| 1. |
- Aguzzi, Jacopo, et al.
(författare)
-
Developing technological synergies between deep-sea and space research
- 2022
-
Ingår i: Elementa. - 2325-1026. ; 10
-
Tidskriftsartikel (refereegranskat)abstract
- Recent advances in robotic design, autonomy and sensor integration create solutions for the exploration of deep-sea environments, transferable to the oceans of icy moons. Marine platforms do not yet have the mission autonomy capacity of their space counterparts (e.g., the state of the art Mars Perseverance rover mission), although different levels of autonomous navigation and mapping, as well as sampling, are an extant capability. In this setting their increasingly biomimicked designs may allow access to complex environmental scenarios, with novel, highly-integrated life-detecting, oceanographic and geochemical sensor packages. Here, we lay an outlook for the upcoming advances in deep-sea robotics through synergies with space technologies within three major research areas: biomimetic structure and propulsion (including power storage and generation), artificial intelligence and cooperative networks, and life-detecting instrument design. New morphological and material designs, with miniaturized and more diffuse sensor packages, will advance robotic sensing systems. Artificial intelligence algorithms controlling navigation and communications will allow the further development of the behavioral biomimicking by cooperating networks. Solutions will have to be tested within infrastructural networks of cabled observatories, neutrino telescopes, and off-shore industry sites with agendas and modalities that are beyond the scope of our work, but could draw inspiration on the proposed examples for the operational combination of fixed and mobile platforms.
|
|
| 2. |
- Dorschel, Boris, et al.
(författare)
-
The International Bathymetric Chart of the Southern Ocean Version 2
- 2022
-
Ingår i: Scientific Data. - : Springer Science and Business Media LLC. - 2052-4463. ; 9:1
-
Tidskriftsartikel (refereegranskat)abstract
- The Southern Ocean surrounding Antarctica is a region that is key to a range of climatic and oceanographic processes with worldwide effects, and is characterised by high biological productivity and biodiversity. Since 2013, the International Bathymetric Chart of the Southern Ocean (IBCSO) has represented the most comprehensive compilation of bathymetry for the Southern Ocean south of 60 degrees S. Recently, the IBCSO Project has combined its efforts with the Nippon Foundation - GEBCO Seabed 2030 Project supporting the goal of mapping the world's oceans by 2030. New datasets initiated a second version of IBCSO (IBCSO v2). This version extends to 50 degrees S (covering approximately 2.4 times the area of seafloor of the previous version) including the gateways of the Antarctic Circumpolar Current and the Antarctic circumpolar frontal systems. Due to increased (multibeam) data coverage, IBCSO v2 significantly improves the overall representation of the Southern Ocean seafloor and resolves many submarine landforms in more detail. This makes IBCSO v2 the most authoritative seafloor map of the area south of 50 degrees S.
|
|
| 3. |
- Purser, Autun, et al.
(författare)
-
The Applicability of Machine-Learning Algorithms for Automated Detection and Classification of Cold-Water Coral Habitats from Video Transect Data
- 2008
-
Ingår i: 4th International Symposium on Deep-Sea Corals, Wellington, New Zealand, December 2008.
-
Konferensbidrag (refereegranskat)abstract
- In recent years, Cold-Water Coral (CWC) ecosystems have been found throughout the world oceans. Reef-building corals, such as the Scleractinian Lophelia pertusa can build large, complex three-dimensional structures providing habitat niches for many other species. The great majority of these reef structures are found at depths exceeding 50m. Therefore, reef health assessments and community surveys cannot be carried using the same methodologies employed at shallow tropical coral reefs. Remotely operated vehicles (ROV’s), submersibles and towed camera sledges are used to obtain image transects of deep CWC reefs. However, the traditional manual analysis of the transects is both time-consuming and labour-intensive. Here, we used video transect data collected by a towed camera sledge and ROV from a selection of Norwegian CWC reefs to test the applicability of machine-learning algorithms to process image data. Individual frames extracted from the video stream were examined by a number of CWC experts and organisms labelled, using the novel Web 2.0 ‘Biigle’ (Bielefeld Image Indexing and Graphical Labelling Environment) interface. Additional annotations were made of areas of notable substrate types, such as ‘coral rubble’ regions. For each image a 30-dimensional texture feature was computed based on Gabor-jet filter responses. These numerical feature vectors were used to train a hierarchically organized hyperbolic self-organizing map (H2SOM). The unsupervised machine learning approach computes a “coral feature map” which is automatically annotated with the expert labels obtained from Biigle. The combination of expert labels and machine learning allows the detection and classification of selected organisms and substrates throughout the unlabelled remainder of the image data. The applicability of this approach for quick location and classification of organisms and habitat types in CWC images is discussed. The usefulness of the approach in long-term environmental monitoring schemes of CWC habitats (such as those under threat from fishery activity) is also addressed.
|
|
| 4. |
- Vornsand, Ina, et al.
(författare)
-
Short and decadal impacts of seafloor physical perturbation on the abundances of Lebensspuren 'traces of life' in the Peru Basin manganese nodule province
- 2024
-
Ingår i: MARINE BIODIVERSITY. - 1867-1616 .- 1867-1624. ; 54:1
-
Tidskriftsartikel (refereegranskat)abstract
- Interest in deep-sea mining for polymetallic nodules as an alternative source to onshore mines for various high-technology metals has risen in recent years, as demands and costs have increased. The need for studies to assess its short- and long-term consequences on polymetallic nodule ecosystems is therefore also increasingly prescient. Recent image-based expedition studies have described the temporal impacts on epi-/megafauna seafloor communities across these ecosystems at particular points in time. However, these studies have failed to capture information on large infauna within the sediments or give information on potential transient and temporally limited users of these areas, such as mobile surface deposit feeders or fauna responding to bloom events or food fall depositions. This study uses data from the Peru Basin polymetallic nodule province, where the seafloor was previously disturbed with a plough harrow in 1989 and with an epibenthic sled (EBS) in 2015, to simulate two contrasting possible impact forms of mining disturbance. To try and address the shortfall on information on transient epifauna and infauna use of these various disturbed and undisturbed areas of nodule-rich seafloor, images collected 6 months after the 2015 disturbance event were inspected and all Lebensspuren, 'traces of life', were characterized by type (epi- or infauna tracemakers, as well as forming fauna species where possible), along with whether they occurred on undisturbed seafloor or regions disturbed in 1989 or 2015. The results show that epi- and endobenthic Lebensspuren were at least 50% less abundant across both the ploughed and EBS disturbed seafloors. This indicates that even 26 years after disturbance, sediment use by fauna may remain depressed across these areas.
|
|
