For the first time, astronomers have detected visible starlight reflecting off a planet orbiting a distant star. The telescope used in the discovery was too small to tell scientists much new about the previously discovered planet. But astronomers say the new technique used promises to reveal much more when combined with better spectrographs and bigger telescopes now in the works. “The ultimate goal is to characterize a planet like Earth,” says team leader Jorge Martins of the Institute of Astrophysics and Space Sciences in Porto, Portugal.
It is now almost impossible to see exoplanets directly, as they get lost in the glare of their parent stars. Astronomers have gleaned some information about exoplanet atmospheres by observing how the atmosphere absorbs starlight when an exoplanet’s orbit carries it between the star and Earth.
The Porto team used a different method to study a familiar exoplanet, 51 Pegasi b, which 20 years ago was the first exoplanet found orbiting a normal star. Researchers employed an instrument called the High Accuracy Radial velocity Planet Searcher (HARPS) attached to a 3.6-meter telescope operated by the European Southern Observatory (ESO) at La Silla in Chile. HARPS is one of astronomy’s most successful planet hunters; it works by scrutinizing stars for wobbles caused by the tug of an orbiting exoplanet’s gravity. That is how 51 Pegasi b—known as a “hot Jupiter” because of its size and closeness to its star—and hundreds of other exoplanets have been found.
To spot the wobble, researchers using instruments like HARPS build up a reference spectrum for the star—how brightly it shines at each wavelength—as if it is at rest. Using that reference, they then look to see if the observed spectrum shifts over time. If the observed spectrum shifts a bit toward the red end, it means the star is moving away—stretching out its light to longer wavelengths. A shift to the blue end means the star is moving closer—bunching up its light to shorter wavelengths. Repeated observations reveal the size and period of the wobble, and from that astronomers can infer some characteristics of the planet that is causing it—without actually “seeing” the planet itself.
The Porto team took their reference spectrum for the star 51 Pegasi and compared it with new, very sensitive observations. In addition to the first set of shifts caused by the star’s wobble, they found a second set of shifts, much fainter and with higher redshifts and blueshifts. These are from starlight that, instead of traveling straight from the star to Earth, first reflects off the planet 51 Pegasi b. It has larger shifts because the exoplanet in its orbit is traveling much faster than its parent star is wobbling.
The team’s observations—published online today in Astronomy & Astrophysics—pin down 51 Pegasi b’s mass (half that of Jupiter’s) and the inclination of its orbit (9° with respect to Earth) more accurately than ever before. They also enabled the researchers to estimate the planet’s reflectivity, or albedo. A more detailed knowledge of the albedo would provide information about possible cloud cover on the planet, or the nature of its surface.
Such discoveries will have to wait until the team gets time on ESO’s Very Large Telescope (VLT)—a battery of four 8.2-meter telescopes, also in Chile. Further improvements will follow when a new spectrograph called ESPRESSO is installed on the VLT in 2016 and when ESO’s 39-meter European Extremely Large Telescope (E-ELT) is completed in 2024. “With E-ELT we hope to be able to characterize an Earth-like exoplanet,” Martins says.
“I am very happy they’ve managed to provide independent evidence of this planet and nail down its mass,” says Didier Queloz of the University of Cambridge in the United Kingdom, co-leader of the team that originally discovered the planet. 51 Pegasi is, he says, “an emblematic star.”
Sara Seager of the Massachusetts Institute of Technology in Cambridge agrees: “Fantastic to hear a version of the literally original technique for exoplanet atmosphere spectroscopic detection has finally succeeded, but with a [relatively weak] detection, more work is warranted.”