This advancement could greatly improve disaster preparedness and save many lives.
GraphCast outperformed the European Center for Medium-Range Weather Forecasts (ECMWF) model in more than 90% of more than 1,300 test areas.
The results were particularly impressive in predicting conditions in Earth’s troposphere, where GraphCast outperformed the ECMWF model on over 99% of weather variables, such as rainfall and air temperature.
Google’s new product uses machine learning to perform calculations in less than a minute, rather than physics equations based on four decades of historical weather data.
The model uses graphical neural networks that map the Earth’s surface at more than a million grid points, predicting temperature, wind speed and direction, and other conditions such as humidity.
The model is currently used by ECMWF and represents a “moment of reinvention” in weather forecasting, showing that predictions can be made using historical data.
But GraphCast still lags behind traditional models in some areas, such as precipitation, and meteorologists will need to use traditional models alongside machine learning models for more accurate forecasts.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
