Analysis of calibrated seafloor backscatter for habitat classification methodology and case study of 158 spots in the Bay of Biscay and Celtic Sea

Type Article
Date 2018-06
Language English
Author(s) Fezzani Ridha1, Berger LaurentORCID1
Affiliation(s) 1 : IFREMER, Underwater Acoust Lab IMN NSE ASTI, Plouzane, France.
Source Marine Geophysical Research (0025-3235) (Springer), 2018-06 , Vol. 39 , N. 1-2 , P. 169-181
DOI 10.1007/s11001-018-9342-y
WOS© Times Cited 13
Keyword(s) Multibeam, Reflectivity, Backscatter angular response, Segmentation

An automated signal-based method was developed in order to analyse the seafloor backscatter data logged by calibrated multibeam echosounder. The processing consists first in the clustering of each survey sub-area into a small number of homogeneous sediment types, based on the backscatter average level at one or several incidence angles. Second, it uses their local average angular response to extract discriminant descriptors, obtained by fitting the field data to the Generic Seafloor Acoustic Backscatter parametric model. Third, the descriptors are used for seafloor type classification. The method was tested on the multi-year data recorded by a calibrated 90-kHz Simrad ME70 multibeam sonar operated in the Bay of Biscay, France and Celtic Sea, Ireland. It was applied for seafloor-type classification into 12 classes, to a dataset of 158 spots surveyed for demersal and benthic fauna study and monitoring. Qualitative analyses and classified clusters using extracted parameters show a good discriminatory potential, indicating the robustness of this approach.

Full Text
File Pages Size Access
13 5 MB Access on demand
Author's final draft 19 3 MB Open access
Top of the page