Scientists Work Around Telescope's Fuzzy Images

Computer processing, prolonged exposure will improve focus, they say

WITH the initial shock of the Hubble Space Telescope's flawed focus behind them, observers are preparing to make the most of the crippled instrument. Their options brighten the promise of Leonard Fisk, associate administrator of the National Aeronautics and Space Administration, that the telescope will do ``excellent and unique science.'' In fact, says Peter Stockman, deputy director of the Space Telescope Science Institute, ``I'm optimistic we can do good science now.''

NASA's quickly assembled Image Processing Working Group expects computer processing to improve by two to three times the ability of Hubble images to resolve detail.

Research that does not depend on razor-sharp images also can proceed. This includes using Hubble's guidance sensors for precision measurement of the position of stars. Many planned studies in ultraviolet light should be relatively unimpeded.

Prolonged exposures may allow some visible-image research to compensate for the fuzzy focus. Also, Dr. Stockman notes, it's still possible to recover essentially the original resolution of the telescope with relatively bright objects.

Peter Jakobsen, project scientist for the European Space Agency (ESA) which is NASA's Hubble partner, also expresses optimism. He says he is certain that ``there is more than plenty of exciting science to do to keep the observatory more than busy until the problem can be fixed with the next generation of [Hubble] instruments.''

These instruments, which use the telescope's images, include a high-speed photometer to measure brightness; spectrometers that spread light into its various colors to reveal the ``fingerprints'' of chemicals; and cameras, notably NASA's Wide Field/Planetary Camera designed to image dozens of distant objects at a time and provide closeups of planets and ESA's Faint Object Camera for imaging stars five times more distant than can ground-based facilities.

The program has planned all along for astronauts to install improved instruments. Now designers will add optical devices to sharpen the images. But it will be several years before any Hubble service mission lifts off. Meanwhile, observers must make the most of what they have.

The telescope's main mirror appears to have spherical aberration. Light rays don't all come to the same focus. Hubble was supposed to focus at least 70 percent of a star's light into a spot 0.1 arc-seconds across. Instead, only about 15 percent of the light makes it into the ``winner's circle.'' The rest smears out into a surrounding halo. This prevents the $1.6 billion observatory from resolving detail 10 times more sharply than present ground-based telescopes as promised.

But the 15 percent of light that is sharply focused does provide high-resolution information with which computer image-enhancing programs can work. University of Arizona computer engineer Bobby R. Hunt, who chairs the image-processing group, notes that ``if you have two point sources that are so close together that they're blurred, ... a factor of two or three in resolution might make it possible to separate the two points.''

He explains that, right now, his group is trying to learn all it can about the Hubble instruments and the image degradation and ``to sort out which algorithms to apply and in what situations.'' He says it is not seeking a miracle cure. But he believes some image improvement is possible.

Another way to beat the blur, in some situations, is to expose images longer to bring up the sharply focused 0.1 arc-second image core. Also a spectrograph, which views images through a slit, can narrow that slit to exclude much of the blurry halo surrounding the image core. Again, much longer exposures would be needed.

Accommodating longer exposures in the observing schedule - as much as three to four times the original observing time in some cases - means curtailing or deferring some other research. Stockman says that about 55 percent of planned observing programs are still feasible in terms of time. The other 45 percent must wait for new Hubble instruments.

He adds that 141 investigators in 82 countries are now restudying their plans and will submit revised proposals. After further review, he expects to have the first year's observing schedule for these scientists ready by next March. He further expects the ``guaranteed time'' observers to begin serious observing in October or November. These are scientists who spent years helping develop the telescope and were guaranteed first priority in its use.

Looking ahead to the second-generation Hubble instruments - and even to a third-generation - Stockman says he expects ``eventually to get almost all the originally planned research done.''

We want to hear, did we miss an angle we should have covered? Should we come back to this topic? Or just give us a rating for this story. We want to hear from you.