Enhancing our ability to predict these extreme weather events swiftly and accurately holds the key to better preparation for natural disasters, potentially saving lives.

Addressing this need, Google DeepMind has introduced a groundbreaking AI model, GraphCast.

In a recent publication in Science, GraphCast demonstrated its capability to forecast weather conditions up to 10 days in advance, surpassing the current gold standard both in accuracy and speed.

Across more than 1,300 test areas, GraphCast outperformed the European Centre for Medium-Range Weather Forecasts (ECMWF) model in over 90% of cases.

In the Earth’s troposphere, where the majority of weather phenomena occur, GraphCast excelled, outperforming the ECMWF model in over 99% of weather variables, including rain and air temperature.

What sets GraphCast apart is its ability to provide meteorologists with precise warnings much earlier than conventional models, MIT Technology Review reported.

In a notable example, GraphCast accurately predicted Hurricane Lee’s landfall in Nova Scotia nine days in advance, outperforming traditional models that identified the hurricane’s trajectory only six days ahead.

This advancement could significantly improve our capacity to anticipate and respond to extreme weather events, offering valuable time for proactive measures and decision-making.

Written by B.C. Begley