← Back
NVIDIA Blog

Pioneers Across Europe Deploy NVIDIA Earth-2 Open Models to Power Real-World Weather and Climate Prediction

•4 min read•
#compute
Pioneers Across Europe Deploy NVIDIA Earth-2 Open Models to Power Real-World Weather and Climate Prediction
Level:Advanced
For:AI Engineers
✦TL;DR

NVIDIA Earth‑2 now includes StormScope EU, a regional short‑term forecasting model trained on European weather data that ingests EUMETSAT satellite observations across 16 Meteosat channels and outputs cloud predictions every 10 minutes. The open‑source PhysicsNeMo framework lets developers fine‑tune Earth‑2 architectures such as StormScope, FourCastNet 3, and CorrDiff with proprietary data for applications ranging from cyclone risk simulation (over 1,000 simulated seasons) to air‑quality and solar‑radiation nowcasting. Real‑time results are served as map tiles via Earthmover’s data platform, enabling interactive demos without bulk dataset downloads. Engineers can thus build end‑to‑end, region‑specific weather pipelines on NVIDIA hardware from DGX Spark to the Isambard‑AI supercomputer.

⚡ Key Takeaways

  • StormScope EU predicts cloud conditions across 16 Meteosat channels at 10‑minute intervals.
  • PhysicsNeMo provides the training and customization toolkit for Earth‑2 model architectures.
  • FourCastNet 3 is used by AXA to simulate over 1,000 seasons for regional cyclone risk analysis.
  • Earth2Studio runs the forecasts while Earthmover serves the outputs as map tiles for instant visualization.
  • Models can be re‑trained on proprietary data (e.g., lightning, air‑quality) to extend to new domains.
💡 Why It Matters

By exposing a modular, open‑source stack that runs on NVIDIA GPUs, teams can rapidly prototype and deploy high‑resolution, real‑time weather services without relying on large national supercomputers, accelerating product development for energy, insurance, and transportation sectors.

✅ Practical Steps

  1. Use Earth2Studio to execute the StormScope EU pipeline and ingest fresh EUMETSAT observations every 10 minutes.
  2. Deploy Earthmover’s data platform to publish forecast outputs as map tiles for downstream applications.
  3. Apply PhysicsNeMo to fine‑tune StormScope, FourCastNet 3, or CorrDiff with your own sensor or domain‑specific datasets.

Want the full story? Read the original article.

Read on NVIDIA Blog ↗

More like this

AI’s GPU focus meets a CPU reality check at Dell and AMD

SiliconANGLE AI•#agents

NVIDIA RTX Spark Compact Desktops Out Now

NVIDIA Blog•#compute

AI’s hidden energy tax: More than 80% of scale-up traffic is just memory moving around

SiliconANGLE AI•#compute

Ian Buck Kicks Off NVIDIA’s Week at SC26 as Science Enters the Age of Agents

NVIDIA Blog•#agents

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING