We'll Find Earth's Twin by May 2011, Experts Predict

Think of it as Moore's Law for exoplanets. Two astronomers have used the current rate of discovery for planets outside our solar system to calculate when we're going to find a habitable planet akin to Earth.
Their answer: mid-2011.

But just like 42, the true meaning of the answer depends a lot on the nature of the question.
At the crux of the issue is, what's the definition of habitable?


A chart of habitable zones around stars that are hot, cold, and just right. Green means "go" for life as we know it.
—Image courtesy NASA


Computer whiz Samuel Arbesman of Harvard and exoplanet expert Gregory Laughlin of UC Santa Cruz say the answer is that a habitable planet will have 1) a surface temperature that would support liquid water, and 2) a planetary mass that indicates it's a rocky world, like Earth.

Based on these criteria, the pair developed what they call a simple habitability metric, H_k, where k is a given planet, and the value of H_k can run from 0 (uninhabitable) to 1 (just like Earth).
[There's some much more detailed math behind the variables involved in calculating H_k ... click here to see the study and all its juicy details.]

Once they had their equations in place, Arbesman and Laughlin ran their metric for 370 exoplanets with known properties.

Not surprisingly, the vast majority came up as H_k = 0. The "most Earthlike planet known," Gliese 581d, scored an H_k of just 0.01.

That's because most of the exoplanets found so far have been more massive than Jupiter, and the few known rocky worlds are so close to their stars they would have painful surface temperatures. I mean you, COROT 7b.



An artist's impression of a Jupiter-like exoplanet.
—Image courtesy NASA


But like the incredible shrinking computer chip, the technology for finding smaller exoplanets farther from their host stars is whizzing along at a breakneck pace. So what we need to do is consider the chances of finding increasingly Earthlike worlds over time.

To that end, the study authors ran what's called a bootstrap analysis, a statistical method that tells you what you *might have* seen, based on what you *have* seen.
A single run of the bootstrap analysis that takes into account H_k and the date of discovery for each of the 370 exoplanets creates a plot of H_k over time.

By running 10,000 sets of bootstrapped data, the pair created a distribution plot for H_k all the way out to the year 3000.

The median date for when a planet with H_k = 1 turns up is early May 2011.
But that's just the median. Chances get better over time, with a 66 percent probability of discovery by 2013, and a 95 percent chance of habitability by 2264.

"Of course, predicting future scientific and technological progress is a slippery and difficult process," the pair write in their study. After all, "history is littered with predictions that are far off the mark."
No. Kidding.

The exoplanet-prediction effort is also complicated by the fact that some of the most promising candidates for Earth twins have yet to be described publicly.
NASA's Kepler mission was designed specifically to look for Earth-size worlds in the habitable zones of sunlike stars. But the Kepler team is waiting until February 2011 to publish any details on 400 of the best planetary candidates it's seen so far.

For now, H_k is really just a thought experiment, rather like the Drake Equation, but one with equally provocative implications.