Prof. Dr. Stephan Juricke

So called mesoscale ocean eddies of the order of 10-100km are approximately circular motions in the ocean that are of fundamental importance for the transport of tracers, heat, and the transfer of energy. They carry a large part of the kinetic energy in the ocean and interact with the mean current, smaller scale structures and the overlying atmosphere. Their spatial distribution and intensity is strongly linked the strength of the mean currents which is why they can be predominantly found in the western boundary currents and the Southern Ocean.

We investigate several aspects of eddies:

• Parameterizations: For ocean models these eddies often need to be parameterized, i.e. approximated via specific equations and methods, in cases where the grid resolution is too coarse to explicitly simulate them. We develop new ways of parameterizing eddies.
• Energy transfers: Eddies transfer energy between smaller and large scales and vice versa. They also are associated to instabilities that transfer energy between potential and kinetic energy. We develop tools to investigate how these energy transfers look like and under which conditions they are most pronounced.
• Large scale circulation: Eddies play a crucial role in the large-scale ocean circulation and climate in general. We investigate how the presence or absence of eddies affects global climate, using lower and high-resolution ocean simulations where eddies are either mostly absent or can be accurately represented.

The following current projects are associated to this topic:

• DFG TRR181 “Energy Transfers in Atmosphere and Ocean”, subproject M3 “Towards Consistent Subgrid Momentum Closures” (until end of June 2024) 
• BMBF WarmWorld project “Enabling Lagrangian Particle Tracking for High-resolution and unstructurEd meshes (ELPHE)” (starting June 2024)
• MarDATA Helmholtz School for Marine Data Science project “Data based Probabilistic Parameter Estimation for Ocean and Earth System Models” (until end of May 2025)

The exchange of material and energy between the atmosphere and ocean occurs on a range of temporal and spatial scales. This interaction does not only define the climate on a regional and global scale, it also affects the weather and the occurrence of extreme events such as heat waves and extreme precipitation.

We focus on the transfer of different forms of energy between the ocean and the atmosphere to identify on which temporal and spatial scales the ocean predominantly drives the atmosphere circulation and vice versa.

The following current project is associated to this topic:

• DFG TRR181 “Energy Transfers in Atmosphere and Ocean”, subproject L4 “Energy-Consistent Ocean-Atmosphere Coupling” (until end of June 2024)

Complex ocean and climate models are used to predict the behaviour of the climate system from days to decades and even centuries. Such models are based on discretized physical equations of motion and mass exchanges and need sophisticated numerical methods to be solved. New methodologies from data science including machine learning are explored to improve our ability to simulate ocean and climate.

We investigate several aspects in the context of ocean modelling:

• High-resolution modelling: Going to km-scale resolutions for the ocean is necessary to resolve comparatively small scale processes such as submesoscale eddies and filament which play an important role in the transfer of energy with the ocean and in atmosphere-ocean interactions. We employ high-resolution models to investigate such transfers in selected regions such as the Atlantic or Southern Ocean.
• Numerical methods: Numerical discretization methods and parameterizations are essential building blocks of any dynamical ocean model. We develop new methodologies to improve the models and their efficiency, to ultimately allow for better predictions.
• Machine learning and data science: As new computing science methods become available and the amount of observational and model data rapidly increases, the classical view on how to develop ocean models is challenged. We develop new tools and model components based on machine learning and advanced data science techniques to improve our model simulations.
• Incorporation of model uncertainty: Random as well as systematic errors are part of any model. One needs to be able to quantify these errors to provide meaningful predictions that include a measure of uncertainty. We investigate the sources for model errors and develop methodologies, including stochastic parameterizations, to both reduce errors as well as quantify them and identify how they affect our predictions.

The following current projects are associated to this topic:

• DFG TRR181 “Energy Transfers in Atmosphere and Ocean” (until end of June 2024)
• BMBF WarmWorld project “Enabling Lagrangian Particle Tracking for High-resolution and unstructurEd meshes (ELPHE)” (starting June 2024)
• MarDATA Helmholtz School for Marine Data Science project “Data based Probabilistic Parameter Estimation for Ocean and Earth System Models” (until end of May 2025)