Seafloor sediment mapping is a requirement in various applications (e.g., defence/naval, environm... more Seafloor sediment mapping is a requirement in various applications (e.g., defence/naval, environmental, maritime industry) . Specifically, the delineation of seafloor sediment types is crucial to the definition of the spatial distribution of benthic Frontiers in Earth Science
Gas seeps and fluid-flow related seabed features are found over the entire Norwegian exclusive ec... more Gas seeps and fluid-flow related seabed features are found over the entire Norwegian exclusive economic zone (EEZ). Multibeam water-column data from c. 136 000 km 2 has revealed more than 5 000 gas seeps. Most of the gas seeps seem to have biogenic, thermogenic or mixed origin; some may be of abiotic origin. The spatial distribution of the gas seeps appears to correlate with: 1-structural highs with associated faulting, exposing hydrocarbon reservoir rocks at or near the seabed; 2-faults serving as conduits for fluid flow; 3-settings where reservoir rocks overlain by less permeable cap rocks sub-crop at the seabed. Other mechanisms involve seepage around abandoned exploration wells, and possible abiotic gas generation from serpentinisation of ultramafic rocks near mid-oceanic ridges. The gas seeping from the Norwegian cold seeps is mostly methane and has, in many places, led to the formation of methane-derived authigenic carbonate crusts, which give evidence for either extensive gas seepage in the past or long-lived seepage. Chemosynthetic communities are commonly associated with cold seeps and may form special habitats together with the carbonate crusts. Methane seepage has been proposed to contribute significantly to the global carbon budget and may be associated with gas hydrates giving rise to potential geohazards. Gas seeps have been identified and spatially mapped as acoustic gas flares, using multibeam echosounder systems, which have the ability to record reflections from both the water column and the seabed. Water-column data have been recorded in the MAREANO seabed mapping program since 2010, covering an area of c. 262 000 km 2 , with a data volume in the order of 210 Tb. The observations of extensive gas flares in the Norwegian EEZ are available to the scientific community and other users through a dedicated MAREANO data and web access system.
<p>Cold seeps are commonly associated with water column and seabed features. Active... more <p>Cold seeps are commonly associated with water column and seabed features. Active seeps form acoustic flares in the water column and can be detected using data from single or multibeam beam echosounders. They may be associated with pockmarks, but the majority of pockmarks on the Norwegian continental shelf have proven to be inactive. Cold seeps are commonly associated with carbonate crust fields exposed at the seabed. <br>Studies using multibeam echosounder water column data in the Håkjerringdjupet region, underlain by the petroleum province Harstad Basin, have revealed more than 200 active gas flares related to cold seeps. We have studied the seabed around some of these, using the HUGIN HUS AUV equipped with HiSAS 1030 Synthetic Aperture Sonar (SAS) from Kongsberg. The SAS gave a 2 x 150 m wide swath. The primary product is the sonar imagery with a pixel resolution up to c. 3 x 3 cm. For selected areas, bathymetric grids with 20x20 cm grids were produced, giving unrivalled resolution at these water depths. The carbonate crust fields have normally a characteristic appearance, with a low reflectivity and a rugged morphology compared to the surrounding sediments. <br>The interpretation of the acoustic data was verified by visual inspection using the TFish photo system on the AUV, and at a later stage by ROV video footage and physical sampling. The integration of hullborne echosounder data with AUV-mounted acoustic and visual tools provides a very powerful approach for studies of cold seep habitats and related seabed features.<br>An important conclusion from the study is that many pockmarks are not associated with active gas seeps today, and that many of the presently active gas seeps are associated with carbonate crust fields which are readily identifiable from synthetic aperture sonar data.</p>
Results from geological mapping within the MAREANO (Marine Areal Database for Norwegian Coasts an... more Results from geological mapping within the MAREANO (Marine Areal Database for Norwegian Coasts and Sea Areas) programme and mapping projects in the coastal zone reveal a rich and diverse seafloor in Norwegian territories. The geomorphology and sediment distribution patterns reflect a complex geological history, as well as various modern-day hydrodynamic processes. By early 2019, MAREANO has mapped more than 200 000 km 2 ( c. 10%) of Norwegian offshore areas, spanning environmental gradients from shallow water to more than 3000 m depth, with ocean currents in places exceeding 1 m s −1 and water temperatures below −1°C. Inshore, along the 100 000 km-long Norwegian coastline, the Geological Survey of Norway (NGU) has conducted a series of seabed mapping projects in collaboration with local communities, industry and other stakeholders, resulting in detailed seabed and thematic maps of seabed properties covering c. 10 000 km 2 (11% of the areas). Bathymetric and geological maps produced ...
Et fiskefelt i Ytre Sklinnadjupet i Norskehavet ble i 2017 stengt for fiske av atlantisk kveite (... more Et fiskefelt i Ytre Sklinnadjupet i Norskehavet ble i 2017 stengt for fiske av atlantisk kveite (Hippoglossus hippoglossus) på grunn av funn av høye nivåer av miljøgifter. De høye nivåene av særlig kvikksølv, men også dioksiner og PCB, kan ikke knyttes til spesielle biologiske egenskaper ved kveitene fra dette området. Det er kjent at det ble dumpet et stort antall tønner med kjemisk avfall ved Haltenbanken og Sklinnabanken fram til 1971, og det er vurdert om dette kunne være en mulig kilde. I et forsøk på å spore eventuelle kilder for miljøgifter i området, har vi i dette arbeidet analysert nivåer av miljøgifter i prøver av sedimenter og sjøpølser (Parastichopus tremulus, sedimentspisende) fra et tokt med MAREANO i juni 2018 samt i sedimentprøver fra flere tidligere tokt. Resultatene er oppsummert sammen med eksisterende kunnskap om fysisk miljø (batymetri, oseanografi, sedimentasjonsmiljø) og nivåer av kvikksølv, dioksiner og PCB i fisk fra tidligere overvåkning i og rundt området...
MAREANO kartlegger dybde, bunnforhold, natur typer og forurensning i norske havområder. Resultate... more MAREANO kartlegger dybde, bunnforhold, natur typer og forurensning i norske havområder. Resultatene gjøres tilgjengelig på MAREANOs nettsider og visualiseres ved hjelp av kart.
<p>Knowing the type and distribution of seafloor sediments is crucial for many purp... more <p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p>
Continental shelf sediments are places of both rapid organic carbon turnover and accumulation, wh... more Continental shelf sediments are places of both rapid organic carbon turnover and accumulation, while at the same 5 time increasingly subjected to human-induced disturbances. Recent research suggests that shelf sediments might have a role to play as a natural climate solution, e.g. by protecting the seafloor against human-induced disturbance. However, we have an incomplete understanding about the centres of organic carbon accumulation and storage on continental shelves. To better constrain the rate of accumulation and the mass of organic carbon that is stored in sediments, we developed and applied a spatial modelling framework that allows to estimate those quantities from sparse observations and predictor variables known 10 or suspected to influence the spatial patterns of these parameters. This paper presents spatial distribution patterns of organic carbon densities and accumulation rates in the North Sea and Skagerrak. We found that organic carbon stocks and accumulation rates are ...
An initiative is currently being taken by several Norwegian organizations to obtain funds to inte... more An initiative is currently being taken by several Norwegian organizations to obtain funds to intensify ongoing investigations on marine sea-floor mapping off Norway. Led by the Geological Survey of Norway and Institute of Marine Research, planning during the last two years has led to the inception of a large-scale mapping project entitled “MAREANO Marine Areal Database for the Norwegian Sea”. The investigation area covers 270 000 km2 of the shelf and deep sea off the central part of western Norway. It is a commercially important region for fisheries and the petroleum industry and includes the world’s largest system of cold-water coral reefs. The aim of MAREANO is to collect new as well as historical data elucidating the physical, chemical and biological characteristics of the seabed along the mid-Norwegian shelf and parts of the deeper Norwegian Sea. The project shall produce maps and/or provide information on seabed bathymetry, marine habitats, biological diversity and resources, m...
Today’s detailed swath bathymetry and backscatter obtained from multibeam sonar data, together wi... more Today’s detailed swath bathymetry and backscatter obtained from multibeam sonar data, together with advanced ground truthing tools have opened up new possibilities for the characterisation and mapping of marine habitats. This has also triggered the need for robust habitat classification systems. Incorporating data and preliminary results in classification schemes can be problematic if the schemes do not explicitly address high-resolution information of seabed morphology and texture. Regional character has been difficult using wellestablished classifications schemes, e.g. EUNIS (http:// eunis.eea.eu.int/index.jsp) which have problems coping with mixed substrates such as till normally found in glaciated areas (Rinde et al., 2004). This problem makes such classifications impractical to use in Norwegian waters, where glacial deposits dominate the seabed. Valentine et al. (2005) designed a habitat classification system for use along the repeatedly glaciated northeast coast of the USA and...
9 Abstract— The deep Norwegian Sea spans depths reaching to nearly 10 6000 m and an area of aroun... more 9 Abstract— The deep Norwegian Sea spans depths reaching to nearly 10 6000 m and an area of around 1 000 000 km extending both sides of 11 the rugged mid-Atlantic Ridge. Armed with coarse regional 12 bathymetric and oceanographic datasets the Norwegian national 13 offshore mapping programme, MAREANO, was tasked with 14 mapping ‘representative areas’ of this vast seabed terrain. Here we 15 introduce the planning process, guided by semi-automatic methods, 16 and present examples from newly acquired ship-borne multibeam 17 bathymetric mapping in this varied deep-sea terrain with a focus on 18 the use of these bathymetry data for onward substrate, 19 geomorphological, and habitat mapping. We discuss the challenges 20 of using these data with existing, lower quality, bathymetry data, as 21 well as highlighting some typical data artefacts which can limit the 22 calculation of meaningful terrain attributes and thereby their use in 23 geological and habitat mapping. 24
Seafloor sediment mapping is a requirement in various applications (e.g., defence/naval, environm... more Seafloor sediment mapping is a requirement in various applications (e.g., defence/naval, environmental, maritime industry) . Specifically, the delineation of seafloor sediment types is crucial to the definition of the spatial distribution of benthic Frontiers in Earth Science
Gas seeps and fluid-flow related seabed features are found over the entire Norwegian exclusive ec... more Gas seeps and fluid-flow related seabed features are found over the entire Norwegian exclusive economic zone (EEZ). Multibeam water-column data from c. 136 000 km 2 has revealed more than 5 000 gas seeps. Most of the gas seeps seem to have biogenic, thermogenic or mixed origin; some may be of abiotic origin. The spatial distribution of the gas seeps appears to correlate with: 1-structural highs with associated faulting, exposing hydrocarbon reservoir rocks at or near the seabed; 2-faults serving as conduits for fluid flow; 3-settings where reservoir rocks overlain by less permeable cap rocks sub-crop at the seabed. Other mechanisms involve seepage around abandoned exploration wells, and possible abiotic gas generation from serpentinisation of ultramafic rocks near mid-oceanic ridges. The gas seeping from the Norwegian cold seeps is mostly methane and has, in many places, led to the formation of methane-derived authigenic carbonate crusts, which give evidence for either extensive gas seepage in the past or long-lived seepage. Chemosynthetic communities are commonly associated with cold seeps and may form special habitats together with the carbonate crusts. Methane seepage has been proposed to contribute significantly to the global carbon budget and may be associated with gas hydrates giving rise to potential geohazards. Gas seeps have been identified and spatially mapped as acoustic gas flares, using multibeam echosounder systems, which have the ability to record reflections from both the water column and the seabed. Water-column data have been recorded in the MAREANO seabed mapping program since 2010, covering an area of c. 262 000 km 2 , with a data volume in the order of 210 Tb. The observations of extensive gas flares in the Norwegian EEZ are available to the scientific community and other users through a dedicated MAREANO data and web access system.
<p>Cold seeps are commonly associated with water column and seabed features. Active... more <p>Cold seeps are commonly associated with water column and seabed features. Active seeps form acoustic flares in the water column and can be detected using data from single or multibeam beam echosounders. They may be associated with pockmarks, but the majority of pockmarks on the Norwegian continental shelf have proven to be inactive. Cold seeps are commonly associated with carbonate crust fields exposed at the seabed. <br>Studies using multibeam echosounder water column data in the Håkjerringdjupet region, underlain by the petroleum province Harstad Basin, have revealed more than 200 active gas flares related to cold seeps. We have studied the seabed around some of these, using the HUGIN HUS AUV equipped with HiSAS 1030 Synthetic Aperture Sonar (SAS) from Kongsberg. The SAS gave a 2 x 150 m wide swath. The primary product is the sonar imagery with a pixel resolution up to c. 3 x 3 cm. For selected areas, bathymetric grids with 20x20 cm grids were produced, giving unrivalled resolution at these water depths. The carbonate crust fields have normally a characteristic appearance, with a low reflectivity and a rugged morphology compared to the surrounding sediments. <br>The interpretation of the acoustic data was verified by visual inspection using the TFish photo system on the AUV, and at a later stage by ROV video footage and physical sampling. The integration of hullborne echosounder data with AUV-mounted acoustic and visual tools provides a very powerful approach for studies of cold seep habitats and related seabed features.<br>An important conclusion from the study is that many pockmarks are not associated with active gas seeps today, and that many of the presently active gas seeps are associated with carbonate crust fields which are readily identifiable from synthetic aperture sonar data.</p>
Results from geological mapping within the MAREANO (Marine Areal Database for Norwegian Coasts an... more Results from geological mapping within the MAREANO (Marine Areal Database for Norwegian Coasts and Sea Areas) programme and mapping projects in the coastal zone reveal a rich and diverse seafloor in Norwegian territories. The geomorphology and sediment distribution patterns reflect a complex geological history, as well as various modern-day hydrodynamic processes. By early 2019, MAREANO has mapped more than 200 000 km 2 ( c. 10%) of Norwegian offshore areas, spanning environmental gradients from shallow water to more than 3000 m depth, with ocean currents in places exceeding 1 m s −1 and water temperatures below −1°C. Inshore, along the 100 000 km-long Norwegian coastline, the Geological Survey of Norway (NGU) has conducted a series of seabed mapping projects in collaboration with local communities, industry and other stakeholders, resulting in detailed seabed and thematic maps of seabed properties covering c. 10 000 km 2 (11% of the areas). Bathymetric and geological maps produced ...
Et fiskefelt i Ytre Sklinnadjupet i Norskehavet ble i 2017 stengt for fiske av atlantisk kveite (... more Et fiskefelt i Ytre Sklinnadjupet i Norskehavet ble i 2017 stengt for fiske av atlantisk kveite (Hippoglossus hippoglossus) på grunn av funn av høye nivåer av miljøgifter. De høye nivåene av særlig kvikksølv, men også dioksiner og PCB, kan ikke knyttes til spesielle biologiske egenskaper ved kveitene fra dette området. Det er kjent at det ble dumpet et stort antall tønner med kjemisk avfall ved Haltenbanken og Sklinnabanken fram til 1971, og det er vurdert om dette kunne være en mulig kilde. I et forsøk på å spore eventuelle kilder for miljøgifter i området, har vi i dette arbeidet analysert nivåer av miljøgifter i prøver av sedimenter og sjøpølser (Parastichopus tremulus, sedimentspisende) fra et tokt med MAREANO i juni 2018 samt i sedimentprøver fra flere tidligere tokt. Resultatene er oppsummert sammen med eksisterende kunnskap om fysisk miljø (batymetri, oseanografi, sedimentasjonsmiljø) og nivåer av kvikksølv, dioksiner og PCB i fisk fra tidligere overvåkning i og rundt området...
MAREANO kartlegger dybde, bunnforhold, natur typer og forurensning i norske havområder. Resultate... more MAREANO kartlegger dybde, bunnforhold, natur typer og forurensning i norske havområder. Resultatene gjøres tilgjengelig på MAREANOs nettsider og visualiseres ved hjelp av kart.
<p>Knowing the type and distribution of seafloor sediments is crucial for many purp... more <p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p><p>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.</p>
Continental shelf sediments are places of both rapid organic carbon turnover and accumulation, wh... more Continental shelf sediments are places of both rapid organic carbon turnover and accumulation, while at the same 5 time increasingly subjected to human-induced disturbances. Recent research suggests that shelf sediments might have a role to play as a natural climate solution, e.g. by protecting the seafloor against human-induced disturbance. However, we have an incomplete understanding about the centres of organic carbon accumulation and storage on continental shelves. To better constrain the rate of accumulation and the mass of organic carbon that is stored in sediments, we developed and applied a spatial modelling framework that allows to estimate those quantities from sparse observations and predictor variables known 10 or suspected to influence the spatial patterns of these parameters. This paper presents spatial distribution patterns of organic carbon densities and accumulation rates in the North Sea and Skagerrak. We found that organic carbon stocks and accumulation rates are ...
An initiative is currently being taken by several Norwegian organizations to obtain funds to inte... more An initiative is currently being taken by several Norwegian organizations to obtain funds to intensify ongoing investigations on marine sea-floor mapping off Norway. Led by the Geological Survey of Norway and Institute of Marine Research, planning during the last two years has led to the inception of a large-scale mapping project entitled “MAREANO Marine Areal Database for the Norwegian Sea”. The investigation area covers 270 000 km2 of the shelf and deep sea off the central part of western Norway. It is a commercially important region for fisheries and the petroleum industry and includes the world’s largest system of cold-water coral reefs. The aim of MAREANO is to collect new as well as historical data elucidating the physical, chemical and biological characteristics of the seabed along the mid-Norwegian shelf and parts of the deeper Norwegian Sea. The project shall produce maps and/or provide information on seabed bathymetry, marine habitats, biological diversity and resources, m...
Today’s detailed swath bathymetry and backscatter obtained from multibeam sonar data, together wi... more Today’s detailed swath bathymetry and backscatter obtained from multibeam sonar data, together with advanced ground truthing tools have opened up new possibilities for the characterisation and mapping of marine habitats. This has also triggered the need for robust habitat classification systems. Incorporating data and preliminary results in classification schemes can be problematic if the schemes do not explicitly address high-resolution information of seabed morphology and texture. Regional character has been difficult using wellestablished classifications schemes, e.g. EUNIS (http:// eunis.eea.eu.int/index.jsp) which have problems coping with mixed substrates such as till normally found in glaciated areas (Rinde et al., 2004). This problem makes such classifications impractical to use in Norwegian waters, where glacial deposits dominate the seabed. Valentine et al. (2005) designed a habitat classification system for use along the repeatedly glaciated northeast coast of the USA and...
9 Abstract— The deep Norwegian Sea spans depths reaching to nearly 10 6000 m and an area of aroun... more 9 Abstract— The deep Norwegian Sea spans depths reaching to nearly 10 6000 m and an area of around 1 000 000 km extending both sides of 11 the rugged mid-Atlantic Ridge. Armed with coarse regional 12 bathymetric and oceanographic datasets the Norwegian national 13 offshore mapping programme, MAREANO, was tasked with 14 mapping ‘representative areas’ of this vast seabed terrain. Here we 15 introduce the planning process, guided by semi-automatic methods, 16 and present examples from newly acquired ship-borne multibeam 17 bathymetric mapping in this varied deep-sea terrain with a focus on 18 the use of these bathymetry data for onward substrate, 19 geomorphological, and habitat mapping. We discuss the challenges 20 of using these data with existing, lower quality, bathymetry data, as 21 well as highlighting some typical data artefacts which can limit the 22 calculation of meaningful terrain attributes and thereby their use in 23 geological and habitat mapping. 24
Uploads
Papers by TERJE THORSNES