Improving the context of and decisions made regarding adaptation and resilience requires a wide range of conditions and actions. For instance, programs in understanding and planning for risks, as well as mobilizing climate finance, have been put forward by dialogues including the Paris Agreement, the Climate Adaptation Summit in January 2021, and the Global Commission on Adaptation (GCA, 2019; UNEP, 2021). Depending on context, a critical element of this process may be integration of information about climate risk into local decisions (WMO, 2007) supported by understanding of the availability, dissemination and usability of weather and climate information.

Weather and climate information is an amalgam of processed data, products and/or evidence-based knowledge about the historical, current, and future climate conditions in a specific location covering various timescales (Figure 1) (Dinku et al., 2011; Lemos et al., 2012; Mason et al., 2015; Singh et al., 2018). A focus on information vis-à-vis data “implies that it has meaning and relevance within a given context” (Singh et al., 2018, p. 390), noting that climate information may complement existing local and traditional knowledge. Climate information is an important element of the design and implementation of climate risk management (Chen et al., 2021; New et al., 2022), although its characteristics and level of uptake vary with different approaches. Effective climate information use can support improved risk assessment based on enhanced temporal and spatial predictions (Dinku et al., 2011). At short timescales, locally relevant climate information including weather forecasts can inform preparedness planning, early warning systems, and response and relief operations within a disaster risk reduction (DRR) remit (see Figure 1). At medium-to-longer timescales, products such as outlooks, predictions and climate scenarios can assist resilience and anticipatory adaptation by identifying potential risks, and guiding investment and planning decisions (Jones et al., 2017).

To narrow the gap between climate science and adaptation action (Kirchhoff et al., 2015), a plethora of research has stressed the importance of tailoring climate information for decision-makers (Goddard et al., 2010; Kirchhoff, 2013; Knapp and Trainor, 2013; Singh et al., 2018; Orlove et al., 2020). This includes using dialogues with decision-makers (Vaughan and Dessai, 2014; Vaughan et al., 2018), considering usability (Dilling and Lemos, 2011; Lemos et al., 2012; Kalafatis et al., 2015), and promoting knowledge-action systems (i.e., comprehensive networks focused on advancing climate change adaptation products and services according to their impact and success) (Weichselgartner and Arheimer, 2019). As well, climate information is often discussed in tandem with climate services (Hewitt et al., 2012, 2017; Lourenço et al., 2016; Chen et al., 2021; Ranasinghe et al., 2021; New et al., 2022), i.e., activities focusing on generating and providing climate information to support ex-ante decision-making and climate-resilient development (Vaughan and Dessai, 2014; Ouédraogo et al., 2018).

Despite this awareness, there remain stubborn challenges to the mobilization and uptake of climate information, especially in climate-vulnerable, including small and low-income, countries (Jones et al., 2017, 2018; Harvey et al., 2021; New et al., 2022). Specifically, there is limited comparative and contextualized evidence for some countries, including Caribbean small island developing states (SIDS), in contrast to, for example, Europe (Porter et al., 2015; Bruno Soares et al., 2018; and sub-Saharan Africa: Vincent et al., 2020; Harvey et al., 2021). To address this research gap, we consider three questions relating to the current context of climate information use in the Caribbean: (1) how are the concepts of disaster risk reduction, adaptation and resilience in relation to the different timescales of climate-related decisions understood within the region?, (2) how are localized perspectives of available sources and current use of climate information manifested?, and (3) what are the perceived barriers and enablers of climate information use?

Recognizing the importance of local context and understanding, in this article we examine these questions with 26 Caribbean-focused respondents, including end-users and decision-makers, who work in a variety of climate-related sectors. The paper reviews their experiences and perceptions and focuses on how this information could be better contextualized for more effective use within the region. We offer thoughts on the support needed to reduce barriers and encourage enablers of climate information use, within the wider context of climate-resilient development. We note that since this research captures only a sub-set of perspectives the findings are intended to serve as indicative and a starting point for regional awareness and dialogue.

Taylor et al. (2016, p. 172) state that “climate sensitivity is both interwoven into and entrenched in all levels of Caribbean existence.” Although not a homogenous group, Caribbean countries feature broad similarities of physical characteristics, including geographical location, land area and topography. These feature alongside parallels of socio-economic and political aspects relating to small population sizes, economic dependency on climate-influenced primary industries, developing economic and governance systems, institutional legacies, high reliance on imports, and broad susceptibility to climate and non-climate-related shocks (Briguglio, 1995; Barnett and Dessai, 2002; ECLAC, 2011; Robinson, 2017; Thomas et al., 2020; Dookie and Osgood, 2021b). Such sensitivities, in turn, contribute to the reality of vulnerability to climate variations. While we note that there are diverse descriptions of Caribbean, from a geographical and ecosystem-based perspective, and as a region encompassing the Caribbean Sea basin and its bordering insular and coastal areas (Debels et al., 2017), we situate this study within the geopolitically aligned Caribbean Community (CARICOM) countries. In addition, we refer to Caribbean Small Island Developing States (SIDS) as a group of countries that face unique social, economic, and environmental vulnerabilities (UNOHRLLS, 2021), though recognizing the increased focus on the wider small islands contexts in gray and academic literature (Mimura et al., 2007; Nurse et al., 2014; Petzold and Magnan, 2019; Mycoo et al., 2022), as well as the Commonwealth Secretariat's attention to small states (Commonwealth Secretariat, 2021).

Historically, Caribbean countries have been heavily impacted by short-term climate variability and extreme weather events including numerous flash floods, droughts, and tropical storms (Pelling and Uitto, 2001; Dookie and Osgood, 2021b), as well as effects of seasonal and inter-annual phenomenon such as the El Niño-Southern Oscillation (Giannini et al., 2000). In some instances, due to the interaction between hazard events and the location's vulnerability (or propensity to be harmed), there have been disasters, with short- and long-term outcomes through human, material, economic and environmental losses, and impacts (Pelling et al., 2002; Rasmussen, 2004; Hochrainer, 2009; Hsiang, 2010; Hsiang and Jina, 2014; Hallegatte, 2015; Dookie and Osgood, 2021a). Given these outcomes, the need to reduce existing and prevent new disaster risk has been and remains important in national and regional capacity building efforts (Dookie and Osgood, 2021b), objectives which can be traditionally encapsulated within the concept of disaster risk reduction (DRR) (UNDRR, 2023). Further, the ongoing impacts and future threat of climate change within the region continues to drive the need for longer-term decisions and policy on adaptation (Giordono et al., 2020). Generally referring to “adjustments in ecological, social or economic systems in response to actual or expected climatic stimuli and their effect,” adaptation can come in many forms and is considered “a critical component of the long-term global response to climate change to protect people, livelihoods and ecosystems” (UNFCCC, 2023). As Caribbean countries look toward the future, the need to build robust or resilient capacity to withstand further potential threats is essential. As such, using an increasingly common concept within the scientific and policy community, regionally and globally, resilience can be seen as “the capacity of social, economic and environmental systems to cope with a hazardous event or trend or disturbance, responding or reorganizing in ways that maintain their essential function, identity and structure while also maintaining the capacity for adaptation, learning and transformation” (Ara Begum et al., 2022, p. 134).

Although these concepts are familiar and frequently used, they should be treated with caution as they may simplify broader discussion on their tenets and evolution of use. To assist, Rajabi et al. (2022) offer a historical narrative literature review of the concept of disaster risk management (DRM), which has evolved from mostly post-disaster response to a systematic and comprehensive risk management process with activities focused on four stages of mitigation, preparedness, response, and recovery. The terminology from UNDRR (2023) offers context on the distinct types of DRM and situates DRR (which includes prevention and mitigation) as the policy objective of DRM; a related term, disaster management, refers to the “organization, planning and application of measures preparing for, responding to and recovering from disasters”. While the above definition on climate change adaptation (CCA) from the United Nations Framework Convention on Climate Change (UNFCCC) is commonly referred to, we are aware of the broader discussions of this structure and the reflections on bringing the two sectors together. Although the concepts of DRR and adaptation may appear separate due to different international frameworks and spheres of action such as the long-term Paris Agreement and the Sendai Framework for Disaster Risk Reduction 2015–2030, there may be many similarities (Mercer, 2010), with a fundamental element of both being the motivation to reduce vulnerability and improve resilience (Kelman et al., 2015; Schipper et al., 2016; Kelman, 2017). The separation of climate change from wider contexts of DRR, as well as sustainable development, seems “counterproductive” (Kelman et al., 2015) and embedding climate change adaptation into DRR would serve a better purpose of enhancing the climate-related disaster resilience process (also see Thomalla et al., 2006; Dwirahmadi et al., 2013; Mercer et al., 2014; Schipper et al., 2016; Kelman et al., 2017; Islam et al., 2020; Flood et al., 2022; Nemakonde and Niekerk, 2022). We highlight that the knowledge on CCA may also require further consideration in the context of adaptation heuristics (Nalau et al., 2021), which is defined as “common sense, rule[s] of thumb guiding the conceptual framing of adaptation, the prioritization of adaptation policies and measures, and/or the pathways by which they are implemented” (Preston et al., 2015, p. 469), as well as “adaptive capacity” (Robinson, 2018; Nalau and Verrall, 2021). Lastly, the complexity of resilience as a concept needs to be considered. Once mostly considered as the opposite of vulnerability, the term resilience has evolved etymologically and conceptually, temporally and across disciplines (Graveline and Germain, 2022).

Appreciating the context of these concepts, we may understand their roles in shaping approaches to resilience in the Caribbean. Noting an estimate of the economic impact of climate change within the region from 5.6% to 34% of gross domestic product (Pulwarty et al., 2010), the region has taken formative action to understand and mainstream adaptation to climate change into national strategies and build local adaptive capacity and resilience (Tompkins, 2005; Pulwarty et al., 2010; Birch and Simpson, 2011; Robinson, 2018; Thomas et al., 2019). This is reflected in a variety of regional/sub-regional donor-funded programs and projects, as noted on the website of the CARICOM Climate Change Centre (5Cs), as well as National Communications and National Adaptation Plans to the United Nations Framework Convention on Climate Change (UNFCCC) (see Figure 2).

A review of limits to adaptation in Caribbean SIDS highlighted persistent issues of finance and organizational capacity at the forefront of challenges, with technical capacity, data, human resources, and understanding of climate effects hindering national-level adaptation (Robinson, 2018, see also Betzold, 2015). Operationalization of a longer-term focus on climate has been limited due to challenges of regional fragmentation, prioritization of current and shorter-term issues, and dependency on externally driven institutions, ideas, capital, and skill (Bishop and Payne, 2012). This is noted in a context where the “climate agenda in the region is almost wholly dependent upon multilateral institutions and international NGOs for impetus, finance and expertise” (Bishop and Payne, 2012, p. 1549). While there are nuances and differences across Caribbean countries, with some (including Dominica, St. Lucia, and Jamaica) currently showing more focus than others, it is important to note that an assessment of 89 adaptation planning documents from Caribbean countries highlighted that adaptation planning and appraisal is a recent focus within the region with room for much improvement (Thomas et al., 2019). This has created requirements for longer-term systemic planning across society (Rhiney, 2015), with implications for the demand for and use of relevant climate information.

There has been an increasing effort within the region to develop, disseminate and utilize climate information. As shown in Figure 1, there is a range of Caribbean-centered climate information sources available across diverse types and various timescales. The development of short-term weather-related climate information has been primarily within the domain of National Meteorological and Hydrological Services (NMHSs) in Caribbean nations (Mahon et al., 2018, 2019; CDB, 2020) with early responsibility of monitoring and forecasting weather for maritime and later aviation outlooks. Such information is largely utilized within watches and warnings as well as forecasts helpful for disaster response and relief operations (Guido et al., 2016, 2020; Dookie et al., 2019; Mahon et al., 2019; CDB, 2020; Wilkinson et al., 2021). With World Meteorological Organization-directed jurisdiction for the Caribbean region, the United States National Hurricane Centre assists in reconnaissance, monitoring and forecasting necessary for understanding tropical cyclone threats within the region, building further knowledge critical for preparedness planning. The regional Caribbean Institute for Meteorology and Hydrology (CIMH) operates at the boundary by supporting NMHSs with sub-seasonal (i.e., around 2 weeks to 2 months) and seasonal climate information in the form of regional climate outlooks including Temperature and Precipitation Outlook Maps, the Drought Outlook, the Wet Days and Wet Spells Outlook, and the Experimental Dry Spell Outlook (CIMH, 2018; Mahon et al., 2018; CDB, 2020). They also provide guidance through projects such as Early Warning Information System Across Climate Timescales (EWISACT) and outputs such as the Caribbean Climate Outlook Newsletter, and offer training on tools at their bi-annual Climate Outlook Forums, which collectively assist strategic planning and scenario building. Longer-term climate information in the form of climate scenarios is often the focus of the Climate Studies Group at the University of the West Indies, Mona, Jamaica. While this group aims to provide details of longer-term projections utilizing General Circulation Model data, it is noted that, as small islands, the Caribbean faces challenges relating to downscaling (UNFCCC, 2007; Campbell et al., 2011; Karmalkar et al., 2013; Stephenson et al., 2014; Taylor et al., 2016; CDB, 2020; WMO, 2020). As the coordinator of the Caribbean's response to climate change, the CARICOM Climate Change Centre (5Cs) situates itself as a resource hub on climate change efforts and serves as the executing agency for projects related to climate change (Rosenberg, 2020). In Figure 2 we also highlight all regional/subregional adaptation-related projects cited on the 5Cs website, noting that 7 of 17 referenced projects specifically reference climate data and information.

Based on these sources, it is considered that “there is a sufficiency of knowledge on which to premise in response to the climate change threat, where action is a must and not an option” (Taylor et al., 2016, p. 193). However, there are few regional studies which focus on whether that knowledge is being used and what may be the ongoing influences, barriers, and enablers. As such, a wide user-driven understanding of and reflection on the demand, mobilization, use and uptake of climate information within the region could be insightful. We note that such an understanding may encompass user experiences on climate information (i.e., their practical contact with and observation of facts or events), their perspectives (i.e., points of view or attitudes toward or way of regarding something), and perceptions (i.e., intuitive understanding, insight, and interpretation). Complementary research on user experiences, perspectives, and perceptions of climate information within the region is noted in Mahon et al. (2021), which analyzed the insights of tourism sector-focused respondents on the use, sources of, and barriers to using weather and climate information, as well as Stewart-Ibarra et al. (2019) which looked at the role of climate/weather data on the public health sector. Although these studies are sector specific, they highlight the importance of knowledge which “facilitates a better understanding of who uses what, how and why, and for what group of users tailored climate information could make the most difference” (Mahon et al., 2021, p. 4).

Given the current availability, it is unlikely that an increased quantity of climate information is a solely sufficient condition to ensure climate information use and uptake. Rather, there will likely be a range of factors that influence the use of climate information. As such, user-led insights can assist our contextual understanding of these influences and may offer useful entry points for interventions and cost-efficiencies of operational learning which can specifically help resource-scarce regions such as the Caribbean (Mahon et al., 2021). They may also shed light on possible challenges of the “who” and the “where” of climate information (Pasgaard and Strange, 2013; Pasgaard et al., 2015; Skelton et al., 2019), which may influence ownership and commitment to local adaptation efforts (Lahsen, 2016; Blicharska et al., 2017; Skelton et al., 2018). An understanding of local factors that hinder and promote the effective use of climate information in decision-making is deemed critical within the wider literature (e.g., Lemos et al., 2012; Kirchhoff, 2013; Vaughan and Dessai, 2014; Brugger et al., 2016; Nissan et al., 2019; Guido et al., 2020). These considerations fall broadly within the wider context of utilizing climate change risk perceptions as a framing tool to understand and evaluate climate services (Jones et al., 2015; Steynor and Pasquini, 2022), as well as discussions relating to the knowledge and reasoning capacities of individual to form relevant risk perceptions (Kahan et al., 2011). We contrast this context with the concept of reception, relating to the manner, way and/or quality in which climate information is received by the user (Hahn et al., 2016), as well as the framework of credibility (relating to scientific adequacy of the information and/or the credentials of the information providers), salience (referring to comprehension and access), and legitimacy (relating to perceived levels of representation, bias and/or participation in the information process) of scientific information (Cash et al., 2003; Guido et al., 2020). Further, we consider the importance of focus on individual thought processes and motivations to act (O'Connor et al., 1999; Wolf and Moser, 2011), as well as the value of understanding heuristics of adaptation (Nalau et al., 2021) using climate information for decision-making. Local voices on the understanding and use of climate information, framed in the theoretical and practical considerations described above, are integral to understanding barriers and encouraging enablers of climate information use within the region.

The objectives of the study were to understand the current context of climate information use within the region through a focus on respondents' understanding of the terms DRR, adaptation and resilience (considering the range of temporal scales of action of climate-related decisions), their perspectives on their current use and sources of climate information, as well as their perceptions of barriers and enablers of using climate information. This aim to capture detailed Caribbean-focused perspectives encouraged the use of semi-structured interviews as the main form of data collection. The qualitative nature of the semi-structured interviews allowed for reasonable depth of insight from a wide range of respondents, based on one-on-one discussion (without limitations in candor if done in a focus group setting). Although we had an interview protocol designed to address the three research questions introduced in Section 1, the interview format encouraged latitude to pursue leads and opportunities for follow-up while still retaining options for some standardized, closed-ended questions to give a basis for systematic comparison (Adams, 2015; Krueger and Casey, 2015).

Recognizing that the knowledge needed for adaptation is highly contextual (Ryan and Bustos, 2019), we first sought to understand the scales of action in terms of what respondents were referring to when they said they work on “climate-related decisions” (see Figure 1). Noting the many different definitions and interpretations of the terms disaster risk reduction, adaptation, and resilience, as described within the literature, we solicited their understanding of these concepts as a potential aspect of how it might affect the demand for and use of climate information. Next, we sought respondents' perspectives of climate information: how it related to different timescales and sources, and their views on its availability and utility (Dookie et al., 2019; CDB, 2020; WMO, 2020). In our initial communication with potential respondents, we included a summary of research interests, and highlighted a definition of climate information as given in the Introduction (Lemos et al., 2012; Singh et al., 2018).

Lastly, we encouraged open-ended discussion about barriers and enablers of climate information use. We asked respondents to share their thoughts on factors which could enable and hinder the use of climate information within the region. Then, to aid comparison, we asked respondents to order a list of eight given factors from most to least influential in terms of their relevance to demand for and use of climate information: financial, technical, capacity/personnel, legislative mandates, loud voices and champions, international drive, local awareness of climate information, donor attention, and “other.” The selection of these factors was based on the literature as outlined in Sections 1 and 2 (particularly Cash et al., 2003; Wolf and Moser, 2011; Lemos et al., 2012) and our understanding of challenges which may typically affect regional decision-making (Pulwarty et al., 2010; ECLAC, 2011; Rhiney, 2015; Mahon et al., 2018, 2019; Dookie et al., 2019). To avoid bias and minimize priming, we did not share this question nor the eight factors in the emailed interview protocol so that respondents were only aware of the identified factors and factor preference process at the interview. During the interview, respondents were asked for their interpretation of each factor to ensure consistency of use. We also requested additional factor suggestions, and asked respondents to order these against the original eight to elicit the relative factor importance.

For sampling, we followed a mainly purposive non-probability design, referring to the iterative “intentional selection of informants based on their ability to elucidate a specific theme, concept, or phenomenon” (Robinson, 2014). While this sampling method does not offer statistical representation of the wider population of interest (of unknown size), it does provide informed insights from a sub-sample that is likely to include a high proportion of what is a small total population. To develop our list of respondents including researchers and decision-makers who could likely be end-users of climate information, we first considered the range of climate-related sectors of interest within the region, including agriculture, climate adaptation, coastal resources and management, disaster risk reduction and management, economics, fisheries, health, hydrology and meteorology, land management, renewable energy, tourism, and water (Robinson and Gilfillan, 2017; Bush, 2018). Noting these sectors, we attended national, regional, and international webinars, reviewed Caribbean-specific academic and gray literature, and utilized personal contacts. To refine the research questions and minimize bias of the respondent sample, we held preliminary interviews with four resource persons from this list. This allowed for an improved informant awareness and selection as we benefited from suggestions of respondents raised by these contacts. As a note, often, some people were repeated suggestions or persons already on the list we had in mind. Such a process resulted in a diverse list of respondents drawn from various levels/areas of local, regional, and international government agencies, academia, local, regional, and international development agencies, private sector interests, chambers of commerce, and civil society. An effort was made to secure both a gender balance, and due to the nature of institutions working on related topics within the region, we also aimed to prioritize local/regional representation.

Using this process, we reached out to 60 persons via email as well as networking platforms such as LinkedIn. We received 32 responses of interest to participate, which we followed up by sharing more details on the research and the interview protocol, as well as information on consent and the modality of the interview. In the end, 26 persons completed our interview process. As agreed by consent form, interviewees have been kept anonymous and a detailed summary of respondents is not offered due to the small nature of the participant base, but Table 1 summarizes a selection of characteristics of the 26 respondents while Table 2 offers some individual insights. As shown, of the 26 respondents, 20 were persons within a regional background, while 6 were of international origin working on Caribbean-related topics (one of them working within the region for at least 10 years). Despite our efforts to ensure a gender balance, 10 of those interviewed presented as a woman. There is a variety in the locality of focus, with 5 persons offering that they worked mainly within national interests, 12 worked within a regional and/or multiple country interests, and 9 working across both national and regional interests. Through the interviews, we discovered that persons worked in, had experience with, and/or were from countries including Antigua and Barbuda, The Bahamas, Barbados, Dominica, Grenada, Jamaica, St. Lucia, St. Vincent and the Grenadines, and Trinidad and Tobago. There was also a variety of sectoral backgrounds, including physical sciences such as meteorology, climate and agriculture, and social sciences such as related to tourism, development studies and economics; they operated in roles relating to research, such as at local universities, as well as policy and practice, such as within NMHSs, and national, regional, and international agencies. In the summary information on respondents, we did not specifically identify the years of capacity at their current place of work since we felt that this did not reflect the culmination of efforts in other capacities. Several respondents hold simultaneous positions, and many were able to offer perspectives based on both current and previous experience within the region. As well, we did not identify their country of origin since many have worked/continue to work in multiple countries, which collectively influence their experiences and perceptions. While some respondents were able to give perspectives both from a regional and national viewpoint based on their background, due to the small sample size this research does not attempt to isolate individual country trends. As noted, we do not purport to incorporate probability sampling in our approach and the results should not be considered representative of national or regional estimates (Ryan and Bustos, 2019), or the sectors represented by the respondents.

Due to the ongoing coronavirus pandemic, we used the Zoom web-based conferencing platform which gave acceptable replicability of in-person interviews, convenience, as well as ease of use and cost-effectiveness (Archibald et al., 2019), in addition to being carbon-efficient. The modal average interview length was 1 hour.

The interviews were transcribed verbatim at the time of recording and corroborated against the recorded audio for confirmation. Transcripts were supplemented with further email communication and sharing of relevant documents; respondents were also emailed their factors of preferences for confirmation and were encouraged to add context. Interview transcripts were then coded manually, with open-ended questions coded according to inductive reasoning through a comparison of recurring points within the interview responses and compilation of emerging themes. This use of thematic analysis was important to us due to the method's relevance and power in “seeking to understand a set of experiences, thoughts, or behaviors across a data set,” and its design to focus on or shared meanings rather than “unique meanings or experiences from a single person or data item” (Kiger and Varpio, 2020). Thematic interpretation was considered within a general social constructivist research paradigm, and a selection of representative narratives was extracted to highlight the range within each theme.

Since there was one closed-ended question regarding factor preferences, this was coded deductively since we had already pre-defined the factors. However, since there was much debate by respondents on how to decide on the factors—often, respondents could not select one factor over another and had “tied” factors—we flexibly decided to integrate this process within our research. As such, a mental mapping of the process was designed to help visualize how decision-makers approach and prioritize the use of climate information within the region. We compiled this mapping using NodeXL, a free network analysis and visualization software package.¹

As this research is meant to offer a starting point on local perceptions of climate information use, we feel confident in the quality and suitability of the respondents who accepted our invite to participate. While maximal efforts were made to conduct further interviews, we feel reassured by the recurring themes and the petering trend of new ideas with each additional response inferring some level of saturation (Saunders et al., 2018). As caveats, we again note that our sampling method meant that we were able to only draw perspectives from a diverse sub-sample and as such we are not able to make generalized conclusions about the Caribbean-wide population of interest nor attain quantitative representation of national nor sectoral interests. As well, we were only focused on the perspectives of a wide range of persons within our sub-sample and have not focused on the inter-comparability of individuals' efforts or backgrounds, the scope of climate information use nor the extent of climate-related decisions taken by these individuals. Lastly, while all efforts were made to gain some insight from persons in many climate-related sectors, we are aware that this was not entirely possible both due to the intersectional nature of climate within different spheres of Caribbean action (there may be areas which we were not aware of), and that there were persons who either did not respond to our contact or could not participate in our research.

In this paper, we have discussed elements of the current context of climate information use in the Caribbean through an analysis of responses by 26 Caribbean-focused sectoral experts to the aforementioned research questions. Overall, there is interest in the role of climate information to boost the region's arsenal of knowledge to prepare for and respond to weather and climate-related impacts. Yet, while even persons with limited direct use are aware of climate information, all persons, regardless of context of use, sectoral or national background, identify a range of barriers of climate information use.

Based on Section 4.1, a first key finding of this study is that respondents' perspectives on the concepts of DRR, adaptation and resilience show various understandings of definitions and agreement of scope, irrespective of respondent background. While we asked respondents to offer their understanding of each of the concepts, few gave definitions and rather opted to discuss wider contexts and the possibility of some level of interchangeability. This understanding is akin to the findings of Saxena et al. (2017) which analyzed the responses of 35 climate change donors and project implementers in the Caribbean and found varying interpretations of “resilience.” While our study differs from Saxena et al. (2017) due to our wider respondent base, the outcome implication is similar in that different values and interpretations could complicate “communication and action across various actors working at different scales and in different local contexts” (Saxena et al., 2017, 157). Increased understanding and harmonization of the definitions and their use could facilitate improved collective communication of gaps and needs within the region and strengthen programmatic focus. This could include better alignment of project aims and outcomes—how they address DRR, adaptation and/or resilience—and potentially affect their eligibility for climate finance.

Considering the respondents' narratives in Section 4.2, there is broad awareness of climate information and the customary sources of information within the region. These findings confirm that efforts by NMHSs, the CIMH, the 5Cs, as well as the Climate Studies Group (as highlighted within Figure 2, and in Section 2), are likely reaching the intended end-users. However, compared to the study by Mahon et al. (2021) which only interviewed persons from the tourism sector, our cohort with a wider thematic/sectoral base did not explicitly identify as many specific information sources. It is uncertain whether this contrast of information sources could be due to the sample or the interview process, but both studies confirm similar barriers to climate information use including challenges relating to awareness, application/usability of information, and a limited capacity to utilize information (Mahon et al., 2021). As such, despite the long history of projects/programs which have supported provision of climate information for its use in the region (see Figures 1, 2), there is yet a presence of stubborn barriers and a failure to look beyond narrow assumptions about what drives information use (Sections 1, 2).

A further key finding and subsequent recommendation is the need for contextualized climate information, i.e., considering both island geography and local climate-related socio-economic influences (e.g., Skelton et al., 2018). This relates to the island setting, as well as the packaging and dissemination of information in ways that are tailored to, and understood, accessed, and utilized by, diverse groups of decision-makers, across a range of timescales. These insights reflect seminal work by Cash et al. (2003) (see Section 2) on the importance of credibility, salience, and legitimacy for effective use of scientific information (Cash et al., 2003, p. 8086). Firstly, interviewees described the lost opportunity of not connecting the direct utility of climate information with local and regional priorities (institutional bottom lines), highlighting a lack of salience. However, a challenge to achieving usability is the limited availability of localized information (with data management identified as a barrier), and the current elevated levels of uncertainty. This stirs elements of a lack of credibility, relating to scientific adequacy of the information and/or the credentials of the information providers. As one official put it, there was a “principle of optimal ignorance … [as it is] better to be roughly right, than precisely wrong.” Further, as expressed by one respondent, “information is only as good as understood by receiver.” This signals the need to better understand challenges relating to the interaction between awareness or observation of climate information, and its reception by the user community. As well, this insight underscores the need for improved translation of climate information into more usable formats (Marx et al., 2007; Lemos et al., 2012) and the specific need for encouraging a space for climate science translators (Enenkel and Kruczkiewicz, 2022) to assist with making climate science findings more accessible or useful for the community of practitioners.

Another key finding relates to the focus on donor attention and international drive, raising questions of legitimacy (i.e., perceived levels of representation, bias and/or participation in the information process). Although the term legitimacy was not specifically mentioned within the discussions and does not explicitly appear as a barrier or an enabler, the factors “donor attention” and “international drive” were very highly preferred (Section 4.3.3.). As well, the mapping of “tied” factors (Figure 4) shows that respondents considered them highly linked particularly with financial resources. Although respondents noted the value of international financial support and expertise, this linkage (manifest as influence) sometimes meant that project focus may not always align with country focus. For example, shifts in the use of the different concepts of DRR, resilience and adaptation could be likely driven by donors, with a recent surge in regional focus on resilience a possible partial outcome of donor agendas. An interesting facet of donor interventions was the apparent presence of “co-editing” (as mentioned by an interviewee) referring to instances where donors had already set out their plan of interventions, with cursory editing and rubberstamping of donor project descriptions and reports made by local counterparts to satisfy the requirement for local input. This contrasts to the more open co-production and co-design ideals of many intervention efforts (Dilling and Lemos, 2011; Jones et al., 2017; Vincent et al., 2018; Bremer et al., 2019), and as such draws attention to well-intentioned but likely misdirected strategies. Several respondents mentioned that perhaps donors did not fully understand the region or local/cultural contexts, with one person offering the recommendation: “do not bring [local partners] in at the end; [they] should be there at the start.”

Greater recognition of this need for localized climate information alongside local professional and cultural considerations is crucial. Toward these efforts, it was suggested that there be renewed attention to highlighting and exchanging best practices and success stories of how climate information is used, alongside information about financial savings (i.e., relevance to the economic bottom line), and other socio-economic benefits. While the general paucity of evaluation studies may make this difficult (Wall et al., 2017), it is an area for future priority. Furthermore, identification of the role of political awareness and changing political alignments as barriers suggests that climate service programs must be positioned more within independent/autonomous policy processes and institutional structures within the region. To minimize future challenges of legitimacy, as regional climate producers develop potential for local improvement and more contextualized information use (Guido et al., 2016, 2020; Mahon et al., 2019; CDB, 2020), it would be beneficial for experts such as those interviewed here to be more involved with climate information and service providers and/or regional boundary organizations to promote Caribbean ingenuity and innovation.

Concerns about donors are also related to a wider challenge of regional and national institutions being fragmented or siloed in their approaches to climate change and climate information use. Figure 2 shows that there has been a range of regional donor-funded efforts across the years, with more than 40% of these including a focus on climate information. However, the initiatives are disjointed and carried out by different donors over time. It is perhaps with this in mind that respondents noted the need for donors to coordinate more effectively and reduce elevated levels of consultation fatigue and duplication (especially given low human resource capacity levels) and minimize the risk of promoting poorly aligned agendas. Some respondents also noted the need for renewed regional collaboration, particularly in the context of SIDS, given opportunities to further develop discussions and alignment with Pacific SIDS. In addition, respondents suggested that states and the region as a whole work to balance climate issues with other development priorities more effectively (given competing interests), and develop local expertise to take ownership of issues, rather than rely on external approaches and consultants.

Respondents also noted the role of human resources as a constraint for using climate information (see Table 3). The sparse numbers of overall staff, especially in climate services, is perhaps distinctive about SIDS/small states. However, human resource requirements also feature in, for example, European countries, where critical enabling factors were identified as “the existing level of resources, capacity, and expertise in the organizations” (Bruno Soares and Dessai, 2016, p. 99). Furthermore, Caribbean interviewees mentioned elevated levels of attrition within government agencies despite significant investments in skills training. Figure 4 highlights the relevance and strong connections of this factor with technological resources, financial resources, and local awareness of climate information. Thus, together with financing and data salience and credibility, there also needs to be a complementary focus on investing in local staffing and their skills.

Collectively, our results highlight numerous barriers and enablers that relate to credibility, salience and legitimacy and the need to address these through improved co-production processes. These goals reinforce the need to focus on strengthening capacity (and in some cases the remit) of NMHSs and regional boundary organizations to work across boundaries, recognizing that a focus just on the information deficit model of uptake will not be enough. This requires climate scientists and the users of climate information to engage more and to create better enabling conditions to bring this about: to listen, understand and respond to the local and sectoral needs of the region (Guido et al., 2016, 2020; Porter and Dessai, 2017; Gerlak et al., 2018; CREWS, 2019; Mahon et al., 2019; Matthews et al., 2020). In this vein, an important initiative is the Consortium of Sectoral Early Warning Information Systems across Climate Timescales (EWISACTs) (see CIMH, 2018; CDB, 2020; Figure 1), designed to improve sectoral alliances for climate resilience and develop a range of sector-specific climate products. While such an alliance requires attention to both the climate information provider and user sides to effectively share what is “credible, usable, and relevant,” it also requires broader social support and institutional design and incentives to change (Porter and Dessai, 2017).

To this end, the key roles of established factors such as loud voices/champions and legislative mandates, as well as respondent-suggested political will and leadership, and the influence of system shocks are noted as significant enabling factors in the form of catalyzing conditions (Ara Begum et al., 2022; New et al., 2022). While respondents did not identify where these loud voices should come from, the prominence of critical climate-sensitive sectors such as agriculture, fisheries and tourism could be a starting point for seeking and encouraging champions. Interestingly, there may be several benefits from a renewed focus on the tourism sector due to its crucial position in the region, its dependence on the climate and private sector perspective on achieving a cost-effective value of climate information, and the high potential for public-private partnerships (for example, see Matthews et al., 2020; Mahon et al., 2021).

In terms of policy drivers, high level political will to promote more informed climate information use could have widespread effects throughout decision-making structures in countries. However, since many of the current conversations are retained within technical groups, this limits the awareness and motivation of politicians to improve the visibility and use of climate information. While the Caribbean Meteorological Organization (CMO) has plans for increased inter-ministerial discussions and sectoral strategic plans (Laing, 2021), there is a need to leverage these existing governance structures, learning from empirical evidence of barriers and enablers, to further coordinate a coherent strategic approach to adaptation in the region (Thomas et al., 2019). It is evident that there is yet a clear need for Caribbean governments to be more direct and detailed in their climate-related projects, programs, and policies.

Lastly, while this research has offered localized experiences and narratives regarding the wide possibilities of barriers and enablers of climate information use within the region, we recognize that having more climate information, or even having such a basic knowledge of experiences, is not sufficient to enable efficient uptake of climate information. This study does not suggest that Caribbean people per se are responsible for the ongoing challenges of climate information use within the region; rather, it reinforces the point that there are many interlinked enabling and catalyzing conditions that play a role in the awareness, applicability, and use of climate information. As well, while many of these conditions may have been exposed within this study, there are wider socio-economic and legacy structures in place within the region which mediate the observed conditions. These discussions merit consideration but are outside the scope of this study.

This research has investigated Caribbean-focused perspectives on the demand for and use of climate information within the region. Empirical evidence from semi-structured interviews was used to examine the understanding of common terms within the scope and timescales of climate-related decisions, the sources of and how climate information is understood, complemented by detailed local perspectives on barriers and enablers which mediate climate information use.

While Caribbean SIDS face severe climate impacts, locally relevant research on how climate information is viewed has been limited. This research provides insights which could not only assist climate adaptation processes within the region, but also offer relevance for other SIDS. We note that this study complements existing research—Caribbean initiatives have been developing and pursuing a range of climate-related efforts—and within this process, the utility and importance of climate information is evident. However, respondents identify numerous barriers, many of which are well-established in the literature and persistent over time. Viewpoints on Caribbean contexts of climate information use underscore the fact that knowledge itself is not tantamount to immediate mobilization and uptake. The perceptions captured in this research highlight the importance of greater focus on developing locally appropriate agendas of action that are clear about their goals, across multiple timescales. Consideration of the range of enabling and catalyzing conditions identified for effective utilization of climate information is necessary, particularly the need for tailored capacity strengthening, focus on roles of political factors and champions, co-production, and donor coordination. As well, it should be noted that since many of the barriers identified and confirmed within this study are similar to findings in other regions, there is a wider research need to reflect on the persistence of certain factors and conditions and the impact that this may have on making global progress on adaptation and resilience.

As the Caribbean continues to face climate-related challenges, greater attention is needed on limiting the persistence of barrier factors and encouraging supporting factors, in tandem with better understanding and attending to the availability and accessibility of locally relevant climate information. Such an improved focus on enabling and catalyzing conditions of climate information use should enhance the design of an effective adaptation agenda and signal the need for increased and improved attention to SIDS as likely canaries in the climate coal mine.

The data that support the findings of this study are available from the corresponding author, DSD, upon reasonable request.

A Research Ethics Review was completed, within which the research was deemed low-risk, and as such did not require formal review and approval by the LSE Research Ethics Committee. The participants provided their written informed consent to take part in this study.

DSD, DC, and SD contributed to conception and design of the study. DSD designed the interview protocol, undertook the interviews, performed the data analysis, and wrote the first draft of the manuscript. DC and SD provided substantial manuscript rewrites, comments, edits to all sections, and elements of the manuscript. All authors contributed to manuscript revision, read, and approved the submitted version.