Comparison of 3D structural metrics on oyster reefs using unoccupied aircraft photogrammetry and terrestrial LiDAR across a tidal elevation gradient

Abstract

Abstract Physical structures generated from ecosystem engineers can have a cascade of impacts on the ecological community and the surrounding landscape. The Eastern oyster Crassostrea virginica can form extensive intertidal reefs, whose three‐dimensional structures provide ecosystem services like nursery and foraging habitat for fishes and invertebrates and shoreline stabilization. Measurements of the structural properties of these reefs provide opportunities to quantitatively assess associated services. There is a growing variety of tools available for measuring three‐dimensional (3D) properties of intertidal habitats, including two remote sensing methods that capture 3D structural metrics in a number of environments. We surveyed reefs using a terrestrial laser scanner (TLS, LiDAR) and imagery from unoccupied aircraft systems (UAS, or drones) processed through Structure from Motion photogrammetry. Comparisons of digital elevation models from repetitive flights over an oyster reef to checkpoints yielded mean horizontal and vertical root mean square errors (RMSE) of −0.54 ± 0.47 cm and 0.97 ± 1.0 cm (Mean ± SD), respectively, indicating high accuracy among UAS surveys. Compared to TLS products, point cloud densities from UAS‐derived products were more consistent across the reef elevation gradient and much denser overall except in the low reef zone, which was proximal to most of the TLS scan locations. Comparisons of structural metrics between UAS and TLS showed similarities in metrics like profile and planform curvatures, yet indicated UAS surveys produced higher values of surface complexity and slope. Results indicate that UAS photogrammetry can produce robust oyster reef structural metrics that can be highly useful in oyster conservation and restoration.

Keywords

• Digital elevation models
• drone
• lidar
• oyster reefs
• rugosity
• structure from motion