The paper evaluates a physics-informed neural network (PINN) for short‑horizon atmospheric temperature forecasting where observations are sparse. Using ERA5 data at three pressure levels, the PINN outperforms persistence, local‑trend, and two neural‑network baselines, with mean RMSE improvements ranging from 8.1 % at one hour to 23.8 % at three hours. The advantage persists under severe observation sparsity and transfers across regions, though it degrades in complex terrain, highlighting limits of a fixed vertical‑coordinate representation.
arXiv:2608. 09971v1 Announce Type: cross Abstract: Over the past few years, the rapid development of machine learning (ML) models for weather forecasting has produced deterministic models whose medium-range skill matches or exceeds that of the European Centre for Medium-Range Weather Forecasts (ECMWF)'s high-resolution forecast (HRES).
Data-driven models now rival numerical weather prediction in the medium range, but extending them to sub-seasonal lead times raises challenges absent at shorter horizons. Errors accumulate over long autoregressive rollouts, systematic biases grow with lead time, and several years of data must be held out for independent verification, even though machine-learning models otherwise benefit from longer training records.
arXiv:2607. 05100v1 Announce Type: cross Abstract: Data-driven models now rival numerical weather prediction in the medium range, but extending them to sub-seasonal lead times raises challenges absent at shorter horizons.
WeatherNext 3 is a new AI‑driven global weather model that improves both spatial and temporal resolution by generating hourly forecasts at 0.1° resolution, matching the best physics‑based models. It incorporates low‑latency geostationary satellite data and learns to predict satellite‑derived precipitation, tropical cyclones, and station observations, enabling 2 m temperature and dewpoint predictions anywhere and anytime. By directly using raw observations instead of relying solely on analysis data, WeatherNext 3 sets a new state‑of‑the‑art for probabilistic medium‑range forecasting skill.
arXiv:2608. 20024v1 Announce Type: new Abstract: District heating energy hubs require reliable heat load forecasts for efficient operational scheduling.
The paper introduces CloudCast v2, a machine‑learning model that forecasts 12‑hour cloud‑cover from satellite‑derived initial conditions. Trained first on the Copernicus European Regional Reanalysis to learn cloud‑evolution dynamics, it is then adapted to real satellite data using conditional flow matching, a generative technique that conditions noise on observed cloud fields and NWP inputs. CloudCast v2 achieves a 10 % reduction in mean absolute error compared to its predecessor and surpasses it in spatial skill after 3–6 hours, extending useful forecasting beyond the typical 1–3‑hour nowcasting window while preserving satellite‑level spatial detail.
District heating energy hubs require reliable heat load forecasts for efficient operational scheduling. Conventional forecasting workflows train system-specific models on historical data, which can become burdensome when networks change through new consumers, retrofits, or changing operating regimes.
arXiv:2605. 02524v2 Announce Type: replace Abstract: Accurate reconstruction of greenhouse climate variables from sparse sensor measurements is essential for intelligent environmental monitoring, automated climate control, and precision agriculture.
The paper introduces the Continuous Spatiotemporal Temperature Forecaster (CSTF), a neural field that predicts 2‑meter temperature (T2M) by treating forecast lead time and output resolution as explicit queries. CSTF encodes ERA5 history into latent states and decodes T2M as a coordinate‑based field, allowing flexible evaluation at any spatial location, lead time, or resolution. Experiments on a Southeast China benchmark show CSTF outperforms existing methods, achieving a 17.0 % bias reduction and demonstrating coherent predictions across varying lead times and resolutions.
The paper introduces Nested‑EAGLE, a 0.25° global weather model with a 6 km refinement over the contiguous United States, designed to merge short‑ and medium‑range forecasts into a single system. It shows lower mean‑squared error for near‑surface and low‑level variables over the U.S. compared to NOAA’s GFS and HRRR, while remaining competitive globally. Although precipitation forecasts are less skillful than HRRR’s deterministic training, Nested‑EAGLE delivers the most accurate storm‑location predictions at longer lead times, with blurred extrema.
arXiv:2609.22702v1 Announce Type: new Abstract: Accurate flood forecasts several days in advance are essential for flood control, water-resource management, and emergency response. Producing them at...
arXiv:2606. 17659v1 Announce Type: new Abstract: This study introduces enhancements to physics-constrained neural networks (PCNNs) that improve the accuracy and stability of hybrid short-term weather forecasting models.