On September 11, 1875, a Jesuit priest in Cuba made what might be the first hurricane forecast. Father Benito Viñes, director of the Meteorological Observatory of the Royal College of Belén in Havana, had recently received telegraphs about a hurricane in the eastern Caribbean Sea. Based on this information, he issued an alert to newspapers and the Havana harbormaster that the storm could hit Cuba the next day. Which it did.

Viñes’ prediction of the storm’s path wasn’t quite right—he thought it would pass through northeastern Cuba, but it ended up hitting the western part of the island. Even so, his forecast gave people a heads-up that a storm was coming, and may have prevented ships from leaving harbor during the storm.

Since then, scientific and technological advancements have significantly increased meteorologists’ ability to predict storms. Today, meteorologists can provide several days’ warning about hurricanes and typhoons (which are the same type of storm, just occurring in different parts of the world). However, even as forecasts have gotten better, the destruction hurricanes wreak has increased.

One of the earliest people to record how hurricanes were different from other storms was William Redfield. After a hurricane hit Connecticut in 1821, Redfield deduced that the storm’s winds moved in a large cyclone based on the different directions that trees had been blown down in the storm’s path.

Although other people had observed unique aspects of hurricanes before Redfield, “he’s kind of regarded as the father of hurricane research,” says Cary Mock, a geography professor at the University of South Carolina.

Redfield corresponded with British engineer William Reid, and their collaboration informed Reid’s 1838 book The Law of Storms. In 1847, Reid—who at various points was governor of Bermuda, Barbados and Malta—set up an early storm warning system in Barbados. He instructed police in the capital of Bridgetown to take regular barometric readings and signal if there was a sudden drop in pressure, suggesting an incoming storm.

By the time Viñes made his hurricane forecast in 1875, the U.S. government had established its first weather service under the Army Signal Service. In 1891, the United States transferred this weather service to the Department of Agriculture and renamed it the Weather Bureau. The bureau’s headquarters were in Washington, D.C., but it received weather observations by telegraph from many regional sources.

The bureau’s most significant failure came in September 1900, when a hurricane hit Galveston, Texas, killing an estimated 8,000 to 12,000 people. The D.C. office had predicted that the hurricane would pass up to New England, but meteorologists at the Belén Observatory in Havana, who had continued Viñes’ work of hurricane observation, warned the bureau that the hurricane was actually barreling toward Texas.

The D.C. office ignored Havana’s warning and issued its own, incorrect forecast. As a result, the people of Galveston didn’t receive proper warning of the in coming storm. The Galveston hurricane remains the most deadly natural disaster in U.S. history.

By the 1920s, forecasters used a variety of methods to try to anticipate hurricanes. They could observe barometric pressure, cloud patterns and ocean swells to predict when a storm might occur locally. They could also receive telegraph and wireless reports about where storms were already occurring, and try to predict where they were going.

Some of these reports came from ships that encountered hurricanes at sea. Once ships received word that a hurricane was in a certain area, they would usually avoid that area—which was good for the ships, but led forecasters to lose track of where a hurricane was. These “lost hurricanes” could then sneak up on islands and coastlines without warning. However, the invention of airplanes gave meterologists a new tool to find them.

In September 1935, Captain Leonard Povey, an American working for the Cuban Army Air Corps, set out in an open-cockpit plane to locate a hurricane that seemed to be moving in a different direction than meteorologists had predicted. He found the hurricane, and observed it by flying around the periphery. According to his observations, it looked like it was heading toward the Florida Keys, and officials issued an hurricane warning to the area.

“At that time, the overseas highway had been built onto Islamorada, and so some people in the northern Keys were able to get out in time because of this aerial reconnaissance,” says Neal Dorst, a meteorologist in the Hurricane Research Division of NOAA’s Atlantic Oceanographic and Meteorological Laboratory who has written about Povey’s historic flight.

Planes became an important part of hurricane tracking in the 1940s and ‘50s. Still, people in a hurricane’s path might only get about 12 to 24 hours notice that a hurricane was approaching—which didn’t provide a lot of time to evacuate. Although Povey’s warning had saved lives, many people had still died when the hurricane struck the Keys.

“Unfortunately, not everybody got out, including a lot of the workers who were building the overseas highway who were stuck in these camps in Islamorada and the northern Keys,” Dorst says.

The introduction of weather-tracking satellites in the 1960s had a huge impact on meteorologists' ability to track hurricanes and forecast their movements. Advancements in computer technology and forecasting models have also allowed meteorologists to predict where a hurricane will land several days in advance, and with better precision.

“In the 40 years I was working, I saw forecast skill improve dramatically,” says Frank Marks, former director of the Hurricane Research Division at NOAA’s Atlantic Oceanographic and Meteorological Laboratory. “You’re talking about a huge jump in our ability, mostly in the last 15 years.”

Despite better tracking and forecasting, the U.S. federal and state governments have struggled to adequately warn and help evacuate people in a storm’s path, or rescue those left stranded by a storm. In 2005, Hurricane Katrina killed at least 1,392 people, according to NOAA’s National Hurricane Center.

Even when people get out in time, the damage that hurricanes wreak has increased dramatically. Climate change plays a role in this, as does increased construction in and migration to hurricane-prone areas.

“If you look at the damage from hurricanes, or any other weather-related peril, the damage is going through the roof,” says Phil Klotzbach, an atmospheric research scientist at Colorado State University. “Most of which is because you have more people, more stuff in harm’s way.”