Multispectral Acoustic Backscatter: How Useful Is it for Marine Habitat Mapping and Management?

Marine managers routinely use benthic habitat maps to make decisions about the ocean, its resources, and associated human uses. Acoustic backscatter from multibeam echosounders (MBES) is often critical for developing these habitat maps. Recent advances now allow MBES to collect backscatter at multiple acoustic frequencies. This type of data may help researchers more accurately map benthic habitats and managers more confidently make decisions. However, new research is needed quantifying how much multispectral backscatter improves the habitat characterization process and identifying which management needs would benefit most from its collection. To begin answering these questions, a case study was conducted opportunistically in Bedford Basin, Canada, with MBES bathymetry and backscatter collected at 100, 200, and 400-kHz frequencies. Underwater photos and boosted regression trees were used to characterize seven dominant benthic habitats and calculate the relative importance of multispectral backscatter to the characterization process. Response curves were generated to identify key relative frequency thresholds for differentiating among habitat types. These results indicated that multispectral backscatter can enhance the discrimination of soft bottom by 17.4%, hard bottom by 5.7%, and all habitats by 9.1%. Topographic information (e.g., depth) contributed the most to the hard-bottom maps (51.8% +/- 4.0), whereas multispectral backscatter contributed the most to the soft-bottom map (46.9%). The 100-kHz frequency was the most important frequency for all habitat types. These findings suggest that multispectral backscatter maybe most useful to management applications focused on soft-bottom habitats. Single-frequency (i.e. 100 kHz) backscatter may be adequate for applications focused on hard bottom, because it only improved the models by a small amount. Researchers and marine managers can start to use this information to decide a priori which backscatter frequency (or frequencies) are best suited to support their research objectives, mapping needs, and management actions.