Three-dimensional models of coral microatolls using structure-from-motion photogrammetry and iPhone LiDAR scanning: A fast, reproducible method for collecting relative sea-level data in the field

Abstract

Coral microatolls, geological proxies commonly used for reconstructing relative sea-level (RSL) in low-latitude regions, are valued for their precision and ability to continuously track RSL changes through the elevation of successive concentric surface rings. The brief low-tide window prevents rigorous methods for replicating field observations, limiting opportunities for reinterpretation of coral morphology. Additionally, while the extraction of a physical coral slab remains the preferred method for RSL reconstruction, logistical constraints can render it non-viable. When slabbing is possible, the reliability of the reconstructed RSL might be questionable. This study introduces three-dimensional models created using structure-from-motion photogrammetry and iPhone LiDAR scans to facilitate rigorous analysis of coral microatolls. These methods result in accurate and high-resolution documentation of the coral surface, enabling comprehensive and simultaneous analysis of ring structures of multiple microatolls while ensuring results are representative and replicable. Where slabbing is feasible, this method guides the selection of optimal corals that contain the most complete record of RSL change and validates slabbing results. Where slabbing is not viable, this approach provides an alternative means to obtaining RSL histories. Integrating this model-based approach into conventional fieldwork enables extensive data interpre tation off-site. Furthermore, the user-friendly nature of these methods enhances accessibility for researchers with limited resources. The benefits and limitations of each technique are also discussed. While photogrammetry derived point clouds are denser, they necessitate additional georeferencing steps to ensure accurate scale and orientation. Conversely, iPhone-derived models possess inherent scale, though they require additional processing steps, carrying a potential risk of data loss.