Earth-2 StormCast
AssessLanguages & Frameworks
A time-dependent forecasting model that directly incorporates observations for environmental prediction.
Why it's here
Placed in Assess: 1 article(s) of evidence from 1 source(s), led by research-stage coverage, with 1 in the last 30 days. Confidence 24%. Low accumulated evidence, so it defaults conservatively pending more signal.
Evidence (1)
- 7NVIDIA GenAI·9/16/2026researchUniversity of Manchester Uses NVIDIA Earth-2 for UK Air Pollution Forecasts
Researchers at the University of Manchester adapted NVIDIA Earth-2 models to forecast air pollution across the U.K. by training CorrDiff on simulated chemistry-climate data and later adding StormCast for time-dependent forecasts using observations. The workflow ran on Isambard-AI and was also demonstrated on DGX Spark, with potential uses in policy analysis, healthcare alerts, and real-time environmental response.