Metadata Standards and FAIR Principles in Astronomical Data Repositories

“ Metadata Standards and FAIR Principles in Astronomical Data Repositories” Beatriz Juátrez Santamaría and María Elena Jiménez Fragozo Universidad Nacional Autónoma de México. Instituto de Astronomía. A.P. 70-264, 04510. Ciudad de México, México. [email protected] Universidad Nacional Autónoma de México. Instituto de Astronomía. A.P. 106, 22800. Ensenada, B.C. , México. [email protected] ABSTRACT: Technological advances have enabled new telescopes to generate vast amounts of data that contribute to a better understanding of the universe. Data science and technology work together to transform the observation of the cosmos. Data science-driven processing systems make it possible to manage, process, and analyze these data. In this context, repositories play a key role in storing, sharing, and reusing data obtained through observations, simulations, and scientific projects. This work explores different types of repositories, classified according to the kind of data they host: those that store data generated by observatories, space missions, or scientific projects; those that encompass data produced by simulations and computational models; and repositories where researchers themselves deposit data. Data repositories in the field of astronomy have adopted interoperable metadata standards, as well as the FAIR principles (Findable, Accessible, Interoperable, Reusable). They have also integrated the Virtual Observatory initiative, which envisions astronomical datasets and other resources functioning as a unified whole. Accordingly, the metadata used by each repository will be analyzed to identify their differences and similarities, and to assess the extent to which they comply with the FAIR principles. . Introduction For this study, the following repositories were considered:MAST, CDS, ZENODO, Harvard Dataverse, ALMA and NED where the metadata they use are described and whether they comply with the FAIR principles (Findable, Accessible, Interoperable and Reusable). The role that metadata plays in the Virtual Observatory (OV) data collections and services will also be analyzed, enabling OV users to easily locate information of interest. 1.Importance of the FAIR Principles The FAIR principles emphasize the ability of computing systems to find, access, interoperate, and reuse data. The first step in reusing data is to locate it.To achieve this, metadata and data must be easy to find for humans and computers.Once the user locates the data they need, they need access to it. The principle of interoperability is important, as it allows data to be integrated with other data.(FAIR, n.d.) Data reuse is another principle;this is achieved through well-described metadata and data. 2. IVOA and the FAIR Principles 3.Repositories in Astronomy Data repositories are essential for storing, sharing, and reusing astronomical observations, simulations, and catalogs, they allow the application of FAIR principles, and they are key pieces for open science. These store data generated by observatories, space missions, or scientific consortia. Astronomical data repositories play a crucial role in managing, preserving, and accessing large amounts of data generated by various astronomical observations and experiments. These repositories are essential to boost open science, facilitate data sharing, and ensure its long-term preservation.Below are some repositories that describe the metadata they use and how they comply with the FAIR principles. As can be seen, the MAST, CDS, and NED repositories work with the standards established by IVOA and Dublin Core, enabling interoperability between repositories. Most use persistent identifiers (DOD, IVOA ID) and implement open APIs, guaranteeing access. However, there are some embargo restrictions.Data reuse is ensured through the use of licenses (CC-BY, NASA Open Data). Conclusion and Recommendations Most astronomical data repositories comply with or are in the process of complying with the FAIR principles, using interoperable standards and semantic models promoted by IVOA. Zenodo complies with the four principles, the metadata it uses allows for the integration of datasets with publications, facilitating the analysis of the impact of the data. Astronomical data repositories and metadata management are fundamental to the advancement of astronomy.By adopting standardized metadata schemes such as ObsCore,leveraging advanced data management systems, and adhering to the guidelines of international astronomical organizations, the astronomical community can ensure data sharing and reuse, long-term preservation, and further scientific discovery. An effective data management plan will allow accurate a correct description of the data, define the metadata standards that will be used, and indicate the licenses that will be assigned and where the data will be stored, all these criteria will allow the data to be retrieved, shared, and reused, complying with the FAIR principles. Table 1 .Metadata schemas and FAIR principles in astronomical repositories The Virtual Observatory (VO) represents the vision that astronomical datasets and other resources function as an integrated whole.The International Virtual Observatories Alliance (IVOA) is an organization that discusses and agrees on the technical standards necessary to make VO possible.(IVOA, n.d.) Its goal is to make all astronomical data available for exploration in asingle system. In 2017, IVOA published a document that will provide a simple, easy-to-understand template for anyone wishing to publish their data in the VO. (Louys, M., Tody, Doug, Dowler, P. et al. 2017) IVOA aligns with the FAIR principles but work still needs to be done to implement persistence identifiers and specify data usage licenses. References 1. Asok, K., et al. (2024). Common Metadata Framework for Research Data Repository: Necessity to Support Open Science. Journal of Library Metadata. 24 (1), 133-145 DOI: 10.1080/19386389.2024.2329370 2. FAIR. (n.d.) FAIR Principles. https://www.go-fair.org/fairprinciples/ IVOA (n.d). What is VO? https://www.ivoa.net/about/what-isvo.html 3. Louys, M., Tody, Doug, Dowler, P. et al.(2017). IVOA Recommendation, May 09, 2017.https://www.ivoa.net/documents/ObsCore/ DOI: 10.5479/ADS/bib/2017ivoa.spec.0509L 5. NASA (n.d.). Astrophysics Data Centers. https://science.nasa.gov/astrophysics/data/multimission-archiveat-stsci-mast/ 6. National Radio Observatory (NRAO) Public Wiki. (2009). ALMA Science Data Model. https://safe.nrao.edu/wiki/pub/EVLA/ScienceDataModelRe sourcesAndDocumentation/asdm_quick.pdf 7. NED NASA/IPC Extragalactic Database (2025). About NED. https://ned.ipac.caltech.edu/Documents/Overview 8. Zenodo (n.d.). About Zenodo. https://about.zenodo.org/ Repository Metadata FAIR Principles ALMA Atacama Large Millimeter and Submillimeter Array The ALMA Science Data Model (ASDM) defines the collection of information recorded during an observation, necessary for scientific analysis. (National Radio Astronomy Observatory, 2009) Observation Information Date and time of observation. Unique identifier for the observation and data. Proposal Information Research proposal associated with the data. Principal Investigator Information. Environmental and Atmospheric Metadata Real-time measurements of humidity, temperature, and atmospheric pressure. Antennas and Observation Metadata Antenna Position Calibration Signal Metadata Observation Wavelength FAIR Principles Findable Unique identifier for the observation and data. There is a data embargo that may limit access. HARVARD DATAVERSE Harvard Dataverse stands out for having the largest number of metadata elements among research data repositories (Asok, K., et al. 2024). This extensive metadata framework allows data to be described and identified in detail, promoting its reuse and interoperability. Metadata schema: Dublin Core+DataCite Descriptive metadata: title, authors, institution Technical metadata: format, size, software Provenance metadata: project, observatory, mission Rights Metadata: Licenses and Restrictions. Findable DOI y metadata OAIPMH Accessible: Access with permission controls or open access. Interoperable: OAI-PMH APIs Reusable: Exportable and citable metadata NED (NASA/IPAC Extragalactic Database) NED accelerates scientific discovery in astrophysics, within NASA's strategic plan, by providing a census of extragalactic objects, complemented by data collected from across the electromagnetic spectrum, for use by the scientific and non-scientific community worldwide, accessible through computerbased query protocols and intuitive interfaces. (NED, 2025) Metadata schema: IVOA+FITS headers Observational metadata: Wavelength, flux, observation time, detector, and exposure Administrative metadata: access policy, DOI Provenance metadata: Institutional source or project. Findable Mission-indexed metadata Accessible: Use of APIs Interoperable: Votable/XML standardization Reusable: Comprehensive documentation, DOI available. MAST (Mikulski Archive for Space Telescopes) It ‘s a NASA-funded project to support and provide the astronomical community with various astronomical data archives, mainly focusing on scientifically related data sets in the optical, ultraviolet, and near-infrared parts of the spectrum. (NASA, n.d.) Metadata schema: Dublin Core+FITS Headers+ VO (Virtual Observatory) Metadata. Descriptive metadata: title, mission, instrument, observed object. Technical metadata: FITS format, resolution, wavelengths. Administrative metadata: date, version, DOI, usage rights. Provenance metadata: special mission. Structural metadata: table, format, ASCII/FITS. Findable It has persistent DOIs and can be searched by mission. Accessible: Open access via APIs and VO services. Interoperable with VO. Reusable: Metadata with citations and clear documentation. CDS (Centre Donnés astronomiques de Strasbourg / VizieR, SIMBAD) Data center dedicated to the collection and global distribution of astronomical data. Metadata schema: IVOA, Dublin Core Observational metadata: coordinates, magnitudes, photometric bands. Contextual metadata: reference, bibliographic, author, catalog. Findable IVOA Identifiers (Bibcode, Object ID) Accessible Open VO and FTP interfaces. Interoperable: Complies with the IVOA model. Reusable: Highly reusable, semantic metadata. ZENODO Zenodo is derived from Zenodotus the first librarian of the Ancient Library of Alexandria and father of the first recorded use of metadata, a landmark in library history. (Zenodo, n.d.) Metadata schema: Dublin Core, DataCite Metadata Schema Descriptive metadata: Title, authors, affiliation, abstract. Technical metadata: format: FITS, CSV, size, software Copyright metadata: CC license version Provenance metadata: Institutional source or project. Findable A persistent identifier is assigned. The data is described with rich metadata. Accessible The metadata can be retrieved via its unique identifier. Interoperable It uses the JSON schema, which offers export to other popular formats such as Dublin Core. Reusable The metadata is published with a clear and accessible data usage license.