Rolling Deck to Repository (R2R): A "Linked Data" Approach for the U.S. Academic Research Fleet
Abstract
The U.S. academic research fleet is an essential mobile observing platform for ocean science. The NSF-funded Rolling Deck to Repository (R2R) program is working with the fleet community to routinely document, assess, and preserve data from the underway sensor systems on each vessel. <br /> The R2R facility maintains a master catalog of vessels, instrument systems, operating institutions, cruises, personnel, and data sets. In less than two years, the catalog has grown to over 2,000 cruises including unique identifiers for vessel deployments, project titles, chief scientists, dates, ports, survey targets, and navigation tracks. This master catalog is of great value to peer data systems, ranging from large national data centers to small disciplinary data offices, as an aid in organizing, extending, and quality controlling their own collections and finding related data from authoritative sources. <br /> R2R breaks with the tradition of stovepipe portals built around complex search interfaces tightly bound to backend databases. Instead, we have adopted a Linked Data approach to publish our catalog content, based on the W3C Resource Description Framework (RDF) and Uniform Resource Identifiers (URIs). Our data model is published as a collection of W3C Simple Knowledge Organization System (SKOS) concepts, mapped to partner vocabularies such as those developed by the Global Change Master Directory (GCMD) and the pan-European SeaDataNet partnership, and our catalog content is published as collections of RDF resources with globally unique and persistent identifiers. The combination of exposing our data model, mapping local terms to community-wide vocabularies, and using reliable identifiers improves interoperability and reduces ambiguity. R2R's metric of success is the degree to which peer data systems routinely harvest and reuse our content. <br /> R2R is working collaboratively with the NOAA National Data Centers and the NSF-funded Biological and Chemical Oceanography Data Management Office (BCO-DMO) on a range of Linked Data pilot applications, including production of ISO-compliant metadata and deployment of a SPARQL search interface. Our objective is to support a distributed, loosely federated network of complementary systems that collectively manage the vast body of ocean science data. We will present results and lessons learned.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2011
- Bibcode:
- 2011AGUFMIN33E..03A
- Keywords:
-
- 1930 INFORMATICS / Data and information governance;
- 1936 INFORMATICS / Interoperability;
- 1970 INFORMATICS / Semantic web and semantic integration;
- 1982 INFORMATICS / Standards