Vegetation and Habitat Classification of Created and Natural Brackish Marshes via Unoccupied Aerial Systems and Potential Monitoring Applications

Author

Seth T. Chapman, Louisiana State University at Baton RougeFollow

Semester of Graduation

Fall 2023

Degree

Master of Oceanography and Coastal Sciences (SOCS)

Department

Department of Oceanography and Coastal Sciences

Document Type

Thesis

Abstract

Louisiana’s coastal marshes are some of the most valuable natural, economic, and cultural state resources that however, have been severely deteriorating over the last century. Federal and state agencies have pooled billions of dollars into coastal restoration projects in Louisiana of which marsh creation is a primary focus. Traditional monitoring of created marshes involves in-situ vegetation sampling and aerial imagery from fixed-wing occupied aircraft, but these methods have limitations and/or can negatively impact sites. My thesis explores the use of Unoccupied Aerial Systems (UAS; i.e., drones) to classify vegetation and habitat in coastal marsh systems, compare habitat and vegetation cover between created and natural marsh sites, and evaluate the ability of in-situ, plot-based sampling to characterize site-wide vegetation community cover. To achieve these objectives, I surveyed created and nearby natural, reference marshes along the eastern side of Barataria Bay, Louisiana. Results indicate that high-resolution UAS-based imagery can accurately classify natural and created brackish marshes in coastal Louisiana. It was observed that marshes created as part of the Lake Hermitage Marsh Creation Project had more heterogeneous vegetation cover relative to reference marshes. In addition, results indicate that the current number of in-situ vegetation plots employed at these sites may be insufficient to reliably provide an accurate representation of site-wide vegetation cover. However, given that the number of plots required to fully characterize site-wide vegetation cover is likely impractical it is recommended that the analyses of UAS-based imagery should be included as a non-invasive, supplemental tool to complement and inform plot-based, in-situ vegetation sampling in future post-construction marsh monitoring plans.

Date

11-14-2023

Recommended Citation

Chapman, Seth T., "Vegetation and Habitat Classification of Created and Natural Brackish Marshes via Unoccupied Aerial Systems and Potential Monitoring Applications" (2023). LSU Master's Theses. 5855.
https://repository.lsu.edu/gradschool_theses/5855

Committee Chair

Polito, Michael J.