Measuring the deflection of light by Earth.

Dedicated to the highly accurate measurement of star positions, distances, and velocities, the Hipparcos satellite for nearly four years gathered data on more than 100,000 stars. But earlier this year, the spacecraft began to show the effects of prolonged exposure to radiation, and the steady stream of data flowing to Earth abruptly stopped. Unable to continue operating the satellite, the European Space Agency (ESA) in Paris ended communication with Hipparcos on Aug. 15.

Although this step marked the end of the satellite's mission, the huge volume of data relayed to Earth by Hipparcos provides the raw material for a variety of astronomical analyses and investigations. Among the possibilities is an opportunity to test Einstein's general theory of relativity in a realm not previously accessible to experiment.

In the Sept. 1 ASTROPHYSICAL JOURNAL LETTERS, Andrew Gould of the Institute for Advanced Study in Princeton, N.J., proposes a scheme by which Hipparcos data on star positions could be analyzed to detect the minuscule deflection of starlight caused by Earth's gravity.

"The question of whether general relativity is the true theory of gravity remains an open one," Gould asserts. "In the final analysis, this is an experimental question. [General relativity] should be subjected to as many tests as possible on as many scales as possible."

When Albert Einstein first proposed his general theory of relativity, he predicted that a light ray passing through a gravitational field would be deflected by an angle that depends on the mass of the body responsible for the field. In 1919, a team led by British astronomer Arthur Eddington dramatically confirmed Einstein's prediction by measuring the shift in apparent star positions caused by the sun as light from these stars grazed the sun's edge during a solar eclipse.

Since then, measurements of light and radio-wave deflections caused by bodies as large as a cluster of galaxies and as small as Jupiter (SN: 11/9/91, p.294) have extended the range of masses over which general relativity appears to hold. Because Jupiter is about 300 times more massive than Earth, detecting Earth's effect on starlight would provide a new test at the lower end of this range.

Gould's calculations indicate that the Hipparcos measurements of the relative positions of pairs of stars were sensitive enough to exhibit the expected gravitational deflection caused by Earth. He estimates that the resulting angular deflection roughly equals the angular width of a baseball on the moon, as seen from Earth. Although the gravitational bending of light from a single, typical star would be too small to discern, combining the measurements for all 100,000 stars would reveal Earth's effect, Gould concludes.

Michael A.C. Perryman of ESA and his co-workers have already used Hipparcos data to determine the light deflection caused by the sun far more precisely than was possible in previous experiments. They found that their measurements agree with theoretical predictions to within 0.7 percent.