3D Spatial Mapping and Immersive Visualisation of Forest Biodiversity

As forests face unprecedented environmental challenges, maintaining ecosystem, species, and genetic diversity is essential to their resilience and the critical functions they provide for humanity. Forest biodiversity serves as a natural buffer against the increasing frequency and intensity of extreme events. Forest biodiversity inventories have traditionally played a vital role in tracking ecological changes and guiding conservation efforts. However, with climate change escalating the scale and frequency of forest disturbances, conventional methods—which primarily rely on visual assessments within a limited number of plots—are becoming increasingly inadequate. The aim of this thesis is to develop a comprehensive methodology aligned with current needs in forest biodiversity monitoring, encompassing data acquisition, processing, and visualisation. To address these challenges, an innovative framework is proposed that leverages advanced technologies across three core stages: 3D spatial mapping for high-resolution data acquisition, machine learning (ML) for efficient data processing, and virtual reality (VR) and mixed reality (MR) for immersive visualisation. In Chapter 1, the motivations, objectives, and structure of this thesis are introduced, which lays the foundation for the investigation. In Chapter 2, essential background information on forest biodiversity concepts are provided and the technologies employed in this study are detailed. In Chapter 3, the first research question is addressed: Which 3D mapping technologies are most suitable for biodiversity monitoring in NFIs? The findings indicate that photogrammetric surveys are particularly effective for biodiversity tracking, offering high-resolution, photorealistic 3D point clouds. In Chapter 4, the acquired photogrammetry dataset—comprising 17 high-resolution 3D point clouds of individual tree stems—is utilised to address the second research question: How can VR and ML contribute to a better understanding of forest biodiversity? It is demonstrated that VR can serve as an interactive tool to annotate biodiversity features on 3D point clouds, which can then be used as a training dataset for ML models. In Chapter 5, a pipeline is developed to reintegrate processed 3D biodiversity data into the physical forest environment where it was initially captured. This approach addresses the third research question: What 3D visualisation tools can MR offer to facilitate forest biodiversity management? This method enables in-situ data verification, providing researchers with an invaluable tool for refining data accuracy over time as new forest disturbances arise. However, enhanced monitoring capacity alone does not guarantee biodiversity preservation; it must be accompanied by public understanding and engagement. Finally, the limitations and outlook of this thesis are presented in Chapter 6. Overall, this thesis demonstrates that 3D spatial mapping and immersive visualisation technologies can be effectively integrated into forest biodiversity inventories, and it offers a comprehensive workflow for their application. By extending the reach of this work beyond scientific circles and merging scientific insights with practice and artistic exploration, this thesis helps to inspire a cultural transformation in humanity’s relationship with nature and promotes the perspective of biodiversity preservation as a shared responsibility.