With the growing dominance of complex and multi-level urban structures, current cadastral systems, which are often developed based on 2D representations, are not capable of providing unambiguous spatial information about urban properties. Therefore, the concept of 3D cadastre is proposed to support 3D digital representation of land and properties and facilitate the communication of legal ownership rights associated with them. On the other hand, Building Information Modeling (BIM) provides a common data environment for integrated management of 3D building lifecycle data in the construction industry. Harnessing BIM for 3D cadastre will provide the ability to capture and manage the ownership and legal status of properties in modern cities, providing an efficient approach to access legal information about urban properties. However, querying the meaning of property information inside BIM models has not been addressed. There are studies which investigated the adoption of the semantic web techniques for a better understanding of concepts of the building lifecycle in BIM; however, these investigations do not provide a semantic approach to query and retrieve property information encoded in the BIM environment.

Therefore, this study aims to adopt a linked data methodology to develop a BIM-based spatial ontology for managing 3D property information. In addition, SPARQL query language will be used to retrieve 3D property information. The appropriate entities for modelling properties the in the open BIM data model were deciphered and the spatial ontology was developed in accordance with the BIM open data model structure. To assess the feasibility of the developed spatial ontology, BIM models of multi-story buildings with different levels of complexity and detail were implemented. These models were transformed into the RDF (Resource Description Framework) structure, and 3D semantic queries related to 3D property information have been done in semantic BIM models. These queries retrieve boundaries and spatial arrangements of properties. The query results were displayed using a WebGL viewer, and retrieval time of each query was measured. The findings of this study indicate semantic querying of 3D property information is faster than spatial queries relying on the geometry of 3D property objects.