Semantics enabled framework for knowledge discovery from Earth Observation data in coastal zones

For millennia, coastal zones of the world have been major centers of human population. There is a wealth of accumulated information about coastal zones, such as data/images in various databases, files, spreadsheets, video and audio data. However optimal harnessing of these resources has long been recognized as an insurmountable task. The major challenges being the heterogeneous nature of the data due to the diverse procedures and techniques used to collect it, and stored in a variety of formats and at different locations. The Earth Observation (EO) satellites have been collecting huge amounts of data over the past decades (Landsat data alone comprises 434 terabytes of archive), however the current methods of searching for useful information is only at the syntactic metadata level, thus the optimal exploitation of the archived data is severely constrained by the lack of content and semantics based knowledge retrieval. In this paper we present a semantics enabled framework for content-based retrieval from remote sensing data and also for integrating heterogeneous resources in coastal zones. We discuss the need for an ontology-driven middleware to achieve such interoperability. A methodology for domain specific qualitative spatial reasoning in coastal wetlands is also presented.