Chasing a cosmic mystery:

The mystery began in early December, when the Burst and Transient Source Experiment on board the Compton Gamma Ray Observatory satellite detected bursts of high-energy gamma rays, arriving at the unprecedented rate of 20 bursts an hour. Research Associate Coleman Miller, an expert on gamma-ray bursts, along with his collaborators -- Research Scientist Jean Quashnock and Don Lamb, Professor in Astronomy & Astrophysics -- puzzled over what could possibly generate such a rapid and persistent flux of such high energy. High-energy cosmic gamma-ray bursts rarely come more than once from the same source; they remain such an enigma that whether they originate from within our galaxy or far out in space is still hotly debated. "This is unlike anything we've ever seen before," Quashnock said. "This is an event of extraordinary intensity and persistence."

Because of this source's persistence, the opportunity was ripe for scientists to look at it with other kinds of telescopes, to try to see corresponding bursts of energy in other regions of the electromagnetic spectrum. The more information they had, the easier it would be to determine what cataclysmic event might be causing it.

Quashnock proposed using the telescope owned by the Astrophysical Research Consortium, of which Chicago is a member. The ARC telescope, which is the world's only ground-based remote-controlled telescope, is the brainchild of Donald York, Professor in Astronomy & Astrophysics. The telescope can be completely controlled via the Internet, which not only allows researchers to work from their desktops, without traveling to the remote mountaintop in New Mexico where the telescope is based, but also means that observers from different universities can all use the telescope in the same night. The ARC telescope is also uniquely flexible -- different instruments for the telescope can be changed in a matter of minutes by one person on site, rather than over several hours and with the multiple-person power required by most major telescopes.

When Miller approached Sloan Research Associate Robert Nichol, a seasoned observer, with the idea of using the ARC telescope, Nichol was skeptical. The object was so low on the horizon, viewing it meant the telescope would be tipped far over on its side. For an ordinary telescope, this would put far too much stress on the mirror, perhaps bending it out of shape. But the ARC telescope has a unique, lightweight honeycomb mirror. There was a chance it might work.

Nichol's other objection was that because the object is so close to the sun, viewing it meant pointing the telescope during astronomical twilight, generally a time when no astronomer would even try to observe because the sky is so bright with the approaching dawn. But Miller persisted.

York, Director of the Apache Point Observatory where the telescope is based, was cautiously optimistic. "This is exactly what this telescope was designed for -- to be able to jump immediately on an event of astronomical significance and track it," he said.

The chase was on.

January 12

A group Lamb came to call the "Chicago Collective" -- consisting of theorists Lamb, Quashnock and Miller, along with observers Nichol, University graduate students David Cole and Daniel Vanden Berk and, at Apache Point, Daniel Long, an observing specialist -- had its first successful night. The object rose above the horizon only minutes before the sun, but the observers snapped a couple of pictures with an infrared camera before the rising sun completely washed out their view of the night sky.

Since a source could not be pinpointed, the Chicago Collective researchers snapped blindly in the region they thought was most likely. York said, "It's as if you were trying to look at an area the size of the full moon, but your field of view is only the size of a crater on the moon and you don't know which crater you should be looking at."

January 18

Another satellite-borne instrument got its first look at the gamma-ray source. This experiment, the X-ray Timing Explorer, had only been launched on Dec. 30 and was still in its calibration phase. But the satellite turned to look and found an X-ray counterpart in the same region of the sky. XTE narrowed the box the ARC telescope would have to search to find an optical or infrared counterpart.

January 21

On Saturday morning, Apache Point was clouded out. But early Sunday morning, Jan. 21, the skies were clear again, and team Chicago sprang into action. Nature had granted the observers precious extra minutes, since the object rises slightly earlier each day. They first used an optical camera, and then switched to the infrared. Observing specialist Long had been practicing changing the instruments. It was designed to take half an hour; Long promised 12 minutes and delivered in nine.

The next step was to find an old picture of the same patch of sky and compare. "It's not as if there aren't objects there," said Lamb. "There are a gazillion of them. We were looking for something that brightened significantly, something new that appeared or something that disappeared." The group obtained a picture that had been taken in the infrared several years earlier by a telescope in Arizona.

The pictures in the optical could be compared with pictures taken in 1948 for the Palomar Sky Survey, the most definitive map of the northern sky to date. Staring through a microscope at tiny black dots on a white background (it's much easier to see black on white than white on black, so standard astronomical images are printed as negatives), each observer in turn began checking off stars one by one against a printout of the Chicago data.

Comparing the plates was far from easy. Nichol, Vanden Berk and Cole tried several techniques. They put the pictures on top of one another on a computer screen and blinked back and forth very fast between the two, hoping their eyes could detect minute changes. They also made transparencies -- one positive and one negative -- and overlaid them on a light table to see if there were any patches where light shone through.

Nothing definitive appeared. "But we chased an awful lot of red herrings," said Lamb.

January 23

They tried to observe again on Monday, but couldn't -- the skies were clear, but winds were too high to risk opening the dome -- and then again on Tuesday morning, Jan. 23, but a dense cloud cover obscured the sky.

Late Tuesday afternoon, Bob Nichol cried "Eureka!" He'd found a tiny black smudge on the Chicago optical picture that didn't appear in the Palomar data. Before celebrating, however, the team checked to see if the smudge was an anomaly, caused by dust or a bad pixel in the computer chip that collected the data. They had chased too many false blips not to question everything. By 10:30 p.m., Nichol had convinced himself the spot was real. Lamb gave it a fifty-fifty chance.

The forecast for Apache Point was for clear skies in the morning. Before heading home for the night, the team discussed its observing strategy. One more look, with the telescope offset to correct for any background anomalies, would have confirmed the detection of the optical counterpart to the X-ray source. Chicago would scoop the entire astrophysical community.

Nichol began listing the number of things that could go wrong. Lamb, already exhausted and excited, laughed. "Bob, if you keep talking like that, I know I won't sleep tonight!" he said. They agreed to meet at 6 a.m. the next day. Lamb promised to bring the champagne -- just in case -- and they headed home for a fitful night's sleep.

January 24

At 6 a.m., the group assembled in the Remote Observing Room in the Astronomy & Astrophysics Center. Everyone was there except for Miller, who was attending a workshop at the Aspen Center for Physics, ironically, on the topic of bright X-ray sources. He was standing by to receive Chicago's report and was trying to glean as much information as he could from other theorists and observers at the meeting who were also tracking the source.

Lamb brought donuts and coffee. York, as thrilled as a father to watch his new telescope in action, came to help out.

At 6:38 a.m., everyone was in position. Nichol controlled the telescope via a Macintosh computer screen. Long, at the observatory, was on the speakerphone. Lamb was poised to keep a log. Vanden Berk and Cole prepared to capture data on a Silicon Graphics terminal. Quashnock, pacing nervously in the background, kept an eye on the brightness of the sky using a chart he carried in his hand, while York provided a running commentary.

At 6:50 a.m., Long called out, "It's all yours, Bob." Nichol moved the telescope into position and snapped a picture. Vanden Berk and Cole grabbed the data and brought the picture up on the screen to see if they were pointing in the right direction. Atmospheric distortion from pointing so low on the horizon meant that it was difficult to tell the telescope exactly where to point. It turned out they were right on target. But it was after 7 a.m., and the New Mexico sky would soon be too bright for any more observations.

Nichol snapped another picture, a long exposure -- 30 seconds. Then a series of glitches, including a crash of the controller Macintosh, threatened to eat up precious time. To save time, Long began taking the pictures himself from Apache Point, with direction from Chicago -- Vanden Berk and Cole shouted coordinates, Nichol determined the exposure times.

At 7:25 a.m., the sky was too blue for any more optical observations. But the window for infrared observations would be available for about 20 more minutes. Nichol instructed Long to change instruments. Lamb set his stopwatch; seven minutes and 30 seconds later, the infrared camera was ready to go.

At 7:48 a.m., and 20 pictures later, the observing run was over. It was 6:48 a.m. in New Mexico, and the sun was minutes away from rising.

A sense of relief settled over the group, but the phase of data reduction and analysis had begun. In order to view the pictures, the brightness of the sky and the uneven response of the computer chip had to be subtracted out.

By 8:30 a.m., they had a preliminary answer. The tantalizing smudge seen on Sunday's shots had disappeared. Exhaustion began to show. "My kingdom for a new star!" Nichol quipped. "It just looked so bloody real."

The cold bottle of champagne waiting in the next room went unopened. "It's not as exciting as if we'd found something, but not finding something is an important result, too," Nichol said. "I think we've proved, to the limits that we can detect, that there's nothing new there. That doesn't mean there's nothing there, but if we go deeper into the sky [with a longer exposure], even if we see something there won't be anything to compare it to. We won't know if it's something new or if it's always been there."

Lamb, ever the jubilant theorist, said, "This doesn't invalidate my model. I'll bet anyone a hundred dollars that I know what it is.

"In the cosmic fabric of things, this is telling us about the death throes of stars. And, to borrow a line from Dylan Thomas, they are not going gentle into that good night, they are raging, raging against the dying of the light."

Lamb thinks the object consists of two stars locked in orbit around each other, what astronomers call a binary. One of the stars is a small but very dense neutron star -- the remnant of a sun-sized star that exploded in a supernova and then collapsed to an incredibly dense, dark core about the size of the city of Chicago. It's being orbited by a smaller star that's dying, puffing up so that its outer hydrogen shell is being pulled to the neutron star. Because of the neutron star's incredible density and high magnetic field, the matter is crashing to the surface, igniting thermonuclear explosions that release bursts of high-energy radiation. Because of the timing of the bursts, scientists have already determined that the younger star orbits the neutron star once every 12 days.

"We're looking at the death throes of a star," said York. "We know a lot about how that happens when a star is off by itself, but we don't know what happens when it's in a binary system." Lamb added, "This happens so rarely, but it gives us so much information about how a star dies."

In addition to racing against the clock, Lamb is in a contest with his identical twin brother, Frederick, a theoretical physicist at the University of Illinois at Urbana-Champaign and a mission scientist for the XTE satellite. Both are racing to be the first to figure out exactly what is going on and to submit their papers to the reigning journals. But they're not talking to each other about their work.

Lamb laughed. "I have a sneaking suspicion that I know what Fred's thinking, and there's not a chance he's right," he said. "Of course, he's probably saying the same thing to his group right now. 'I know exactly what Don thinks this is, and there's not a chance he's right.' " But he added, confidently, "The evidence is resonating with my ideas -- it has the ring of truth about it."

In the end, if astronomers have enough time to collect more data before the source fades -- which Lamb expects will take only a matter of weeks -- truth will out.

-- Diana Steele