University of Lausanne

The University of lausanne is a leading international teaching and research institution, with over 5,000 employees and 17,000 students split between its Dorigny campus, CHUV and Epalinges. As an employer, UNIL encourages excellence, individual recognition and responsibility.

We offer a dynamic and inclusive work environment, fostering a strong sense of community and providing dedicated support for developing an academic career. Our team members are part of international research networks, and PhD students are encouraged and supported to participate in these networks, including conferences and workshops.

We offer a nice working place in a multicultural, diverse and dynamic academic environment. Opportunities for professional training, a lot of activities and other benefits to discover.

unil.ch/work

Homepage: https://www.unil.ch/fgse/en/home/menuinst/recherche/idyst.html

OVERVIEW

A PhD student position in spatio-temporal modelling of extreme wildfires events funded by the Swiss National Science Foundation (SNSF) is available in Dr. Marj Tonini's lab at the Institute of Earth Surface Dynamics (IDYST), Faculty of Geosciences and Environment (GSE), University of Lausanne (Unil), in collaboration with Prof. Valérie Chavez-Demoulin (co-PI), Professor of Statistics at the Faculty of Business and Economics (HEC Lausanne), Unil. The lab focuses on developing innovative machine learning and statistical approaches for complex spatiotemporal environmental datasets. We are looking for a candidate to join ExtFire, an SNSF-funded project that aims to develop the first pan-European framework combining non-stationary extreme value theory (EVT), causal inference, and tail-sensitive deep learning to quantify and predict extreme wildfire risk in a reproducible and transferable way. In addition to the two PIs, the ExtFire team brings together three external partners with complementary expertise : Prof. Marco Turco (University of Murcia, Spain) contributes specialized expertise in fire-climate interactions; Dr. Paolo Fiorucci (CIMA Research Foundation, Italy) brings expertise in fire-spread simulation; and Dr. Erwan Koch (Expertise Centre for Climate Extremes, Unil) provides expertise in spatial EVT and max-stable random fields. The successful candidate will join Dr. Tonini's lab at IDYST and work closely with the postdoctoral researcher recruited within the ExtFire project, benefiting from the expretise and support of the entire ExtFire team.

Your responsibilities

The successful candidate will develop research leading to a PhD in environmental sciences under the main supervision of Dr. Marj Tonini. The PhD student will contribute to ExtFire by developing the data-driven core of the project.

The main tasks include :

• harmonising and integrating large open datasets on wildfires, meteorology and land surface into a consistent pan-European data infrastructure;
• develping statistical and spatio-temporal models to characterise extreme fire events, such as those with very large burned areas or rapid spread;
• building the datasets and baseline machine learning models used to evaluate predictions of extreme events;
• applying explainable AI methods (e.g. SHAP) to identify the main environmental drivers of extreme fires across regions;
• writing scientific papers, presenting results at international conferences and assisting in teaching (up to 10% of the time).

Your qualifications

We are looking for a motivated candidate with the following profile :

• Desire to write a thesis in spatio-temporal modelling of extreme events (wildfires);
• Master's degree in environmental sciences, geosciences, statistics, applied mathematics, physics, computer science, or a closely related quantitative field, with a strong academic record;
• Hands-on experience with machine learning, ideally deep learning (e.g. PyTorch, TensorFlow), and familiarity with explainable AI;
• Solid background in statistics and data analysis (e.g. probability, regression, time series or spatial statistics). Knowledge of extreme value theory is an asset;
• Solid programming skills in Python and/or R, demonstrated by experience in processing and analysing large datasets (e.g. gridded climate or remote sensing data, NetCDF/raster formats);
• Fluency in English, written and spoken, with the ability to write scientific reports. Knowledge of French is an asset;
• Proven experience in the quantitative analysis or modelling of natural hazards (e.g. wildfires, floods, landslides, earthquakes), as shown by a thesis, project or publication, is an asset.

Please submit the following documents in a single PDF file :

• Motivation letter (max. 1 page) outlining your motivation and explaining how your backround and experience align with the project.
• Curriculum vitae, including a list of publications, if any.
• Copies of academic qualifications, transcripts of grades and Master's thesis.
• Names, e-mail addresses and phone numbers of two referees.
• Short statement (max. 1 page) describing a previous project involving the quantitative or computational analysis of environmental or natural hazard data (e.g. your Master's thesis, a research project, or an internship), including your personal contribution and a link to your code (e.g. GitHub), if available.

Please note: Only applications submitted through the official UNIL Career Opportunities website will be considered.
Full job description and application form here