ALBERT

Description: Summary of Ilse Seibold's vita Ilse Seibold, née Usbeck, was born May 8, 1925 in Breslau, Silesia, and went to school in Halle/Saale during WW2. She started her studies of geology and paleontology at the University of Halle and at the Humboldt University in Berlin, and later at the University of Tübingen, where she received her doctorate as micropaleontologist in 1951 with Otto Schindewolf as her supervisor. She remained active as productive scientist over many decades. In 1952, she married Dr. Eugen Seibold, who in 1958 became professor at Kiel University, founded one of Europe's most important institutes for marine geology, and later became president of the German Science Foundation (DFG), and subsequently of the European Science Foundation (ESF). Being a scientist herself Ilse Seibold soon evolved to a deeply reflective insider of geological sciences. She followed her husband during his scientific career from his appointments in Tübingen, Bonn, Karlsruhe, Kiel, to Bonn and Strasbourg/Freiburg i.Br. She accompanied Eugen on his sabbatical leave at Scripps Institution of Oceanography in La Jolla, CA. She participated in countless international scientific meetings. Together with Eugen she published many papers that document her independence and autonomy as scientist. She gained deep insights into the origins of the geosciences and their historical evolution, up to the ideas of fine arts. We are happy that she documented in her publications a broad range of her scientific and distinguished-humane impressions.

Description: Nowadays underwater vision systems are being widely applied in ocean research. However, the largest portion of the ocean - the deep sea - still remains mostly unexplored. Only relatively few image sets have been taken from the deep sea due to the physical limitations caused by technical challenges and enormous costs. Deep sea images are very different from the images taken in shallow waters and this area did not get much attention from the community. The shortage of deep sea images and the corresponding ground truth data for evaluation and training is becoming a bottleneck for the development of underwater computer vision methods. Thus, this paper presents a physical model-based image simulation solution, which uses an in-air texture and depth information as inputs, to generate underwater image sequences taken by robots in deep ocean scenarios. Different from shallow water conditions, artificial illumination plays a vital role in deep sea image formation as it strongly affects the scene appearance. Our radiometric image formation model considers both attenuation and scattering effects with co-moving spotlights in the dark. By detailed analysis and evaluation of the underwater image formation model, we propose a 3D lookup table structure in combination with a novel rendering strategy to improve simulation performance. This enables us to integrate an interactive deep sea robotic vision simulation in the Unmanned Underwater Vehicles simulator. To inspire further deep sea vision research by the community, we release the source code of our deep sea image converter to the public (https://www.geomar.de/en/omv-research/robotic-imaging-simulator).

Description: Since the sunlight only penetrates a few hundred meters into the ocean, deep-diving robots have to bring their own light sources for imaging the deep sea, e.g., to inspect hydrothermal vent fields. Such co-moving light sources mounted not very far from a camera introduce uneven illumination and dynamic patterns on seafloor structures but also illuminate particles in the water column and create scattered light in the illuminated volume in front of the camera. In this scenario, a key challenge for forward-looking robots inspecting vertical structures in complex terrain is to identify free space (water) for navigation. At the same time, visual SLAM and 3D reconstruction algorithms should only map rigid structures, but not get distracted by apparent patterns in the water, which often resulted in very noisy maps or 3D models with many artefacts. Both challenges, free space detection, and clean mapping could benefit from pre-segmenting the images before maneuvering or 3D reconstruction. We derive a training scheme that exploits depth maps of a reconstructed 3D model of a black smoker field in 1400 m water depth, resulting in a carefully selected, ground-truthed data set of 1000 images. Using this set, we compare the advantages and drawbacks of a classical Markov Random Field-based segmentation solution (graph cut) and a deep learning-based scheme (U-Net) to finding free space in forward-looking cameras in the deep ocean.

Description: Seagrass meadows deliver important ecosystem services such as nutrient cycling, enhanced biodiversity, and contribution to climate change mitigation and adaption through carbon sequestration and coastal protection. Seagrasses, however, are facing the impacts of ocean warming and marine heatwaves, which are altering their ecological structure and function. Shifts in species composition, mass mortality events, and loss of ecosystem complexity after sudden extreme climate events are increasingly common, weakening the ecosystem services they provide. In the west coast of Australia, Shark Bay holds between 0.7 and 2.4% of global seagrass extent (〉4300 km2), but in the austral summer of 2010/2011, the Ningaloo El Niño marine heatwave resulted in the collapse of ~1300 km2 of seagrass ecosystem extent. The loss of the seagrass canopy resulted in the erosion and the likely remineralization of ancient carbon stocks into 2–4 Tg CO2-eq over 6 years following seagrass loss, increasing emissions from land-use change in Australia by 4–8% per annum. Seagrass collapse at Shark Bay also impacted marine food webs, including dugongs, dolphins, cormorants, fish communities, and invertebrates. With increasing recurrence and intensity of marine heatwaves, seagrass resilience is being compromised, underlining the need to implement conservation strategies. Such strategies must precede irreversible climate change-driven tipping points in ecosystem functioning and collapse and result from synchronized efforts involving science, policy, and stakeholders. Management should aim to maintain or enhance the resilience of seagrasses, and using propagation material from heatwave-resistant meadows to restore impacted regions arises as a challenging but promising solution against climate change threats. Although scientific evidence points to severe impacts of extreme climate events on seagrass ecosystems, the occurrence of seagrass assemblages across the planet and the capacity of humans to modify the environment sheds some light on the capability of seagrasses to adapt to changing ecological niches.

Description: In deep water conditions, vision systems mounted on underwater robotic platforms require artificial light sources to illuminate the scene. The particular lighting configurations significantly influence the quality of the captured underwater images and can make their analysis much harder or easier. Nowadays, classical monolithic Xenon flashes are gradually being replaced by more flexible setups of multiple powerful LEDs. However, this raises the question of how to arrange these light sources, given different types of seawater and-depending-on different flying altitudes of the capture platforms. Hence, this paper presents a rendering based coarse-to-fine approach to optimize recent multi-light setups for underwater vehicles. It uses physical underwater light transport models and target ocean and mission parameters to simulate the underwater images as would be observed by a camera system with particular lighting setups. This paper proposes to systematically vary certain design parameters such as each LED’s orientation and analyses the rendered image properties (such as illuminated image area and light uniformity) to find optimal light configurations. We report first results on a real, ongoing AUV light design process for deep sea mission conditions.

Description: The vast majority of active volcanism that is located at plate boundaries can be easily explained by plate tectonic processes. Intraplate volcanism, which incorporates some of the smallest and largest volcanic events on Earth, cannot be successfully explained by a single process or model. The most volumetrically significant intraplate volcanic events are associated with the arrival of the head of a thermo-chemical anomaly rising from the deep mantle and impacting the base of the lithosphere. This event generates massive and short-lived magmatic activity over a wide area (up to 2000 km across), forming a large igneous province. A long-lived hotspot track can form over the mantle plume tail and is best illustrated by the formation of age progressive volcanic chains, such as the famous Hawaiian-Emperor chain. Millions of smaller solitary volcanic edifices, non-age progressive volcanic chains and provinces, on the other hand, have other potential mechanisms of origin. Potential models comprise decompression melting due to lithospheric extension, destabilization of fusible lithologies in the lithospheric mantle, small scale sub-lithospheric convection, or lithospheric delamination.

Description: Landslide deposits offshore many volcanic islands provide evidence of catastrophic lateral collapses. These deposits span a larger volume range than their continental equivalents, and can generate devastating tsunamis. All historical volcanic-island lateral collapses have occurred in arc settings, and have been characterised by rapid failure and efficient tsunami generation. The varied morphology of their deposits is influenced both by lithological properties and the nature of the substrate. Many deposits show evidence of extensive seafloor erosion and transformation into debris flows, and the propagation of frontally-confined sediment deformation beyond and beneath the primary deposit. Mobilised volumes can far exceed that of the initial failure, and accurate deposit interpretation requires internal geophysical imaging and sampling. Around intraplate ocean-island volcanoes, multi-unit turbidites suggest that lateral collapses may occur in discrete stages; although this would reduce their overall tsunamigenic potential, the volumes of individual stages of collapse remain very large. Numerical models of both landslide and tsunami processes in ocean-island settings are difficult to test, and the smaller collapses that typify island arcs are an important focus of research due to their higher global frequency, availability of direct failure and tsunami observations, and a need to better understand the signals of incipient collapse to develop approaches for tsunami hazard mitigation.