Abstract:
Dozens of terabytes of Earth observation data and their derived science products are
generated each day. For the tremendous and precious Earth science data and information
to be fully explored and utilized, it is required that advanced technologies be developed
and deployed for data providers to easily publish and deliver their products and data
consumers to easily find and obtain needed data, preferably in customized forms. The
purpose of this thesis research is to develop a service- and data-casting architecture for
publication and subscription of Earth science data, information, and services. The
architecture will not only enable generation of feeds for any updates and changes
occurring to Earth Science data and Web Services from a particular provider but also
allow aggregation of feeds from other data and service providers, thus significantly
enhancing the visibility of available data and services and easing the searching and access
of data. A prototypical system based on the architecture has been developed and deployed
to publish the data and services provided by GeoBrain, a distributed interoperable webbased
geospatial information system. The terms "serviceCasting" and "dataCasting" were
coined because of their similarities with the "Podcasting" concept; they all describe the
delivery of metadata over the Internet by information feeds based on Really Simple
Syndication (RSS) or ATOM standard. GeoBrain data and services are published via
ATOM feeds to the cloud and to be syndicated in the to-be-created geographic feed
reader. In order to satisfy the special needs of geographers, two namespaces and
additional elements are added to enhance geographic features descriptions. The
prototypical system demonstrates that syndication and aggregation of feeds can
decentralize the repository of earth science data and make information more discoverable
and accessible.
