Acoustic monitoring of Undaria pinnatifida forests: first 2 approach using broadband echosounders in Patagonia 3 Argentina
Undaria pinnatifida (hereafter Undaria) was first recorded in the Southwest Atlantic in 1992 in Golfo Nuevo (GN), Argentina. Since its introduction, in Golfo Nuevo, Argentina, in 1992, Undaria pinnatifida has become a species of significant ecological and economic importance, prompting interest in m...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2025 |
| País: | España |
| Institución: | Universidad Nacional de Educación a Distancia |
| Repositorio: | e-spacio. Repositorio Institucional de la UNED |
| Idioma: | inglés |
| OAI Identifier: | oai:e-spacio.uned.es:20.500.14468/31513 |
| Acceso en línea: | https://hdl.handle.net/20.500.14468/31513 |
| Access Level: | acceso embargado |
| Palabra clave: | 1209 Estadística Random forest Classification trees Undaria pinnatifida Acoustic data Broadband signal |
| Sumario: | Undaria pinnatifida (hereafter Undaria) was first recorded in the Southwest Atlantic in 1992 in Golfo Nuevo (GN), Argentina. Since its introduction, in Golfo Nuevo, Argentina, in 1992, Undaria pinnatifida has become a species of significant ecological and economic importance, prompting interest in monitoring its spread and exploitation. This has generated significant interest in assessing its exploitation and monitoring its spread. Acoustic methods have been widely used for underwater vegetation mapping; however our study is the first in this area and the first to use broadband technology in this region. The aim of this study is to assess, for the first time, the performance of a broadband echosounder in detecting Undaria using both a method based on the height and acoustic intensity of Undaria fronds above the seafloor, and a method based on the acoustic signatures across multiple frequencies for target characterization. Supervised classifications were made in both cases using ground-truth data from diver surveys: using random forest and classification trees. The accuracies obtained for above-bottom features were higher than 75 % (kappa = 0.6); and for multi spectral response were even higher (close to 95 %, kappa = 0.89). These results prove broadband echosounders are valuable tools for underwater vegetation detection and promising for benthic habitat classification. |
|---|