What happened
Google DeepMind and Google Research introduced WeatherNext 3, a global weather AI model. It now provides hourly, high-resolution forecasts by learning from real-time satellite data, instead of relying solely on physics simulations.
Why it matters
The model forecasts key variables like temperature at a 5-kilometer resolution, described as roughly five times sharper than its predecessor WeatherNext 2. For precipitation, medium-range forecasts show accuracy improvements of up to 60% against NASA's IMERG data, 30% for MRMS, and 10% against rain gauge measurements.
What to watch
WeatherNext 3’s practical impact hinges on whether its hourly, real-time updates translate into consistently better forecasts on the ground, especially where rain gauges are sparse. Watch for accuracy improvements of up to 60% against NASA’s IMERG data to hold in real-world use.
Summaries like this, in your inbox every morning.
The integration of WeatherNext 3 across Google's ecosystem signals an effort to make advanced forecasting widely accessible. The stated goal is to provide more accurate long-term forecasts, with precipitation predictions reportedly up to 50% more accurate when planning a day or more ahead. However, the company acknowledges the atmosphere retains an inherent degree of unpredictability.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
