New AI Weather Forecasting Model Improves Accuracy by 20%
The European Centre for Medium-Range Weather Forecasts (ECMWF) has launched the Artificial Intelligence Forecasting System (AIFS), which it claims is the first fully operational open AI model for weather forecasting.
ECMWF states that AIFS outperforms traditional physics-based models by up to 20% in some measures. The agency tested the system with Member States and users for 18 months before deployment. One key finding was that AIFS could predict hurricane tracks 12 hours further ahead than existing models.
The AI model is also significantly more energy-efficient. ECMWF reports that AIFS uses approximately 1,000 times less computing energy than physics-based simulations. It also generates forecasts much faster, potentially reducing supercomputing emissions and costs for meteorology.
ECMWF Director-General Florence Rabier called AIFS a milestone that will make high-quality forecasts freely available. She highlighted its potential to improve disaster preparedness and renewable energy planning, such as predicting surface solar radiation levels and wind speeds at turbine heights.
AIFS is not replacing existing models but will run alongside them. It uses the same data sources as the traditional Integrated Forecasting System (IFS), including 60 million quality-controlled observations from satellites, planes, boats, and buoys. However, AIFS differs by using historical data to evaluate real-time conditions, whereas IFS relies solely on physics-based simulations.
ECMWF’s Director of Forecasts and Services, Florian Pappenberger, emphasised that AIFS and IFS are complementary. He stated that making AIFS operational ensures it is openly available and supported 24/7 for the meteorological community.
ECMWF launched AIFS, an AI-based weather forecasting model.
AIFS improves accuracy by up to 20% and predicts hurricane tracks 12 hours further ahead.
The AI model uses 1,000 times less computing energy than traditional physics-based simulations.
Source: FORBES