The emergence of climate migration research as an academic field

1 Introduction

When does an academic topic of interest become a research field? This question is central to understanding how knowledge is formed, organized, and institutionalized over time, particularly around complex problems that emerge faster than established disciplines can fully address them. Climate change exemplifies such a challenge: a complex and rapidly evolving phenomenon, intersecting various contexts, which keeps generating new research questions to understand its many implications. Among these, climate migration has emerged as a revelatory case due to its historical knowledge formation. What started in the 1980s¹ as a dispersed multidisciplinary inquiry to study how environmental degradation affects human mobility, has transformed into a more structured knowledge infrastructure, marked by expanding scholarly networks, aligning theoretical concepts and methodologies, emerging institutional venues (conferences, research centers) and growing funding in research.

While previous studies² have mapped the climate migration scholarship with its networks, less attention has been paid to the dynamics of institutionalizing knowledge, scholarly alignment, building networks, sustaining support and legitimacy, which contribute to the field's emergence. This article builds on existing analyses by adding an analytical framework on transforming the scholarly inquiry into an academic field.

By asking how and why the study of climate migration has evolved from an emerging topic of interest into a distinct academic field, this article addresses a broader set of questions about knowledge infrastructures–how scholarly communities organize around shared problem spaces, how epistemic boundaries are drawn and maintained, how resources are mobilized to exercise authority and legitimacy. Climate migration is therefore not only an empirical case but also lens through which we examine the mechanisms that lead to enduring fields of study.

Understanding whether climate migration constitutes a research field carries both analytical and practical significance. Analytically, recognizing an academic field sustains its epistemic boundaries (what is and is not climate migration research), and allows others to study it as a coherent entity. Scholars who cite others and collaborate closely within the field also create a knowledge-production structure that is easier to trace, especially in terms of how concepts emerge, paradigms shift, and the boundaries of debate are redrawn. Practically, the label of a ‘field' brings visibility, funding, support, and legitimacy for its members who are recognized as belonging to its structures (Hambrick and Chen, 2008; Tight, 2020). Fields do more than aggregate research. They mobilize people, resources and ideas toward institutionalization (Frickel and Gross, 2005). Examining these processes provides insight into how emerging area of inquiry gains stability and influence.

The article is structured around four dimensions of field formation drawn from Hambrick and Chen (2008) and subsequent scholarship: (1) differentiation and boundary work, (2) coherence of community, (3) social infrastructure, and (4) scientific legitimacy. The following section introduces the theoretical framework underpinning these dimensions. The methods section then explains the mixed-method design combining bibliometric analysis, document and policy review, and expert interviews. Each empirical section analyses one dimension separately, before the article closes by reflecting on what the climate migration case reveals about how knowledge infrastructures emerge around complex global problems.

2 Studying research fields

Understanding how research fields emerge and demarcate their boundaries has attracted much academic attention (Fleck, 1982; Hambrick and Chen, 2008; Krishnan, 2009; Mitnick, 2019; Tight, 2020). This has revealed a variety of conditions that may play a role in the rise of research fields. We define a research field to be a consistent and concentrated research effort that mobilizes support and resources, builds infrastructure, draws and maintains epistemological boundaries and achieves recognition as a legitimate and distinctive area of study (Whitley, 2000; Frickel and Gross, 2005; Mitnick, 2019; Levy et al., 2020). In contrast, a research topic is understood as an area of scientific inquiry characterized by blurred epistemological boundaries, limited coordination among scholars, and the absence of stable organizational structures that could amass resources.

In this paper, ‘climate migration' refers to the research examining the links between climate change and human mobility. We study the scholarly community, its outputs and the knowledge infrastructures. We use ‘climate migration' as the most widely used term among researchers since this scholarly inquiry gained traction. This does not mean that we empirically substantiate the existence of climate migration, rather we treat it as a description of a field of research.

To map the emergence of climate migration research as an academic field, we sharpened the model of Hambrick and Chen (2008) by integrating other perspective from the literature on field's ascendence (Whitley, 2000; Krishnan, 2009; Mitnick, 2019). This was synthetised into four analytical dimensions that guide our empirical analysis:

(1) Differentiation and boundary work: Hambrick and Chen (2008) argue that to demonstrate ‘distinctiveness' aspiring academic fields need to prove that their inquiry cannot be fully captured by other existing fields (e.g. migration, population or climate studies). Differentiation means carving out new area of expertise, while boundary work requires convincing others that drawing these boundaries have merit (Benford and Snow, 2000). Boundaries could be recognized internally within academia, and externally, by e.g. policymakers and funders, whose support leads to more visibility and resources. Hambrick and Chen (2008) call this “political opportunity structure” which signals to other actors that the issue is important and therefore, requires more research.

(2) “Coherence of community” is a term used by Mitnick (2019), for scholars concentrated around one shared problem space, engaging in ongoing debates and drawing on common paradigms. We do not equate coherence with uniformity or maturity of the field because diversity of ideas, internal debates and challenging established paradigms are common features of academic fields³. We view emerging fields as dynamic arenas influenced by two opposing forces: centripetal forces, leading to more coherence by pulling scholars toward an epistemological core, and centrifugal forces, embracing more diversity and new ideas. We consider the synchrony of these forces (not a dominance of one over the other) to be a pivotal element of the field's dynamics because they mark the boundaries of their own internal debate (Mitnick, 2019). We use paradigms as a main indicator of how a community is organized around a shared problem space because paradigms define what is to be studied and how (Kuhn, 1962; Bird, 2010), and they foster a group's coherence and identity (Davis and Thompson, 1994).

(3) Social Infrastructure encompasses research collaborations, networks, and institutions built around the shared problem space. These structures provide a platform for mobilization, knowledge transfer, and exerting influence (Marwell et al., 1988; Conell and Voss, 1990). Interconnected infrastructure is crucial for emerging fields as it leads to better coherence of community and dissemination of ideas (Wade et al., 1998). We analyse social infrastructure through co-authorship networks, and establishing research centers, conferences and other venues of scholarly engagement to map the collaboration patterns.

(4) Scientific Legitimacy: An aspiring academic field needs to demonstrate that its knowledge production is scientifically credible and aligned with academic standards (Whitley, 2000). Hambrick and Chen (2008) argue that legitimacy can be built through intellectual argumentation, producing evidence and unique contributions to science (Merton, 1973) and emulation, replicating methodologies, and theories of well-established fields (Cole, 1983). Thus, academic fields need to be not only recognized by academia but also have capacity and authority to produce sound knowledge.

The main limitation of theory Hambrick and Chen's (2008), already noted by Mitnick (2019), is its ambiguity on defining the field's emergence as a process (field-building) or an outcome (field-establishment). We are aligned with Bourdieu and Wacquant (1992) in recognizing field as dynamic arena of which boundaries and core priorities are constantly being reshaped by those who hold economic, social, cultural, and symbolic capital, such as well-funded institutions, established scholars with wide networks, and authors whose work is widely recognized as authoritative. Studying the field-building is crucial in analyzing topic- or problem-based fields, such as climate migration, which unlike discipline-based fields, tend to fluctuate in focus, purpose, and size (Abbott, 2001). Similarly to Levy et al. (2020) who uncovered the field-building as a gradual process, we do not expect to identify one moment in history in which the field has suddenly emerged.

3 Materials and methods

This article employs a mixed-method design, combining bibliometric analysis, expert interviews, policy and document reviews to examine the development of climate migration research. The metadata was collected from the Web of Science database in January 2024, covering publications from 1990–2023 (Figure 1).

Although we use ‘climate migration' as the label throughout this paper, scholars do not use consistent terminology to describe climate-migration links. We therefore kept the search terms broad, combining “environmental” and “climate” and their variations (e.g. climate-driven, climate-induced) with mobility, displacement, relocation, migration, and refugee. Older literature tends to use environmental framing, while more recent work has shifted toward climate migration or climate mobility. Capturing this diversity of terms is crucial to get an approximate picture of the field's developments.

The selected keywords were tested on a sample of influential papers to ensure the relevant language was captured. The final search string, included in the Supplementary material, returned 1,385 results. After removing duplicates, 1,329 remained. We then screened these against a set of inclusion and exclusion criteria (see Supplementary material), excluding 505 sources where climate migration was not the central focus, such as in papers mentioning migration only in passing⁴. Screening was conducted by the first author based on title and abstract, resulting in a final dataset of 824 sources.

As seen in Table 1, bibliometrics⁵ was selected as a main method to map emerging trends in research, assess international cooperation, and evaluate the outputs of authors (Li and Zhao, 2015; Wang et al., 2018). While citation analysis measures an author's impact on the field based on the citations, co-authorship analysis shows collaboration patterns in the field (Fonseca et al., 2016). We use co-occurrence analysis to visualize how language and framing have evolved by connecting keywords that appear together in the same title and abstract. The more two words co-occur, the stronger is their link, clustering them closer together (Callon et al., 1991). Document and policy reviews assess how climate migration is acknowledged by policymakers, funders, and academia, and how it builds its legitimacy. In addition, expert interviews provide insight into the social dynamics of research collaboration. This triangulation uncovers the processes through which climate migration gained visibility, mobilized resources, and developed knowledge structures characteristic of a research field.

Table 1. Methodology explained.

The data collection strategy, although aimed at being exhaustive, has limitations. The Web of Science database provides only a snapshot of academic knowledge, given that local and indigenous knowledge, gray literature, and knowledge written in languages other than English are not indexed. Interviews with 20 climate migration experts, across every continent (see Supplementary material), alongside with an additional review of the literature, were conducted to broaden the diversity of voices and experiences represented. We also deliberately excluded most research from the natural and physical sciences, where ‘migration' usually refers to non-human processes such as the movement of species, sediments, or water. Filtering the few relevant articles out of this volume was not feasible, and most of them did not treat the climate-migration link as their primary focus, which placed them outside our selection criteria. The visualized networks therefore serve only as an approximation of research developments in climate migration, drawn mainly from English-speaking and social-science scholarship.

4 Mapping climate migration studies

4.1 Differentiation and boundary work

Differentiation as a process of defining and acknowledging research problem as distinct from existing domains (Hambrick and Chen, 2008; Mitnick, 2019) is traced through three domains: academia, policy and funding. Publishing research in peer-reviewed journals (Figure 2) indicates that the topic has undergone academic scrutiny and gained scholarly validation of its distinctiveness. Policy attention (Figure 3) leads to visibility around issues they regard as relevant, while funding mobilizes financial support, necessary for the field's long-term growth (Hambrick and Chen, 2008). Together, they are seen as co-producers of the field. Policy defines urgent societal problems that need to be researched, funders translate policy interests into resources and scholars generate evidence and theories, reinforcing its importance and distinctiveness to all domains.

Differentiating climate migration from adjacent fields first required distinguishing it from the longer, discontinuous history of environment-migration scholarship from which it emerged. The effects of climate on human mobility were documented as early as the nineteenth century (Ravenstein, 1885; Ratzel, 1882), before fading as migration theory grew more complex over the twentieth century and then re-emerging toward its end in the context of climate change anxiety (Piguet, 2013). This return was initially shaped by alarmist framing, as environmental advocates and some researchers (Myers, 1993, 1997) sought to draw attention to climate change by warning of “floods of environmental refugees” (Flavel et al., 2020). Although these claims were criticized as deterministic, they brought visibility to the security implications of climate change (European Commission, 2022), with migration as one facet.

While the topic has deep roots in history, studying climate migration as a distinct object of inquiry has increased in particular since the mid-2000′s (Figure 2). This trend holds across several independent bibliometric mappings of the field, which, despite using different sources and scopes, all show the same upward trajectory (Piguet et al., 2018; Milán-García et al., 2021; Ghosh et al., 2022). Comparing our data with these mappings helps offset the limitations of relying on a single database and offers a fuller picture of how knowledge production evolved⁶. This convergence strengthens the evidence that climate migration moved from a peripheral concern to a recognizable area of study. Albeit the growth alone does not mean that the field has formed, it signals that a growing community of scholars has concentrated around a shared problem space.

Policy and other actors have also played an active role in producing and legitimizing climate migration research, reinforcing its boundaries beyond academia. Growing policy attention has increased the demand for more research (Figure 2), often blurring the boundary between policy and knowledge production (Nash, 2018). Actors such as the Intergovernmental Panel on Climate Change (IPCC) (2007), UN agencies (United Nations Environment Programme (UNEP), 2007; United Nations Framework Convention on Climate Change (UNFCCC) 2010; 2015), European Commission (2013), the World Bank (Rigaud et al., 2018), the European Parliament (Kraler et al., 2020) commissioned and published their own research, and funded research initiatives such as EACH-FOR (2009)⁷. A key development was the establishment of UN Advisory Group on Climate Change and Human Mobility in 2013, which brought policy and academics together to translate emerging research within the UNFCCC negotiations (Nash, 2017, 2018). Moreover, key policy documents (Figure 3) such as the Cancun Adaptation Framework (United Nations Framework Convention on Climate Change (UNFCCC) 2010), Sendai Framework for Disaster Risk Reduction (United Nations Office for Disaster Risk Reduction (UNDRR) 2015), and Global Compact for Migration (United Nations General Assembly (UNGA), 2019), explicitly called for better understanding of climate-induced migration. Climate migration was also a topic at high-level international meetings, such as G20 and G7, and various COP events focused on biodiversity and desertification (Hahn and Fessler, 2023).

While policy has often relied on simplified, mono-causal interpretations of climate migration, some scholars have adopted these frames⁸, while others systematically reshaped them by introducing new policy narratives such as ‘migration as an adaptation', climate justice, and immobility (European Commission, 2022; Hahn and Fessler, 2023). These interactions constitute a form of boundary work, with scholars and policy negotiating how a phenomenon should be conceptualized, thereby drawing clear epistemic boundaries. Policy attention amplified the urgency and attracted resources, whereas scientific research brought greater nuance and reinterpreted the conceptual framing. Nonetheless, close research-policy collaboration can pose potential pitfalls. Scholars have warned that proximity to the policy can privilege certain actors and particular research ideas, narrow critical debate by framing climate migration as a technical problem and by creating echo chambers among the same policy and research actors (Baldwin and Bettini, 2017; Nash, 2018; Veronis et al., 2018; Boas et al., 2019).

Growing policy attention has been accompanied by increasing research funding from diverse sources, which made the steep rise in publications (Figure 2) possible. Funding for climate migration projects, not necessarily research alone⁹, flows through bilateral government donors, multilateral actors¹⁰, development banks, national research councils, and private foundations (Huang and Davidoff-Gore, 2025). Durand-Delacre and Bunchuay-Peth (2026) found 46 research projects between 2005 and 2030 with an overall budget of €47.5 million, which were funded from the EU, Asia-Pacific, national, and international sources. Their data shows funding growing over time in both the number of projects and their scale, although some funders might be underrepresented¹¹. Early initiatives such as EACH-FOR (2007–2009) were modest, while recent projects e.g. HABITABLE, an EU-funded consortium led by the Hugo Observatory, received funding of 6.8 million EUR. Their analysis indicates that the number of simultaneously active projects has risen to a peak in recent years, signaling an intensifying research effort. The funding is international in scope, though it is pre-dominantly funded by the EU (cf. Piguet et al., 2018). Overall, the upward spending trend across the EU, multilateral, and bilateral funders points to a field that is increasingly able to mobilize resources, reinforcing its consolidation as a distinct area of study.

In summary, climate migration research has begun to be recognized as a distinct object of study across scholarship, policy and funding. As climate change became politicized in the 1990s, elevating the debates into the global high-level policy arenas, climate-migration nexus began to gain more visibility. This shift could be seen in the expansion of gray literature after mid-2000s (Figure 2), increase in policy documents (Figure 3) and expansion and diversification of funding, all of which contributed to the field's differentiation.

4.2 Coherence of community

To understand whether a scholarly community is becoming a coherent field, oriented around a common shared space¹², we use co-occurrence and citation analysis. Together, they capture epistemic coherence, showing how language and ideas evolve, and social coherence, pointing to how scholars cite and engage with each other's research.

Through co-occurrence analysis we study paradigms and more gradual discursive shifts. Paradigms guide the questions scholars ask and the methods, theories, and languages they use, thereby leading to temporary consensus (Kuhn, 1962; Bird, 2000; Hoyningen-Huene and Kincaid, 2023). Alongside paradigm shifts, the discourse is continuously being reshaped by smaller changes in terminology, methods, and research questions, which we call discursive shifts (Peci and Vieira, 2009; Beddoes, 2014). We regard paradigms as centripetal forces, stabilizing the discourse, and subtler shifts as centrifugal forces, expanding its boundaries.

A deterministic paradigm, seeing environmental changes as the sole driver of migration¹³, has been replaced since the 2000s by a multicausal paradigm, considering migration to be driven by climate, political, economic, and social factors. Early scholars such as (Black 2001), Castles (2002), and Hunter (2005) advanced this approach, and the Foresight Report (2011)¹⁴, one of the most widely cited works in the field, further consolidated the paradigm. It is a pivotal example of a centripetal force, bringing the elements toward one shared problem space.

Figure 4 illustrates the discursive shifts between 2009–2023, which brought more diverse terminology and geographical perspectives and thus stretched the boundaries of the field with their centrifugal forces. In the earliest period (2009–2013), the discourse was shaped around terms such as “environmental migration” and “climate refugees”, reflecting the initial focus on the direct links between climate and migration. By 2014–2018, clusters such as “adaptation,” “resilience,” and “vulnerability” signaled that the understanding of climate mobility had become anchored in the broader social, political and other contexts. The most recent map expands further, adding concepts such as “justice,” “(im)mobility,” “protection” and “rights”, cementing the shift toward climate migration that engages with legal, social, and policy dimensions. Additionally, we see more geographical terms (e.g., Philippines, Morocco, Pacific) which point to growing number of case studies, which study the local effects of climate on migration.

While paradigms and discourse contribute to mapping epistemic coherence, a scholarly community also needs to show signs of social coherence, which we study through citation analysis. Since the scholars are the driving force behind the strengthening of certain arguments, the expansion of the vocabulary, and the contestation of one another's research, citations indicate whether the field has developed a shared intellectual core or split into separate schools of thought (cf. Hoyningen-Huene and Kincaid, 2023).

In our citation analysis¹⁵ the number of links represents the breadth of an author's connections (how many others they cite or are cited by) and the link strength represents the intensity of the connections, with authors who are cited repeatedly having stronger citation links (Table 2). Richard Black, for instance, is connected to 142 authors through 450 citations. On average, authors are connected to him by 3.2 citations, placing him at the network's center. Lori Hunter, by contrast, is connected to fewer authors (126) but has the highest link strength (503), indicating fewer authors engage with her work but those who do cite it more frequently.

Table 2. Most interlinked authors in 2023 (as per citation analysis).

The citation analysis revealed increasing interconnectedness, mainly around early contributors to the climate migration research (Table 2; Figure 5)¹⁶. This pattern reflects cumulative advantage, also called the Matthew Effect (Merton, 1973; Hess, 1997), in which scholars who pioneer ideas in emerging fields tend to attract attention of later scholarships building on their work, making them more cited and prominent as the field expands (Hess, 1997; Barabasi et al., 1999; Frickel and Gross, 2005). Table 2 does not provide a definite list of “early contributors”¹⁷, however, it points to a broader dynamic, in which sustained early engagement with the topic tend to generate cumulative visibility, citations, and network centrality over time.

Although the citation network has expanded substantially in 15 years (see Figure 5)¹⁸, we observe no fragmentation into separate clusters. Instead, we observe a centralization pattern around early contributors, who are increasingly interconnected with newer scholars. Interviewed experts confirmed that early studies significantly shaped how the field has developed, which explains why these publications continue to be widely cited. For example, Richard Black, co-author of the Foresight Report, is still one of the most cited authors (Table 2), despite no longer being active in the field. His contribution to building the multicausality paradigm remains a key reference point that newcomers in the field frequently cite (de Sherbinin et al., 2022).

The coherence of the citation network reflects how we defined the field's boundaries in our search. We focused on scholarship that treats the climate-migration link as its central concern, since these are the main actors drawing and defending the field's boundaries (see Methods). This focus excluded much of natural science research, where migration pre-dominantly appears as a peripheral outcome of environmental change, and which might otherwise have formed a separate citation cluster. The network therefore maps the scholarship for which climate migration is the primary object of study, rather than the full research space where climate and human mobility intersect.

Figure 6 on network density¹⁹ further supports the interpretation of well-connected rather than fragmenting network, despite the number of citation links increasing. The early peak around 2004 reflects the field's formative years. As knowledge began to accumulate after 2000, a small group of scholars became aware of one anothers work and started citing it, producing high density precisely because the network was still small. From 2008 to 2020, density remained at roughly 10%, even as the network grows substantially in authors and links²⁰. Despite an overall expansion, the network has been absorbing new literature into an interconnected body of work, rather than splintering into parallel communities.

The scholarly community of climate migration has shown growing epistemic and social coherence around a shared intellectual space. The multicausality paradigm and the interconnected citation core acted as centripetal forces, anchoring the field, while the growth of published knowledge and conceptual and geographic diversity indicated centrifugal forces, broadening the field's scope. We view the co-existence of both forces as a signal of a field- building, since established fields are shaped by structured pluralism organized around shared reference points (Whitley, 2000; Abbott, 2001; Frickel and Gross, 2005; Mitnick, 2019).

Yet the same centripetal force (multicausality paradigm) stabilizing a shared problem space has become so dominant that it limits the deeper development of theories, concepts, and methodologies (Hoffmann et al., 2021; de Sherbinin et al., 2022). Moving beyond this paradigm has proven difficult for three reasons. First, because the field is young, the early scholars who introduced the paradigm remain active and influential actors, often limiting critical engagement with their work (Mayer, 2013). Second, the paradigm is so broad that it is difficult to disprove. As one interviewee put it, “most researchers agree that climate migration is multicausal.” Third, policy framings of adaptation and justice, which have grown more prominent over time, continue to reinforce it (Hahn and Fessler, 2023; European Commission, 2022). Together, these dynamics entrench multicausality as a shared foundation, leaving little room for new directions that could push the field forward.

4.3 Social infrastructure

To understand the social fabric of the field, we conducted a co-authorship analysis²¹, which maps international research collaboration. In emerging fields, such collaborations help mobilize scholars, disseminate knowledge and stabilize collective inquiry (Marwell et al., 1988; Frickel and Gross, 2005; Hambrick and Chen, 2008). The co-authorship links show how climate migration research has moved from dispersed scholarship toward a globally organized research network.

The co-authorship network has grown more interconnected and polycentric over time²², but it remains unequally distributed across regions (Figure 7). Although the network in 2013 appears relatively fragmented, the network in 2023 shows more interconnected and globally dispersed ties, with the number of co-authorship links almost tripling. Throughout, the USA, England, Australia, and Germany remained central due to their high publication volume. Yet the largest contributors tend to co-author research more among domestic universities. For example, USA has produced by far the most publications (250) but only 35% is co-authored internationally, while Australia and England sit at around 50% (Table 3)²³. Smaller producers²⁴ show the opposite pattern. Countries such as Bangladesh, Ghana, and Thailand, as well as overall regions of Africa and Asia rely heavily on international collaboration to publish research in the field (Tables 3, 4)²⁵. This asymmetry is what gives the network its polycentric but uneven structure. A few high-output countries center the field through their domestic research capacity, while lower-output countries connect to the network largely through ties to these hubs²⁶. Interviewees described high-output countries as ‘Hubs of Excellence', where established experts attract collaborators and early-career scholars.

Table 3. Countries with >5 international co-authorships (1990–2023).

Table 4. Proportion of international links per regions.

Several factors shape the co-authorship networks. As in Piguet et al. (2018), we found stronger ties between countries with shared language or colonial histories, such as England connected with India, Kenya, and Ghana, or France with Côte d'Ivoire and Tunisia. Geographical proximity (EU countries), international projects (e.g. HABITABLE) as well as funding structures have also influenced research cooperation. Several interviewees noted that funding tends to favor established researchers in certain countries, and that EU funding schemes require projects to include at least one EU partner (European Research Executive Agency, 2024), which can hinder South–South collaboration.

While co-authorship networks illustrate the observable ties among the researchers, development of institutional platforms shows how these ties are organized into more durable forms of collaboration within the field's social infrastructure²⁷. In climate migration, research networks, such as the Environment and Climate Mobilities Network (ECMN), the Population Environment Research Network (PERN), the Climate Mobility Network at the Columbia University, the South American Network for Environmental Migrations (RESAMA), the Asia-Pacific Migration and the Climate Mobility Africa Research Network (CMARN) have contributed to fostering research collaboration among the scholars, sharing knowledge across regions, and coordinating research initiatives. They contributed to anchoring the field more securely and outlasting individual projects and researchers, an essential indicator of the durable social infrastructure of the field. For example, interviewees highlighted that research initiatives such as HABITABLE, led by the Hugo Observatory, not only shape future citation and co-authorship links but also create long-term research networks that persist beyond the project lifecycle.

Aside from research networks, conferences also play a role in the institutionalization of the field, as they provide a platform for scholars to bring new ideas and co-create a shared problem space. Scholars began gathering around climate and mobility before the field had fully taken shape. In 2008, the Environment, Forced Migration and Social Vulnerability conference brought together around 400 international experts (IISD, 2008). The European research network COST Action IS1101 held workshops in Brighton (2012) and Neuchâtel (2013) on mountain regions and policy frameworks (University of Neuchâtel, 2013). More recently, the first Environmental and Climate Mobilities Network Conference (2023) was organized for scholars to discuss their current work and future directions for the field. Dedicated conference series have since emerged on the other side of the Atlantic. Columbia Climate School (2025) has hosted biennial conferences on managed retreat, mobility and resilience since 2019, drawing up to 800 participants. Furthermore, major established conferences, including the International Geographical Congress (2016) and IMISCOE (2023), have also incorporated climate migration sessions. Together, these examples show climate migration gradually gaining a foothold within broader scholarly debates.

Co-authorship analysis and institutional evidence indicate that climate migration scholars have developed a distinctive, albeit geographically uneven, social infrastructure. Expanding research networks and the establishment of research hubs suggests that the field can mobilize more scholars and resources globally. However, this mobilization is concentrated around a relatively small set of universities, often in high-income countries that had sufficient resources to support research in an emerging topic like climate migration before it had a clear academic payoff. Their early involvement built their reputations as the leaders of the field, and attracted funding, collaboration and influence over research agendas²⁸. This uneven concentration of resources is also seen in the citation patterns, in which early contributors accumulate recognition, funding, and citations. While such developments can stabilize an emerging field, they may also narrow alternative paradigms and their proponents, which some interviewed scholars identified as a challenge.

4.4 Legitimacy-building

In this paper, legitimacy can be achieved through intellectual argumentation (whereby scholars convince others that their claims are relevant and sound) and emulation, by replicating theories and methods of more established fields to align with the standards of scientific inquiry (Hambrick and Chen, 2008). We use academic journals as the primary indicator of intellectual argumentation because they act as gatekeeping arenas in which knowledge claims are collectively evaluated (Merton, 1973; Lamont, 2009). Specifically, we analyse the three journals that most frequently published climate migration research in our dataset to show how a topic persists alongside other research themes. However, legitimacy is built across multiple arenas (Frickel and Gross, 2005; Hambrick and Chen, 2008), with journals being only one avenue. We therefore combine journal analysis with the evidence of other field-building dimensions. In contrast, emulation cannot be reliably captured through bibliometrics since theories and methods are not systematically indexed. We therefore rely on document review and interviews.

Although climate migration has gained attention in recent years, its representation in leading journals remains limited (Figure 8). Across the three journals most frequently appearing in our dataset, climate migration articles constitute only a small portion of total publication output over time, with 1–3% in Climate and Development, 1–2 % in Global Environmental Change and 5–12% in Population and Environment, despite the overall publication volume increasing. This pattern does not automatically indicate weak intellectual argumentation but rather reflects structural challenges typical of emerging fields. First, climate migration is a highly specific topic, competing for attention with broader, and more established topics, such as climate adaptation²⁹. Second, there is no flagship journal that would focus on climate migration. Our dataset (N=824 sources) included 333 journals, but approximately 200 of them published only a single article on the topic between 1990–2023. This fragmented publication landscape is common in emerging fields, in which multiple disciplines engage with the same problem and infrastructure is not yet fully developed (Abbott, 2001; Frickel and Gross, 2005). However, some journals³⁰ published special issues on climate migration, and Frontiers of Climate even established a dedicated Climate Mobility Section in 2022, indicating that a publication infrastructure is beginning to take shape, a meaningful step toward the field-building.

Legitimacy-building does not depend solely on journal recognition but also on broader institutional acknowledgment. The establishment of research centers, networks and conferences (see Social Infrastructure) reflects the consolidation of intellectual communities that present themselves as legitimate experts, while engagement with policy and funders (see Differentiation and Boundary Work) captures external recognition beyond academia. Scholars used a range of tools to convince others that their research is sound and relevant, which could be more visible in the growth of published research and policy interest than in journal presence. However, the field's legitimacy often develops unevenly across arenas (Frickel and Gross, 2005; Hambrick and Chen, 2008), suggesting that it remains in early stages.

In the second pathway of legitimacy-building (emulation), methods have been replicated from established disciplines more successfully than theories. In terms of theory building, de Sherbinin et al. (2022) found that nearly 40% of the climate migration literature made little or no reference to theory. While unifying theories (McAdam, 2013; Baldwin, 2014; Hunter et al., 2015) and coherent conceptual frameworks (McAdam, 2012; Piguet, 2018; de Sherbinin et al., 2022) have not been fully developed, Hunter and Simon (2022) argue that climate migration research can enrich migration theory more broadly by incorporating environment as one of the drivers of migration. However, methodologies are being more successfully emulated, mostly from the scholars' home disciplines, such as geography, demography, or economics. Common methods include qualitative approaches such as ethnography (Morrissey, 2012; Tschakert et al., 2013), quantitative techniques such as modeling (Rigaud et al., 2018), and mixed methods such as agent-based modeling (Kniveton et al., 2011; Cai and Oppenheimer, 2013) and scenario planning (Foresight Report, 2011). The methodological borrowing creates an analytical toolkit that helps new scholars entering the field and aligns the field with recognized standards of scientific inquiry, thereby strengthening its legitimacy.

The limited emulation in climate migration research reflects structural features of the field. Its interdisciplinary position between migration and climate studies complicates collaboration and slows the field-building because the scholars pre-dominantly rely on disciplinary toolkits that influence terminology and concepts³¹. Therefore, research uses simplified models and deterministic concepts (Morrissey, 2012; Hoffmann et al., 2021), which limits the understanding of why some people migrate, and others do not (de Sherbinin et al., 2022). Interviewees also noted that using the same theories and methods restricts the potential to innovate and ‘reimagine' studies of climate migration³² by integrating new ideas from e.g. feminist literature. Additionally, policy actors with their specific agendas also influence research priorities and undermine theoretical cohesion³³. Such dynamics are typical of emerging fields, which need time to consolidate, as seen in higher education and migration studies (Levy et al., 2020; Tight, 2020). However, some interviewed experts view this conceptual diversity as a strength, arguing that interdisciplinarity keeps the field open to new perspectives and prevents the formation of intellectual echo chambers. Nevertheless, the reluctance of scholarly community to pursue theoretical and conceptual consolidation also helps explain why emulation has not been yet fully adopted.

5 Conclusion

Our findings show that the study of climate migration has evolved from an emerging topic of interest into a distinct field of academic research (Table 5). We trace how epistemic, social, and institutional structures of the field have developed over time, by examining differentiation and boundary work, coherence of community, social infrastructure, and legitimacy-building, (Hambrick and Chen, 2008; Mitnick, 2019). The field's emergence was a cumulative process that accelerated after the 2010′s, as seen in the rapid growth of publications, increasing policy recognition, interconnected scholarly community, and emerging institutional and collaborative structures.

Table 5. Indicators of climate migration as an academic field.

^aWe are interested not only in the final stage (if the group has reached each level), but also in the dynamics preceding that.

Across these dimensions, field formation has been driven by the interplay of centripetal and centrifugal forces. Centripetal dynamics could be seen in creating a shared problem space anchored in the multicausality paradigm and reinforced through dense citation networks around early contributors. Recognition of the new boundaries by policy and funders further underpinned the consolidation of climate migration as a distinctive and legitimate area of inquiry. Moreover, social infrastructure became increasingly interconnected across geographically dispersed communities, with its polycentric co-authorship networks and establishment of new platforms for research collaboration (networks, conferences, research centers), enabling knowledge circulation and coordination across regions. Simultaneously, centrifugal dynamics such as new terminology, diverging ideas, interdisciplinarity, expanding geographical participation have broadened the field's thematical and regional boundaries without fragmenting it. The coexistence of stabilizing and diversifying dynamics is common for emerging fields, in which structural pluralism emerges around shared reference points (Whitley, 2000; Abbott, 2001; Frickel and Gross, 2005).

Legitimacy-building is the least consolidated dimension of field formation. While climate migration has attracted attention from policy, funders, and other researchers who entered the field, intellectual argumentation remains dispersed across many journals. Although theoretical frameworks are underdeveloped, the breadth of research has a potential to advance migration theories (Hunter and Simon, 2022). On the other hand, scholars reached some gains in methodological borrowing from other disciplines.

The field's heterogeneity does not indicate fragmentation or dysfunctionality. Many of our interviewees in fact viewed it as a strength, since it fosters inclusivity and openness to new ideas and perspectives, which are crucial for addressing the complex and multifaceted nature of climate migration. This heterogeneity reflects (a) the short existence of climate migration as a wider academic topic (in comparison, migration studies are around 50 years old Levy et al., 2020), (b) diversity in disciplinary ontologies, (c) asymmetrical distribution of power among researchers, and (d) siloed policy thinking, with actors approaching the issue from their own vantage points (Hahn and Fessler, 2023).

Studying how climate migration became a research field has offered insights into how knowledge infrastructures emerge around complex global problems. Analytically, the paper shows that field formation is best understood as a gradual process shaped by interactions between a shared problem space, social organization and external recognition. In practical terms, calling an academic effort ‘research field' recognizes its legitimacy to operate as a separate area of study, which leads to more resources, visibility and authority. Our findings show that the scholars have managed to draw clear epistemic boundaries around the topic. Although this process might not have been coordinated effort to establish a field per se, its cumulative publishing has generated visibility and interest from other actors. In this sense, defining climate migration as a research field is not merely descriptive but performative. It can consolidate its authority, shape institutional expectations and reinforce the position of its scholarly community.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fclim.2026.1825882/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Alex de Sherbinin, Columbia University, United States

Reviewed by: Kees Van Der Geest, Institute for Environment and Human Security (UNU), Germany

Etienne Piguet, Université de Neuchâtel, Switzerland

Footnotes

El-Hinnawi, 1985; Jacobson, 1988; Myers, 1993, 1997.

Scholars studied ‘what' research is produced ‘where' and ‘who' the main authors are (Obokata et al., 2014; Maretti et al., 2019; Milán-García et al., 2021; Ghosh and Orchiston, 2022; Ghosh et al., 2022), but also the reasons for specific patterns in the research geography (Piguet et al., 2018).

Levy et al. (2020) discovered both fragmentation and coherence in studying the evolution of migration studies.

This includes e.g. the environmental security literature (Homer-Dixon, 1994), in which migration typically appears as one intervening factor in a pathway toward conflict rather than as the central object of study. Although this body of work can overlap with climate migration, it predominantly addresses the intersection of climate and security.

All networks are visualised through VOSviewer (van Eck and Waltman, 2009). The position of nodes is determined by a layout algorithm that places strongly associated items close together. The axes themselves carry no independent meaning and only relative distance and clustering are interpretable. The size of each node reflects its importance in the network: the more often a keyword is used, an author is cited or a country is involved in co-authored research, the larger its node.

The trajectories differ across datasets due to the various scope and data collection methods. Piguet et al. (2018), top line, include grey and non-English literature, which Web of Science or Scopus do not index. That is why they appear to capture the most knowledge, peaking in 2012, when grey literature started to get published. Milán-García et al. (2021) gather data from Web of Science and Scopus, while Ghosh et al. (2022) limit their mapping to the Pacific region, which is why their data appears at the bottom.

EACH-FOR (Environmental Change and Forced Migration Scenarios) was an EU-funded project (2007–2009) that studied the links between environmental and forced migration and developed future scenarios of environmentally induced displacement across several regions.

The initial policy framing of climate migration as a security threat has shaped some research (Reuveny and Moore, 2009; Ross, 2010) as well as funding schemes (Boas et al., 2019).

The scale of operational funding far exceeds research funding. The EU's RE2CLID programme, for instance, commits 100 million EUR to climate displacement responses across eight Sub-Saharan African countries, which is the largest climate mobility programme globally (Huang and Davidoff-Gore, 2025).

Belmont Forum (2024) committed €6.2 million in 2022 to climate migration.

For example, US-funded projects such as the Convergence Research initiative (2019–2022) funded by the National Science Foundation (2019) and the Department of Defense project (2022–2025) on migration drivers in Central America and West Africa (Columbia Climate School, 2022) are not captured in the dataset.

Frickel and Gross (2005) see the field as a movement, whose participants are consciously oriented towards its shared knowledge core, even if they understand it in different ways.

The discourse was promoted by environmental researchers and advocates (El-Hinnawi, 1985; Jacobson, 1988; Myers, 1993), and other groups (e.g. Friends of the Earth Australia, 2005; CNA Military Advisory Board, 2007).

The report introduced a theoretical framework, describing climate migration as a matrix of micro, meso and macro levels, influencing the decisions to migrate or stay. Since the research field lacks a unified theory, Foresight framework is used as one of the most influential theoretical lenses.

Citation analysis examines impact and connections by tracking citations between publications. We derived these links from the 824 sources in our dataset, visualising a connection between two authors whenever one cites or is cited by the other. This reveals whose research is most influential, and which work later scholars build upon.

For example, Farbotko (2010), McLeman and Hunter (2010), Black et al. (2011), Gemenne (2011), Gray and Mueller (2012), Warner (2012).

Early contributors are identified based on their publications explicitly studying climate-migration links, captured by our data strategy. Scholars researching adjacent topics might be underrepresented if mobility was not their primary analytical focus. As with any bibliometric analysis, the patterns should be understood as an approximation of evolving scholarly networks.

While in 2013, we found only 27 ‘consistent' scholars (those who publish at least two documents in the studied period), 10 years later, the number expanded eightfold.

Network density is a way to understand how interconnected a field is. It is calculated as a number of citation links divided by the number of potential citation links (if all authors were connected with each other, it would mean 100% density because the number of real and potential links is the same). The formula from Levy et al. (2020) is included in the Supplementary material.

Since newer publications need time to accumulate citations, the number of citation links is lower in the last 3 years.

Co-authorship analysis maps research collaborations between countries, which helps us understand which countries are working closely and e.g. sharing knowledge and resources. Nevertheless, the data cannot capture other types of collaboration such as publishing grey literature, working with local policymakers and civil society, which interviewees also noted as important.

Unlike in the field of migration studies (Levy et al., 2020), EU funding has not led to the Europeanization of the climate migration field. Instead, EU countries appear scattered across the network, indicating weaker connections. Experts suggest that this might be due to the persistence of the security narrative in EU policy, also seen in European funding calls (European Commission, 2023). It mainly funds partnerships between European and migration-origin countries, particularly in Asia and Africa to gain deeper insights into the drivers of migration and develop strategies to mitigate them.

These figures reflect institutional affiliation rather than country of birth. The countries may host a diverse author pool, which collaborates mainly within its own institutions. The analysis therefore captures whether research capacity extends outward across borders, particularly to under-resourced regions, rather than the diversity of individual researchers.

Many are lower-income countries, which co-author research with another higher-income country. This might be the reason why South-South co-authorships remain rare.

The percentage is calculated as the number of international co-authorships divided by the number of documents per each country, then aggregated by region. For example, 67% of papers with a MENA-affiliated author were co-authored with at least one researcher from another country.

This is also because our data representation privileges citations of literature written in English and by scholars from certain universities - a clear limitation of the Web of Science database, which does not allow to capture all published research.

Emerging fields rely on mobilisation to sustain collaboration, recruit new scholars and enable collective agenda-setting (Frickel and Gross, 2005; Hambrick and Chen, 2008).

Interviewed experts noted that some universities are seen as the hubs for climate migration with more visibility and influence. It is often up to them who they bring in as partners into their research projects.

Population and Environment journal explores the link between demography and environment, in which climate migration, as a demographic change, fits more neatly and thus, embodies the highest proportions of the published literature in the selected sample.

Such as Journal of Demographic Economics (2021), Journal of Ethnic Migration Studies (2022) and Journal of Operations Management (2023).

Fussell (2012) noted that the rigid categorization of migration into voluntary and forced has impeded cooperation between migration and refugee studies.

Nash (2018) and Mayer (2013) argue that deeper reflections within the field are needed.

Actors such as IOM influence the field by promoting specific discourses or methods (International Organization for Migration (IOM), 2008, 2014).