Skip to main content

Inside the new technology transforming hurricane forecasts this hurricane season

Why AI is a new piece of the forecasting toolbox

ORLANDO, Fla. – Hurricane forecasting has come a long way over the last few decades.

Since the 1990s, forecast track errors have been cut by more than half, while intensity forecasting has steadily improved. This is giving meteorologists a much clearer picture of where storms are headed and how strong they could become several days in advance.

That accuracy was on display during the 2025 hurricane season with Hurricane Melissa, when forecasters made the aggressive call nearly three days ahead of landfall that the storm would rapidly intensify from a Category 1 hurricane into a powerful Category 5 before striking Jamaica.

Much of that improvement is being driven by better data, including information collected from areas inside of hurricanes that was impossible to safely reach just a few years ago.

National Hurricane Center Director Michael Brennan says new uncrewed aircraft systems are now helping forecasters gather data just feet above the ocean surface inside hurricanes.

“The small UAS ... can fly down near the ocean surface within just a few feet of the ocean surface where you can’t take a crew aircraft,” Brennan explained. “It’s not safe to fly down that far.”

That information helps scientists better understand the structure and intensity of storms in real time.

Forecast technology is also expanding outside the storm itself.

Brennan says NOAA is preparing to launch new tools and aircraft that will improve data collection around hurricanes.

“There’s always new tools coming on board,” Brennan said. “NOAA has new aircraft that will be coming online later this year ... that’ll help us gather data around and ahead of the storm.”

Artificial intelligence is also becoming part of the forecasting process. In 2025, AI based hurricane models ranked as one of the most accurate guidance tools available to forecasters.

But Brennan says AI is just one piece of the forecasting toolbox.

“AI models are another part of that toolbox,” Brennan said. “If they are all pointing in the same direction, that’s going to help us be more confident to make the more aggressive forecasts.”

Even with rapidly improving technology, Brennan says the final forecast still depends heavily on human expertise.

“There’s a lot of care and a lot of thought and consideration that goes into every aspect of that forecast,” he said.

It’s that combination of cutting edge technology and experienced forecasters that continues to improve hurricane predictions, giving people more confidence in the forecast and, most importantly, more time to prepare.