Ocean Data Gets a FAIR Upgrade to Power Sustainable Seas

The ocean has always been difficult to see. It hides its chemistry, its currents, and its creatures behind kilometers of water, and for centuries scientists have pieced together its behavior from scattered ship logs, tide gauges, and expeditions that might take years to complete. Today the picture is changing at extraordinary speed. Satellites map sea surface temperature in near real time, autonomous gliders drift through deep basins relaying profiles of salinity and oxygen, and genomic sequencers reveal entire microbial ecosystems from a single liter of seawater. Yet for all this technological abundance, a quieter crisis persists beneath the waves of information: the data itself remains fragmented, locked in incompatible formats, and scattered across national repositories, research cruises, and institutional hard drives that may outlive the projects that created them.

A new perspective article published in npj Ocean Sustainability argues that the next leap in ocean science will not come from another sensor or satellite, but from the infrastructure that connects the data we already have. The authors examine how marine spatial data infrastructures, the networks of standards, repositories, and governance frameworks that underpin ocean data sharing, can be aligned with the principles of open science and with the FAIR guidelines, which hold that research data should be findable, accessible, interoperable, and reusable. The central claim is deceptively simple: sustainable ocean management is only as good as the knowledge systems that feed it, and those systems are currently too disjointed to support the decisions the ocean urgently needs.

The timing of this argument is not accidental. The United Nations Decade of Ocean Science for Sustainable Development has entered its second half, and the ambitions set out in 2021 are colliding with the practical realities of data governance. The Kunming-Montreal Global Biodiversity Framework commits nations to protecting thirty percent of land and sea by 2030, the high seas treaty known as the Biodiversity Beyond National Jurisdiction agreement is moving toward implementation, and marine spatial planning initiatives are expanding across every continent. Each of these commitments depends on the ability to ask hard, quantitative questions: Where are the biodiversity hotspots? How do shipping lanes overlap with whale migration corridors? Which seafloor habitats will be most resilient as waters warm and acidify? Answering them requires integrating physical oceanography, biology, geology, human-use data, and socioeconomic information at resolutions and accuracies that no single institution can produce alone.

Marine spatial data infrastructures, often abbreviated as MSDIs, emerged over the past two decades as a response to exactly this integration problem. Modeled in part on terrestrial spatial data infrastructures developed for land management, MSDIs provide the technical scaffolding for storing, cataloging, and serving geospatial ocean data. They define metadata standards that describe where a measurement was taken, when, by what instrument, and under what quality-control procedures. They establish web services that allow a map of coral reef extent in one national database to be layered seamlessly onto shipping density data held by an international maritime body. In principle, an MSDI lets a marine planner in a coastal ministry assemble a complete picture of their waters without ever downloading a raw file. In practice, the authors contend, the promise remains unevenly realized.

The technical barriers are real but solvable, and the article is refreshingly specific about them. Ocean data arrives in a bewildering array of formats: netCDF files from oceanographic models, Darwin Core archives from biodiversity surveys, Shapefiles and GeoJSON from habitat mapping, and proprietary formats from acoustic and optical sensors. Vocabulary mismatches compound the format problem, as one agency’s benthic habitat classification may bear little resemblance to another’s even when describing the same seafloor. Semantic interoperability, the ability of machines to understand not just that two datasets exist but that they describe comparable phenomena, demands shared ontologies and controlled vocabularies such as those maintained by the community behind the NERC Vocabulary Server and the Marine Regions gazetteer. FAIR alignment pushes these requirements further, insisting that data carry persistent identifiers, that metadata be rich enough for independent reuse, and that licenses clearly state what others may do with the material.

What elevates the article beyond a technical audit is its insistence that infrastructure is inseparable from culture and equity. Open science, the authors emphasize, is not merely the removal of paywalls from publications. It is a systemic commitment to transparency across the entire research lifecycle, from preregistered sampling designs to openly licensed datasets to reproducible analysis code. Applied to the ocean, this commitment carries a justice dimension that is easy to overlook from well-funded laboratories in the Global North. Much of the world’s marine data is collected in the waters of developing nations and small island states, yet the analytical capacity to exploit it is concentrated elsewhere. A genuinely aligned knowledge system would ensure that coastal communities who steward coral reefs, mangroves, and fisheries can access, understand, and contribute to the data describing their own seascapes, rather than serving as passive subjects of extraction by international research programs.

The authors sketch what this alignment looks like in operational terms. National and regional MSDIs would adopt FAIR-compliant repositories with persistent identifiers for every dataset and sample. Metadata would follow internationally recognized standards such as ISO 19115 for geospatial information, supplemented by community schemas for biological occurrences. Application programming interfaces, built on open protocols like OGC standards and the SPARQL query language for linked data, would allow researchers and decision-makers to compose datasets on the fly rather than negotiating bilateral data-sharing agreements that can take years. Crucially, the vision includes provenance tracking, so that a sea surface temperature anomaly used in a fisheries closure decision can be traced back through every processing step to the original instrument deployment. In an era when contested science can stall policy for a decade, that traceability is not bureaucratic decoration; it is the foundation of trust.

The article also confronts the hard economics of open infrastructure. Repositories cost money to run, data curation is skilled labor, and long-term stewardship of ocean observations requires funding models that outlast individual grants. The authors point to the tension between mandates for immediate open data and the legitimate needs of researchers to publish from their own collections, arguing for community norms that balance openness with recognition. They highlight the role of coordinated bodies, including the Intergovernmental Oceanographic Commission’s International Oceanographic Data and Information Exchange program and the emerging Ocean Data and Information System, as the connective tissue that can harmonize national efforts without imposing one-size-fits-all solutions on institutions with wildly different capacities.

Perhaps the most compelling thread in the analysis concerns artificial intelligence and the coming wave of machine-driven discovery. Foundation models trained on ocean observations, digital twins that simulate entire ocean basins, and automated anomaly detection for illegal fishing all presuppose data that machines can read, validate, and combine at scale. FAIR infrastructure designed for human users a decade ago is not automatically fit for this purpose. Machine-actionable metadata, semantically annotated datasets, and interoperable services are the raw substrate on which the next generation of ocean intelligence will be built. The authors make clear that this is not a distant hypothetical; it is happening now, and knowledge infrastructures that fail to adapt risk becoming irrelevant bottlenecks in an otherwise accelerating field.

The stakes could hardly be higher. The ocean regulates the planet’s climate, feeds billions of people, and harbors the largest reservoir of life on Earth, yet it is warming, acidifying, deoxygenating, and losing biodiversity at rates that outpace many models. The article’s conclusion is ultimately an optimistic one: the pieces of a global, equitable, and technically rigorous ocean knowledge system already exist in scattered form. Standards bodies, data centers, and open science movements have each built parts of the puzzle. What remains, the authors argue, is deliberate alignment, the patient diplomatic and technical work of making these components interoperable, FAIR-compliant, and responsive to the communities who depend on ocean knowledge most. If that work succeeds, the invisible ocean becomes visible not just to satellites and supercomputers, but to every planner, scientist, and citizen with a stake in its future.

Subject of Research: Integration of marine spatial data infrastructures with FAIR principles and open science for sustainable ocean knowledge systems

Article Title: Aligning Marine Spatial Data Infrastructures and Open Science for FAIR and sustainable ocean knowledge systems

Article References: Aligning Marine Spatial Data Infrastructures and Open Science for FAIR and sustainable ocean knowledge systems. (n.d.). https://doi.org/10.1038/s44183-026-00246-6

Image Credits: AI Generated

DOI: 10.1038/s44183-026-00246-6

Keywords: marine spatial data infrastructure, FAIR principles, open science, ocean sustainability, ocean data governance, semantic interoperability, marine spatial planning, ocean observations, data infrastructure, biodiversity data, ocean decade, digital ocean