The Global Multi-Resolution Topography (GMRT) Synthesis is a multi-resolution Digital Elevation Model (DEM) developed at the Lamont-Doherty Earth Observatory of Columbia University. The data synthesis is maintained in three projections and is managed with a scalable global architecture and tiling scheme.  Primary content assembled into GMRT includes a curated multibeam bathymetry component that is derived from processed swath files and is gridded at ~100m resolution or better. These data are seamlessly assembled with other publicly available gridded data sets, including bathymetry and topography data at a variety of resolutions.  GMRT delivers the best resolution data that have been curated for a particular area of interest, and allows users to extract custom grids, images, points and profiles.

Most data processing and curation effort for GMRT is focused on cleaning and reviewing ship-based multibeam sonar data to facilitate gridding at their full spatial resolution. In addition to  performing standard data cleaning and applying necessary corrections to data, GMRT tools are used to review and assess swath data in the context of the existing data synthesis. This approach ensures that data are fit for purpose and will integrate well with existing content, and is especially well-suited for ensuring the quality of data acquired during transits. GMRT tools and workflows used for data cleaning and assessment have recently been adapted for distributed use to enable the broader community to leverage this approach, streamlining the data pipeline and ensuring high quality processed swath data can be delivered to public archives. This presentation will include a summary of GMRT tools, opportunities, and lessons learned.

How to cite: Ferrini, V., Morton, J., Drennon, H., Uribe, R., Miller, E., Martin, T., Nitsche, F., Goodwillie, A., and Carbotte, S.: Global Multi-Resolution Topography (GMRT) Synthesis – Tools and Workflows for Processing, Integrating and Accessing Multibeam Sonar Data, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-13349, https://doi.org/10.5194/egusphere-egu22-13349, 2022.

Bathymetry inversion using remote sensing techniques is a topic of increasing interest in coastal management and monitoring. Freely accessible Sentinel-2 imagery offers high-resolution multispectral data that enables bathymetry inversion in optically shallow waters. This study presents a framework leading to a generalized Satellite-Derived Bathymetry (SDB) model applicable to vast and diversified coastal regions utilizing multi-date images. A multivariate regression random forest model was used to derive bathymetry from optimal Sentinel-2 images over an extensive 210 km coastal stretch along southwestern Florida (United States). Model calibration and validation were done using airborne lidar bathymetry (ALB) data. As ALB surveys are costly, the proposed model was trained with a limited and practically feasible ALB data sample to expand the model’s practicality. Using multi-image bands as individual features in the random forest model yielded high accuracy with root-mean-square error values of 0.42 m and lower for depths up to 13 m.

How to cite: Mudiyanselage, S., Abd-Elrahman, A., and Wilkinson, B.: Bathymetry inversion with optimal Sentinel-2 imagery using random forest modeling, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10426, https://doi.org/10.5194/egusphere-egu22-10426, 2022.

More than 60% of the world's population lives near coastal zones. These are the most productive as well as the most vulnerable ecosystems in the world. Considering these, among other factors, the study of coastal zones is a matter of vital importance, so that it is necessary to have accurate information for an appropriate coastal management. The shallow bottom topography is considered one of the most critical parameter in coastal studies, because of its significance in different areas such as industry, navigation, defense, aquaculture, tourism, maritime planning, and environmental management, among others. The bathymetry is one of the biggest challenges for coastal engineers and scientists, since it is quiet complex to gather accurate data and to keep it updated because it is a time-consuming and very expensive process. In recent years, satellite-derived bathymetry (SDB) has emerged as an alternative to the most common survey techniques. In the present case study, a recently developed multi-temporal SDB model is applied to overcome problems associated with turbidity and noise effects. This model had been applied in many areas of the Caribbean and EEUU coasts with outstanding performance, providing an accurate bathymetry of the selected areas. In this case, it has been analyzed the bottom topography changes in the Cala Millor beach (Mallorca Island, Spain) between 2018, 2019 and 2020, using images from the Sentinel-2A/B twin mission of the Copernicus Programme. ACOLITE processor has been applied to Sentinel-2 L1A images for atmospheric and sunglint correction. The study aims at demonstrating the effectiveness of this model in the Mediterranean region to show its consistent performance on distinct geographic zones around the world, in addition to improving the results with a composited multi-temporal image selected automatically. Showing the confidence of this capability to be applied in any micro-tidal coast around the world may enhance the existing survey methods and highly contribute to the scientific knowledge by providing scientists and engineers with new science-based tools to better understand coastal zones.

How to cite: Viana-Borja, S., Fernández-Mora, A., Stumpf, R. P., Navarro Almendros, G., and Caballero de Frutos, I.: Satellite-based coastal bathymetry for annual monitoring on the Mediterranean coast: A case study in the Balearic Islands, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4754, https://doi.org/10.5194/egusphere-egu22-4754, 2022.

Terrain attributes are increasingly used in seabed mapping to describe the shape of the seabed. In recent years, many calls have been made to move seabed mapping practices towards multiscale characterization to better capture the natural geomorphic patterns found at different spatial scales. However, the community of practice lacks computationally efficient, user-friendly, and open-source tools to implement multiscale analyses, preventing multiscale analyses from gaining traction for seabed mapping and characterization. Here we present a new R package that enables the calculation of multiple terrain attributes like slope, curvature, and rugosity from bathymetric data. The user-friendly package allows for a repeatable and well-documented workflow that can be run using open-source tools. We also introduce a new measure of rugosity that ensures decoupling from slope. Examples of the performance of the package, including the new rugosity metric, will be presented using bathymetric datasets presenting different characteristics.

How to cite: Ilich, A., Misiuk, B., Lecours, V., and Murawski, S.: A New Toolset for Multiscale Seabed Characterization, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-10268, https://doi.org/10.5194/egusphere-egu22-10268, 2022.

Multifrequency single- and multibeam echosounders are today mature technologies for underwater mapping and monitoring of the seafloor and water column. However, the current scarcity of reference models (checked with field measurement results including detailed geoacoustical groundtruthing) for seafloor backscatter angular response and suspended sediment scattering hampers the generation of applicable information. In this context, defining heuristic models derived from measurements made in a well-controlled environment should optimize the use of backscatter data for ocean observation and management. Such reference measurements could be conducted in flumes designed for hydrodynamics and sedimentology experimental studies, since such facilities constitute well-dimensioned and equipped infrastructures adapted to the deployment of echosounders over controlled sedimentary targets. In order to check the feasibility of this concept in terms of acoustical measurement quality, a preliminary experiment was conducted in the Delta Flume (dimensions 291 x 5 x 9.5 m), as a preparation for more comprehensive systematic measurement campaigns. Multifrequency single- and multibeam echosounder data were recorded from the flume floor at various angles and from in-water fine sand plumes. The results reveal that reverberation caused by the flume walls and infrastructure does not interfere significantly with bottom targets and that fine sand plumes in the water column can be detected and measured for various particle concentrations. Future comprehensive experiments (in preparation) will feature multi-frequency multi-angle measurements both on a variety of sediment types and interface roughness, and on plumes of various sediment grain size, shape and concentration.

How to cite: Lurton, X., Roche, M., van Dijk, T., Berger, L., Fezzani, R., Fietzek, P., Gastauer, S., Klein Breteler, M., Mesdag, C., Simmons, S., and Parsons, D.: Measurements of sediment backscatter in a flume: preliminary experiment results and prospective , EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9050, https://doi.org/10.5194/egusphere-egu22-9050, 2022.

Geogenic reefs are hotspots for benthic organisms including fish. Given their ecosystem importance, the European Union has protected them by law and demands an area-wide mapping. The German federal agency for nature conservation together with scientific experts has lately published a guideline to map reefs in the Baltic Sea. Reef delineation is based on hydroacoustic backscatter mosaics which are divided and interpreted in 50x50 m cells. Each cell is categorized according to the number of boulders present:  none, 1-5, and more than 5 boulders. The categorization is strongly dependent on the data quality, hydroacoustic frequency used and technique of boulder identification (manual or automatic). By comparing data with different frequencies interpreted each manually and automatically we will demonstrate the importance of appropriate data for reef delineation.

How to cite: Papenmeier, S. and Feldens, P.: Hydroacoustic mapping of geogenic reefs, a matter of technique: a practical example from the Baltic Sea, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4684, https://doi.org/10.5194/egusphere-egu22-4684, 2022.

Bottom trawling is a fishing technique in which a net held open by otter boards is dragged across the seafloor to harvest bottom living resources. This action induces high levels of stress to ecosystems by overturning boulders, disturbing and resuspending surface sediment, and plowing scars into the seabed. In the long term the trawling impact on benthic habitats becomes problematic when the time between trawls is shorter than the time it takes for the ecosystem to recover. Since quantitative information on the intensity of bottom fishing is particularly important but rarely available, our study is crucial to reveal the extent and magnitude of the anthropogenic impacts to the seafloor. As part of the MGF Baltic Sea project, a multibeam-echosounder was used to record high-resolution bathymetric data in a small, heavily fished focus area at a 1-year interval. Based on bathymetric data, we present an automated workflow for extracting trawlmark features from seafloor morphology and deriving parameters that qualitatively characterize trawlmark intensity. We also demonstrate how the seafloor surface of an exploited area develops within a year and what can be derived from this for regeneration indicators.

How to cite: Schönke, M., Clemens, D., and Feldens, P.: How fast do Trawlmarks degenerate? A field study in muddy sediments near Fehmarn Island, German., EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-8766, https://doi.org/10.5194/egusphere-egu22-8766, 2022.

Knowing the type and distribution of seafloor sediments is crucial for many purposes, including marine spatial planning and nature conservation. Seabed sediment maps are typically obtained by manually or automatically classifying data recorded by swath sonar systems such as multibeam echosounders (MBES), aided with ground-truth data.

While progress has been made to map the seafloor based on acoustic data in an automated way, such methods have not advanced enough to become operational for routine map production in geological surveys. Mapping seafloor sediments is therefore still a manual and partly subjective process, which may imply a lack of repeatability.

In recent years, deep learning using convolutional neural networks (CNNs) has shown great promise for image classification applied in domains such as satellite or biomedical image analysis, and there is an increasing interest in the use of CNNs for seabed image classification.

In this work, we evaluate the performance of semantic segmentation using a U-Net CNN for the purpose of classifying seafloor acoustic images into sediment types.

Our study site is an area of 576 km2, located in the Søre Sunnmøre region, where seafloor sediments have been manually mapped by the Geological Survey of Norway (NGU). For our initial investigation, we simplified the NGU map into two classes – soft sediment and hard substrate – and trained multiple U-Net networks to predict the sediment classes using an MBES bathymetry grid and seabed backscatter image mosaic as source datasets. Our training reference was the expertly delineated sediment map, and the method thus seeks to mimic the human observer. Our initial analysis derived features directly from acoustic backscatter and bathymetry data but also derived slope and hillshade images from the bathymetry grid.

The MBES imagery was pre-processed and divided into patches of 256 m x 256 m (where 1 m = 1 image pixel). We evaluated models using a single input layer, e.g., backscatter mosaic, bathymetry grid, hillshade or slope respectively, and three models that used two input layers, hillshade & depth, hillshade & backscatter, slope & backscatter. Performance was evaluated using the Dice score (DS), a relative measure of overlap between the predicted and reference map.

Interestingly, results showed that for models using a single data source, the hillshade and slope models produced the highest performance with a DS of approximately 0.85, followed by the backscatter model (DS = 0.8) and the depth model with a DS of 0.7. Models using dual data sources showed improved results for the backscatter/slope & depth model (DS = 0.9) while showing a lower DS (0.7) for the hillshade & depth model.

Our preliminary results demonstrate the potential of using a U-Net to classify seafloor sediments from MBES data, thus far using two sediment classes. Assuming here that the human observer has correctly annotated the seabed sediments, such an approach could help to automate seafloor mapping in future applications. Further work will provide an in-depth analysis on feature importance, further improve the models by using additional input layers, and use data where several relevant sediment classes are included.

How to cite: Garone, R. V., Birkenes Lønmo, T. I., Tichy, F., Diesing, M., Thorsnes, T., Schimel, A. C. G., and Løvstakken, L.: Deep Learning for seafloor sediment mapping: a preliminary investigation using U-Net, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-11654, https://doi.org/10.5194/egusphere-egu22-11654, 2022.

The spatial distribution of deep-sea polymetallic nodules (PMN) is of high interest due to increasing global demand in metals (Ni, Co, Cu), and their significant contribution to deep-sea ecology as hard-substrate. The spatial mapping is based on a combination of multibeam echosounders and underwater images in parallel to traditional ground-truth sampling by box coring. The combined analysis of such data has been advanced by using machine learning approaches, especially for automated image analyses and quantitative predictive mapping. However, the presence of spatial autocorrelation (SAC) in PMN distribution has not been extensively studied. While SAC could provide information regarding the patchy distribution of PMN and thus enlighten the variable selection before machine learning modeling, it could also result in an over-optimistic validation performance when not treated carefully. Here, we present a case study from a geomorphologically complex part of the Peru Basin. The local Moran’s I analysis revealed the presence of SAC of the PMN distribution, which can be linked with specific seafloor acoustic and geomorphological characteristics such as aspect and backscatter intensity. A quantile regression forests (QRF) model was developed using three cross-validations (CV) techniques: random-, spatial-, and feature space cluster-blocking. The results showed that spatial block cross-validation is the least unbiased method. Opposite the commonly used random-CV overestimates the true prediction error. QRF predicts well in morphologically similar areas, but the model uncertainty is high in areas with novel feature space conditions. Therefore, there is the need for dissimilarity analysis and transferability assessment even at local scales. Here, we used the recently proposed method “Area of Applicability” to map the geographical areas where feature space extrapolation occurs.

How to cite: Gazis, I.-Z. and Greinert, J.: Machine learning-based modeling of deep-sea polymetallic nodules spatial distribution: spatial autocorrelation and model transferability at local scales, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-4495, https://doi.org/10.5194/egusphere-egu22-4495, 2022.