Science

Scientists say at least one in six stars has an Earth-sized planet

Scientists say at least one in six stars has an Earth-sized planet
A recent analysis of the data gathered from the Kepler telescope has revealed that Earth-sized planets across our Milky Way are much more common than previously thought (Image: C. Pulliam & D. Aguilar/CfA)
A recent analysis of the data gathered from the Kepler telescope has revealed that Earth-sized planets across our Milky Way are much more common than previously thought (Image: C. Pulliam & D. Aguilar/CfA)
View 2 Images
A recent analysis of the data gathered from the Kepler telescope has revealed that Earth-sized planets across our Milky Way are much more common than previously thought (Image: C. Pulliam & D. Aguilar/CfA)
1/2
A recent analysis of the data gathered from the Kepler telescope has revealed that Earth-sized planets across our Milky Way are much more common than previously thought (Image: C. Pulliam & D. Aguilar/CfA)
A new analysis examined the frequencies of planets of different sizes based on findings from NASA's Kepler spacecraft, correcting for both incompleteness and false positives (Image: F. Fressin/CfA)
2/2
A new analysis examined the frequencies of planets of different sizes based on findings from NASA's Kepler spacecraft, correcting for both incompleteness and false positives (Image: F. Fressin/CfA)

The latest analysis of data coming from NASA's Kepler telescope has revealed that nearly all the Sun-like stars in our galaxy appear to have planets orbiting them, and that at least 17 percent of them – about one in six – are hosting a planet the size of our own in close orbit. Because the Milky Way is estimated to contain some 100 billion stars, this means that our galaxy alone could have at least 17 billion Earth-sized planets, some of which may harbor the conditions for life.

Launched in March 2009, the Kepler telescope is detecting planets in faraway stars using the transit method – that is, detecting the periodic, tiny decrease in the amount of light we receive from a star that we can detect when a planet eclipses its own star from our observation point.

Despite Kepler's technical prowess, this method is still fairly limited. For the planet to be detected, it must happen to transit exactly between the star and our observation point. Moreover smaller planets, which block out little light, are generally harder to discern, while planets in wide orbit require, on average, a much longer observation period to be detected as they won't eclipse their star nearly as often.

After a thorough 16-month survey, Kepler was nonetheless able to identify some 2,400 planetary candidates. And now, an international group of researchers has carefully analyzed the data gathered from the telescope and concluded that only about ten percent of Kepler's planetary candidates appear to be false positives, while all the others are actual planets.

When tallying up the data and extrapolating it to our entire galaxy, accounting for the planets the transit method simply can't detect, a surprising picture emerges. Half of the stars in our galaxy appear to have a planet at least as big as our Earth in close orbit; when also including larger planets in wider orbits, the number shoots up to 70 percent. If we add to this figure the data obtained with other planetary detection techniques, it seems that nearly all Sun-like stars have planets orbiting them.

A new analysis examined the frequencies of planets of different sizes based on findings from NASA's Kepler spacecraft, correcting for both incompleteness and false positives (Image: F. Fressin/CfA)
A new analysis examined the frequencies of planets of different sizes based on findings from NASA's Kepler spacecraft, correcting for both incompleteness and false positives (Image: F. Fressin/CfA)

Of these stars, about 17 percent have a planet 0.8 to 1.25 the size of Earth in an orbit of 85 days or less; about 25 percent of stars have a planet 1.25 to 2 times the size of Earth in an orbit of 150 days or less; a further 25 percent of stars has a planet 2 to 4 times the size of Earth in orbit up to 250 days; and finally, only one star in twenty has a gas giant (6 to 22 times Earth's size) in an orbit of 400 days or less.

In direct contradiction with previous results, the researchers also found that there seems to be no clear relationship between the type of star and the size of the planet orbiting it. In other words, super-Earth or Earth-like exoplanets could be it orbiting nearly any star, be it a red dwarf or a Sun-like star.

Back in 2011, data from Kepler had brought astronomers to estimate the number of planets in our galaxy at 50 billion, of which 500 million would be in the so-called "habitable zone" – with the right temperature and pressure to maintain liquid water on their surface, opening the door for life to evolve.

Scientists, however, don't quite yet know exactly what the right conditions for life are. To the best of our knowledge, planets must orbit in the habitable zone, but other conditions may have to be met as well. For instance, it is thought that the presence of a gas giant (Jupiter) in our solar system could have acted as a shield against most passing asteroids, protecting the Earth's fragile biochemistry.

Kepler's ongoing mission won't help astronomers test out such theories directly but, by helping characterize the incidence of gas giants and other types of planets across our galaxy, it could help pinpoint how many solar systems virtually identical to our own are out there – a good place to start our search.

The new analysis was recently presented to the American Astronomical Society by Francois Fressin from the Harvard-Smithsonian Center for Astrophysics (CfA).

A paper detailing the research has been accepted for publication in The Astrophysical Journal.

Sources: CfA, NASA Kepler site

3 comments
3 comments
Bruce H. Anderson
So.....maybe we're not alone. Trouble is, it would take a very long time for anyone to pay us a visit, or for us to visit them. And once we got there, whatever we were headed for may not exist any more. And if it did, would we even know what it/they are? It seems the more we learn, the more we understand how much we don't know. What a fascinating place this Universe is!
Fritz Menzel
Whenever I read about data extrapolated from planetary transits I wonder how accurate those extrapolations might be. Here's why:
What are the chances that the planetary orbital plane (accretion disk) of a given star is within the plane of our viewing angle?
I suspect the chances are very low. Just a 1° planar tilt would make distant transits of all but improbably large/near planets unobservable. Accepting that logic, only 1 out of every 180 (degrees) planetary orbital planes provide observable transits, meaning we see only a tiny fraction (fewer than .6%) of actual transits. In fact, I suspect it takes far less than a 1° planar tilt to make a transit unobservable, making that a very conservative estimate.
I understand that wobbling also provides a planetary observation method, albeit far less reliable for smaller planets around more distant stars. But perhaps NASA could extrapolate wobble data to at least determine what percentage of planetary orbital planes (ergo transits) are within our viewing angle, using that to more accurately extrapolate the probability of earth-size planets.
Bottom line: There could be many times more than just 17% of stars featuring earth-size planets.
Jon A.
We have absolutely no idea what is out there.
We have only been able to catalog those planets that happen to be detectable using our current technology, which is very limited.
Trying to issue a definitive statement based on available data is either foolish, or downright dishonest.