Max Planck Gesellschaft

Machine Learning & Data Science

Earth system models are the basis for understanding and projecting climate change. Combining machine learning with physical models of the atmosphere and land will improve climate models and the way how we can analyze and interpret complex Earth system data.

Activities & collaborations

Machine Learning for Earth and Climate Sciences

Model and understand the Earth system with Machine Learning and Process Understanding

more about ELLIS Unit Jena and Ellis Society

ERC Synergy Grant “Understanding and Modelling the Earth System with Machine Learning” (USMILE)

Earth System Data Lab

Information from diverse and heterogeneous data streams must be interpreted jointly to understand complex extreme event impacts such as e.g. the Russian Heatwave 2010.

Methodological challenges

… more about causal discovery in the complex Earth system

Runge, J. et al. (2019). Inferring causation from time series in Earth system sciences. doi:10.1038/s41467-019-10105-3

DeepCube

Big data technologies and Artificial Intelligence for Copernicus

Explore and develop new deep learning approaches to address relevant Earth system dynamics related to localized impacts of extreme events through AI-based understanding, quantification and prediction

…more at https://deepcube-h2020.eu/

XAIDA

eXtreme events : Artificial Intelligence for Detection and Attribution

to better assess and predict the influence of climate change on extreme weather using novel artificial intelligence methods

…more about XAIDA

Machine learning challenges

… more about ML challenges for Earth System Sciences

Reichstein, M. et al. (2019). Deep learning and process understanding for data-driven Earth system science. doi:10.1038/s41586-019-0912-1

Key publications

Reichstein, M., Camps-Valls, G., Tuia, D., Zhu, X. X. (2021). Outlook. In G. Camps-Valls, D. Tuia, X. X. Zhu, M. Reichstein (Eds.), Deep Learning for the Earth Sciences: A Comprehensive Approach to Remote Sensing, Climate Science, and Geosciences (pp. 328-330). Hoboken, New Jersey: John Wiley & Sons Ltd.

Mahecha, M. D., Gans, F., Brandt, G., Christiansen, R., Cornell, S. E., Fomferra, N., Kraemer, G., Peters, J., Bodesheim, P., Camps-Valls, G., Donges, J. F., Dorigo, W., Estupinan-Suarez, L. M., Gutierrez-Velez, V. H., Gutwin, M., Jung, M., Londoño, M. C., Miralles, D. G., Papastefanou, P., Reichstein, M. (2020). Earth system data cubes unravel global multivariate dynamics. Earth System Dynamics, 11(1), 201-234. doi:10.5194/esd-11-201-2020

Reichstein, M., Camps-Valls, G., Stevens, B., Jung, M., Denzler, J., Carvalhais, N., Prabhat (2019). Deep learning and process understanding for data-driven Earth system science. Nature, 566(7743), 195-204. doi:10.1038/s41586-019-0912-1

Kraft, B., Jung, M., Körner, M., Requena Mesa, C., Cortés, J., Reichstein, M. (2019). Identifying dynamic memory effects on vegetation state using recurrent neural networks. Frontiers in Big Data, 2: 31. doi:10.3389/fdata.2019.00031


Talks, videos & interviews

Seminar with Markus Reichstein within the Machine Learning seminar series, ECMWF, 24.11.2020

Interview with Markus Reichstein in BR2 radio series "IQ - Science and Research" (German), 17.01.2019

Stockholm Seminar with Markus Reichstein, 21.08.2019

Latest Thinking video with Markus Reichstein, 04.06.2020


back to main page

Directions | Disclaimer | Data Protection | Contact | Internal | Webmail | Local weather | PRINT | © 2011-2022 Max Planck Institute for Biogeochemistry