How We Beat Hurricanes’ Deadliest Element: Surprise

Before radar, satellites, and 24-hour news, meteorologists faced impending hurricanes with swell measurements and a wet finger.

How We Beat Hurricanes’ Deadliest Element: Surprise

Instead of 24-hour updates on Sandy’s movements, what if you had turned on the news last week and heard this:


“The usual signs which herald the approach of hurricanes was not present in this case. The brick-dust sky was not in evidence to the smallest degree…There were cirrus clouds moving from the southeast during the forenoon of the 7th, but by noon only alto-stratus from the northeast were observed.”

In 1900, that’s how Weather Bureau employee Isaac M. Cline described the forecast of the Galveston Hurricane, which is still known as the deadliest storm in American history. In addition to 130-mph winds and 20-foot storm surges, storms of this era often arrived with a disastrous element of surprise.

“They might have a barometer and some surf indications and a wet finger,” explains Jim Fleming, the editor of peer-reviewed journal History of Meteorology. “But they couldn’t have known where the cone of the hurricane was.”

Hurricane Sandy was devastating. But the accurateness of its forecasted path was, from a technological perspective, remarkable. It involved satellite data, computer models, and planes equipped with radar, sensors, and parachute-equipped measurement devices called dropwindsones.

Ten days before Sandy made landfall, the Weather Channel’s senior hurricane specialist, Bryan Norcross, knew there was a chance that a big storm would be hitting the U.S. Then, with eight days before the storm hit, he began blogging about the unusual possibility that a tropical storm in the Caribbean could turn into a huge weather event in New England, specifically. Three days before Sandy devastated the coast, he had pinpointed “New York City, where Long Island and New Jersey make an “L” as the place where the storm was likely to cause the most trouble.


Had we been in a different era, we may not have been introduced to Sandy before she had arrived. Here’s how we got from forecasting with a barometer and a wet finger to tracking potential hurricanes more than a week before they make landfall.

1900: The Galveston Hurricane

Without satellite technology, radar, or even knowing much about the structure of hurricane, there was no way to predict that the Galveston Hurricane would be the deadliest in American history.

Forecasting at the time relied on measurements such as barometer pressure, tide and swell changes, and wind direction. Though the U.S. Weather Bureau issued warnings about the storm, which had been sighted at sea and made landfall in Cuba, they had no way of pinpointing where it would hit.


“The way that forecasts were made then is that the way things moved yesterday is probably the way they’ll move today,” says Norcross, who is also the author of Hurricane Almanac: The Essential Guide to Storms Past, Present, and Future.

Wrote Galveston-based U.S. Weather Bureau official Cline after the storm: “No one ever dreamed that the water would reach the height observed in the present case.”

1938: The New England Hurricane

Before satellite imagery, ships often provided the first reports of a storm.

A Brazilian freighter, for instance, spotted the storm that became the New England Hurricane of 1938 as it passed by Puerto Rico. It alerted the U.S. Weather Bureau, which in turn warned the citizens of Florida to stock supplies and board up their homes.

But while sailors passed on information about how the storm appeared to be moving, hurricanes often change course. Instead of hitting Florida, as predicted, the storm instead worked its way up the New England coast.


According to the NOAA, about 600 people died in the storm, and the coast suffered $308 million in damage–about $5 billion in today’s money.

1944: The Great Atlantic Hurricane

Though at least one daredevil had already flown a plane into a hurricane by the time The Great Atlantic Hurricane threatened the East Coast in 1944, Major Harry Wexler and Lloyd Woods became the first to take the trip for meteorological purposes. They flew directly into the hurricane, kicking off a data-collection method that remains an important component of hurricane forecasting today.

“Until we actually sent airplanes into a hurricane, we really didn’t have any indication of how large the circulation was or how it was distributed,” explains Norcross. “Once we started flying into hurricanes with airplanes, we could get a sense of how much coast would be affected.”

1960: Hurricane Donna

Hurricane Donna holds the distinction of being the only storm in U.S. history to sustain hurricane-force winds throughout Florida, the Mid-Atlantic, and New England. NOAA calls it “one of the great all-time hurricanes.”

It also narrowly missed becoming the first major East-Coast hurricane to arrive with NASA’s first weather satellites in the air. The first experimental weather satellite, Tiros-1, went out of service about three months before Donna arrived. The second launched about two months later.


1965: Hurricane Betsy

Images from satellites such as Tiros snap photos of the Earth as they orbit it. They don’t, however, show meteorologists how a storm system is moving.

A quickly shifting storm called Hurricane Betsy best demonstrated the significance of this limitation. First it appeared to be heading toward the Carolinas. Then it abruptly changed directions, plowed through the Bahamas, hit Florida and moved on to Louisiana and Mississippi.

A Tiros satellite helped track Betsy as a tropical cyclone.

“People knew the hurricane was there,” Norcross says. “They were flying hurricane hunters through it, but we didn’t have the moving pictures.”

It wouldn’t be until 10 years later that forecasters got their moving pictures with the first Geostationary Operational Environmental Satellites. These satellites orbit at the same speed as the Earth, with the effect that they seem to hover over the same area continuously. They provide 24-hour constant storm surveillance to meteorologists and, eventually, their computer models.


1992: Hurricane Andrew

Though computer simulations of weather systems had been used in forecasting as far back as the 1950s, Norcross says that they weren’t used in a significant way in hurricane forecasting until Hurricane Andrew in 1992.

He remembers using a computer model for the first time as a guide to his own forecast of the storm (he had used models for other types of forecasts since the start of his career in the early ’70s).

Computer models take all that that data being collected by weather balloons and augmented by satellite, radar, and aircraft data from hurricane hunter flights and turn it into a simulation of where a storm is likely to move in the future. “In the interim time, computer models have become better and better,” Norcross says, “and more than anything, that is what is responsible for storm forecasts getting better and better.”

[Image: Flickr user Amber Wilkie]

About the author

Sarah Kessler is a senior writer at Fast Company, where she writes about the on-demand/gig/sharing "economies" and the future of work.