And now here’s Tony with the weather, brought to you by artificial intelligence …
Google’s DeepMind AI wing has created a weather forecasting model it claims “provides better forecasts of both day-to-day weather and extreme events than the top operational system, the European Centre for Medium-Range Weather Forecasts’ ENS, up to 15 days in advance”.
Known as GenCast, it “predicts weather and the risks of extreme conditions with state-of-the-art accuracy”.
In a paper published in Nature, DeepMind states: “Weather forecasts are fundamentally uncertain, so predicting the range of probable weather scenarios is crucial for important decisions, from warning the public about hazardous weather to planning renewable energy use.
“Traditionally, weather forecasts have been based on numerical weather prediction (NWP), which relies on physics-based simulations of the atmosphere.
Google DeepMind GenCast.
“Recent advances in machine learning-based weather prediction have produced machine learning-based models with less forecast error than single NWP simulations.”
The paper says that these machine-learning advances have, until now, “focused primarily on single, deterministic forecasts that fail to represent uncertainty and estimate risk”, and that machine learning-based predictions have “remained less accurate and reliable than state-of-the-art NWP ensemble forecasts”.
But now, with GenCast, Google DeepMind says it has built a “probabilistic weather model with greater skill and speed than the top operational medium-range weather forecast in the world, ENS”.
Now we’re going to get a little technical. Google says “GenCast is … trained on decades of reanalysis data. GenCast generates an ensemble of stochastic 15-day global forecasts, at 12-h steps and 0.25° latitude–longitude resolution, for more than 80 surface and atmospheric variables, in 8 min”.
The result is a forecast that “has greater skill than ENS on 97.2 per cent of 1,320 targets we evaluated and better predicts extreme weather, tropical cyclone tracks and wind power production”.
The upshot, as we enter “the next chapter in operational weather forecasting”, is that “crucial weather-dependent decisions [will be] made more accurately and efficiently”.
You can read the Nature paper here.
