Leveraging Automated Image Analysis Tools to Transform Our Capacity to Assess Status and Trends of Coral Reefs

Digital photography is widely used by coral reef monitoring programs to assess benthic status and trends. In addition to creating a permanent archive, photographic surveys can be rapidly conducted, which is important in environments where bottom-time is frequently limiting. However, substantial effort is required to manually analyze benthic images; which is expensive and leads to lags before data are available. Using previously analyzed imagery from NOAA’s Pacific Reef Assessment and Monitoring Program, we assessed the capacity of a trained and widely used machine-learning image analysis tool – CoralNet (coralnet.ucsd.edu) - to generate fully-automated benthic cover estimates for the main Hawaiian Islands (MHI) and American Samoa. CoralNet was able to generate estimates of site-level coral cover for both regions that were highly comparable to those generated by human analysts (Pearson’s r > 0.97, and with bias of 1% or less). CoralNet was generally effective at estimating cover of common coral genera (Pearson’s r > 0.92 and with bias of 2% or less in 6 of 7 cases), but performance was mixed for other groups including algal categories, although generally better for American Samoa than MHI. CoralNet performance was improved by simplifying the classification scheme from genus to functional group and by training within habitat types, i.e., separately for coral-rich, pavement, boulder, or ‘other’ habitats. The close match between human-generated and CoralNet-generated estimates of coral cover pooled to the scale of island and year demonstrates that CoralNet is capable of generating data suitable for assessing spatial and temporal patterns. The imagery we used was gathered from sites randomly located in <30 m hard-bottom at multiple islands and habitat-types per region, suggesting our results are likely to be widely applicable. As image acquisition is relatively straightforward, the capacity of fully-automated image analysis tools to minimize the need for resource intensive human analysts opens possibilities for enormous increases in the quantity and consistency of coral reef benthic data that could become available to researchers and managers.

[1]  E. Edinger,et al.  Reef degradation and coral biodiversity in indonesia: Effects of land-based pollution, destructive fishing practices and changes over time , 1998 .

[2]  A. DesRochers,et al.  Long-term monitoring of coral reef fish assemblages in the Western central pacific , 2017, Scientific Data.

[3]  Tim W. Nattkemper,et al.  BIIGLE 2.0 - Browsing and Annotating Large Marine Image Collections , 2017, Front. Mar. Sci..

[4]  Rafael García,et al.  Image-Based Coral Reef Classification and Thematic Mapping , 2013, Remote. Sens..

[5]  Grant B. Deane,et al.  Automated processing of coral reef benthic images , 2009 .

[6]  Oscar Pizarro,et al.  The Catlin Seaview Survey – kilometre‐scale seascape assessment, and monitoring of coral reef ecosystems , 2014 .

[7]  M. González-Rivero,et al.  An evaluation of semi‐automated methods for collecting ecosystem‐level data in temperate marine systems , 2017, Ecology and evolution.

[8]  Stefan B. Williams,et al.  Autonomous underwater vehicle–assisted surveying of drowned reefs on the shelf edge of the Great Barrier Reef, Australia , 2010, J. Field Robotics.

[9]  P. Marshall,et al.  Building resilience into practical conservation: identifying local management responses to global climate change in the southern Great Barrier Reef , 2010, Coral Reefs.

[10]  D. Altman,et al.  STATISTICAL METHODS FOR ASSESSING AGREEMENT BETWEEN TWO METHODS OF CLINICAL MEASUREMENT , 1986, The Lancet.

[11]  G. De’ath,et al.  The 27–year decline of coral cover on the Great Barrier Reef and its causes , 2012, Proceedings of the National Academy of Sciences.

[12]  K. Fabricius.  Effects of terrestrial runoff on the ecology of corals and coral reefs: review and synthesis. , 2005, Marine pollution bulletin.

[13]  M. Dethier.  Tidepools as refuges: Predation and the limits of the harpacticoid copepod Tigriopus californicus (Baker) , 1980 .

[14]  P. Glynn.  Some Physical and Biological Determinants of Coral Community Structure in the Eastern Pacific , 1976 .

[15]  A. Beckett,et al.  AKUFO AND IBARAPA. , 1965, Lancet.

[16]  Brett Schumacher,et al.  Analysis of benthic survey images via CoralNet : a summary of standard operating procedures and guidelines. , 2017 .

[17]  Ryan J. Lowe,et al.  Spatial and temporal patterns of mass bleaching of corals in the Anthropocene , 2018, Science.

[18]  Stuart R. Phinn,et al.  Scaling up Ecological Measurements of Coral Reefs Using Semi-Automated Field Image Collection and Analysis , 2016, Remote. Sens..

[19]  S. Dollar.  Wave stress and coral community structure in Hawaii , 1982, Coral Reefs.

[20]  T. Hughes,et al.  Global warming transforms coral reef assemblages , 2018, Nature.

[21]  D. Delparte,et al.  Distributed under Creative Commons Cc-by 4.0 Integrating Structure-from-motion Photogrammetry with Geospatial Software as a Novel Technique for Quantifying 3d Ecological Characteristics of Coral Reefs , 2022 .

[22]  Maricor Soriano,et al.  Classification of coral reef images from underwater video using neural networks. , 2005, Optics express.

[23]  A. Heenan,et al.  Coral Reef Ecosystem Division standard operating procedures data collection for rapid ecological assessment fish surveys , 2011 .

[24]  Chris Roelfsema,et al.  Towards Automated Annotation of Benthic Survey Images: Variability of Human Experts and Operational Modes of Automation , 2015, PloS one.

[25]  J. H. Burns,et al.  Mass coral bleaching due to unprecedented marine heatwave in Papahānaumokuākea Marine National Monument (Northwestern Hawaiian Islands) , 2017, PloS one.

[26]  Gregory Dudek,et al.  Robotic Coral Reef Health Assessment Using Automated Image Analysis , 2017, J. Field Robotics.

[27]  A. Hedayat,et al.  Statistical Methods in Assessing Agreement , 2002 .

[28]  M. Patterson,et al.  Coral disease following massive bleaching in 2005 causes 60% decline in coral cover on reefs in the US Virgin Islands , 2009, Coral Reefs.

[29]  Scott F. Heron,et al.  Warming Trends and Bleaching Stress of the World’s Coral Reefs 1985–2012 , 2016, Scientific Reports.

[30]  Key herbivores reveal limited functional redundancy on inshore coral reefs , 2013, Coral Reefs.

[31]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[32]  Vasilis Trygonis,et al.  photoQuad: A dedicated seabed image processing software, and a comparative error analysis of four photoquadrat methods , 2012 .

[33]  Chris P. Tsokos,et al.  Detection of Coral Reef Change by the Florida Keys Coral Reef Monitoring Project , 2001 .

[34]  C. Roman,et al.  Characterizing the deep insular shelf coral reef habitat of the Hind Bank marine conservation district (US Virgin Islands) using the Seabed autonomous underwater vehicle , 2006 .

[35]  Scott F. Heron,et al.  Prioritizing Key Resilience Indicators to Support Coral Reef Management in a Changing Climate , 2012, PloS one.

[36]  Jennifer M. Durden,et al.  Integrating “Big Data” into Aquatic Ecology: Challenges and Opportunities , 2017 .

[37]  C. Couch,et al.  Ecosystem Sciences Division Standard Operating Procedures: Data Collection for Rapid Ecological Assessment Benthic Surveys, 2018 Update , 2018 .

[38]  Guy Cox,et al.  Fluorescent pigments in corals are photoprotective , 2000, Nature.

[39]  Arthur C. R. Gleason,et al.  Large-area imaging reveals biologically driven non-random spatial patterns of corals at a remote reef , 2017, Coral Reefs.

[40]  E. Fisher,et al.  The Coral Reef , 1964 .