An ontology for the representation of Earth Observation data: a step towards semantic interoperability

Earth observation (EO) consists in collecting information related to the Earth's physical, chemical and biological systems. Data are gathered through remote sensing technologies, mainly consisting of satellites, and represent an essential source of information, which can be associated with administrative, social and economic issues in order to support policy analysis and decision-making. In recent years, characterized by an intensification of climate change phenomena and of extreme environmental events, the amount of EO data collected has significantly increased. Nevertheless, their huge volume and heterogeneity do not allow the data to be easily and promptly used by the scientific community. The different methods adopted for collecting, processing, cataloguing and describing data through metadata introduce an additional high level of variability. In this sense, it is important to guarantee technical, syntactic and semantic interoperability among data. This paper focuses on semantic interoperability issues in the EO domain and introduces an ontological representation of knowledge tailored to the concept of Essential Variables (EVs). The ontological model has been defined within a European research program which is also oriented towards the development of a Knowledge Base in the specific domain. Its general intent, however, is to provide a framework concerning a set of EVs, identified and characterized by a community of experts in order to guarantee information and knowledge generation from observable environmental data.