scieee AI-readable full text Open interactive document viewer

The Role of CDS in Data Curation and Standardization

Eisele, Aline

Abstract

The Strasbourg Astronomical Data Center (CDS), has for over 50 years focused on collecting, improving, and disseminating astronomical data. It adapts to technological advancements and engages in open science, collaborating across disciplines and institutions. The CDS team, including astronomers, IT specialists, and data curators, maintains key services like SIMBAD, VizieR, and Aladin. Data curation at CDS adheres to FAIR principles, ensuring data are Findable, Accessible, Interoperable, and Reusable. The DJIN tool aids in metadata creation for SIMBAD, enriching information and calculating relevance scores for references. The curators of VizieR manage and standardize astronomical data, handling tables up to billions of rows, using FAIR principles and standards to organize data with keywords, descriptions, and metadata, enabling specific searches, with each catalogue having a unique DOI for direct access. The Aladin service uses HiPS as an efficient method that enables access, optimal visualisation and analysis of large image data surveys. All of the services employ metadata and standards ensure interoperability and accessibility. This presentation will highlight CDS's role in managing and standardizing astronomical data across SIMBAD, VizieR, and Aladin underscoring the importance of metadata and standards in improving data accessibility and interoperability, and highlighting the significance of manual curation for quality data.

Full text

The Role of CDS in Data Curation and Standardization [email protected] Aline Eisele on behalf of the CDS team LISA X Santiago, 05 November 2025 Collecting, Improving, and Distributing Astronomical Data 2/24 24 Different levels of data/metadata Different professional profiles SIMBAD database : Djin point of view and example Role of the different documentalist teams 3/24 3 Different levels of data/metadata Metadata Publishers Metadata Bibliographic Metadata Computer/IT Metadata Sales, marketing, distribution, rights management Discovery, cataloging, longterm preservation Interoperability between data 4/24 4 Different levels of data/metadata data librarian): organizing, managing and supporting the management of research data in a digital environment, with a role focused on the exploitation and structuring of data throughout their life cycle Different professional profiles data scientist/data analyst : Analyze / Represent / Model / Predict ( statistics, algorithm, machine learning) => analysis output data curator): Clean, refine, convert data metadata curator): describe data (by scientific metadata) Documentalists at CDS): VizieR): data and metadata curator Simbad): bibliographic metadata curator Aladin): metadata curator 5/24 5 Different levels of data/metadata Scientific data/scientific metadata Documentary (bibliographic) metadata Computer metadata 6/24 6 Different levels of data/metadata Scientific data/scientific metadata Documentary (bibliographic) metadata Computer metadata 7/24 7 Different levels of data/metadata Scientific data/scientific metadata Documentary (bibliographic) metadata Computer metadata Unified Content Descriptors UCDs are a controlled vocabulary for describing astronomical data, ensuring interoperability and clarity for both humans and machines 8/24 8 Different levels of data/metadata Scientific data/scientific metadata Documentary (bibliographic) metadata Computer metadata Unified Content Descriptors UCDs are a controlled vocabulary for describing astronomical data, ensuring interoperability and clarity for both humans and machines Created by Astronomers 9/24 9 Different levels of data/metadata Scientific data/scientific metadata Documentary (bibliographic) metadata Computer metadata Unified Content Descriptors UCDs are a controlled vocabulary for describing astronomical data, ensuring interoperability and clarity for both humans and machines structure determined by computer Scientists 16/24 16 SIMBAD database": Djin example 17/24 Role of CDS identify verify homogenize standardize describe add metadata Homogenisation and standardisation to improve the data quality and interoperability 18/24 Simbad are we talking about the same astronomical objet identify verify homogenize the names through the Dictionary of nomenclature standardize same unit of measure index/ reference link an astronomical object with an article add metadata DJIN software adds -the location of the object in the article -the occurrence (number of appearances in the text) -the identifier used 19/24 Simbad identify verify homogenize standardize index link an astronomical object with an article add data Data extraction and formatting Compare what is already present in Simbad same unit of measure Like for DJIN): location, occurrence, identifier used add metadata Basic data / measures 20/24 VizieR identify verify describe standardize add metadata from where are coming the measurements Identifiers (SIMBAD, NED, other catalogs) the format of the table, Units & Labels homogenize redaction of a ReadMe file: description of tables content (measurement explanations, origins, etc.) UCDs, filters, keywords add data Links, Columns, Tables (references, summary,...), Files (spectra...), plots index Main ID, coordinate, dates 21/24 Aladin homogenize index describe verify Ensure that the image header contains valid astrometric WCS (World Coordinate System) keywords sky surveys through the HiPS IVOA standard (Hierarchical Progressive Surveys) link the standardized image sky survey with the original images short description, origin of the original images, rights, associated bibliography, wavelength, observation period add metadata ObsCore IVOA vocabulary and syntax manually or automaticaly by the HiPSgen tool 22/24 FAIR Findable Accessible Interoperable Reusable through descriptions, key-words, indexation via standard protocols and different search methods: : by identifier, coordinate, bibliographic reference… through notebooks, tutorials and associated data standards enable the comparison and combination of data — and enables links bewteen CDS and other databases and archives (e.g. IVOA standards, DOIs) 23/24 Conclusion Research Scientists Data curators Software engineers 24/24 Conclusion