Extreme Weather Storylines via Generative AI
This is a fully-funded 4-year PhD position based in the HetSys Centre for Doctoral Training at the University of Warwick.
Project outline
Generative machine learning techniques, as known for example from large language models or image generation, have recently also taken the atmospheric dynamics and weather forecast communities by storm.
At the same time, due to antropogenetic climate change, society faces elevated risk from extreme weather events such as flash floods, heatwaves, or tropical cyclones.
This project aims to combine generative weather prediction models with rare event techniques to generate plausible trajectories of the weather system exhibiting extreme weather events.
The ECMWF machine learning division has agreed to provide data and support the project in adapting their in-house intermediate-resolution generative weather model.
Supervisors
Primary: Dr Tobias Grafke (Maths)
Prof. Sandra Chapman (Physics)
Project Partner: ECMWF Machine Learning Group
The main aim of the project is to extend the ECMWF generative weather forecast model with a conditional path sampling algorithms to produce a versatile software stack for producing statistically faithful weather trajectories leading to extreme weather events.
The potential outcomes of the project encompass both a concrete software application to generate extreme weather scenarios on demand, as well as publications in theoretical/numerical scientific journals about techniques to combine generative machine learning models with rare event algorithms.
The primary skill a potential student on this project will develop is hands-on experience in developing, implementing, training, and applying generative machine learning algorithms to real-world problems in environmental science and atmosphere dynamics. Additionally, a student working on this project will acquire a broad range of skills, ranging from rigorous mathematical foundations in probability theory and stochastic analysis, over algorithmic and numerical knowledge concerning rare event sampling algorithms and software engineering practice, to skills in geophysical and environmental sciences concerning extreme weather events, meteorological prediction and atmosphere/ocean dynamics.
These skills position you for careers in AI research, computational materials science, national laboratories, tech industry or academic research. The HetSys training provides a foundation for these skills through dedicated courses and cohort activities.
We require at least a II(i) honours degree at BSc or an integrated masters degree (e.g. MPhys, MChem, MSci, MEng etc.) in a physical sciences, mathematics or engineering discipline. We do not accept applications from existing PhD holders.
If you are an overseas candidate please check here that you hold the equivalent grades before applying.
For postgraduate study in HetSys, the term “overseas” or “international” student refers to anyone who does not qualify for UK home fee status. This includes applicants from the European Union (EU), European Economic Area (EEA), and Switzerland, unless they hold settled or pre-settled status under the UK’s EU Settlement Scheme.
If you are a European applicant without UK residency or immigration status that qualifies you for home fees, you will be classified as an overseas student.