For nearly 15 years, the hubble space telescope has provided breath-taking images and other data from our solar system and beyond. It has given astronomers a view into the depths of space that hadn't been possible previously. In fact, it has been said that Hubble's impact on astronomy has been greater than that of any other instrument since Galileo's astronomical telescope.

Yet Hubble, for all its impressive results, faces an unclear future.

The renowned telescope is in need of a serious tune-up to repair and replace aging components. In the past, astronauts did this work during space shuttle missions, but the shuttle fleet is currently grounded, and the National Aeronautics & Space Administration has been forced to look for alternatives to extend the telescope's life span, including the possibility of a robotic mission.

Named for famed astronomer Edwin P. Hubble and launched in 1990 aboard the space shuttle Discovery, the Hubble mission was intended to last 15 years. Servicing missions were planned for every three to five years to replace the telescope's batteries and gyroscopic stabilizers, as needed, as well as to upgrade its scientific instruments and repair other components.

Although Hubble suffered some initial setbacks, most seriously a flawed mirror that required a shuttle repair mission in 1993 to outfit its instruments with corrective lenses, the mission has been an astounding success. The telescope, which circles Earth every 97 minutes, now has amassed a series of discoveries not even dreamed of, including the acceleration of the universe's expansion. Hubble has shown us close-ups of the early universe, black holes at the centers of galaxies, and planetary details not visible with ground-based telescopes.

With the prospect of more dazzling discoveries from Hubble, NASA decided to extend its mission lifetime by adding another shuttle servicing mission. The mission, originally scheduled for this year, would have added years to the telescope's life.

Those plans came to an abrupt halt on Feb. 1, 2003, when the space shuttle Columbia broke apart during its reentry into the atmosphere, resulting in the deaths of all seven astronauts on board (C&EN, Feb. 10, 2003, page 6). The agency immediately grounded the space shuttle fleet until it could determine the cause of the tragedy and figure out how to keep it from happening again. The group charged with this task, the Columbia Accident Investigation Board, released its report seven months after the accident (C&EN, Sept. 1, 2003, page 6).

After reviewing the report, NASA Administrator Sean O'Keefe announced in January that the scheduled Hubble shuttle servicing mission would be canceled because of safety risks. That decision meant that the telescope would have just two to three years of scientific life remaining before its components--mainly, the gyroscopes--would start to break down, leading eventually to the scope's demise.

According to Michael L. Weiss, deputy program manager for the Hubble program at Goddard Space Flight Center, Hubble has six gyroscopes but needs only three to be effectively stabilized. Now, two gyroscopes are down because of normal wear and tear. A third is predicted to fail in mid-2006.

STANDING TALL These pillarlike structures are columns of cool interstellar hydrogen gas and dust and are part of the Eagle Nebula, a star-forming region that is 6,500 light-years away from Earth.

NASA PHOTO

This potential loss of Hubble's scientific capability caused many of its supporters to cry foul. "Canceling the final servicing mission for Hubble is major surgery. It is extreme. It is irreversible," said Sen. Barbara A Mikulski (D-Md.) at a hearing following O'Keefe's announcement.

The biggest concern from the scientific community is that allowing Hubble to die before the next-generation space telescope, the James Webb Space Telescope, is launched in 2011 will result in the loss of a window into deep space. Moreover, the Webb telescope will image the infrared regions of the spectrum, whereas Hubble's instruments are tuned primarily to the visible regions. That makes Hubble unique among orbiting space telescopes.

"I don't think we have an easy replacement," says Steven Beckwith, director of the Space Telescope Science Institute (STScI), which manages Hubble's science program. "Hubble continues to provide the best angular resolution [of any telescope.] To give up that capability would be a tragic loss."

"It's always hard to see a very good, useful tool go away," says William B. Latter, an astronomer at California Institute of Technology who uses Hubble to study the late-stage evolution of stars. "It's a little difficult to accept that it could just be let go."

On the basis of an outpouring of support for Hubble's potential for continued scientific discovery, NASA reevaluated its decision to cancel the servicing mission in March and began looking at alternatives to accomplishing such a mission without the shuttle.

The agency also asked the National Academies for help in reviewing the plausibility of a shuttle servicing mission and evaluating the possibility of using robotics and other ground operation activities to extend the telescope's life. In the interim report, released in July, the committee conducting the study underscored the importance of Hubble and advised NASA to keep the option of a shuttle servicing mission on the table until robotic missions and the shuttle return-to-flight program are better assessed (C&EN, July 19, page 15).

Response by Hubble supporters to the recommendations of the interim report was optimistic. And as concerned parties await the final report--due late November or early December--the outlook remains positive. "I think saving the Hubble should be a priority, and I look forward to hearing from the National Academies and NASA as the process goes forward regarding the feasibility of a robotic servicing mission or other means to preserve the telescope's scientific missions," House Science Committee Chairman Sherwood L. Boehlert (R-N.Y.) tells C&EN.

NASA, too, is encouraged about the National Academies' support for a Hubble servicing mission, notes Michael R. Moore, program executive for Hubble. In addition to exploring the possibility of a robotic servicing mission, the agency is also keeping its options open for a shuttle mission if it's ultimately deemed safe, he says. NASA is also moving forward with a "deorbiting" mission that would bring the telescope back down to Earth in a controlled way at the end of its lifetime.

NASA'S MAJOR focus, however, is on determining if it can put together a robotic mission on a short timescale, Moore says. According to him, the estimated cost for such a mission is $1.3 billion, although independent estimates say this figure is closer to $2 billion. Because such a mission would have to take place by the end of 2007, this money would be needed relatively quickly.

The situation would be different if NASA decides to perform only the deorbiting mission. NASA estimates that this mission would cost $400 million to $500 million. The timescale for this mission would also be longer because the batteries--which control the vehicle life--aren't projected to expire until 2009, Moore points out. He didn't provide the cost estimates for a potential shuttle servicing mission because it is currently not an active option.

In any case, coming up with the necessary funding quickly may not be easy for NASA, because the fiscal 2005 budget is already set and does not include extra funds for a Hubble rescue mission. Moore points out, however, that NASA's lack of funds has not caused any slowdown in work on a rescue mission so far. "We will find the money somewhere in the agency to continue this work until we make a decision to go forward further or to end this particular exercise," he says.

NASA is aggressively pursuing its options. The agency has already awarded a pair of contracts for work on the project. Lockheed Martin was commissioned to develop the deorbit module, Weiss says, which will be necessary regardless of whether a servicing mission is done.

A second contract was awarded to Canada-based MD Robotics to design both a robotic arm to capture Hubble and a dexterous robot (called Dextre) to service the telescope. The joined system of the robotic components, the replacement hardware, and the deorbit module is called the Hubble Robotic Vehicle.

According to Weiss, the Hubble Robotic Vehicle will be lofted on an expendable launch vehicle and, once in space, will rendezvous with and capture Hubble. This would be the first demonstration of a robot capturing an uncontrolled spacecraft in orbit--an important capability for future space exploration objectives. At the end of the servicing mission, anything that is not needed to do the controlled reentry will be ejected, he says. The telescope will then remain in orbit until its scientific capabilities expire.

A preliminary design review is scheduled for March, Weiss says. Between now and then, he notes, those involved are continuing to refine mission designs and cost estimates.

If NASA decides to move forward with a robotic mission and finds itself short on funds, most people believe Congress will step up and fill the gap. "Hubble is an easy sell in Congress," says Alex Roland, professor of history at Duke University. "If NASA had a realistic plan and said to Congress, 'We're a billion short to save the Hubble,' I think they could probably get that money," he notes.

This current crisis is not the first hurdle that Hubble has faced. From the time the project was given the green light by Congress in 1977, this telescope has pushed the envelope. "NASA started down the path knowing that building a large telescope would be a great thing," Beckwith says. "As the agency started building the telescope, it became more difficult and more expensive than NASA had predicted, primarily because there was a lot of technology development to be done," he explains.

After several delays and at a construction cost of $1.5 billion, Hubble was ready to launch in 1986. Unfortunately, the shuttle flight that preceded its launch was the ill-fated Challenger. After an additional four-year delay, Hubble was finally launched in April 1990, having already cost taxpayers $2.5 billion, according to NASA.

IN THE FEW WEEKS following Hubble's launch, all the instruments onboard were put through their paces. But when Sandra Faber, astronomy professor at the University of California, Santa Cruz, and her colleagues took a picture with Hubble's Wide Field & Planetary Camera, each star had a large, fuzzy halo. The image "was released by NASA as a great success, but privately we thought the picture didn't look right," Faber said.

A series of test images showed a fatal optical flaw. "It was an absolutely classic spherical aberration," Faber said. "At that point, we knew what was going on."

The source of the flaw was mismanufactured test optics. Faber's team designed corrections for the Wide Field & Planetary Camera using a duplicate they had in the lab. Corrective "glasses" were also made for the other three Hubble instruments. Then, astronauts aboard the space shuttle Endeavour undertook a difficult and delicate space walk, installing the new camera and a module with the other instruments' corrective mirrors.

After the repair mission, Faber's then-postdoc Jon Holtzman unveiled a crystal-clear image of a star at the American Astronomical Society's annual meeting. "It was a great moment of glory," Faber says, and "2,000 people got to their feet and roared."

Since then, Hubble has redeemed itself as one of the premier astronomical achievements of humankind. Astronomers are still lined up to use it.

The science projects that Hubble does are chosen annually via peer review. According to Beckwith, STScI gets about 1,000 proposals per year but can only afford to give time to between 150 and 200 of them.

Sometimes, the discoveries are completely unexpected. "About 50% of Hubble's major accomplishments were not predicted at the time Hubble was launched," says David Leckrone, senior project scientist for Hubble at NASA's Goddard Space Flight Center, who has worked on the telescope since 1976.

Perhaps the most startling example is the finding that the universe is expanding at an accelerating rate. Until the 1990s, scientists largely believed that the universe was decelerating. Measuring the universe's expansion rate had been one of Hubble's goals, even before launch. To that end, astronomers used both ground-based telescopes and Hubble to study a range of supernovas, or exploding stars, at different distances from Earth.

HUBBLE'S ABILITY to image supernovas extremely far away enabled astronomers to determine that the most distant supernovas are speeding away much faster than the closer ones. Hubble's law, named after the astronomer, who derived it in 1929, holds that the speed at which galaxies recede from Earth increases linearly as they become more distant. But the new Hubble telescope findings show that this speed is increasing even faster than that.

Leckrone likens it to throwing a baseball in the air and, instead of having it come down, watching it speed away.

To explain this phenomenon, cosmologists are invoking a "dark energy," an unknown repulsive force. At one point billions of years ago, the universe's expansion was actually decelerating, they believe, as gravitational attraction of the universe's mass slowed the expansion. But over time, this mysterious repulsive force began to overpower the attractive gravitational energy. The nature of this energy remains one of the biggest puzzles in cosmology, Leckrone says.

Another major accomplishment of Hubble was being the first telescope to spot the telltale signatures of supermassive black holes at the centers of galaxies. Black holes are dense collections of matter with gravitational fields so intense that light can't escape from them. Faber's team discovered the majority of black holes. Identifying black holes is very difficult, as they can't be seen directly. But they can be inferred from the behavior of nearby stars, which zip around faster than would normally be predicted. "The high resolution of Hubble allowed us to see the excess motion," Faber says.

Unencumbered by atmospheric interference, Hubble also has been able to peer farther into deep space than any other telescope. A series of surveys--intense scans of relatively unexplored regions of the sky--culminated this year in the Ultra Deep Field exposure, which pushed Hubble to the limit of optical observation. The telescope's Near Infrared Camera & Multiobject Spectrometer and its Advanced Camera for Surveys focused on a dark section of sky for a total of 1 million seconds over four months. Hundreds of objects in a chaotic jumble--galaxies that came into being when the universe was 400 million to 800 million years old--revealed themselves (C&EN, March 15, page 9).

Hubble's pioneering discovery with the deep-field survey will be followed by more surveys even further back in time. The Webb telescope will take over where Hubble leaves off, exploring deeper fields in the infrared regions of the spectrum, going back to the earliest days of the universe.

The spatial resolution of Hubble has also made possible the study of planetary nebulae, which are gas clouds expelled by dying stars. The outer envelopes of stars such as these have exhausted their nuclear fuel and are puffing out into space.

Until recently, it was believed that dying stars ejected this material in a uniform, spherical way. But Hubble has found a myriad of bizarrely shaped planetary nebulae, leading astronomers to give them colorful names, such as the "Rotten Egg Nebula" and the "Dumbbell Nebula." As a result of these discoveries, scientists are rethinking physics. "Once we had a good look, we found out it's not simple at all," Latter says. "We're scrambling to re-create a model of planetary nebula formation."

EYE OF THE STORM A black hole likely sits at the core of this galaxy in the Virgo cluster, sucking up material that swirls around it.	GREAT SPOTS! Hubble captured a rare triple-moon eclipse on Jupiter, imaging the shadows of Io, Ganymede, and Callisto as they passed across Jupiter's face.
	NASA PHOTOS

ASTRONOMERS HAVE long theorized that planets are formed from accreting disks of dust and gas orbiting a star. Hubble confirmed that prediction by directly imaging disks around young stars in the Orion nebula.

In addition to discovering more than 100 extrasolar planets, Hubble was also the first telescope to identify an atmosphere on a planet outside our solar system, detecting the spectral lines of sodium, carbon, and hydrogen around the massive planet HD209458, which is 150 light-years away.

Closer to home, the telescope has captured some of the most memorable solar system events of the decade. When the fragmented comet Shoemaker Levy crashed into Jupiter in 1994, Hubble was there to capture the bull's-eye-shaped impacts generated by the comet pieces as they plunged through Jupiter's atmosphere. "That was just so exciting--I got about two hours of sleep a night for a week, and it was worth every second," Leckrone says.

Most recently, Hubble captured a rare triple-moon eclipse on Jupiter, imaging the shadows of Io, Ganymede, and Callisto as they passed across Jupiter's face.

Astronauts have visited Hubble four times for makeovers, including the optics repair mission in 1993. During the last service mission, in 2002, astronauts installed the Advanced Camera for Surveys, an instrument that is more than 10 times more powerful than its predecessor.

Engineers purposely designed Hubble to be repeatedly refurbished. "Generally, it's very hard to build new big telescopes," Beckwith explains. "It's much easier to build smaller things, like instruments to go in existing telescopes. The pace of improvement of technology is such that you can often improve an observatory by a huge factor just by putting better instruments on an old telescope," he notes.

And that's just how Hubble has evolved. "The capabilities of the telescope now are several orders of magnitude more than when the telescope was first launched," Beckwith says, "and it's because of these refurbishing missions."

In fact, working with the telescope since the last servicing mission in 2002 was "like starting all over again," states Keith S. Noll, an astronomer at STScI who studies Kuiper Belt objects, which are icy bodies that orbit beyond Neptune. With increased resolution of four pixels for each previous one, there was "almost no comparison in quality," he says.

Hubble's serviceability is unique in the space telescope world. Other space-based observatories, like the Chandra X-ray Observatory, the Spitzer Space Telescope, and the upcoming Webb telescope, orbit much farther from Earth: Hubble's orbit is currently about 350 miles, whereas the others' orbits are orders of magnitude higher. Those telescopes have less interference from Earth's warmth and atmosphere, but it isn't possible to send a shuttle out to fix or improve them.

ART GALLERY The Universe Is A Pretty Picture HOLY CATS The Hubble Heritage Project produces striking images, such as this 'Cat's-Eye Nebula,' for posterity and science. NASA PHOTO The big thrill of the Hubble telescope is, of course, the pictures. Glowing, opulent images of stars and galaxies are tangible evidence of the astronomical wonders that Hubble is uncovering. This steady stream of arty images provided for our enjoyment is generated largely by the Hubble Heritage Project. Headed by Keith S. Noll, astronomer at the Space Telescope Science Institute in Baltimore, the project once a month releases an image designed to elicit that 'ooh, aah' response. 'We think of ourselves as nature photographers using the world's coolest camera,' Noll says. The project's genesis six years ago stemmed from the realization that a lot of the cutting-edge science done with Hubble, such as spectroscopy and imaging of faint objects, doesn't necessarily produce great images. So the group members, not all of whom are scientists, find exposures, most culled from an archive of Hubble data, that have potential to be great art. They then process and edit them, adding color and contrast. 'We all take turns suggesting things,' Noll says. The group also gets a small amount of coveted time each year so they themselves can take images of objects that not only are beautiful but also can be used to do science.

SOME EXPERTS question the wisdom of a serviceable telescope. The Hubble mission has essentially been hostage to the shuttle program, so the telescope has perhaps accomplished less than it could have, some believe. "Because Hubble was designed for the shuttle, it actually made the space telescope program more vulnerable than it needed to be," Roland says.

Hubble could have been launched on an expendable launch vehicle, Roland says, and could have been placed in a better orbit and therefore have produced better science. Additionally, for the price of repairing the space telescope, NASA could have built multiple space telescopes, he says. "If one deteriorated or malfunctioned, we could simply replace it with another one more cheaply than going to repair it with the shuttle," he points out.

NASA is now looking at what procedures from the shuttle servicing mission's manifest could be done in a robotic servicing mission. Priority is given to those procedures related to the health and safety of the vehicle, according to Weiss. The next level of priority goes to items that enhance the scientific capabilities of the telescope. And finally, fixing failed or degraded components is considered.

Based on this set of priorities, the first objective would be to replace the batteries and the gyroscopes, followed by the installation of the two new scientific instruments: Wide Field Camera 3 and Cosmic Origins Spectrograph. At this point, replacing the telescope's Fine Guidance Sensors and fixing components like the Space Telescope Imaging Spectrograph, which failed in August (C&EN, Aug. 16, page 17), would be considered, Weiss explains.

"If we can pull off this robotic mission, Hubble will be even more powerful than it has been before," Leckrone says.

And that's what the folks at NASA are working toward and are optimistic will happen. "We have a team at Goddard that has pulled things off in the past that seemed to be almost impossible," Moore says. "If anyone can do it, they can," he notes.

"People really have a sense of ownership of Hubble, globally," Noll says. "So I hope that we keep this going as long as it's feasible. We've really only started to scratch the surface of what we can do."

INTO THE DEEP Hubble captured this deep-field exposure—which took 1 million seconds—of a collection of oddly shaped galaxies that existed when the universe was 400 million to 700 million years old.

NASA PHOTO