This study introduces the first controlled benchmark of generative models for weather data assimilation using real station observations from 11,849 NOAA MADIS stations across the U.S. It evaluates key design choices—diffusion vs. flow matching, pixel vs. latent-space formulations, and inference-time conditioning strategies—against a classical 3D-Var baseline. The benchmark finds that learned generative priors and full-gradient guidance improve RMSE over ERA5, while other design variations offer minimal benefit, especially under sparse observation conditions.
By Ruizhe Huang, Qidong Yang, Jonathan Giezendanner, Sherrie Wang
arXiv:2610.00728v2 Announce Type: replace
Abstract: Weather reanalysis products rely on computationally intensive numerical weather predictions followed by data assimilation that corrects the forecas...
By Ruizhe Huang, Qidong Yang, Jonathan Giezendanner, Sherrie Wang
arXiv:2607. 19147v1 Announce Type: cross Abstract: Data-driven methods have revolutionized ocean modeling, yet current approaches rely heavily on complete reanalysis datasets, imposing computational constraints and limiting model performance to that of the training data.
By Yangyang Kong, Yutong Jiang, Yanhai Gan, Junyu Dong, Feng Gao, Xiaopei Lin
arXiv:2512. 03606v2 Announce Type: replace Abstract: Accurate marine wind forecasts are essential for safe navigation, ship routing, and energy operations, yet they remain challenging because observations over the ocean are sparse, heterogeneous, and temporally variable.
By Matteo Peduto, Qidong Yang, Jonathan Giezendanner, Devis Tuia, Sherrie Wang
WindINR is a latent‑state implicit neural representation that enables fast, continuous high‑resolution local wind queries and corrections in complex terrain using sparse observations. It decouples reusable representation learning from sample‑specific latent correction, training a privileged encoder to infer latent states from high‑resolution data and a deployable predictor to estimate initial states from inference‑time inputs. During inference, only the low‑dimensional latent state is updated via a regularized objective, yielding a 14‑fold speedup over full‑network fine‑tuning while maintaining continuous spatial queries.
By Yi Xiao, Qilong Jia, Hang Fan, Pascal Fua, Robert Jenssen, Xiaosong Ma, Wei Xue
arXiv:2606. 17413v1 Announce Type: new Abstract: Space-based monitoring of atmospheric carbon dioxide (CO2) is essential for constraining the global carbon budget.
By Alejandro Calle-Saldarriaga, Felix Jimenez, Jack Grosskreuz, Jiazheng Wang, Jonathan Hobbs, Matthias Katzfuss
arXiv:2606. 11140v1 Announce Type: cross Abstract: Data assimilation (DA) in subsurface flow entails calibrating model parameters to match observed data, typically at wells, while preserving geological realism.
By Guido Di Federico, Wenchao Teng, Louis J. Durlofsky
arXiv:2509. 21751v2 Announce Type: replace Abstract: Four-dimensional variational data assimilation (4DVAR) is a cornerstone of numerical weather prediction, yet it remains computationally intensive and sensitive to initialization due to the non-convexity of its objective function.
By Jaemin Oh
arXiv:2610.28327v1 Announce Type: new
Abstract: We present the Observation Assimilation Transformer (OAT), a geometry-aware neural process that corrects and refines a prior numerical weather predicti...
By Christian Donner, Michele Cattaneo, Maxim Samarin, Francesco Zanetta, Ic\'iar Llor\'ens Jover, Garbiela Aznar Siguan, Michele Volpi
The paper introduces FREESIA, a training‑free, covariance‑aware posterior transport method for data assimilation that embeds forecast cross‑covariance into a flow‑based transport to recover unobserved states while preserving non‑Gaussian posterior structure. It combines an observation‑adaptive proposal with posterior correction, providing an asymptotically exact approximation of nonlinear posteriors and a Wasserstein error bound. Experiments on Double‑Well, Lorenz‑96, and Kolmogorov flow demonstrate that FREESIA captures complex posterior structures and achieves up to a 56% reduction in RMSE compared to the best baseline in sparse, nonlinear, non‑injective observation scenarios.
By Shiwei Ni, Yangwen Zhang, Hang Qi, Xiaofei Guan, Lili Ju
SimCast‑S2S is a generative latent‑diffusion framework designed for probabilistic subseasonal‑to‑seasonal precipitation forecasting. It tackles three key challenges: it uses a diffusion‑based generative pipeline for uncertainty quantification, operates in a compact latent space learned by VAEs for efficient large‑ensemble generation, and employs transfer learning with LoRA to overcome limited training data. On reanalysis data, it outperforms deep‑learning baselines and competes with or surpasses state‑of‑the‑art operational systems such as ECMWF‑S2S.
By Hiep V. Dang, Antonios Mamalakis
arXiv:2601. 17074v5 Announce Type: replace-cross Abstract: Accurate estimation of unobserved quantities in time-varying inverse problems remains challenging when observations are sparse and only indirectly related to the target variable.
By Akila Sampath, Vandana P. Janeja, Jianwu Wang