A Quarter of Sun-Like Stars Host Earth-Size Worlds 105
astroengine writes "Although there appears to be a mysterious dearth of exoplanets smaller than Earth, astronomers using data from NASA's Kepler space telescope have estimated that nearly a quarter of all sun-like stars in our galaxy play host to worlds 1-3 times the size of our planet. These astonishing results were discussed by Geoff Marcy, professor of astronomy at the University of California, Berkeley, during a talk the W. M. Keck Observatory 20th Anniversary Science Meeting on Thursday. '23 percent of sun-like stars have a planet within (1-2.8 Earth radii) just within Mercury's orbit,' said Marcy. 'I'll say that again, because that number really surprised me: 23 percent of sun-like stars have a nearly-Earth-sized planet orbiting in tight orbits within 0.25 AU of the host stars.'"
Re:Great! (Score:5, Informative)
Re:But... (Score:4, Informative)
2 problems with your assertion:
1: the majority of stars are smaller and dimmer than the sun, .25 AU is not necessarily out of the "green" zone for the most common dwarf stars.
2: earth sized planets further out from stars can not be reliable detected using current technology and processes. The fact that the earth sized planets that we can detect are plentiful does indicate that the earth sized planets we can not detect are not plentiful. Recall that the first few exo-planets were much larger than Jupiter and much closer than earth. We are constantly expanding the lower limit of mass and higher limit of distance that we can detect effectively.
Re:Streetlight effect? (Score:5, Informative)
The answer is that it's not much more difficult, but a lot more time consuming (gleaned from going to talks on the subject, not my area of expertise).
There are two basic ways that these planets are observed: They make the stars they orbit wobble (the basic 2 body problem - each body orbits the center of mass of the pair) and they dim the light from the star when they pass in front (like an eclipse).
The time problem comes from the fact that orbits are longer for objects more distant from the star. If we make the simplification that the orbit of the planet is basically circular, the time period for an orbit increases as radius^(3/2). (Insert semi-major axis for radius for non-circular). The standard is about three events separated by equal times to count as an observation - you have to wait to see an event at least twice to know the time period and so infer the radius of orbit, and once again to remove some flukes. Hence you're having to wait a long time looking at a star to see this happen.
Now, on top of that you've got the possibility that there's more than one planet, that the earth-like planet isn't the dominant mass, etc etc. This can all be cleverly dealt with (multiple wobbles, multiple eclipses) but it adds time to the confirmation process.
To give an example: Suppose you were somewhere near Proxima Centauri, and making the relevant observation looking for Earth. It would take at least three years to detect Earth, even if your telescope was amazing. Dynamics of the system would pick up the effect of Jupiter on the sun first, for the wobble detection (you wouldn't get much eclipse given the angle between the plane of the solar system and the position of PC) and it might take quite some analysis to pick up Earth at all given the effects of all the other planets.
Anyway, I'm sure some astro people can give a much better version of all this. Suffice to say that we aren't looking for Earth like planets at Earth like radii yet, but I imagine over the next ten to twenty years there will be a lot of poor graduate students analyzing data desperately looking for Gallifrey.
Re:But... (Score:4, Informative)
Most people don't understand that we can only find (with the current way how we do detection) very very few planets. Perhaps 1/300 or even less (more likely 1/900) of the systems can be observed in a way that reveals planets.
We can only detect a planet if his orbit plane is cutting the star like this: -o-
Ofc you can turn this now clockwise or counterclockwise, the cut does not need to be horizontal.
However we can not detect any planet in a solar system that looks like a cut up onion to us: the star in the middle and the planets orbiting on the rings around it (because the planerts are to dim to see directly, and they never obscure the star)