New Planet, Magnified

For the first time, a planet has been discovered using "gravitational microlensing."

Raw image data used to discover new planet.
Credit:Bond, et al.

The gravity of a star can act as a lens, focusing and intensifying the light of a star behind it. The combined light from the two stars causes that point in the night sky to suddenly appear much brighter.

Stars are constantly on the move, so astronomers vigilantly monitor the night skies, looking for the chance event of one star passing in front of another. These events are brief – lasting only a few days or weeks – so astronomers must be ready to record and analyze the resulting light curve. In this case, extra spikes in the light curve indicated a planet is orbiting the foreground "lens" star.

Ian Bond of the Institute for Astronomy in Edinburgh, Scotland is the lead author of the paper reporting the discovery. The paper, due to be published in the May 10 edition of Astrophysical Journal Letters, describes how the discovery was made through the cooperation of the Microlensing Observations in Astrophysics (MOA) in New Zealand and the Optical Gravitational Lensing Experiment (OGLE) in Chile.

The newly discovered planet is about one-and-a-half times more massive than Jupiter. It orbits a red dwarf star at 3 AU, or three times the Earth-Sun distance. The star-planet system is 17,000 light years away, in the constellation Sagittarius. The magnified background star is 24,000 light years away, near the center of the Milky Way galaxy.

This is only the second time a red dwarf star has been found to harbor planets. The other red dwarf-planet system, Gliese 876, was discovered using the radial velocity technique. It takes years for the radial velocity technique to confirm a planet orbiting at 3 AU, but the microlensing event occurred over 30 to 40 days. The spikes in the light curve that hinted at the existence of the planet appeared for just a few hours.

Albert Einstein predicted 70 years ago that a star’s gravity could act as a lens to focus distant light, but he never thought we would see this effect. The first microlensing events were observed about 10 years ago. In 1991, Bohdan Paczynski of Princeton University, an OGLE team member, proposed using gravitational microlensing to detect planets. Since then, there have been a few reports of finding planets with this technique, but more observations were needed for these claims to be definitively proven.

Paczynski says the microlensing technique will likely yield more planet discoveries within the year, and he predicts that over the next few years they may even find Earth-sized planets around distant stars.

"In principle, if you have observations of enough stars, you can detect Earth-like planets," says Paczynski. "It’s not so difficult to see it when it happens, the difficulty comes from monitoring enough stars."

OGLE observes 200 million stars nearly every night, and Paczynski says they find one microlensing event for every 100 million stars monitored, or about 500 microlensing events each year. MOA monitors far fewer stars, and finds about 70 microlensing events each year. Most of these events are "very mundane," he says, just one star lensing a second star, with no indication of orbiting planets.

Even with over 500 events witnessed each year, microlensing events are relatively rare – stars must be perfectly aligned from our viewpoint on Earth for the lensing event to occur. After 10 years of observing, several thousand events have been documented.

"To witness more events, it might be necessary to observe them continuously and uninterrupted from a space station," says Bond. "One problem (with finding planets) is that it’s hard to know which events should be followed up."

Microlensing discoveries are listed on the OGLE and MOA web sites, so interested scientists can investigate them in greater detail. That’s just what happened with this event: OGLE posted the microlensing event, and a few weeks later MOA followed up on the observation and noticed something unusual about the light curve.

"When it was detected in MOA, we looked at the light curve and it just didn’t look quite right for a single lens," says Bond. "At the time it was detected in the MOA survey, it was actually deviating from a single light curve, a single lens. We then got additional observations to try to map out this feature."

By combining the high sensitivity of space telescopes with the sharply detailed pictures from an interferometer, TPF will be able to reduce the glare of parent stars to see planetary systems.
Credit: NASA

Bond says that a microlensing star could have a planet that is missed by observers. Having the planet appear in the light curve depends on the geometrical alignment of the star-planet system. Scientists could also miss evidence of the planet because it shows up only briefly during the lensing event.

Paczynski says multiple planetary systems can in principle be detected with this method, but since finding even one planet is such a difficult task, he doesn’t think finding multiple planets around one star is likely. While this particular star system could have more than one planet, other planet-finding methods would more easily detect them.

Philippe Crane, Origins theme scientist from NASA Headquarters in Washington, says microlensing events could be used to target stars for more intensive searches by other programs.

"One of the critical things we need to know is what is the frequency of planetary systems and what is their pedigree, what shapes and sizes do they come in," says Crane. "This looks like the kind of system where we might expect to find a full planetary system, as opposed to the planets we found before – the hot Jupiters that closely orbit their stars. So for us this is an indication of the frequency of planetary systems where we might be able to find Earths."

Crane says that while the upcoming Terrestrial Planet Finder will hunt for Earth-like planets, it won’t be able to look out beyond 100 light years from Earth. That puts this new star-planet system far out of range for that particular mission.

The science team says that amateur astronomers could find planets by looking for microlensing effects. While detecting Earth-like planets would be beyond their capabilities, amateur astronomers might be able to detect the more massive Jupiter planets because the signal is comparatively huge.

Bond says to observe microlensing events, amateurs would need a telescope size of at least a half-meter. More importantly, an observer would need a scientific-grade CCD camera (the cheaper CCD cameras get too much noise). Amateurs can check the OGLE and MOA web sites for ongoing events, or sign up for their mailing lists.

"It is going to require dedication on their part because of the time-critical nature," says Bond. "Amateur astronomers are going to have to be prepared to drop whatever they are doing in their normal life and then go out and observe these events."

Andrzej Udalski of Poland’s Warsaw University Observatory and Bohdan Paczynski in the U.S lead the Optical Gravitational Lensing Experiment (OGLE). It operates at Las Campanas Observatory in Chile, run by the Carnegie Institution of Washington, and includes the world’s largest microlensing survey on the 1.3 meter (51-inch) Warsaw Telescope. NASA and the National Science Foundation funds OGLE in the U.S., and the Polish State Committee for Scientific Research and Foundation for Polish Science funds it in Poland. Microlensing Observations in Astrophysics (MOA) is primarily a New Zealand/Japanese group, with collaborators in the United Kingdom and U.S., New Zealand’s Marsden Fund, NASA and National Science Foundation, Japan’s Ministry of Education, Culture, Sports, Science, and Technology, and the Japan Society for the Promotion of Science