the complexity of identifying limits to adaptation and the great sociological and geographical variation to be expected.

The argument presented in this book responds to these four strands of enquiry starting from a perspective of wishing to understand, rather than measure adaptation. This requires a broad lens, close to Fankhauser's comment that adaptation as a research field can be interpreted as a revision of sustainable development. Following the critical literature on sustainable development (for example, Grin et al., 2010), climate change adaptation is seen as a process not an object, with discrete capacities, actions and outcomes offering windows for observation. Elements that are subject to being contested in discourse (as different explanations for events and situation are presented) as well as materially (as different actors compete for the control and use of a.s.sets and resources). This approach also builds on a belief that the limits of adaptation are rooted in culture and society; they can be subjective but are mutable (Adger et al., 2009c). The primary aspiration of this work is to open debate on adaptation as a critical process. It uses adaptation as resilience, transition and transformation as a basis for this contribution. In one sense this can be seen as adding a novel line of categorisation to those discussed in the typology above. But a more fundamental aim is to highlight adaptation to climate change as a multi-layered process, with observed acts of adaptation potentially concealing or denying opportunities for alternative pathways that could lead to different social and socio-ecological futures. Making these three levels of adaptation transparent is an initial step in supporting actors at risk and managing risk in questioning the power relations that give shape to adaptation as observed. These are tools for a critical consciousness in climate change adaptation.