Authors: Amy Li*, Harvard University
Topics: Coastal and Marine, UAS / UAV, Pacific Islands
Keywords: coral, hawaii, drone, coral bleaching, classification
Session Type: Poster
Presentation File: No File Uploaded
The breadth of remote sensing technology applications in studying coral reefs still has not been fully explored. To better understand the realm of possibilities of this suite of technologies, we used imagery of Kāne‘ohe Bay in northeastern O‘ahu, Hawai‘i collected from an unmanned aerial system (Firefly6 Pro) and an airborne remote sensing system to preliminarily assess coral reef health following the massive 2015 statewide bleaching event. Coral bleaching events, exacerbated by increasing ocean temperatures, are a growing threat to coral reef survival and biodiversity. By comparing two sets of ultra-high resolution (2 cm) true-color imagery of the same shallow reef ecosystem from 2015 and 2018, we could quantify the state of coral bleaching and evaluate subsequent recovery using image classifiers. We identified bleached coral colonies using supervised and unsupervised image classification techniques, as well as a series of nested classifications. This resulted in findings comparable to bay-wide in-water transect surveys of coral reefs conducted following the 2015 bleaching event. Challenges and areas of improvement for collecting, processing, and analyzing high resolution UAV imagery were also discovered. It serves as a proof-of-concept for applications of remotely sensed imagery to study and respond to future coral bleaching events in the Hawaiian Islands and shallow reef systems around the world.
To access contact information login