NVIDIA announced on 2026-10-22 that StormScope EU, a regional version of its StormScope model for short-term forecasting, has joined the Earth-2 family of open AI models (NVIDIA). The model was trained on European weather and climate data and uses EUMETSAT satellite observations to predict evolving cloud conditions across 16 Meteosat channels at 10-minute intervals (NVIDIA). At GTC Berlin, the company is showcasing a live StormScope EU pipeline in an interactive weather demo built in partnership with Earthmover (NVIDIA).
Key Facts
| Item | Detail | Source |
|---|---|---|
| New model | StormScope EU, a regional version of StormScope | NVIDIA |
| Input data | EUMETSAT satellite observations across 16 Meteosat channels | NVIDIA |
| Forecast interval | 10 minutes | NVIDIA |
| Live demo | GTC Berlin, built with Earthmover | NVIDIA |
| Training framework | Open source NVIDIA PhysicsNeMo | NVIDIA |
Timeline
- 2026-10-22: NVIDIA published its announcement of StormScope EU and of European organizations building on Earth-2 (NVIDIA).
Details (as Reported)
- Earth-2 gives developers and researchers a family of open AI models and tools they can customize with their own data, running on NVIDIA computing (NVIDIA).
- The demo uses Earth2Studio to run forecasts and Earthmover’s data platform to manage and serve the real-time results, which update every 10 minutes as new satellite observations come in (NVIDIA).
- Earthmover serves the results as map tiles, which lets visitors explore predictions across Europe without downloading entire datasets (NVIDIA).
- Developers can combine StormScope for short-term forecasting with CorrDiff to add local detail to coarser weather data (NVIDIA).
- The open source PhysicsNeMo framework provides training and customization tools for applications that incorporate proprietary data and domain expertise (NVIDIA).
- AXA is using FourCastNet 3 to study regional cyclone risk beyond the historical record across more than 1,000 simulated seasons (NVIDIA).
- Vaisala Xweather is combining Earth-2 with its proprietary observations and lightning data, and is adapting StormScope with lightning data to advance short-term forecasting lead time (NVIDIA).
- University of Manchester researchers adapted CorrDiff for U.K. air-quality modeling, trained it on the Isambard-AI supercomputer and ran it locally on NVIDIA DGX Spark systems (NVIDIA).
- TotalEnergies is exploring Earth-2 for solar-energy forecasting and is evaluating StormScope-based solar-radiation nowcasts at two photovoltaic sites (NVIDIA).
- Space42 developed SARDiff4Height using CorrDiff and PhysicsNeMo, a project that estimates surface heights from a single radar image and produces uncertainty maps (NVIDIA).
What Is Not Yet Confirmed
The sources do not list open questions.