Most of the extrasolar planets discovered up to now are giant gaseous planets, as big as, or even bigger than our Jupiter. Don’t imagine these planets look like Jupiter. Most of them are much hotter, with atmosphere temperatures ranging from 500 to 2000 K, because they are very close to their star. I guess these so-called ‘hot-Jupiters’ would probably look quite different from our solar system giant!

But why are known extrasolar planets so big? It is mainly an observational bias. The radial velocity method detects Doppler shift of the stellar wobble due to the planet. The more massive the planet, the larger the wobble of the star ( see Lisa Kaltenegger’s contribution ). With a given instrument, it is thus obvious that the most massive planets are discovered first in the current programs scanning hundreds of stars with appropriate metallicity (Sun-like or close to).
 

Figure 1: Minimum mass of the known extrasolar planets orbiting solar-type stars, in Earth masses as a function of the year of their discovery. The masses of Jupiter, Saturn, Neptune, and the Earth are marked for comparison. A few benchmark cases are labelled, as well as the masses of the two planets orbiting the pulsar PSR 1257+12 discovered by Wolszczan & Frail in 1992 (from Mayor & Queloz 2006). (Updated the 9th of February).

I like very much Figure 1 above. It shows the mass of planets (vertical axis) versus the year of their discovery (Mayor & Queloz 2006). The planets with the lowest masses are plotted in red and labelled with their name. It is amazing to see how the mass of the smallest discovered planet decreased during the last decade. Of course this is due to the constant improvement of the radial velocity method: new spectrographs are built that are always more sensitive to that of the previous generation. Also data collected over several years help to improve detectivity.

I cannot resist to the temptation (and you?) to extrapolate with my finger the apparent straight line passing through the red points, and suggesting that we should detect an Earth-mass planet around 2011 (in fact probably a ‘hot-Earth’ close to its star)! Of course the relevance of such an extrapolation depends on how instrumentation will yet progress in the coming years. New spectrographs dedicated to planet search are being built and are expected to provide unprecedented sensitivity. HARPS, a spectrograph installed in Chile at ESO/La Silla observatory showed superb sensitivity as soon as it got first light in early 2003, and discovered a 14 Earth-mass planet around the star mu Ara in 2004, just one year after it started routine operation. And last year, an even smaller planet was discovered by a team using HIRES on one of the 10-m Keck telescopes: a 7.5 Earth-mass planet orbiting dwarf star Gliese 876 (of course, the smaller the star, the smaller the planet that can be detected by a given instrument).

But instrumentation cannot do everything. There are fundamental limitations to the detection by radial velocity. The limitation comes from stellar activity, i.e. stellar pulsations, surface activity (bright or dark spots), which generates ‘noise’ in the data and hides the small wobble due to a low-mass planet. This ‘noise’ put the detection of an Earth-mass planet near a Sun-like star at the very limit of the radial velocity capability.

The transit method (see again Lisa Kaltenegger’s contribution) will also allow detecting small planets from space, by detecting the weak drop of stellar luminosity due to the planet transit in front of its star. It has been shown that the CNES/Corot mission, to be launched this year, should allow to see planets with radius as small as 2 Earth-radius, thus a mass of 8 Earth-mass, assuming a rocky body. The NASA/Kepler mission, equipped with a larger telescope, is expected to detect Earth-like planets, i.e. one-Earth-radius planets. Kepler is even designed to detect one Earth-radius planet in the habitable zone around Sun-Like stars, in that region far enough from the star to have liquid water rather than vapour, and also close enough to not have a frozen planet.

While Kepler and Corot will not be able to provide a direct image of these Earth-like planets, we are waiting for two more ambitious missions, ESA/Darwin and NASA/Terrestrial Planet Finder (TPF). These missions should deliver us in 2015-2020 the first image of a terrestrial planet. The planet will appear as a single dot near the mother star (the planet will not be resolved as a disk - this will come much later). Spectroscopy of the light coming from this single-dot-planet hopefully will allow the analysis of the planet’s surface chemistry, and ultimately will allow to look for the signature of life, like combination of gas like 0₂, ozone 0₃, methane CH₄ and others. Be sure that this single dot will be a milestone in astronomy history, as the discovery of the first extrasolar planet around Sun-like star 51 Peg was ten years ago, in 1995!

So we are en route for the discovery of an Earth-mass planet, and even maybe an inhabited Earth-like one. That’s the dream behind the search – at least for me.

Reference

Mayor, M. & Queloz, D. 2006, “The discovery of 51 Pegasi at Haute-Provence Observatory -- The quest for precise radial velocity” in the proceedings of the “Tenth anniversary of 51 Peg-b” colloquium, 22-25 August 2005, Observatoire de Haute-Provence, France, eds. L. Arnold, F. Bouchy & C. Moutou, Platypus Presse (in press)

By Send a private message to the author
Follow this author