From mobile LiDAR to neural radiance fields: practical pathways for accessible ecological 3D reconstruction

Henry Cerbone, Sruthi M. Krishna Moorthy, Rob Salguero-Gomez, Graham Taylor

Ecological monitoring increasingly depends on high resolution three dimensional data, yet many reconstruction workflows remain constrained by the cost and logistical complexity of terrestrial laser scanning (TLS). At the same time, advances in mobile device imaging have enabled accessible three dimensional reconstruction using mobile LiDAR, structure from motion (SfM) photogrammetry, and neural radiance fields (NeRFs). Here, we present an interdisciplinary perspective on mobile phone based reconstruction workflows for ecology and compare their tradeoffs in accessibility, geometric fidelity, realism, and computational demand. Using a case study in Wytham Woods, UK, we compare reconstructions of trees generated using TLS, mobile LiDAR, SfM photogrammetry, app-based NeRFs, and NeRFs. Rather than positioning radiance fields as replacements for TLS or established photogrammetric workflows, we argue that they occupy a complementary niche by combining accessibility, visual realism, and reusable digital scene representations for ecological monitoring. We conclude with practical recommendations for selecting workflows across ecological applications including forest inventories, understory monitoring, and long term ecosystem archives.