February 17, 1997

Estimated ages of oldest stars probably won't fall below 15 billion years

A mystery starring globular clusters, the Hubble Telescope, inflation, and the Seven (Sub)dwarfs

By Robert Irion

It's a curious cosmic conundrum: The most ancient stars appear older than the universe itself. Embarrassing headlines aside, this situation has sparked intense research by astronomers, cosmologists, and stellar modelers, all trying to turn back the clock to reconcile the histories of the universe and its bright denizens.

Today, many astronomers favor a universe between 8 and 12 billion years old, based on Hubble Space Telescope data and other measurements of how quickly space expands. Stellar modelers, however, point to stars in our Milky Way that don't look an aeon younger than 13 billion years, and likely are a few billion years older than that. There's the merest hint of potential overlap at the extremes of those ranges, but not enough to make anyone comfortable--especially cosmologists, whose careful theories of how the universe has grown can't account for the eldest galactic geezers.

Something has to give, but it probably won't be the stellar models, Michael Bolte believes. He'll explain why today (February 17) at a session on "New Milestones in Cosmic Distance Measurements" at a meeting of the American Association for the Advancement of Science in Seattle.

"I think there's no question that our models for stars are extraordinarily good," says Bolte, an assistant professor of astronomy and astrophysics at UCSC. "All the physics of this century goes into stellar modeling. Every time we have a chance to test the models, they give us the right answers."

Stellar models start with the star we know best: our sun. Researchers call upon complex equations to describe how this ball of hydrogen, helium, and a soup�on of heavier elements cooks over time. Among other things, the models predict the density of the gas and the speeds of the sun's acoustic "vibrations" with exquisite accuracy--to within 1/2 percent throughout most of the sun's interior. "We understand how the sun works," Bolte says.

These successes make modelers confident that they can predict how stars of a given mass will evolve. Our sun is a middle-aged suburbanite, quietly living on the "main sequence" of stars that burn hydrogen for fuel. When that tank runs dry in about five billion years, the sun's core will assume a far hotter and more hectic lifestyle. The inner turmoil will force the sun's outer layers to swell grotesquely, creating a bright and short-lived "red giant."

This concept of when stars drift off the main sequence, it turns out, is critical for gauging the ages of stars elsewhere. Big stars live fast and die young; small stars conserve their energy and linger on the main sequence. Bolte compares this process to a candle burning down: As a cluster of stars ages, its main sequence steadily erodes as more and more stars wax into red giants, then wane.

Globular clusters, compact swarms of stars that freckle space around our galaxy, are ideal natural labs in which to observe this process. All cluster stars were born simultaneously as the Milky Way coalesced, which makes each cluster a fossil relic of the galaxy's birth. Bolte and others have scrutizined the stars in many globular clusters to pinpoint how much of each cluster's main sequence has eroded. When they convert those data into ages, the results are consistent: between 13 and 18 billion years.

There's a catch, naturally. Astronomers must calculate the distance of a cluster from earth to derive the ages of its stars. Such distance measurements, says Bolte, are "the dominant sources of error in our estimates of stellar ages"--far larger, indeed, than any errors that may lurk within the stellar models themselves. An unusual class of stars, the "subdwarfs," may help solve that problem.

Subdwarfs are solitary old stars that zoom in and out of the galaxy's spiral arms on huge looping orbits. Astronomers have identified seven subdwarfs within a stellar stone's throw of earth. As our planet revolves around the sun, these nearby subdwarfs appear to move subtly in relation to more distant objects. This effect, called trigonometric parallax, provides a solid distance scale. Then, the researchers compare the brightnesses of these stars to the similarly ancient stars they observe in globular clusters. With the known distances to subdwarfs as stepping stones, the distance estimates--and hence age estimates--improve for cluster stars as well. Early results, says Bolte, serve only to strengthen his preferred age for globular clusters of about 15 billion years. (Hipparcos satellite data, on tap at the same AAAS session, appear consistent with this distance scale, he notes.)

As for what the "age problem" means for cosmology, Bolte notes that the Big Bang is alive and well. However, antediluvian stars may deflate the most popular versions of a theory known as cosmic inflation. Inflation holds that the infant universe experienced an unimaginably rapid burst of growth in its first fraction of a second. It also holds that the universe contains just enough matter to perpetually slow down its expansion via gravity--the so-called "critical density." Stars as old as 15 billion years, says Bolte, virtually rule out that scenario. But if the universe contained some fraction of that mass, say 20 to 30 percent, astronomers might have some hope of bridging the gap between the expansion age and the stellar ages.

Return to the Currents home page

Go to UCSC's home page