Advancing Seabed Mapping Techniques for Deep-water Habitat Classification in Submarine Canyons (ASMaT)
Cold-water corals represent an important share towards deep-sea biodiversity due to their three-dimensional structure and reef building capacity. Submarine canyons act as conduits for sediments, nutrients and organic matter supporting high biomass communities. There is an increasing demand for new methods able to efficiently capture fine-scale changes in these environments. To date, scientists have used the rendering of 2D images from 3D objects to analyse seabed features. However, advancing from commonly employed 2D image analysis techniques to 3D methods can provide more realistic representations of coral reefs and submarine environments. This project focuses on the development of novel mapping techniques using ROV data, 3D photogrammetry and computer vision to allow us to understand, in detail, CWC habitats.
The aims of this project are to 1) Advance seabed mapping techniques for habitat classification in submarine canyons and cold-water coral habitats 2) Develop a novel methodology to create an efficient 3D analysis workflow for deep-water habitats 3) Determine the appropriate scale for seabed habitat characterisation. Our goals are to find a robust methodological framework that provides high-resolution information from 3D and multiscale seabed mapping to help us explore, understand, and manage key segments of our largely unexplored seabed.
This project funded in conjunction with the Science Foundation of Ireland (SFI) through the project MMMonKey_Pro and Irish Research Council (IRC) through the Government of Ireland Postgraduate Scholarship 2020 program.
Larissa Oliveira (PhD Researcher)
Dr. Aaron Lim (Supervisor)
Prof. Andy Wheeler (Supervisor)
Prof. Luis Conti (Supervisor)
David Mark Dalton “Semantic coral reef mapping through image annotation and analysis: An example from cold-water corals in the Northeast Atlantic” – Masters in Bioinformatics (UCC; CO-supervised by Aaron Lim, Andy Wheeler and Larissa Oliveira)