PhD Scholarship in Causal-Informed ML for Climate Modelling — DTU Compute

About the role

We invite applications for a fully funded PhD scholarship (3-year employment) at DTU Compute, part of the research project “IcyAlert – Intelligent Climate Early Warning Alert for Arctic Ice-Free Summers”. The project is funded under the NNF Grand AI Challenge programme and aims to develop causal-informed machine learning methods for climate modelling. You will work closely with collaborators at the Danish Meteorological Institute (DMI) and the Royal Meteorological Institute of Belgium (RMI) to advance neural network-based predictions of Arctic sea ice.

Project background

Arctic summer sea ice extent has declined by nearly 13% per decade since 1979, with a 90% reduction of older ice. These changes pose increasing risks to ecosystems, food security, and navigation. The IcyAlert project combines dynamical models, satellite data, causal analysis, and explainable AI to identify critical thresholds, enhance prediction accuracy, and generate automated alerts for seasonal Arctic and extratropical climate events. The project will use the Gefion supercomputer for analysing complex datasets and training large-scale predictive models.

Responsibilities

• Develop causal-informed machine learning methods for Arctic sea-ice predictions at multi-seasonal timescale.
• Work with datasets including CMIP6, ERA5, and CARRA2.
• Quantify causal links between winter climate drivers (November–April) and summer Arctic sea ice area (May–October).
• Develop causal-informed AI/ML models for an ice-free Arctic predictor.
• Generate probabilistic predictions of ice-free conditions from 2030 to 2050.

Required qualifications

• Strong background or interest in causal modelling (mathematics, physics, or computer science background).
• Demonstrated experience in machine learning, including exposure to graph modelling and diffusion models.
• Experience implementing ML methods in Python using PyTorch or Flax.
• Familiarity with managing larger code bases and training models on HPC systems.
• High level of motivation and creative problem-solving skills.
• Excellent communication and writing skills in English.
• A two-year master’s degree (120 ECTS points) or equivalent academic level.

Preferred qualifications

Familiarity with climate modelling is considered a strong plus.

What we offer

• A vibrant interdisciplinary research environment at DTU, a leading technical university globally recognised for research and innovation.
• Close collaboration with climate and AI experts from DMI and RMI.
• Access to state-of-the-art computing resources including the Gefion supercomputer.
• Employment and salary in accordance with the collective agreement with the Danish Confederation of Professional Associations. Allowance is agreed with the relevant union.
• Full-time position for 3 years, starting 1 November 2026 or by mutual agreement.
• For international applicants: free monthly “PhD relocation to Denmark and startup Zoom seminar” covering practical matters of moving and working at DTU.

About DTU Compute

DTU Compute – Department of Applied Mathematics and Computer Science – is an internationally recognised academic environment with over 400 employees across 10 research sections. The department covers digital technologies including artificial intelligence, machine learning, data science, computer engineering, cybersecurity, and human-computer interaction. The research is rooted in basic science with a strong ethical, human, and sustainable approach, contributing to health, green transition, energy supply, and life science. DTU Compute values diversity, inclusion, and a flexible work-life balance.

How to apply

Your complete online application must be submitted by 9 August 2026 (23:59 Danish time). Applications must be compiled into a single PDF file (in English) containing:

• A letter motivating the application (cover letter)
• Curriculum vitae
• Grade transcripts and BSc/MSc diploma including an official description of the grading scale

You may apply before obtaining your master’s degree but cannot begin the position before receiving it. Applications received after the deadline will not be considered.

All interested candidates irrespective of age, gender, disability, race, religion, or ethnic background are encouraged to apply. As DTU works with research in critical technology subject to special rules for security and export control, open-source background checks may be conducted on qualified candidates.

Assessment of applicants will be made by Associate Professor Tommy Sonne Alstrøm and colleagues in the IcyAlert project. For further information, you may contact Tommy Sonne Alstrøm ([email protected]).

Original vacancy