Journal article · 2025 · Open access

Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet

Melissa Adrian, Daniel Sanz-Alonso, Rebecca M. Willett

Artificial Intelligence for the Earth Systems, 4(3) · Published 22 April 2025 · DOI 10.1175/aies-d-24-0050.1

Free PDFView at the publisherCite this paper

In brief

“Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet” is a 2025 journal article by Melissa Adrian and 2 co-authors, published in Artificial Intelligence for the Earth Systems. OpenAlex counts 21 citations as of 16 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
21
Most cited in
2025 (12)
Field-weighted impact
5.22
References
42
Authors
3

Abstract

Modern data-driven surrogate models for weather forecasting provide accurate short-term predictions but inaccurate and nonphysical long-term forecasts. This paper investigates online weather prediction using machine learning surrogates supplemented with partial and noisy observations. We empirically demonstrate and theoretically justify that, despite the long-time instability of the surrogates and the sparsity of the observations, filtering estimates can remain accurate in the long-time horizon. As a case study, we integrate the Fourier Forecasting Neural Network (FourCastNet), a weather surrogate model, within a variational data assimilation framework using partial, noisy ERA5 global reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF). Our results show that filtering estimates remain accurate over a year-long assimilation window and provide effective initial conditions for forecasting tasks, including extreme event prediction. Significance Statement Data-driven weather models have recently shown substantial success in acting as surrogates of physics-based numerical weather prediction models. However, this success has been measured using high-resolution reanalysis data, which are not available in real time. This work assesses data-driven weather models’ use in more practical settings, where we observe noisy, partial measurements of a high-dimensional weather state across time. Our experiments show that integrating FourCastNet into a data assimilation algorithm, 3DVar, using low-resolution observations, allows us to 1) accurately estimate high-dimensional weather states across a long time horizon and 2) effectively use these estimates for initializing forecasts, particularly for 3) the extreme event Typhoon Mawar in 2023. This work provides evidence that data-driven weather models can successfully perform in realistic settings.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Adrian, M., Sanz-Alonso, D., & Willett, R. M. (2025). Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet. Artificial Intelligence for the Earth Systems, 4(3). https://doi.org/10.1175/aies-d-24-0050.1

MLA 9

Adrian, Melissa, et al. “Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet.” Artificial Intelligence for the Earth Systems, vol. 4, no. 3, 2025. https://doi.org/10.1175/aies-d-24-0050.1.

Chicago

Adrian, Melissa, Daniel Sanz-Alonso, and Rebecca M. Willett. 2025. “Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet.” Artificial Intelligence for the Earth Systems 4 (3). https://doi.org/10.1175/aies-d-24-0050.1.

Harvard

Adrian, M., Sanz-Alonso, D. and Willett, R. M. (2025) ‘Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet’, Artificial Intelligence for the Earth Systems, 4(3). doi:10.1175/aies-d-24-0050.1.

IEEE

M. Adrian, D. Sanz-Alonso, and R. M. Willett, “Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet,” Artificial Intelligence for the Earth Systems, vol. 4, no. 3, 2025, doi: 10.1175/aies-d-24-0050.1.

BibTeX

@article{adrian2025data,
  title     = {{Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet}},
  author    = {Adrian, Melissa and Sanz-Alonso, Daniel and Willett, Rebecca M.},
  journal   = {Artificial Intelligence for the Earth Systems},
  year      = {2025},
  volume    = {4},
  number    = {3},
  doi       = {10.1175/aies-d-24-0050.1},
  url       = {https://doi.org/10.1175/aies-d-24-0050.1}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
201811
202433
20251212
202655

Its field-weighted citation impact is 5.22: it has been cited 5.22 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Earth and Planetary Sciences › Atmospheric Science › Meteorological Phenomena and Simulations

  • data assimilation
  • machine learning surrogate models
  • FourCastNet
  • variational data assimilation
  • 3D-Var
  • weather forecasting
  • long-term forecasting
  • extreme event prediction

Work with this paper

Most-cited papers in Earth and Planetary Sciences

All highly cited Earth and Planetary Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Adrian, M., Sanz-Alonso, D., & Willett, R. M. (2025). Data Assimilation with Machine Learning Surrogate Models: A Case Study with FourCastNet. Artificial Intelligence for the Earth Systems, 4(3). https://doi.org/10.1175/aies-d-24-0050.1

How many times has this paper been cited?

21 times, according to OpenAlex on 16 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: read the free full text at arXiv (Cornell University).

Who wrote this paper?

Melissa Adrian, Daniel Sanz-Alonso, Rebecca M. Willett.

What is the DOI of this paper?

10.1175/aies-d-24-0050.1. The permanent link is https://doi.org/10.1175/aies-d-24-0050.1.

Bibliographic data from OpenAlex (CC0), last checked 16 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.