Once struck, a church bell ringing somewhere in the octave below middle C will resonate for about half a minute, says Penn State oceanographer John Spiesberger. The same sound underwater lasts 40 minutes—long enough to carry 4,000 kilometers, from the deeps north of Hawaii to waters off San Francisco.
Spiesberger and his colleagues at the Woods Hole Oceanographic Institute sent scores of such sounds during the 1980s. Comparing the time each of these sounds spent in transit, they demonstrated, can be a powerfully accurate way to monitor changes in the ocean's temperature.
What's involved is a basic physical truth: Sound bends underwater.
"If sound bounced from the top and bottom of the ocean," Spiesberger writes, "it would die out after a few reflections like a billiard ball ricocheting from soft rails."
But it doesn't.
Most of the heat in the ocean lies beneath the surface, invisible to space-based instruments. To measure it, acoustics professor John Spiesberger and his colleagues designed "satellites" for the sea, with microphones that dangle 500 meters down. These acoustic satellites relay data for plots such as the one above, in which the speed of sound can be used to determine the water temperature.
The speed of sound increases with temperature, and also with pressure. Because temperature is higher at the ocean's surface and water pressure increases with depth, sound travels slowest somewhere in the middle—actually at a depth of about one kilometer.
By the laws of refraction acoustic waves bend away from regions where they travel faster, continually turning back toward the depth of minimum speed.
So sound, traveling underwater, is effectively contained in a channel centered around that one kilometer depth.
"This principle was discovered in 1919," Spiesberger says, "by a German named Lichte who was interested in using acoustics to clear mines from German harbors after World War I."
In 1983, Spiesberger hoped to exploit Lichte's discovery for a very different purpose.
Since water temperature bears directly on speed, he reasoned, if you could precisely measure the time it took for a sound pulse to get from one underwater point to another, you could get a reading of the average water temperature between those points. And since sound could be sent over very long distances, those points could be very far away. With acoustics, you could derive an average temperature reading for the entire Pacific basin.
Furthermore, by comparing the travel times of successive sound pulses over the same distances, you could track changes in temperature—at basin scale—over days, months, even years. "This kind of averaging is critical for looking at global warming," Spiesberger says.
The Hawaii-to-San Francisco experiments demonstrated just how sensitive acoustics can be: after adjusting for slight fluctuations in currents, salinity, and pressure, the thermometer picked up temperature variations as little as .02 degrees C.
The next step, Spiesberger explains, is to map ocean temperatures at a much finer resolution—about 500 kilometers. "We can't really understand climate variability without understanding the ocean at this scale," he says. And without understanding "ordinary" climate variability, we can't untangle its effects from those of global warming.
A collaboration involving six research institutions—Penn State along with Woods Hole, Florida State, the Naval Research Laboratory, the University of Alaska, and the University of Texas—was formed to develop the necessary technology. Dubbing itself the GAMOT group (for Global Acoustic Mapping of Ocean Temperatures), the consortium in 1993 won an $11.2 million grant from the Defense Department's Advanced Research Projects Agency. Spiesberger is overseeing the project.
Temperature mapping at this smaller scale will require far more data than has been collected so far. To this point, Spiesperger says, it's been prohibitively expensive. Placing just one set of bottom-mounted, cabled-to-shore acoustic devices like the ones used in the Hawaii-San Francisco trial costs over a million dollars.
"What we're doing," says Spiesberger, "is developing new instruments for collecting this data."
The acoustic signal emitters they are devising will be attached to underwater moorings, deployed and serviced by research vessels. Receivers will be free-floating, self-contained devices set out to drift. "They measure about five feet across at the surface," Spiesberger explains. "They dangle microphones to a depth of about 500 meters, and they're equipped with very fast computers. They're satellites in the ocean." Data from the receivers will be beamed via space-borne satellites back to shore.
The first receivers were deployed in the Atlantic Ocean for testing last September. By the end of 1995, Spiesberger says, "We hope to have demonstrated that this technology works." If it does, he expects, it will provide real-time ocean temperatures at 500-kilometer resolution for a cost "cheaper than what we routinely spend to measure the temperature of the atmosphere."
John L. Spiesberger, Ph.D., is associate professor of meteorology and senior research associate at the Applied Research Laboratory, 512 Walker Building, University Park, PA 16802; 814-863-8601. The Global Acoustic Mapping of Ocean Temperatures (GAMOT) project is supported by a grant from the Advanced Research Projects Agency of the Department of Defense.
Other Penn State researchers involved in the GAMOT project include Carter L. Ackerman, Ph.D., associate professor engineering research; Daniel Merdes, Ph.D. research associate; Martin Woodard, Ph.D., research associate; Alexandr Draganov, Ph.D., research associate; Bruce Einfalt, M.S., research assistant; Mark Keller, M.S., research assistant; John Kenny, M.B.A., associate research engineer (GAMOT program manager), all from the Applied Research Laboratory; George Parides, M.S., and David Norris, M.S., both Ph.D. students in acoustics; and Michael Zelman, B.S., a master's student in acoustics.
Researchers from other institutions include Harley Hurlburt, Ph.D., and Joseph McCaffrey, Ph.D., both at the Naval Research Center, Stennis Space Center; and Mark Leach of the University of Texas at Austin.