Paradata in Virtual Reconstruction: Towards a Conceptual Model of Transparent Reconstruction
Abstract
Virtual reconstruction has become a central method in digital heritage research, particularly for documenting destroyed, altered, or unrealized architectural heritage. However, reconstruction workflows remain inherently interpretative, especially in archive-based modeling, where incomplete or contradictory sources require inferential decisions. This review examines how recent literature conceptualizes and operationalizes paradata as a mechanism for documenting interpretative reasoning, provenance, and uncertainty in digital heritage workflows. Based on the thematic synthesis, the paper further proposes a conceptual model that integrates the identified research directions into a unified representation linking conceptual principles, reconstruction workflows, and implementation technologies. Using a focused qualitative thematic literature review, the paper analyses recurring themes, including the distinction between metadata and paradata, semantic provenance modeling, CIDOC-CRM-based documentation, HBIM interoperability, FAIR data principles, explainable reconstruction workflows, and the longterm preservation of born-digital heritage assets. The review identifies a consistent shift from purely visual reconstruction toward machine-readable transparency frameworks that connect sources, interpretative assumptions, workflows, and resulting visualizations. At the same time, the literature reveals persistent methodological fragmentation, limited standardization, and insufficient integration of uncertainty documentation into operational reconstruction pipelines. Recent studies increasingly emphasize semantic interoperability, knowledge graphs, provenance ontologies, and lightweight paradata schemas that support reuse, scholarly validation, and digital preservation. The proposed three-layer conceptual model extends the literature by integrating previously fragmented perspectives into a coherent structure that may support future archive-based virtual reconstruction projects and interoperable paradata implementations.