When the 10-meter Keck I telescope first became fully operational in 1992, the only planets known to exist were those in our own solar system. Models of planet formation dutifully reproduced the nine known examples, and most astronomers thought that when other planets were finally seen, those solar systems would look like our own, with giant Jovian planets in the outer regions and small rocky planets in the inner parts.
Less than four years later, before even the second Keck II telescope saw starlight, the first planet had been discovered orbiting the sunlike star 51 Pegasus (1). Since then, more than 400 planets have been detected outside our solar system and our picture of the frequency, nature, and formation of planetary systems has been upended and not yet fully recovered.
Almost all of those planets have been found indirectly, most through measurements of the Doppler shift of the parent star or through detecting a transit as the planet passes in front of the star. These techniques are powerful, but also have fundamental limits—they are biased toward planets in small short-period orbits. For Doppler detections, the only information provided is the planets orbital parameters and a lower limit on its mass.
Directly imaging a planet—separating its light from the parent star—opens up new possibilities to detect planets in wide orbits and characterize them spectroscopically. This potential was finally realized with the announcement in November 2008 of images of a single planet orbiting the star Fomalhaut and a family of three planets orbiting the young A-type star HR8799. HR8799 is a massive star accompanied by three planets, 5-10 times the mass of Jupiter, in apparently circular, coplanar orbits—a scaled-up version of our solar system. This latter detection, using adaptive optics on the Keck and Gemini telescopes, sets the stage for the even more powerful capabilities of TMT.
TMT’s first-light adaptive optics system (NFIRAOS) and integral field spectrograph (IRIS) will be able to provide high-quality spectra of planets like these. Such spectra will allow measurements of the planet’s temperature, gravity, composition, and cloud structure—the HR8799 planets show evidence of hot clouds of dust in their upper atmosphere. In turn, this will illuminate their formation. Did these planets form (like Jupiter, or the Earth) through a gradual accretion in a protoplanetary disk? Or did they form rapidly as instabilities broke the disk up into fragments?
The next step in this field will be new, powerful adaptive optics systems designed specifically for the study of extrasolar planets. The Gemini Planet Imager (GPI)—http://gpi.berkeley.edu—is one example. With a 2000-actuator AO system and a sophisticated coronagraph, GPI will be able to see planets the size of Jupiter orbiting young stars in our solar neighborhood.
The TMT equivalent is the Planet Formation Imager (PFI). PFI would combine advanced wavefront sensors, microchip deformable mirrors with more than ten thousand actuators, starlight-canceling interferometers and a sensitive imaging spectrograph. This enormously powerful instrument will be able to study anything from a mature solar system—seeing the light reflected by Jupiter or even Neptune-sized planets – to the planets that are forming even now near young stars in the Taurus and Ophiucus regions. PFI may even be able to detect a handful of “Super-earths”—watery planets only twice the size of our own—around the very nearest stars.
1. In 1992, astronomers announced the discovery of planets orbiting a pulsar—the dead remnant of a once-massive star. Their exact origin and nature are still a mystery, and no other pulsar planets have been found.
By Bruce Macintosh, Lawrence Livermore National Laboratory