Blue innovation and ecological transitions: a realist synthesis of mechanisms and knowledge asymmetries

Purpose This study aims to synthesize how scientific literature defines and operationalizes blue innovation and to explain the context-dependent mechanisms linking blue innovation to ecological transitions and blue economy outcomes. The study also examines how geographical asymmetries in knowledge production limit the external validity of dominant blue economy models, particularly for Latin America, Africa and the South Atlantic. Design/methodology/approach The study adopts a systematic literature review supported by semiautomated computational screening and realist synthesis. The review follows preferred reporting items for systematic reviews and meta-analyses logic and uses the sample, phenomenon of interest, design, evaluation and research type framework to define the research question and search strategy. Bibliometric mapping, Latent Dirichlet Allocation (LDA) topic modeling and a context–mechanism–outcome (CMO) realist synthesis were applied to a final corpus of 120 studies selected from 11,465 unique records consolidated from 20,082 database hits. Findings The literature on blue economy and blue innovation is rapidly expanding, but remains conceptually fragmented, methodologically heterogeneous and geographically concentrated in the Global North and Indo-Pacific. The review identifies three recurrent mechanisms associated with blue innovation: regulatory innovation, interorganizational coordination and institutional learning. These mechanisms are not universally activated. Their outcomes depend on state capacity, governance arrangements, knowledge-network density, monitoring systems and socio-environmental vulnerability. The underrepresentation of Latin America, Africa and the South Atlantic is therefore not only a geographical gap, but also a limitation for theory development and policy transfer. Research limitations/implications The reliance on international databases may underrepresent regional publications, gray literature and non-English outputs. The use of titles, abstracts and keywords for computational screening and topic modeling may also miss conceptual nuances available only in full texts. Future research should test and refine the proposed CMO framework through comparative, longitudinal and implementation-oriented studies in underrepresented coastal regions. Practical implications The findings suggest that blue innovation agendas should begin with institutional diagnosis rather than technological optimism. Policymakers and international organizations should assess regulatory capacity, monitoring systems, coordination arrangements, scientific networks and participatory mechanisms before transferring blue economy instruments across contexts. Originality/value This study offers a theory-informed realist synthesis of blue innovation. Rather than treating blue innovation as a unified concept or as a synonym for the blue economy, the study positions it at the intersection of socio-technical transition theory and institutional theory. Its core contribution is a CMO framework explaining how regulatory innovation, interorganizational coordination and institutional learning operate as context-dependent mechanisms. Its supporting contribution is methodological, demonstrating how computational mapping and realist synthesis can be combined to review fragmented interdisciplinary policy domains.