Semantics in Ecosystem Research and Monitoring

Abstract

The field of ecology in general, and environmental assessment in particular, demands the sharing of knowledge, information and data. On the European level –on the legal basis of the INSPIRE directive - a framework has been established which enables public access to geo-data in a structurally harmonized way. However, for ecological data the temporal dimension is just as important as the spatial dimension. Some of the existing data integration approaches show that efforts are needed to extend structural harmonization and include semantic harmonization. The sharing of knowledge, information and data implies a common understanding of the meaning of terms and concepts. This requirement has been met by controlled vocabularies, such as species lists and other taxonomies or catalogues of domain terminologies, long before the first computer was built. Current IT technologies have adopted these concepts of controlled vocabularies, and established and published them in digital form, mostly via the world wide web. This has resulted in a lot of benefits, such as accessibility, shared editing and the usability of controlled vocabularies in all sorts of applications. Some of the most prominent vocabularies are GEMET, CORINE Land Cover classes, EUNIS habitat list, Catalogue of Life, SERONTO, OBOE, Observation and Measurement, just to name a few. Those controlled vocabularies can be used in various ways: - As reference lists for scientific publications: e.g.: looking up GEMET concepts in different European languages on the site of the EEA (EIONET) - To tag metadata with keywords using controlled vocabularies by e.g. inserting keywords into an ISO19115 compliant XML document, as demanded by the INSPIRE directive, using SoilThes as a source for the keywords. - Semantic based data management linking data and semantically enriched metadata, e.g. of Integrated Monitoring Austria using the information system MORIS. However, efficient use of these resources is still hampered by the lack of a standardized framework for their interlinkage. The need for such a framework is not specific to the field of ecology or science in general. It is a requirement for all domains dealing with the sharing of information, knowledge and data. Technologies based on internet technologies such as the emerging Linked Data approach are trying to meet this challenge. This article first focuses on the specific needs for the use of semantics in ecological monitoring and gives a rough overview of how these have been met so far, independent of IT solutions. Secondly, we describe some technical approaches to meet these requirements and outline how these approaches are applied to specific solutions. Then we give an outlook on how these solutions could become part of a larger network of linked ecological data.