Graph Neural Networks for Ice Sheet Modeling and Sea Level Rise Projections
(ref. BAP-2025-483)
Laatst aangepast: 17/07/25
Laatst aangepast: 17/07/25
Are you excited about using cutting-edge AI to tackle one of the most pressing challenges of our time - sea level rise? Do you want to work at the intersection of deep learning, climate science, and computational modeling? We invite applications for a fully funded PhD position at KU Leuven to develop graph-based AI emulators for Antarctic ice sheet modeling.This interdisciplinary project combines graph neural networks, high-performance computing, and glaciology to improve projections of Antarctica’s contribution to sea level rise. The PhD position is part of the FWO-funded IceGraph project, a collaboration between KU Leuven (Prof. Stef Lhermitte) and ULB (Prof. Frank Pattyn), with additional collaborations with IGE (France), TUDelft (Netherlands) and the eScience center (Netherlands). The project is embedded in the Leuven.AI initiative and offers opportunities for international collaboration and research visits.
Project
Antarctica’s future contribution to sea level rise remains one of the largest uncertainties in climate projections. A key challenge lies in simulating the complex processes of ice damage and calving, which are computationally expensive and difficult to scale. The IceGraph project aims to overcome this by developing IceGraph, a novel graph neural network (GNN) emulator based on MeshGraphNets, capable of learning the dynamics of high-resolution ice sheet models on irregular and adaptive meshes.
The PhD candidate will focus on developing and training this GNN-based emulator to replace traditional physics-based models, enabling faster and more scalable simulations of Antarctic ice dynamics. This will support improved projections of sea level rise and inform global climate adaptation strategies.
Responsibilities:
- Data Preparation & Preprocessing: Translate high-resolution ice sheet model outputs (e.g., from ISMIP6) into graph-based data structures suitable for IceGraph.
- Model Development: Design, implement, and train the IceGraph emulator using state-of-the-art GNN architectures to simulate ice flow dynamics.
- Validation & Testing: Evaluate the emulator’s performance against traditional models and satellite observations.
- Integration & Application: Integrate the emulator with the Kori-ULB ice sheet model to assess damage-driven mass loss and uncertainty in sea level rise projections.
- Dissemination: Publish results in peer-reviewed journals and present at international conferences.
- Collaboration: Work closely with interdisciplinary teams at KU Leuven, ULB, and IGE, and co-supervise MSc students.
- Teaching support: co-support teaching activities within the project.
Profile
We are looking for a highly motivated candidate with:
- A Master’s degree in computer science, AI, physics, geoscience, or a related field.
- Strong programming skills (e.g., Python) and ideally experience with deep learning frameworks (e.g., PyTorch, TensorFlow).
- Interest in graph neural networks, scientific machine learning, or physics-informed AI.
- A solid background in numerical modeling, computational physics, or climate science is a plus.
- Excellent analytical and problem-solving skills.
- Strong written and verbal communication skills in English.
- Willingness to engage in interdisciplinary research and international collaboration.
Candidates with a strong analytical background but limited experience in GNNs or glaciology are encouraged to apply if they are eager to learn.
Offer
- A fully funded 4-year PhD position in a dynamic, internationally recognized research environment.
- Supervision by leading experts in AI, glaciology, and remote sensing.
- Access to high-performance computing resources and cutting-edge AI tools.
- Opportunities for international collaboration and research visits (e.g., to international partners).
- Support for publishing in top-tier journals and attending international conferences.
- A vibrant academic community in Leuven, a historic university town in the heart of Europe.
Earthmapps home page
Interested?
Interested candidates should submit the following documents:
CV (including transcripts and, if available, a link to the MSc thesis).- Motivation Letter, including your earliest possible start date.
- References: Names and contact details of two referees (no letters required at this stage).
For more information please contact Prof. dr. Stef Lhermitte, tel.: +32 16 37 25 61, mail: stef.lhermitte@kuleuven.be.
You can apply for this job no later than August 15, 2025 via the online application tool
KU Leuven strives for an inclusive, respectful and socially safe environment. We embrace diversity among individuals and groups as an asset. Open dialogue and differences in perspective are essential for an ambitious research and educational environment. In our commitment to equal opportunity, we recognize the consequences of historical inequalities. We do not accept any form of discrimination based on, but not limited to, gender identity and expression, sexual orientation, age, ethnic or national background, skin colour, religious and philosophical diversity, neurodivergence, employment disability, health, or socioeconomic status. For questions about accessibility or support offered, we are happy to assist you at this email address.
Heb je een vraag over de online sollicitatieprocedure? Raadpleeg onze veelgestelde vragen of stuur een e-mail naar solliciteren@kuleuven.be
av_timer
Tewerkstellingspercentage: Voltijds
location_city
Locatie: Leuven
timer
Solliciteren tot en met:15/08/2025 23:59 CET
bookmarks
Tags: Bio-ingenieurswetenschappen