ExoPlanet Hunting with Kepler 2

The Kepler “Planet Finder” mission has published new results, describing over 1,000 extrasolar planets. Cool!

First of all, the current mission is dubbed “Kepler 2”, because the orbiting telescope had to be reconfigured when two of its four reaction mass flywheels failed, eliminating the means to steer and aim the instrument. The “K2” concept salvaged several more years of life, with the really cool use of solar wind to steer the telescope.


The big splash this week (NASA did a press conference, reported by main stream media) is the report of over 1200 planets in orbit around other stars, including 100 that are approximately the size of Earth.

We’ve long assumed that there are zillions of planets out there, and, sure enough, Kepler is finding them.

The new paper ([1]) is actually mostly about how these discoveries were made.  The telescope identifies signatures that may indicate planets transiting a distant star. These candidates must be examined to weed out false positives, such as binary stars. Kepler sucks in thousands of such observations, so there needs to be automated processes for analyzing these cases.

(Note that this only detects planets in a limited range of orientations to Earth—it is likely that there are many, many more planets that Kepler simply can’t detect because they aren’t visible form here, or their orbit is not “just right”. For that matter, Kepler has only sampled a tiny segment of the sky.)

The paper by Morton and colleagues describes a probabilistic approach, by which they estimate the probability of various possible scenarios, to assign a probability that this is a “false positive”, likely not a planet. The 1200 reported planets are the observations that they calculate a low probability (less than .01) of being a false positive.

This is nice work (though I have not actually worked through the math myself.) They have done a careful job of explaining the approach and identifying areas of uncertainty. Naturally, this kind of argument will benefit from additional empirical support in the form of future observations to confirm the probabilities.

For the record, their software is available so you can check their work, or use it for another project. [source code]

  1. Timothy D. Morton, Stephen T. Bryson, Jeffrey L. Coughlin, Jason F. Rowe, Ganesh Ravichandran, A. Petigura Erik, Michael R. Haas, and Natalie M. Batalha, False Positive Probabilities for all Kepler Objects of Interest: 1284 Newly Validated Planets and 428 Likely False Positives. The Astrophysical Journal, 822 (2):86, 2016. http://stacks.iop.org/0004-637X/822/i=2/a=86


Space Saturday

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s