TOWARDS DETECTING FLOATING OBJECTS ON A GLOBAL SCALE WITH LEARNED SPATIAL FEATURES USING SENTINEL 2

Marine litter is a growing problem that has been attracting attention and raising concerns over the last years. Significant quantities of plastic can be found in the oceans due to the unfiltered discharge of waste into rivers, poor waste management, or lost fishing nets. The floating elements drift on the surface of water bodies and can be aggregated by processes, such as river plumes, windrows, oceanic fronts, or currents. In this paper, we focus on detecting big patches of floating objects that can contain plastic as well as other materials with optical Sentinel 2 data. In contrast to previous work that focuses on pixel-wise spectral responses of some bands, we employ a deep learning predictor that learns the spatial characteristics of floating objects. Along with this work, we provide a hand-labeled Sentinel 2 dataset of floating objects on the sea surface and other water bodies such as lakes together with pre-trained deep learning models. Our experiments demonstrate that harnessing the spatial patterns learned with a CNN is advantageous over pixel-wise classifications that use hand-crafted features. We further provide an analysis of the categories of floating objects that we captured while labeling the dataset and analyze the feature importance for the CNN predictions. Finally, we outline the limitations of trained CNN on several systematic failure cases that we would like to address in future work by increasing the diversity in the dataset and tackling the domain shift between regions and satellite acquisitions. The dataset introduced in this work is the first to provide public large-scale data for floating litter detection and we hope it will give more insights into developing techniques for floating litter detection and classification. Source code and data are available at https://github.com/ESA-PhiLab/floatingobjects.

Hydrographic shifts south of Australia over the last deglaciation and possible interhemispheric linkages

Northern and southern hemispheric influences—particularly changes in Southern Hemisphere westerly winds (SSW) and Southern Ocean ventilation—triggered the stepwise atmospheric CO2 increase that accompanied the last deglaciation. One approach for gaining potential insights into past changes in SWW/CO2 upwelling is to reconstruct the positions of the northern oceanic fronts associated with the Antarctic Circumpolar Current. Using two deep-sea cores located ~600 km apart off the southern coast of Australia, we detail oceanic changes from ~23 to 6 ka using foraminifer faunal and biomarker alkenone records. Our results indicate a tight coupling between hydrographic and related frontal displacements offshore South Australia (and by analogy, possibly the entire Southern Ocean) and Northern Hemisphere (NH) climate that may help confirm previous hypotheses that the westerlies play a critical role in modulating CO2 uptake and release from the Southern Ocean on millennial and potentially even centennial timescales. The intensity and extent of the northward displacements of the Subtropical Front following well-known NH cold events seem to decrease with progressing NH ice sheet deglaciation and parallel a weakening NH temperature response and amplitude of Intertropical Convergence Zone shifts. In addition, an exceptional poleward shift of Southern Hemisphere fronts occurs during the NH Heinrich Stadial 1. This event was likely facilitated by the NH ice maximum and acted as a coup-de-grâce for glacial ocean stratification and its high CO2 capacitance. Thus, through its influence on the global atmosphere and on ocean mixing, “excessive” NH glaciation could have triggered its own demise by facilitating the destratification of the glacial ocean CO2 state.

Maintenance of mid-latitude oceanic fronts by mesoscale eddies

Oceanic fronts associated with strong western boundary current extensions vent a vast amount of heat into the atmosphere, anchoring mid-latitude storm tracks and facilitating ocean carbon sequestration. However, it remains unclear how the surface heat reservoir is replenished by ocean processes to sustain the atmospheric heat uptake. Using high-resolution climate simulations, we find that the vertical heat transport by ocean mesoscale eddies acts as an important heat supplier to the surface ocean in frontal regions. This vertical eddy heat transport is not accounted for by the prevailing inviscid and adiabatic ocean dynamical theories such as baroclinic instability and frontogenesis but is tightly related to the atmospheric forcing. Strong surface cooling associated with intense winds in winter promotes turbulent mixing in the mixed layer, destructing the vertical shear of mesoscale eddies. The restoring of vertical shear induces an ageostrophic secondary circulation transporting heat from the subsurface to surface ocean.