Weather forecasts move to ten-minute updates: Europe's StormScope turns satellite cloud images into regional nowcasts
Synopsis
NVIDIA has added StormScope EU to its Earth-2 family of open models, a regional version trained on European weather and climate data that uses EUMETSAT satellite observations to predict evolving cloud conditions across 16 Meteosat channels at 10-minute intervals. A global forecast can show a storm taking shape, but decisions about solar farms, shipping routes or insured property depend on what happens locally. At GTC Berlin, the model runs in a live pipeline with Earthmover's data platform, updating as new observations arrive and serving results as map tiles so visitors can explore European predictions without downloading whole datasets.
Interpretation
Regional short-term forecasting gains an open model that teams can customize with their own data: StormScope EU targets Europe and predicts cloud evolution at 10-minute intervals. Earth-2 previously offered a family of open AI models and tools for global weather prediction, detailed regional forecasts and environmental research; StormScope EU is the latest regional model to join that family. The model is trained over European weather and climate data and takes EUMETSAT satellite observations as input, predicting cloud conditions across 16 Meteosat channels.
The forecasts are wired into a live pipeline that updates as soon as new satellite observations arrive and serves results as map tiles. Instead of downloading entire datasets before analysis, visitors can explore predictions across Europe without pulling the full data. In the interactive weather demo at GTC Berlin, Earth2Studio runs the forecasts and Earthmover's data platform manages and serves real-time results, refreshed every 10 minutes with new observations.
Earth-2 models can be combined into workflows tailored to regional needs and retrained for other environmental applications. Short-term forecasting and local detail were separate tasks; they can now be paired inside one workflow. StormScope handles short-term forecasting while CorrDiff adds local detail to coarser weather data, and the open source NVIDIA PhysicsNeMo framework provides training and customization tools so teams can incorporate proprietary data and domain expertise.
Perspective
The report covers the release and demonstration of the Earth-2 open models and tools and their regional member StormScope EU, for researchers and engineers in weather, energy and environmental modeling. The stated operating setting is European coverage, 16 Meteosat channels, a 10-minute update interval, and the split of work between Earth2Studio and Earthmover's platform. The work of AXA, Vaisala Xweather, the University of Manchester, TotalEnergies and Space42 is described as in use, exploration or adaptation within their own settings, not as a common benchmark result.
No forecast error metrics for StormScope EU are given, nor a quantitative comparison with existing regional forecasting methods. The organizational applications are framed as in use, exploration or evaluation; TotalEnergies, for instance, is evaluating StormScope-based solar-radiation nowcasts at two photovoltaic sites. The Manchester air-quality adaptation and Space42's surface-height estimation are research efforts whose accuracy and applicable range are not stated.