By: Shawn  Wang

In early July, a top European weather agency predicted Hurricane Berly would most likely strike Mexico. This prediction was based on global observations by planes, spacecraft, and buoys. On the same day, experts were running an AI program that predicted the hurricane would hit Texas. The forecast drew on nothing more than what the machine had previously learned about the planet’s atmosphere. Four days later, the hurricane flew into Houston, Texas, killing 36 people and wiping out electricity for millions. This prediction offers a peek into the growing world of artificial intelligence.

“This is a really exciting step,” said Matthew Chantry, an A.I. specialist at the European Center for Medium-Range Weather Forecasts. On average, he added, GraphCast (Google’s AI driven forecasting system) and its smart cousins can outperform his agency in predicting hurricane paths. “It’s letting us look for fundamental processes,” Dr. McGovern said.

Importantly, A.I. models can run on much smaller devices, making them much easier to use than a room-sized computer. “You don’t need a supercomputer to generate a forecast. You can do it on your laptop, which makes the science more accessible.” said Maria Molina, a research meteorologist at the University of Maryland who studies A.I. programs for extreme-event prediction.

Some weather experts don’t think we should fully rely on the A.I. systems because each method has its own particular strengths. AI may be able to predict where a hurricane may hit, but what about rain, maximum winds and storm surges?

Jamie Rhome, second director of the National Hurricane Center said “With A.I. coming on so quickly, many people see the human role as diminishing, but our forecasters are making big contributions. There’s still very much a strong human role.”

Sources: https://www.nytimes.com/interactive/2024/07/29/science/ai-weather-forecast-hurricane.html