Mesophotic coral reef community structure: the constraints of imagery collected by unmanned vehicles

There is a critical need to quantify and monitor mesophotic coral reef community structure and function at multiple spatial and temporal scales. Because accessing these habitats is costly in terms of infrastructure and effort, often for a modest return in data, many investigators collect digital imagery using transect techniques from unmanned platforms. Specifically, remotely operated vehicles and autonomous underwater vehicles are used because they operate at deeper depths for extensive periods of time, can carry an array of oceanographic and imaging instruments, and can collect and archive extensive amounts of video and still imagery. However, substrate angle, camera angle, and vehicle position above the benthos creates varying degrees of error in the imagery due to parallax and geometric distortion. Photogrammetry conducted on 2D photographs from uncorrected 3D imagery can over- or under-estimate the percent cover, biomass estimates, and abundance of the benthic groups of interest. Here we illustrate these errors and emphasize the requirement for post-processing of imagery to ensure that these data can be used for valid quantitative ecological descriptions of mesophotic benthic communities in the future.

Benthic and coral reef community field data for Heron Reef, Southern Great Barrier Reef, Australia, 2002–2018

This paper describes benthic coral reef community composition point-based field data sets derived from georeferenced photoquadrats using machine learning. Annually over a 17 year period (2002–2018), data were collected using downward-looking photoquadrats that capture an approximately 1 m2 footprint along 100 m–1500 m transect surveys distributed along the reef slope and across the reef flat of Heron Reef (28 km2), Southern Great Barrier Reef, Australia. Benthic community composition for the photoquadrats was automatically interpreted through deep learning, following initial manual calibration of the algorithm. The resulting data sets support understanding of coral reef biology, ecology, mapping and dynamics. Similar methods to derive the benthic data have been published for seagrass habitats, however here we have adapted the methods for application to coral reef habitats, with the integration of automatic photoquadrat analysis. The approach presented is globally applicable for various submerged and benthic community ecological applications, and provides the basis for further studies at this site, regional to global comparative studies, and for the design of similar monitoring programs elsewhere.