Subtidal Natural Hard Substrate Quantitative Habitat Mapping: Interlinking Underwater Acoustics and Optical Imagery with Machine Learning

Subtidal natural hard substrates (SNHS) promote occupancy by rich benthic communities that provide irreplaceable and fundamental ecosystem functions, representing a global priority target for nature conservation and recognised in most European environmental legislation. However, scientifically validated methodologies for their quantitative spatial demarcation, including information on species occupancy and fine-scale environmental drivers (e.g., the effect of stone size on colonisation) are rare. This is, however, crucial information for sound ecological management. In this investigation, high-resolution (1 m) multibeam echosounder (MBES) depth and backscatter data and derivates, underwater imagery (UI) by video drop-frame, and grab sediment samples, all acquired within 32 km2 of seafloor in offshore Belgian waters, were integrated to produce a random forest (RF) spatial model, predicting the continuous distribution of the seafloor areal cover/m2 of the stones’ grain sizes promoting colonisation by sessile epilithic organisms. A semi-automated UI acquisition, processing, and analytical workflow was set up to quantitatively study the colonisation proportion of different grain sizes, identifying the colonisation potential to begin at stones with grain sizes Ø ≥ 2 cm. This parameter (i.e., % areal cover of stones Ø ≥ 2 cm/m2) was selected as the response variable for spatial predictive modelling. The model output is presented along with a protocol of error and uncertainty estimation. RF is confirmed as an accurate, versatile, and transferable mapping methodology, applicable to area-wide mapping of SNHS. UI is confirmed as an essential aid to acoustic seafloor classification, providing spatially representative numerical observations needed to carry out quantitative seafloor modelling of ecologically relevant parameters. This contribution sheds innovative insights into the ecologically relevant delineation of subtidal natural reef habitat, exploiting state-of-the-art underwater remote sensing and acoustic seafloor classification approaches.

Acoustic Seafloor Classification Using the Weyl Transform of Multibeam Echosounder Backscatter Mosaic

The use of multibeam echosounder systems (MBES) for detailed seafloor mapping is increasing at a fast pace. Due to their design, enabling continuous high-density measurements and the coregistration of seafloor’s depth and reflectivity, MBES has become a fundamental instrument in the advancing field of acoustic seafloor classification (ASC). With these data becoming available, recent seafloor mapping research focuses on the interpretation of the hydroacoustic data and automated predictive modeling of seafloor composition. While a methodological consensus on which seafloor sediment classification algorithm and routine does not exist in the scientific community, it is expected that progress will occur through the refinement of each stage of the ASC pipeline: ranging from the data acquisition to the modeling phase. This research focuses on the stage of the feature extraction; the stage wherein the spatial variables used for the classification are, in this case, derived from the MBES backscatter data. This contribution explored the sediment classification potential of a textural feature based on the recently introduced Weyl transform of 300 kHz MBES backscatter imagery acquired over a nearshore study site in Belgian Waters. The goodness of the Weyl transform textural feature for seafloor sediment classification was assessed in terms of cluster separation of Folk’s sedimentological categories (4-class scheme). Class separation potential was quantified at multiple spatial scales by cluster silhouette coefficients. Weyl features derived from MBES backscatter data were found to exhibit superior thematic class separation compared to other well-established textural features, namely: (1) First-order Statistics, (2) Gray Level Co-occurrence Matrices (GLCM), (3) Wavelet Transform and (4) Local Binary Pattern (LBP). Finally, by employing a Random Forest (RF) categorical classifier, the value of the proposed textural feature for seafloor sediment mapping was confirmed in terms of global and by-class classification accuracies, highest for models based on the backscatter Weyl features. Further tests on different backscatter datasets and sediment classification schemes are required to further elucidate the use of the Weyl transform of MBES backscatter imagery in the context of seafloor mapping.

Mapping gravel beds combining multibeam bathymetry, underwater video, and grab samples with continuous spatial modeling by random forest

<p>In this study, high-resolution (1 m) multibeam echosounder system (MBES) bathymetry data and derivatives, optical images by underwater video drop-frame, and Hamon grab sediment samples, all acquired within 170 km2 of seafloor in offshore Belgian Waters, were integrated to produce a random forest spatial model targeting the prediction of the continuous surficial distribution of gravel %, i.e., a substrate category whose known detailed distribution is central to the environmental stewardship of natural gravel bed habitat. MBES bathymetry reveals explicit details of the seafloor topography, allowing the derivation of geomorphometric variables that are important in the classification process. Underwater video and grab samples provide the means to directly observe the nature and distribution of the response variable. The model output is presented along with a protocol of error and uncertainty estimation, providing detailed information of the gravel spatial distribution that would otherwise remain undetected by categorical-type classifications, focused on predefined habitat classification schemes. Targeting the methodological improvement of this mapping approach, an overview of the limitations identified at the various steps of the acoustic seafloor classification (ASC) pipeline is presented.</p>

Assessing the Repeatability of Automated Seafloor Classification Algorithms, with Application in Marine Protected Area Monitoring

The number and areal extent of marine protected areas worldwide is rapidly increasing as a result of numerous national targets that aim to see up to 30% of their waters protected by 2030. Automated seabed classification algorithms are arising as faster and objective methods to generate benthic habitat maps to monitor these areas. However, no study has yet systematically compared their repeatability. Here we aim to address that problem by comparing the repeatability of maps derived from acoustic datasets collected on consecutive days using three automated seafloor classification algorithms: (1) Random Forest (RF), (2) K–Nearest Neighbour (KNN) and (3) K means (KMEANS). The most robust and repeatable approach is then used to evaluate the change in seafloor habitats between 2012 and 2015 within the Greater Haig Fras Marine Conservation Zone, Celtic Sea, UK. Our results demonstrate that only RF and KNN provide statistically repeatable maps, with 60.3% and 47.2% agreement between consecutive days. Additionally, this study suggests that in low-relief areas, bathymetric derivatives are non-essential input parameters, while backscatter textural features, in particular Grey Level Co-occurrence Matrices, are substantially more effective in the detection of different habitats. Habitat persistence in the test area between 2012 and 2015 was 48.8%, with swapping of habitats driving the changes in 38.2% of the area. Overall, this study highlights the importance of investigating the repeatability of automated seafloor classification methods before they can be fully used in the monitoring of benthic habitats.

3D Scanning bathymetry applied for assessment and monitoring of protected marine habitats: El Cachucho case study (Spanish Cantabrian margin)

<p>El Cachucho, also known as Le Danois bank, is the first and only marine zone declared as Marine Protected Area (MPA) in Spain since 2008. This bank consists on a 72 km-long E-W trending marginal platform located at Spanish Cantabrian margin (southern Bay of Biscay) and interpreted as horst block separated from the Spanish continental shelf by an interior basin. The bank seafloor has an almost flat-topped morphology with minimum water depth of 424 m, having only local structural and erosive features. During last decades researchers have highlighted the importance of the bathyal ecosystem developed in this geological formation. As a result, significant efforts are being carried out to asses and monitor the evolution of this MPA in order to ensure the conservation of its biodiversity, applying new techniques as 3D scanning bathymetry.</p><p>Since 2013 the Spanish Institute of Oceanography (IEO) is leading the ESMAREC project (founded by the Spanish Government) for the monitoring of El Cachucho in order to guarantee the continuity as MPA based European Union regulations. This monitoring mainly consists on repeated multibeam seafloor bathymetries to assess the geomorphological evolution and reflectivity mosaics in order to map and classify the seafloor that can be related to the different types of marine habitats. The survey plan of the ECOMARG-2019 oceanographic cruise included four different locations that were chosen along the El Cachucho for sampling stations with the remotely operated towed vehicle (ROTV) POLITOLANA in order to identify various species of gorgonians and sponges with video images. Furthermore, in those locations, a multi-parametric platform system (lander) was anchored to study the oceanographic dynamics of the Benthic Boundary Layer (BBL). Both ROTV operations and lander anchorages require a detailed knowledge of the seafloor morphology for instrumental safety and optimize efforts. Existing multibeam bathymetry along El Cachucho before the ECOMARG-2019 cruise was only 75 meter and then inadequate to carry out those seafloor operations.</p><p>With the aim to improve the existing bathymetry, during the ECOMARG-2019 cruise was used the Kongsberg EM710 multibeam echo-sounder using the 3D Scanning technique. In this technique the vessel navigates to 0.5 knots and 250 beams sweep the bottom with an 45º opening angle and 10º horizontal movement. Higher point density was achieved, so it was possible to increase the average resolution of bathymetry and reflectivity up to 5 meters. New high resolution data provided a precise image of the geomorphology and allowed a more detailed seafloor classification. In this way, potential risks were reduced during ROTV operations and anchorages. In addition, the locations for ROTV operations were optimized based on the reflectivity mosaics that allowed to identify hard seafloor zones, preferred typology of seabed for gorgonians and sponges. Using the 3D Scanning in El Cachucho has resulted in an essential tool for safety and to optimize the seafloor operations. This technique allows to achieve a detailed knowledge of the seafloor in order to better assess and monitor MPA.</p>

High-Silica Lava Morphology at Ocean Spreading Ridges: Machine-Learning Seafloor Classification at Alarcon Rise

The oceanic crust consists mostly of basalt, but more evolved compositions may be far more common than previously thought. To aid in distinguishing rhyolite from basaltic lava and help guide sampling and understand spatial distribution, we constructed a classifier using neural networks and fuzzy inference to recognize rhyolite from its lava morphology in sonar data. The Alarcon Rise is ideal to study the relationship between lava flow morphology and composition, because it exhibits a full range of lava compositions in a well-mapped ocean ridge segment. This study shows that the most dramatic geomorphic threshold in submarine lava separates rhyolitic lava from lower-silica compositions. Extremely viscous rhyolite erupts as jagged lobes and lava branches in submarine environments. An automated classification of sonar data is a useful first-order tool to differentiate submarine rhyolite flows from widespread basalts, yielding insights into eruption, emplacement, and architecture of the ocean crust.