Investigating the spatial distribution of growth anomalies affecting Montipora capitata corals in a 3-dimensional framework

Abstract

Diseases have caused significant reductions in coral populations throughout the global ocean. Despite a substantial effort to thoroughly characterize the epizootiology and etiology of coral diseases, little is known about the distribution and spatial clustering of disease lesions on affected coral colonies. This study investigated spatial clustering of the coral disease, growth anomaly (GA), which exhibits high levels of prevalence and severity in Montipora capitata and other corals at Wai'ōpae, southeast Hawai'i Island. Like many other coral diseases, the patterns of disease spread and transmissibility of GA remains unknown. We utilized cutting-edge 3D reconstruction techniques to map the precise spatial distribution of GAs on affected coral colonies. Three statistical measures, Ripley's K, Moran's I, and the Kolmogorov-Smirnov test were used to determine if the GA lesions were distributed in a non-random pattern. Each measure showed the GA lesions exhibited distinct spatial clustering on all ten affected colonies analyzed in this study. Our study is not only the first 3D analysis of intra-colony disease clustering, but also provides a novel approach for investigating and quantifying levels of disease clustering in order to improve our understanding of coral disease epizootiology, transmission, and etiology.

Photogrammetry as a tool to improve ecosystem restoration

“…3D models allow quantification of even small amounts of growth over short periods of time [36], facilitating the assessment of early progress towards the desired trajectory and enabling adaptive management if necessary. Photogrammetry also enables novel metrics which can capture more subtle changes in growth and fitness, but also alterations to coral structure that may indicate unexpected outcomes of phenotype-environment interactions [36,37] (Figure 1). Some of the most exciting metrics relate to coral fitness (e.g., accretion, surface area, and volume) [15,38,39] and habitat function (e.g., fractal dimension, surface curvature, branch packing, interstitial space, field of view, and object distribution) [40][41][42][43].…”

Data for spatial analysis of growth anomaly lesions on Montipora capitata coral colonies using 3D reconstruction techniques

“…Each table is associated with an accompanying 3D VRML model and.jpg image of the annotated 3D model showing all lesion locations. The tables can be integrated with the VRML models in Matlab in order to spatially analyze the distribution of the GA lesions [1] . Matlab scripts are provided in order to conduct Ripley׳s K, Moran׳s I, and the Kolmogorov–Smirnov test in a 3D framework.…”

“…Burns, T. Alexandrov, E. Ovchinnikova, R.D. Gates, M. Takabayashi, 2016) [1] , for further interpretation and discussion of the data.…”

Automated classification of three-dimensional reconstructions of coral reefs using convolutional neural networks

“…3D reconstructions of reefs can be generated from images using Simultaneous Localization and Mapping (SLAM) or Structure-from-Motion (SfM) techniques [26,27] and employed to assess structural complexity in a more comprehensive fashion [28][29][30][31]. 3D reconstructions have also been used to provide better insight into spatial clustering of species on reefs [32], the effects of disturbances on reef complexity and community structure [33], and to study disease prevalence and spatial distribution [34].…”