Luca Schmidt

Dept. of Computer Science

Maria-von-Linden-Straße 1

3.OG/ Level 5, A-544

Tübingen, Germany

I am a PhD student in the Climate, Energy and Machine Learning Systems and the Methods of Machine Learning research groups at University of Tübingen. I am also part of the International Max Planck Research School for Intelligent Systems (IMPRS).

My research lies at the intersection of Machine Learning and Climate Science. I analyze climate and weather models to better understand local climate change impacts and support decision-making. On the methodological side, I work with deep generative models, e.g., diffusion models, for statistical downscaling and related climate modeling tasks. Furthermore, I am particularly interested in downstream applications such as wind power forecasting.

From May to October 2025, I joined Rolnick Lab as a visiting researcher at Mila - Quebec Artificial Intelligence Institute in Montréal, Canada.

Feel free to reach out if you are interested in any of these or related topics!

news

Jul 13, 2026	How can machine learning emulators best support climate science? Together with 7 other researchers at the intersection of machine learning and climate science, we have been investigating this question. Check out our updated pre-print on arXiv!
Jul 06, 2026	I will give a guest lecture on Machine Learning for Climate Science at the Department of Geosciences.
May 02, 2026	I’ll be at EGU in Vienna, presenting two posters on geographical generalization and climate emulators.
Apr 27, 2026	I’ll be at Climate Informatics in Lausanne, presenting a poster
Mar 20, 2026	ML emulators can reduce the high computational costs of climate models, but using them effectively remains challenging. In our new arXiv preprint, we explore how to bridge the gap between ML and climate science. If you work in either field, we’d be happy to connect.

selected publications

arXiv

Bridging the Gap Between Climate Science and Machine Learning in Climate Model Emulation

Luca Schmidt^* and Nina Effenberger^*

arXiv preprint , Mar 2026

DOI Bib HTML

@misc{schmidt2026bridging,
  title = {Bridging the Gap Between Climate Science and Machine Learning in Climate Model Emulation},
  author = {Schmidt, Luca and Effenberger, Nina},
  year = {2026},
  month = mar,
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  doi = {10.48550/arXiv.2603.22320},
}

Sci. Rep.

Benchmarking the geographic generalization of deep learning models for precipitation downscaling

Paula Harder^*, Luca Schmidt^*, Francis Pelletier, and 5 more authors

Scientific Reports, Jan 2026

DOI Bib HTML

@article{harder2026benchmarking,
  title = {Benchmarking the geographic generalization of deep learning models for precipitation downscaling},
  author = {Harder, Paula and Schmidt, Luca and Pelletier, Francis and Ludwig, Nicole and Chantry, Matthew and Lessig, Christian and Hernandez-Garcia, Alex and Rolnick, David},
  journal = {Scientific Reports},
  volume = {16},
  number = {1},
  pages = {3733},
  year = {2026},
  month = jan,
  publisher = {Springer Science and Business Media {LLC}},
  doi = {10.1038/s41598-025-34557-4},
}

ERL

Climate data selection for multi-decadal wind power forecasts

Sofia Morelli^*, Nina Effenberger^*, Luca Schmidt^*, and 1 more author

Environmental Research Letters, 2025

DOI Bib HTML

@article{morelli2025climate,
  title = {Climate data selection for multi-decadal wind power forecasts},
  author = {Morelli, Sofia and Effenberger, Nina and Schmidt, Luca and Ludwig, Nicole},
  journal = {Environmental Research Letters},
  volume = {20},
  number = {4},
  pages = {044032},
  year = {2025},
  publisher = {IOP Publishing},
  doi = {10.1088/1748-9326/adc01f},
}