Results show that observation-initialized machine-learning forecasts can extend beyond the usual 1-3-hour nowcasting range while retaining spatial detail from satellite cloud fields, and overtake CloudCast v1 in fractions skill score, a neighborhood-based measure of spatial agreement, after approximately 3-6 h, depending on the cloudiness category.
Abstract
Accurate cloud-cover forecasts are important for temperature prediction, radiation forecasting, and solar-power operations. Short-range forecasting methods can preserve observed cloud placement during the first forecast hours, but their skill decreases when cloud fields evolve through formation, dissipation and deformation. Longer lead times require accounting for atmospheric evolution, but operational numerical weather prediction (NWP) forecasts may not accurately represent the satellite-observed cloud state at initialization. We develop CloudCast v2, a machine-learning model for 12-hour cloud-cover forecasting from observation-based initial conditions. The model is first trained on the Copernicus European Regional Reanalysis (Ridal2024) to learn cloud-evolution dynamics, and is then adapted to satellite-derived cloud fields using conditional flow matching (Lipman2023), a generative method that transforms noise into cloud-cover forecasts conditioned on the observed initial cloud fields and NWP inputs. CloudCast v2 reduces mean absolute error by 10% relative to its predecessor, CloudCast v1 (Partio2025), over the 1-12 h range. It also overtakes CloudCast v1 in fractions skill score, a neighborhood-based measure of spatial agreement, after approximately 3-6 h, depending on the cloudiness category. These results show that observation-initialized machine-learning forecasts can extend beyond the usual 1-3-hour nowcasting range while retaining spatial detail from satellite cloud fields.
The results show that the threat score for hourly precipitation is generally improved in the forecasts updated by URDA relative to the pre-existing baseline forecasts, through the ensemble-based error covariance.
Fumitoshi Kawasaki, K. Kurosawa, Atsushi Okazaki et al.· 0 citations
Microsoft Weather Nowcast is presented, a six-hour ensemble radar nowcasting model that jointly learns a deterministic predictor to capture organised precipitation structure shared across ensemble members, and a generator to produce diverse local residuals around this shared prediction.
Ning Wang, Zu-Liang Fang, Wei-Xin Jin et al.· arXiv.org· 0 citations
Data-driven weather forecasting models are commonly trained on ERA5, yet operational deployment must use timely initial states from physics-based numerical weather prediction (NWP) systems whose distributions differ substantially from ERA5. Direct initialization with these states can degrade autoregressive forecasts, m...
Mu-Tian Xi, Xinyu Su, Xiao-Hui Zhong et al.· 0 citations
AI weather forecasting models are commonly trained on the ERA5 reanalysis, which is unavailable in real time. Operational deployment therefore relies on initial conditions produced by numerical or AI analysis systems that differ from those encountered during training. This mismatch can degrade forecast skill, while ret...
Minseok Seo, Noah D. Brenowitz, Do-Yun Kim et al.· 0 citations
Dynamics-Precipitation Decoupling is studied, a plug-and-play framework that uses a separately trained generator to map non-precipitation forecasts from a frozen weather model to observation-aligned daily precipitation.
Guo-Wen Li, Yang Liu, Meng-Xuan Chen et al.· 0 citations
Short-term photovoltaic (PV) power and global horizontal irradiance (GHI) forecasts are essential for effective dispatch, reserve scheduling, and grid operations. At these forecasting horizons, errors are predominantly driven by cloud induced ramps: relying solely on historical numerical data may struggle to anticipate...
Ken Chen, M. Perera, Wei Wang et al.· arXiv.org· 0 citations