Metraipolis › archive › dc_89cb3c4be620

Benchmarking Generative Models for Weather Data Assimilation on Real Station Observations — arXiv:2610.00728

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.00728
License (as archived)
CC0-1.0
Archived
2026-10-06T12:10:55.748Z
SHA-256 of the archived text
8303b1a424818dba8d5b9a96792691b245851901f99f3c40fc4c39c8e3d4fafb
Stable URL of this page
https://metraipolis.com/documento/dc_89cb3c4be620

How to cite

Benchmarking Generative Models for Weather Data Assimilation on Real Station Observations — arXiv:2610.00728 — arXiv (en), https://arxiv.org/abs/2610.00728 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_89cb3c4be620 (Metraipolis archive dc_89cb3c4be620, sha256 8303b1a424818dba8d5b9a96792691b245851901f99f3c40fc4c39c8e3d4fafb)
BibTeX
@misc{dc89cb3c4be620,
  title = {Benchmarking Generative Models for Weather Data Assimilation on Real Station Observations — arXiv:2610.00728},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.00728},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_89cb3c4be620; sha256 8303b1a424818dba8d5b9a96792691b245851901f99f3c40fc4c39c8e3d4fafb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00728. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Weather reanalysis products rely on computationally intensive numerical weather predictions followed by data assimilation that corrects the forecast toward observations. Deep generative models offer a cheaper alternative that shifts much of this cost from inference to offline training. However, existing generative approaches have been evaluated on synthetic observations or under different datasets and evaluation schemes, making it unclear which design choices actually improve real-world data assimilation. We present the first controlled benchmark of generative weather data assimilation on real weather station observations. Using 11,849 NOAA MADIS stations across the contiguous United States and four weather variables, we evaluate methods while holding the dataset, observation operator, and deep learning architecture fixed. The benchmark compares the major design choices, including diffusion versus flow matching, pixel versus latent-space formulations, and multiple inference-time conditioning strategies, against a classical 3D-Var baseline. The benchmark reveals three clear conclusions. First, learned generative priors outperform the Gaussian prior of 3D-Var (35.7% vs. 33.3% RMSE reduction over ERA5) despite using no ERA5 background field at inference. Second, full-gradient guidance consistently outperforms stop-gradient and initial-noise optimization. Third, other choices provide little measurable benefit: diffusion and flow matching perform nearly identically under matched conditions, and latent-space variable mixing does not help. We further evaluate both dense and sparse station settings and find advantages from generative AI and full-gradient guidance more pronounced under sparsity. Together, these results identify which components of generative weather data assimilation improve performance on real station observations and establish a standardized benchmark for future work.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.00728 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T10:21:28.105Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).