Adaptive Optics Ushers In A New Era In Ground-Based Astronomy

The technology still has limitations, however. For example, today's adaptive optics systems on the largest telescopes are not able to correct visible-light images. Advanced AO systems now in development are expected to greatly expand the applications for adaptive optics, and will be essential for the next generation of extremely large telescopes now in the planning stages.

"Adaptive optics is working well today on several large telescopes, but for the giant telescopes of the future, the adaptive optics systems will have to be significantly more sophisticated than they are now," said Jerry Nelson, director of the Center for Adaptive Optics (CfAO), a National Science Foundation (NSF) Science and Technology Center based at the University of California, Santa Cruz. Established in 1999, the CfAO plays a key role in the advancement of adaptive optics technology through a network of partners that includes academic institutions, national laboratories, and companies in related industries.

Already, astronomers have been thrilled by the results achieved with adaptive optics systems operating at some of the world's major observatories. At the W. M. Keck Observatory in Hawaii, for example, adaptive optics technology has produced eightfold improvements in image quality, said the obervatory's director, Frederic H. Chaffee.

"When the Keck II Telescope was first 'fitted' with adaptive optics in 1999, the effect was as dramatic as someone who has had 20/150 vision all his life getting fitted with glasses and seeing the world with 20/20 eyes for the first time," Chaffee said. "With adaptive optics, the Keck Telescopes are giving astronomers unprecedented views of the planets and their moons, nearby stars, and distant galaxies. It's a whole new universe out there."

Andrea Ghez, an associate director of the CfAO and professor of astronomy and physics at UCLA, is using the AO system on the Keck Telescopes to study the black hole at the center of our home galaxy, the Milky Way. Ghez first demonstrated the existence of a supermassive black hole at the galactic center using a technique called speckle interferometry, a precursor to adaptive optics.

"It was sort of poor-man's adaptive optics, and it provided very limited information compared to what we are now able to gather using adaptive optics," she said. "For example, we can now start to use spectroscopy to understand the types of stars that are located in the vicinity of the black hole, and we are getting some very surprising results."

An adaptive optics system uses a point source of light as a reference beacon to measure the effects of the atmosphere. The light is analyzed by a detector, called a wavefront sensor, that measures how the light waves were distorted as they passed through the atmosphere. The light collected by the telescope is then bounced off a deformable mirror that changes shape to counteract the distortions measured by the wavefront sensor. A high-speed computer calculates the necessary corrections several hundred times per second, enabling the system to respond to the constantly changing turbulence of the atmosphere.

The light source for the reference beacon can be a bright star in the sky--either the same star being studied or a star adjacent to the object of interest, which might be a faint, distant galaxy. Relying on these "natural" guide stars, however, limits AO observations to the small fraction of the sky that is close to relatively bright stars--only about 1 percent of the sky. So researchers have devised ways of creating artificial guide stars using powerful lasers.

Laser guide stars are still being perfected, however, and most AO observations are still done using natural guide stars. Searching for planets and other dim objects around nearby stars is a natural application for this approach, said CfAO director Nelson.

"A bright star gives off a lot of light that gets spread out by the atmosphere, producing a haze that blocks your ability to see faint things nearby, such as a planet, a red dwarf, or a disk of dust where planets may be forming," Nelson said.

Adaptive optics can be used to concentrate the starlight into a smaller region, while at the same time concentrating the light from faint objects, making them easier to detect. This can be done to some extent with current AO systems, as was demonstrated early this year by astronomers at the University of Hawaii and UC Berkeley who used the AO systems on the Gemini North and Keck Telescopes to obtain images of a brown dwarf in a close orbit around its parent star. Researchers are also working to develop specialized "extreme AO" systems that are specifically tailored for such high-contrast observations.

"You need a lot of light to do this, so it will only work around bright stars. But that's where we want to look anyway to find planets," Nelson said. "I think in the next two to three years we can expect dramatic improvements in this area."

Currently, only the 3-meter Shane Telescope at UC's Lick Observatory has an operational laser guide star that astronomers are using for observations. Within the next few months, however, researchers expect to begin using a laser guide star system recently installed on the 10-meter Keck II Telescope. CfAO researchers at UCSC and Lawrence Livermore National Laboratory (LLNL) are working with the observatory's staff to integrate the laser with the Keck AO system and optimize its performance.

"The laser guide star allows us to make observations anywhere in the sky using adaptive optics," said CfAO associate director Claire Max.

Max, a professor of astronomy and astrophysics at UCSC with a joint appointment at LLNL, led the LLNL teams that developed laser guide stars for the Lick and Keck Observatories. The Lick AO system was also built by her group at LLNL, and the Keck AO system was built in a partnership between LLNL and the Keck Observatory.

At both Lick and Keck, a powerful laser tuned to the wavelength at which sodium atoms absorb and emit light is used to create a glowing spot in a sodium-rich layer of the upper atmosphere. This artificial star is not visible to the naked eye, but it provides enough light for the AO system's wavefront sensor to analyze. Building the specialized lasers is expensive, however, and operating them can be challenging.

"They are really prototypes, and we need to develop better laser technology to make it broadly useful to more observatories," Max said.

The cutting edge of adaptive optics technology is an area called multi-conjugate AO, which requires multiple laser beacons in the sky. There are many challenges to be overcome, but Nelson predicted that a multi-conjugate AO system will be in place on the Gemini South Telescope in about 5 years.

"Designing multi-conjugate AO systems is a high priority for the CfAO," Nelson said. "At a qualitative level, we understand how to go about building one of these systems, but it will not be easy. The analytical tools are far from complete, and the technology for the mirrors and the lasers is also very challenging. But it's the only reasonable way to build AO systems for the giant telescopes with 30-meter or 50-meter mirrors that are being planned for the future."

In the meantime, researchers are eager to begin using the Keck laser guide star--especially those studying extremely faint, distant galaxies. Astronomers James Larkin at UCLA and David Koo and Eric Steinbring at UCSC have used the Keck AO system with natural guide stars to study a handful of distant galaxies, some of which had already been imaged by the Hubble Space Telescope (HST). HST and Keck provide complementary information, because the HST works best for imaging visible light, or "optical" wavelengths, whereas adaptive optics works best at near infrared wavelengths.

"With adaptive optics, the Keck Telescopes can outperform the Hubble in the near infrared, which means we can now look at these very distant galaxies not only in the optical range with the Hubble, but also with the same precision and sharpness in the near infrared with Keck," said Koo, a professor of astronomy and astrophysics at UCSC.

This is important because the redshift effect, whereby light from distant objects is shifted to longer wavelengths by the expansion of the universe, causes the visible light originally emitted by these distant galaxies to be shifted into the near infrared by the time it reaches Earth.

"When you look in the near infrared with adaptive optics, you're seeing the original visible light from these galaxies, while the optical images from Hubble are actually the original ultraviolet light," said Steinbring, a postdoctoral researcher working with Koo.

The sharpness, or resolution, of the Keck AO images is just good enough to distinguish important structural features of the galaxies, such as spiral arms and central bulges, Koo said. With the laser guide star, astronomers will get that same resolution all over the sky, which will enable them to study many more distant galaxies, gathering valuable clues about the processes involved in galaxy formation. The small sample studied so far doesn't provide much information about galaxy formation, but it does show that adaptive optics is a very powerful tool for studying faint galaxies, Steinbring said.

"It wasn't obvious at first that this would work, so it's exciting that we can do this at all from ground-based telescopes," he said.