Place of work
Leipzig, mobile working partially possible
Working time
Up to 19h/week, but at least 8 h/ week
Contract limitations
limited contract / 6 months
Contact
Your contact for any questions you may have about the job:
[email protected]
[email protected]
Your application
To ensure a fair selection process, please submit your application (cover letter, CV, and relevant supporting documents) via our online portal without a photo, age information, or details about your marital status.
Diversity and Inclusion
The UFZ values diversity and is actively committed to ensuring equal opportunities for all employees, regardless of their origin, religion, beliefs, disability, age or sexual identity.
We welcome people who represent diverse backgrounds, identities and perspectives. We therefore particularly encourage people who are affected by structural discrimination to apply to us.
The UFZ
The Helmholtz Centre for Environmental Research (UFZ) is a world-leading institution in environmental research and a member of the Helmholtz Association, Germany’s largest scientific organisation. With approximately 1,200 employees across Leipzig, Halle, and Magdeburg, we have been conducting research since 1991 using a transdisciplinary approach to address the most pressing challenges of our time: biodiversity loss, climate change, and environmental pollution. Our goal is to translate excellent research into practical solutions for policymakers, business, and society, and to serve as a reliable partner in supporting transformation processes toward a sustainable and just future for current and future generations. We foster a culture of collaboration, openness, and diversity within a work environment that actively promotes creativity and personal development.
The job
Join our team, starting on October 1st, to explore how machine learning can improve the bias adjustment of climate model outputs, with a focus on downstream hydrological applications.
Your tasks
The student assistant will implement existing deep learning-based bias correction methods (e.g., CycleGANs, diffusion models) for climate model outputs and compare their performance against existing tools and reference datasets. Tasks include setting up and applying the workflows, evaluating corrected outputs against reference datasets, and conducting sensitivity analyses across different regions, time periods, and model settings.
-
Implement and adapt existing spatial deep learning-based bias correction workflows for climate model outputs (e.g., methods based on CycleGANs and diffusion models).
- Evaluate and compare climate model outputs before and after correction against reference datasets.
- Conduct sensitivity analyses of the developed correction framework across different data periods, regions, model settings, and climate conditions.
- Develop reproducible scripts and workflows for data processing, model training, evaluation, and visualization.
We offer
- Excellent supervision that supports your personal and professional development
- Exciting insights into the work of a leading research institute
- The chance to work in interdisciplinary, international teams and benefit from a wide range of perspectives
- The opportunity to contribute and actively shape your own ideas and impulses
right from the start
- Modern technical equipment and IT service to optimally support your work
Your profile
- Student in climate science, hydrology, Earth system science, geosciences, environmental data science, computer science, statistics, or a related field.
- Good programming skills in Python; familiarity with Bash scripting is a plus.
- Interest in or basic understanding of machine learning and deep learning concepts.
- Motivated to contribute to climate science research and enjoys working with clean, reproducible workflows
- Good written and spoken English skills.
