New Seismology Device Yields More Accuracy

As Jodie Foster showed us in Contact, scientists around the world are listening to signals from space to learn about the universe.

But while the search for aliens continues, another band of scientists is pointing their instruments downward to see what the signals from the earth's interior can tell us about our own planet.

Although many seismologists have all but forsaken their "Holy Grail"--accurately predicting when and where an earthquake will strike--a new generation of scientists has taken on the challenge of deducing the precise structure and motion of the earth's mantle and core. They are using data collected by 100 stations around the world and shared in digital form over the Internet.

And Harvard is leading the way.

Seismology has come a long way from wobbly pens recording earthquakes as spikes on a roll of paper. Today's equipment, using technology designed by Dr. Joseph M. Steim '78 for his Harvard Ph.D. thesis, measures the motion of the earth as an electric signal which is recorded into a computer.

The primary advantage of Steim's Very Broad Band (VBB) technology is that it allows a single instrument to accomplish the work of dozens of specialized devices.

When Steim's seismograph was installed at Caltech in 1988, it replaced "something like over 40 specialized seismographs," according to Adam Dziewonski, professor of geophysics at Harvard.

The device brings together a spectrum of signals much as a piano combines a range of pitches of varying volumes.

In contrast to a tuning fork which can produce only one pitch at one volume, Steim's seismograph works like a grand piano by harmonizing various signals from one source to create a more polished and rich composition.

"It's as if there were previously one instrument to measure very dim blue light and another to measure very bright red light," Steim says. "You can imagine it was very difficult to put that together to get a clear rendition of the signal, for example, if the source were white light" which is composed of all colors.

Another crucial advantage, Steim says, is the ability to store data in digital form--a practice that has revolutionized the way scientists share and analyze findings. Back in the days when seismic waves were recorded on paper, "it took years to gather data--it was literally mailed from researcher-to-researcher," Steim says.

But with the advent of the Internet, the exchange of data has been facilitated, Steim says, but he adds that it was not crucial to the early success of the project.

"The Internet has made processes that we already found ways to deal with much easier and less expensive, but the project was in no way waiting for the Internet," Steim says.

Using the Internet, seismologists routinely send each other the latest data in a matter of seconds, and can even take remote readings from outlying stations.

"It's much easier now for a researcher to send their preliminary results to another," Steim says. "There are stations in eastern Siberia where the Internet connection exists, but physically it's very difficult to get there."

A VBB Digital Seismograph is so precise that the unit at Harvard's Oak Ridge Observatory in Harvard, Mass., can detect "marine microseisms"--the "motion of the ground generated by wave action against the entire coastline," Steim says. Units in other locations have also detected volcanic explosions, the sonic wave generated by landing space shuttles and even trucks on I-495.

Steim has founded a company named Quanterra that is based in Harvard, Mass. The company was founded to develop and market his technology. Although a seismographic station using his technology currently costs over $150,000, Steim says he hopes to bring the price down to $50,000.

The Big Picture

Harvard has also contributed to the "big picture"--encouraging seismic stations around the world to collaborate and exchange data.

Although there were several attempts to create a global seismograph network in the 1960s and 1970s, Steim says the efforts focused less on science and more on the pressing Cold War need to detect nuclear explosions by measuring their shock waves.

All of that began to change 15 years ago at a meeting at Harvard with the birth of Incorporated Research Institutions for Seismology (IRIS)--a consortium of international seismic research stations funded by the National Science Foundation.

Dziewonski was the first chair of IRIS, and was reelected to serve again in that capacity this year.

"We hope to create, with increasing resolution, a map of the earth's interior," Dziewonski says. "To do this, we essentially have to observe how waves propagate through different parts of the earth."

The signal from an earthquake in China, for example, can travel through the center of the earth at up to 22,000 miles per hour and be felt on the other side of the earth within 20 minutes, according to Dziewonski. But he says that the signals will be slightly different in Paris, New York and Caracas. Analyzing these differences can shed light onto the composition of the earth's interior.

"We think the variations in how the waves propagate in various places relates to the temperature within the earth," Dziewonski says.

To make these calculations possible, IRIS planned and constructed a network of unprecedented scale.

"The idea was to cover the world with a more or less evenly distributed network of some 128 stations...to study the earth's structure in three dimensions," Dziewonski says.

IRIS is close to its goal, with 100 stations already operational in its Global Seismic Network, using state-of-the-art technology 10 to 100 times better than the previous generations of seismographs.

Two-thirds of the IRIS stations are using Steim's equipment, and the data is pouring in. Scientists have already made several discoveries using the network seismographs.

Using digital data from a station in Alaska, scientists at Columbia University last year discovered that the Earth's inner core rotates faster than its mantle.

The network has also confirmed speculations of giant "upwellings and downwellings some 500 to 2,000 miles inside the earth," Dziewonski says.

The Holy Grail

Although scientists have determined which regions of the earth are statistically prone to earthquakes, Dziewonski says they are far from being able to predict when and where the next earthquake will occur.

"It doesn't do you any great good to be able to say there's going to be an earthquake five seconds from now, or five years from now," Steim says, nothing that the most effective prediction would be to pin-point an earthquake within five days. "The goal of being able to do that with that kind of accuracy is, right now, unachievable."

Most of earthquake prediction programs have been dismantled, Steim says.

The focus now "is to gather basic knowledge of where there might be a like-lihood of earthquakes, but also to have enough instruments deployed so that the exact nature and extent of an earthquake, once it happens, can be accurately assessed," he says.