How astronomers find real (and fictional) planets around other stars

Scott Sutherland

Ever since I first heard of, I've been going there at least a few times each week to help search for exoplanets in the Kepler telescope data.

What you look for, on the site, are transits — dips in the brightness of a star's light that signal that something has passed between us and that star. It's hard to pick these out sometimes, because the dips can be so small that they get lost in the 'noise' — the normal variation seen from data-point to data-point in the star's light curve.

[ More Geekquinox: Astronomers detect water in exoplanet’s atmosphere ]

You can go through a fair number of stars — 'quiet', 'variable', 'pulsating', 'eclipsing binaries' — before you see some obvious signs of a planet, but it's rewarding when you find one, especially when it turns out to be one of the Kepler team's favourites.

The Planet Hunters site has some great examples to look at, but scientists can also simulate transits on the computer, to give them a better idea of what to look for in the Kepler data. Since I follow Professor Abel Méndez on Twitter (@ProfAbelMendez), I spotted a neat post on his blog, showing some simulated transits of planets and their moons. Prof. Méndez is a planetary scientist and Director of the Planetary Habitability Laboratory at the University of Puerto Rico at Arecibo.

The simulations are made with a program known as SPHERE-SIM, which is used by the SPHERE project (Search for Potentially Habitable Exoworlds Resembling Earth).

The first he shows us is a simulation of what Earth would look like to some hypothetical alien race with a hypothetical telescope like Kepler that was pointed at our Sun. The cool animated image that he has is too big for this post, but here's a sample from it showing when Earth (the tiny dark speck) is nearly at the centre of the solar disk. (See his post for the full animated image and other Earth light curves.)

It's easy to pick out the dip in the light with the "expected solar flux" line included, but when you're looking at Kepler light curves, you really only see the dots, so picking out that the dip caused by the planet is actually a dip caused by a planet, and not just some normal variation in the star's light, isn't easy.

It's easier for larger planets, though. He includes a simulated light curve from an actual confirmed exoplanet, Kepler-22b, a potentially-habitable planet about two and a half times the size of Earth, orbiting a Sun-like star (Kepler-22) around 600 light years away. As you can see, the larger planet causes a deeper dip in the light curve, so it's easier to pick out, even without the 'expected solar flux' line.

A fun example he slips in is the fictional planet and moon from the James Cameron movie Avatar (I love that the image even includes the 'Papyrus' font the movie-makers used for the title and the subtitles).

The light curve really shows off the planet, though. Polyphemus, the massive gas giant that Pandora orbits around, causes a huge dip in the light curve, compared to Earth and Kepler-22b. It's hard to pick out Pandora itself in the light curve, but if you look closely at the expected solar flux line at around -20 on the time scale, there is a very slight dip there.

Of course, as Prof. Méndez points out: "We already know that it is unlikely that a planet such as Polyphemus exist around Alpha Centauri A since we already have the capability to detect such large planets (sorry Jim). However, smaller Earth-size ones are still possible, such as Alpha Centauri B b."

[ More Geekquinox: Could a comet impact produce a more habitable Mars? ]

If you find this kind of thing interesting (like I do), you can join the search for exoplanets on the Planet Hunters website, or help examine the properties of confirmed exoplanets at the Agent Exoplanet site.

(Images courtesy: NASA/Kepler mission/Wendy Stenzel and SPHERE project, PHL @ UPR Arecibo)

Geek out with the latest in science and weather.
Follow @ygeekquinox on Twitter!