Google Unveils Most Advanced Global Weather Model: WeatherNext 3

Google DeepMind and Google Research have introduced WeatherNext 3, their most advanced global weather model yet. According to independent evaluations by Brightband, this new model provides timely and more localized predictions for critical weather events.

WeatherNext 3 uses real-time satellite data to produce hourly forecasts at high resolution, making reliable forecasts accessible across Google products worldwide. This model learns directly from observations, allowing it to generate a new forecast every hour, each one grounded in the most recent satellite data available.

The model's biggest leap forward is its ability to train on sparse weather station observation data, enabling global forecasts at 5-kilometer resolution that account for regional details like topography. This breakthrough is particularly vital for regions across Latin America, Africa, and Asia-Pacific that have historically been underserved by high-resolution forecasting.

WeatherNext 3 also introduces predictions specifically engineered for renewable energy production, including 100-meter wind speeds, cloud cover, and sun radiation levels to help solar farms estimate power generation. This data is crucial for global clean energy planning, allowing grid operators and renewables developers to accurately predict power output and match it with consumer demand.