Presentation on theme: "Observations Data Model 2.0"— Presentation transcript:
1Observations Data Model 2.0 A community information model for interoperability among feature-based earth observationsJeff Horsburgh, USU. Project PI.Anthony K. Aufdenkampe, Stroud Water Research CenterKerstin Lehnert, IEDA/ColumbiaEmilio Mayorga, UW-APLIlya Zaslavsky, SDSCDavid Valentine, SDSCDavid Tarboton, USUDavid Lubinski, UC-BoulderI’ll present a combination of the history behind this effort, our achievements and our vision – and along the way explain how the complexities of multiple disciplines and data types has required a community building and engagement effort that serves as an excellent prototype for EarthCube.
2Critical Zone ScienceEarth's permeable near-surface layer from the tops of the trees to the bottom of actively cycling groundwater.Where rock, soil, water, air, and living organisms interact and shape the Earth's surface.Critical to sustaining the earth’s sustaining servicesClean waterProductive soilBalanced atmosphereAtmosphereBiosphereHydrosphereLithosphereMinutesDecadesMilleniaEonsData management needs for critical zone science – seamlessly integrate sensor data logged every few minutes to complex geochemical sample preparation and analysis and sample sharingHillslope Catchment Watershed
6ODM2: Common to Most Data Types SensorExtensionEquipment & LabExtensionsObservationsCoreFeatureModelGenericExtensionCommon Semantics for Earth ObservationsA common information model is critically important to the effectiveness and interoperability of domain Cyberinfrastructure.CUAHSIHISEarthChemCZODataIOOSDomain Cyberinfrastructures
7ODM2: Common to All Components DatabaseEncodingXML SchemaLegendData and Metadata TransferCatalogMetadataCatalogMetadataTransferMetadataTransferMetadata HarvestingData DiscoveryInformation ModelIt forms the conceptual foundation for each component of a data system. In practice, however, information models for each component are often arbitrarily different, limiting capabilities.Data StorageData DeliveryData ServerClientsDataTransfer
8ODM2: Additional Goals Driven by Community & Use Cases: 3 workshops + ~12 data models + much feedbackuse cases: CZOData, Little Bear River, PetDB, IOOSBalance between general vs. understandableExternal unique identifiers, vocabularies & taxonomiesRich Specimen, Site & other Sampling FeaturesGranular Methods, Data Quality & EquipmentDataset publishing & archiving via:Result “packages”, Versions, Citations, ProvenanceStrong Annotations & general extensibility
9ODM2CoreShowing Entity Relationship Diagrams, but Class-Object Model is equally important and we’re actively developing an Object Relation Map
10ODM2CoreAt the very center of ODM2 is the concept of Observations; taken from OGC O&M. We consider ODM2 a profile of O&M.“An Observation is an action whose result is an estimate of the value of some property of the feature-of-interest, obtained using a specified procedure”Use cases pointed out the need to separate Action from Results, and to allow a single Action on many SamplingFeatures.
11ODM2SamplingFeaturesRelationships between “Specimens” and the “Site” at which they were collected are captured in “FeatureParents”, which may also include other feature types
20NSF Scientific Software Integration BiG CZ SSI project ( ):The community-driven BiG CZ software system for integration and analysis of bio- and geoscience data in the critical zoneCommunity Engagement in Software Design through co-design, training & testing workshops.BiG CZ Portal web application for high-performance map-based discovery, visualization, access & publication of data on critical zone structure & functionBiG CZ Toolbox to enable cyber-savvy CZ scientists & data managers to manage and publish the data they produce through a single scientist-focused toolkitBiG CZ Central software stack to bridge data systems developed for multiple critical zone domainsRequired all software developed be Open Source, and required us to name the specific licenses that we will use.
21Thank You Funded by the National Science Foundation EAR 1224638 ACIODM2 is on GitHUB:https://github.com/UCHIC/ODM2
22ODM2: Object-Relation Map Our project objectives requires that we create both object models and relationtional models
23What can we do with ODM2? (that we couldn’t do before) Add multiple comments/annotations to any entityRepresent Actions and sequences of Actions that lead to observation ResultsMore granularly represent people and organizationsStore information about Actions that do not have Results
24What can we do with ODM2? (that we couldn’t do before) Separate Results from ResultValues – enables multiple ResultTypesMove DataValues out of the Core – better facilitates catalogingAdd taxonomic classifiers to Results, adding an additional dimension to observationsCreate relationships among Results and store provenanceGroup Results into Datasets
25What can we do with ODM2? (that we couldn’t do before) Store information about the equipment used to create observationsAdd extension properties to any record in any entityLink many entities to external identifier systemsSupport SamplingFeatures of multiple types - Sites and Specimens, among othersNot limited to a single spatial offsetNot Limited to a single qualifier
26Observation Data Model 2.0 NSF funded project: PI. Jeff Horsburgh“Developing a Community Information Model and Supporting Software to Extend Interoperability of Sensor and Sample Based Earth Observations”To achieve interoperability between IEDA, EarthCHEM, CUAHSI HIS, and other data systemsBetter support for samples and unique identifiers (IGSN/SESAR)Extensibility to table attributesBetter annotation and provenanceEnable integrated web service based publication of a broader class of CZO data
27ODM2 Functional Use Cases Information Model(All)Storage Encoding(USU/LDEO)Catalog Encoding(SDSC)Web Service Interface(UW)Archival Encoding(USU)XML Schema Encoding
28Future Directions for CZO Science Report prepared by CZO community, Dec. 2010Develop a unifying theoretical framework of CZ evolution;Develop coupled systems models to explore how CZ services respond to anthropogenic, climatic, and tectonic forcings;Develop four dimensional data sets thatdocument differing CZ geologic and climatic settings,inform our theoretical framework,constrain our conceptual and coupled systems models,test model-generated hypotheses.Report Prepared by the CZO CommunityDecember 29, 2010
29EarthCube Critical Zone Domain Workshop Engaging the Critical Zone community to bridge long tail science with big dataConvened by A.K. Aufdenkampe, C.J. Duffy, G.E. TuckerUniv. of Delaware: Jan , 2013Organizing Committee:Kerstin Lehnert, IEDA/Columbia.Ilya Zaslavsky, SDSC.David Tarboton, USUJeff Horsburgh, USU.Emilio Mayorga, UW-APLJames Syvitski, CSDMS.Susan Brantley, PSU & SH-CZO.Susan Gill, SWRC.
30103 Participants from 16 Disciplines Biogeochemistry (30)Biology/Ecology (15)Biology/Molecular (3)Climatology/ Meteorology (15)Data Management/CyberInfra structure (46)Engineering/Method Development (8)Geochemistry/Mineralog y (13)Geology/Chronology (14)Geomorphology (15)Geophysics (8)GIS/Remote Sensing (31)Hydrology (46)Modeling/ Computational Science (36)Outreach/ Education Research (7)Soil Science/Pedology (16)Water Chemistry (14)Early-Career (28)