By Guy Gugliotta
Washington Post Staff Writer
Friday, October 22, 2004; Page A03
By measuring variations in satellite orbits, scientists have found the first direct evidence of one of the hallowed tenets of Albert Einstein's theory of general relativity -- that the Earth and other large celestial bodies distort space and time as they rotate.
Researchers reporting yesterday in the journal Nature said improved satellite data had enabled them to show the effect known as "frame-dragging" with a degree of precision never previously possible.
"We improved our accuracy by orders of magnitude," said geodesist Erricos C. Pavlis of NASA's Goddard Space Flight Center in Greenbelt and the University of Maryland at Baltimore. "In a while, we should be able to do even better."
Scientists expect that the results of the experiment, by Pavlis and Ignazio Ciufolini of Italy's University of Lecce, will be reinforced by NASA's ongoing Gravity Probe B, a satellite mission designed to measure frame-dragging and another Einsteinian effect by a different method -- calculating gyroscope deviations over time.
"Gravity Probe B is less systematic, but will provide higher accuracy -- within a margin of error of less than 1 percent," said Michael Salamon, NASA's discipline scientist for fundamental physics. "What this research [yesterday's report] means is that GPB may not in fact provide the first direct evidence of frame-dragging."
In the early 20th century, Einstein theorized that the gravity of large bodies such as the Earth distorts space and time, much the way a bowling ball would stretch a rubber sheet held aloft on all four corners.
Frame-dragging occurs, he said, because the Earth's rotation pulls space-time along with it. Salamon likened the effect to dipping a spoon into a cup of honey and turning it. Close to the spoon the honey twists, but the effect dissipates with distance.
Scientists have wanted to prove Einstein's theory since the dawn of the space age. Gravity Probe B, conceived more than 40 years ago, is measuring frame-dragging from a satellite by focusing a telescope on a distant "guide star" and measuring how the axes of gyroscopes deviate from their original positions pointing directly at the star.
Pavlis and Ciufolini used satellites in a completely different way. They closely tracked the orbits of LAGEOS and LAGEOS2, passive satellites covered with "retroreflectors" that reflect laser beams from ground stations, giving precise measurements of distance from the station to the satellite.
The satellites' orbits are slightly distorted -- not perfectly circular or elliptical -- because irregularities in Earth's surface jog them. But even after subtracting this surface-caused "noise," the researchers were still left with orbits that deviated slightly from what they should have been. The difference, they said, reflected frame-dragging.
"The satellite orbits are not perfect because the Earth is not perfect," Salamon said. "So subtract them out, and what you're left with are the effects of space time. The results are better with two satellites, and three would have been even better."
The key to the experiment's success was better data on Earth's gravity field -- a better map of the Earth-induced orbital distortions. This information, collected by another new satellite, enabled Ciufolini and Pavlis to shrink their margin of error dramatically from the 20 percent they obtained from an earlier attempt.
"There was a tremendous amount of criticism then, and a lot of people said 20 percent was on the edge of being acceptable," Salamon said. "This result, between five and 10 percent, is a lot cleaner."