PhD on Conditional Generative Modelling of Local High-Impact Events under Structured Scenarios

PhD on Conditional Generative Modelling of Local High-Impact Events under Structured Scenarios 100%, 4 years, Bern

UniBE is ambitious. With us, you deliver top performance and test your skills against the best.

Research and Collaborative Environment

The PhD student will be hosted at the Institute of Mathematical Statistics and Actuarial Science (IMSV) of the University of Bern, joining the Uncertainty Quantification and Spatial Statistics group led by Prof. David Ginsbourger. IMSV is an engaged, open-minded institute where mathematical statistics and probability theory meet data and models arising from various scientific and societal challenges.

The position is part of the NCCR CLIM+ programme on Climate Extremes and Society, funded by the Swiss National Science Foundation (SNSF). More specifically, the PhD project will contribute to the work package “Local Risks: Impacts and Adaptation” and involve participation in tasks covering topics such as the downscaling and debiasing of input data for hydrological models. Regular interactions and potential collaborations with the teams of Prof. Manuela Brunner (ETH Zurich, WSL), Dr. Christian Grams (MeteoSwiss), Prof. Michael Lehning (EPFL Valais Wallis, WSL), and Prof. Olivia Romppainen-Martius (University of Bern) are planned.

The NCCR CLIM+ supports Switzerland in the transformation towards a climate-resilient society. Together with stakeholders, the interdisciplinary NCCR CLIM+ research community tackles unexplored solution spaces to the climate crisis and develops a blueprint for actionable climate research worldwide. NCCR CLIM+ broadly communicates and shares knowledge, and trains a new generation of experts with the necessary domain and transdisciplinary knowledge.

The NCCR CLIM+ strives to implement equal opportunity at hiring. At NCCR CLIM+ we believe that diversity of thought, background and experience creates better research

Profile of the candidate

• A MSc degree in statistics or a closely related field with strong mathematical background
• Strong programming skills (ideally in Python and/or R)
• Knowledge and ideally practical experience of generative machine learning
• Experience in working with large datasets, ideally hydrological, meteorological or climate observations/simulations
• Very good oral and written communication skills in English
• Motivation to work in an interdisciplinary and international working environment and a collaborative mindset

International reputation

Application and Contact

We look forward to receiving your online application with the following elements:

• your CV,
• a cover letter,
• diploma certificates and transcripts,
• a link to your MSc thesis (draft or final version),
• and the contact information of two professional references.

Applications submitted by October 1, 2026 (inclusive) are guaranteed full consideration.

Please submit your application as a single pdf document to office.stat@unibe.ch

PhD on Conditional Generative Modelling of Local High-Impact Events under Structured Scenarios

UniBE is ambitious. With us, you deliver top performance and test your skills against the best.

Start Date November 1, 2026, or by arrangement

Employment Relationship 100%, 4 years

Institution / Workplace Bern

About The Project We invite applications for a PhD position in Statistics focusing on conditional generative modelling of local high-impact events in a changing climate. Generative models are increasingly used in weather forecasting and climate research, notably to downscale global predictions and projections to local scales. Their use for studying future high-impact events raises fundamental statistical and computational challenges, including the integration of heterogeneous data sources, conditioning on complex information, and generalisation to climate conditions beyond those represented in the training data.

In this PhD project, we will investigate existing and develop new conditional generative modelling approaches for studying future high-impact events at local scales under prescribed, structured conditioning scenarios. These may include large-scale dynamical storylines and event-prone regimes. The project will address methodological questions in conditional generation, the representation and integration of conditioning information, and the generation of plausible local outcomes under changing and potentially unprecedented climate conditions.

Research and Collaborative Environment The PhD student will be hosted at the Institute of Mathematical Statistics and Actuarial Science (IMSV) of the University of Bern, joining the Uncertainty Quantification and Spatial Statistics group led by Prof. David Ginsbourger. IMSV is an engaged, open-minded institute where mathematical statistics and probability theory meet data and models arising from various scientific and societal challenges.

The position is part of the NCCR CLIM+ programme on Climate Extremes and Society, funded by the Swiss National Science Foundation (SNSF). More specifically, the PhD project will contribute to the work package “Local Risks: Impacts and Adaptation” and involve participation in tasks covering topics such as the downscaling and debiasing of input data for hydrological models. Regular interactions and potential collaborations with the teams of Prof. Manuela Brunner (ETH Zurich, WSL), Dr. Christian Grams (MeteoSwiss), Prof. Michael Lehning (EPFL Valais Wallis, WSL), and Prof. Olivia Romppainen-Martius (University of Bern) are planned.

The NCCR CLIM+ supports Switzerland in the transformation towards a climate-resilient society. Together with stakeholders, the interdisciplinary NCCR CLIM+ research community tackles unexplored solution spaces to the climate crisis and develops a blueprint for actionable climate research worldwide. NCCR CLIM+ broadly communicates and shares knowledge, and trains a new generation of experts with the necessary domain and transdisciplinary knowledge.

The NCCR CLIM+ strives to implement equal opportunity at hiring. At NCCR CLIM+ we believe that diversity of thought, background and experience creates better research

Profile of the candidate

• A MSc degree in statistics or a closely related field with strong mathematical background
• Strong programming skills (ideally in Python and/or R)
• Knowledge and ideally practical experience of generative machine learning
• Experience in working with large datasets, ideally hydrological, meteorological or climate observations/simulations
• Very good oral and written communication skills in English
• Motivation to work in an interdisciplinary and international working environment and a collaborative mindset

More Information Job-Abo

Application Process

Your Benefits International reputation

Strong research infrastructure and international network

Collaborative environment and ambitious team

Individual career support

Other Benefits Your Benefits

• International reputation
• Strong research infrastructure and international network
• Collaborative environment and ambitious team
• Individual career support

Working at the University of Bern The University of Bern not only offers exciting tasks but also an environment that actively promotes development, diversity, and equal opportunities. Discover what makes us stand out as an employer and how you can grow with us.

Learn more Career page

Open positions

Application and Contact We look forward to receiving your online application with the following elements:

• your CV,
• a cover letter,
• diploma certificates and transcripts,
• a link to your MSc thesis (draft or final version),
• and the contact information of two professional references.

Applications submitted by October 1, 2026 (inclusive) are guaranteed full consideration.

Please submit your application as a single pdf document to office.stat@unibe.ch

Questions about the position? Prof. Dr. David Ginsbourger (david.ginsbourger@unibe.ch)

Questions about the application? Prof. Dr. David Ginsbourger (david.ginsbourger@unibe.ch)

More interesting job opportunities

Junior Controller/in Finanzabteilung (FIN), 80 - 100%, unbefristet

Wissenschaftliche/r Mitarbeiter/in Institut für Medizinische Lehre (IML), 70 - 80%, befristet

Follow the University of Bern

With us, you create value.

For yourself and for society.

© 2026 University of Bern

• Imprint
• Legal Notice

With us, you create value.

For yourself and for society.

apply