Reuters Edge

Hurricane intensity forecasting remains a puzzle

FORT LAUDERDALE, Florida (Reuters) - People living in hurricane danger zones could have a long wait -- as much as 10 years -- for improved forecasts of rapid intensity changes in a hurricane, the top U.S. hurricane forecaster said on Tuesday.

In the past 15 years, forecasters have cut their errors in predicting the track of a hurricane by half. But there’s been almost no improvement in their ability to foresee a storm quickly revving up from a mild Category 1 to a destructive Category 5, National Hurricane Center director Bill Read said.

U.S. researchers are making a major effort to figure out the complex interactions inside a hurricane that might lead to explosive growth.

“Unless there’s some Einstein out there that knows something none of us do, the prevailing thought among the researchers is that a good 24-hour forecast of intensity change is a goal, a stretch goal, that might be attainable (in 10 years),” Read told Reuters.

“But to say, here’s a seedling off Africa that three days from now is going to blow up into a major hurricane, we’re just not there,” he said in an interview at the Florida Governor’s Hurricane Conference in Fort Lauderdale.

Hurricane track forecasts have a five-year average error rate of just under 100 miles 48 hours in advance. Last year, track forecast errors averaged 87 miles.

But intensity forecasting has long been a puzzle to hurricane researchers.

They know tropical cyclones draw energy from warm sea water and can predict some intensification if a storm is passing over a particularly warm area.

But for a more accurate forecast, they need better information from inside the storm about the complex interaction between the eyewall, where the strongest winds are located, the sea below and the atmosphere above, Read said.


The weakness in intensity forecasting shows up nearly every hurricane season. Last year, for example, the hurricane center thought both hurricanes Gustav and Ike would reintensify when they moved off Cuba into the Gulf of Mexico, where they would meet an area of superheated water, Read said.

The reason Ike didn’t rev up is still a puzzle, Read said. In Gustav’s case, the storm encountered dry air and wind shear, a difference in wind speeds at different altitudes that tends to disrupt hurricanes.

“At the time we did not anticipate either shear or dry air to interfere,” Read said.

It’s the number of variables and their unpredictability that make intensity forecasting so difficult, Read said.

Intensity forecasts are critical to public safety. In 2004, Hurricane Charley powered up from a relatively benign Category 2 to Category 4, from 110 mph to 145 miles, in just six hours, only a few miles off the Florida coast.

Many residents were furious, saying they had been caught off guard, with no time to prepare.

Cyclone Nargis, the killer storm that hit Myanmar’s Irrawaddy Delta last year, killing nearly 140,000 people, intensified from a Category 1 to Category 4 in the 24 hours before it hit land.

NASA scientists found that it passed over an abnormally thick warm water layer that had formed about a month earlier. It kept deeper and colder water away from the surface, where it could have weakened the cyclone.

“An explosive change in intensity can occur very close to land, and thus anything we can do to build some lead time onto that will help people prepare,” Read said.

While he admits intensity forecasting is a tricky business, Read noted that for a long time weather forecasters were reluctant to try to predict the weather more than a couple of days in advance.

“Now here we are producing usable five-day track forecasts and we’re probably three to five years away from some minimum skill to go out farther,” he said. “Six or seven days, who knows?”

“Ten years from now the intensity forecasts will be better,” Read said.

Editing by Eric Walsh