Past Observing Campaigns:
Arecibo, Puerto Rico

Diary March 1999

Mar. 16, 1999

"Project Phoenix is Back at Arecibo!"

The Project Phoenix team, about a dozen scientists and engineers, has returned to the corrugated, chlorophyl-clad countryside of Puerto Rico. Their scrutiny of our cosmic back yard using the imposing Arecibo Radio Telescope is once more underway. In today's hi-tech search for intelligent extraterrestrials, this is where the rubber meets the road.

Project Phoenix was last at Arecibo six months ago, and if you haven't already done so, you may wish to check out the dispatches from our September run.

John and Jane As regular mouse-clickers to this site are undoubtedly aware, Phoenix differs from other professional SETI experiments in that it is a targeted search. Clearly, we don't know where E.T. might be hanging out until we've detected his signals, but some locales are more promising than others. The neighborhoods of stars similar to the Sun are, we think, more likely to shelter worlds like the Earth. So as the Ides of March arrived in the tropics, so did the Californians from the SETI Institute, eager to aim Arecibo's 18 acres of aluminum mesh in the direction of Sol's nearby cousins, and perhaps find the signal that would permanently change the homo sapiens mindset.

For the next several years, Phoenix will be using the Arecibo instrument in September and March, during the equinoxes, spending 12 hours a night in its hunt for extraterrestrials. Although radio telescopes can generally operate throughout the day, the Phoenix team has opted to observe without the Sun's noisy celestial presence. We avoid observing within 60 degrees of the Sun because the solar wind would destroy the coherence of any signals.

Although the observing equipment is complex and occasionally temperamental, the scheme used for the search is simple. The telescope is pointed in the direction of one of the thousand, nearby sun-like stars on the Phoenix hit list. Then the receiver, with its 28 million channels (each roughly 1 Hz wide) begins observations at the low end of the band, typically 1,200 MHz. If nothing interesting is found after about four minutes, then the receiver is reset 20 MHz higher in frequency, and another 28 million channels (between 1,220 and 1,240 MHz, say) are examined. After four minutes, the receiver is again stepped up the dial. Currently, only L-band frequencies, between 1,200 and 1,750 MHz, are being observed in Puerto Rico, and it takes several hours to cover these for each star. Sometime soon, when the confirming telescope in Jodrell Bank, England is outfitted with an S-band receiver, these stars will be looked at again, but at frequencies between 1,750 and 3,000 MHz.

The first day or two of observing time is generally consumed by tests and efforts to determine the so-called "RFI environment." The latter refers to the cacophony of radar and other interference that seems to find Puerto Rico especially attractive. The Phoenix team has so far spent many hours looking at blank sky, and the signals found during these scans are added to an interference database. Soon, the real observing will be underway, and the database will be of critical importance in separating radio chaff from possible wheat. Until then, the work here is merely interesting. Once observing is underway, interest turns to excitement.

Mar. 19

When you're reporting on news from the front, it's not always good news. For the past 24 hours, the Phoenix team has been wrestling with a gnarly problem. As the technically inclined know, radio astronomers typically observe with two receivers simultaneously, each sensitive to a different polarization of the incoming waves. By doing so, they capture all of the energy collected by their large metal dishes. Needless to say, Project Phoenix also observes with two (circularly) polarized receivers, both at the main site, in Arecibo, and at the confirmation site - the Lovell Telescope, at Jodrell Bank, England. Information about interesting signals in each polarization is sent from Puerto Rico to England, and the two polarizations are checked out independently.

This yin and yang separation of incoming signals into two polarizations is necessary if you're hunting for E.T., as it is in the nature of deliberately built transmitting stations to make polarized signals. Think about it: the rooftop antennas used for TV reception generally have their elements in a horizontal plane, since the transmitter on the hill outside of town is producing horizontally polarized signals. This is different from naturally produced radio noise, such as from interstellar gas, which is usually only slightly polarized.

The problem confronting the Phoenix team was that they were not sure that the receivers at the two telescopes line up: that the left circularly polarized system in Arecibo was "talking" to its similarly left circularized buddy at Jodrell Bank. This may sound like a trivial matter to resolve - simply look at the plumbing at each telescope. But the reality is otherwise, and somewhat akin to sexing chickens: simple in principle but difficult in practice.

Kiriaki Xiouri It would be easy to sort out the polarizations if we could find a calibration source in the sky. Phoenix regularly tunes in the Pioneer 10 spacecraft for just this kind of test. Pioneer 10, launched in the 1970s, is presently 6 billion miles from Earth (one-and-a-half times the distance of Pluto), and moving outward. Its transmitter is barely squawking, whispering with a watt of power. But it's still easily detectable at both Arecibo and Jodrell Bank. Alas, its transmitter is tuned to S-band, nearly 3 GHz, whereas the Phoenix observations are currently being conducted using L-band receivers, between 1.2 and 1.7 GHz. The bottom line? Pioneer 10 can't resolve the L-band polarization puzzle.

On the night of 18 March, we tried pointing the telescopes at masers, dense clouds of interstellar molecules that emit polarized radio signals, in an attempt to determine left from right. It didn't work: the signals were too weak to be easily discerned with the Lovell Telescope. As I went to bed, a few hours after midnight, the situation was discouraging. Our yin and yang were still ambiguous, and there seemed little hope for a quick resolution.

By the time I stumbled in for lunch, the problem had been resolved. Kiriaki Xilouri, an astronomer at Arecibo, and Andrew Lyne, at Jodrell Bank, had made some observations that cleared everything up. They had taken a bead on a few pulsars - collapsed stellar corpses that flash polarized light and radio into space. And they could assure us that the Arecibo and Jodrell Bank receivers were lined up.

The bad news was that the observations of the last 2-1/2 days were compromised, and would have to be tossed. The good news was that the problem was solved, and Phoenix could now scrutinize our cosmic neighbors with confidence.

Mar. 24

Kevin Dalley was on the phone from Jodrell Bank, trying to be matter-of-fact. "Well, there's good news, and there's bad news..." I listened quietly, thinking to myself that "good news, bad news" has become the leitmotif of this observing run. "The good news," Kevin intoned, "is that tomorrow morning, I'm taking off for a one week vacation in Scotland." I offered a mildly enthusiastic response, waiting for the other shoe to drop. Kevin dropped it: "The bad news is that the telescope is broken."

Jodrell Bank Early in the evening of March 24, the 250 foot Lovell Telescope had begun making ugly sounds, like bowling balls in a clothes dryer. Within a few hours, the elevation thrust bearing had ground itself into small bits, and the telescope was shut down. Stopped. Out of commission. This failure has occurred before, so the astronomers at Jodrell had a pretty good notion about how long the repair would take. The estimate was five days, minimum.

Jodrell Bank's role for Project Phoenix is important. Fortunately, it's not absolutely essential. When everything is working, the candidate signals found at Arecibo are automatically sent to the rolling countryside south of Manchester, where their fate is quickly determined. Either they are confirmed as interesting signals or rejected as earthly interference. Now our confirmation telescope was deader than Yorick, alas.

However, even though Jodrell was temporarily out of the game, play continued. Arecibo switched to "two-star" mode. The way this works is quite simple, and similar to a technique long employed by radio astronomers. To find a faint signal, one looks first at the source, then at a so-called "off" position in which there is only "blank" sky. The two results are differenced, and if the result is not zero, then a signal is present. Most of the interference that gives migraines to Project Phoenix researchers doesn't come from the directions of the stars under study. However, these noxious terrestrial signals are so strong that they can "leak" into the receivers no matter where the dish is aimed. They will appear in the "off" position as well as the "on."

So the two-star mode works like this: The telescope is pointed at Star A, and a four-minute observation is made. The interesting signals are written to a file, and Arecibo is then redirected to Star B. As Star B is being searched for tantalizing emissions, the telescope is simultaneously checking out the filed signals from A. The 'scope then reverts to Star A, which serves as the "off" source for the signals just collected from B. Each star serves as the "off" position for the other. The implicit assumption here is that aliens are not broadcasting from both star systems (which might be separated by 100 light-years or more) at exactly the same frequency and same strength - a not unreasonable assumption!

Two star mode works. But it has its down side. Ping-ponging the telescope back and forth from A to B takes time. Furthermore, the sorting out of interference both takes longer and is less efficient than when the two-telescope scheme is used. In addition, the software that automates so much of the observing tedium is largely optimized for two-telescope operation, so the astronomers work harder when stuck with only one.

Still, even though the ship of discovery has hit a squall, the researchers continue to tack and turn, making slow but steady progress. Maybe it's just bad luck or Murphy's Law. More likely, it's the inevitable fate of any exploration into the unknown.

Mar. 30

"Oh, right!" Kevin Day exclaims with a simultaneous smile. He, Alan Patrick, Peter Backus and I are shooting the breeze in the mellow part of the evening. It's a rare break from the high-pressure environment of the past few days, when getting the telescope to work was Job Number One, Two and Three.

Monitoring The control room for Project Phoenix is not much more than an alcove off the main Arecibo command center. There's space for three long tables covered with computers, a smaller table for miscellaneous junk, and enough floor for five people to wheel around in their blue-and-black swivel chairs. For the past several days, Peter and I have been observing during the early evening, from 6:00 pm to midnight or 1:00. After that Jill Tarter takes over. Jill prefers the graveyard shift - she says the most interesting things happen then. But few people are around in the wee hours to verify this claim, so no one's sure whether Jill might not be razzing us.

The bull session is unusual. Most evenings are grimly frenetic, with the observer locked in a non-stop struggle to keep the system afloat as powerful radar interference or other glitches crash both hardware and code. But tonight everything's working as well as it can, and the crew is relaxed. Peter nods quietly as Alan, whose Virginia upbringing is reflected in a soft voice and rare good manners, modestly offers an opinion. It occurs to me that I have never seen these people on edge or angry. They're easy to work with.

The phone rings, and the Arecibo telescope operator, Willie Torres, picks it up. There are Arecibo operators here 24 hours a day, and normally it is they who drive this mighty instrument. But Project Phoenix is like a paranoid pharoh, controlling everything. For twelve hours each night, Phoenix software takes charge of the telescope, leaving the operators with little to do. So they usually put on some music and surf the Web. But now Willie is barking emphatic Spanish at the caller. Mildly intrigued, I wait until he's done, then ask what the call was about. "The meteors," he says. A flurry of meteors has lit up the Caribbean skies, and some of the locals are concerned that the strange lights might be invading saucers. The caller has asked if the Arecibo Telescope tracks the lights. Willie tells them we can't and we don't, and that the invaders are pebble-sized rocks.

Donna Kubik, a blonde wisp of a woman from Michigan and an Arecibo engineer, wanders in. Donna likes to work nights, and she's using the kitchenette abutting the Phoenix alcove to make some coffee. Everyone asks for coffee around here, but I've never seen anyone actually drink it. Kevin Day tried, but after one sip, he declared it strong enough to pickle steel, and set it aside for the toxic waste pickup. We continually wonder at the unpalatable coffee, as Puerto Rican coffee is reknown. Something wrong in the proportions, we figure. The night wears on, and our voices shuffle against the endless cheep of the tracking control, an audible feedback to the Arecibo operator that the motors are working. The Phoenix system itself is relatively quiet: even the occasional "interesting signal" - one requiring an extra scan or two to confirm as interference or otherwise - only elicits a faint beep from the computers. No klaxons, and no flashing red lights. If we pick up an alien signal tonight, or any other night, the announcement will be soft and quiet, like my companions on this rare, "routine" evening.

more about Past Campaigns

SETI