PhD Researcher in Machine Learning for Climate Modelling

Job summary

This position offers a fully funded PhD scholarship focused on causal-informed machine learning methods for climate modelling, specifically aimed at predicting Arctic sea-ice conditions. The role is part of the IcyAlert project, which seeks to enhance climate predictions and support sustainable solutions in response to climate change.

Qualifications

• Strong background or interest in causal modelling (Math/Physics/Computer Science).
• Demonstrated experience in machine learning, including exposure to graph modelling and diffusion models.
• Proficiency in implementing machine learning methods in Python using frameworks like Pytorch or Flax.
• Familiarity with managing larger code bases and training models on HPC systems.
• Knowledge of climate modelling is advantageous.
• High motivation and creative problem-solving abilities.
• Excellent communication and writing skills in English.

Responsibilities

• Develop causal-informed machine learning methods for predicting Arctic sea-ice at multi-seasonal timescales.
• Work with datasets such as CMIP6, ERA5, and CERRA2.
• Conduct three subprojects: quantifying causal links between winter climate drivers and summer Arctic sea ice area, developing AI/ML models for ice-free Arctic predictions, and generating probabilistic predictions for ice-free conditions from 2030 to 2050.

Education

• A two-year master's degree (120 ECTS points) or an equivalent academic level is required.

Tools

• Python, Pytorch, Flax, and high-performance computing systems.