Questioning the assumptions: the role of vulnerability assessments in climate change adaptation

Abstract

A number of challenges prevent climate change adaptation activities from contributing to vulnerability reduction. Some of these relate to the nature and quality of vulnerability assessments, while others concern how well assessments link with adaptation actions. By highlighting four persistent assumptions about the role of vulnerability assessments in adaptation, this paper seeks to identify ways in which the practice of vulnerability assessment can better support progressive adaptation actions. This paper argues that, for adaptation actions identified within vulnerability assessments to be successful, there is a need for assessment to: (i) engage with marginalized perspectives; (ii) draw upon diverse knowledge domains; (iii) integrate scales of analysis with appropriate scales of action; and (iv) build political will. This requires vulnerability assessment to be considered as more than just a method for information generation on who is most vulnerable to climate change, where they are located and the underlying reasons for their vulnerability. In addition, improving the rigor and transparency of assessments and engagement with stakeholders during the assessment process can better facilitate vulnerability-focused adaptation.

Keywords:

• vulnerability assessment
• climate change adaptation
• participation
• stakeholder engagement
• scale

Introduction: distinguishing adaptation and vulnerability reduction

Addressing the root causes of vulnerability to climate change underpins the success of adaptation strategies, yet reducing vulnerability is proving to be one of the greatest challenges confronting governments, agencies and communities in their efforts to tackle climate change. In this paper we draw a clear distinction between adaptation and vulnerability reduction to climate change, and demonstrate how the latter is sometimes subsumed into the former, assuming they mean the same thing when this may not necessarily be the case. Through a discussion of the role of vulnerability assessments in climate change adaptation, and the linkages between assessments and adaptation, we examine some key assumptions apparent in the theory, policy and practice of adaptation and argue that there is a need to give increased attention to appropriate ways and methods of identifying the most vulnerable people, places, sectors and countries. This requires attention not only on the quality of the outcomes of vulnerability assessments, in terms of the information generated and options identified, but also on the process of undertaking vulnerability assessments, as how this occurs greatly determines how strongly assessments link with adaptation decision-making and actions, and the extent to which sensitive issues associated with the equity dimensions of adaptation are anticipated and addressed.

This paper shares a common understanding of vulnerability to that identified by Smit and Wandell (2006), where:

the vulnerability of any system (at any scale) is reflective of (or a function of) the exposure and sensitivity of that system to hazardous conditions and the ability or capacity or resilience of the system to cope, adapt or recover from the effects of those conditions. (p. 286)

Smit and Wandell (2006) consider adaptation ‘to be a process, action or outcome in a system (household, community, group, sector, region, country) in order for the system to better cope with, manage or adjust to some changing condition, stress, hazard, risk or opportunity’ (p. 282). Proceeding on the basis that vulnerability is strongly influenced by processes of marginalization and inequality (Bohle et al. 1994; Ribot 1996; Kelly & Adger 2000; Wisner et al. 2004; Mearns & Norton 2010), we recognize that actions that seek to reduce vulnerability entail some negotiation of the distribution of the costs, risks and benefits of development. This issue is rarely acknowledged and is probably part of the explanation as to why adaptation strategies appear to be failing to address persistent and intractable vulnerabilities, as identified by Nelson et al. (2007).

The concept and assessment of vulnerability

Vulnerability assessment has emerged as a specialized methodology for assessing how stresses are likely to have differential impacts and how systems are likely to respond to such stresses in terms of resistance, coping, adaptation and resilience. Vulnerability assessment is a structured approach to the collection and analysis of information that is largely complementary to other assessment methodologies such as environmental impact assessment (EIA), social impact assessment (SIA), strategic environmental assessment and health impact assessment. It is similarly focused on how the process and outputs of such assessments inform policy, planning, projects and programmes to prevent harm. Unlike efforts to address climate change in EIA and SIA (for a discussion see Burdge 2008 and Sok et al. 2011) vulnerability assessment takes the threats associated with climate change, rather than a proposed project or programme, as the entry point for analysis.

Much of the literature on vulnerability is concerned with how to define the concept and how to measure and assess vulnerability (Miller et al. 2010), although an appropriate definition is key to robust and transparent assessment. Some consideration is given to analysing vulnerability in particular contexts, with much of this work being made up of local case studies with limited theorizing (Zou & Thomalla 2008). Very little work exists that explicitly evaluates the success of vulnerability assessments in generating appropriate information and efforts to reduce vulnerability (van Aalst et al. 2008). So, how can the practice of vulnerability assessment better meet the demands of policy-and decision-makers, and communities, in finding ways to effectively reduce vulnerability?

As a result of the considerable academic work and policy debate regarding the definition of vulnerability, especially within the context of the United Nations Framework Convention on Climate Change (UNFCCC), greater clarity has emerged regarding the unifying features of vulnerability studies in terms of their focus on: the stress (otherwise known as threat, risk or hazard), the unit or system on which such stress is focused (people, ecosystems, localities, sectors, economies, etc.), the ways that stress is experienced (livelihood impacts, health outcomes, etc.) and responded to (actions, programmes, etc.) (Downing et al. 2005). Füssel adds another important dimension, that of time (Füssel 2007) – whether the focus is on present or future vulnerability.

The concept of vulnerability considers the underlying causes of the differential impacts of climate change and natural hazards on people, environments and economies. Key contributions to the concept of vulnerability have emerged from work in the area of disasters (Burton et al. 1978; Hewitt 1983; Blaikie et al. 1994; Oliver-Smith 1996; Bankoff 2001; Pelling 2003; Wisner et al. 2004), livelihoods and poverty (Chambers 1989; Chambers & Conway 1992; Moser 1998; Prowse 2003), food security (Sen 1981; Watts 1983; Watts & Bohle 1993; Bohle et al. 1994) and climate change (Klein & Nicholls 1999; Smit et al. 1999; Kelly & Adger 2000; Barnett 2001; Barnett 2003; Füssel & Klein 2006; Carter et al. 2007). This work reflects different epistemological and theoretical contributions that can be categorized according to geophysical, human ecology, political economy, constructivist and political ecology perspectives (McLaughlin & Dietz 2007). Each perspective allots different explanatory emphasis to the role of environmental processes, culture, power and political economic structures. The field of public health has its own approach to vulnerability predominantly referring to the vulnerability of individuals and populations to disease. Some areas of public health (such as HIV/AIDS) are now referring to ‘key affected populations’ – the change came from key affected populations themselves, not only for the shift to more empowering language but also because the terminology is inclusive of those who are positive and those affected by HIV, for example, orphans and widows.

Some of the challenges vulnerability assessment confronts, as outlined here, are common to social impact assessment in general, as documented by Burdge and Vanclay (1996) and Esteves et al. (2012). However, the complex nature of the concept of vulnerability creates particular challenges in terms of its assessment. In addition to the severe environmental and other stresses communities and countries currently confront, the threat of future climate change has highlighted the temporal nature of vulnerability in terms of past, present and future trends and shocks (Blaikie et al. 1994; Burton et al. 2002; Wisner et al. 2004). The challenge from an assessment point of view is to appreciate the interaction between present vulnerability, to current shocks and stresses, for example, current climate-related hazards, and future vulnerability associated with predicted climate change. Current climate-related impacts, such as the impacts of natural hazards and disease, already exact a considerable humanitarian, economic and environmental toll throughout the world, especially in developing country contexts (see IFRC, various). Addressing a theoretical and distant future vulnerability to climate change is unlikely to gain much support (particularly in developing country contexts) unless the empirical and observed vulnerabilities people currently confront form the initial step in vulnerability assessment (van Aalst et al. 2008). Thus a key challenge in assessment is to address the causes of current and future vulnerability.

Vulnerability is recognized as being shaped by multiple stresses (Turner II et al. 2003), which may include a combination of climate-related hazards, environmental degradation, food insecurity, poor health, economic change and social pressures. The reality of people's lives is that they are often concurrently responding to multiple stresses. This makes the task of assessment much more complex than the focus of early-generation vulnerability assessment; these generally focused on single stresses, such as flood risk. This evolution in thinking has contributed to a (re-)framing of vulnerability as a condition, rather than an outcome of a particular stress or hazard (Leichenko & O'Brien 2002; Downing et al. 2005; Eakin & Luers 2006).

Associated with the recognition of multiple stresses is that social and environmental systems together shape the vulnerability context, and thus an integrated approach to assessment is necessary to capture the complex interplay between social and environmental systems (Kasperson & Kasperson 2001; Turner II et al. 2003; Eakin & Luers 2006; Burdge 2008). This is often termed a systems approach and reveals the influence of resilience thinking in the field of vulnerability assessment (Eakin & Luers 2006; Miller et al. 2010). As such, assessment cannot rely on static indicators of physical systems alone, such as the projected frequency and magnitude of extreme events or sea-level rise, but rather the mapping of vulnerability requires a consideration of the social relations, institutions and context and how these might change over time, as highlighted by Füssel (2007). For example see Adger's (1999) pioneering study of the role of collective institutions and income inequality in vulnerability to climate change in Vietnam. Such factors may not necessarily directly correlate with physical space (Liverman 2001), making them difficult to map or represent spatially. This presents a challenge in assessment to be able to capture and communicate these factors in appropriate ways.

According to the Inter-governmental Panel on Climate Change (IPCC), vulnerability has three key dimensions: exposure, sensitivity and adaptive capacity (Carter et al. 2007). The capacity of societies, at multiple scales from the individual to the global scale, to cope and adapt to change is thus seen as a key determining factor of vulnerability. Assessments need to go beyond consideration of issues of exposure to also investigate issues that shape capacity in the face of threats. These issues might include values (for instance, the level of support for equity within and between generations), agency of individuals and communities to pursue adaptations, that is, their capacity to act (including claim of entitlements or advocacy of legal reforms), and the structural limitations of adaptation (which might include economic restraints, entrenched vested interests) (Sen 1981; Chambers 1989; Adger & Kelly 1999; Kelly & Adger 2000; Adger et al. 2009). For those involved in assessment, apportioning explanatory power to different processes influencing vulnerability is a significant challenge, but one that is crucial in determining where subsequent strategies should be focused and prioritized. Explanations of vulnerability require a focus on deeper, underlying structural causes rather than focusing on proximate causes (Blaikie 1994; Bohle et al. 1994; Wisner et al. 2004). This is one of the distinguishing features of a political ecology approach, whereby explanations for environmental change are historically situated with reference to wider scale processes of political economic change (Blaikie 1985; Blaikie & Brookfield 1987).

Vulnerability thus concerns the differential impacts upon and capacities of people to cope with and recover from change, whether that be slow onset or abrupt changes. Exact definitions, and associated approaches to assessment, are shaped by the purpose of each study and policy and practice domains they seek to inform. While vulnerability assessment has been applied in the context of natural disasters, food insecurity and health epi/pandemics for many years, its application is increasing rapidly in the field of climate change adaptation.

Adaptation assumptions

While policy attention on the mitigation of climate change has tended to overshadow adaptation, from the first Conference of Parties meeting of the UNFCCC the value of studies of vulnerability in guiding adaptation was recognized. According to Burton et al. (2002), studies of vulnerability were seen as a crucial first stage in adaptation within the UNFCCC, in order to allow subsequent capacity building for the preparation of adaptation, and then the facilitation of adaptation measures. Yet, strengthening the links between vulnerability assessment and adaptation planning and decision-making has proven to be a difficult challenge (Mastrandrea et al. 2010).

We wish to interrogate four assumptions apparent in both academic and policy debates on climate change adaptation through a focus on the role of vulnerability assessment. These assumptions are:

1.	Adaptation reduces vulnerability to climate change.
2.	Vulnerability assessments contribute to an improved understanding of vulnerability.
3.	Vulnerability assessments result in effective actions to reduce vulnerability and build adaptive capacity.
4.	Vulnerability assessments are central to the process of adaptation priority setting, targeting and resource allocation.

These assumptions were identified through a structured review of literature on climate change vulnerability and the emerging body of knowledge on adaptation in practice, as discussed below, as well as research undertaken by the authors in Cambodia, Vietnam and Fiji (Bowen et al. 2012a, 2012b, 2012c; Miller et al. 2013a, 2013b, 2013c), exploring the connections between vulnerability assessments and the development of adaptation strategies in the health and water sectors.

Assumption 1: adaptation reduces vulnerability to climate change

Adaptation to climate change is no easy matter; actions may fail to meet their objectives and may even increase vulnerability, resulting in ‘maladaptation’ (Barnett and O'Neil 2010). This may occur owing to poor conceptions of vulnerability as well as resistance to realizing the aims of adaptation strategies. Barnett and O'Neil (2010) provide an example of how investments in desalination plants and large inter-basin transfers, ostensibly to adapt to a drier climate, can increase vulnerability by raising the cost of water for those least able to afford such increases.

Adaptation actions tend to concentrate on where immediate benefits can be gained and actors can mobilize resources, but persistent and intractable vulnerabilities often remain despite adaptation actions (Nelson et al. 2007). This can occur because investors, whether governments, donors or the private sector, require tangible outcomes that are fairly immediate in order to justify their actions. Longer-term outcomes (such as improvements in a community's health status) are much more difficult to demonstrate and require a greater level of commitment. For instance, investment in structural or engineering measures to respond to climate change is currently growing, such as the construction of higher flood levees, coastal protection dykes, desalination plants, dams and barrages, with diverse non-structural measures, such as investment in public health and agricultural diversification, receiving less consideration and resources (Ford et al. 2011, p. 330). Hard engineering investments are likely to play an important role in adaptation, yet their benefits need to be weighed against the considerable ecological and social impacts they entail, including displacement. Moreover, consideration should be given to their potential to lock in inflexible responses rather than building for adaptive management responses that provide opportunities for an evolving understanding of changes from unknown future feedbacks to be responded to.

Assumption 2: vulnerability assessments contribute to an improved understanding of vulnerability

As vulnerability is socially and spatially differentiated, any vulnerability assessment requires clear and specific information on who is the most vulnerable to climate change, where they are located, when they are vulnerable as well as the underlying reasons for their vulnerability. Some reasons why the quality of information on vulnerability generated by an assessment may be poor relate to generic issues influencing the quality of research, such as: the poor quality or availability of data; lack of conceptual clarity on key terms; the use of inappropriate analytical frameworks; and application of inappropriate methods for the purpose of the assessment. The above issues may result in the findings of vulnerability assessments being unclear, illegitimate, contested or rejected for epistemological, political or economic reasons. Another methodological issue of concern regarding vulnerability assessment is that they may not be undertaken at a scale appropriate for action. This might occur if the scope is too locally based and specific for broad application, as is the case for many community-based and participatory assessments, or too macro to generate practical and applicable insights for particular settings, as observed in national- and regional-level assessments that tend to identify priority sectors or regions.

In the field of vulnerability studies there is a great deal of interest in the development of vulnerability indicators and indexes (Barnett et al. 2008). An over-reliance on indicators and indexes can potentially neglect important contextual factors, such as historical causes of vulnerability or the ways in which vulnerability is differentiated at more local scales, such as the individual or collective scale. While mapping vulnerability indicators can be useful in identifying countries or areas to target efforts, such as the attempt to map vulnerability to climate change in Southeast Asia (Yusuf & Francisco 2010), in isolation, macro studies such as these have limited efficacy in helping explain the underlying reasons for patterns of vulnerability, which are generally related to structural factors (Wisner et al. 2004).

Quantitative data used in vulnerability assessments, including climate model predictions and baseline health indicators, may be of poor quality, especially in developing country or remote area settings. However such data are often still used as a key source for the assessment. In the absence of quality baseline data, alternative approaches to data that draw upon community-based knowledge systems and analyse these against clear conceptual frameworks can be usefully employed. This mixed-methods approach is long evident in the practice of SIA in Indigenous contexts where, for example, oral histories can usefully provide an understanding of historical changes in human and environmental conditions (see Ross 1990; Lane et al. 1997). In addition, climate models may be extrapolated from other countries, leading to inaccurate predictions. These factors present a misleading picture of vulnerability owing to this emphasis on ‘collecting the numbers’. Conversely, assessments based on unstructured qualitative data may help give an understanding of a particular context but make monitoring and comparison over space and time difficult. Instead, it has been suggested by the World Health Organization (World Health Organization (WHO) 2012) that a mix of qualitative and quantitative data be collected. The advantages of using both types of data need to be clearly explained to stakeholders so as to advocate their use and thus enhance the reliability of vulnerability assessments.

Other factors contributing to poor quality of information on vulnerability may be the focus on proximate rather than underlying causes of vulnerability, such as considering issues of exposure (location of people and assets) without considering the role of historical processes (poor tenure, lack of regional planning, lax building codes) that contribute to this exposure. To gain a robust understanding of the causes of vulnerability, there is a need to engage with diverse stakeholders and gain an appreciation of their agency and capacity.

A final concern regarding the knowledge base on vulnerability is that the results of vulnerability assessments are not always easily available, especially those assessments conducted by non-governmental organizations (NGOs) or government departments.

Assumption 3: vulnerability assessments result in effective actions to reduce vulnerability and build adaptive capacity

In addition to problems associated with the quality of information contained in vulnerability assessments, there are other reasons why the linkages between vulnerability assessment and adaptation actions are often quite weak. Vulnerability assessments are not always embedded in processes that ensure that the necessary funds and resources required for adaptation are allocated. A case in point is the UNFCCC National Adaptation Programme of Action (NAPA) process, where difficulties accessing funds mean there is an outstanding number of adaptation projects that remain unfunded (LDC Expert Group 2011). A related issue is that assessments may also fail to properly engage with relevant actors to allow ‘buy-in’ or support of actions identified in assessments.

One of the biggest challenges is that the agency undertaking the assessment may not have responsibility for implementing the identified actions or the recommended strategies lie outside their organizational mandate. This issue can partly be overcome through engagement with the necessary actors responsible for implementation during the process of undertaking a vulnerability assessment. This is a common strategy employed by the International Federation of the Red Cross and Red Crescent Societies when they undertake their Vulnerability and Capacity Assessments (VCA) (IFRC 2007).

A related issue is that the agency undertaking the assessment may lack the necessary resources, both human and financial, to invest in the strategies identified. For instance, a very small number of the projects identified in the Cambodian NAPA (Royal Government of Cambodia 2006) have been funded. This is a growing concern that is likely to intensify with the continuing gap between global adaptation funds promised and actual adaptation funds disbursed (Caravani et al. 2010; Heinrich Böll Stiftung 2012). As noted in the practice of SIA and other assessments (Vanclay & Esteves 2011; Esteves et al. 2012), the involvement of key regulatory and implementing agencies and communities in the assessment process, from the scoping and definition of the terms of reference stages through to the management, monitoring and evaluation stages, is key to ensuring buy-in.

Assumption 4: vulnerability assessments are central to climate change adaptation decision making

Assessments can play a critical role in allowing actors, especially decision-makers, to identify particularly vulnerable social groups, regions and sectors, and identify and evaluate adaptation actions. In an international context, funding for adaptation activities is seen as an important source of additional development assistance for developing countries, especially less developed countries. Adaptation funds, as supported through the UNFCCC and related processes, are seen as additional to existing official development assistance contributions and are part of a wider process of redistributive justice associated with the burden of climate change (Adger et al. 2006). Yet there is considerable variation in the extent adaptation actions and policies actually emerge from a transparent and structured process of vulnerability assessment whereby priorities and targets are rigorously identified, justified and to some extent negotiated with stakeholders.

With the estimated US$28–67 billion a year required for adaptation in developing countries by 2030 (UNFCCC figures in Caravani et al. 2010), there are obviously big business opportunities in adaptation – new programmes, new markets, new products, new infrastructure, and in many cases old projects with a new justification now jostling for a slice of the ‘adaptation pie’. Large-scale adaptation projects are being proposed for financing in the absence of detailed assessments on vulnerability to climate change, especially the social dimensions, as seen in the case of the National Target Program to Respond to Climate Change in Vietnam, which so far has primarily proposed ‘hard’ adaptation options (McElwee 2010). This situation reflects a wider concern related to the marginalization of assessment in decision-making, as documented in the case of EIA and SIA. We contend that vulnerability assessments that are rigorous in the data, methods and engagement, as shown in the area of SIA (Esteves et al. 2012), are likely to support more transparent decision-making on adaptation options, targeting and priority setting in ways that help reduce the risk of political bias and manipulation.

Towards vulnerability-focused adaptation

As outlined above, there are several challenges concerning the extent to which vulnerability assessments link with effective, equitable and sustainable adaptation measures. Based on the above interrogation of key assumptions regarding adaptation, certain pre-requisites are identified as key to vulnerability-focused adaptation, as summarized in Table 1.

Table 1 Towards vulnerability-focused adaptation

Adaptation assumption	Pre-requisite for vulnerability-focused adaptation
Adaptation reduces vulnerability to climate change	Engage with marginalized perspectives
Vulnerability assessments contribute to an improved understanding of vulnerability	Draw upon diverse knowledge domains
Vulnerability assessments result in effective actions to reduce vulnerability and build adaptive capacity	Integrate scales of analysis with appropriate scales of action
Vulnerability assessments are central to the process of adaptation priority setting, targeting and resource allocation	Build political will

Engage with marginalized perspectives

Engaging with the knowledge and perspectives of the most marginalized or vulnerable members of society is critical to ensuring that vulnerability assessments understand the key determinants of vulnerability and identify appropriate options to reduce vulnerability. Such perspectives cannot be gathered through a reliance on quantitative methods alone. Owing to various social and economic barriers, as well as biases in the way research is undertaken (Chambers 1983), it is often hard to access the knowledge and experience of highly vulnerable people. Greater attention to this issue is needed in the design and implementation of vulnerability assessment, and there is much that can be gained from the experience of NGOs in undertaking participatory poverty assessments as well as the body of work on SIA in Indigenous settings (see Howitt 2005; O'Faircheallaigh 2009). However experience engaging marginalized social groups needs to be better documented. Sharing vulnerability assessments at local scales of action, nationally, and internationally, such as through information sharing initiatives under the UNFCCC and other initiatives such as WeADAP (see www.weadapt.org) is key to improving the practice of vulnerability assessment in this regard.

Draw upon diverse knowledge domains

Recognizing the high degree of uncertainty associated with future climate change, appropriate responses require drawing on multiple domains of knowledge, from the science of climate change modelling to indigenous knowledge of ecological change to community-based knowledge of coping strategies. As such, the adoption of multiple methods within vulnerability assessment is necessary to incorporate these different sources of knowledge where relevant. A toolkit approach that seeks to match appropriate methods with the purpose of vulnerability assessments, in each adaptation context, as exemplified by the International Federation of the Red Cross and Red Crescent Societies when they undertake their VCA approach (IFRC 2007), rather than predetermined selection of methods, is a constructive way forward. Such an approach better ensures that assessment is driven by the issues of significance in each context rather than preferences around methods or the availability of data.

Integrate scales of analysis with appropriate scales of action

In light of experiences from other environmental policy domains, recommendations for policy-makers from vulnerability assessments need to be of sufficient specificity to inform actual adaptation operations and implementation schemes (Mastrandrea et al. 2010). This raises the need for scale to be explicitly considered in the way data are collected and analysed, and that engagement occurs with communities and appropriate stakeholders involved in decision-making and action at different levels as part of the vulnerability assessment process.

Approaches to vulnerability assessment continue to be polarized between local, place-specific and community-based approaches, and macro-scale, indicator-, model- and scenario-based approaches. Innovative ways to integrate the findings and processes associated with each approach (data, methods, frameworks, engagement) is key to effectively addressing the multiple scales in which climate change impacts and adaptation actions occur. Identifying ways in which data, findings and experience generated from across studies, in relation to specific sectors, environmental risks or social groups, can be synthesized is an important element of integration.

Build political will

Assessments need to be properly embedded within appropriate social and institutional settings of accountability, as vulnerability assessments are as much artefacts of knowledge and the data captured at a particular time as they are reflections of dynamic social values, norms and power relations. Following through on identified needs of adaptation can be considered an ethical responsibility, whether at the scale of the agency undertaking a community-based assessment or at the scale of adaptation financing where the global community needs to urgently disburse its promised funding commitments.

In order to support the follow-through on adaptation activities, the involvement of policy-makers and key decision-making stakeholders is required from the beginning of a vulnerability assessment process. In many cases it is too late and too difficult to engage with policy-makers once the vulnerability assessment is complete. One way to support this goal is through the linking of adaptation activities to current national policy documents and strategies where possible in order to dovetail adaptation and other related policies. Such strategies have been proposed for adaptation in the health sector, see Bowen and Friel (2012). This helps to reduce duplication and competition between actors and sectors, and also demonstrates better value for money.

Conclusion

Change, whether environmental, political, social or economic, involves a redistribution of costs, risks and benefits, socially and spatially. The changes associated with adaptation, particularly deliberate, planned adaptation are the subject of intense negotiations as some are likely to benefit from particular adaptation options more than others. Civil society is also increasingly vocal in demanding more transparency on the justifications used for action (and non-action) on climate change, whether in the realms of mitigation, adaptation or negotiation. Vulnerability assessments can play a useful role in this process by clarifying how the costs, risks and benefits of adaptation actions (and non-action) are distributed socially, spatially and over time.

As vulnerability is largely determined by processes of marginalization and inequality, the extent to which equity is an explicit goal of adaptation is a key measure of its success. As Adger et al. (2005) argue, the legitimacy of adaptation measures rests heavily upon the authority of information drawn upon and the extent to which this reflects the values and perspectives of relevant stakeholders. So, while the legitimacy of the science of climate change has been well debated and tested in the public sphere, notably through the IPCC process, the social, spatial and temporal differentiation of vulnerability to climate change has received less attention. However, the legitimacy of findings on vulnerability is likely to gain greater scrutiny as limited adaptation funds and resources are prioritized and allocated between countries, and between regions and communities within countries.

In conclusion, no single approach to vulnerability assessment is appropriate for understanding a threat as complex as climate change. Rather, what is required is an awareness of the diverse approaches that exist, their strengths and limitations, and the multiple methods underpinning them. In this paper, by highlighting persistent assumptions about vulnerability assessment in the context of climate change adaptation we have sought to identify key issues of process, engagement and resource allocation that are vital to vulnerability-focused adaptation. Selecting the most appropriate approach to vulnerability assessment to match the intended purpose, whether that be knowledge generation, targetting and prioritization of options, and/or development of appropriate interventions to reduce vulnerability, needs to be driven by context. We contend that the success of assessments in supporting vulnerability-focused adaptation is contingent upon engaging with marginalized perspectives, drawing on diverse knowledge domains, integrating scales of analysis with appropriate scales of action, and a commitment to building political will. Strong engagement with both policy makers and affected stakeholders and communities as part of the vulnerability assessment process is key to the legitimacy of adaptation outcomes. This reinforces the point that vulnerability assessment is as much about the process of how assessment is conducted as the outcome of such assessment.

Acknowledgements

We thank the two anonymous reviewers for their constructive comments on an earlier draft of this paper. The usual disclaimers apply.

The research underpinning this paper was partly funded by an AusAID Australian Development Research Award (ref. no. 49834) and supported by a University of Melbourne Future Generation Fellowship formerly held by Fiona Miller.