Reconstructing Hurricane History

Global climate change may lead to more frequent and more severe hurricanes, climate researchers say, as a result of rising ocean surface temperatures. In order to tell if hurricanes are happening more often — and, in turn, what actions should be taken to prepare for or protect against future storms — scientists have to know how frequent hurricanes were in the past. Unfortunately, they’re not sure they do.

Before the advent of satellite hurricane detection in the 1960s, the storms were only noticed if a land station or an airplane spotted them, or if they intercepted a ship at a sea. Because of these constraints, it’s possible that a number of hurricanes were never recorded. But the farther back scientists can trace patterns, the better they’re able to understand if and how those patterns are changing.

“If climate change really is increasing the frequency of hurricanes, we need to have a better statistical sample to really answer that question,” says Carl Ebeling, a geophysicist at Northwestern University in Evanston, Illinois.

To help find that answer, Ebeling and fellow Northwestern researcher Seth Stein have come up with a technique that may allow them to fill in gaps in hurricane history. They’re analyzing the seismic record, the data collected about vibrations in the earth’s crust, to glean information about past storms. They presented their work at the Geological Society of America conference in October.

The winds over the ocean make waves, and those waves pound against the ocean floor, creating vibrations in the earth’s crust. Seismographs record those vibrations as what scientists call “seismic noise,” an ever-present, low amplitude background signal in the seismic record.

Certain characteristics of the seismic noise, such as its frequency and amplitude, reflect the speed of the winds that created it. Ebeling and Stein looked at seismic records from stations in Puerto Rico and Massachusetts to see whether they could work backward from patterns in the seismic noise to differentiate old hurricanes from calm days or non-hurricane storms.

“We wanted to start out with a large storm to see if we could actually see it, so we started out with Hurricane Andrew,” the Category 5 hurricane that wreaked havoc on the southern United States in 1992, Ebeling says. “From the preliminary results, it does look like there’s a correlatable signal.” The information contained in the seismic noise was enough for the researchers to pick out Andrew in the seismic record.

While the technique is still in its early stages, many researchers think it has potential to answer lingering questions about historical hurricanes. “I think it’s very promising, and it’s [produced] pretty much the only solid measurements that there are,” says Luciana Astiz, a geophysicist with the Scripps Institution of Oceanography at the University of California, San Diego.

Before the technique can be widely applied, it must be tested further, Ebeling says. He’s hoping to see if it works in reconstructing Katrina and other recent, well-documented storms.

Another major obstacle is that, until 1990, most seismometers were analog. Many North American seismometers were built in the early twentieth century, and for decades they recorded their data on enormous sheets of paper. Researchers interested in analyzing the seismic noise captured by a particular seismometer would have to wade through “one sheet per day, times 365, times however many years,” Ebeling says. “The problem becomes one of manpower in the digitization of these records.” For now, he’s working only with already-digital data.

Astiz agrees that there are limitations to the technique, but thinks the researchers are looking in the right direction. “Seismic noise is like the heartbeat of the earth — it’s always there in the background,” she says. But, much the way a doctor can extract valuable diagnostic information from the heartbeat of a patient, “it’s a signal that can answer other questions.”

Ebeling hopes his technique will eventually give climate researchers a new way to study changes in hurricane patterns. “I’d like [to find] that this really does work and that it’s able to give the climatology community a different tool to help answer this big question they’re looking at.”