Assessing the effects of engineered oyster reefs on shoreline change using drones

Introduction Coastal infrastructure and property, as well as intertidal wetlands, are increasingly being threatened by shoreline erosion; a consequence of human activities and climate change. Nature-based solutions, such as intertidal engineered oyster reefs, can reduce erosion and promote sediment accretion, thereby promoting the restoration and persistence of salt marshes and preventing the loss of coastal lands. Engineered oyster reef substrate and design options have rapidly expanded in the last decade, yet our understanding of how these approaches influence ecosystems and intertidal morphology is limited. Drones (or small uncrewed aerial systems sUAS) coupled with structure-from-motion (SfM) photogrammetry have recently been suggested as a low-cost method that offers optimal spatial coverage, fine-scale resolution, and high vertical accuracy for monitoring changes around living shorelines. Methods We evaluated how using different vertical and horizontal uncertainty thresholds for detection of drone-based shoreline change can influence interpretation of performance of engineered oyster reefs on coastal morphology and vegetation. We monitored three sites with engineered oyster reefs installed in 2020 and one reference site located on Carrot Island along Taylor Creek in Beaufort, NC, USA. Results Comparisons of the Digital Elevation Models (DEMs) and orthomosaics derived from the drone imagery revealed all sites saw marsh edge retreat from 2022 to 2023 (2-3 years post-restoration), and all sites except one low-relief oyster reef site saw elevation loss. Elevation loss was highest at the control site, but marsh edge retreat was highest at one of the engineered oyster reefs. Discussion While horizontal thresholds did not yield statistically different results, vertical thresholds did. Our results support using a 95% confidence interval for conservative volumetric estimates and recommend that future studies consider aligning uncertainty thresholds with monitoring goals and timelines.