News blog

Citizen scientists beat Kepler team to candidate planets

phpwW2iGKPM.jpgAmateur scientists combing through data released by NASA’s planet-finder, Kepler, may have discovered as many as 47 new candidates, says Debra Fischer, an astronomer at Yale University in New Haven, Connecticut, who is involved with the citizen science project at The site asks internet users to flag dips on light-curves – graphs recording the brightness of Kepler’s 150,000 target stars — that could be caused by orbiting planets passing in front. It went live December 16th, 2010, and has been using raw data from Kepler’s first release, which took place in June 2010.

Fischer says a couple of hours ago, she got word from a post-doc tasked with cross-checking candidates identified by amateurs on the site with candidates released by Kepler February 1st. The post-doc, Meg Schwamb, found that 47 are novel finds. The Kepler team uses software to identify dips on light-curves.

There is a big difference between candidates and confirmed planets, and the Kepler team goes through a painstaking process before announcing and naming planets, as a feature in Nature this week explains. Candidates can also be vetted to different levels of thoroughness and the mere listing of a planet as a candidate is far from a guarantee that it will turn out to be real. But there seems little doubt that the amateurs are finding some genuine possibilities. Fischer says that 83 of 130 systems listed on before February 1st were independently announced by the Kepler team February 1st, at the same time as the NASA project’s second data release. The Kepler team has claimed its rate of false positives in its sample may be as low as 10%. The 47 candidates not announced by Kepler suggest amateur efforts to sift the Kepler data may lead to planets that the Kepler team would not on its own ever have found, says Fischer. “This is an out-of-the-box approach that makes use of the pattern recognition ability humans have,” Fischer says, “this is going to be the place where humans beat the computers.”

Fischer says some amateur planet hunters have been asking whether they deserve priority over the Kepler team for announcing some systems before Kepler, even if Kepler data was involved. Being beaten to its own discoveries may be one risk that Kepler has taken by holding its candidates back for some months while it confirms them. But Fischer downplays the competition, saying the most value comes from finding planetary candidates that Kepler does not or cannot find, rather than in seeking priority for those that Kepler finds as well.

Natalie Batalha, deputy team lead for Kepler, says she’s tremendously excited about the amateur effort. “I am absolutely expecting that they will successfully identify planet candidates that our pipeline misses” she says, “you bet they’ll get credit for the discovery.”

Of the 1200 planetary candidates announced by Kepler, 54 are in the “habitable zone”, meaning that the temperature is suitable for liquid water to exist on their surface, or on the surface of any moons they might have. NASA actually announced that these 54 are the first habitable zone candidates, but Batalha clarifies that, in fact, five candidates were reported as part of the June 2010 data release.

Image: Searching for dips/


  1. Report this comment

    david johnson said:

    i hope to god that we never discover a planet similar to earth.

    lets pray that IF we do, the inhabitants, if possessing intelligence, have not already sown the seeds of their own destruction as we obviously have.

    Dave Johnson

Comments are closed.