More than 700 planets have been discovered to date, but astronomers are seeking help from the public to find still more. Planet Hunters is part of the very successful Zooniverse family of projects.
This particular spin-off relies on citizen scientists to classify light curves from NASA’s Kepler spacecraft. Kepler was launched in March 2009 with the aim to stare intently at 150,000 stars for three and a half years. Kepler watches as it waits for its targets to blink, in an event known as a transit. A planet passing in front of a star will cause the light to dim ever so slightly and this periodic dip in the light curve can be detected.
From its orbit above the Earth, Kepler is free from the interference of the Earth’s atmosphere and it is able to get pristine data of its target stars. But 150,000 stars is quite a lot to monitor. A computer algorithm, known as a pipeline, first glues together observations taken at different times in order to form a light-curve. Another algorithm then finds any interesting blips in the light-curves and flags them as objects of interest for astronomers to look at.
How can the Kepler scientists be 100 per cent sure that the computer has spotted all the potential planets? In short they can’t, which is why they’ve asked the public to help them with the search. While the computer programs are incredibly efficient, it is hoped that the human brain’s capability to recognise patterns will help in identifying transits that the computer has missed.
Debra Fischer from Yale University, a member of the Planet Hunters team, explains to Skymania News how the citizen science project can compliment the computer pipeline. “Overall, the Kepler pipeline does an outstanding job of finding planets. The Planet Hunters simply provide human intervention to catch some of the planets that the algorithms missed,” she said.
Whether it is a computer or a person doing the planet-hunting, most detected transits end up being false positives, such as eclipsing binaries. In these cases, two stars orbiting each other will also produce a light curve with a periodic dip as one star passes in front of the other. These eclipsing binaries often have transit dips that are 10 per cent deeper than those caused by a planet. But sometimes the eclipsing binaries like to play a frustrating game by teaming up with a third star to mimic a planetary transit. The transit dip in an eclipsing binary can become diluted if a brighter, foreground star is present, and the third star isn’t always obvious if all three stars appear to be in the same position in the sky.
Can someone using the Planet Hunters website hope to discover the next Earth? “I think that this will be very difficult,” says Fischer. “As time goes on, I expect that we will do a better job of figuring out where computers excel and where human brains and pattern recognition is better. My guess is that humans may beat computers for long-period planets (orbits that are a year or more in length) but that these will still need to be at least three times the radius of the Earth.”
Several planet candidates have already been detected by Planet Hunters participants, including one that was found during BBC Stargazing Live. Fisher and her colleagues have just published a paper detailing two new candidates. Both of these potential planets were originally flagged by the pipeline, but later rejected. One of the stars was rejected because the type of stellar variability suggested that the star might be an evolved star, which would make observing transits too difficult. However, Planet Hunters saw beyond the variability and noticed the potential transit. When the star was analysed further by the Kepler team they realised that it was actually a main sequence star and that the transit indicates a viable candidate.
So why all this talk of “candidates” and not actual planets? A planet candidate can’t be positively identified as a planet until a Doppler follow-up is performed. The Doppler technique is responsible for discovering most of the exoplanets so far, and it relies on measuring the wobble of the star due to the effect of the planet’s gravity. It enables astronomers to calculate the mass of the object, and thus tell if it is a planet or a sneaky binary system. However this technique is limited, and the small planets being found by Kepler are difficult to follow up on with the Doppler method.
“We should be able to detect the gas giants at most orbital radii, and Neptune-sized planets in short period orbits with Doppler observations,” explains Fischer. “That will provide a mass for some of these planets. However, the Doppler technique does not have sufficient precision to detect planets the size of Earth. We are working on instrumentation to push this technique along, but the first successes are likely to be on bright nearby stars rather than Kepler stars.”
You can become one of the Planet Hunters at www.planethunters.org.