by Mark Bowen
It was as if the light particles, the photons, were bouncing around randomly like pinballs off millions of tiny mirrors, describing what are called random walks through the ice. This meant that any kind of light, whether it was a pulse from the end of an optical fiber or a Cherenkov cone from a muon, would diffuse in all directions, rather than travel in straight lines. Demonstrating his knack for vivid, and in this case aromatic metaphors, Bob writes, “It was not like the sound of a fart crossing the room, which is what we hoped for and expected, but it was like the diffusing smell of the fart arriving much later.”
He, Per Olof, and Steve did some back-of-the-envelope calculations that showed that the light was hitting a bubble every twenty or thirty centimeters. It was as if their instrument was looking at the sky through a pane of frosted glass (another of Bob’s metaphors). There was plenty of light, but no information. By the time any photon reached an optical module it had lost memory of its original direction. There was no way they would be able to build a telescope—in this ice, anyway.
Although the scattering was definitely a show-stopper, the three physicists’ quick analysis also revealed a “bright” side: as the average photon followed its random path through the ice, it was traveling almost ten times farther than the actual distance between strings, and the simple fact that it made it from one string to the next indicated that it was not being absorbed over that long a distance. This told them that intrinsically the ice was an excellent medium for neutrino detection, if only they could get away from the bubbles. And that meant drilling deeper.
After deploying four strings, they stopped near the end of January, figuring it wasn’t worth the effort to deploy any more in view of the scattering.
And in spite of the bad weather, it was a long season. Bob’s logbook tells him that he arrived at Pole on October 29 and left on February 21. His notes also include “a lot of wonderful and sometimes really snide comments about a host of people … because we had to drink that shitty, twice-frozen Bud Light beer at the Pole.”
10. A Supernova of Science
This was the first time Francis thought the project was going to die. (There would be three more.) They had done a fine job with the drilling, and they had managed to deploy working hardware, but they had “goofed up” by placing their instrument in shallow ice. They were also in debt, and a thought that had occurred to him and Bob the previous fall came back to haunt them. During the tense financial discussions with Buford and Steve, they had suggested that it would be “better to borrow than to fail, for any type of failure will hamper future funding.” Now some could argue that they had failed—and a few of their rivals did just that.
This was probably the most competitive moment that the small world of neutrino astronomy will ever see. There were four neutrino telescopes in development at the time, DUMAND, Baikal, AMANDA, and NESTOR, the last led by Leo Resvanis, the Wisconsin-style physicist from Greece. NESTOR’s proposed site was in the Ionian Sea, off the Peloponnesian town of Pylos, and Leo had gone a good distance out of his way to name his instrument after the mythical king of the town, who appears in the Iliad: NEutrinos from Supernovae and TeV sources, Ocean Range. Nothing much of a practical nature had transpired on NESTOR by that time, and not much more has transpired since, although Leo has come up with some creative ideas.
The most prominent of the projects, DUMAND, had just begun to well and truly flail. In mid-December, six days before Bruce and his crew had begun drilling the first hole for AMANDA-A, the DUMAND collaboration had failed miserably in an attempt to deploy the first string for DUMAND-II. After numerous mishaps, they had lost contact with the string about five hours into their complex deployment operation, having obtained only two minutes of seriously compromised data.
The way they deployed a string was to attach anchors to the top and bottom ends, lower it to the ocean floor, and release the top end from its “sacrificial anchor” by remote control. A buoy at that end would float up and hold the string vertical. Well, the anchor at the top of the first string failed to release, so what data they did get came from an arch-shaped, inclined string—which proceeded to go silent two minutes after they began listening to it. They managed to retrieve it late in January, whereupon an inspection revealed a small leak in one of the nearly sixty connectors on the “string controller.” (That many connectors was asking for trouble.) In February, another piece of hardware failed: the “junction box” on the ocean floor, which transferred the electrical signals from the light detectors to the optical cable that carried the signals to shore.
Typifying their high-tech approach, they had placed video cameras three miles down on the ocean floor in order to watch the deployment from their boat. Sometime later, an AMANDA post-doc somehow got ahold of some especially damaging footage from a different deployment, showing a robot picking something up, dropping it, and stirring up all kinds of dust and muck from the ocean floor. “It just looked horrible,” he said. “It kind of explained how the … it didn’t look good.”
The general picture was one of daunting objective challenges, excessively complicated hardware, and inadequate operating procedures.
* * *
Baikal by contrast had continued to make steady progress in the face of daunting political challenges.
When we last visited the lake, 1988, the Soviet Academy of Sciences had elevated the status of the project and increased its level of funding, and Christian Spiering’s group from East Germany had joined the collaboration. The strengthened collaboration had then stepped back and designed a small array that they called NT-200 (Neutrino Telescope with “about” 200 optical modules; it ended up with something like 192).
Then, in November 1989, the Berlin Wall came down. Funding shrank rather than grew, and the Soviet scientific establishment, which had once been one of the greatest in history, began to unravel.
Christian writes,
The construction of NT-200 coincided with the decay of the USSR and an economically desperate period. Members of the collaboration and even some industrial suppliers had to be supported by grants from Germany; nevertheless many highly qualified experimentalists left the collaboration and tried to survive in the private sector. Over a period of three years, a large part of the food for the winter campaigns at Lake Baikal had to be bought in Germany and transported to Siberia. Still, a nucleus of dedicated Russian physicists heroically continued to work for the project. Under these circumstances, the construction of NT-200 extended over more than five years.
He is understating his group’s influence: they saved the day. The crumbling of the wall signaled the reunification of Germany. Christian’s small institute in Zeuthen became part of the former West Germany’s high-energy physics establishment, Deutsches Elektronen Synchrotron, or DESY (“Daisy”), and they found themselves flush with cash. They also gained access to western markets for electrical components, glass pressure spheres, and the like, which they could now transfer to Baikal. Christian himself smuggled tens of thousands of deutschemarks across the border in his pockets on his visits to the lake, and the food that his group sent to their Siberian colleagues was packed in crates labeled as scientific equipment.
The Russian government is notorious for paying its workers late. This wasn’t so bad in Moscow, where paychecks generally arrived within a month of their due date, but at the University of Irkutsk, the closest collaborating institution to the lake, it sometimes took six months, by which time deflation of the ruble had rendered the checks nearly worthless. And Russian scientists weren’t paid handsomely to begin with. They earned less than the engineers who worked at their sides and only 20 or 30 percent more than the people who cleaned their labs.
Those who remained in Irkutsk supplemented their income by tutoring the children of wealthy neighbors, installing security systems in the same people’s homes, or working in the construction industry. Even Nikolaij Budnev, the leader of the group (he of the ingenious device for rescuing the string from the bottom of the lake), came nowhere near s
urviving on his university salary, which amounted to about twenty U.S. dollars a month. Zeuthen scientist Ralf Wischnewski gave him his old car, which Budnev drove as a taxi for three hours every morning and evening, before and after his academic work. Zeuthen also paid him a stipend.
Despite these hardships, in March and April of 1993, Baikal established a beachhead in neutrino astronomy by deploying the world’s first three-string Cherenkov detector. (You need at least three strings to reconstruct a three-dimensional muon track, and one of them needs to be out of the plane defined by the other two. In other words, they need to be arranged in a triangle.) This was about nine months before DUMAND’s disastrous deployment attempt and ten months before Bob Morse, Per Olof Hulth, and Steve Barwick realized that they had placed four strings in bubbly ice. John Learned and Francis Halzen demonstrated the collaborative nature of all good science, and especially of this field at that point in time, by sending e-mails to Christian, congratulating Baikal for “winning the three-string race.”
Not long after Christian received Francis’s e-mail, his group commenced a more direct collaboration with AMANDA. They employed software they had developed for Baikal to run computer simulations of its icy sibling. They refrained from officially joining AMANDA just yet, however.
* * *
In late February 1994, with Baikal’s success and DUMAND’s failure as a backdrop, and hot on the heels of AMANDA’s second seemingly disastrous field season, the Sixth International Workshop on Neutrino Telescopes took place in Venice.
Perhaps you’ve noticed that four years into the project Francis Halzen still had not visited Pole. Well, more than twenty years later, he still hasn’t. He’s always said that it would be a farce, since there would be nothing of a practical nature for him to do (although he does add, jokingly, that he’ll go when there’s a good French restaurant there). Still, one would think that once or twice in almost thirty years the leader of a project might find the time for a symbolic visit at least, or had some curiosity about seeing his creation at first hand. I suspect that there’s a deeper reason for Francis’s reticence, and that is that he wants to send a message that the real adventure is in the science, not the derring-do of traveling and working in Antarctica. His job—and it’s been a harrowing one—has been to keep an eye on the big picture and the highest levels of the physics.
Francis was in Venice, of course. He was scheduled to give his usual theoretical talk about high-energy neutrino astronomy: sources, how many neutrinos they should produce, and so on, and he planned to include a few remarks about AMANDA, by the by. Bob Morse had been scheduled to talk about recent progress, but he was still traveling north from his long season at Pole, so Adam Bouchta, the young Swedish graduate student, took his place. It was Adam’s first big conference.
This was a delicate moment, obviously. It is important to proceed deliberately in science, and the data from Pole was very fresh. Even if it had been clean and clear and just what they had expected, they might not have presented it at this meeting (well, probably they would have). It would turn out that the scientists on the Ice were correct about the bubbles, but even they admitted that their calculations were crude and that other explanations, including hardware malfunctions, had not been be ruled out.
They had not shared the bubble hypothesis with Francis. He was aware of the confusing data—it had been sent north by e-mail—but he had not spoken to Bob, Per Olof, or Steve, and neither had he found time to focus on the data himself. He claims that he did not hear the word “bubbles” before flying to Venice.
The two AMANDA talks were scheduled for the last day of the meeting. On the previous day, Francis happened to chat with Grigorii Domogatsky, the leader of the Baikal project, and Domogatsky told him about a paper about the Vostok ice core he had recently run across in an obscure Russian glaciology journal. It reported that between eight hundred and a thousand meters down—the depths of AMANDA-A—the bubbles at Vostok were about fifty microns in diameter, a twentieth of a millimeter. It seems that the American glaciologist Tony Gow had been right about the density of the bubbles, how many there were per unit volume, but wrong about their size; he’d figured they were only about one micron in diameter. A fifty micron bubble would scatter about fifty times more light.
This got Francis to thinking. In his hotel room that night, overlooking the Grand Canal, he worked out the random walk problem by hand on a piece of hotel stationary. This was before the days of the internet and Google. He “had no books,… no library, no nothing,” (Neither did his colleagues on the Ice.) At about four in the morning, he came up with a formula, plugged in the diameter he’d gotten from Domogatsky and the density from Tony Gow, and “everything fell in place.” He constructed a probability distribution for the arrival times of the laser pulses between strings that matched the experimental data well enough.
Now Francis is an excitable boy. The thrill of his insight left him so exhilarated that he didn’t sleep at all that night—or take the next logical step and realize that it was very bad news. “I’m a theorist,” he says. “I’d solved a problem!”
When he caught his first glimpse of Adam Bouchta the next morning, he ran up and told him he “understood everything. We have big bubbles in our detector!”
Adam hadn’t heard about the bubbles either. He’d taken a vacation after returning from Pole, and as a grad student focused mainly on hardware, had not been privy to the conversations between the senior scientists as the data came in. Despite his wetness behind the ears, however, Adam understood the implications right away—and swiftly turned white.
“What do I do now?” he asked.
This brought Francis to his senses. He thought for a moment and advised, “Nothing. You just give your talk, pretending you didn’t talk to me.”
He was reasonably confident in his all-night calculation, but again, it was far too new to be shouted from the rooftops. He drilled Adam on sticking to his script and what to say if uncomfortable questions arose.
Adam remembers his talk as being “pathetically short” and “pure hardware.” It had nothing to do with data. He was there to review the recent deployment and announce that it had gone very well. They had shown that they could drill to a thousand meters and that the hardware would survive refreezing. (Even the strings installed two years earlier were still sending up signals.) These simple achievements demonstrated after only one year of real effort that it would probably be easier—and definitely cheaper—to build a neutrino telescope in Antarctic ice than in the tropical ocean or even a relatively shallow Siberian lake.
The young man had already been dealing with some aggressive questioning in the halls by John Learned, who had heard through the rumor mill that something was up with AMANDA and was understandably anxious, since he would be giving a bleak status report on DUMAND later in the same session. During the question period at the end of Adam’s talk, John asked a few more nasty questions, but Adam kept his poise.
“We have some fresh data, and we’re looking at it,” he said. “Here we’re presenting the hardware and the status, and everything is working very well.”
John congratulated him afterward for doing a fine job.
* * *
They were off and running, but they were running scared. And in response to these manifold pressures, not unlike a star collapsing under force of gravity, they produced a supernova of fine science. They now had a small amount of data. And data, as long as it’s good data, is always a good thing, especially in capable hands.
The Swedes took the lead on the “flasher data.” They developed both a computer model and a physical model, based on calculations from first principles, for the propagation of laser flashes in the ice. (In physics parlance, this type of computer model is known as a “Monte Carlo,” because it relies on a random number generator to accumulate statistics about how an instrument or a physical system will behave. Such a method was quite appropriate in this case, since the photons were doing random walks off the bubbles.) The model
s not only agreed with each other, they fit the data precisely. This put the bubble hypothesis beyond dispute and gave solid numbers for the optical parameters that the scientists on the Ice and Francis in Venice had roughed out by hand. The scattering length, the average distance a photon was traveling before bouncing off a bubble, turned out to be about ten centimeters at the top of the array, eight hundred meters down, and twenty-five centimeters at the bottom of the array, a thousand meters down: there were fewer bubbles in the deeper ice. The trend indicated that the bubbles might disappear altogether at about 1,150 meters, and their diameter also seemed to be shrinking with depth.
The models also gave an estimate for the absorption length of the laser light, and this too confirmed earlier hunches. It turned out to be about sixty meters—it was growing as their measurements got better. The ice at the South Pole appeared to be at least as pure as the ultra-pure water that the Kamiokande and IMB collaborations had used in their detectors.
This fine work demonstrates just how powerful the methods of physics can be, and it was also deemed interesting enough to find publication in Science. All in all, the paper concluded, “the results of this study suggest that the ice cap is indeed an ideal medium for a neutrino telescope.”
* * *
Meanwhile, Buford Price approached the bubble question from a more fundamental perspective, based on his background in the physics of solids. In the process, he not only helped the collaboration decide how deep they needed to place their next array, he also made an impact on the fields of glaciology and climatology.