June 6, 1999

Building a Cosmic Tape Measure

By GEORGE JOHNSON

hroughout the century, scientists have had to rely on maddeningly oblique methods, laden with assumptions, for measuring the size of the universe. They've had to guess, from purely theoretical considerations, how bright a star or galaxy really is. Then from its apparent brightness, dimmed by the journey of the light through space, they judge its distance.

These interpretations have been so open to debate that the universe seems to expand and contract like an accordion. With uncertainty long considered the norm, it was remarkable when astronomers boldly announced on Tuesday that they had gauged the distance of a galaxy a full 23.5 million light-years away using a stunningly direct approach that relied on little more than simple trigonometry.

"This measurement is the golden meter stick in the glass case," Morris Aizenman, executive officer of the National Science Foundation's Division of Astronomical Sciences proclaimed in a news release. The results came from Dr. James Herrnstein of the National Radio Astronomy Observatory in Socorro, N.M., and a team of scientists from the Harvard-Smithsonian Center for Astrophysics and other institutions.

Scientists are already confident that they can measure how fast galaxies are receding from the primordial big-bang explosion. If they can more precisely know the distances, they can finally close in on the long-disputed number called the Hubble constant, the key to how rapidly the universe is expanding and to calculating its age and size.

It was a week for dramatic -- and conflicting -- announcements. The previous Tuesday, astronomers from the Carnegie Observatories in Pasadena, Calif., made headlines by announcing that using the orbiting Hubble Space Telescope, they had narrowed the age of the universe to 12 billion to perhaps 14 billion years.

Estimates from competing astronomers, which used to be off a full 100 percent, were said to have been pared down to a mere 10 percent. In a widely quoted line, one astronomer said that it was as if scientists, accustomed to arguing over whether the universe had one foot or two, were now quibbling about a single toe.

Then came the measurement by the Herrnstein team, using the Very Long Baseline Array, a network of dish antennas that capture stellar radio waves instead of visible light. If their results hold up, the Carnegie estimate could be 15 percent (a toe and a half) too high.

There is still plenty of wiggle room in the calculations. Determining the universe's age also depends on how much gravity slows the expansion and how much something called the cosmological constant speeds it up. But with this splash of cold water, the earlier euphoria suddenly seemed premature.

Since the late 1920s, when the astronomer Edwin Hubble began observing the expansion from the big bang, there has been nothing very constant about his constant.

According to Hubble's law, the farther a celestial object is from Earth, the faster it will appear to be receding. Pick an object in the sky and measure its velocity and distance. Dividing one by the other gives the Hubble constant -- and with that you have a grip on the rate at which the galaxies are flying apart. From there you can work backward, estimating how long ago everything was compacted into the primordial bomb.

Astronomers feel certain that they can directly determine one of these numbers, the velocity of an object's recession, by its red shift: The faster something is moving away, the more its light waves should be stretched toward the red end of the spectrum -- a phenomenon based on the well-known Doppler effect. It is the other number, the distance of an object, that causes most of the consternation.

Since ancient times, astronomers have measured the distance of nearby planets and stars by using trigonometry. Gauge how much their positions in the sky seem to shift as one moves from one part of the Earth to another, or as the Earth moves around the sun. From this "parallax" one can calculate how far away something is.

But estimating distances beyond the range of parallax -- recently set at a mere few hundred light-years -- has required more imagination. The Carnegie observations rely on stars called Cepheid variables, which blink at a rate believed to vary with their brightness. If two of these beacons are pulsating at the same pace and one appears dimmer, then it is assumed to be farther away.

But the system is fraught with uncertainty. To confidently measure absolute as opposed to relative distance, you have to calibrate the yardstick by directly measuring how far away the nearest Cepheids are. Astronomers have tried to do this using parallactic observations by the European Space Agency satellite Hipparcos. But the measurements are tricky and the data unsure.

And there are other problems. A Cepheid's rhythm may be thrown off by its metallic content. More guesswork comes in when astronomers try to judge how much of the dimming of a Cepheid's light comes not from distance but intervening cosmic dust.

The results from the Very Long Baseline Array cut through these assumptions, measuring the distance of a galaxy (NGC 4258 in the constellation Ursa Major) with old-fashioned parallax. By using computers to coordinate the data from 10 radio telescopes, ranging from Hawaii across North America to the Virgin Islands, the astronomers essentially simulated a dish antenna thousands of miles wide. This let them measure the galaxy's radio wave emissions with a resolution so fine that it puts even the Hubble telescope to shame.

The astronomers focused on a rotating disc of gas at the galaxy's core. First they calculated how fast it was spinning by measuring how much the Doppler effect stretched and squeezed its radio waves. Then they compared this intrinsic velocity with how fast the disc appears to spin from Earth. The farther something is, the more slowly it seems to move -- parallax again. Think of how a jet plane seems to inch across the sky. Trigonometry then yields the distance.

Though this new method greatly expands the power of parallax, it still has its limits. The galaxy was measured at 23.5 million light-years away; the universe is billions of light-years wide. Reaching farther will still require Cepheids and other indirect methods. But now there may be a better way to ensure the accuracy of these yardsticks, putting a more solid foundation under astronomers' feet.