Utilizing Repeat UAV Imagery to Evaluate the Spatiotemporal Patterns and Environmental Drivers of Wrack in a Coastal Georgia Salt Marsh

Wrack, comprised of dead marsh grass, occurs naturally in salt marshes. Wrack can reduce biomass in underlying vegetation and affect salt marsh function. Unmanned aerial vehicles (UAV) provide a more efficient and cost-effective method than traditional field sampling for characterizing the distribution of wrack at a fine spatial scale. We used a DJI Matrice 210 UAV with a MicaSense Altum to collect a total of 20 images from January 2020-December 2021 in a salt marsh on Sapelo Island, GA. Wrack was classified using principal component analysis. Classified images were then used to characterize the size-frequency distribution, landscape position, and potential environmental drivers of wrack. We observed similar to 2100 wrack patches over the course of the study, most of which were present for only a single month. Wrack was found most frequently at the mean higher high water line (similar to 1 m), although the areas with the highest frequency of wrack as a proportion of available marsh area were at a higher elevation (> 1.3 m) and closer to creeks or shorelines (similar to 40-50 m). High tide events were found to decrease the distance to water of wrack and increase the standard deviation of wrack elevation. This study provides a methodology for understanding wrack dynamics at a landscape scale using frequent, high-resolution UAV data.

Automated summary

This study utilized UAV technology to collect images from January 2020 to December 2021 and used classification analysis to characterize the distribution and environmental drivers of wrack. The study found that wrack patches were mainly located near the mean higher high water line and were influenced by tidal events.