Go to Contents Go to Navigation

Google improves weather predictions with AI technology

All News 14:30 February 04, 2020

By Kim Han-joo

SEOUL, Feb. 4 (Yonhap) -- Google Inc. could greatly improve weather predictions by using machine learning, a type of artificial intelligence (AI) technology, an official from the company claimed Tuesday.

Accurately predicting weather can be particularly challenging for localized events that evolve on hourly timescales, such as thunderstorms.

Google, however, said it is currently conducting research into the development of a machine learning model that addresses this challenge by making predictions that are applied to the immediate future.

This photo, provided by Google Inc., shows scientist Carla Bromberg talking to reporters in Seoul via video conference on Feb. 4, 2020. (PHOTO NOT FOR SALE) (Yonhap)

Carla Bromberg, a program lead at Google's AI for Social Good, said the tech company's precipitation "nowcasting" model focuses on weather forecasts within between one and three hours.

"A significant advantage of machine learning is that inference is computationally cheap, allowing forecasts that are nearly instantaneous," according to the research blog post.

Unlike conventional measurements used by weather agencies around the globe, Google's program uses physics-free approach, meaning that the neural network will learn to approximate the atmospheric physics from the training examples alone, not by incorporating a prior knowledge of how the atmosphere actually works.

The program lead said the Google's results were compared with the High Resolution Rapid Refresh (HRRR) numerical forecast model by the U.S.' weather agency, National Oceanic and Atmospheric Administration.

She claimed that Google's model was about roughly 10 times the spatial resolution made by HRRR, only taking between five and 10 minutes for the forecast.

Google said it currently has no plan to commercialize the research product but is sharing the research product with the general public.


Send Feedback
How can we improve?
Thanks for your feedback!