Google’s DeepMind team has unveiled an AI model for weather prediction this week called GenCast. In a paper published in Nature, Google presents GenCast, a new high-resolution (0.25°) AI ensemble model. Google said GenCast 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 (ECMWF) ENS, up to 15 days in advance.
Google said to achieve this remarkable AI innovation, GenCast was trained by providing it with four decades of historical weather data from ECMWF’s ERA5 archive. This data includes variables such as temperature, wind speed, and pressure at various altitudes.
Advanced Forecasts with AI Help Save Lifes
Forecasts of extreme weather threats that are more accurate can help governments save money, prevent damage, and save more lives. According to Google When GenCast’s ability was tested to predict extreme heat and cold, and high wind speeds, GenCast consistently outperformed ENS.
Additionally, the DeepMind team provided a more approachable explanation of the technology in a blog post: GenCast “consists of an ensemble of 50 or more predictions, each representing a possible weather trajectory,” producing a “complex probability distribution of future weather scenarios,” in contrast to its prior weather model, which was “deterministic, and provided a single, best estimate of future weather.”
