The convergence of machine learning and data assimilation in Earth system science
Abstract Data assimilation (DA) combines observations with numerical models to estimate evolving Earth system states for forecasting and monitoring. Machine learning (ML) enables surrogate modeling, pattern recognition and Bayesian inference. These fields are converging: ML accelerates DA, while DA provides uncertainty quantification and physical constraints. Hybrid DA-ML systems are promising, yet challenges persist in generalization, consistency and reproducibility. These approaches are increasingly integrated, shaping next-generation prediction systems and observing networks.
Authors
- Nedjeljka Žagar (ORCID: https://orcid.org/0000-0002-7256-5073)
- Takemasa Miyoshi (ORCID: https://orcid.org/0000-0003-3160-2525)
- Clara Draper (ORCID: https://orcid.org/0000-0002-8299-4939)
- Sid‐Ahmed Boukabara (ORCID: https://orcid.org/0000-0002-1857-3806)
- Alberto Carrassi (ORCID: https://orcid.org/0000-0003-0722-5600)
- María Eugenia Dillon (ORCID: https://orcid.org/0000-0002-9187-2837)
- Juan José Ruiz
- Daryl Kleist (ORCID: https://orcid.org/0000-0001-8188-8053)
- Peter Lean (ORCID: https://orcid.org/0000-0002-3662-5382)
- Anthony Weaver (ORCID: https://orcid.org/0000-0002-8952-338X)
- Peter Jan van Leeuwen (ORCID: https://orcid.org/0000-0003-2325-5340)
- Niels Bormann (ORCID: https://orcid.org/0000-0001-5302-6093)
- Azadeh Gholoubi
- Rochelle Schneider (ORCID: https://orcid.org/0000-0002-2905-0154)
- Andrew Collard
- Andrew M. Moore
- Haroldo Fraga de Campos Velho
- Nahidul Samrat
- Nico Caltabiano
- Massimo Bonavita
- Marc Bocquet
- Lili Lei
- Sean Healy
- Simon Lang
- Peter Bauer
- Gemma Ralton
- Peter Dueben
- Xuguang Wang
- Russ Treadon
- Rossella Arcucci
- Sibo Cheng
- Alan Geer
- Eviatar Bach
- Katia Lamer
- Sarah Dance
Institutions
- Bureau of Meteorology (AU)
- National Oceanic and Atmospheric Administration (US)
- World Meteorological Organization (CH)
- European Centre for Medium-Range Weather Forecasts (GB)
- Universität Hamburg (DE)
- University of California, Santa Cruz (US)
- Brookhaven National Laboratory (US)
- Universidad de Buenos Aires (AR)
- NOAA Oceanic and Atmospheric Research (US)
- Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique (FR)
- National Meteorological Service (AR)
- Earth System Science Interdisciplinary Center (US)
- Max Planck Institute for Meteorology (DE)
- European Space Research Institute (IT)
- Imperial College London (GB)
- University of Reading (GB)
- Instituto Nacional de Pesquisas Espaciais (BR)
- University of Oklahoma (US)
- Nanjing University (CN)
- Colorado State University (US)
- University of Bologna (IT)
Publication Details
- Journal
- npj Artificial Intelligence
- Published
- 2026-04-23
- DOI
- https://doi.org/10.1038/s44387-026-00107-0
- Primary Topic
- Meteorological Phenomena and Simulations
- Type
- article
- Field-Weighted Citation Impact
- 0.00