ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Articles | Volume II-2/W1
13 Sep 2013
 | 13 Sep 2013


M.-O. Löwner and T. Becker

Keywords: 3D Data Models, Landform Development, 4D Modelling, Dual Graph Structures, Poincaré Duality, GML

Abstract. Natural landforms are of great importance for a variety of scientific and engineering disciplines. Investigation of landforms can be improved by comparison of features that have similar characteristics, structure and genesis. We propose a novel framework for the representation of the temporal aspects of landform development that simplifies the complex spatial relationships between 3D objects and the modelling of geological processes over time (4D) applying the Poincaré Duality. Single landform layers are represented as nodes (DualStructures) and the neighbourhood of these layers are represented as edges (DualStructureRelations). Finally, a DualStructureState represents a whole landform of stable conditions over a period of time. Change of a landform is represented as additional edges (Abstract_GeoProcess) between the nodes of different layers. The overall structure constitutes a multilayer graph, where all the nodes from all N layers are included but are separated into N partitions of time. All dual representations may be associated with geometric and semantic representations, if available. A formal data model on natural landforms focusing on topological representation is a major step towards the interoperable exchange and comparison of scientific results on landforms. Concerning existing models, our framework can be considered as a superset with regard to model expressivity. This will improve the possibilities to exchange or link data between different application fields.