Global Cryosphere Watch GCW web portal Structure and demonstration Øystein Godøy
The high level purpose The GCW web portal will make GCW data and information available to WMO Members, their partners, and users while providing the ability to exchange data and information among a distributed network of providers of data and products. The portal, as a part of WIS, will allow for rapid exchange of data, metadata, information, and analyses.
The GCW portal Two main components Editorial component Addressing editorial content prepared by the community Distributed by nature Community contributions Push information through Catalogue component Addressing metadata from various catalogues preferably linking directly to data Distributed by nature Harvest/distributed search across relevant existing catalogues Pull information through
The current situation looking for data Fragmented view Multiple portals/catalogues to check Limited availability of data online Multiple or even incompatible file formats/interfaces for products Much resources spent reformatting/preparing data Highly variable documentation standards e.g. positioning information e.g. quality information e.g. traceability
Why is data management important for GCW? A major issue for GCW is improved exchange of observations and products improved access to observations and products improved utilization of observations and products from WMO and other observing systems Establishing interoperability between data management systems supporting cryospheric data directly addresses GCW objectives. This is addressed through the catalogue component mainly
GCW will link infrastructure GCW aims to utilise relevant information no matter where it is located by linking and bridging existing infrastructure. GCW will not reinvent anything
Distributed data management in practise and why... Utilise existing infrastructure and efforts Avoid duplication of data Integrate existing infrastructure through Machine interfaces to metadata and data Exchange of metadata Sustainable solution Datasets are maintained by owners/creators Information on datasets are maintained by owners/creators Subsystems work...
Necessary steps for distributed data management Standardisation of documentation, file formats/access mechanisms etc supporting system interoperability Implementation of interfaces supporting the standards Willingness to utilise standards Willingness to properly document data Willingness to properly share data and information Translation between various interoperability standards/dialects Establish rules of operation
Distributed data management
Preliminary functional requirements Users of cryospheric data, products and information should be able to easily find what they need. The portal should support both human and machine interfaces. While discovery and access to data is the primary functionality of any portal, experience shows that users soon request more advanced functionality, like visualisation, transformation, reformatting, and re-projection of datasets
GCW portal, catalogue component roles and preliminary priority It should support both Data producers Data consumers Data discovery has first priority Higher order functionality should be enabled wherever possible Visualisation Transformation
Technological considerations for catalogue interoperability Expose metadata using standard interoperability interfaces and documentation standards e.g. OAI-PMH, OGC CSW, ISO23950 e.g. ISO19115, GCMD DIF Preferably keep data online Implement machine interfaces in addition to human interfaces May enable integrated services across cryosphere datasets Self describing data formats/interfaces are highly preferred If authentication/authorisation is required, consider mechanisms like CAS, OpenSSO etc
Relevant frameworks for catalogue interoperability WMO Information System (WIS) ICSU World Data System (WDS) Group on Earth Observation (GEO) Regional frameworks E.g INSPIRE Domain specific frameworks Global Change Master Directory International Permafrost Association IPYDIS
Challenges developing the GCW portal Content management for editorial information How to handle the open data space Impacts on catalogue component What is a GCW dataset? How to address the diverse subsystems How many interfaces to support? Interoperability standards How to achieve a sustainable solution? For the editorial component For the catalogue Ensuring future relevance of GCW
Demonstrator The demonstrator is available at http://gcwdemo.met.no/ Addresses only the catalogue component of the GCW portal Filtering of information is not implemented for all streams It contains currently information on IPY datasets documented in Norway Through metadata exchange between Norwegian Polar Institute, Institute of Marine Research and Norwegian Meteorological Institute EU FP7 DAMOCLES and EUMETSAT OSISAF datasets NSIDC data
Relation to WIS components
Preliminary plans for the demonstrator Update the software Improve filtering Move demonstrator to a new computer and database, linking with METNO WIS nodes Add regular harvest from CCIN Harvest has been tested, but ingestion into the database has yet not Examine interoperability towards e.g. BAS Examine interoperability towards IPA Examine sustainable visualisation and transformation of remote datasets
Questions for discussion Desired functionality of the portal Editorial component Catalogue component Organisational structure required To support the portal components Interaction with existing systems Which frameworks/systems to contact? Defining GCW datasets? Cost / benefit analysis Addressing requested functionality And resources available What is achievable with the resources available? How to consolidate a system description How to consolidate an implementation plan