The paper introduces a new diffusion-based approach for probabilistic weather forecasting that incorporates auxiliary conditional denoising tasks. By predicting future states from both clean and corrupted contexts, the model reduces ambiguity and learns conditional distributions across noise levels using a proper scoring rule. Experiments demonstrate improved one‑step forecasting, especially at longer horizons, and better calibration in high‑dimensional global weather scenarios.
By Joonhyeong Park, Giung Nam, Hyungi Lee, Kyunghyun Cho, Byoungwoo Park, Juho Lee
The study compares the Weather Research and Forecasting (WRF) dynamical model with an unpaired diffusion-based generative model for downscaling extreme precipitation events up to three weeks ahead. Both models outperform raw European Centre for Medium-Range Weather Forecasts forecasts when evaluated against Swiss rain gauge-radar observations, but their strengths differ by atmospheric regime: WRF excels in a multicell, non‑stationary event, while the diffusion model performs more consistently and better in a stationary supercell event.
By Mauricio Lima, Marika Koukoula, Romain Pilon, Monika Feldmann, Erwan Koch, Daniela I. V. Domeisen, Tom Beucler
The paper presents Laxmi, a retrained version of the AIFS weather model that uses satellite-based precipitation observations instead of ERA5 reanalysis data. Laxmi achieves a 19% improvement in global probabilistic accuracy, reduces drizzle overprediction by 33%, and boosts the 95th percentile Brier skill score by 57%. In a case study of 10 Indian tropical storms, Laxmi accurately forecasted 150 mm event-total precipitation in 7 events, outperforming both the original AIFS and the leading physical model IFS.
By Julian F. Schmitt, Bertrand Delorme, Robert C. King, Yashica Patodia, Tapio Schneider, Aditi Sheshadri, Ravi Jain
arXiv:2603. 11229v2 Announce Type: replace-cross Abstract: Machine learning forecast systems are moving beyond point predictions to full predictive distributions for future outcomes y conditional on complex inputs x.
By Elizabeth Cucuzzella, Rafael Izbicki, Ann B. Lee
The study evaluates probabilistic AI weather models—GenCast, FourCastNet 3, and FGN—against the ECMWF IFS-ENS ensemble for tropical cyclone forecasting. Using new probabilistic metrics, the authors find that AI models, particularly FGN, outperform IFS-ENS in cyclone tracking, achieving better location scores for 90% of storms in 2025. The analysis also identifies intensity and spread limitations, model‑specific failure modes, and the influence of pretraining data, highlighting opportunities for targeted improvement.
By Nathan Chalumeau, Xavier Renard, Remi Meynadier, Hugo Rakotoarimanga, Marcin Detyniecki, Emmanuel de Bezenac
The study evaluates three conditioning strategies for a denoising diffusion probabilistic model to downscale daily precipitation for the Colorado River Basin. Channel concatenation of upsampled coarse predictors yields the lowest point‑wise CRPS and MSE but tends to over‑smooth high‑intensity events. Cross‑attention conditioning—both with a learned encoder and with the frozen encoder of the pretrained Prithvi‑WxC weather foundation model—provides better distributional realism, improved spectral fidelity, and stronger performance on extreme events, especially when data are limited.
By Victor Nascimento Ribeiro, Jorge Guevara, Jorge Sebastian Moraga, Chris Lucas, Natalie Lord, Andrew Taylor, Edward Lockhart, Will Trojak, Johannes Schmude, Anne Jones
arXiv:2608. 11951v1 Announce Type: cross Abstract: Extreme events in air transport, such as severe arrival delays and abnormal air times, cause cascading network disruptions with substantial operational, economic, and safety costs.
By Karim Aly, Alexei Sharpanskykh, Jacco Hoekstra
arXiv:2509. 25017v2 Announce Type: replace Abstract: Wildfires are among the most severe natural hazards, posing a significant threat to both humans and natural ecosystems.
By Spyros Kondylatos, Nikolas Papadopoulos, Gustau Camps-Valls, Ioannis Papoutsis
arXiv:2606. 08587v1 Announce Type: cross Abstract: Statistical post-processing has proven to be an effective tool in improving ensemble forecast of different weather variables.
By \'Agnes Baran, M\'at\'e Mihalina
arXiv:2607. 04862v1 Announce Type: new Abstract: Accurate extreme precipitation forecasting is critical for disaster mitigation but remains challenging for numerical weather prediction (NWP) models due to systemic intensity underestimation and spatial displacement.
By Yu Wang, Yong Cao, Kan Dai, Yue Shen, Xiaoqing Zeng, Ruixia Zhao
The paper introduces Hurdle‑RMIL, a two‑stage model that first separates zero‑inflated rainfall from the long‑tailed distribution of positive rainfall and then applies a Bayes‑based transformation to learn a balanced‑distribution model from natural data. Experiments over multiple Chinese regions show that Hurdle‑RMIL reduces systematic underestimation of rare high‑intensity rainfall, improves detection of extreme events, and achieves higher equitable threat scores without significantly harming lower‑threshold accuracy.
By Fangjian Zhang, Xiaoyong Zhuge, Wenlan Wang, Haixia Xiao, Yuying Zhu, Siyang Cheng, Ali Mamtimin
arXiv:2606. 02886v1 Announce Type: cross Abstract: Deep learning weather models now match numerical weather prediction accuracy while running orders of magnitude faster, but produce deterministic forecasts without uncertainty estimates, a critical gap for high-stakes decisions during extreme weather events.
By Jose Marie Antonio Mi\~noza, Rex Gregor Laylo, Sebastian C. Iba\~nez