New computer model gives hope for forecasting hurricane intensity

Published 6:53 pm Friday, May 18, 2007

Meteorologists have spent decades drastically improving predictions on where a looming hurricane could hit — warnings that potentially drive millions of people from their homes. Now, they aim to better determine how powerful those storms actually will be.

Forecasters are debuting their new Hurricane Weather Research and Forecasting model next month which, for the first time, will take into account most data from within the storm and use it in real time to better determine its strength.

“The processes at the inner core are not well informed and not well predicted,” senior hurricane specialist Richard Pasch said at the National Hurricane Center. “With the HWRF, we’re hoping that we can analyze that middle core.”

Sign up for our daily email newsletter

Get the latest news sent to your inbox

Until now, experts have mostly relied on the Geophysical Fluid Dynamics Laboratory model which, like those before it, mainly depended on initial storm information paired with historical data for similar storms. The higher-resolution new model will consider conditions over the oceans that have never been plugged into models before.

It could take years, and some tinkering, for the new model to realize its full potential. But forecasters hope the result will be a greater understanding of storms like hurricanes Charley and Wilma, which grew substantially stronger in a matter of hours. Wilma went from being a tropical storm to the strongest Atlantic hurricane on record in a day.

The hope is that the model will better predict the strength of a hurricane when it eventually makes landfall and ultimately save lives.

The National Hurricane Center has cut its average forecast error on storm tracks in half over the past 15 years. Average track errors last year were about 55 miles on one-day forecasts, about 111 miles on two-day predictions and 169 miles on three days’ ahead.

In the same period, two-day forecasts for the intensity of all tropical cyclones improved from an average of about 18 mph to about 17 mph.

“We’ve made those improvements in track but we’ve made little improvement in forecasting intensity,” Pasch said. “It was what the science allows. We understand more about hurricane track than intensity.”

With the HWRF, that should change. Information from hurricane hunter aircraft, satellites and other sources will immediately relay wind conditions in and around the storm, temperature, pressure, humidity and other oceanic and atmospheric conditions and analyze them to better determine the track and intensity.

Naomi Surgi, who coordinates the hurricane modeling program at the National Center for Climate Prediction in Camp Springs, Md., said using real-time data provides the most accurate forecasts.

“You have to with as much accuracy as possible describe what that hurricane is doing now,” she said, adding HWRF shows great promise. “It’s getting the storm right, it’s getting the ocean underneath the storm right, it’s getting the environment around the storm right.”

The model has been in development since 2001. Surgi said while improvements will begin the day after it goes operational next month, the model is expected to be used for the next 10 to 15 years.

Forecasters will also test a new radar technique that allows meteorologists to determine wind speeds and barometric pressures in all parts of a storm every few minutes, not the several hours it takes to get readings from hurricane hunter aircraft, said Colin McAdie, a hurricane center meteorologist.

The National Center for Atmospheric Research’s method of analyzing radar images gives a three-dimensional view of an approaching hurricane and will warn if a storm is strengthening unexpectedly, he said.

The hurricane center’s new director, Bill Proenza, has warned there is at least one major threat to forecasters’ accuracy. The QuikScat weather satellite, designed to last five years, is in its seventh year of operation, and it is only a matter of time until it fails. The device gives forecasters basic storm information like wind speed and Proenza has said he is unaware of any plans to allocate an estimated $400 million to replace it.

That aside, even with the expected improvements the HWRF could bring, Surgi said meteorologists still concede they will never deliver error-free hurricane forecasts.

“We have stopped thinking in terms of 100 percent accuracy,” she said, “because I don’t think it will ever be realistic to expect that.”

On the Net:

National Hurricane Center: