Amateur astronomers discover a planet with four suns | Hacker News Reader

Amateur astronomers discover a planet with four suns

io9.com

116 pointsboh37 comments

I tracked down the original paper (http://arxiv.org/abs/1210.3612) and dug up some useful details.

The planet in question is a gas/ice giant similar to Uranus with a radius 6 times that of Earth (nearly 80,000 km in diameter) with an as yet unknown mass (although it must weigh less than half of Jupiter's mass at the most). It orbits a pair of binary stars with a period of 138 days. The binary it orbits is an F dwarf star with 1.5 the Sun's mass and an M dwarf with 40% of the Sun's mass, they orbit each other with a period of 20 days. For reference, this translates to the two stars orbiting much closer than Mercury's orbit (around 0.17 AU) and the planet with an orbit close to that of Venus (around 0.65 AU, according to my calculations).

This system is bound to another binary system of similar total mass (a G2 star similar in mass to our Sun and an M2 dwarf star with around half or less the mass of our Sun) at a distance of around 1,000 AU. At that distance the second binary system would merely be the brightest stars in the night sky of the planet. The two binary systems would orbit each other with a period of tens of thousands of years.

I've tried about 8 different ways to modelling that and so far all of them have the planet being ripped apart as soon as its orbit is elongated by the traversing between the two pairs of binaries. Bizarre doesn't begin to describe it.

Maybe the orbit is perpendicular to the stars' plane of orbit?

This doesn't make sense to me. The two pairs of binaries are around 6 light-days apart. The gravitational forces they have on each other are miniscule, but they are persistent enough to cause the two systems to orbit each other with a period longer than the duration of human civilization.

Yes, and if you model the force on the planet when it travels between them it has a small tidal effect. Depending on the relative orbital planes of the planet and the second binary star system, that may occur any time from once per orbital period to once every 10,000 years. But now try to see how the planet stays in a stable orbit over a billion years. At least in the models I've been able to come up with the tidal effect is assymetric, which is to say the tweak to orbit of the planet is not counter acted by a symmetric tweak on the far side. Even starting with the 'suns' being effectively two point sources 1000AU apart in a bilaterally stable orbit. Trying to find a way to reliably orbit a planet around one of them stably is eluding me. If you can come up with some orbital parameters that work I'd love to see them.

Apologies for my skepticism, my estimate is that the distant binary star system would impart a force on a planet less than 5% that of Jupiter on Earth. Could you provide more details on your n-body simulations? Such as what code base you are using, etc.

Wow that´s great, I have always wondered how astronomers come to calculate the orbits of objects like asteroids or comets just knowing a small part of their movement. Modeling a system as complex as this double double, is just awesome (even if it´s not holding water yet). What programs are you using? is there a book or web that explains to a layman how this is done?

They were using Planet Hunters - which appears to be a crowd sourced "find interesting bits in all this data" website. (http://planethunters.org/)

I work for the group that made Planet Hunters. We have a variety of similar projects at http://zooniverse.org, including four others we launched in the last month that have people classifying sea floor life, galaxies, cyclones, and bat calls.

They're very cool projects! I've been noodling around for a bit.

I'll register if I think I'll be any good, and I'll certainly be passing the URLs around to people if I think they'll be interested.

Could someone explain how that site works?

Are the amateur astronomers mentioned in the article user of planethunters.org who just happen to have been served the right pictures to work on? Is this Mechanical Turk type work or does it require serious scientific effort?

Any chance someone could make a diagram of the orbit for us? I really can't picture this? Does it make a figure 8 around the first two stars?

It's a pair of binary stars with a planet orbiting one pair.

To give you a conceptional model, imagine that Mercury is a star and then imagine that, say, Saturn and one of its moons are also stars. Now, obviously, orbital mechanics are going to change when you have all of these much more massive bodies in the system, but hopefully you get the picture.

Edit: obviously not to scale, but maybe this helps: http://i.imgur.com/RJrOw.png

Your parent is asking how the planet orbits two stars, and probably how the 2-star-1-planet system orbits the two other stars. That is, "how does any body orbit two stars?"

And I explained. How does the Earth orbit the Sun-Mercury system? It's no different than how a planet orbits a binary system from afar. The two stars orbit each other and then the planet orbits around both stars. And then that whole system orbits around another binary star system at a tremendous distance (1,000 AU, ten times farther away than Voyager 2 is from the Earth now).

It's a little different; in our solar system, the Sun is one of the foci of the ellipse, whereas in a binary system that focus would instead be the center of gravity between the two stars.

Yes, though in this case the big star weighs nearly 4x as much as the little star, so it's a matter of one star shuffling around in a small ellipse with the smaller star in orbit, rather than a more intimate orbital dance if the stars had been closer in mass.

Although Mercury in this instance is extremely insignificant, technically the earth orbits the center of gravity of the Sun and Mercury. Actually, the center of gravity of the Solar system as a whole, but I'm just going along with the GP.

Gosh, how incredibly helpful.

We still have an analogy in the Solar system, Hydra and Nix orbit the Pluto-Charon binary. The barycenter of the Pluto system is outside the surface of the dwarf planet.

http://upload.wikimedia.org/wikipedia/commons/thumb/d/d6/Plu...

Normally, it doesn't orbit two stars, it orbits their combined center of gravity (which, if the stars were the same mass, would lie exactly between them). Or to be more precise, both the stars and the planet together orbit their combined center of mass.

Its complicated. You can approximate it using the 2-star-center-of-mass, but thats skipping over important stuff.

See, the stars are orbiting one another. At any time one may be closer/further away than the other from the planet. That makes the COG appear to 'wobble'.

IF the planet is in something like a 'lagrange point' where it always sees both stars in the same relative position, then the COG approximation is true. Otherwise the planet will behave in a more complicated manner, and in fact may not have a stable orbit at all.

Thanks. That is helpful. So there's no scenario where a planet could be in a figure 8 orbit? That would be amazing.

It's probably similar to other two-sun planets. The planet orbits around both stars at the same time, which are close together. http://galileospendulum.org/2011/09/15/a-planet-with-two-sun...

The point the earth actually orbits around isn't actually in the middle of the sun. It is at the center of mass between us and the sun. A planet orbiting a dual star system will do the same thing, but the center of mass winds up being somewhere between the two stars.

Their link to human vs machine is interesting:

http://www.planethunters.org/science#human

I wonder how good their machine learning is? I'd love to throw some support vector machines or neural networks at this stuff!

A team at McGill made a "game" to find patterns in I think protein-folding or genome data, to leverage humans to identify patterns that their algorithms could not.

http://en.wikipedia.org/wiki/Phylo_(video_game)

Man, everything has to be quad core these days.

Overheating too, I guess.

I wonder if it's a planet such that the surrounding stars' lights prevent any other starlight from being seen, except for once every few thousand years.

One could only imagine a civilization that lives here and what would happen to society whenever the full night sky is occasionally seen...

I know that you're being fatuous but in case anyone doesn't get the reference, check out Nightfall by Asimov - it's a classic sci-fi short story.

http://www.uni.edu/morgans/astro/course/nightfall.pdf

The planet is a Neptune-like gas giant, so no civilization there. It probably has big moons though, and with all the suns around they might be warm enough to develop life. I haven't done the math, but I'm guessing that when the moon is on the night side of the planet relative to the nearby stars, the location of the distant stars would cycle in basically the same way Jupiter does in our sky. It moves slowly relative to the background stars, but still rises and sets every day.

I don't know the details and couldn't do the math if I did, but I'm guessing those stars at 1000AU can't be much brighter than our full moon is, maybe combined with the light pollution of a big city. I've lived outside of New York all my life, and my night sky still has stars in it. Not many, but enough to know they're there, and sometimes on a clear night I can see quite a few.

> I'm guessing those stars at 1000AU can't be much brighter than our full moon is

They may well be fainter. According to Wikipedia, the Sun is about 400,000 times as bright as the full moon, on average:

http://en.wikipedia.org/wiki/Apparent_magnitude

So if the Sun were 1,000 AU away from us instead of 1 AU, it would be a million times fainter (1,000 squared), or less bright than the full moon.

But there are two of them, so that would mean closer to the moon's brightness, right? If I'm following, the sun would be about 40% as bright ast he full moon at 1000AU, so with the second start being smaller than, it may be 60%?

(Mock my math if need be, I'm not really trying here. :)

Yes, the second star would add to the brightness and make things closer. Also, I didn't see in the article what the actual brightnesses of the two stars 1,000 AU away are relative to the brightness of the Sun. I just used the Sun because it was easy to look up the relative brightness of the Sun vs. the Moon. :-)

Edit: I see from looking at more recent posts that the two distant stars are thought to be spectral class G and M. The Sun is a G, so it should be about the same brightness as the brighter of the two.

The planet is a Neptune-like gas giant, so no civilization there.

The gas creatures beg to differ.

Brian Aldiss referenced a similar kind of system in his Helliconia science fiction trilogy, where a planet orbited a small sun, which in turn orbited a much larger sun in a much longer orbit.

Ironically, truth turns out to be stranger than fiction, seeing as this two binary systems are involved.

Reply on news.ycombinator.com