Keck Observatory to team up with Kepler on identifying other Earths

Subaru and Keck observatories at sunset, getting ready to start work.

I wrote a while back that I suspected once Kepler was up other Earth-bound observatories would suddenly be interested in donating a great deal of their time to finding out more about these newly discovered planets, simply for the fact that finding out more about another possible Earth is way more interesting from a human point of view than simply observing yet another galaxy. It seems that one of these observatories is the Keck Observatory:

For nearly a decade, Cal-Berkeley astronomer Geoff Marcy and his colleagues have been using the W. M. Keck telescopes to discover giant planets orbiting distant stars. Now, with the successful launch of NASA's Kepler mission, they will be using Keck I's ten-meter astronomical eye to discover distant Earths. Kepler will pick out Earth-like candidates. Keck will then zero in on them and determine, with certainty, if they are at all similar to our home planet.

and:

In the Kepler-Keck duo, once Kepler team members find an Earth candidate and determine as best they can that they're not looking at two stars orbiting each other, they will hand the object off to Marcy and his colleagues. The team will use Keck I telescope and its instrument HIRES, the High Resolution Spectrometer, to monitor how the light coming from the parent star changes as the planet candidate orbits.

HIRES is an instrument that spreads light collected from the telescope mirrors into its component wavelengths or colors. This is called a spectrum. When the planet candidate orbits around the back of the star, its gravity will ever so slightly pull on the star causing the star's spectrum to shift toward redder wavelengths. When the planet comes around in its orbit to cross the face of the star, it will pull the star in the other direction, and the star's spectrum will shift toward bluer wavelengths. HIRES will detect these shifts and give astronomers the star's radial velocity, or the speed at which the star moves toward or away from Earth. Based on this speed, Marcy and his team will be able to calculate the mass of planet candidate.

"Keck's HIRES is the only game in town that can measure spectral shifts caused by an Earth-sized planet. No other telescope is big enough," Marcy said. "That is why NASA is really heavily dependent on the Keck telescopes right now."

Since Kepler is observing 100,000 stars at the same time it will be able to point out the existence of these planets with much greater speed than any Earth-bound observatory could. Since time is precious at these huge observatories it's not really possible to spend a great deal of time looking at star after star to see whether there are Earth-like planets in orbit, but once Kepler has made the confirmation the solar system then becomes an extremely valuable target; when that happens we then know that there are planets similar to ours orbiting the system's star and now it's just a matter of refining what we know.

Also note this last part here:

Marcy and his colleagues plan to start studying Kepler's candidate Earths with Keck I and HIRES during the last three night of July 2009.

After Kepler was launched it was noted that it wouldn't be until 2012 that Earth-like planets would be confirmed because they aren't going to announce these discoveries until the planet has been seen three times to make sure that they haven't observed something else (sunspots, etc.). However, even after the first pass there will certainly be a strong suspicion that they've identified a planet. Also, note that there will also be Earth-like planets around red dwarf stars, which have habitability zones much further in, usually with orbits of only a few weeks instead of a year.

Finally, the three year number (2012) isn't exactly correct, because that would assume one full year before the first passage of a planet in front of its star. It's more likely though that any planets with approximately 1-year orbits will already be on their way towards the point where they come in between us and their star, meaning that the first observation will come within a few months (and if we're lucky within a few days), and then two years after that for observations two and three. So let's say Kepler first observes one of these planets in May 2009, then observation two will be May 2010, and three will come in May 2011. So don't despair and think that we're going to have to wait until 2012 before we start finding these planets because it'll take much less time than that.