The Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (AWI) is a member of the Helmholtz Association (HGF) and funded by federal and state government. AWI focuses on polar and marine research in a variety of disciplines such as biology, oceanography, geology, geochemistry and geophysics thus allowing multidisciplinary approaches to scientific goals.
Scientist in AI-based emulator for High-Resolution Sea Ice Modeling (f/m/d)

BACKGROUND

The Climate Dynamics section of the Alfred Wegener Institute Helmholtz Centre for Polar and Marine Research (AWI) in Bremerhaven, Germany, is seeking applications for a position focused on development of AI-based emulator for high-resolution sea ice modeling in climate research.

The position includes the following responsibilities:

• Developing an AI-based sea ice model emulator based on high-resolution FESOM2 simulations (https://github.com/FESOM/fesom2) and potentially incorporating observational data
• Adapting the AI-based sea ice emulator for various downstream tasks
• Extending the emulator to include advanced sea ice thermodynamics
• Publishing the results of these developments in scientific literature

You will collaborate with leading European climate modelers, computational scientists, and software engineers to develop Digital Twins of the climate system, supporting climate change adaptation efforts. The project has a strong AI component, and there will be close cooperation with other national and international projects developing AI solutions for geoscience.

YOUR TASKS

You will be part of a large, collaborative team working on the development of the IFS-FESOM system, a next-generation kilometer-scale climate model that is a key component of the Destination Earth Climate Digital Twin initiative and several other national and international projects.
Your primary responsibilities will include the development of an AI-based sea ice emulator using data generated by the kilometer-scale setup of the unstructured ocean-sea ice model FESOM2. Observational data may also be considered as an additional source. The emulator will then be fine-tuned for a range of downstream tasks, including: short-term forecasting, temporal interpolation between time steps, regional downscaling to higher resolutions, observation-constrained model correction.
The emulator will also be tested under climate simulation scenarios. Furthermore, additional variables provided by ICEPACK thermodynamics will be integrated into the emulator to assess their impact on the emulator’s accuracy and solution quality.
These tasks will be carried out in close collaboration and continuous dialogue with model developers and another project member responsible for development of the sea ice model. You will be expected to publish the results of these developments in the scientific literature.

The position includes the following responsibilities:

• Developing an AI-based sea ice model emulator based on high-resolution FESOM2 simulations (https://github.com/FESOM/fesom2) and potentially incorporating observational data
• Adapting the AI-based sea ice emulator for various downstream tasks
• Extending the emulator to include advanced sea ice thermodynamics
• Publishing the results of these developments in scientific literatur