Paczynski 'OGLEs' the sky for answers to cosmic questions

Steven Schultz

Princeton NJ -- To most of us, the night sky is a steady backdrop. The only change we know is the seasonal rotation of the constellations. For Princeton astrophysicist Bohdan Paczynski, however, the sky teems with fast-changing events.

Astrophysics professor Bohdan Paczynski, pictured here near the observatory on the roof of Peyton Hall, is part of a group that has pioneered a method of surveying large patches of sky.


Paczynski, the Lyman Spitzer Professor of Theoretical Astrophysics, has found that subtle, but often highly regular, variations in the light of stars can provide insights into longstanding problems in astronomy, from finding planets around other stars to estimating the age of the universe.

For the past 10 years, Paczynski and a small group of collaborators here and in his native Poland have pioneered a method of surveying large patches of the sky many times a night in search of anything that changes.

The idea for such a survey was "completely novel," said astrophysics chair Scott Tremaine. "It is one of the big growth areas of astronomy for the next decade, and has risen -- I think it's not an exaggeration to say -- entirely out of Paczynski's idea.

"One of the really impressive things about him," said Tremaine, "is that if you look at dramatic developments in astronomy over the last decade, he has played an extremely important role in many of them."

As a technical feat alone, the project is remarkable. Last July and August, the group tracked 5 million stars over a six-week period, making 800 observations of each star. During the peak viewing periods, the astronomers recorded data on all 5 million stars every 12 minutes.

Their ability to perform such a rapid search grows out of years of refinements the researchers have made to their equipment, which includes their own telescope at the Las Campanas Observatory in Chile. A large digital camera records detailed information over a wide area and feeds it to software that automatically sorts out interesting events worthy of further study.

Earlier this year, the group, including Princeton graduate student Przemyslaw Wozniak and the lead member of the team in Poland, Andrzej Udalski of Warsaw University, published a free Internet database of 250,000 variable stars from a survey conducted from 1997 to 1999. The project, known by the acronym OGLE, has become an important source of data to scientists working in many aspects of astronomy.

"There is so much data that there is no way our small team can do all the science with it," said Paczynski. "Now other people, instead of asking for expensive telescope time, can go the Internet and get the data they need."

One of the most recent uses of the OGLE data has been in looking for planets around other stars. Astronomers first discovered such planets in 1995, but the standard techniques for finding them are slow and require time on big telescopes. The OGLE survey, however, can find them quickly by looking for stars that periodically dim as planets pass before them, even though the planets themselves are far too small to see.

In February, the group reported that it had searched for planets among 50,000 relatively bright and nearby stars out of the 5 million they monitored. These showed almost no light variability, except for 46 of them, which exhibited very small periodic dimmings, indicating that small dark objects are orbiting them and periodically passing in front. Some of those objects might be very faint stars known as brown or red dwarfs, but others are almost certainly planets, said Paczynski.

Either way, he will be happy. "You could say that the project OGLE is focused not on a specific scientific goal, but on a specific technology, and uses this technology for whatever science can be done with it," Paczynski said. "Whenever anything new pops up, we can switch on a dime."

That was true even at the beginning of the project in 1992 when the main scientific justification for doing the survey was to search for the invisible and mysterious material called "dark matter," which is believed to make up the majority of the matter in the universe.

Paczynski realized that scanning for flickers among stars might reveal an effect called gravitational lensing in which the gravity of a massive body bends light that passes by. If a great mass of dark matter passed in front of a star, so the thinking went, it would not block the light. It would have the opposite effect, momentarily focusing and brightening the starlight like a lens (hence the OGLE acronym, which stands for Optical Gravitational Lensing Experiment).

OGLE and several other microlensing surveys showed almost no examples of gravitational lensing by dark matter, which was a major factor in convincing most astronomers that dark matter does not cluster. In the meantime, however, the surveys have found more than 500 examples of lensing by stars and other massive objects, which have turned out to be interesting as well.

Astronomers have found that they can use lensing and other flickering events to calculate very accurately the distances and sizes of faraway objects, which can then be used as a yardstick for calibrating other cosmic features such as the shape and distribution of mass in our galaxy, said Tremaine.

Udalski, who founded OGLE with Paczynski, has even used the data to propose a recalibration of the basic measure by which astronomers gauge the size of the universe. Although the idea is still controversial, he has argued that the universe is really 10 to 15 percent smaller and thus a billion or two years younger than previously believed.

Paczynski noted, however, that he and his colleagues try not to become too absorbed in following up specific applications of the data. Rather, he said, they focus on developing the observational techniques and optimizing them to produce the best data for the most important questions of the day.

Paczynski's drive for efficiency is evident in the dramatic growth of OGLE over the last decade. In its first few years, it discovered several hundred variable stars -- something that would take just days or weeks with the current technology. Paczynski likes to put it another way: "How many measurements can you do for a dollar?" Before the last equipment upgrade, the group made 100,000 measurements per dollar of funding; now they produce 1 million per dollar.

"That is how we can afford to give data away," he said. "We have just so much data."

The other reason is that the real currency of Paczynski's work is ideas. Apart from the success of OGLE, Paczynski is respected among astronomers for his theoretical insights and knack for forecasting the most important new directions for the field.

"He is generating ideas for people around the world," said Tremaine, noting that one of Paczynski's valuable roles at Princeton has been to build a strong relationship with colleagues in Poland, bringing gifted students here and helping build strong research programs there.

Indeed, Paczynski defers much of the credit for OGLE to his collaborators, particularly Udalski, whom he describes as "the real doer" who sees the project through its many technical and scientific hurdles.

Paczynski came to Princeton in 1982 after a 20-year career at what is now called the Copernicus Astronomical Center in Warsaw. Starting during his graduate research at Warsaw University, Paczynski has kept up a continuous circuit of visits to astronomical centers around the world, planting ideas and building collaborations.

At the same time, his theoretical research has had a substantial impact on the field. For many years, Tremaine said, Paczynski was a "lone voice in the wilderness" in proposing the correct distance for an odd cosmic phenomenon called gamma ray bursts, which are something like giant nuclear explosions. Most astronomers thought these events came from within our own galaxy. Paczynski argued that they emanated from the very edge of the visible universe, an idea that was ultimately vindicated by recent observations.

"Everyone had access to the same data," said Tremaine. "It's just that Paczynski has a real genius for looking at fragmentary and sometimes confusing data and having insight into what it really is telling you."

For Paczynski, it often is possible to guess what insights a new round of experiments will bring, but the most exciting times are when new data triggers ideas that even he did not anticipate. With any luck, he said, the OGLE group is likely to discover a number of planets much like our own Jupiter in the next few years. "And if we are unbelievably lucky, we could find lower-mass planets like our Earth," he said.

"However, if the past is used as a guide, something even more attractive will pop up," he said, "Something we can't even imagine."

[an error occurred while processing this directive]