Combining soundscape analysis with in situ observations and oceanographic data for future ecosystem evaluation techniques.

Coral reef ecosystems face many anthropogenic threats. There are urgent requirements for improved monitoring and management. Conventional assessment methods using SCUBA are costly and prone to bias and under-sampling. Here, three approaches to understanding coral reef ecology are combined to aid the goal of enhanced passive monitoring in the future: statistical analysis of oceanographic habitats, remote cameras for nocturnal surveys of benthic fauna, and soundscape analysis in the context of oceanographic setting and ecological metrics collected in-situ. Hawaiian reefs from Kure Atoll to the island of Hawaii, an area spanning two oceanographic habitats, are assessed. Multivariate analysis of acoustic, remote camera, and in-situ observational data showed significant differences in more than 20 percent of ecological and acoustic variables when grouped by oceanic regime, suggesting that large-scale oceanography substantially influences local ecological states and associated soundscapes. Acoustic variables further delineated sites by island, suggesting local conditions influence the soundscape to a greater degree. While the number of invertebrates (with an emphasis on crustaceans and echinoderms) imaged using remote cameras correlated with a number of acoustic metrics, an increasingly higher correlation between invertebrate density and spectral level was observed as acoustic bands increased in frequency from 2 to 20 kHz. In turn, correlation was also observed between the number of predatory fish and sound levels above 2 kHz, suggesting a connection between the number of invertebrates, sound levels at higher frequencies, and the presence of their predators. Comparisons between sound recordings and diversity indices calculated from observational and remote camera data indicate that greater diversity in fishes and benthic invertebrates is associated with a larger change in sound levels between day and night. Interdisciplinary analyses provide a novel view to underwater ecology, and can reveal new quantitative metrics that may be more efficiently sampled. These techniques may be used to detect subtle yet important shifts in ecosystem function, critical for effective marine resource management in the face of environmental changes that occur over multi-year timescales.