DeepMind’s new AI promises world’s most accurate 10-day weather forecasts

Key Takeaways:

– Google DeepMind has developed a new AI model called GraphCast, which is the world’s most accurate 10-day global weather forecasting system.
– GraphCast promises medium-range weather forecasts of unprecedented accuracy and is more precise and faster than the industry gold standard for weather simulation, the High-Resolution Forecast (HRES).
– The system can predict extreme weather further into the future than was previously possible and has been deployed on the ECMWF website.
– GraphCast can identify dangerous weather events without being explicitly trained to find them, and it predicted cyclone movement more accurately than the HRES method.
– The model combines machine learning with Graph Neural Networks (GNNs) to process spatially structured data and was trained on decades of weather information.
– The system makes predictions at a spatial resolution of 0.25-degrees latitude/longitude, covering various Earth-surface and atmospheric variables.
– GraphCast significantly outperformed other operational deterministic systems in tests, and it is highly efficient, taking under a minute to complete a 10-day forecast on a single Google TPU v4 machine.
– Further refinement is needed, particularly in measuring the intensity of cyclones, but the model code has been open-sourced for global organizations and individuals to experiment with and make improvements.
– The potential applications of GraphCast are vast and could impact renewable energy production, air traffic routing, and other yet-to-be-imagined tasks.

The Next Web:

A new AI model from Google DeepMind is the world’s most accurate 10-day global weather forecasting system, according to the London-based lab.

Named GraphCast, the model promises medium-range weather forecasts of “unprecedented accuracy.” In research published today, GraphCast was found to be more precise and faster than the industry gold standard for weather simulation, the High-Resolution Forecast (HRES).

The system also predicted extreme weather further into the future than was previously possible. These insights were analysed by the European Centre for Medium-Range Weather Forecasts (ECMWF), an intergovernmental organisation that produces the HRES.

A live version of Graphcast was deployed on the ECMWF website. In September, the system accurately predicted that Hurricane Lee would make landfall in Nova Scotia around nine days in advance.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

In contrast, traditional forecasting methods only spotlighted Nova Scotia around six days ahead of time. They also provided less consistent predictions of the time and location of landfall.