Earth System Foundation Model (ESFM)

ESFM is an open Earth system foundation model designed to ingest heterogeneous geoscience data, including dense gridded reanalysis, climate-model output, sparse satellite observations, and irregular station measurements, using a shared backbone for forecasting and downstream adaptation.

The project focuses on making foundation models more flexible for climate and weather applications, especially when input data are incomplete, sparse, or drawn from different observing systems.

Project page: swiss-ai.github.io/ESFM
Code: github.com/swiss-ai/ESFM
Model weights: huggingface.co/ESFM

Highlights

Handles missing data by design, including partially observed variables, pressure levels, regions, and sparse observations.
Uses one backbone across multiple data modalities, including ERA5, CMIP6, satellite swaths, and station data.
Supports probabilistic forecasting through ensemble conditioning.
Releases code, training scripts, preprocessing tools, and pretrained weights for community use.

This project is part of the Swiss AI Initiative and is developed with collaborators across SDSC, ETH Zurich, EPFL, MeteoSwiss, CSCS, and partner institutes.

(Ozdemir et al., 2026)

References

2026

arXiv
Earth System Foundation Model (ESFM): A unified framework for heterogeneous data integration and forecasting

Firat Ozdemir, Yun Cheng, Salman Mohebi, and 11 more authors

arXiv preprint arXiv:2605.00850, 2026

Abs arXiv Bib HTML Code

An open foundation model extending Aurora’s 3D Swin UNet to integrate heterogeneous Earth system data — satellite observations, station data, and gridded climate data — under a single backbone. Adds axial attention, individual-variable tokenization, and adaptive layer-norm ensembles for probabilistic forecasting; demonstrated on Super Typhoon Doksuri and stratospheric warming cases.
@article{ozdemir2026esfm, title = {Earth System Foundation Model ({ESFM}): A unified framework for heterogeneous data integration and forecasting}, author = {Ozdemir, Firat and Cheng, Yun and Mohebi, Salman and Lehmann, Fanny and Adamov, Simon and Zhang, Zhenyi and Trentini, Leonardo and Grund, Dana and Fuhrer, Oliver and Hoefler, Torsten and Mishra, Siddhartha and Schemm, Sebastian and Soja, Benedikt and Salzmann, Mathieu}, journal = {arXiv preprint arXiv:2605.00850}, year = {2026}, }