Quantifying the Loss of Coral from a Bleaching Event Using Underwater Photogrammetry and AI-Assisted Image Segmentation

Abstract

Detecting the impacts of natural and anthropogenic disturbances that cause declines in organisms or changes in community composition has long been a focus of ecology. However, a tradeoff often exists between the spatial extent over which relevant data can be collected, and the resolution of those data. Recent advances in underwater photogrammetry, as well as computer vision and machine learning tools that employ artificial intelligence (AI), offer potential solutions with which to resolve this tradeoff. Here, we coupled a rigorous photogrammetric survey method with novel AI-assisted image segmentation software in order to quantify the impact of a coral bleaching event on a tropical reef, both at an ecologically meaningful spatial scale and with high spatial resolution. In addition to outlining our workflow, we highlight three key results: (1) dramatic changes in the three-dimensional surface areas of live and dead coral, as well as the ratio of live to dead colonies before and after bleaching; (2) a size-dependent pattern of mortality in bleached corals, where the largest corals were disproportionately affected, and (3) a significantly greater decline in the surface area of live coral, as revealed by our approximation of the 3D shape compared to the more standard planar area (2D) approach. The technique of photogrammetry allows us to turn 2D images into approximate 3D models in a flexible and efficient way. Increasing the resolution, accuracy, spatial extent, and efficiency with which we can quantify effects of disturbances will improve our ability to understand the ecological consequences that cascade from small to large scales, as well as allow more informed decisions to be made regarding the mitigation of undesired impacts.

Corals that survive repeated thermal stress show signs of selection and acclimatization

“…In Pocillopora , size-dependent patterns of bleaching and survivorship reversed over time, with more bleaching and mortality in larger genets in 2015, followed by more bleaching and mortality in smaller genets in 2019 (S5 Fig in S1 File ). While we would expect to see the highest mortality in smaller size classes in Pocillopora [ 61 , 108 ], higher mortality in large Pocillopora colonies in response to bleaching has been observed before [ 93 , 109 ]. In Moorea, the coexistence of multiple cryptic species of Pocillopora led to the appearance of a positive relationship between colony size and mortality after a bleaching event [ 114 ].…”

“…However, these corals can persist via regrowth of tissue that survived bleaching in shaded or cryptic regions of the colony [ 60 ]. Due to these complexities, most studies tracking coral growth and survivorship focus on taxa with identifiable, discrete colonies that display little fission or fusion, such as Pocillopora [ 108 , 109 ]. Our results demonstrate that such approaches may be inadequate for describing demographic and recovery dynamics in locations like Hawai‘i, where fusion and fission predominate.…”

Potentialities of the combined use of underwater fluorescence imagery and photogrammetry for the detection of fine-scale changes in marine bioconstructors

“…On the other hand, the 3D models provided detailed corallite imaging, with fine definition at a polyp scale suitable for morphological observations and change detection analysis. Highquality texture mapping in the digital mesh allowed not only the identification of corallites, Kopecky et al (2023) also estimates living tissues, epiphyte colonization, and bleaching events thanks to the implementation of color calibration procedures. Overall, the developed measuring system is easier to use, and more effective in saving time and providing detailed corallite imaging for change detection analyses than 2D photos.…”

“…Due to modern technological advances in computer vision, change-detection techniques for investigating the health of corals permit the analyses of live cover, counting species presence/absence and decay, and the assessment of benthic assemblages (Gutierrez-Heredia et al, 2016;Kersting et al, 2017;Kopecky et al, 2023). The photogrammetric approach allows non-intrusive measurements from digitalized 3D models of organisms, relying on an array of two-dimensional camera images taken from different points of view (Figueira et al, 2015).…”

“…During the last few years, this technique has been largely employed to map marine habitats on a large scale and study benthic assemblages (Marre et al, 2019). Close-range photogrammetry has been successfully utilized to obtain measures of coral colony dimensions, surface area and 3D complexity (Courtney et al, 2007;Burns et al, 2015;Lavy et al, 2015;Guo et al, 2016;Gutierrez-Heredia et al, 2016;Ferrari et al, 2017) and is widely adopted in marine investigations from habitat scale (Bongaerts et al, 2021) to individual phenotyping (Million et al, 2021;Kopecky et al, 2023). These tools provide accurate measurements of lengths, areas, and volumes that are difficult or even impossible to get in situ with traditional methods (Bayley et al, 2019).…”