Scientists discover planetary system orbiting Kepler-47

Sep 13, 2012 by Dr. Tony Phillips
This diagram compares our own solar system to Kepler-47, a double-star system containing two planets, one orbiting in the so-called "habitable zone." This is the sweet spot in a planetary system where liquid water might exist on the surface of a planet. Credit: NASA/JPL-Caltech/T. Pyle

News flash: The Milky Way galaxy just got a little weirder. Back in 2011 astronomers were amazed when NASA's Kepler spacecraft discovered a planet orbiting a double star system.  Such a world, they realized, would have double sunsets and sunrises just like the fictional planet Tatooine in the movie Star Wars.  Yet this planet was real.

Now Kepler has discovered a whole system of planets orbiting a double star.

The star system, known as Kepler-47, is located 4,900 light-years from Earth in the constellation Cygnus. Two stars one another at the center of the system: One is similar to the sun in size, but only 84 percent as bright. The second star is smaller, only one-third the size of the sun and less than 1 percent as bright. Kepler found two planets orbiting this mismatched pair.

"The presence of a full-fledged planetary system orbiting Kepler-47 is an amazing discovery," says Greg Laughlin, professor of Astrophysics and at the University of California in Santa Cruz. "This is going to change the way we think about the formation of planets."

The , Kepler-47b, closely circles the pair of stars, completing each orbit in less than 50 days. Astronomers think it is a sweltering world, where the destruction of methane in its super-heated atmosphere might lead to a thick global haze.  Kepler-47b is about three times the size of Earth.

The , Kepler-47c, orbits every 303 days.  This puts it in the system's , a band of orbits that are "just right" for liquid water to exist on the surface of a planet. But does this planet even have a surface?  Possibly not.  The astronomers think it is a slightly larger than Neptune.

The discovery of planets orbiting means that planetary systems are even weirder and more abundant than previously thought.

"Many stars are part of multiple-star systems where two or more stars orbit one another. The question always has been—do they have planets and planetary systems?" says William Borucki, Kepler mission principal investigator at NASA's Ames Research Center. "This Kepler discovery proves that they do."

Our own sun is a single, isolated star, with a relatively simple gravitational field that rules the motions of the planets orbiting it.

But, as Borucki points out, not all stars are single.  Astronomers estimate that more than half of the stars in the galaxy have companions.  There are double, triple and even quadruple star systems.  Any planets in such systems would have to navigate a complex gravitational field, tugged in multiple directions by multiple stars.  In fact, for many years, astronomers doubted that planets could even form in such an environment.

Kepler-47 erases those doubts—and poses a conundrum: "These planets are very difficult to form using the currently accepted paradigm," says Laughlin. "I believe that theorists, myself included, will be going back to the drawing board to try to improve our understanding of how planets are assembled in the dusty gaseous disks that surround many young stars."

The Kepler spacecraft is on a mission to find Earth-like planets that might support life.  Says Borucki: "In our search for habitable worlds, we have just found more opportunities for life to exist."

Explore further: Partial solar eclipse over the U.S. on Thursday, Oct. 23

Related Stories

An exoplanet orbiting a double star

Oct 03, 2011

(PhysOrg.com) -- The Kepler satellite, which has now reported the detection of 1781 candidate exoplanets (a planet around a star other than the sun), has also discovered that at least one of them orbits a ...

Kepler announces 11 planetary systems hosting 26 planets

Jan 26, 2012

(PhysOrg.com) -- NASA's Kepler mission has discovered 11 new planetary systems hosting 26 confirmed planets. These discoveries nearly double the number of verified Kepler planets and triple the number of stars ...

Kepler Set to Launch Tonight on Planet Finding Mission

Mar 06, 2009

(PhysOrg.com) -- The Kepler spacecraft and its Delta II rocket are "go" for a launch tonight that is expected to light up the sky along Florida's Space Coast at 10:49 p.m. EST as the rocket lifts off from ...

Recommended for you

Partial solar eclipse over the U.S. on Thursday, Oct. 23

Oct 17, 2014

People in most of the continental United States will be in the shadow of the Moon on Thursday afternoon, Oct. 23, as a partial solar eclipse sweeps across the Earth. For people looking through sun-safe filters, from Los Angeles, ...

A newborn supernova every night

Oct 17, 2014

Thanks to a $9 million grant from the National Science Foundation and matching funds from the Zwicky Transient Facility (ZTF) collaboration, a new camera is being built at Caltech's Palomar Observatory that ...

Scientists build first map of hidden universe

Oct 16, 2014

A team led by astronomers from the Max Planck Institute for Astronomy has created the first three-dimensional map of the 'adolescent' Universe, just 3 billion years after the Big Bang. This map, built from ...

User comments : 35

Adjust slider to filter visible comments by rank

Display comments: newest first

geokstr
1.7 / 5 (13) Sep 13, 2012
I am uncertain why this is so inexplicable. This system is not much different than if Jupiter and Mercury switched places.

Jupiter is only a little bit smaller than it would take to become a star. If it had, why would the rest of the planets not be able to form?

I wouldn't doubt that eventually they will find double and even triple stars with planetary systems, and the stars being at various places in those systems, with the largest star in the center, and planets outside and inside the orbits of the others.

El_Nose
4.2 / 5 (17) Sep 13, 2012
There is no way to say this nicely so I am just going to say it. The reason you have no reason thinking planetary systems should be able to form in systems with multiple stars is simply because you are uneducated in this field.

Humans don't know enough to model this behavior. We (humans) have felt that this is impossible because we could not come up with a model that did not fall apart to create planets in a binary system. All the models said the planets would fall into their stars early in development.

This observation proves they exist so we need to try to make new models.

Or -- the planets were captured rouge planets and this is not the norm.

We need more observations to be sure.
chardo137
3.9 / 5 (9) Sep 13, 2012
We do not understand the dynamics of protoplanetary discs any more than we understand the dynamics of accretion discs around black holes. There is a Nobel prize waiting for the person who first makes the models match the observations. No one really doubts the nebular hypothesis, or that accretion discs around black holes power Active Galactic Nuclei. There is much exciting science still to be done in this area. While we, of course, need more observations, solving this problem is in the domain of the theorists. All hail the guys with the pencils!
eachus
2.8 / 5 (5) Sep 13, 2012
There is much exciting science still to be done in this area. While we, of course, need more observations, solving this problem is in the domain of the theorists. All hail the guys with the pencils!


Gee, I would say it is in the hands of applied mathematicians. Even with supercomputers, you have to make many compromises in your model to run it for any substantial period of (simulated) time. And for these models, one million (simulated) years is short.

One way to fix it is with brute force: bigger and faster computers. But the real progress in this area is when someone comes up with a simulation method that allows for bigger time steps, and still getting the same (or close to it) answer.

Incidentally, for the solar system, the rough rule of thumb is that projections of planet, moon, and asteroid locations starts going wonky about 100 years out. It might be possible to push very accurate predictions out to a thousand years, but a million? Forget it.
Husky
5 / 5 (8) Sep 13, 2012
i venture to say that a neptune sized kepler 47c could harbour a host of habitable moons.
Argiod
2 / 5 (12) Sep 13, 2012
Searching for a humanly viable planet is like Goldilocks trying to find the stuff that's 'just right'. We search endlessly, and come up with planetary systems; but they all seem to be too big, too small, too close, too far, to cold, too hot... Looks like we'll be searching for a long time before we find one that's 'just right'.

Maybe we should learn how to take proper care of this one, while we still have it...
MrVibrating
3.7 / 5 (6) Sep 13, 2012
For anyone who played Elite 2 back in the day, these types of systems were a dime a dozen. Eventually Kepler will probably resolve ternary, quaternary, quintuple & sextuple (etc.) systems too...
nuge
4.3 / 5 (3) Sep 13, 2012
Despite the complexity, I'd have thought that ultimately far enough away from the stars it would be for all intents and purposes equivalent to a single star at the centre of mass of the two. Expecially since one is so small anyway. It might cause a slight wobble. The calculations for solving the system exactly are obviously very complex, and perhaps impossible to solve with our current methods....but surely we realise that the part we can't solve is really just dotting the i's and crossing the t's - we get that the general gist of it is a big heavy thing in the middle?
antialias_physorg
4.5 / 5 (8) Sep 14, 2012
Looks like we'll be searching for a long time before we find one that's 'just right'.

Since not even the Earth was 'just right' for human life when it started out (no oxygen in the atmosphere) - what do you expect?
Trewoor
1.5 / 5 (8) Sep 14, 2012
…The reason you have no reason thinking planetary systems should be able to form in systems with multiple stars is simply because you are uneducated in this field.


El Nose, I do entirely agree with geokstr.

This is not the meter of being educated in a subject. Even more I would say, by being educated in it, they (or you) should realise how much we do not know and don't understand! 99% of astronomy is based on theories and assumptions and we are creating new theories biased on other theories. This can only lead to fault results. We cont even model our own whether condition on our own planet!
The best example of fake assumptions are your own words

…the planets were captured rouge planets and this is not the norm.

How in the world can you be so sure about that?
Therefore when I hear that some astronomer are surprised on things like this discovery it just makes me laugh on their ignorance…
El_Nose
1.3 / 5 (6) Sep 14, 2012
@trewoor

You are right astronomy is a purely theoretical field. We don;t know much about it at all. And as for the rogue planet... well this is after all only one observation, I was just giving the thousands of astronomers a convenient out...
GSwift7
2.6 / 5 (5) Sep 14, 2012
I am uncertain why this is so inexplicable. This system is not much different than if Jupiter and Mercury switched places.

Jupiter is only a little bit smaller than it would take to become a star. If it had, why would the rest of the planets not be able to form?


I agree with this, as do many in the field of astronomy these days. Finding binary systems with more than one planet has been expected by many. We have found several binary stars with one planet already. This is just the first one where there happen to be two detectable planets.

Keep in mind that Kepler has barely had enough time to get results. Planets with longer orbits wouldn't be detected by Kepler yet, so there should be more to come.

The Kepler team estimated that 1 in 20 stars has an earth sized planet and slightly less than that have multiple planets. That was last year. The above finding should increase that estimate though.
blazingspark
not rated yet Sep 14, 2012
I don't know anything about solar system formation... Just wondering: If our models don't account for planets forming around binary systems, what about having the star be a single star as the planets developed, then a rogue brown dwarf comes along at high speed and smashes into the sun knocking enough material into a close orbit to create a binary system?

If a brown dwarf slightly below the mass required to ignite stable fusion grazed the sun could it trade some velocity in exchange for extra mass and a stable orbit?

Earth's moon was created in a similar fashion early in our solar system's history.
packrat
1 / 5 (4) Sep 14, 2012
Couldn't the larger sun already have had the planets and the little sun come drifting in at a relatively slow speed and get caught by the big one's gravity and then begin to orbit it?
Fleetfoot
5 / 5 (2) Sep 15, 2012
Couldn't the larger sun already have had the planets and the little sun come drifting in at a relatively slow speed and get caught by the big one's gravity and then begin to orbit it?


It would pass on an hyperbolic orbit, To be captured it has to lose momentum so typically it would need to eject a previous planet for the system.
Torbjorn_Larsson_OM
5 / 5 (2) Sep 15, 2012
Planets around binaries are not new, having a habitable zone planet around a binary is.

@ geokstr:

"I am uncertain why this is so inexplicable."

The main reason is that models of protoplanetary disks, predicting planet formation in the first place, have problems with binaries. But they have problems with more mundane Kepler observations (hot Jupiters), so they are under reassessment. It was just a few weeks ago that they naturally predicted the ice line for our own system!

Another reason is that formed planets are pulled by more gravity sources. Observations show that these systems are less densely populated (see the figure), because increased distances between planets compensate for the varying pull of orbiting binary stars.
Torbjorn_Larsson_OM
2.5 / 5 (4) Sep 15, 2012
@ Treewoor:

You may want to supply a reference for that figure of yours.

Another word for a tested theory is "solid fact", and another word for "assumptions" (which science in fact doesn't do) is constraints.

Astronomy is our oldest empirical science, and it has been tremendously successful. Together with cosmology we can now predict structure formation on all scales down through galaxies (where cosmology lets off) to stars and their Main sequence. Modeling protoplanetary disks is for the first time meeting the bulk of observations, give it time and it will sort itself out.

Are you trolling climate science denial? Climate science has done very well in prediction indeed (see IPCC -07), and similarly we can predict weather within weather science over a few days despite the chaotic nature of parts of it.
baudrunner
1 / 5 (4) Sep 15, 2012
"I believe that theorists, myself included, will be going back to the drawing board...
yah, yah...whatever. HAH!
VendicarD
1 / 5 (1) Sep 15, 2012
Three body gravitational problems are impossible to solve. Period!

No mathematical solutions exist.

Simulations however are possible, going forward billions of years if necessary.

Our planetary system has been simulated forward several billion years using computers designed specifically for the task.

"The calculations for solving the system exactly are obviously very complex, and perhaps impossible to solve with our current methods..." - nuge
VendicarD
1 / 5 (1) Sep 15, 2012
God does not play Elite 2.

"For anyone who played Elite 2 back in the day" - MrVibrating
MrVibrating
1.8 / 5 (5) Sep 16, 2012
Believe what you like, my faith is strong that David Braben IS God, and that this universe is actually Elite 4, which we've been playing all along - the endless wait for its release is but a test; one that you've clearly failed, along with everybody else, apart from me, the last sane player in the greatest video game never released. :P
yaridanjo
1 / 5 (3) Sep 16, 2012
"These planets are very difficult to form using the currently accepted paradigm," says Laughlin.
Then try a new one that actually works on our own solar system.
Google: ASTRO-METRICS Of Undiscovered Planets And Intelligent Life Forms http://www.yarida...sel.com/

PS, It predicts the Bode law.
TABLE 8-2
PREDICTED PLANETARY ORBITS
SOLAR PLANETS

and there is another little surprise there too.

Torbjorn_Larsson_OM
not rated yet Sep 17, 2012
@ varidanio:

It is obvious for scientists that a generic pattern search data fishing is not helpful and in most cases pseudoscience. The "Bode law" is a non-generic pattern for some planets in our system, the discovery of exoplanets and their diversity put a stop to any lingering ideas to predict it.

Instead the Nice model for our system with its extensions predict the planets from Mars to Neptune together with a lot of other details, a recent update of the inner protoplanetary disk formation model predicts the remaining terrestrials and our ice line, and a recent supernova molecular cloud model predict many of the Kuiper Belt Objects traits together with a lot of other details.

The over all problem isn't to understand our system, the problem is that other systems have behaved very differently and as I understand it there are too many reasonable models at the moment.
GSwift7
1 / 5 (3) Sep 17, 2012
Just wondering: If our models don't account for planets forming around binary systems,


That's a bit of an exageration, and not all of the models agree on that point. It's more of a probability or lack of probability, rather than the models excluding the posibility entirely. Besides, some of the mainstream models actually suggest that a binary system may be more likely to form planets rather than less likely, though keeping those planets after they form might be a different case. Take a look at the wiki page on binaries if you want to read some cool stuff: http://en.wikiped...ary_star

In the case of our nearest neighbor, Alpha Centauri, there's plenty of room for Earth to orbit either member of the A-B pair, and the second star would only appear as a faint dot in the night sky, as Sol would look from Pluto. Most binary systems are far enough apart to allow S-type orbits around one or the other member of the pair. Note: an "S-type" orbit is not "S-shaped".
Bog_Mire
1 / 5 (3) Sep 19, 2012
we could never even begin to dream of getting there - right?
antialias_physorg
1 / 5 (1) Sep 19, 2012
Depends of in what form and over what kind of timeframe.

5000 lightyears is still within our galactic suburbs.
GSwift7
1 / 5 (3) Sep 19, 2012
we could never even begin to dream of getting there - right?


There are many technical barriers. Even a robotic probe is not reasonable with current technology. There's a wiki page about interstellar travel. It's interesting. They bring up the point that our ability to improve propulsion systems means that if you launched a probe with today's state-of-the-art propulsion, it would take so long to get there that you could have just as easily waited and built one with more advanced propulsion, and have the second one get there first. The best we could do right now is still probably hundreds of years travel time, and that's assuming a few advances we don't have yet, but probably could if we tried.

Some engineers don't think humans will ever leave the solar system unless there's a way to bend the laws of physics.
Fleetfoot
not rated yet Sep 20, 2012
we could never even begin to dream of getting there - right?


Even a robotic probe is not reasonable with current technology. .. The best we could do right now is still probably hundreds of years travel time, ..


Ships carrying people look untenable with any credible technology, self-replicating robot probes are the obvious approach. Solar sail propulsion is one of the fastest and has a significant advantage in not needing on-board fuel. I've done some analysis and ~2% of c is about the limit for aluminium launched from a Sun-like star.

Some engineers don't think humans will ever leave the solar system unless there's a way to bend the laws of physics.


Only if we can develop a method using email ;-)

The robotic probe fleet could create the star-to-star communications network.

However, it would be possible to send the DNA for humans and have clones of them grown at the destination, the race might leave the planet even if individuals could not.
antialias_physorg
1 / 5 (1) Sep 20, 2012
There's several ways on could envision humans leaving the solar system.

a) we figure out how immortality works (e.g. by understanding so much about the body that we can either repair all damage on the cellular level as it happens or bioengineer ourselves to the point where our bodies are able to do so themselves). i find this unlikely to happen
b) As stored seed (frozen fertilized egg) which gets incubated at the destination. I also find this unlikely as we'd be missing anyone to raise the baby. Even with sentint robots that would probably not work out well.
c) As 'mind-copied-into-machine'. This seem most likely to me. If not 'really' human it would be something so close that you couldn't tell the difference.

All three of these make the timefactor for reaching any destination pretty much irrelevant. So 'slow' travel would be acceptable.
GSwift7
1 / 5 (3) Sep 20, 2012
a)
b)
c)


True, but if you are willing to imagine technology that far ahead of where we are now, then there are certainly other possibilities that aren't imaginable. Like a 5th century person imagining microelectronics.

No matter the method, thermodynamics might be the ultimate limit on travel. Any propulsion system that obeys the laws of physics as we know them must deal with waste heat that would accumulate in the spacecraft. With the power required to reach relativistic speeds, even a small amount of inefficiency in your propulsion system would be a show stopper.

The problem with a sail is that most of your trip would be unpowered.

Microscopic interstellar debris may be another show stopper no matter how clever we become.
Fleetfoot
not rated yet Sep 20, 2012
The problem with a sail is that most of your trip would be unpowered.


That's not a problem, it's a benefit. If debris limits your speed, get up to that limit as quickly as possible then cruise at the maximum. The sail also uses the destination star to slow, no braking system needed.

Microscopic interstellar debris may be another show stopper no matter how clever we become.


Even without debris, there would be degradation over time. An essential technology to be developed is therefore self-repair, equivalent to self-replication, that's probably the hardest problem.

The second key technology is on-board power for the repair system. You can't use an RTG because you can't find native fuel but spinning the sail and generating power by interacting with the galactic magnetic field has been proposed. Spin is easily imparted by angling vanes on the sail edge and EMF can be induced through the aluminium so again, no extra mass is needed.
GSwift7
1 / 5 (2) Sep 21, 2012
Even without debris, there would be degradation over time. An essential technology to be developed is therefore self-repair, equivalent to self-replication, that's probably the hardest problem


If you are moving at a reasonable portion of C, like 10 or 20 percent or more, hitting a piece of sand isn't going to leave anything to repair.

That's not a problem, it's a benefit. If debris limits your speed, get up to that limit as quickly as possible then cruise at the maximum. The sail also uses the destination star to slow, no braking system needed


If you are okay with travel times over 500 years then that's okay. I don't really think that's reasonable though. A multistage hybrid system is more likely. If you used a sail, it would only be to get started. You'd want to discard it before you leave the solar system. The gains from using it to brake at the other end wouldn't equal the cost to acceleration for the bulk of the trip.
Fleetfoot
not rated yet Sep 21, 2012
If you are moving at a reasonable portion of C, like 10 or 20 percent or more, hitting a piece of sand isn't going to leave anything to repair.


As I said, around 2% is the theoretical max, 1% may be achievable. Most attrition will be ions but a cone shape flying base first would retain vapourised material from dust for repair.

A multistage hybrid system is more likely.


If you think you can beat 1% of c, I'd like to see your calculations.

If you used a sail, it would only be to get started. You'd want to discard it before you leave the solar system. The gains from using it to brake at the other end wouldn't equal the cost to acceleration for the bulk of the trip.


There is no cost for acceleration, the sail is your propulsion system! The trick is making sure the payload is distributed and small compared to the mass of the sail. Terminal speed scales with the square root of areal density, it halves if the payload is three times the mass of sail per unit area.
antialias_physorg
3 / 5 (2) Sep 21, 2012
I'm not sure solar sails will do the trick. For a sizeable craft they'd have to be huge and lightweight (which means flimsy). At high speeds impacts will shred them badly.

I'm really not seeing a viable drive technology yet for interstellar travel. Exhaust driven drives are certainly a no-go.
Fusion drives which simply put out photons the back, might.
Fleetfoot
not rated yet Sep 22, 2012
I'm not sure solar sails will do the trick. For a sizeable craft they'd have to be huge and lightweight (which means flimsy). At high speeds impacts will shred them badly.


My analysis suggests the optimum thickness for aluminium is about 45nm so "flimsy" is accurate, it would need to be on a support structure. The best technology would look like carbon nanotube scaffolding with embedded graphene computing forming a honeycomb wall structure.

Impacts would punch holes through rather than tearing it, especially if the structure was an array of small, adjoining plates, hexagonal probably. The sail would need to reconfigure its shape so electrostatic actuators like those on MEMS video projector chips could be used along the plate edges.

Most of that technology is in commercial use or in the lab already.

The size is governed by that needed to create a synthetic aperture transmitter/receiver capable of interstellar communications, say Rayleigh length in excess of a light year.