High resolution optical remote sensing of tree damage in South Africa
Category: Research Poster
Author(s): Miranda Chin, Gloria Tait
Presenter(s): Miranda Chin
Mentors(s): Jody Vogeler, Steven Filippelli
Elephants act as ecosystem engineers that can cause large impacts on their surroundings through browsing and toppling woody vegetation. In Thornybush Reserve, South Africa, elephant populations recently increased, causing a subsequent rise in these behaviors. However, methods to track and identify elephant-caused tree damage are still under-developed. This leads into the question of if it is possible to use remote sensing imagery and spectral indices to detect tree damage status. Methods consisted of tree digitizing from high resolution satellite imagery, in addition to ground data collected in 2023-2024 detailing tree location and damage. For analysis, 4-meter PlanetScope and 10-meter Sentinel imagery were compared to determine if there were trends in surface reflectance corresponding to tree damage. Results show that there is a small correlation of spectral bands and indices, specifically near-infrared, with tree damage status, but an ANOVA did not indicate significant differences in spectral values between tree damage classes. Implications include that more research is needed to determine how and when remote sensing imagery can identify tree damage. Some recommendations include more detailed classification for damage status, improved geospatial registration, and research into how remote sensing timing affects damage detectability.