News & AnalysisAstronomy

Rival Detectors Prepare to Take Snapshots of Distant Worlds

See allHide authors and affiliations

Science  21 Feb 2014:
Vol. 343, Issue 6173, pp. 833
DOI: 10.1126/science.343.6173.833
Vantage point.

Europe's Very Large Telescope will host the SPHERE planet imager.


In the coming months on two mountaintops in Chile, two new state-of-the-art instruments will start scanning the skies for planets around other stars. The vast majority of the roughly 1000 exoplanets identified so far have been found using indirect methods because starlight swamps their faint optical signals. But the new instruments, one North American and one European, will see planets directly. Fixed to two of the world's biggest telescopes, they push optical technology to the limit. "After 10 years building it, to see it on the sky is fantastic," says Bruce Macintosh of the Lawrence Livermore National Laboratory in California, principal investigator for the Gemini Planet Imager (GPI).

GPI, built by a consortium of U.S. and Canadian institutions, is already mounted on the 8-meter Gemini South telescope on Chile's Cerro Pachón. "The integration went very smoothly, which was a pleasant surprise," says Macintosh, who will soon be moving to Stanford University. "It worked on the first star we pointed it at." On 7 January, the GPI team published its first image of an exoplanet: Beta Pictoris b, a young planet several times the size of Jupiter that an indirect detection strategy had previously spotted. Commissioning of GPI will continue for a few months.

Meanwhile, the SPHERE instrument, built by the Institute of Planetology and Astrophysics of Grenoble in France and a group of other European institutions, is in transit to Chile. Once it arrives at Cerro Paranal later this month, it will be mounted on one of the four 8.2-meter scopes of the European Southern Observatory's (ESO's) Very Large Telescope. Co–principal investigator Markus Feldt of the Max Planck Institute for Astronomy in Heidelberg, Germany, acknowledges that there is a friendly rivalry between the two teams. "GPI is way ahead now," he says. SPHERE is not expected to see its first light until 12 May.

Trying to see an exoplanet is often likened to trying to spot a firefly while staring into a billion-times-brighter searchlight. Astronomers use a mask called a coronagraph to block out the star's light, but it takes sophisticated optics to stop glare that can hide the planet and to correct for various optical imperfections.

Using these techniques, astronomers have directly imaged seven exoplanets, all supergiants both larger and farther from their parent star than Jupiter. About 10 years ago, researchers began working on instruments optimized for detecting exoplanets; GPI and SPHERE are the first to be completed. To remove the distorting effect of Earth's atmosphere, both teams have added another technology: extreme adaptive optics, in which deformable mirrors are reshaped in real time to correct for atmospheric distortion. GPI and SPHERE both boast more than a thousand actuators able to make adjustments a thousand times a second. GPI's system is built on a microchip 3 centimeters across sporting 4000 microelectromechanical actuators spaced 0.4 millimeters apart.

First light.

Beta Pictoris b (bright dot), the first extrasolar planet snapped by the Gemini Planet Imager.


Both instruments use multistage systems to eliminate stray light and correct the light path. GPI cuts out diffracted light by passing the beam through a small circular window fringed with a pattern of 10-micrometer dots carefully designed to prevent scattering. In SPHERE, after the coronagraph blocks a star's light, other instruments split the remaining beam into two, shift one of them by half a wavelength, and then recombine them so that any remaining starlight is removed by destructive interference. SPHERE also has multiple detectors at different wavelengths so that starlight can be removed electronically.

Despite these efforts, the new instruments will see only a few of the hordes of undiscovered exoplanets thought to lurk in space. Reflected light is just too dim, so they will focus on young planets still glowing from the heat of their formation. They won't be able to see planets much smaller than Jupiter, and they work best for planets orbiting far from their star. (The other main exoplanet detection methods—picking up a star's wobble as an orbiting planet pulls on it or the star's dimming as it passes in front—favor close-in planets.)

Later this year, each instrument will embark on a large-scale survey of about 500 nearby young stars to spot orbiting planets and analyze their light to measure their temperatures, the composition of their atmospheres, and even the structure of their cloud cover.

GPI and SPHERE will likely be the prime instruments for exoplanet imaging until the next generation of 30- to 40-meter extremely large telescopes start observing next decade. To see rocky planets the size of Earth, however, will require a telescope in space. Last decade, researchers at NASA and the European Space Agency drew up plans for ambitious planet imaging missions, but neither got off the drawing board. "It's a very costly endeavor. None of the proposed concepts were funded," says ESO instrument scientist Markus Kasper. There is, however, a more modest proposal on the table: adding a coronagraph instrument to WFIRST, a NASA astronomy mission that may fly around 2024 (Science, 15 February 2013, p. 748).

View Abstract

Stay Connected to Science

Navigate This Article