NVIDIA Earth-2 has added a regional forecasting model trained on European weather and climate data, extending its open AI tools for localized prediction. In an October 22 report, NVIDIA detailed StormScope EU alongside work by insurers, energy companies and researchers adapting Earth-2 models to specialized tasks, with applications ranging from cyclone-risk studies to solar-energy forecasting.
Key Takeaways
- StormScope EU predicts evolving cloud conditions across 16 Meteosat channels at 10-minute intervals using EUMETSAT satellite observations.
- AXA is studying cyclone risk across more than 1,000 simulated seasons, while TotalEnergies is evaluating solar-radiation forecasts at two photovoltaic sites.
- Other projects adapt Earth-2 models for lightning-informed warnings, U.K. air-quality modeling and surface-height estimation from radar imagery.
Table of Contents
- NVIDIA Earth-2 adds a European forecasting pipeline
- Insurance and energy projects target specific decisions
- Researchers extend the models beyond weather
- Conclusion
NVIDIA Earth-2 adds a European forecasting pipeline
The new StormScope EU model targets short-term regional forecasting. At GTC Berlin, NVIDIA is showcasing it through an interactive demonstration developed with Earthmover, connecting incoming satellite observations with forecasts that visitors can explore on a map.
The pipeline uses Earth2Studio to run forecasts and Earthmover’s platform to manage and serve the results. Updates arrive with new satellite observations every 10 minutes. Earthmover delivers the output as map tiles, allowing users to examine predictions across Europe without downloading complete datasets.
NVIDIA’s broader approach allows developers to combine models for different tasks: StormScope handles short-term prediction, while CorrDiff adds local detail to coarser weather data. The open-source PhysicsNeMo framework provides tools for training and customization with organizations’ own data and expertise.
Insurance and energy projects target specific decisions
AXA is using Earth-2’s FourCastNet 3 to examine regional cyclone risk beyond the historical record. Its work spans more than 1,000 simulated seasons, generating insights that NVIDIA says can inform risk assessments and help customers focus prevention efforts.
TotalEnergies’ work remains exploratory. The company is evaluating StormScope-based solar-radiation nowcasts at two photovoltaic sites and developing models intended to provide weather information at subhourly intervals. The project examines whether more detailed forecasts could help anticipate changes in solar-power generation.
Vaisala Xweather is combining Earth-2 with proprietary observations, lightning data and forecasting expertise for its weather intelligence platforms. Its work includes adapting StormScope with lightning data to improve short-term forecasting lead time and make warnings accessible through software-as-a-service products and AI-assistant integrations.
Researchers extend the models beyond weather
University of Manchester researchers adapted CorrDiff for U.K. air-quality modeling. They trained the model on the NVIDIA-powered Isambard-AI supercomputer and ran it locally on NVIDIA DGX Spark systems, demonstrating a workflow spanning national supercomputing infrastructure and desktop hardware.
Space42 developed SARDiff4Height using CorrDiff and PhysicsNeMo. The research estimates surface heights from a single radar image and also produces uncertainty maps, indicating where analysts may need to scrutinize estimates more closely.
Conclusion
StormScope EU adds a Europe-focused forecasting option to Earth-2. The accompanying projects show how organizations are adapting the models to particular risks and environmental questions, while several applications remain under evaluation or development.
Read More News
Disclaimer
This FinanceVelix article provides general information, not personalized financial, investment, legal, or tax advice. Market prices, interest rates and economic conditions can change; investments carry a risk of loss. Facts reflect the material available when published and may change. Check important information independently and consult a qualified professional when needed. Mentioning an asset, company, or service is not an endorsement.
