Taiwan employs AI to forecast typhoon

tracks as storm Bebinca approaches.

Artificial intelligence (AI) is being used by weather forecasters in Taipei to track the route of tropical storm Bebinca as it approaches the waters off the northern Taiwan, potentially intensifying into a typhoon.
Typhoon tracks have so far been predicted more accurately by AI-generated forecasts than by conventional methods. Some of these forecasts are powered by software from tech giants such as Nvidia (NVDA.O), opens new tab, whose chips are built by Taiwan’s domestic semiconductor champion TSMC (2330.TW), opens new tab.

For the first time, Taiwan was able to better anticipate the path and effects of Typhoon Gaemi in July, which was the largest storm to hit the island in eight years and brought with it unprecedented levels of rainfall thanks to the application of AI-based meteorological models.
Forecasters in Taiwan were astonished by the new technology, which was able to anticipate a direct strike up to eight days ahead of Gaemi’s landfall. This technology far outperformed traditional approaches, which are still the foundation of prediction planning.

According to Chia Hsin-sing, director of Taiwan Integrated Disaster Prevention of Technology Engineering Consulting Company Ltd., a weather service provider, “people are starting to realize AI indeed delivered some stunning performances compared to conventional models.”
The same AI algorithms are currently being used by others to track Bebinca, such as Lin Ping-yu, a forecaster at Taiwan‘s Central Weather Administration (CWA),who said AI has given them a higher degree of confidence there will not be a direct hit.

“We should be happy about this (AI). Lin stated, “It’s like having another practical instrument at my disposal.
GraphCast from Google (GOOGL.O), Pangu-Weather from Huawei (HWT.UL), Nvidia’s FourCastNet, and a deep learning-based system from the European Centre for Medium-Range Weather Forecasts are among the AI weather programs available.
It is a competition that is widely watched. We’ll find out who wins soon,” Chia remarked.
Forecasters and researchers claim that these artificial intelligence algorithms have also started to be used to accurately anticipate storms and hurricanes in other areas.

In just a few minutes, the AI-based software can forecast hundreds of weather variables days ahead of time by learning the cause and effect links of meteorological systems by training on historical weather data.
According to statistics produced by the CWA, AI was roughly 20% more accurate than conventional models in forecasting storm trajectories over a three-day period for all of this year’s typhoons in the Western Pacific up until mid-September.
Ahead of Gaemi, AI helped the administration foresee an unusual loop in its path that prolonged its impact on Taiwan and prompted them to swiftly issue a rare warning for rainfall of 1.8 metres (5.9 feet), which was later proven accurate, according to CWA’s deputy head Lu Kuo-Chen.

“(AI) boosted the confidence for forecasters to make that prediction,” Lu said, adding the early warning gave extra time for authorities to carry out preparations.
Lu is also pinning hopes on a partnership with Nvidia, which this year announced a generative AI tool called CorrDiff that aims to forecast more precise locations of typhoon landfall and provide higher resolution images inside a storm.
“We are seeing the potential,” Lu said.
For now, however, experts say the AI tools were not able to deliver quality forecasts for more detailed impact of a typhoon, such as its strength and winds, and more time is needed for the new technology to solidify its lead over more traditional ways.

Chia questioned, “Was it just good luck?” while highlighting AI’s outstanding work on Gaemi. “We must give AI some further time. It’s something to anticipate.”