How reanalysis became essential training data for AI weather forecasting
This article explores how reanalysis — the reconstruction of past weather and climate data by combining observations with models — has become increasingly vital in the era of AI/ML weather forecasting. Far from being made obsolete by AI models, reanalysis serves as the essential training data foundation for these systems. ECMWF's ERA5 reanalysis dataset has been widely used to train data-driven weather models globally, and the upcoming ERA6 promises even greater capability and resolution. The piece positions reanalysis as the "memory" of the Earth system, bridging past, present, and future climate understanding.
Key quotes
The era of artificial intelligence (AI) and machine learning (ML)-powered weather forecasting is here to stay.
Far from making traditional approaches such as reanalysis irrelevant, AI/ML weather models have instead brought reanalysis back to the spotlight.
An AI/ML model is only as good as the data it is trained with, and ECMWF's fifth-generation reanalysis (ERA5) has been widely used to train data-driven weather models around the world.
From the article
You might also wanna read
Recommended Practices for Data Preprocessing in AI-Driven Climate Prediction
This article establishes protocols and recommended practices for preprocessing input data for AI/ML models used in climate prediction across
journals.ametsoc.org·22d agoMachine Learning Eliminates Reanalysis Warm Bias and Reveals Weaker Winter Surface Cooling Over Arctic Sea Ice
DSpace
AI startup WindBorne Systems outperforms government weather forecasts with greater efficiency
WindBorne Systems, an AI weather forecasting startup, has outperformed world-leading government-run systems. Their five-day forecasts match
semafor.com·1mo ago
AI-boosted rare-event sampling framework improves extreme weather prediction accuracy
Researchers have developed an AI-boosted rare-event sampling framework that combines artificial intelligence with physical climate modelling
AI-boosted rare-event sampling framework improves extreme weather prediction accuracy
Researchers have developed an AI-boosted rare-event sampling framework that combines artificial intelligence with physical climate modelling
climate-dashboard/forecast_skill at main · hausfath/climate-dashboard
github.com·27d ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.