DigiTech Update

More Accurate, Taiwan Starts Using AI Technology for Weather Forecasts

The use of technology, largely backed by Nvidia and whose chips are made by Taiwan’s local semiconductor champion TSMC, has impressed forecasters, outperforming conventional methods, which remain the mainstay.

“People are starting to realize that AI does produce some amazing performance compared to conventional models,” said Chia Hsin-sing, director at Taiwanese weather service provider Integrated Disaster Prevention of Technology Engineering Consulting Company Ltd, as quoted by Reuters, Friday (13/9).

Last July, an AI-based weather model, used for the first time, successfully helped Taiwan better predict the path and impact of Typhoon Gaemi.

Several members, including Lin Ping-yu, a weather forecaster at the Central Taiwan Weather Agency (CWA), are again tapping into AI sophistication.

“(AI) is a good thing for us. It’s like having a useful tool to use,” Lin said.

AI weather programs on offer include Nvidia’s FourCastNet, Google’s GraphCast and Huawei’s Pangu-Weather, as well as a deep learning-based system by the European Centre for Medium-Range Weather Forecasts.

“This is a highly anticipated competition. We will know the winner soon,” Chia said.

Such AI models have also begun to be used to predict hurricanes and typhoons in other regions with good accuracy, according to weather forecasters and academics.

This AI-based software is trained using historical weather data to learn the cause-and-effect relationships of meteorological systems, allowing it to predict hundreds of weather variables days in advance – a process that takes just minutes to complete.

For all typhoons in the Western Pacific this year through mid-September, AI’s accuracy in predicting storm tracks over a three-day period was nearly 20 percent higher than conventional models, according to data collected by the CWA.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button