Using semantic-based spatial reclassification for interoperable data management in Natura 2000 monitoring

In various fields of spatial research, semantic heterogeneity remains an unsolved problem in terms of data comparability. Interoperability of biodiversity monitoring information is especially difficult because of its thematic complexity and the high variability of acquisition methods and national differences in nomenclatures. Each EU member state must compile comprehensive information on protected areas every six years to comply with the reporting obligations of the habitats directive. Since data collection methods and interpretation manuals broadly vary by member state and manual delineation of protected areas can never be neutral, there is a need for automatized, objective methodologies for the generation of comparable datasets. Comparable datasets derived objectively would further support decision making on a European level. Ontology-based applications offer vast opportunities in data management regarding the interoperability of this kind of information. Basis of this study are two datasets of protected heathlands in Germany and Belgium which are derived from remote sensing and semantically formalized in an OWL2 ontology. The proposed methodology uses semantic relations of the two datasets, which are (semi-) automatically derived from remote sensing imagery, to generate objective and comparable information about the status of protected areas by utilizing kernel-based spatial reclassification (methods/algorithms/techniques). The method therefore suggests a generalization approach, which is able to generate delineation of protected areas of the Natura 2000 network in an entirely automatized procedure. Furthermore, it is able to transfer generalization rules between areas surveyed with varying acquisition methods in different countries by taking into account automated inference of the underlying semantics.