High-Resolution Vertical Terrain of a Deep-Sea Cliff offshore Greenland and the Associated Epibenthic Fauna
Fuente:
PANGAEA
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Dataset Open Access |
| Language: | en |
| Published: |
PANGAEA
2021
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1867168574023401472 |
|---|---|
| author | Van Audenhaege, Loïc Broad, Emmeline Hendry, Katharine R Huvenne, Veerle A I |
| author_facet | Van Audenhaege, Loïc Broad, Emmeline Hendry, Katharine R Huvenne, Veerle A I |
| collection | Datos científicos de ciencias marinas y ambientales |
| contents | We used a multibeam echosounder (Reson7125) front-mounted onto the ROV Isis (Dive D333, DY081 expedition) to map the terrain of a vertical feature marking the edge of a deep-sea glacial trough (Labrador Sea, [63°51.9'N, 53°16.9'W, depth: 650 to 800 m]). After correction of the ROV navigation (i.e. merging of USBL and DVL), bathymetry [m] and backscatter [nominal unit] were extracted at a resolution of 0.3 m and different terrain descriptors were computed: Slope, Bathymetric Position Index (BPI), Terrain Ruggedness Index, Roughness, Mean and Gaussian curvatures and orientations (Northness and Eastness), at scales of 0.9, 3 and 9 m. Using a Principal Component Analysis (PCA), the terrain descriptors enabled to retrieve 4 terrain clusters and their associated confusion index, to investigate the spatial heterogeneity of the terrain. This approach also underlined the presence of geomorphic features in the wall terrain. The extraction of the backscatter intensity for the first time considering vertical terrains, opens space for further acquisition and processing development. Using photographs collected by the ROV Isis (Dive D334, DY081 expedition), epibenthic fauna was annotated. Each image was linked to a terrain cluster in the 3D space and pooled into 20-m² bins of images. A Bray-Curtis dissimilarity matrix was constructed from morphospecies abundances. This enabled to test for differences of assemblage composition among clusters. Few species appeared more abundant in particular clusters such as L. pertusa in high-roughness cluster. However, nMDS suggested differences in assemblage composition but these dissimilarities were not strongly delineated. Whereas the design of this study may have limited distinctive differences among assemblages, this shows the potential of this cost-effective method of top-down habitat mapping to be applied in undersampled benthic habitat in order to provide a priori knwoledge for defining appropriate sampling design. |
| format | Dataset Open Access |
| id | pangaea_https___doi_org_10_1594_PANGAEA_931687 |
| institution | PANGAEA |
| language | en |
| publishDate | 2021 |
| publisher | PANGAEA |
| record_format | pangaea |
| spellingShingle | High-Resolution Vertical Terrain of a Deep-Sea Cliff offshore Greenland and the Associated Epibenthic Fauna Van Audenhaege, Loïc Broad, Emmeline Hendry, Katharine R Huvenne, Veerle A I acoustic data; CLASS; Climate Linked Atlantic Sector Science; deep-water vertical cliff; DY081; fine-scale spatial patterns; Greenland glacial trough; High-resolution terrain; iAtlantic; ICY-LAB; Integrated Assessment of Atlantic Marine Ecosystems in Space and Time; Isotope CYcling in the LABrador Sea; Labrador Sea; marine habitat mapping; Multibeam Echosounder; ROV; suspension-feeding community; terrain point cloud; top-down habitat mapping; underwater exploration; unsupervised terrain clustering; Vertical mapping We used a multibeam echosounder (Reson7125) front-mounted onto the ROV Isis (Dive D333, DY081 expedition) to map the terrain of a vertical feature marking the edge of a deep-sea glacial trough (Labrador Sea, [63°51.9'N, 53°16.9'W, depth: 650 to 800 m]). After correction of the ROV navigation (i.e. merging of USBL and DVL), bathymetry [m] and backscatter [nominal unit] were extracted at a resolution of 0.3 m and different terrain descriptors were computed: Slope, Bathymetric Position Index (BPI), Terrain Ruggedness Index, Roughness, Mean and Gaussian curvatures and orientations (Northness and Eastness), at scales of 0.9, 3 and 9 m. Using a Principal Component Analysis (PCA), the terrain descriptors enabled to retrieve 4 terrain clusters and their associated confusion index, to investigate the spatial heterogeneity of the terrain. This approach also underlined the presence of geomorphic features in the wall terrain. The extraction of the backscatter intensity for the first time considering vertical terrains, opens space for further acquisition and processing development. Using photographs collected by the ROV Isis (Dive D334, DY081 expedition), epibenthic fauna was annotated. Each image was linked to a terrain cluster in the 3D space and pooled into 20-m² bins of images. A Bray-Curtis dissimilarity matrix was constructed from morphospecies abundances. This enabled to test for differences of assemblage composition among clusters. Few species appeared more abundant in particular clusters such as L. pertusa in high-roughness cluster. However, nMDS suggested differences in assemblage composition but these dissimilarities were not strongly delineated. Whereas the design of this study may have limited distinctive differences among assemblages, this shows the potential of this cost-effective method of top-down habitat mapping to be applied in undersampled benthic habitat in order to provide a priori knwoledge for defining appropriate sampling design. |
| title | High-Resolution Vertical Terrain of a Deep-Sea Cliff offshore Greenland and the Associated Epibenthic Fauna |
| topic | acoustic data; CLASS; Climate Linked Atlantic Sector Science; deep-water vertical cliff; DY081; fine-scale spatial patterns; Greenland glacial trough; High-resolution terrain; iAtlantic; ICY-LAB; Integrated Assessment of Atlantic Marine Ecosystems in Space and Time; Isotope CYcling in the LABrador Sea; Labrador Sea; marine habitat mapping; Multibeam Echosounder; ROV; suspension-feeding community; terrain point cloud; top-down habitat mapping; underwater exploration; unsupervised terrain clustering; Vertical mapping |
| url | https://doi.org/10.1594/PANGAEA.931687 |