Publications/CV
Publications
Kochkov, D., Yuval, J., Langmore, I., Norgaard, P., Smith, J., Mooers, G., Klöwer, M., Lottes, J., Rasp, S., Düben, P., Hatfield, S., Battaglia, P., Sanchez-Gonzalez, A., Willson, M., Brenner, M.P. & Hoyer, S., 2024. Neural general circulation models for weather and climate. Nature. https://doi.org/10.1038/s41586-024-07744-y
Rasp, S., Hoyer, S., Merose, A., Langmore, I., Battaglia, P., Russell, T., Sanchez‐Gonzalez, A., Yang, V., Carver, R., Agrawal, S., Chantry, M., Ben Bouallegue, Z., Dueben, P., Bromberg, C., Sisk, J., Barrington, L., Bell, A. and Sha, F., 2024. WeatherBench 2: A benchmark for the next generation of data‐driven global weather models. Journal of Advances in Earth System Modeling. https://doi.org/10.1029/2023MS004019
Beucler, T., Pritchard, P., Yuva, J., Gupta, A., Peng, L., Rasp, S., Ahmed, F., O’Gorman, P., Neelin, D., Lutsko, N. and Gentine, P., 2021. Climate-Invariant Machine Learning. Science Advances. https://doi.org/10.1126/sciadv.adj7250.
Blanusa, M., Lopez-Zurita, C. and Rasp, S., 2022. Internal variability plays a dominant role in global climate projections of temperature and precipitation extremes. Climate Dynamics. https://doi.org/10.1007/s00382-023-06664-3, ArXiv
Robertson, A., et al., 2022. Outcomes of the WMO Prize Challenge to Improve Sub-Seasonal to Seasonal Predictions Using Artificial Intelligence. Bulletin of the American Meteorological Society. https://doi.org/10.1175/BAMS-D-22-0046.1
Price, I. and Rasp, S., 2022. Increasing the accuracy and resolution of precipitation forecasts using deep generative models. AISTATS 2022. https://arxiv.org/abs/2203.12297.
Garg, S., Rasp, S. and Thuerey, N., 2022. WeatherBench Probability: A benchmark dataset for probabilistic medium-range weather forecasting along with deep learning baseline models. Preprint
Kriegmair, R., Ruckstuhl, Y., Rasp, S., Craig, G., 2021. Using neural networks to improve simulations in the gray zone. Nonlinear Processes in Geophysics Discussions. https://doi.org/10.5194/npg-2021-20
Beucler, T. Ebert-Uphoff, I., Rasp, S., Pritchard M. and Gentine, P., 2021. Machine Learning for Clouds and Climate (Invited Chapter for the AGU monograph Clouds and Climate). Preprint
Ruckstuhl, Y., Janjic, T., and Rasp, S., 2021. Training a convolutional neural network to conserve mass in data assimilation. Nonlin. Processes Geophys, https://doi.org/10.5194/npg-28-111-2021
Beucler, T., Pritchard, M., Rasp, S., Ott, J., Baldi, P. and Gentine, P., 2021. Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems. Physical Review Letters, accepted. https://doi.org/10.1103/PhysRevLett.126.098302
Rasp, S. and Thuerey N., 2020. Data-driven medium-range weather prediction with a Resnet pretrained on climate simulations: A new model for WeatherBench. Journal of Advances in Earth System Modeling. https://doi.org/10.1029/2020MS002405
Seifert, A. and Rasp, S., 2020. Potential and Limitations of Machine Learning for Modeling Warm‐Rain Cloud Microphysical Processes. Journal of Advances in Earth System Modeling. https://doi.org/10.1029/2020MS002301
Rasp, S., Dueben, P. Scher, S., Weyn, J., Mouatadid, S., Thuerey, N., 2020. WeatherBench: A benchmark dataset for data-driven weather forecasting. Journal of Advances in Earth System Modeling. https://doi.org/10.1029/2020MS002203
Rasp, S., Schulz, H., Bony, S. and Stevens, B., 2020. Combining crowd-sourcing and deep learning to explore the meso-scale organization of shallow convection. Bulletin of the American Meteorological Society. https://doi.org/10.1175/BAMS-D-19-0324.1
Rasp, S., 2020. Coupled online learning as a way to tackle instabilities and biases in neural network parameterizations: general algorithms and Lorenz 96 case study (v1. 0). Geoscientific Model Development. https://doi.org/10.5194/gmd-13-2185-2020
Zeng Y., Janjić T., de Lozar A., Rasp S., Blahak U., Seifert A., Craig G., 2020. Comparison of methods accounting for subgrid-scale model error in convective-scale data assimilation. Monthly Weather Review. https://doi.org/10.1175/MWR-D-19-0064.1
Hirt, M., Rasp, S., Blahak, U. and Craig, G., 2019. Stochastic parameterization of processes leading to convection initiation in kilometre-scale models. Monthly Weather Review. https://doi.org/10.1175/MWR-D-19-0060.1
Keil, C., Baur, F., Bachmann, K., Rasp, S., Schneider, L. and Barthlott, C., 2019. Relative contribution of soil moisture, boundary layer and microphysical perturbations on convective predictability in different weather regimes Quarterly Journal of the Royal Meteorological Society. https://doi.org/10.1002/qj.3607
Rasp, S. and Lerch, S., 2018. Neural networks for post-processing ensemble weather forecasts. Monthly Weather Review. https://doi.org/10.1175/MWR-D-18-0187.1
Rasp, S., Pritchard, M. and Gentine, P., 2018. Deep learning to represent sub-grid processes in climate models. Proceedings of the National Academy of Sciences. https://doi.org/10.1073/pnas.1810286115
Gentine, P., Pritchard, M., Rasp, S., Reinaudi, G. and Yacalis, G., 2018. Could machine learning break the convection parameterization deadlock? Geophysical Research Letters. https://doi.org/10.1029/2018GL078202
Rasp, S., Selz, T. and Craig, G.C., 2018. Variability and clustering of mid-latitude summertime convection: Testing the Craig and Cohen (2006) theory in a convection-permitting ensemble with stochastic boundary layer perturbations. Journal of the Atmospheric Sciences. DOI: http://dx.doi.org/10.1175/JAS-D-17-0258.1.
Rasp, S., Selz, T. and Craig, G.C., 2016. Convective and Slantwise Trajectory Ascent in Convection-Permitting Simulations of Midlatitude Cyclones. Monthly Weather Review. DOI: http://dx.doi.org/10.1175/MWR-D-16-0112.1
Italics denote people I supervised.
Talks and conference presentations
Invited talks
ISDA Online, January 2024. The second revolution in numerical weather prediction. Video
AI for Good, November 2023. WeatherBench 2 and the second revolution of weather prediction. Video
Joint IS-ENES3/ESiWACE2 Virtual Workshop on New Opportunities for ML and AI in Weather and Climate Modelling, Online, March 2021
The optimization dichotomy: Why is it so hard to improve climate models with machine learning?
Scaling Cascades in Complex Systems, Online, March 2021
The optimization dichotomy: Why is it so hard to improve climate models with machine learning?
CWI SEMINAR FOR MACHINE LEARNING AND UQ IN SCIENTIFIC COMPUTING, Online, February 2021
Hybrid machine learning-physics climate modeling: Challenges and potential solutions, Slides
One World Mathematics of Climate, Online, February 2021
The optimization dichotomy: Why is it so hard to improve climate models with machine learning?, Video
AI for Earth Sciences Workshop, NeurIPS, Online, December 2020
Keynote: Why Benchmarks are Crucial for Progress in AI and How to Design Good Ones for Earth Science, Video
Machine Learning for Weather and Climate Modelling, Oxford, UK, September 2019
Improving Models with Machine Learning Parameterizations: Status Quo vs. Real Progress Slides
Convection parameterization: Progress and challenges, Exeter, UK, July 2019
Machine Learning: Building parameterizations and understanding clouds Slides
Workshop on Data Science and Machine Learning, AGU Fall Meeting, Washington DC, USA, December 2018
Opportunities and challenges for machine learning parameterizations
Columbia University, New York, USA, December 2018
Deep learning to represent subgrid processes in climate models
Harvard University - ClimaTea, Cambridge, MA, USA, December 2018
Deep learning to represent subgrid processes in climate models
Massachusetts Institute of Technology, Cambridge, MA, USA, December 2018
Deep learning to represent subgrid processes in climate models
Computer vision seminar, TU Munich, Munich, Germany, August 2018
Deep learning in weather and climate science
Max-Planck-Institute for Meteorology, Hamburg, Germany, July 2018
Machine learning to represent atmospheric sub-grid processes
Conferences
American Geophysical Union Fall Meeting, Online, December 2020
Data-driven Medium-range Weather Prediction Achieves Comparable Skill to Dynamical Models. But What Does It Mean?, Virtual Poster
Understanding Clouds and Precipitation, Berlin, Germany, February 2019 Sugar, Flower, Fish or Gravel - Cloud classification from humans to machines, Talk
American Geophysical Union Fall Meeting, Washington DC, USA, December 2018
Deep learning to represent subgrid processes in climate models, Poster
Conference on Predictability and Multi-Scale Prediction of High Impact Weather, Landshut, Germany, October 2017
Variability and clustering of mid-latitude summertime convection, Talk
Future of Cumulus Parameterization Workshop, Delft, Netherlands, July 2017
Variability of Mid-Latitude Summertime Convection, Poster
COSMO/CLM/ART User Seminar, DWD Offenbach, Germany, March 2017
COSMO-KENDA experiments with a stochastic boundary layer perturbation scheme, Talk
DACH, Berlin, Germany, March 2016
Der Einfluss eines physikalischen, stochastischen Störungenschemas auf das Störungswachstum in einem konvektiv-skaligen Modell, Talk
COSMO/CLM/ART User Seminar, DWD Offenbach, Germany, March 2016
Stochastic boundary layer perturbations: Systematic impact and perturbation growth, Poster
PANDOWAE Final Symposium, Karlsruhe, Germany, May 2015
High-resolution trajectory analysis of vertical motions in different weather situations, Poster
Workshops
Waves 2 Weather Early Career Scientists, Online, October 2020
Three-day workshop on machine learning and AI Link
Waves 2 Weather Early Career Scientists, Zugspitze, Germany, September 2018
Three-day workshop on machine learning and neural networks in Python Link
Max-Planck-Institute for Meteorology, Hamburg, Germany, July 2018
One-day workshop on neural networks in Python
Earth System Science department, UC Irvine, April/May 2018
Workshops on machine learning in Python, Jupyter notebooks and the xarray package
Experience/Education
2022 - now
Senior Research Scientist, Google Research
Munich, Germany
2022 - 2022
Lead Data Scientist, ClimateAi
Remote
2020 - 2021
Senior Data Scientist, ClimateAi
Remote
2019 - 2020
Postdoc, TU Munich
Group: Physics-based deep learning, Prof. Nils Thuerey
Topic: Figuring out how machine learning can actually improve weather and climate prediction
Jul - Aug 2019
Visiting Researcher, Vulcan, Inc., Seattle
Group: Climate modeling group led by Prof. Chris Bretherton
Feb - May 2018
Visiting Researcher, UC Irvine
Advisor: Prof. Mike Pritchard
Topic: Improving climate model physics with neural networks.
2015 - 2019
PhD in Meteorology, LMU Munich
Project A6 of the DFG Collaborative Research Project Waves to Weather
Advisor: Prof. George C. Craig
Topic: Representing forecast uncertainty using stochastic physical prarameterizations
2012 - 2015
Master of Science in Meteorology, LMU Munich
Thesis supervisors: Prof. George C. Craig, Dr. Tobias Selz
Thesis topic: High-Resolution Trajectory Analysis of Vertical Motions in Different Weather Situations
2009 - 2012
Bachelor of Science in Physical Geography, University of Hull, UK
Thesis supervisors: Dr. Stuart McLelland
Thesis topic: The Effects of Relative Submergence on the Hydrodynamics of an Open Vegetated Channel