Extending the Interoperability of Sensor and Sample Based Earth Observations using a Community Information Model (Invited)

With recent initiatives within many scientific communities toward trans-disciplinary, synthetic research using both new and existing data resources, our ability to manage, share, discover, interpret, and integrate data is paramount to scientific progress. Indeed, many new scientific advances are dependent on the synthesis of observations from multiple measurements, at multiple scales, across scientific disciplines, across environmental observatory or other experimental sites, and from multiple sources. We are at a point where the inadequacy of tools available for describing and sharing data leads to heterogeneity in the way data are organized, annotated, and encoded that hinders its discovery and interpretation. Several systems have emerged within geoscience communities for sharing earth observations, including the Consortium of Universities for the Advancement of Hydrologic Science, Inc. (CUAHSI) Hydrologic Information System (HIS), EarthChem, the Integrated Ocean Observing System (IOOS), and, more recently, the Critical Zone Observatory (CZO) Integrated Data Management System (CZOData). These systems focus on publishing or sharing data using web services in domain specific encodings or markup languages. While these systems have made considerable progress in making data available, it still takes a knowledgeable investigator considerable effort to discover, access, and integrate datasets from multiple domain-specific repositories because of inconsistencies in the way the different domain systems describe, encode, and share data. These inconsistencies also limit the ability of scientists from outside the domain to unambiguously interpret the data. With input from the broad geoscience community, we are developing a community information model ('ODM2') to extend interoperability of earth observations and improve the capture, management, sharing, and archival these data. We have focused on the domain of spatially discrete, feature-based earth observations resulting from in situ sensors and from environmental samples, as well as data products directly derived from them. Using the existing CUAHSI HIS Observations Data Model (ODM), EarthChem's database structure, and the Open Geospatial Consortium's Observations & Measurements specification as starting points, our multidisciplinary, community-focused effort has been aimed at building consensus about the elements of the information model and addressing deficiencies in data interoperability both within and among existing geoscience cyberinfrastructures. The supporting software infrastructure we are developing includes storage, transfer, and catalog encodings of the information model and additional software tools aimed at improving data capture, validation, verification, sharing, and archival. We are using diverse data use cases from existing repositories and observatories to demonstrate how this advanced information model can support federation of earth observational data across multiple data publication systems within the geosciences. We anticipate that this information model and its prototype implementations can also serve as a common conceptual foundation for the next generation of geoscience cyberinfrastructure. In this presentation we describe our draft designs for the ODM2 information model and how ODM2 is foundational in achieving deeper interoperability across multiple disciplines and systems to support powerful data discovery, access, publication and analysis capabilities.