ORIGINAL ARTICLE

Climate change adaptation trends in small islanddeveloping states

Stacy-ann Robinson1

Received: 26 May 2015 /Accepted: 3 November 2015 /Published online: 18 December 2015# Springer Science+Business Media Dordrecht 2015

Abstract Small island developing states (SIDS) are among the countries in the world that aremost vulnerable to climate change and required to adapt to its impacts. Yet, there is littleinformation in the academic literature about how SIDS are adapting to climate change, acrossmultiple countries and geographic regions. This paper helps to fill this gap. Using a sample of16 countries across the Atlantic, Indian Ocean and South China Sea, Caribbean and Pacificregions, this study has two main aims, to identify (1) national-level adaptation trends acrossclimate, climate-induced and non-climate-induced vulnerabilities, sectors and actors, as re-ported in National Communications (NCs) to the United Nations Framework Convention onClimate Change (UNFCCC), and (2) typologies of national-level adaptation actions in SIDS. Itidentifies, codes and assesses 977 adaptation actions. These actions were reported as address-ing 47 climate and climate-induced vulnerabilities and 50 non-climate-induced vulnerabilitiesand were undertaken in 37 sectors by 34 actors. The paper proposes five typologies ofadaptation actions for SIDS, based on actions reported by SIDS. It specifically explores theimplications of its findings for global adaptation strategies. As this work establishes a baselineof adaptation action in SIDS, it can assist national governments to gauge their adaptationprogress, identify gaps in their adaptation effort and, thereafter, develop appropriate strategiesfor filling the gaps. It can also assist donors, whether bilateral or multilateral, to make more‘climate-smart’ investment decisions by being able to identify the adaptation needs that are notbeing met in SIDS.

Keywords Adaptation . Climate change . National Communications . Non-Annex I Parties .

Small island developing states (SIDS) . Sustainable development . Trends . United NationsFrameworkConvention on Climate Change (UNFCCC)

Mitig Adapt Strateg Glob Change (2017) 22:669–691DOI 10.1007/s11027-015-9693-5

* Stacy-ann Robinsonstacy-ann.robinson@anu.edu.au

1 Fenner School of Environment and Society, The Australian National University, 48 Linnaeus Way,Acton, ACT 2601, Australia

1 Introduction

Small island developing states (SIDS) are 571 countries that are particularly vulnerable to theimpacts of climate change. These impacts include increased temperatures, precipitation and sealevel rise (SLR) (Nurse et al. 2014). Since the landmark 1992 United Nations (UN)Conference on Environment and Development, SIDS have been recognised as a distinct groupof developing countries, based on their comparatively greater vulnerabilities (UN-OHRLLS2011). Chapter 29 of the Fifth Assessment Report of the Intergovernmental Panel on ClimateChange (IPCC) (also called ‘AR5’), which focusses on small islands, confirms that islandvulnerability is often ‘a function of four key stressors: physical, socio-economic, socio-ecological and climate-induced, whose reinforcing mechanisms are important in determiningthe magnitude of impacts’ (Nurse et al. 2014, p. 1635). Benamara (2013) and Kelman (2014)also highlight the social, economic and environmental vulnerabilities of small islands, includ-ing (1) rapidly increasing populations, (2) weak governance structures, (3) remoteness fromworld markets, (4) sensitivity to external and global shocks, (5) small size and (6) fragileecosystems that are susceptible to natural hazards. As a result of these and other vulnerabilities,climate change adaptation is a high priority in SIDS and continues to increase in importance onthe agendas of policy-makers (following McGray et al. 2007).

Climate change adaptation, according to AR5, is a process—‘of adjustment to actual orexpected climate and its effects’ (emphasis added) (IPCC 2014, p. 1758). This process, aconceptualisation also iterated by Klein et al. (1999) and Moser and Ekstrom (2010), can takeplace in natural or human systems and is either incremental or transformational. Incrementaladaptation aims ‘to maintain the essence and integrity of a system or process at a given scale’(IPCC 2014, p. 1758). It is an extension of actions and behaviours that already exist in order toavoid the disruption of a system (Berrang-Ford et al. 2011). Transformational adaptation aimsto ‘change the fundamental attributes of a system in response to climate and its effects’ (IPCC2014, p. 1758). It is more radical, involving significant changes in, for example, how countriesorganise, develop and manage their assets and resources (Kates et al. 2012). Little academicwork has been done on transformational adaptation, which is even more relevant in contextswhere change is more rapid or extreme, or where people are more vulnerable to climate change(Kates et al. 2012), as is the case with SIDS. However, this paper is more concerned withincremental adaptation—the changes within human systems at the national level that aresimilar to the changes that have already been occurring.

Identifying and implementing appropriate and/or effective adaptation responses in SIDS isa complex issue. The short- and long-term impacts of climate change in SIDS are notcomprehensively understood. AR5 acknowledges that the science is inexact, that there is great

1 Although the United Nations does not maintain an official list of SIDS (Bruckner 2013), the United NationsOffice of the High Representative for the Least Developed Countries, Landlocked Developing Countries andSmall Island Developing States (UN-OHRLLS) identifies 57 SIDS (UN-OHRLLS 2015). The nine SIDS in theAtlantic, Indian Ocean and South China Sea (AIMS) region are Bahrain, Cape Verde, Comoros, Guinea-Bissau,Maldives, Mauritius, Sao Tome and Principe, Seychelles and Singapore. The 28 SIDS in the Caribbean areAnguilla, Antigua and Barbuda, Aruba, Bahamas, Barbados, Belize, Bermuda, British Virgin Islands, CaymanIslands, Cuba, Curacao, Dominica, Dominican Republic, Grenada, Guadeloupe, Guyana, Haiti, Jamaica,Martinique, Montserrat, Puerto Rico, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Suriname,Trinidad and Tobago, Turks and Caicos Islands and United States Virgin Islands. The 20 SIDS in the Pacific areAmerican Samoa, Cook Islands, Fiji, French Polynesia, Guam, Kiribati, Marshall Islands, Micronesia (FederatedStates of), Nauru, New Caledonia, Niue, Northern Marianas Islands, Palau, Papua New Guinea, Samoa, SolomonIslands, Timor-Leste, Tonga, Tuvalu and Vanuatu.

670 Mitig Adapt Strateg Glob Change (2017) 22:669–691

difficulty in ‘generating formal climate scenarios at the scale of small islands’, and that‘vulnerabilities and adaptation needs are as diverse as the variety of islands between regionsand even within nation states’ (Nurse et al. 2014, pp. 1626 & 1635). These hinder robustadaptation decision-making and the willingness of national governments to embark on trans-formational adaptation programs. No single adaptation response will, however, be sufficient,and appropriate and/or effective adaptation responses, in the context of SIDS, will need to bemulti-scale, multi-component, multi-sector, multi-actor initiatives that address the four keystressors (physical, socio-economic, socio-ecological and climate-induced) identified in AR5.For the purposes of this paper, these stressors have been categorised as climate, climate-induced and non-climate-induced vulnerabilities.

AR5 recognises the dearth of academic literature on ‘the probability, frequency, severity orconsequences of climate change risks […] – or associated adaptation measures’ (emphasisadded) (Nurse et al. 2014, p. 1634). As a consequence, the status of national-level climatechange adaptation in SIDS, across multiple countries and geographic regions, is also under-explored in the academic literature, although much work has been done on developed and non-SIDS developing countries (e.g. Adger et al. 2003; Ford et al. 2011; Preston et al. 2011;Tompkins et al. 2010; Webb et al. 2013). This paper helps to fill this gap. Using a sample of 16SIDS across the Atlantic, Indian Ocean and South China Sea (AIMS), Caribbean and Pacific,the three main geographic regions in which SIDS are located, this study is a secondaryqualitative enquiry into climate change adaptation trends in SIDS. It has two main aims, toidentify (1) national-level adaptation trends across climate, climate-induced and non-climate-induced vulnerabilities, sectors and actors, as reported in National Communications (NCs) tothe United Nations Framework Convention on Climate Change (UNFCCC) (also called the‘Convention’), and (2) typologies of national-level adaptation actions in SIDS.

Divided into five main sections, the remainder of this paper (1) reviews seminal and otheracademic work related to climate change adaptation in SIDS, (2) describes the methods used,(3) presents the results of the analysis and (4) discusses them, particularly in view of theirimplications for global adaptation strategies, and (5) in concluding, situates the findings in theliterature and recommends that, in order to develop adaptation strategies that are based on asound, scientific understanding of climate change and its impacts, national governmentsshould focus on improving adaptation framing and reporting, identifying ‘good’ and appro-priate adaptation actions, and conceptualising adaptation as a multi-stage, iterative process.

2 Literature review

SIDS are not homogenous in nature and differ in ‘population, area, international powerposition, economic performance, and the extent of governmental involvement’ (Anckar2013, p. 13), which increases the complexity of climate change adaptation in these coun-tries—both from the perspective of understanding the diversity of vulnerabilities and identi-fying appropriate and/or effective adaptation actions, which will vary from country to country.There is a reasonable amount of literature on adaptation actions being undertaken at differentscales in individual SIDS. This approach, however, limits comparative analyses and theidentification of trends across multiple SIDS in different regions. Where comparative studiesare done, many are carried in the grey literature and case study countries are typically from thesame region (e.g. Dohan et al. 2011; Medeiros et al. 2011). Exceptions include Mavrogeniset al. (2014) who researched ecosystem-based adaptation (EbA) in Barbados, Seychelles and

Mitig Adapt Strateg Glob Change (2017) 22:669–691 671

Tonga, and Berrang-Ford et al. (2014) and Lesnikowski et al. (2015) who explored the driversof adaptation in 117 UNFCCC State Parties and the status of adaptation globally, respectively,both incorporating samples of SIDS but not discussing the findings for SIDS in any significantdetail.

One of the earliest academic studies into national-level responses to climate change inSIDS, by Pernetta (1992, p. 29), found sectoral approaches in the Maldives along with theGovernment pursuing international financing, raising awareness through Ministerial meetings,encouraging small islands to ‘band together’, and advocating for international environmentalprotection support for small islands. In the tourism sector, for example, Belle and Bramwell(2005) documented the preference of policy-makers in Barbados for policy formulation,collaborative cross-sectoral work and increased public awareness as appropriate adaptationpolicy responses to climate change. Mycoo (2014), also studying Barbados, highlighted a longhistory of coastal setback policies. Klint et al. (2012) found a number of implicit and explicitpolicies relevant to climate change adaptation in Vanuatu. Wong et al. (2013), however,concluded that the policy environment in Vanuatu is less conducive to supporting adaptation,when compared to Samoa.

Nunn et al. (2014) observed a reliance on top-down approaches to adaptation in PacificIsland Countries (PICs). Simpson et al. (2008) noted building regulations in Fiji being appliedto help prevent storm surge and SLR damage. Kuruppu (2009) cited the incorporation ofadaptation into the National Development Strategy in Kiribati. Guillotreau et al. (2012, p. 287)named institutional changes as ‘critical elements in the adaptive capacity and exposure/sensitivity of SIDS’—in fisheries, industrial and trade policy at various scales in countriessuch as Fiji and Papua New Guinea (PNG) (and in Mauritius and Seychelles). In theCaribbean, Tompkins (2005) found institutional and organisational changes, e.g. the passageof constitutional orders and regulations, and the creation of a formal Environment Departmentas evidence of adaptation to storm risk in the Cayman Islands.

Babinard et al. (2014) catalogued infrastructure projects in Kiribati, including the KiribatiClimate Change Adaptation Project III, which constructed coastal protection and otherfacilities. Duvat (2013) found a prevalence of longitudinal coastal stabilisation structures inKiribati with seawalls being the most common type of coastal protection structure. Studyingthe Federated States of Micronesia (FSM), Monnereau and Abraham (2013) also found thebuilding of seawalls along with the raising of houses as household-level responses to coastalerosion. Simpson et al. (2008) cited the Government of Maldives’ Safer Island DevelopmentProgramme, which, from 2004 onwards, provided infrastructure to adapt to natural disastersand the impact of climate change, including the establishment of building codes that wouldfacilitate vertical evacuation. Jones et al. (2012) later valuated adaptation options such asbuilding seawalls and other hard coastal infrastructure in the Maldives. The authors also notedthe contribution of public–private partnerships in establishing marine protected areas in PNG.In the Caribbean, Mycoo (2014, p. 51) noted that the generic adaptation strategies relating tosustainable tourism and integrated coastal zone management (ICZM) in Barbados focussed on‘protection and accommodation, as opposed to retreat’ along with hard beach protection, e.g.seawalls and breakwater structures.

Huang (1997), in exploring community-based ICZM in the FSM, noted the use ofconservation programs, requiring continuous data collection, public participation and trainedpersonnel. Fletcher et al. (2013) found traditional coping methods being used in response toclimate change and disaster risk in the Cook Islands, Fiji, Samoa and Vanuatu. Nunn et al.(2014) also found traditional systems of environmental governance in PICs. Mercer et al.

672 Mitig Adapt Strateg Glob Change (2017) 22:669–691

(2012) and Mercer et al. (2014) identified EbA initiatives in Caribbean and AIMS SIDS,respectively. Mycoo (2014) further documented the use of EbA to prevent coral reef damageand coastal erosion from SLR in Barbados. Hiwasaki et al. (2014) explored processes forintegrating local knowledge with science in reducing climate-induced risk and improvingdisaster preparedness, based on a project undertaken in Timor-Leste.

The above referenced studies provide a snapshot of adaptation actions being undertaken atdifferent scales in individual SIDS, grouped according to the intent and form of the action.They confirm the diversity of adaptation actions being undertaken to meet the diversity ofadaptation needs in SIDS. This literature review, however, suggests that some regions (e.g.Caribbean and Pacific) and islands (e.g. Barbados and Fiji) are more studied than others andthat some vulnerabilities (e.g. coastal erosion, SLR, storm risk or surge and water shortages)and sectors (e.g. tourism and coastal zone) either are typically the subject of academic enquiryor are those to which SIDS are undertaking the greatest number of adaptations. It also suggeststhat the gap in the literature relates to adaptation action across multiple SIDS and in differentregions, and to a typology of action that is appropriate for SIDS, given their special case.

3 Conceptual framework

This study draws upon elements of a conceptual framework by Moser and Ekstrom (2010),developed to diagnose barriers to climate change adaptation. That framework’s key structuralelements comprise the ‘interacting actors’, the ‘governance and larger socio-economic context’and the ‘system of concern’ to be managed for climate change (Moser and Ekstrom 2010, p.22028). Following the AR5 conclusion that climate change policies are best addressed byintegrating them with sustainable development strategies (Denton et al. 2014), this studysituates climate change adaptation within a broader sustainable development context. It furtheridentifies national governments as the actors of concern and SIDS as the system of concern. Itconceptualises adaptation as ‘a multi-stage and iterative process’ (emphasis added) (Klein et al.1999, p. 240), considered necessary for the realisation of sustainable development in SIDS.

3.1 Theoretical framework

This study’s design and methods are influenced by three theoretical perspectives—hermeneu-tics, resource dependence theory and systems theory. Hermeneutics is the ‘art of interpretation’and is ‘especially suitable for work of a textual and interpretive nature’ (Kinsella 2006, p. 1). Itfacilitates the isolation of themes in an effort to ‘make sense’ of the data (Sloan and Bowe2014, p. 1292). Resource dependence theory, as espoused by Pfeffer and Salancik (1978),purports that organisations (or countries) respond to the demands of actors exogenous to thembecause of their dependence on the resources of those actors. This perspective is applied to thestudy through an understanding that, as non-Annex I Parties (developing countries) to theUNFCCC, SIDS will (1) endeavour to satisfy their obligations under the Convention, and (2)acquiesce to the guidelines of the Convention, regarding the production and submission ofNCs. Systems theory, which ‘offers a framework for interpreting modern society’ (Lechner2000, p. 114), is relevant to questions of policy as it analyses how a system adapts to itsenvironment (Bevir 2007). These three perspectives support the analysis of policy-relevantdocuments such as NCs in an effort to understand how national systems respond to interna-tional obligations and how human systems adapt to actual or expected climate and its effects.

Mitig Adapt Strateg Glob Change (2017) 22:669–691 673

4 Methods

Following Lesnikowski et al. (2015), NCs submitted to the UNFCCC were used to identifynational-level adaptation actions. In accordance with Articles 4 and 12 of the Convention,State Parties are required to ‘[f]ormulate, implement, publish and regularly update nationaland, where appropriate, regional programmes [sic] containing measures to […] facilitateadequate adaptation to climate change’ and to ‘[c]ommunicate … information related toimplementation [of the Convention through NCs]’ (United Nations 1992, pp. 10 & 11). NCsare intended to be the primary platform for ‘sharing information, assessing implementation andmonitoring progress’ with respect to the UNFCCC (Breidenich 2011, p. 5; Ellis et al. 2011, p.10). Non-Annex I countries such as SIDS are encouraged to include in their NCs, among otherthings, information on uncertainties, their vulnerability to climate and climate-induced risksand impacts, and high-priority adaptation strategies and measures (Lesnikowski et al. 2015).NCs are, therefore, considered a consistent, standardised data source across countries(Gagnon-Lebrun and Agrawala 2007). As policy documents, they are an iteration of thepriorities of governments, particularly in relation to adaptation (Lesnikowski et al. 2015),thereby providing a basis for comparing actions, priorities and commitments across SIDS.

4.1 Sample selection

Sixteen of the 57 SIDS identified by the UN Office of the High Representative for the LeastDeveloped Countries, Landlocked Developing Countries and Small Island Developing States(UN-OHRLLS 2015) are covered in this study: Antigua and Barbuda, Bahrain, Belize, CookIslands, Dominica, Fiji, Guyana, Jamaica, Kiribati, Mauritius, Samoa, Seychelles, Singapore,St. Lucia, Tonga, and Trinidad and Tobago. Of the 57 SIDS, those that are not a Party to theUNFCCC (N=17) and have not submitted at least two NCs (N=18) in English (N=6) wereexcluded (see Fig. 1).

4.2 Identification of adaptation trends

Content analysis, as described by Elo and Kyngas (2008), was used to identify national-levelactions included in the most recently submitted NC of each of the 16 SIDS—either a Secondor Third National Communication submitted between January 1, 2010 and December 31,2014. This involved reading through adaptation-related chapters in-depth and using manualopen coding procedures to determine (1) what is being adapted to (climate, climate-inducedand non-climate-induced vulnerabilities), who or what is adapting (actors and sectors) and (3)how adaptation is occurring (actions undertaken to achieve national climate change adaptationobjectives, excluding those described as having ‘add on benefits’ for adaptation). Investigatingthese three key framing questions, based on Eisenack and Stecker (2012) and Funfgeld andMcEvoy (2011) who build on Smit et al. (2000), allows for the ‘unpacking’ of the criticalelements of adaptation framing in SIDS. Zeroing in on adaptation at one spatial scale—thenational level—coded text was extracted and actions counted and analysed.

4.3 Identification of adaptation typologies

Note was made of emerging patterns, categories and themes, further used to derive adaptationtypologies. Following Biagini et al. (2014, p. 102), all codes extracted were grouped into

674 Mitig Adapt Strateg Glob Change (2017) 22:669–691

‘families of like adaptation actions’, which then allowed for the identification of ‘a broad,higher order adaptation typology’ (Biagini et al. 2014, p. 102) for SIDS.

5 Results

5.1 Adaptation trends

5.1.1 Vulnerabilities

Forty-seven non-exclusive climate or climate-induced vulnerabilities were reported as beingaddressed across the 16 SIDS. Like vulnerabilities were grouped (e.g. saltwater intrusion and

Fig. 1 Sample selection criteria and list of small island developing states covered in this study

Mitig Adapt Strateg Glob Change (2017) 22:669–691 675

salinity) for a revised total of 28 (see Table 1a). There were 358 adaptation actions (or 37 % ofthe total number of actions identified) for which the climate or climate-induced vulnerabilities

Table 1 Count of adaptation actions by vulnerability in 16 small island developing states

NO. CLIMATE OR CLIMATE-INDUCED VULNERABILITY TOTAL

- Not specified 358

1 Rainfall, precipitation, runoff 271

2 Cyclones, hurricanes, monsoons, storms, winds 239

3 Temperature (air and sea surface), heat, humidity 232

4 Storm surge, coastal inundation, flood 223

5 Drought, dry conditions, fire 213

6 Sea-level rise, tide 197

7 Global change (general), global warming 180

8 Water quality and/or availability, turbidity, siltation, sedimentation 177

9 Soil/coastal erosion, landslide 134

10 Resource/species/biodiversity loss 74

11 Diseases - air-, food-, rodent-, vector-, water-borne 68

12 Coral bleaching/health 58

13 Wave action, tsunami 58

14 Saltwater intrusion, salinity 50

15 Water vapour, evaporation including evapotranspiration 42

16 Ocean acidification 20

17 Pests, insects 15

18 Air pressure 13

19 Seasonality 12

20 Solar radiation 8

21 Ocean current 7

22 Earthquake 7

23 Oceanic circulation 6

24 Cloud cover 5

25 Volcanic eruptions 4

26 Ice/glacier melting 3

27 Visibility 3

28 Lightning 1

NO. NON-CLIMATE-INDUCED VULNERABILITY TOTAL

- Not specified 578

1 Economic, poverty 245

2 Social 105

3 Development, industralisation, urbanisation, construction, infrastructure 79

4 Human activities (general), recreational activities 79

5 Pollution, waste 67

6 Population growth, demographic changes, displacement, migration 65

7 Food insecurity, malnutrition 48

8 Deforestation, desertification, mining 43

9 Agricultural activities, soil (in)fertility 42

10 Over-exploitation of resources, overfishing 35

11 Human settlements, housing 34

12 Technological 33

13 Land use changes 32

14 Environmental degradation 26

15 Elevation - low 25

16 Lack of data/information gaps/awareness 25

17 Stress (psychological, mental, physiological) and well-being 24

18 Cost of adaptation 14

19 Mortality 12

20 Dredging, land reclamation 10

21 Political 6

22 Energy 5

23 Location, insularity 5

24 Size 5

25 National security 3

26 Ownership of land, complex tenure arrangements 3

27 Absence of appropriate policy 1

28 Lack of insurance 1

29 Limited credit availability 1

30 Low/lack of entrepreneurial skills 1

a. Count of Adaptation Actions by Climate or Climate-Induced Vulnerability in 16 Small Island Developing States

b. Count of Adaptation Actions by Non-Climate-Induced Vulnerability in 16 Small Island Developing States

676 Mitig Adapt Strateg Glob Change (2017) 22:669–691

being addressed were not specified. Of the vulnerabilities specified, the most commonlyreported was rainfall (including precipitation and runoff; 271 actions or 12 % of specifiedclimate and climate-induced vulnerabilities) and the least commonly reported was light-ning (1 action). By region, of the climate or climate-induced vulnerabilities specified, thefour AIMS SIDS sampled (Bahrain, Mauritius, Seychelles, Singapore) reported respondingmost to air and sea surface temperature (including heat and humidity; 69 actions or 17 %of specified climate and climate-induced vulnerabilities for the region), SLR (includingtides) (54 actions or 13%) and rainfall (50 actions or 12%), and least to pests and insects(1 action), among others. The seven Caribbean SIDS sampled (Antigua and Barbuda,Belize, Dominica, Guyana, Jamaica, St Lucia, Trinidad and Tobago) reported respondingmost to hurricanes (including winds; 138 actions or 12 % of specified climate andclimate-induced vulnerabilities for the region), rainfall (130 actions or 11 %) and drought(including dry conditions and flood; 117 actions or 10%), and least to ice/glacier melting(2 actions). The five Pacific SIDS sampled (Cook Islands, Fiji, Kiribati, Samoa, Tonga)reported responding most to rainfall (91 actions or 12 % of specified climate and climate-induced vulnerabilities for the region), drought (83 actions or 11 %) and storm surge(including coastal inundation and flood; 80 actions or 10%), and least to air pressure (2actions).

By main island type, based on classifications used in the United Nations EnvironmentProgramme Island Directory (see Dahl 1991), atolls such as Kiribati, which have smallland areas and are susceptible to king tides, high waves, storm surges, water shortages,droughts and health risks (Campbell 2006), reported responding most to these vulnerabil-ities—SLR, storm surge, drought, water quality and/or availability (including sedimenta-tion, siltation and turbidity), and global change (general)/global warming. Jamaica, acontinental SIDS with high elevations and susceptibility to orographic or relief rainfalland river flooding (Campbell 2006), reported responding most to hurricanes (24vulnerability-identified actions or 18 %) and second-most to rainfall (22 actions or17 %). Volcanic SIDS are prone to cyclones and hurricanes; their rivers and streamsare vulnerable to flash flooding; they may also experience volcanic eruptions (Campbell2006). The highest number of adaptations in Dominica, a volcanic SIDS, for example,addressed hurricanes (38 vulnerability-identified actions or 16 %), drought (35 actions or15 %) and third-most, rainfall (25 actions or 10 %). Dominica also identified volcaniceruptions as a vulnerability in one instance.

Fifty non-exclusive non-climate-induced vulnerabilities were reported as being ad-dressed across the 16 SIDS. Like vulnerabilities were grouped (e.g. human settlementsand housing) for a revised total of 30 (see Table 1b). There were 578 actions (or 59 % ofthe total number of adaptation actions identified) where the non-climate-induced vulner-abilities being addressed were not specified. Of the vulnerabilities specified, economicvulnerability (including poverty) was the most commonly reported across all countries(245 actions or 23 % of specified non-climate-induced vulnerabilities) and all regions,averaging 26 %. The least commonly reported across all countries were the absence ofappropriate policy, lack of insurance, limited credit availability and low or lack ofentrepreneurial skills. Across AIMS and Caribbean SIDS, the second-most commonlyaddressed vulnerability was social (38 actions or 24 % and 59 actions or 9 %, respec-tively). For Pacific SIDS, human activities (30 actions or 10 %) posed the second-mostnon-climate-induced challenge, followed closely by pollution (including waste) and pop-ulation growth (28 actions or 10 % each) and food insecurity (26 actions or 9 %).

Mitig Adapt Strateg Glob Change (2017) 22:669–691 677

5.1.2 Actors

In addition to national governments, 33 non-exclusive actors were identified as being involvedin adaptation actions across the 16 SIDS. Like actors were grouped (e.g. private sector andindustry) for a revised total of 14 (see Table 2a). National governments undertook 45 % ofactions as sole actors. Behind national governments, other country governments (includingdonors and development banks), and regional organisations were involved in the most numberof adaptations (219 and 204 actions or 22 and 21 % of total actions, respectively). By region,in AIMS SIDS, other country governments were reported as being more involved (38 actionsor 18 %) than regional and intergovernmental organisations (IGOs), involved in 35 actions (or17 %) each. The Government of Seychelles reported the greatest involvement with othercountry governments; Singapore with IGOs. Caribbean SIDS reported being most engagedwith other country governments (121 actions or 26 %) and regional organisations (101 actionsor 22 %) with the Government of St. Lucia reporting almost equal engagement with regional

Table 2 Count of adaptation actions by actor and sector in 16 small island developing states

NO. ACTOR TOTAL

1 National governments 977

2 Other country governments (bilateral)/donors, development banks 219

3 Regional organisations - including sub-regional organisations 204

4 Intergovernmental organisations 180

5 Public, civil society, communities 165

6 Non-governmental organisations - including women's, youth and church groups 139

7 Researchers, universities, scientists, consultants 110

8 Private sector, businesses, industry, developers 68

9 Schools, teachers - primary, secondary 49

10 Home/land owners 25

11 Media 14

12 Farmers, fishermen 13

13 Local governments, district officers 10

14 Doctors, health workers 3

NO. SECTOR TOTAL

- Not specified 187

1 Coastal zone 163

2 Agriculture, forestry 160

3 Water 157

4 Environment (general), natural resources - including minerals, mining 156

5 Ecosystem, biodiversity (marine, terrestrial) 151

6 Education, information 114

7 Health, sanitation 103

8 Tourism 94

9 (Sustainable) development 85

10 Energy, utility 69

11 Finance, insurance, investment 66

12 Planning - urban, spatial, land use 62

13 Fisheries, aquaculture 54

14 Industry (general), economy, food production 36

15 Human settlements, housing 27

16 Construction 23

17 Early warning 18

18 Waste management - including sewage 17

19 Transport 14

20 Technology/innovation, communication - including telecommunications 11

21 Foreign affairs/relations, trade 8

22 Social, culture, heritage 8

b. Count of Adaptation Actions by Sector in 16 Small Island Developing States

a. Count of Adaptation Actions by Actor in 16 Small Island Developing States

678 Mitig Adapt Strateg Glob Change (2017) 22:669–691

organisations and other country governments. Antigua and Barbuda reported the greatestinvolvement with the private sector, particularly hoteliers, and Guyana with researchers anduniversities. Non-governmental organisations (NGOs) were a key partner of Pacific SIDSgovernments—involved in 102 actions (or 33 %). Youth groups were identified as a majorNGO actor—involved in 27 % of NGO action. Only in the Pacific SIDS was the churchidentified as an actor in climate change adaptation. Fiji reported the greatest donorengagement.

5.1.3 Sectors

Adaptation actions in 37 non-exclusive sectors were reported as being addressed across the 16SIDS. Like sectors were grouped (e.g. health and sanitation) for a revised total of 22 (seeTable 2b). There were 187 actions (or 19 % of the total number of adaptation actionsidentified) where sectors were not specified. Of the sectors specified, adaptations in the coastalzone, agriculture and forestry, and water accounted for 30 % of all sector-specified actions.Adaptations in the social, culture and heritage sector were the least commonly reported witheight actions. By region, AIMS SIDS reported the highest number of adaptations in theeducation and information (45 actions or 16 % of sector-identified adaptations for the region),coastal zone (39 actions or 14 %) and water (35 actions or 12 %) sectors. Seychelles reportedthe highest number of adaptations in these three sectors. Singapore reported the highestnumber of adaptations in the construction, health and sanitation, and transport sectors. Noadaptations in the waste management sector in AIMS SIDS were reported. Caribbean SIDSreported the highest number of adaptations in the coastal zone (93 actions or 12 % of sector-identified adaptations for the region), environment (general) (91 actions or 12 %) andagriculture and forestry (85 actions or 11 %) sectors. Guyana reported the highest number ofadaptations in the coastal zone. Its adaptations in the agriculture and forestry sector accountedfor 47 % of the region total. St. Lucia reported the highest number of adaptations in the healthand sanitation, energy, industry and finance sectors and, together with Antigua and Barbuda,accounted for 86 % of the adaptations in the tourism sector in the region. Of the adaptations inthe finance sector across all regions, 77 % were reported by Caribbean SIDS. No adaptationsin the social sector in Caribbean SIDS were reported. Pacific SIDS reported a similar numberof adaptations across four sectors—ecosystems (59 actions or 11 % or sector-identifiedadaptations for the region), agriculture and forestry (58 actions or 11 %), education andinformation (57 actions or 11 %) and water (55 actions or 10 %), with Cook Islands accountingfor 35 % of the adaptations in the water sector. Cook Islands also reported the highest numberof adaptations across several sectors, including education and information, environment(general) and ecosystems. Fiji and Samoa reported the highest number of adaptations in theagriculture and forestry, and planning sectors, respectively.

5.1.4 Actions

A total of 977 adaptation actions, including 191 unique actions, were identified in the mostrecently submitted NCs of the 16 SIDS. Like actions were merged (e.g. species database andbiodiversity inventory) and grouped (e.g. capacity needs assessments, technology needsassessments and gap assessments, irrespective of focus sector or natural resource, compriseneeds and gaps assessments), where logical, for a revised total of 64 actions (see Table 3). Themost commonly reported actions were vulnerability and impact assessments (including risk

Mitig Adapt Strateg Glob Change (2017) 22:669–691 679

assessments and profiles; 100 actions or 10 % of total actions). Where countries reportedundertaking sector-by-sector assessments, each sector covered was counted as one assessment.Where countries reported undertaking assessments covering multiple sectors, these werecounted as one assessment. Adaptation research, and workshops and training each accounted

Table 3 Count of adaptation actions in 16 small island developing states

NO. ACTION TOTAL

1 Vulnerability and impact assessments 100

2 Adaptation research - including surveys (resources, household, knowledge) 68

3 Workshops and training 64

4 Public awareness activities - including activities related to World Environment Day etc. 62

5 Monitoring of climate variables and resources 48

6 Project/program (without activity descriptions) 40

7 Coastal protection infrastructure 36

8 Climate modelling, predictions, scenarios 34

9 Data storage - national databases 28

10 Systematic observation 27

11 Information production and dissemination (print, radio, TV, videos) 26

12 Action plans 24

13 Official communications to international organisations 23

14 Stakeholder consultations and meetings 23

15 Other policy with climate change considerations 21

16 Needs and gaps assessments 21

17 Position/concept document - approach, strategy 18

18 Restoration and replenishment - beaches, littoral drifts, mangrove replanting 18

19 Creation of national climate change team, committee, task force etc. 17

20 Mainstreaming of climate change 17

21 Hard law - laws, bills, regulations 15

22 Management systems - improvements and upgrades 15

23 Implementation of policy 13

24 Conservation activities (general) 13

25 Adaptation assessments - sector, resources 11

26 Other assessments - including estimation of adaptation costs 10

27 Data collection 10

28 Designation of climate change focal point/responsible authority 10

29 Approved climate change policy 10

30 Participation in regional and international climate change meetings, lobbying 10

31 Institutional capacity building/strengthening activities 9

32 Reviews - of adaptation actions, institutions, policies, policy subprograms 9

33 Soft law - codes, guidelines 8

34 Establishment of protected areas 8

35 Mapping 7

36 National climate change divisions/units (within Government structures) 7

37 Management plans - species, natural/capital resources 7

38 Data sharing - regional and international databases and platforms 6

39 Infrastructure and other repairs/maintenance 6

40 National funding allocation 6

41 New/improved planting, farming, agricultural practices 6

42 Other hard infrastructure - including desalination plants 5

43 Other plans 5

44 Early warning systems 5

45 Spatial and other planning 5

46 Recommendations - of adaptation actions, policies, policy subprograms 4

47 Grants and scholarships 4

48 Postgraduate education programs 4

49 Propagation of new species 4

50 Non-governmental organisations - foundations, research centres 3

51 Implementation framework 3

52 Relocation - of people, communities 3

53 Regional/international funding allocation 3

54 Financial instruments and benefits - loans, insurance, subsidy 3

55 Adoption of planning documents 2

56 Sustainable development plans 2

57 Project/program proposals 2

58 Establishment of partnerships 2

59 Biodiversity inventory 2

60 Signatory to non-UNFCCC agreement 1

61 Draft climate change policy 1

62 Implementation of traditional methods 1

63 Voluntary schemes - buy-back 1

64 Inspections - schools 1

680 Mitig Adapt Strateg Glob Change (2017) 22:669–691

for 7 % of reported actions (68 and 64 actions, respectively), completing the top three mostcommonly reported actions across the sampled countries. Coastal protection infrastructureranked seventh. Among the various types reported, seawalls accounted for 28 %, followed bybreakwater structures and coastal protection units (11 % each).

By region, the 209 actions reported by AIMS SIDS accounted for 21 % of actionsundertaken across the 16 SIDS, averaging 52 actions per AIMS country sampled. Besidesvulnerability and impact assessments (19 actions or 9 %), the highest number of adaptationswere in the monitoring of climate variables and resources (16 actions or 8 %), climatemodelling, predictions and scenarios (13 actions or 6 %), data storage and adaptation research(11 actions or 5 % each). Seychelles’ involvement in data storage was the highest among thesampled AIMS countries. In the NCs of the Caribbean SIDS, 462 actions were identified.These accounted for 47 % of actions undertaken across the 16 SIDS, averaging 66 actions perCaribbean country sampled. Besides vulnerability and impact assessments (44 actions or10 %), adaptation research constituted 9 % of actions undertaken in the region withGuyana’s involvement in this area accounting for 43 %. Workshops and training (42 actionsor 9 %) completed the top three actions for the region—St. Lucia’s involvement was thehighest across all three regions, 37 %. The Pacific SIDS accounted for 31 % of actionsundertaken across the 16 SIDS, averaging 61 actions per Pacific country sampled. Of the 306actions counted for the region, 37 (or 12 %) were classified as vulnerability and impactassessments, 22 (or 7 %) as public awareness campaigns and 19 (or 6 %) as projects/programs.Cook Islands reported implementing the largest number of projects/programs among thesampled Pacific countries—63 %.

5.2 Adaptation typologies

From the revised total of 64 adaptation actions identified in Sect. 7.1 above, 20 families of likeadaptation actions were identified: (1) assessments, (2) research and systematic observation,(3) climate modelling, predictions and scenarios, (4) planning documents and activities, (5)finance, (6) laws, (7) organisations*, (8) responsible authorities*, (9) policies*, (10) policysubsystems*, (11) implementation, (12) hard infrastructure, (13) management, (14) conserva-tion, (15) technology and innovation, (16) monitoring programs, (17) reviews and recommen-dations, (18) information and awareness-raising, (19) education and training and (20)knowledge management (see Table 4). Definitions for the families marked with an asteriskare based on Dovers and Hezri (2010). Organisations are manifestations of laws and processesthat can be dissolved or changed. An example is a Climate Change Department in a nationalgovernment structure. Responsible authorities are decision-makers—organisations or personshaving delegated authority and/or legal jurisdiction to act. Examples include a national climatechange committee or a senior climate change officer in the government. Policies are ‘positionstaken and communicated by governments’ (Dovers and Hezri 2010, p. 222) and include, forexample, a national climate change adaptation strategy that outlines the adaptation challengealong with steps that will be taken to address it. Policy subsystems recognise that, within thepublic policy sphere, there are processes and structures for specific issues or sectors. A climatechange policy or a water policy with mainstreamed climate change considerations is anexample.

From these 20 families, five broad, higher-order adaptation typologies were detected andare being proposed: (1) observation and assessments, (2) planning, institutions* and policies,(3) implementation and management, (4) monitoring and evaluation and (5) education and

Mitig Adapt Strateg Glob Change (2017) 22:669–691 681

Tab

le4

Proposed

typologies

ofadaptationactio

nsin

smallisland

developing

states

NO

.A

CT

ION

FA

MIL

Y O

F L

IKE

AD

AP

TA

TIO

N A

CT

ION

SSU

B-T

OT

AL

NO

.P

RO

PO

SED

AD

AP

TA

TIO

NT

YP

OL

OG

YT

OT

AL

1V

ulne

rabi

lity

and

impa

ct a

sses

smen

ts2

Nee

ds a

nd g

aps

asse

ssm

ents

3A

dapt

atio

n as

sess

men

ts -

sec

tor,

res

ourc

es4

Oth

er a

sses

smen

ts -

incl

udin

g es

timat

ion

of a

dapt

atio

n co

sts

5A

dapt

atio

n re

sear

ch -

incl

udin

g su

rvey

s (r

esou

rces

, hou

seho

ld, k

now

ledg

e)6

Syst

emat

ic o

bser

vatio

n7

Clim

ate

mod

ellin

g, p

redi

ctio

ns, s

cena

rios

8M

appi

ng

9Im

plem

enta

tion

fram

ewor

k10

Act

ion

plan

s11

Sust

aina

ble

deve

lopm

ent p

lans

12O

ther

pla

ns13

Proj

ect/p

rogr

am p

ropo

sals

14Sp

atia

l and

oth

er p

lann

ing

15N

atio

nal f

undi

ng a

lloca

tion

16R

egio

nal/i

nter

natio

nal f

undi

ng a

lloca

tion

17Fi

nanc

ial i

nstr

umen

ts a

nd b

enef

its -

loan

s, in

sura

nce,

sub

sidy

18H

ard

law

- la

ws,

bill

s, r

egul

atio

ns19

Soft

law

- c

odes

, gui

delin

es20

Nat

iona

l clim

ate

chan

ge d

ivis

ions

/uni

ts (

with

in G

over

nmen

t str

uctu

res)

21N

on-g

over

nmen

tal o

rgan

isat

ions

- f

ound

atio

ns, r

esea

rch

cent

res

22D

esig

natio

n of

clim

ate

chan

ge f

ocal

poi

nt/r

espo

nsib

le a

utho

rity

23C

reat

ion

of n

atio

nal c

limat

e ch

ange

team

, com

mitt

ee, t

ask

forc

e et

c.24

Posi

tion/

conc

ept d

ocum

ent -

app

roac

h, s

trat

egy

25O

ffic

ial c

omm

unic

atio

ns to

inte

rnat

iona

l org

anis

atio

ns26

Sign

ator

y to

non

-UN

FCC

C a

gree

men

t27

Ado

ptio

n of

pla

nnin

g do

cum

ents

28D

raft

clim

ate

chan

ge p

olic

y29

App

rove

d cl

imat

e ch

ange

pol

icy

30O

ther

pol

icy

with

clim

ate

chan

ge c

onsi

dera

tions

31M

ains

trea

min

g of

clim

ate

chan

ge32

Proj

ect/p

rogr

am (

with

out a

ctiv

ity d

escr

iptio

ns)

33R

eloc

atio

n -

of p

eopl

e, c

omm

uniti

es34

Impl

emen

tatio

n of

trad

ition

al m

etho

ds35

Impl

emen

tatio

n of

pol

icy

36In

stitu

tiona

l cap

acity

bui

ldin

g/st

reng

then

ing

activ

ities

37C

oast

al p

rote

ctio

n in

fras

truc

ture

38O

ther

har

d in

fras

truc

ture

- in

clud

ing

desa

linat

ion

plan

ts39

Man

agem

ent s

yste

ms

- im

prov

emen

ts a

nd u

pgra

des

40E

stab

lishm

ent o

f pa

rtne

rshi

ps41

Infr

astr

uctu

re a

nd o

ther

rep

airs

/mai

nten

ance

42V

olun

tary

sch

emes

- b

uy-b

ack

43B

iodi

vers

ity in

vent

ory

44C

onse

rvat

ion

activ

ities

(ge

nera

l)45

Est

ablis

hmen

t of

prot

ecte

d ar

eas

46M

anag

emen

t pla

ns -

spe

cies

, nat

ural

/cap

ital r

esou

rces

47R

esto

ratio

n an

d re

plen

ishm

ent -

bea

ches

, litt

oral

dri

fts,

man

grov

e re

plan

ting

48N

ew/im

prov

ed p

lant

ing,

far

min

g, a

gric

ultu

ral p

ract

ices

49Pr

opag

atio

n of

new

spe

cies

50M

onito

ring

of

clim

ate

vari

able

s an

d re

sour

ces

51In

spec

tions

- s

choo

ls52

Ear

ly w

arni

ng s

yste

ms

53R

evie

ws

- of

ada

ptat

ion

actio

ns, p

olic

ies,

pol

icy

subp

rogr

ams

54R

ecom

men

datio

ns -

of

adap

tatio

n ac

tions

, pol

icie

s, p

olic

y su

bpro

gram

s

55St

akeh

olde

r co

nsul

tatio

ns a

nd m

eetin

gs56

Info

rmat

ion

prod

uctio

n (p

rint

, rad

io, T

V, v

ideo

s) a

nd d

isse

min

atio

n vi

a w

ebsi

tes

etc.

57Pu

blic

aw

aren

ess

activ

ities

- in

clud

ing

cont

ests

/com

petit

ions

, act

iviti

es r

elat

ed to

Wor

ld E

nvir

onm

ent D

ay58

Part

icip

atio

n in

reg

iona

l and

inte

rnat

iona

l mee

tings

e.g

. UN

FCC

C C

onfe

renc

e of

the

Part

ies,

lobb

ying

59G

rant

s an

d sc

hola

rshi

ps60

Post

grad

uate

edu

catio

n pr

ogra

ms

61W

orks

hops

and

trai

ning

62D

ata

colle

ctio

n63

Dat

a st

orag

e -

natio

nal d

atab

ases

64D

ata

shar

ing

- re

gion

al a

nd in

tern

atio

nal d

atab

ases

and

pla

tfor

ms

Tot

al97

7T

otal

977

278

Res

earc

h an

d sy

stem

atic

obs

erva

tion

95

Clim

ate

mod

ellin

g, p

redi

ctio

ns, s

cena

rios

41

Ass

essm

ents

142

1O

bser

vatio

n an

d A

sses

smen

ts

Polic

y su

bsys

tem

s*32

Plan

ning

, Ins

titut

ions

and

Pol

icie

s2

Law

s23

Org

anis

atio

ns*

10

Res

pons

ible

aut

hori

ties*

41

Fina

nce

12

Polic

ies*

4427

Impl

emen

tatio

n83

Impl

emen

tatio

n an

d M

anag

emen

t20

6

Har

d in

fras

truc

ture

41

Man

agem

ent

24

Con

serv

atio

n48

Tec

hnol

ogy

and

inno

vatio

n10

3

189

Plan

ning

doc

umen

ts a

nd a

ctiv

ities

Mon

itori

ng p

rogr

ams

544

Mon

itori

ng a

nd E

valu

atio

n67

Rev

iew

s an

d re

com

men

datio

ns13

237

Edu

catio

n an

d tr

aini

ng72

Kno

wle

dge

man

agem

ent

44

Info

rmat

ion

and

awar

enes

s-ra

isin

g12

1

5E

duca

tion

and

Kno

wle

dge

Man

agem

ent

*Definitionsas

perDoversandHezri(2010,

p.222)

682 Mitig Adapt Strateg Glob Change (2017) 22:669–691

knowledge management. The definition for ‘institutions’ is also from Dovers and Hezri (2010,p. 222)—‘laws, processes or customs serving to structure political, social, cultural or economictransactions and relationships in a society’. Based on these five typologies, observa-tion and assessment actions account for 28 % of adaptations across the 16 SIDS,followed by education and knowledge management (24 %) and implementation andmanagement (21 %), constituted primarily by on-the-ground adaptation activities and(to a lesser extent) mainstreaming. When countries are grouped according to region,Caribbean SIDS had the highest percentage of observation and assessment (30 %), theleading type of adaptation across all three regions, and implementation and manage-ment actions (23 %). A quarter of the actions undertaken in AIMS SIDS wereeducation and knowledge management–related, marginally higher than theirCaribbean and Pacific counterparts (24 % each). Planning, institutions and policiesaccounted for 24 % of actions in the Pacific, the highest across the three regions.

6 Discussion

6.1 Implications of results for global adaptation strategies

Considering the foregoing results, in order to develop appropriate adaptation strategies forSIDS and other equally or differently vulnerable countries that are based on a scientificunderstanding of climate change and its impacts, there is a need for national governments tofocus on: (1) improving adaptation framing and reporting in order to better support progressmeasurement and decision-making, (2) identifying ‘good’ and appropriate adaptation actionsand (3) conceptualising adaptation as a multi-stage, iterative process in order to help ensureongoing adaptation action that reduces vulnerabilities, bolsters resilience and achieves sus-tainable development.

6.1.1 Improving adaptation framing and reporting

Several gaps in SIDS’ adaptation framing were identified. Chief among these is the insufficientframing of the vulnerabilities and vulnerable populations being targeted and those not beingtargeted by a specific action. Framing is a process that allows ‘actors [to] construct andrepresent meaning [in order] to understand a particular event, process or occurrence’(McEvoy et al. 2013, p. 281). It is ‘[c]rucial in all policy practice’ as it determines ‘specificallywho and what is actually included, and who and what is ignored and excluded’ (cited inCornwall and Brock 2005, p. 1047). It involves decision-makers answering four key questions(Funfgeld and McEvoy 2011, p. 24), three of which are as follows:

1. Adaptation to what? – This study finds 47 climate and climate-induced and 50 non-climate-induced vulnerabilities being addressed; 37% of actions reported by SIDS did notspecify the climate or climate-induced vulnerabilities being addressed; 8% of actions werereported as responding to climate change (in general)/global warming.

2. Who or what adapts? – This study finds negligible mention of the vulnerable populationsand/or subpopulations being targeted; these were deduced, in part, from the actorsengaged in an action, alongside national governments; 19% of actions were not associ-ated with a sector.

Mitig Adapt Strateg Glob Change (2017) 22:669–691 683

3. How does adaptation occur? – This study identifies 977 adaptation actions; 10% of whichwere vulnerability and impact assessments, constituting a subset of observation andassessment actions, which accounted for 28% of all actions reported across the 16 SIDS.

The reasons for the above gaps may be many. However, in analysing the NCs of the 16SIDS, there appears to be an underlying, yet directing, ‘whole-of-country vulnerability to theimpacts of climate change’ notion, which overrides the need to link actions to vulnerabilitiesand vulnerable populations, implicitly or explicitly, particularly when reporting. It is importantfor SIDS to understand the relevance of the actions being undertaken and their implications forvulnerability reduction, resilience building and sustainable development, and be able toaccurately communicate these in NCs.

Another gap relates to the level of framing. SIDS are significantly engaged in observationand assessment actions, which are theoretical in nature (see McEvoy et al. 2013). Framing, inthe context of the 16 countries sampled, is, therefore, largely occurring at the conceptual leveland at the operational level where practical, tangible adaptation decisions and actions are madeand taken, but to a lesser extent. While there is no yardstick for measuring optimal framinglevels, it is critical for adaptation framing to also occur at the meta-level, which helps to shapethe values and beliefs influencing public discourse on climate change issues (see McEvoy et al.2013). As ‘[a]udiences rely on frames to make sense of and discuss an issue’ (Nisbet 2009, p.15), insufficient and/or inappropriate framing is a challenge for communicating climate changeand engaging stakeholders (Lakoff 2010; Nisbet 2009). The media, for example, plays asignificant role in shaping values and beliefs and in influencing public discourse, which shouldmake it a key stakeholder in national climate change adaptation efforts. Yet, the media wasnamed an actor in only 14 actions (or 1 %) undertaken across the three regions. Along with thechallenges stemming from framing focused at one or two levels, the extent to which the rightstakeholders are being engaged in national-level climate change adaptation action in SIDSshould also be examined and consequently addressed.

The deficiencies in how adaptation is framed in SIDS can, in part, be rectified by explicitlylinking adaptation actions with vulnerabilities, vulnerable populations, actors and sectors, andcommunicating these in NCs. In view of a global commitment to achieving sustainable devel-opment by 2030, NCs provide a platform for information sharing, implementation assessmentand progress monitoring, which is useful as SIDS, actors, sectors and populations are differentlyvulnerable and face a diversity of adaptation challenges. While the lower capacity of developingcountries to report in a ‘detailed fashion’ and the ‘weaker’ reporting requirements for non-AnnexI countries, including SIDS, have been noted as impacting Report quality (Breidenich 2011, p. 7),rectifying framing issues has flow-on benefits for sustainability, which is threatened by climatechange. Changes to the formal reporting requirements may, however, not be required in order forcountries to better communicate their adaptation framing and progress. What is required istweaking. Many SIDS NCs already contain lists of adaptation actions undertaken, and those thatwould ideally be undertaken should sufficient funding be forthcoming. These lists (and, in someinstances, tables) account for a significant portion of adaptation-related chapters. Tweaking theselists to reflect inventories of actions, projects and programs, organised by the scale of the action(e.g. local, national and regional), the vulnerabilities being addressed (climate, climate-induced ornon-climate induced), vulnerable populations being targeted, actors involved and sectors of focuswould allow national governments to quickly identify gaps in their adaptation effort and,thereafter, develop appropriate strategies for filling the gaps. It would also assist donors, whetherbilateral or multilateral, to make more ‘climate-smart’ investment decisions by being able to

684 Mitig Adapt Strateg Glob Change (2017) 22:669–691

identify the adaptation needs not being met in SIDS, thus making NCs more useful for nationaland international decision- and policy-making.

6.1.2 Identifying ‘good’ but appropriate adaptation actions

Following on the three of four framing questions discussed in Sect. 6.1.1 above, the fourth is,‘What is good adaptation?’ The answer to this question is neither simple nor straightforward,especially in a SIDS context where ‘vulnerabilities and adaptation needs are as diverse as thevariety of islands between regions and even within nation states’ (Nurse et al. 2014, p. 1635).Eriksen et al. (2011) concurred that not every adaptation response is a good one. This paperasserts that, in view of its results, good adaptation responses should fit the adaptation problemsbeing addressed, reduce the targeted vulnerabilities, manage uncertainty and increase adaptivecapacity, all while meeting a country’s sustainable development goals and objectives. Based onthis, a number of follow-on questions can be asked of each adaptation response. These includebut are not limited to the following:

1. Does the adaptation response address the critical vulnerabilities?2. Is sufficient attention being paid to a particular vulnerability?3. Will the adaptation response have any future, negative unintended consequences?

The first two questions relate to the extent to which an adaptation response is appropriate.The third relates to maladaptation, which is understood to be adaptation actions that can‘positively increase the vulnerability of other groups and sectors in the future’ (Barnett andO’Neill 2010, p. 211) as well as the failure or reluctance of governments to adopt uncertaintymanagement methods.

The results of this study show that adaptation actions in SIDS, in general, are addressing criticalvulnerabilities. When islands are grouped according to geophysical characteristics, atoll nations,for example, are responding most to their most critical vulnerabilities (e.g. SLR, storm surge,drought and water quality and/or availability). Vulnerabilities, however, vary between islandswithin nation states. Seychelles, for example, is constituted by 15 islands—12 continental, 2coralline and 1 atoll (as per Dahl 1991). An adaptation response developed forMahe, a continentalisland susceptible to orographic or relief rainfall and river flooding (Campbell 2006), may not beappropriate for Aldabra, an atoll with four main islands and susceptible to king tides, high waves,storm surges, water shortages, droughts and health risks (Campbell 2006). Additionally, the extentto which sufficient attention is paid to a particular vulnerability will also vary between islandswithin nation states, island countries and regions. Adaptation priorities differ and may not fit wellwith assumptions regarding the particular climate, climate-induced and non-climate-inducedvulnerabilities that should be addressed. Kiribati, for example, is classified as a lower–middle-income economy by the World Bank and has a per capita gross national income (GNI) rangingbetween US$1046 and US$4125 (World Bank 2015). It may be assumed that Kiribati wouldprioritise addressing economic (including poverty) vulnerabilities. Yet, its NC reported only 3actions (or 9% of instances where a non-climate-induced vulnerability was specified and 7%of itstotal number of adaptation actions) addressing these. Based on the number of associated actions, ahigher priority may have been managing human activities, for which seven actions were reported.On the contrary, Antigua and Barbuda, a high-income economy with a per capita GNI of overUS$12,736 (World Bank 2015), reported 66 actions (or 27 % of instances where a non-climate-induced vulnerability was specified and 87 % of its total number of adaptation actions) aimed at

Mitig Adapt Strateg Glob Change (2017) 22:669–691 685

addressing economic vulnerabilities. These two examples illustrate differences in adaptationpriorities and possibly deficiencies in adaptation framing. Whichever the case, it is important fornational governments to ensure that adaptation priorities are based on scientific and/or othercredible evidence and that any deficiencies in framing are corrected.

It is a hard task to determine whether adaptation responses will have any future, negativeunintended consequences, especially in view of the ‘prevailing uncertainties about climatechange’ (Fankhauser et al. 1999, p. 67) and the ‘time lag between changes in climate andchanges in institutions’ (Barnett and O’Neill 2010, p. 213). There is little evidence in the SIDSNCs that the potential maladaptive nature of adaptation responses was taken into consideration.However, maladaptation can be avoided. Barnett and O’Neill (2010) suggest that policy-makersscreen proposed adaptation actions against five maladaptation criteria and make a determina-tion, before committing resources. The five criteria are actions that ‘increase [the] emissions ofgreenhouse gases, disproportionately burden the most vulnerable, have high opportunity costs,reduce incentives to adapt, or set paths that limit the choices available to future generations’(Barnett and O’Neill 2010, p. 213). Although maladaptation may still be possible despitescreening, it is important for national governments to recognise this and take the necessarysteps to mitigate any future, negative unintended consequences of today’s adaptation response.

6.1.3 Conceptualising adaptation as a multi-stage, iterative process

Conceptualising adaptation as a multi-stage, iterative process that begins with observation andassessments and progresses into planning, implementation and management, monitoring andevaluation and which continuously integrates education and knowledge management activitiesas well as accords equal importance to each stage has a number of benefits. These includeensuring ongoing adaptation action, and identifying strategies for increasing the quantity,quality and success of actions associated with each stage, all while mitigating the bias againstobservation and assessment.

Sampled SIDS are most engaged in observation and assessment actions, which do notdirectly improve their adaptive capacity or the ‘ability of systems, institutions, humans, andother organisms to adjust to potential damage, to take advantage of opportunities, or to respondto consequences’ (IPCC 2014, p. 1758). Lesnikowski et al. (2015) classify actions such asvulnerability and impact assessments, classified as ‘observation and assessments’ in this study,as ‘groundwork’, which are preparatory in nature and which differ from ‘adaptation’ actions,which tangibly improve a country’s adaptive capacity. Observation and assessment actions are,however, a ‘[s]cientifically sound, theoretical approach’ that ‘can lead to the identification andimplementation of actions that specifically respond to identified vulnerabilities’ (Biagini et al.2014, p. 105). Given the difficulty in developing vulnerability indicators (Hinkel 2011), inarriving at agreement on a metric for national-level vulnerability (Füssel 2010), and in‘generating formal climate scenarios at the scale of small islands’ (Nurse et al. 2014, p.1626), heavy engagement in observation and assessment actions has its merits and shouldnot be interpreted, in absolute terms, as a non-contributor to improving adaptive capacity.Additionally, absence of long-range, quality-controlled climate data, observational records,satellite observations, and of the application of dynamical downscaling techniques (i.e.observation and assessment actions) in SIDS can hinder adaptation decision-making that isrobust to climate uncertainties (following Wilby and Dessai 2010).

Despite a focus on observation and assessments, the proportion of adaptation actions classifiedas implementation and management in SIDS is low—21 % of total actions reported by the 16

686 Mitig Adapt Strateg Glob Change (2017) 22:669–691

SIDS sampled. Notwithstanding this, countries are required to and can ‘make effective adaptationdecisions in the absence of accurate and precise climate predictions’ (Dessai et al. 2009, p. 111). Inthe case of SIDS where the impacts of climate change are certain to be experienced, therecommendations of Hallegatte (2009, p. 240) are appropriate: (1) choosing ‘no-regret’ strategiesthat will nevertheless be beneficial, (2) identifying and implementing adaptation actions that arereversible and flexible, (3) referencing the available climate models and projections in order todetermine safety or security margins for new investments and (4) using institutional or financialtools such as insurance schemes and early warning systems, all while incorporating a long-rangeperspective. SIDS and other equally or differently vulnerable countries can and should move onfrom the ‘traditional predict and provide’ modus operandi and towards planning, implementing,managing, monitoring and evaluating more concrete, ‘no-regret’ adaptation action, while contin-uously integrating education and knowledge management activities (Wilby and Dessai 2010, p.180), which will reap tangible benefits in current and future climates.

6.2 Future research agenda

This study, in establishing a baseline of adaptation actions across SIDS, creates opportunities forfuture research. In addition to exploring the co-occurrence of codes (e.g. what actions are used toaddresswhat vulnerabilities), a potential agenda could include,without being limited to the following:

1. Testing the findings of this study against an increased sample size and an expandednumber of data sources from which adaptation actions are identified, coded and assessed.

2. Finalising a coding template for SIDS, and exploring the extent to which it can be appliedto non-SIDS developing countries.

3. In the context of SIDS, defining and determining appropriate measures of ‘good’adaptation.

7 Conclusion

SIDS are a suite of developing countries that are uniquely but not equally vulnerable to theimpacts of climate change. The adaptation literature, though using individual and samples ofSIDS as lenses, does not provide sufficient data about the state of adaptation across multipleSIDS jurisdictions, across the three main geographic regions—AIMS, the Caribbean and thePacific. As a secondary enquiry into climate change adaptation trends across SIDS, this paperidentifies, codes and assesses adaptation actions reported in the most recently submitted NCsof 16 SIDS: Antigua and Barbuda, Bahrain, Belize, Cook Islands, Dominica, Fiji, Guyana,Jamaica, Kiribati, Mauritius, Samoa, Seychelles, Singapore, St. Lucia, Tonga, and Trinidadand Tobago. It finds 977 actions addressing 47 climate and climate-induced vulnerabilities and50 non-climate-induced vulnerabilities in 37 sectors undertaken by 34 actors. The mostcommonly reported vulnerabilities were rainfall (including precipitation and runoff; climate-induced) and economic (including poverty; non-climate-induced). Along with national gov-ernments, the most engaged actors were other country governments (including donors anddevelopment banks). The highest number of adaptations was undertaken in the coastal zonewhile the most commonly reported actions were vulnerability and impact assessments.

This study demonstrates and further confirms the diverse geophysical, economic and socialmakeup of island countries and the islands within these countries as well as the diverse

Mitig Adapt Strateg Glob Change (2017) 22:669–691 687

vulnerabilities, actors, sectors and adaptation actions across countries and regions. It providesevidence that strengthens the arguments made by Babinard et al. (2014), Duvat (2013) andMycoo (2014) on the dominance of seawalls as hard protection structures, Belle and Bramwell(2005) on the importance of public involvement in Pacific adaptation, Klint et al. (2012) on therange of implicit and explicit adaptation policies being developed and implemented, Kuruppu(2009) on the significance of climate change mainstreaming into existing policies, and Pernetta(1992) on the multi-sectoral approaches to adaptation in SIDS. It also provides a number ofnew insights into the negligible participation of the media in adaptation across the threeregions, the extent of the involvement of the church in Pacific adaptation, and the risingimportance of other country governments (as opposed to regional organisations) in national-level climate change adaptation efforts. Based on these diversities, this paper further demon-strates that a one-size-fits-all approach to climate change adaptation in SIDS may neither beappropriate nor effective.

The paper proposes five typologies of adaptation actions for SIDS, based on actionsreported by SIDS: (1) observation and assessments, (2) planning, institutions and policies,(3) implementation and management, (4) monitoring and evaluation and (5) education andknowledge management. In order to develop adaptation strategies, based on a scientificunderstanding of climate change and its impacts, for SIDS and other equally or differentlyvulnerable countries, it recommends that national governments focus on (1) improvingadaptation framing and reporting in order to better support progress measurement and deci-sion-making, (2) identifying ‘good’, robust adaptation actions for implementation now, allwhile avoiding maladaptation, and (3) conceptualising adaptation as a multi-stage, iterativeprocess in order to help ensure ongoing adaptation action that reduces vulnerabilities, bolstersresilience and achieves sustainable development.

Acknowledgments The author wishes to acknowledge constructive feedback on earlier drafts from A/Prof.Jamie Pittock, Drs. Ian Fry and Bob Webb and Profs. Ian White and Stephen Dovers at The Australian NationalUniversity, Australia, Prof Karen Edyvane at Charles Darwin University, Australia, and the three anonymousreviewers; and proofreading by Candice Gordon Williams and Tiffany Taylor. The author takes responsibility forall errors.

References

Adger WN, Huq S, Brown K, Conway D, Hulme M (2003) Adaptation to climate change in the developingworld. Prog Dev Stud 3:179–195

Anckar D (2013) Legislatures in small polities. In: Baldwin N (ed) Legislatures of small states: a comparativestudy. Routledge, Oxon, pp 12–20

Babinard J, Bennett CR, Hatziolos ME, Faiz A, Somani A (2014) Sustainably managing natural resources andthe need for construction materials in pacific island countries: the example of south Tarawa, Kiribati. NatResour Forum 38:58–66

Barnett J, O’Neill S (2010) Maladaptation. Glob Environ Chang 20:211–213Belle N, Bramwell B (2005) Climate change and small island tourism: policy maker and industry perspectives in

Barbados. J Travel Res 44:32–41Benamara H (2013) The special case of small island developing States. United Nations Conference on Trade and

Development, GenevaBerrang-Ford L, Ford JD, Paterson J (2011) Are we adapting to climate change? Glob Environ Chang 21:25–33Berrang-Ford L, Ford JD, Lesnikowski A, Poutiainen C, Barrera M, Heymann SJ (2014) What drives national

adaptation? A global assessment. Clim Chang 124:441–450Bevir M (2007) Encyclopedia of governance. Sage Publications, Inc., Thousand Oaks

688 Mitig Adapt Strateg Glob Change (2017) 22:669–691

Biagini B, Bierbaum R, Stults M, Dobardzic S, McNeeley SM (2014) A typology of adaptation actions: a globallook at climate adaptation actions financed through the global environment facility. Glob Environ Chang 25:97–108

Breidenich C (2011) Improving reporting of national communications and GHG inventories by Non-annex I parties under the climate convention. Natural Resources Defense Council, Washington,DC

Bruckner M (2013) Effectively addressing the vulnerabilities and development needs of small island developingstates. United Nations, New York

Campbell J (2006) Traditional disaster reduction in Pacific Island communities. Institute of Geological andNuclear Sciences Ltd, Avalon

Cornwall A, Brock K (2005) What do buzzwords do for development policy? A critical look at ‘participation’,‘empowerment’ and ‘poverty reduction’. Third World Q 26:1043–1060

Dahl AL (1991) Island directory: UNEP regional seas directories and bibliographies No. 35. United NationsEnvironment Programme, Nairobi

Denton F et al (2014) Climate-resilient pathways: adaptation, mitigation, and sustainable development. In: FieldCB et al (eds) Climate change 2014: impacts, adaptation, and vulnerability part a: global and sectoral aspectscontribution of working group II to the fifth assessment report of the intergovernmental panel of climatechange. Cambridge University Press, Cambridge, pp 1101–1131

Dessai S, Hulme M, Lempert R, Pielke R (2009) Do we need better predictions to adapt to a changing climate?Eos-Trans Am Geophys Union 90:111–112

Dohan R, Hove H, Echeverría D, Hammill A, Parry J-E (2011) Review of current and planned adaptation action:the pacific. International Institute for Sustainable Development, Winnipeg

Dovers SR, Hezri AA (2010) Institutions and policy processes: the means to the ends of adaptation. Wires ClimChang 1:212–231

Duvat V (2013) Coastal protection structures in Tarawa Atoll, Republic of Kiribati. Sustain Sci 8:363–379Eisenack K, Stecker R (2012) A framework for analyzing climate change adaptations as actions. Mitig Adapt

Strateg Glob Chang 17:243–260Ellis J, Moarif S, Briner G (2011) Core elements of national reports. Organisation for Economic Co-operation

and Development/International Energy Agency, ParisElo S, Kyngas H (2008) The qualitative content analysis process. J Adv Nurs 62:107–115Eriksen S et al (2011) When not every response to climate change is a good one: Identifying principles for

sustainable adaptation. Clim Dev 3:7–20Fankhauser S, Smith JB, Tol RSJ (1999) Weathering climate change: some simple rules to guide adaptation

decisions. Ecol Econ 30:67–78Fletcher SM, Thiessen J, Gero A, Rumsey M, Kuruppu N, Willetts J (2013) Traditional coping strategies

and disaster response: examples from the South Pacific region. J Environ Public Health 2013:1–9Ford JD, Berrang-Ford L, Paterson J (2011) A systematic review of observed climate change adaptation in

developed nations. Clim Chang 106:327–336Funfgeld H, McEvoy D (2011) Framing climate change adaptation in policy and practice. Victorian Centre for

Climate Change Adaptation Research, MelbourneFüssel H-M (2010) How inequitable is the global distribution of responsibility, capability, and

vulnerability to climate change: a comprehensive indicator-based assessment. Glob EnvironChang 20:597–611

Gagnon-Lebrun F, Agrawala S (2007) Implementing adaptation in developed countries: an analysis of progressand trends. Clim Policy 7:392–408

Guillotreau P, Campling L, Robinson J (2012) Vulnerability of small island fishery economies to climate andinstitutional changes. Curr Opin Environ Sustain 4:287–291

Hallegatte S (2009) Strategies to adapt to an uncertain climate change. Glob Environ Chang 19:240–247Hinkel J (2011) “Indicators of vulnerability and adaptive capacity”: towards a clarification of the science–policy

interface. Glob Environ Chang 21:198–208Hiwasaki L, Luna E, Syamsidik, Shaw R (2014) Process for integrating local and indigenous knowledge with

science for hydro-meteorological disaster risk reduction and climate change adaptation in coastal and smallisland communities. Int J Disaster Risk Red 10(Part A):15–27

Huang JCK (1997) Climate change and integrated coastal management: a challenge for small island nations.Ocean Coast Manag 37:95–107

IPCC (2014) Annex II: glossary. In: Barros VR et al (eds) Climate change 2014: impacts, adaptation, andvulnerability part B: regional aspects contribution of working group II to the fifth assessment report of theintergovernmental panel on climate change. Cambridge University Press, Cambridge, pp 1757–1776

Jones HP, Hole DG, Zavaleta ES (2012) Harnessing nature to help people adapt to climate change. Nat ClimChang 2:504–509

Mitig Adapt Strateg Glob Change (2017) 22:669–691 689

Kates RW, Travis WR,Wilbanks TJ (2012) Transformational adaptation when incremental adaptations to climatechange are insufficient. Proc Natl Acad Sci U S A 109:7156–7161

Kelman I (2014) No change from climate change: vulnerability and small island developing states. Geogr J 180:120–129

Kinsella EA (2006) Hermeneutics and critical hermeneutics: exploring possibilities within the Art of interpreta-tion. Forum: Qual Soc Res 7:1–16

Klein RJT, Nicholls RJ, Mimura N (1999) Coastal adaptation to climate change: can the IPCC technicalguidelines be applied? Mitig Adapt Strateg Glob Chang 4:239–252

Klint LM, Wong E, Jiang M, Delacy T, Harrison D, Dominey-Howes D (2012) Climate change adaptation in thePacific Island tourism sector: analysing the policy environment in Vanuatu. Curr Issues Tour 15:247–274

Kuruppu N (2009) Adapting water resources to climate change in Kiribati: the importance of cultural values andmeanings. Environ Sci Policy 12:799–809

Lakoff G (2010) Why it matters how we frame the environment. Environ Commun 4:70–81Lechner FJ (2000) Systems theory and functionalism. In: Turner BS (ed) The Blackwell companion to social

theory, 2nd edn. Blackwell Publishers, Inc., Malden, pp 112–132Lesnikowski AC, Ford JD, Berrang-Ford L, Barrera M, Heymann J (2015) How are we adapting to climate

change? A global assessment. Mitig Adapt Strateg Glob Chang 20:277–293Mavrogenis S, Kelman I, Mercer J, Kurvits T (2014) Comparing tools and methodologies for climate change

adaptation in Small Island developing States. National Technical University of Athens, AthensMcEvoy D, Fünfgeld H, Bosomworth K (2013) Resilience and climate change adaptation: the importance of

framing. Plan Pract Res 28:280–293McGray H, Hammill A, Bradley R, Schipper EL, Parry J-E (2007) Weathering the storm: options for framing

adaptation and development. World Resources Institute, Washington, DCMedeiros D, Hove H, Keller M, Echeverría D, Parry J-E (2011) Review of current and planned adaptation action:

the Caribbean. International Institute for Sustainable Development, WinnipegMercer J, Kelman I, Alfthan B, Kurvits T (2012) Ecosystem-based adaptation to climate change in Caribbean

Small Island developing States: integrating local and external knowledge. Sustainability 4:1908–1932Mercer J, Kurvits T, Kelman I, Mavrogenis S (2014) Ecosystem-based adaptation for food security in the AIMS

SIDS: integrating external and local knowledge. Sustainability 6:5566–5597Monnereau I, Abraham S (2013) Limits to autonomous adaptation in response to coastal erosion in Kosrae,

Micronesia. Int J Glob Warming 5:416–432Moser SC, Ekstrom JA (2010) A framework to diagnose barriers to climate change adaptation. Proc Natl Acad

Sci 107:22026–22031Mycoo M (2014) Sustainable tourism, climate change and sea level rise adaptation policies in Barbados. Nat

Resour Forum 38:47–57Nisbet MC (2009) Communicating climate change: why frames matter for public engagement 0065nvironment.

Environ: Sci Policy Sustain Dev 51:12–23Nunn PD, Aalbersberg W, Lata S, Gwilliam M (2014) Beyond the core: community governance for climate-

change adaptation in peripheral parts of Pacific Island Countries. Reg Environ Chang 14:221–235Nurse LA et al (2014) Small islands. In: Barros VR et al (eds) Climate change 2014: impacts, adaptation, and

vulnerability part B: regional aspects contribution of Working Group II to the Fifth Assessment Report of theIntergovernmental Panel of Climate Change. Cambridge University Press, Cambridge and New York, pp1613–1654

Pernetta JC (1992) Impacts of climate change and sea-level rise on small island states: national and internationalresponses. Glob Environ Chang 2:19–31

Pfeffer J, Salancik GR (1978) The external control of organizations: a resource dependence perspective. Harper& Row, New York

Preston BL, Westaway RM, Yuen EJ (2011) Climate adaptation planning in practice: an evaluation of adaptationplans from three developed nations. Mitig Adapt Strateg Glob Chang 16:407–438

Simpson MC, Gössling S, Scott D, Hall CM, Gladin E (2008) Climate change adaptation and mitigation in thetourism sector: frameworks, tools and practices. United Nations Environment Programme, Paris

Sloan A, Bowe B (2014) Phenomenology and hermeneutic phenomenology: the philosophy, the methodologies,and using hermeneutic phenomenology to investigate lecturers’ experiences of curriculum design. QualQuant 48:1291–1303

Smit B, Burton I, Klein RJT, Wandel J (2000) An anatomy of adaptation to climate change and variability. ClimChang 45:223–251

Tompkins EL (2005) Planning for climate change in small islands: insights from national hurricane preparednessin the Cayman Islands. Glob Environ Chang 15:139–149

Tompkins EL, Adger WN, Boyd E, Nicholson-Cole S, Weatherhead K, Arnell N (2010) Observed adaptation toclimate change: UK evidence of transition to a well-adapting society. Glob Environ Chang 20:627–635

690 Mitig Adapt Strateg Glob Change (2017) 22:669–691

United Nations (1992) United Nations framework convention on climate change. United Nations, Rio de JanieroUN-OHRLLS (2015) About SIDS: Country Profiles. United Nations. http://unohrlls.org/about-sids/country-

profiles/. Accessed June 18, 2015UN-OHRLLS (2011) Small island developing states: small island, Big(ger) stakes. United Nations office of the

high representative for the least developed countries. Landlocked Developing Countries and Small IslandDeveloping States, New York

Webb RJ, McKellar R, Kay R (2013) Climate change adaptation in Australia: experience, challenges andcapability development Australas J Environ 20:320–337

Wilby RL, Dessai S (2010) Robust adaptation to climate change. Weather 65:180–185Wong E, Jiang M, Klint LM, Dominey-Howes D, DeLacy T (2013) Evaluation of policy environment for climate

change adaptation in tourism 13:201–225World Bank (2015) Country and lending groups. World Bank. http://data.worldbank.org/about/country-and-

lending-groups. Accessed 20 Oct 2015

Mitig Adapt Strateg Glob Change (2017) 22:669–691 691