Energy-use strategies on a degraded coral reef: insights from remote underwater stereo-video and AI-driven 3D tracking

Abstract Unveiling the intricate relationships between animal movement ecology, feeding behaviour, and internal energy budgeting is essential for a comprehensive understanding of ecosystem functioning. Particularly in degraded habitats like coral reefs, which are under significant anthropogenic stress, these interactions are crucial yet largely unexplored. Employing advanced methods such as underwater stereo-video measurements, AI-driven object recognition, and 3D tracking, our study focuses on the feeding preferences, bite rates, inter-bite distances, and foraging energy expenditure (EE) of two dominant reef grazers, the Brown surgeonfish ( Acanthurus nigrofuscus ) and the Yellowtail tang ( Zebrasoma xanthurum ). Our findings reveal significant variations in functional feeding traits and EE, showcasing how metabolic traits shape individual and community foraging behaviours in degraded habitats. On a microhabitat scale, A. nigrofuscus adopted a specialized feeding strategy with higher energy demands, while Z. xanthurum exhibited a more generalist approach, covering larger distances between food patches but with lower EE. These divergent energy-use strategies indicate key metabolic adaptations, allowing both species to co-exist and maintain high feeding pressures in this degraded ecosystem. The study underscores the transformative potential of technologies like stereo-video and AI-generated 3D tracking in elucidating functional ecology and energy dynamics. Our insights not only contribute to targeted restoration interventions but also accentuate the pivotal role of metabolic traits as possible indicators for community responses to environmental changes. This understanding is crucial in formulating effective conservation strategies for ecosystems globally that are impacted by human disturbances and climate change.