Just Weather Scoring (JWS) is a single‑stage, end‑to‑end diffusion model that forecasts precipitation directly in radar space, eliminating the need for separate compression or deterministic components. By combining Masked Asynchronous Diffusion with a scoring‑rule objective, JWS adapts diffusion training to high‑dimensional spatio‑temporal data and enables few‑step generation. On SEVIR and MeteoNet benchmarks, JWS achieves state‑of‑the‑art probabilistic forecasting while using fewer parameters and delivering over 17× faster inference than existing methods.
arXiv:2609.37038v1 Announce Type: cross Abstract: Precipitation nowcasting demands accurate short-term forecasts under strong spatiotemporal variability. Diffusion models are well suited to modeling...
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.
arXiv:2608.30205v1 Announce Type: new Abstract: Localized extreme precipitation is a major trigger of urban flash floods and landslides, yet producing nowcasts that combine fine spatial detail with p...
The paper introduces a plug‑and‑play framework called Dynamics‑Precipitation Decoupling that employs a diffusion generator to convert non‑precipitation forecasts from a frozen weather model into observation‑aligned daily precipitation. The generator is conditioned on atmospheric states and satellite‑gauge precipitation estimates, using a multi‑scale recurrent encoder and cross‑attention to capture heterogeneous rainfall distributions and align daily accumulations with evolving atmospheric signals. Experiments on global and regional datasets, including a heavy‑rainfall case study in China, show that the method improves daily precipitation accuracy, event detection, and localized accumulation patterns compared to several baseline weather models and ensemble forecasts.
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.
arXiv:2606. 09959v1 Announce Type: cross Abstract: Precipitation nowcasting is increasingly being approached with deep learning models that learn directly from recent radar observations.
arXiv:2609.38926v1 Announce Type: cross Abstract: Long-term precipitation nowcasting requires modeling radar-echo evolution while preserving localized high-intensity structures. Recent radar-specific...
arXiv:2510. 16031v2 Announce Type: replace-cross Abstract: Machine learning-based precipitation nowcasting relies on high-fidelity radar reflectivity sequences to model the short-term evolution of convective storms.
arXiv:2605. 16163v2 Announce Type: replace-cross Abstract: Skillful medium-range precipitation forecasting at kilometer scale remains challenging over complex terrain because precipitation arises from multiscale nonlinear processes that global models cannot explicitly resolve at affordable cost.
arXiv:2608.12685v2 Announce Type: replace-cross Abstract: An ``Attention Residual U-Net'' method is described for probabilistic quantitative precipitation forecasting (PQPF) that predicts the hourly...
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.
IRENE is a deep learning model that provides probabilistic short‑range precipitation nowcasts over Italy at 1 km spatial and 5‑minute temporal resolution. It uses an encoder–forecaster architecture built on multi‑scale Convolutional Gated Recurrent Units (ConvGRUs) and is trained on national radar composites, with an importance‑sampling scheme and the almost‑fair Continuous Ranked Probability Score as its primary loss. Three training variants—standard, adversarial (IRENE‑GAN), and spectrally constrained (IRENE‑GAN‑RAPSD)—outperform benchmark methods STEPS and DGMR in probabilistic skill, though the advantage in mean absolute error is limited to the first 90 minutes.