Developing quantitative and objective approaches to integrate multibeam echosounder (MBES) data with ground observations for predictive modelling is essential for ensuring repeatability and providing confidence measures for benthic habitat mapping. The scale of predictors within predictive models directly influences habitat distribution maps, therefore matching the scale of predictors to the scale of environmental drivers is key to improving model accuracy. This study uses a multi-scalar and hierarchical classification approach to improve the accuracy of benthic habitat maps. We used a 700-km 2 region surrounding Cape Otway in Southeast Australia with full MBES data coverage to conduct this study. Additionally, over 180 linear kilometers of towed video data collected in this area were classified using a hierarchical classification approach. Using a machine learning approach, Random Forests, we combined MBES bathymetry, backscatter, towed video and wave exposure to model the distribution of biotic classes at three hierarchical levels. Confusion matrix results indicated that greater numbers of classes within the hierarchy led to lower model accuracy. Broader scale predictors were generally favored across all three hierarchical levels. This study demonstrates the benefits of testing predictor scales across multiple hierarchies for benthic habitat characterization.
Kelp forests worldwide are under increasing pressure from anthropogenic impacts including kelp harvesting, pollution, and higher sea surface temperatures due to climate change. Accurate mapping of these habitat types is required to inform effective conservation policies. We show that adding the acoustic energy from the multibeam echosounder water-column data immediately above the seafloor as a variable in models of the distribution of benthic marine habitats can significantly improve the accuracy of the resulting maps, specifically for habitat categories that include large species of macroalgae on shallow subtidal reefs. Observations from a comprehensive towed-video survey were used as ground-truth and classified according to a hierarchical marine biotope classification scheme. The multibeam echosounder water-column data were processed into a 2D layer akin to a seafloor backscatter mosaic, but instead representing the average acoustic energy in a layer 0-1 m above the seabed, excluding the echo from the seafloor itself and after filtering the specular artifact. Including this "watercolumn mosaic" in models increased the maps' overall accuracy by up to 1.18 percent points and improved producer accuracies for habitats that contained macroalgae by up to 2.95 percent points. With increasing pressure on temperate macroalgal communities arising from warming oceans, our work provides a timely advance for mapping these critical habitats and monitoring changes in their distribution.Climate change increasingly impacts biodiversity in terrestrial and aquatic ecosystems. To inform management decisions that could mitigate further biodiversity loss, it is necessary to establish baselines of habitat distributions and monitor change in these distributions over time. For marine ecosystems, a commonly used method to achieve this is to create benthic habitat maps (Shapiro and Rohmann 2006;Godet et al. 2009). Benthic habitat maps are used in a variety of management applications such as management of fisheries, spatial marine environmental management and design of Marine Protected Areas (Bax et al. 1999;Costello 2009;Harris and Baker 2011). Generating reliable and timely habitat maps at a regional level is critical for detecting and quantifying changes in habitat distributions, and thus, informing management strategies (Coleman and Wernberg 2017;Vergés et al. 2019;Gurgel et al. 2020).Nearshore benthic communities in southern Australia and at similar latitudes globally are mainly defined by macroalgae, which are key primary producers that support a high biodiversity by providing food and protection from predators for many species of invertebrates and fish (Costanza et al. 1997;Teagle et al. 2017). However, macroalgae species are highly vulnerable to ecosystem regime shifts (e.g., urchin barrens), pollution, and the increase in sea surface temperature (Wernberg et al. 2016;Martínez et al. 2018;Carnell et al. 2020). The loss of local macroalgae communities compromises the productivity and biodiversity of their ecosystem, which can resul...
Accurate maps of biological communities are essential for monitoring and managing marine protected areas but more information on the most effective methods for developing these maps is needed. In this study, we use Wilsons Promontory Marine National Park in southeast Australia as a case study to determine the best combination of variables and scales for producing accurate habitat maps across the site. Wilsons Promontory has full multibeam echosounder (MBES) coverage coupled with towed video, remotely operated underwater vehicle (ROV) and drop video observations. Our study used an image segmentation approach incorporating MBES backscatter angular response curve and bathymetry derivatives to identify benthic community types using a hierarchical habitat classification scheme. The angular response curve data were extracted from MBES data using two different methods: 1) angular range analysis (ARA) and 2) backscatter angular response (AR). Habitat distributions were predicted using a supervised Random Forest approach combining bathymetry, ARA, and AR derivatives. Variable importance metrics indicated that ARA derivatives, such as grain size, impedance and volume heterogeneity were more important to model performance than AR derivatives mean, skewness, and kurtosis. Additionally, this study investigated the impact of segmentation software settings when creating segmented surfaces and their impact on overall model accuracy. We found using fine scale segmentation resulted in the best model performance. These results indicate the importance of incorporating backscatter derivatives into biological habitat maps and the need to consider scale to increase the accuracy of the outputs to help improve the spatial management of marine environments.
<p>Kelp forests worldwide are under ever increasing pressure from anthropogenic impacts including kelp harvesting, pollution, and higher sea surface temperatures due to climate change. Marine spatial planning requires accurate mapping of these habitat types to inform effective policy. Key data needed for benthic habitat maps to inform policy are acquired by multibeam echosounders, which collect high resolution bathymetry and backscatter of the seafloor. An additional and previously little used product of high resolution MBES are mid-water backscatter data, termed water-column data, that have been used to identify and map kelp species that extended above the seafloor. We show that incorporating water-column data as a variable for modeling benthic marine habitat distributions can significantly improve the accuracy of benthic habitat maps, specifically where habitat categories include large species of macroalgae on shallow (2-34m) subtidal reefs. The study site has full coverage multibeam bathymetry, backscatter and water-column data, alongside comprehensive observation surveys of benthic habitats using towed video. All towed video observations were classified using a hierarchal marine biotope classification scheme. Water-column data were processed into a mosaic-like product representing the acoustic energy in a layer 0-1m above the seabed. This processing included filtering of the sidelobe artefact. The volumetric water-column mosaic along with bathymetric and backscatter derivatives combined with towed video observations were used as input variables in a supervised random forest classification algorithm to create habitat maps for the study site. Variable importance was assessed for all variables and water-column performed well as it was retained in all models. Including water-column data increased overall map accuracy up to 1.18% and improved producer class accuracies that contained macroalgae up to 2.95%. With increasing pressure on temperate macroalgal communities due to a synergy of pressures arising from warming oceans, our work provides a timely advance for mapping and monitoring changes.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.