Description: A special focus in data mining is to identify agglomerations of data points in spatial or spatio-temporal databases. Multiple applications have been presented to make use of such clustering algorithms. However, applications exist, where not only dense areas have to be identified, but also requirements regarding the correlation of the cluster to a specific shape must be met, i.e. circles. This is the case for eddy detection in marine science, where eddies are not only specified by their density, but also their circular-shaped rotation. Traditional clustering algorithms lack the ability to take such aspects into account. In this paper, we introduce Vortex Correlation Clustering which aims to identify those correlated groups of objects oriented along a vortex. This can be achieved by adapting the Circle Hough Transformation, already known from image analysis. The presented adaptations not only allow to cluster objects depending on their location next to each other, but also allows to take the orientation of individual objects into considerations. This allows for a more precise clustering of objects. A multi-step approach allows to analyze and aggregate cluster candidates, to also include final clusters, which do not perfectly satisfy the shape condition. We evaluate our approach upon a real world application, to cluster particle simulations composing such shapes. Our approach outperforms comparable methods of clustering for this application both in terms of effectiveness and efficiency.

Description: Mesoscale eddies can be strengthened by the absorption of submesoscale eddies resulting from mixed-layer baroclinic instabilities. This is shown for mesoscale eddies in the Agulhas Current system by investigating the kinetic energy cascade with a spectral and a coarse-graining approach in two model simulations of the Agulhas region. One simulation resolves mixed-layer baroclinic instabilities and one does not. When mixed-layer baroclinic instabilities are included, the largest submesoscale near-surface fluxes occur in winter-time in regions of strong mesoscale activity for upscale as well as downscale directions. The forward cascade at the smallest resolved scales occurs mainly in frontogenetic regions in the upper 30 m of the water column. In the Agulhas ring path, the forward cascade changes to an inverse cascade at a typical scale of mixed-layer eddies (15 km). At the same scale, the largest sources of the upscale flux occur. After the winter, the maximum of the upscale flux shifts to larger scales. Depending on the region, the kinetic energy reaches the mesoscales in spring or early summer aligned with the maximum of mesoscale kinetic energy. This indicates the importance of submesoscale flows for the mesoscale seasonal cycle. A case study shows that the underlying process is the mesoscale absorption of mixed-layer eddies. When mixed-layer baroclinic instabilities are not included in the simulation, the open-ocean upscale cascade in the Agulhas ring path is almost absent. This contributes to a 20 %-reduction of surface kinetic energy at mesoscales larger than 100 km when submesoscale dynamics are not resolved by the model.

Description: Mining spatio-temporal correlation patterns for traffic prediction is a well-studied field. However, most approaches are based on the assumption of the availability of and accessibility to a sufficiently dense data source, which is rather the rare case in reality. Traffic sensors in road networks are generally highly sparse in their distribution: fleet-based traffic sensing is sparse in space but also sparse in time. There are also other traffic application, besides road traffic, like moving objects in the marine space, where observations are sparsely and arbitrarily distributed in space. In this paper, we tackle the problem of traffic prediction on sparse and spatially irregular and non-deterministic traffic observations. We draw a border between imputations and this work as we consider high sparsity rates and no fixed sensor locations. We advance correlation mining methods with a Sparse Unstructured Spatio Temporal Reconstruction (SUSTeR) framework that reconstructs traffic states from sparse non-stationary observations. For the prediction the framework creates a hidden context traffic state which is enriched in a residual fashion with each observation. Such an assimilated hidden traffic state can be used by existing traffic prediction methods to predict future traffic states. We query these states with query locations from the spatial domain.

Description: Marine heatwaves along the coast ofWestern Australia, referred to as Ningaloo Niño, have had dramatic impacts on the ecosystem in the recent decade. A number of local and remote forcing mechanisms have been put forward, however little is known about the depth structure of such temperature extremes. Utilizing an eddy-active global Ocean General Circulation Model, Ningaloo Niño and the corresponding cold Ningaloo Niña events are investigated between 1958-2016, with focus on their depth structure. The relative roles of buoyancy and wind forcing are inferred from sensitivity experiments. Composites reveal a strong symmetry between cold and warm events in their vertical structure and associated large-scale spatial patterns. Temperature anomalies are largest at the surface, where buoyancy forcing is dominant and extend down to 300m depth (or deeper), with wind forcing being the main driver. Large-scale subsurface anomalies arise from a vertical modulation of the thermocline, extending from the western Pacific into the tropical eastern Indian Ocean. The strongest Ningaloo Niños in 2000 and 2011 are unprecedented compound events, where long-lasting high temperatures are accompanied by extreme freshening, which emerges in association with La Niñas, more common and persistent during the negative phase of the Interdecadal Pacific Oscillation. It is shown that Ningaloo Niños during La Nina phases have a distinctively deeper reach and are associated with a strengthening of the Leeuwin Current, while events during El Niño are limited to the surface layer temperatures, likely driven by local atmosphere-ocean feedbacks, without a clear imprint on salinity and velocity.