The latest study published in npj Climate and Atmospheric Science reports that AI models of the mega-corporations including Google, Nvidia, and Huawei exhibited virtue in their lightning-fast and accurate prediction of the paths and intensity of dangerous storms.
The paper focuses on the fallout from Storm Ciaran, which brought havoc to northern and central Europe countries in November 2023.
As a result, we can see the significance of ML-based meteorological forecasts, with high accuracy, speed, cost-effectiveness, and fewer computational requirements as compared to traditional methods of forecasting.
“AI is transforming weather forecasting before our eyes. Two years ago, modern ML techniques were rarely being applied to make weather forecasts. Now, we have multiple models that can produce 10-day global forecasts in minutes,” Professor Andrew Charlton-Perez, who led the study, said.
The scientists employed four AI models to be produced by Google, Nvidia, and Huawei that were used to result in physics-based models. Astoundingly, they forecasted the meteorological storm’s quick intensification and its way more precise tracking 48 hours straight.
Moreover, the AI model proficiently grasped the favorable larger-scale conditions such as the short distance to the subtropical jet stream and low-level large vortices—traps of highly dense air influencing the development.
The experts pointed out that artificial intelligence models could generate predictions no less accurate than traditional forecasting models.
Also, they noticed that the models developed the desired accuracy in forecasting the important atmospheric conditions that impacted the growth patterns of Storm Ciaran. This meant precisely pinpointing the location of its jet stream, a strong high-speed air current that layers high up in the sky.
the AI technology for weather prediction be expanded further to elicit precautions to reduce the effects of raging events from heath like Storm Ciarán.
They predict that there might be the continuous refinement of machine learning models soon such that AI might start routinely contributing to it in weather forecasting, thus reducing the burden on forecasters in terms of resources and time.
