Story posted August 31, 2004
Somewhere in the universe, two black holes are orbiting each other. In their cosmic binary dance, scientists believe they emit gravitational radiation - small waves, or ripples, that travel through the fabric of space and time.
Due to the emission of gravitational radiation, the orbit of the dark stars becomes smaller until they eventually collide and merge.
Waves of gravitational radiation expand throughout the universe. What happens inside the imploded binary no one knows: the gravitational fields within black holes are so strong not even light can escape. But that doesn't stop Thomas Baumgarte from trying to get a peek. Relatively speaking.
The Bowdoin College assistant professor of physics and astronomy is one of the nation's leading numerical relativists - a small group of theoretical astrophysicists trying to solve Einstein's theory of general relativity on the computer. Armed with supercomputers (one at Bowdoin), a National Science Foundation grant and a Guggenheim Fellowship, Baumgarte is attempting to mathematically model violent cosmic events, such as colliding black holes, to predict the gravitational waves they would emit.
This information is critical to the international bank of scientists and engineers intent on detecting, analyzing, and ultimately verifying the existence of gravitational waves, and thus proving Einstein's theory of general relativity - a universal theory of gravity.
"The problem with gravitational radiation is that the wave signals are very difficult to detect," says Baumgarte. "They're extremely small, which is why we haven't yet 'seen' them directly. Very elaborate techniques are being developed to figure out how you can identify an astrophysical signal - which is where my work comes in. If you know what you're looking for, you'll have a better chance of finding it."
While Einstein predicted the existence of gravitational waves as early as 1915, it took recent advances in laser technology to make it possible to build apparatus for detecting them.
Two Laser Interferometer Gravitational Wave Observatories (LIGO) recently were completed in the U.S. - one in Hanford, Washington, and another in Livingston, Louisiana. Each observatory has two L-shaped antennae, or tubes, through which laser beams are reflected back and forth. "You shoot laser beams up and down these L-shaped antennae and you measure how long it takes the laser beam to complete a return trip," says Baumgarte. "That way, you can measure very small distortions in the length of the antenna. If a gravitational wave comes through then it will distort the length of the arms."
And how big might that distortion be? "The amount will be less than the nucleus of a hydrogen atom," says Baumgarte, grinning.
In order to read a faint signal from an event millions of light years away, LIGO's antennae are highly sensitive, which makes them prone to earthly noise interference. "There is noise from small earthquakes, traffic, even waves on a coast that is 200 miles away," says Baumgarte. "You're trying to look in this noise for something that represents a tiny astrophysical signal."
Baumgarte is trying to model events large enough to produce strong gravitational radiation that might be more clearly detectable on Earth. Big events, such as colliding neutron stars and black holes. "Source simulators like myself try to predict what kind of gravitational wave signals these sources emit so that the data analysts can develop techniques for identifying those signals in the detector," he says.
It's a two-step process. First, Baumgarte takes a numerical "snapshot" that describes a binary system at one instant in time. These initial data provide the launching point for step two, a mathematical model of the orbit - or what Baumgarte calls "the movie." He pops those data into a supercomputer, where they mathematically evolve.
"Our program moves the binary forward in time, which allows us to calculate the emission of gravitational radiation," he says. "They spiral towards each other over many, many millions of years until they're at a small separation. That's where we pick up the story. The rest of the movie, the merger and coalescence of these stars, takes only a few milliseconds."
The form of equation used by many source simulators to evolve, or predict, such star systems is one that Baumgarte himself refined. The BSSN Formulation -- short for Baumgarte, Shapiro, Shibata, Nakamura - allows researchers to take Einstein's equations from general relativity and configure them for the supercomputer.
Two Bowdoin undergraduate research assistants co-authored a publication with Baumgarte illustrating the enhanced abilities of the BSSN formulation, a fact that gives him obvious pride: "Usually it takes several years at graduate school until students can do things at this high level," says Baumgarte, who also is an adjunct professor at the University of Illinois, Urbana-Champaign, one of several hotbeds of gravitational radiation research in the country. "Bowdoin has very good undergrads with whom I am able to do very useful parts of my research."
There are roughly 100 physicists worldwide modeling binary systems, says Baumgarte, each working on a slightly different piece of an elusive puzzle, all trying to see into the consequences of Einstein's theory of astrophysical systems. What is it that compels them to create theoretical models of huge, unknown events filtered across infinity? Events that, when they may or may not reach us across millions of light years, register as smaller than an atom?
Baumgarte smiles broadly at this: "Originally we looked at the universe only with the naked eye. Then Galileo used the telescope to look more deeply into the universe. We added to that infrared light, Gamma rays, ultraviolet, radio waves. We now are observing the sky in all different parts of the spectrum. With each new discovery we saw something new.
"If we detect gravity waves, we will not only prove Einstein's theory, we will open up a new window to the universe. It would give us a completely new means of understanding the structure, or geometry, of space. It may help us understand how the universe evolved. Are we interested in our universe? I guess we are."