Daejeon, July 27–31
According to the IUCN, natural heritage is not a mere natural object but a complex body of knowledge requiring environmental, historical, and geographical contexts (UNESCO 1972). Animal specimens function as physical mediators that embody and transmit this value. Yet existing archives in Korea remain largely preservation-oriented, offering only taxonomic information, 2D images, and minimal histories—insufficient to convey ecological and historical contexts (NIBR 2025a; NIBR 2025b). In this study, the Anthropocene serves as a theoretical lens for reinterpreting natural heritage specimens as evidence of human-environment interactions. From an Anthropocenic perspective (Möllers et al. 2022), specimens are not merely natural history materials preserving morphological and taxonomic information about species, but complex knowledge objects at the intersection of habitat change, climate crisis, human collecting practices, and institutional preservation systems. Accordingly, this study connects 3D specimen models, collection sites, current holding institutions, movement trajectories, and climate data, enabling specimens to be explored not as isolated natural objects but within the interwoven contexts of ecology and history.
Existing platforms address aspects of natural heritage digitization but leave critical interpretive gaps. The Global Biodiversity Information Facility (GBIF) aggregates millions of biological records, yet its list-based, 2D interface is optimized for macro-level statistical ecology rather than contextual interpretation. Sketchfab and the Smithsonian 3D Digitization Program offer compelling three-dimensional specimen renderings, but present models as isolated artifacts divorced from geographical provenance and collection history. No existing platform integrates WebGL-based 3D rendering with GIS habitat mapping, movement trajectory visualization, and climate overlays within a unified semantic architecture. The proposed platform addresses this gap by adopting a Deep Mapping approach (Bodenhamer et al. 2015) that fuses biological and cultural heritage standards to enable spatiotemporal exploration of natural heritage.
To protect specimens at risk of physical damage, this study conducted 3D digitization at four partner institutions: the Natural Heritage Center (153 species; 229 specimens; 229 3D models), Seoul Grand Park Zoo (26 species; 33 specimens; 33 3D models), Hannam University Natural History Museum (85 species; 100 specimens; 100 3D models), and Kyungpook National University Natural History Museum (22 species; 22 specimens; 22 3D models), totaling 384 specimens and 384 3D models. Digital twins were generated via photogrammetry and structured-light 3D scanning. An integrated management structure of three entities—Nature, Specimen, and Project—links biological information, physical management data, and digitization records, systematizing scattered information as digital assets (Figure 1).
Figure 1 The Data Schema and Process Flow of the 3D Natural Heritage Archive.
The Nature entity applies Darwin Core (Wieczorek et al. 2012), which is well suited for describing species classification, occurrence, habitat, and observation and collection data. The Specimen entity adopts CIDOC-CRM (Le Boeuf et al. 2021), which excels at representing event-based histories such as acquisition, movement, conservation, management, and digitization of specimens. The Project entity applies CRMdig (Doerr / Doerr 2011)—the CIDOC-CRM extension for digital provenance—formally capturing the digitization process, scanning equipment, software parameters, and output file metadata in a machine-readable and reproducible form. This study combines DwC and CIDOC-CRM complementarily to model natural heritage specimens as both biological entities and heritage objects that are collected, managed, and interpreted by institutions.
The platform uses Next.js for the front end, Three.js/WebGL for plug-in-free 3D rendering, the Google Maps API for GIS layers, and Supabase/PostgreSQL on the back end. A web-based admin interface enables curators to manage specimens, taxa, and protection categories through a tree-structured taxonomy view without direct database access. The open-source stack ensures long-term sustainability in three practical ways. First, all tools are freely available, so institutions do not need a large budget to adopt the platform. Second, the libraries are maintained by active global developer communities, meaning the system continues to receive updates and improvements without relying on any single company or product. Third, the platform stores data in widely accepted open formats—glTF for 3D models and GeoJSON for spatial data—so any institution can access, transfer, or rebuild the system without being tied to specific software.
The platform implements three visualization layers. First, habitat mapping plots discovery and death locations onto a GIS interface, revealing a species' spatial range and ecological characteristics. Second, movement trajectory visualization connects past collection sites with present holding institutions, tracing the historical journey from wild animal to museum specimen. Third, proposed climate overlays will contrast past climatic conditions—drawn from Korea Meteorological Administration (KMA) public data—with the present, prompting users to recognize disappearing biodiversity rather than perceiving specimens as static objects. To implement this feature, live KMA API integration will be added in the next development phase. Users can rotate and zoom 3D models at specimen level, then navigate the map to explore distribution and movement patterns across regions and periods.
Figure 2 User Interface of the Map-based 3D Natural Heritage Visualization Platform.
This platform offers distinct contributions across five stakeholder groups. For specimen managers, the three-entity schema and CRMdig-based Project records provide a rigorous, reproducible standard for tracking digitization provenance. For ecology managers, habitat and climate overlay layers transform isolated records into tools for monitoring long-term environmental change. For researchers, the fusion of DwC and CIDOC-CRM within a single architecture opens new pathways for interdisciplinary inquiry at the intersection of biodiversity informatics and spatial humanities. For educators, the interactive 3D and map-based interface makes abstract ecological histories tangible and accessible. For the general public, the platform democratizes access to natural heritage knowledge, inviting reflection on biodiversity loss in real time. By treating specimens not as relics of a stable natural world but as witnesses to its ongoing transformation, the project advances DH's data-modeling capabilities and provides a practical framework for Ecological Humanities in the AI era (Cameron 2015).