GM6.8

In 2016, through a collaboration between marine mapping programmes in Norway, Ireland, and the UK, we published a new classification scheme to aid the characterisation of seabed geomorphology (Dove et al., 2016). The classification scheme was developed to address shared objectives and challenges in seabed mapping, particularly to enable more consistent classification where required. The novel aspect of this framework was the effort to independently describe seabed features according to their observed physical 1-Morphology, and the more subjective interpretation of their origin and evolution (2-Geomorphology). Initial application of the approach within our own groups and externally proved promising, and through the welcome involvement of colleagues from Geoscience Australia, we continued to progress and improve the approach.

We are now within the second phase of the project, which involves the development of glossaries for both parts of the classification scheme. The glossary for part-1 Morphology was recently completed and published (Dove et al., 2020). This glossary includes a revised list of feature names, with definitions and representative diagrams for each feature. Feature definitions are in-part drawn from the International Hydrographic Organization (IHO) guide for undersea feature names, which were modified and augmented with additional terms to ensure the final feature catalogue and glossary encompasses the diversity of morphologies observed at the seabed.

Part-2 Geomorphology glossary is now in development. We anticipate it to be more complicated than the Morphology glossary due to the (often) variable meaning of different terms between different fields and individual scientists. But as for Part 1, our primary objective is to produce a useful and robust framework (applicable from the coastal zone to the abyss), that minimises duplication and/or ambiguity as much as possible. The Geomorphology glossary will include example bathymetry images to add further value.

Dove, D., Bradwell, T., Carter, G., Cotterill, C., Gafeira Goncalves, J., Green, S., Krabbendam, M., Mellett, C., Stevenson, A., Stewart, H. and Westhead, K., Scott, G., Guinan, J., Judge, M., Monteys, X., Elvenes, S., Maeten, N., Dolan, M., Thorsnes, T., Bjarnadottir, L., Ottesen, D., 2016. Seabed geomorphology: a two-part classification system. British Geological Survey, Open Report OR/16/001.

Dove, D., Nanson, R., Bjarnadóttir, L.R., Guinan, J., Gafeira, J., Post, A., Dolan, M.F.J., Stewart, H., Arosio, R. and Scott, G., 2020. A two-part seabed geomorphology classification scheme:(v. 2). Part 1: morphology features glossary.

How to cite: Dove, D., Nanson, R., Bjarnadóttir, L., Guinan, J., Gafeira, J., Post, A., Dolan, M. F. J., Stewart, H., and Arosio, R.: Development of a Seabed Geomorphology classification approach; aspiring towards a robust tool to support comprehensive and consistent seabed mapping, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-7751, https://doi.org/10.5194/egusphere-egu21-7751, 2021.

It is increasingly recognized that environmental variables must be considered at multiple spatial scales to produce maps of the seabed that better capture and represent geomorphic features and marine habitats. In this paper, the ability of new multiscale geomorphometric variables to classify different types of seabed habitats is tested. A digital terrain model of an area of coastal Florida with different types of intertidal habitats was used in the geospatial data analysis platform Whitebox Tools to generate multiscale measures of roughness, maximum deviation from mean elevation, maximum anisotropy in elevation deviation, maximum difference from mean elevation, and maximum spherical standard deviation. Results show that oyster reefs, muddy areas, and areas with aquatic vegetation have different multiscale terrain signatures, highlighting the potential of multiscale terrain attributes to inform seabed classification.

How to cite: Lecours, V.: Developing multiscale terrain signatures for seabed classification, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-1414, https://doi.org/10.5194/egusphere-egu21-1414, 2021.

The Golfe d’Arguin (NW Mauritania) is a 15,000 km² shelf, 150 km in N-S, and 150 km offshore.  The Golfe is subdivided into inner and outer shelves.  The shallow (<10m) inner shelf, also known as Banc d’Arguin, is part of a UNESCO World Heritage Site and renowned for a richly diverse marine ecosystem.   A steep escarpment separates the deeper (20-200 m) outer shelf.    The bottom morphology of both zones is a complex system of shoals, canyons, and channels. 

              The water is too turbid for traditional optical remote sensing bathymetry.  The alternative is bathymetry derived from ocean wave celerity as waves shoal over the shelf.

              The celerity method has been practiced for several decades, but only with small sensor footprints (typically 10-100 km²).    The Sentinel 2 satellites (now two, with plans to expand to four) have a 290 km push-broom image swath, capturing 2100 km² per second.  The entire Earth is imaged every 4-5 days.  The imagery is free, very accessible, and easy to process.   The Sentinel 2 image archive is five years and growing daily.  Mapping large regions such as the Golfe d’Arguin is made possible.

              This presentation describes the Sentinel 2 bathymetry workflow.   The Golfe d’Arguin is an excellent case study.  Long swell waves from mid-Atlantic storms are frequent.  Long waves are best for sensing the bottom depth.  (Conversely, the method does not work in locations such as the Persian Gulf shielded from oceanic wave systems.)

              For the test case, the horizontal resolution is 200m, covers depths 0 to 35 m, and accuracy is 5% of depth.   Comparisons are made with the General Bathymetric Chart of the Oceans (GEBCO), EMODnet DTM, and various literature sources.    GEBCO compares very poorly.  EMODnet is better but with much lower spatial resolution and does not capture the morphological detail seen in Sentinel 2 derived bathymetry.

              There is very limited ground truth bathymetry, e.g., Multibeam, in this area for rigorous validation.   Other validation methods are presented.  One is Sentinel 2 bathymetry of the Perth Australia area,  with wave conditions similar to Mauritania, and a LiDAR survey for ground truth.   Another is repeatability with images from different dates.

              Below links preview Golfe d’Arguin bathymetry with Sentinel 2 and EMODnet for comparison.  GoogleEarth can be used to view.

                                       http://jOmegak.com/vEGU/28QBH.Sentinel2.kmz
                                       http://jOmegak.com/vEGU/28QBH.EMOD.kmz

How to cite: Abileah, R.: The utility of Sentinel 2 for bathymetry in turbid waters:  a case study in the Gulf d’Arguin, Mauritania, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11157, https://doi.org/10.5194/egusphere-egu21-11157, 2021.

The Charlie Gibbs offsetting by ~340 km the Mid Atlantic Ridge (MAR) axis between 52°-53° N is one of the main transform systems of the North Atlantic. Located between long mid-ocean ridge segments influenced to the south by the Azores and to the north by the Iceland mantle plume, this transform system has been active since the early phases of North Atlantic rifting. Object of several surveys in the ‘70 and ‘80, Charlie Gibbs received great attention for its unique structure characterized by two long-lived right-lateral transform faults linked by a short ~40 km-long intra-transform spreading centre (ITR) with parallel fracture zone valleys extending continuously towards the continental margins. In October 2020 expedition S50 of the R/V A.N. Strakhov surveyed an area of 54552 km² covering the entire Charlie Gibbs transform system and the adjacent MAR spreading segments. We collected new bathymetric, magnetic and high-resolution single channel seismic data, along with basaltic, gabbroic and mantle rocks from 21 dredges. In this contribution we present preliminary data from cruise S50 and discusses the large-scale architecture of this unique, long-lived transform system.

How to cite: Sanfilippo, A., Skolotnev, S., and Peyve, A. and the A.N. Strakhov Expedition S50 Science Party: Geology of the Charlie Gibbs transform system (52-53ºN, Mid Atlantic Ridge): preliminary results from Akademik Nikolaj Strakhov Expedition S50, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11160, https://doi.org/10.5194/egusphere-egu21-11160, 2021.

New hydro-oceanographic data were collected in the Arctic Ocean during HIGN NORTH20 marine geophysical campaign performed in July 2020, in a COVID-19 pandemic period. HIGH NORTH20 was developed as part of the IT-Navy HIGH NORTH program, a Pluriannual Joint Research Program in the Arctic devoted to contribute to oceans knowledge in order to ensure ocean science improving conditions for sustainable development of the Ocean in the aim of United Nations Decade of Ocean Science for Sustainable development and the GEBCO - SEABED 2030 project. In order to contribute in exploration and high-resolution seabed mapping new data was collected using a multibeam echosounder (EM 302 - 30 kHz). The particular sea ice environmental condition with open-sea allowed to survey and mapping the Molloy Hole, the deepest sector of the Arctic Ocean, a key area in the global geodynamics and oceanographic context. A 3D model of the Molloy Hole (804 km²) and the detection of the deepest seafloor (5567m - 79° 08.9’ N 002° 47.0’ E) was obtained with a 10x10m grid in compliance to the IHO standards.

How to cite: Ivaldi, R., Demarte, M., Nannini, M., Aquino, G., Brancati, C., Ercolani, M., Fares, A., Guideri, M., Labella, L., Marro, M., Maurantonio, F., Nardini, R., and Niccolini, A.: High resolution mapping of the Arctic Ocean deepest area: Molloy Hole, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-16227, https://doi.org/10.5194/egusphere-egu21-16227, 2021.

Bubbling reefs are submarine structures formed by aggregating carbonate resulting from leaking gases. The reef formations can form pillars rising several meters above the sea floor. They support a high diversity of benthic communities, and in the EU Habitat Directive they are specifically mentioned as a natural habitat type that require conservation.

Knowledge about the presence, locations and shape of bubbling reefs are usually obtained by geophysical surveying using multibeam echosounder (MBES), sidescan sonar and/or seismic acquisition systems, combined with ground truth verification. However, this traditional survey method is time consuming, especially for full coverage surveys in shallow water. Full coverage surveys are a requirement to capture the bubbling reefs due to their relatively small spatial extent. Besides, traditional geophysical vessel borne surveys have their limitations in shallow water due to low spatial coverage and vessel draft.

In recent years, airborne topobathymetric (green wavelength) lidar has emerged as a new possible surveying method in shallow water (e.g. Andersen et al., 2017). Compared to vessel borne MBES, full coverage lidar surveys can be conducted within hours instead of days/weeks, while also including full coverage in the shallow water and a seamless transition between land and water. Thus, topobathymetric lidar may be a good choice for carrying out full coverage surveys in large shallow water areas. However, the accuracy and the resolution of the collected dataset are important in these surveys, not least when mapping small scale features such as bubbling reefs.

In this study, we investigated the potential of mapping bubbling reefs in shallow water (<10 m) using topobathymetric lidar. The main objective was to assess the performance of airborne topobathymetric lidar to detect and resolve small scale objects, i.e. bubbling reefs, by comparison to MBES data. Both MBES and lidar data were acquired in spring 2019 in a designated Natura 2000 area close to Hirsholmene in the northern Kattegat region in Denmark. The comparison of the two datasets included a quantification of the accuracy, and an assessment of the performance for mapping bubbling reefs.

Reference:

Andersen M.S., Gergely A., Al-Hamdani Z., Steinbacher F., Larsen L.R., Ernstsen V.B. (2017). Processing and performance of topobathymetric lidar data for geomorphometric and morphological classification in a high-energy tidal environment. Hydrology and Earth System Sciences, 21: 43-63, DOI: 10.5194/hess-21-43-2017.

How to cite: Andersen, M. S., Hansen, L. Ø., Al-Hamdani, Z., Hansen, S. S., Niederwieser, M., Baran, R., Steinbacher, F., and Ernstsen, V. B.: Mapping shallow water bubbling reefs – a method comparison between topobathymetric lidar and multibeam echosounder, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-15899, https://doi.org/10.5194/egusphere-egu21-15899, 2021.

This study, conducted under the auspices of the Commonwealth Marine Economies Programme (CMEP), involved the characterisation of geomorphology and benthic assemblage groups around the previously unmapped western St. Lucian coastline. A high-resolution (2 x 2 m) seabed map was then produced, specifically focussing on the economically important regions.

Two hundred twelve video tow transects were acquired by CEFAS and associated to Multibeam echosounder (MBES) data collected by the United Kingdom Hydrographic Office (UKHO). Through object-based image analysis, MBES data and derivatives were categorized into eight basic morphological classes (e.g. slope, flat), with manual intervention required to further discriminate more complex forms (e.g. anthropogenic scours). Percentage coverage of epibenthic biota and substrate type data were extracted from still seabed images using the CATAMI[1] morphological classification system and a randomised point count approach. Benthic community assemblages were then defined based on K-means clustering. A random forest model was used to predict benthic community groups. Predictive layers included MBES-derived physical properties, geomorphological classes and wave exposure (using GREMO[2]). The resulting model had a predictive accuracy of 80%.

St. Lucia is characterised by gently sloping exposed plateaus (~30 m deep), extending for several km offshore in the north and south of the island. These appear to be mostly covered by an algal-dominated gravelly substratum. Bioturbated mud and sand with pockmarks occur typically in sheltered bays, near river mouths or within paleochannels that incise the seabed. Seagrass patches were more difficult to predict (50% producer’s accuracy) but are generally limited to very shallow coastal waters habitats. Finally, coral and sponge-dominated reef communities appear to generally be associated with raised “staircase” platforms and bommie features close to the coast, however, these were also observed further from shore and in deeper waters.

The seabed habitat maps produced will support the St. Lucian government to manage their shallow seabed resources. In particular, the maps will assist the systematic characterisation of the coral reef habitats and, therefore, improve delineations of marine reserves, especially around the town of Soufrière and the Pitons UNESCO world heritage site.

[1]           Althaus, F., N. and many others. 2015. A standardised vocabulary for identifying benthic biota and substrata from underwater imagery: The CATAMI classification scheme. PLoS ONE 10:1–18.

[2]           Pepper, A. and Puotinen, M. L., 2009. GREMO: A GIS-based generic model for estimating relative wave exposure. The 18th World IMACS Congress and MODSIM09 Int. Congress on Modelling and Simulation (pp. 1964-1970). Cairns, Australia.

How to cite: Arosio, R., Mitchell, P., Hawes, J., Bolam, S., Benson, L., and Sperry, J.: Small Island Developing States (SIDS) and the sea: creating high resolution habitat maps to support effective marine management in St. Lucia, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-102, https://doi.org/10.5194/egusphere-egu21-102, 2021.

Recent studies on seafloor mapping have presented different modelling methods to map and classify marine sediment distribution. However, are these methods classify different sediment classes the same way? And how do we choose the right model for a certain set of sediment classes? In this study, we aim to address these issues by using ensemble modelling to map the distribution of different sediment class on a dynamic, shallow continental shelf. Our data were derived from side-scan mosaics and multibeam data repeatedly collected from 2016 to 2018 in the Sylt Outer Reef (German Bight). We used a probabilistic approach for each class separately and then compared the predicted probability for each class, to see which class is more likely to be assigned to the location. Each sediment class was predicted using a combination of different classification modelling techniques, and then the result of these models was ensembled to produced one final prediction. This approach avoids selecting one single method, limits model selection bias and can provide information on the trends and variation across models.  Furthermore, we also looked on the temporal changes in sediment distributions by comparing the sediment class predictions from 2016 to 2018.

Our analysis suggest that combining different modelling techniques  (i.e. random forest, boosting regression trees etc.) provide higher predictive accuracy than using one single modelling method. The resulting sediment distribution maps are more objective and are produced faster than manual delineated maps often considered by stakeholders. We also identify some limitations in having small sample size and we proposed that by combining certain models and choosing the proper amount of pseudo-absence or background data can address this issue.

How to cite: Galvez, D., Papenmeier, S., Bartholomä, A., and Wiltshire, K. H.: Seafloor sediment classification of the Sylt Outer Reef, German Bight from 2016 -2018 using ensemble modelling, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-11688, https://doi.org/10.5194/egusphere-egu21-11688, 2021.

In 2017, the Israel Oceanographic and Limnological Research (IOLR) started an annual seafloor monitoring program. The aim of the program is to evaluate the rate of erosion/deposition and the influence of man-made infrastructures on the seabed along the Israeli continental shelf south of Akko. The survey program onboard R/V Bat-Galim includes a multibeam (Kongsberg EM2040), sub-bottom (Knudsen 3260 Chirp) mapping and box-core sediment sampling along 13 transects across the shelf, from WD 10-100 m. The multibeam was operated at 400-kHz yielding a horizontal resolution of 0.25-1.0 m (depending on water depth), and vertical uncertainty of several centimeters. Using the QPS FMGT software, both angular response curves (ARA) and 0.5 m horizontal resolution of Backscatter data (BS) were derived. The multibeam acoustic return intensities (BS) were locally calibrated at selected reference areas using in-situ sediment sampling. 
The main source of sediments along the Israeli continental shelf is the Nile Delta which undergoes erosion since 1960 when the Aswan dam was constructed. Along the Israeli inner-shelf, these sediments are transported northward and westward by wind-derived currents and storms. The analysis of the bathymetric surfaces from the consecutive years 2017-2020 shows that the shelf is stable in terms of sediment processes except along the marine infrastructures and natural seafloor features (e.g. rocky bottom outcrops) where patterns of sediment accumulation and erosion are observed. The variability along the marine infrastructures is mostly seen in the shallow water (less than 30 m) where yearly changes of up to +/-0.4 m of sediment accumulation/erosion in the vertical axis were measured.
The locally calibrated multibeam BS enabled grain size mode evaluation ranging from very fine gravel (-1 phi) to clay (9 phi). Additional in-situ sampling validated the reliability of the grain size classification method for the Israeli, continental shelf. Accordingly, we show that the Israeli continental shelf south of Haifa Bay is characterized by a sandy seafloor strip at WD 0-35 m and a muddy strip that extends west up to WD 100 m (in agreement with previous studies). Gravelly areas are identified at the coast-parallel Kurkar outcrops (Calcareous sandstone rocky ridges or rock patches) in water depths of 10-15m and 35-40m and in some places even at WD of 90 m. This demonstrates that grain size classification by locally calibrated multibeam BS is likely to be a very useful and fast method for monitoring changes in seafloor characteristics over large areas over time.

How to cite: Giladi, A., Kanari, M., katz, T., and Tibor, G.: Monitoring sediment transport and grain size dynamics along the Israeli continental shelf with multibeam bathymetry and backscatter data, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-6257, https://doi.org/10.5194/egusphere-egu21-6257, 2021.

With less than 20% of the seafloor mapped at a sufficiently high resolution for geological and biological studies (<50m), there is a need for new technological approaches to map and characterize the seafloor environment at higher resolutions. Here, we present preliminary results of an investigation into the use of interferometric synthetic aperture sonar (InSAS) as a new approach to help fill this gap. InSAS can provide very high-resolution acoustic imagery (3cm/pixel) and bathymetry (25 cm/pixel) as well as large coverage area (up to 150m across track per side while flying at a 15m altitude, at 6 knots). Compared to traditional sidescan sonars, high-resolution imagery in both along and across track directions is achieved by the synthetic aperture of the sonar array, which uses a large number of receiver arrays and a cm-size spacing between individual elements. This technique has so far mostly been used for military and industrial purposes.

Onboard the Atlantic Kingfisher in October 2020, we used Kraken Robotic Systems’ InSAS system on a Katfish towed vehicle to survey 85 km² of the Tail of the Grand Banks, the southernmost extremity of the continental shelf offshore Newfoundland, Canada. During a survey, the sonar is set at a center frequency of 337 kHz and survey planning included data coverage overlap for 140% coverage of the seafloor. Kraken Robotic’s processing algorithm and the towing optic cable allowed for real-time processing of the data and initial post processing using Caris Onboard.

How to cite: Gini, C., Robert, K., Jamieson, J., and Dillon, J.: Interferometric Synthetic Aperture Sonar as a tool for seafloor geological mapping on the Grand Banks offshore Atlantic Canada: preliminary results., EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-3939, https://doi.org/10.5194/egusphere-egu21-3939, 2021.

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.

How to cite: Porskamp, P., Schimel, A., Young, M., Rattray, A., Ladroit, Y., and Ierodiaconou, D.: Integrating multibeam echosounder water-column data into benthic habitat mapping, EGU General Assembly 2021, online, 19–30 Apr 2021, EGU21-10389, https://doi.org/10.5194/egusphere-egu21-10389, 2021.