Contact: Tim Stephens
University of California, Santa Cruz

SANTA CRUZ, CA–The desire for ever larger telescopes has driven astronomers relentlessly for nearly 400 years. In 1610, Galileo discovered the moons of Jupiter using a telescope with a lens not much more than an inch in diameter. Today, the largest optical telescopes are the twin Keck Telescopes in Hawaii, with 10-meter mirrors that gather the faint light from distant galaxies. But, of course, astronomers still want bigger telescopes, knowing they can yield new insights into the nature and origins of the universe.

The next milestone in telescope size is likely to be one with a primary mirror 30 meters in diameter, which would provide ten times the light-gathering area of each of the Kecks. The University of California and the California Institute of Technology (Caltech) have teamed up to design and build a 30-meter telescope, dubbed the California Extremely Large Telescope (CELT). The project is still in the early planning stages, but researchers led by UC Santa Cruz astronomers are making steady progress on the conceptual design for CELT.

Project leaders hope to build the massive telescope within the next ten to 15 years, said Joseph Miller, director of UC Observatories/Lick Observatory (UCO/Lick). Miller estimated the total cost of the project would be around $500 million, but it’s not clear yet where that money would come from. UCO/Lick and Caltech are funding the current design studies.

“The scientific potential for a 30-meter telescope is so great that I’m confident we can obtain the necessary funding to build it once we have a detailed proposal,” Miller said.

CELT would have a broad range of applications, according to Jerry Nelson, professor of astronomy and astrophysics at UCSC. It would enable astronomers to probe deeply into nearby star-forming regions to study the births of stars and possibly even to directly image planets around nearby stars. At the other extreme, astronomers could use CELT to study the most distant galaxies, looking far back in time to understand how galaxies and stars formed early in the history of the universe. “And you name it everywhere in between,” Nelson said.

The task for researchers now is to come up with a solid design and a credible cost estimate, said Nelson, who chairs a CELT working group focusing on telescope design issues. There are six working groups made up of UC and Caltech scientists focusing on different aspects of the project. Miller cochairs the project steering committee with Caltech’s Chuck Steidel. Most of the design work to date has taken place at UCSC, where UCO/Lick is headquartered.

CELT will use much of the same technology that has proven so successful in the Keck Telescopes, said Terry Mast, a research physicist at UCO/Lick. Like the Kecks, for example, CELT will have a segmented primary mirror made up of many small, hexagonal mirrors arranged in a honeycomb pattern. (This mirror design, a revolutionary innovation for the Kecks, was Nelson’s brainchild.) Instead of the 36 segments in each of the Keck mirrors, CELT’s mirror will have 1,080 segments.

But CELT will not be simply a scaled-up version of the Kecks. That would be far too expensive, according to Mast.

“The major challenge is cost,” Mast said. “We don’t have to invent new technology, because we think much of the same technology used for the Kecks will work for CELT. But if you just scale up in a naive way you will end up spending a lot of money. We have to come up with a design that is practical and cost-effective.”

Cost and practicality figured into the decision to design CELT with a 30-meter mirror, rather than, say, a 50-meter or 100-meter mirror. The European Southern Observatory, based in Germany, is pursuing a plan to build a 100-meter telescope called the OWL (for “Overwhelmingly Large”) telescope. Based on the historical relationship between telescope size and cost, however, Nelson figured that a 30-meter telescope could be built for 5 percent of the cost of a 100-meter telescope.

“It would be foolish not to build a 30-meter telescope first,” he said.

With the segmented-mirror design, there is no fundamental limit to telescope size. It’s just that as the telescope gets bigger it becomes more difficult and more expensive to build.

“The sky’s the limit, but my logic has been that if you take steps that are too large you run the risk of making mistakes and using money ineffectively,” Nelson said. “Going from 10 meters to 30 meters is an attractive step size, and a 30-meter telescope is going to be fabulous scientifically.”

Although astronomers have a good idea of what CELT’s capabilities will be, many of its most exciting applications probably won’t be discovered until after it has been built, said Michael Bolte, a professor of astronomy and astrophysics at UCSC.

“Who knows what we’re going to discover? Previous experience with major advances in telescopes suggests that many of the most exciting new discoveries will be in unanticipated areas,” Bolte said.

Clearly, however, one of the major applications will be to explore the most distant galaxies. The Hubble Deep Field, a revolutionary image of deep space obtained by the Hubble Space Telescope, revealed a wealth of very faint objects. Observations with the Keck Telescopes showed some of them to be extremely distant galaxies, but a more powerful telescope is needed to study them in detail. UC and Caltech astronomers have used the Keck Telescopes to study about 10 percent of the objects in the Hubble Deep Field, but the rest are too faint to probe with the Kecks. With CELT, astronomers will be able to study the faintest smudges of light in the Hubble Deep Field, some of which may show the earliest stages of galaxy formation in the universe.

“We will be able to see the evolution of galaxies from when they were born, more than 10 billion years ago, right up to today,” Bolte said.

A critical component of CELT will be an adaptive optics system to correct for the blurring effects of the atmosphere. “That’s probably the hardest part, because adaptive optics has a long way to go,” said Nelson, who directs the Center for Adaptive Optics based at UCSC.

Adaptive optics (AO) relies on extremely rapid and precise measurements of atmospheric distortion coupled with equally rapid and precise compensation by the telescope. This compensation is achieved with a deformable mirror (separate from the primary mirror) that changes shape to correct the distortion of the image. As the telescope gets larger, the requirements for an AO system become increasingly demanding. For example, the deformable mirror for the Keck AO system has about 250 actuators–tiny pistons that control the mirror’s shape–whereas the CELT deformable mirror will need about 5,000 actuators. Existing AO technology will not be adequate for CELT, Nelson said.

“There are a lot of issues with adaptive optics and a lot of ideas for how to resolve them. As long as there is sufficient funding for adaptive optics research, the technology will be here when CELT is ready for it,” he said.

Other issues researchers are grappling with during the design phase include the fabrication process for the primary mirror segments; how to contain the cost of the 6,000 sensors and 3,000 actuators needed to keep the mirror segments aligned; and how to design the telescope to withstand the wind forces typical of mountaintop observatories.

Making the mirror segments will be one of the most costly parts of the project, according to Mast. Although the mirror is symmetrical overall, the individual segments are not, making the polishing of their optical surfaces extremely challenging. “It will require a very close collaboration with the manufacturer to ensure that they can deliver what we need,” Mast said.

Interestingly, the computers and electronics needed to control the huge telescope are not a major concern. “The technology we think of as in the forefront, such as electronics, will be easier and less expensive than the older technology, such as polishing mirrors. That’s still part art and part science, and it is still very expensive,” Mast said.

Project leaders have not yet determined where CELT will be built. According to Nelson, the possibilities include Mauna Kea, where the Keck Telescopes are located, and several sites in Chile.


Editor’s note: Reporters may contact Miller at (831) 459-2991 or and Nelson at (831) 459-5132 or