Our moon would be a planet under new definition of planethood

[jwl]

When is the moon not a moon? If a new proposal for defining planethood is adopted, the moon could be considered a planet in its own right.

The meaning of the word “planet” has been a sore point since 2006, when, after a hurried argument, the International Astronomical Union (IAU) adopted a definition demanding that a planet be nearly round and be massive enough to have accreted or flung away other objects in its orbital neighbourhood.

Famously, this excludes Pluto, recently revealed by the New Horizons mission to be a stunningly complex world.



Even astronomers who are quite happy with Pluto’s dwarf-planet status have misgivings about the definition. Its criteria are vague, and it refers only to our solar system – so it excludes the thousands of worlds detected around other stars since the definition was written.

“I want a classification that applies both to the solar system and to exoplanets,” says Jean-Luc Margot at the University of California, Los Angeles.

Margot has now proposed a mathematical definition, which he outlined at a meeting of the Division of Planetary Sciences in National Harbor, Maryland, this week. “I wanted it to be rigorous, and easy to implement, so we don’t have to wait for interstellar travel to get high resolution images.”

He has worked out how massive a body must be to conform to the IAU’s orbit-clearing criterion. Above this critical mass, a planet’s gravity should be powerful enough to sling away or pull in any smaller bodies within a precisely-defined territory called the feeding zone. Given the mass of the star and the size of the planet’s orbit, you can work out the critical mass using a fairly simple formula.

Double trouble
That formula yields one peculiar consequence. Margot defines a pair of orbiting objects that are both above the critical mass as a double planet. So, New Scientist asked him at the meeting, what about Earth and the moon? With a quick glance at a graph, Margot confirmed that the moon is above the critical mass. So by his proposed definition, it’s a planet too.

“But we should be careful here,” he adds. “The IAU has not defined the term ‘satellite’. When they do, that will affect what they might decide about double planets versus satellites.” The next opportunity for the IAU to reopen the case would be their general assembly in Vienna in 2018.

As for the rest of the solar system, Margot’s criterion leaves a gulf between planets and dwarf planets. Pluto would retain its dwarf status, because it still has so much company in the Kuiper belt. The least planet-like planet, Mars, has more than 50 times the orbit-clearing mass; whereas the most dominant dwarf, Ceres, has only a few per cent of the mass required. It also means that all known exoplanets are indeed planets, except in the few cases where measurements aren’t yet good enough to tell. And conveniently, the proposal makes the iffy matter of “roundness” redundant – anything above orbit-clearing mass is so big that its gravity must pull it into a round shape.

“Of course it’s just a proposal,” says Margot. “I don’t know whether it will stick, whether people will love it, hate it or be indifferent.”

His suggestion certainly won’t satisfy those who think the IAU was wrong to require orbit clearance in the first place. “I’m sympathetic with what he’s trying to do,” says Richard Binzel of the Massachusetts Institute of Technology. “But to me, it’s about the body itself, not its location.” Binzel prefers an earlier suggestion that roundness be the main criterion – which would planetise many more objects.

https://www.newscientist.com/article/dn ... lanethood/
original paper:

A simple metric can be used to determine whether a planet or exoplanet can clear its orbital zone
during a characteristic time scale, such as the lifetime of the host star on the main sequence. This
criterion requires only estimates of star mass, planet mass, and orbital period, making it possible to
immediately classify 99% of all known exoplanets. All 8 planets and all classifiable exoplanets satisfy
the criterion. This metric may be useful in generalizing and simplifying the definition of a planet.

http://arxiv.org/pdf/1507.06300v4.pdf

[jwl]

Interestingly, while the graph in the paper clearly excludes the dwarf planets from planethood status (which is unsurprising, since the model is wholly based on clearing the orbit), other moons which would be part of "double planet" systems under this model include possibly ganymede, europa, callisto, io, and titan (they fall between the dotted lines).

[Purple]

Oh for crying out loud. Shouldn't these people be doing something useful instead of bickering over nomenclature? With all the size of the universe you'd think they'd find something better to spend research grants on. Like, oh I don't know, discovering new stuff to name?

[SpottedKitty]

Interesting. I've come across arguments in favour of calling the Earth-Moon system a double planet before, despite where the centre of rotation is — IIRC the Moon isn't big enough for the centre to be above the surface of the Earth.

[jwl]

Oh for crying out loud. Shouldn't these people be doing something useful instead of bickering over nomenclature? With all the size of the universe you'd think they'd find something better to spend research grants on. Like, oh I don't know, discovering new stuff to name?

He has found a formula which tells you if an exoplanet can clear its neighbourhood of smaller bodies within the lifetime of the star, based solely on its mass, the star's mass, and its semi-major axis of orbit. He also found out that there is a sizeable gap in this clearing mass criterion between planets and dwarf planets (if you exclude moons), there are no known exoplanetoids, and anything that can clear its orbit also fulfils the roundness criterion. That seems kind of useful to me, even without the whole definitions thing.

[Purple]

Basically what I take issue with is the need to at all classify things in that way if it violates common sense. I can see its use for determining what is a planet and what is a "dwarf planet" or how ever else you want to call it. But the whole satellite thing is just plain stupid. If X orbits Y than X is a satellite of Y. Plain and simple. No need to reinvent the wheel.

[Simon_Jester]

Except that it's very much possible we will one day encounter two large celestial bodies, close to each other in size, that orbit each OTHER. Individually either one is a planet. If the two of them are together, does it make one of them a moon?

Pluto and Charon are already pretty much like this. And the Earth's moon would be considered a planet if we found it orbiting one AU from a random star, if it weren't in the orbit of some other celestial body.

So no, this is not just quibbling.

[Purple]

Except that it's very much possible we will one day encounter two large celestial bodies, close to each other in size, that orbit each OTHER. Individually either one is a planet. If the two of them are together, does it make one of them a moon?

That's totally different.

Pluto and Charon are already pretty much like this. And the Earth's moon would be considered a planet if we found it orbiting one AU from a random star, if it weren't in the orbit of some other celestial body.

But it is.

[Jub]

That's totally different.

Not really. If we hadn't grown up thinking of the Moon as a moon but as a binary planet you wouldn't be offended by this. This is like the bitching over Pluto all over again.

But it is.

It won't always be. Eventually, the Earth will eject the Moon, it probably won't settle into a stable orbit within our solar system but if it did would it then suddenly become a planet in your eyes?

[Purple]

You don't understand what I am talking about. I'll try and abstract it away from any examples. Basically I think that there is absolutely nothing wrong with the notion that our classification of an object should depend not just on its inherent properties but also its relation with other objects around it.

Our current classification method is based on this principal and it has worked for us so far quite nicely. And if at any point we should reach a special case where it stops we can simply introduce a special case that covers it. And I find this approach to be far more sensible than uprooting everything and trying to think up a new comprehensive system from scratch. If for no other reason than because any system we think up is inevitably going to hit a new unexpected special case and if we restart every time that happens we'll be 5000 systems down the line by the time I reach 50.

---

So to return to the example of satellites and planets this is how I'd do it. If X orbits Y X is a satellite of Y. If Y orbits X Y is a satellite of X. If both orbit each other they are both "something something new term we made up for it".

[Jub]

You don't understand what I am talking about. I'll try and abstract it away from any examples. Basically I think that there is absolutely nothing wrong with the notion that our classification of an object should depend not just on its inherent properties but also its relation with other objects around it.

Our current classification method is based on this principal and it has worked for us so far quite nicely. And if at any point we should reach a special case where it stops we can simply introduce a special case that covers it. And I find this approach to be far more sensible than uprooting everything and trying to think up a new comprehensive system from scratch. If for no other reason than because any system we think up is inevitably going to hit a new unexpected special case and if we restart every time that happens we'll be 5000 systems down the line by the time I reach 50.

---

So to return to the example of satellites and planets this is how I'd do it. If X orbits Y X is a satellite of Y. If Y orbits X Y is a satellite of X. If both orbit each other they are both "something something new term we made up for it".

That's an abjectly horrid way of doing things. It all but ensures that the whole of science will eventually be nothing but special cases rather than a unified series of easily indexed and understood classifications that are updated as new evidence and was of thought come to the fore.

In this same vein, are you against changing the way animals are classified based on new genetic evidence? Or should something stay in the wrong taxonomic family simply because that is where it was first place? How about the larger scale move away from the Linnaean system into a cladistic system that better fits our current understanding of biology?

[Purple]

That's an abjectly horrid way of doing things. It all but ensures that the whole of science will eventually be nothing but special cases rather than a unified series of easily indexed and understood classifications that are updated as new evidence and was of thought come to the fore.

So what? The only context in which what you say would make sense is if you were designing these systems with computer data searches in mind. Than it would make sense to make the position of each object calculable based on its properties so as to minimize the time needed to reach its node in a search tree. But we are not doing that. At least I don't think we are. If we did than we'd use some sort of complex indexing system like the GRAU index for maximum efficiency.
For a system to be good enough for regular human use it really does not matter how it is composed as long as everyone knows and agrees to use the same system. It does not matter why X is called X just as long as everyone agrees to call it X.

[Jub]

So what? The only context in which what you say would make sense is if you were designing these systems with computer data searches in mind. [strike]Than[/strike] Then it would make sense to make the position of each object calculable based on its properties so as to minimize the time needed to reach its node in a search tree. But we are not doing that. At least I don't think we are. If we did [strike]than[/strike] then we'd use some sort of complex indexing system like the GRAU index for maximum efficiency. And for a system to be good enough for regular human use it really does not matter either way as long as everyone knows and agrees to use the same system.

Given that it's the people in the field making and voting on these changes, I would say that you're grossly unqualified to speak on this subject. To my way of thinking, this is just an idiot who's too stupid to spell the word never correctly whining about having to learn a new way of thinking.

[Purple]

Given that it's the people in the field making and voting on these changes, I would say that you're grossly unqualified to speak on this subject. To my way of thinking, this is just an idiot who's too stupid to spell the word never correctly whining about having to learn a new way of thinking.

Fact is that if you look at a lot of fields you'll see the same thing happen over and over again. Like back in the day indexing systems were all the rage in engineering. The idea being that you could take a part or assembly and design a derived index from its properties and that somehow the end user could decode information about the object based on its name. That's how we got stuff like the GRAU index. Turns out that approach does not work because people don't ever do that. So they fell out of use. And now everyone just calls things by various ear catching names that are easy to remember. People, especially smart ones like designing things that seem "efficient". But they really aren't. They are nice and neat and consistent and all that. But they don't actually improve the capacity of the human being to memorize and operate with the information any more than just using random words because the human mind is not an indexing machine. We remember concepts not search trees.

[Imperial528]

That's totally different.

Not really. If we hadn't grown up thinking of the Moon as a moon but as a binary planet you wouldn't be offended by this. This is like the bitching over Pluto all over again.

I'd say it is different. I'm not an astronomer or astrophysicist, but as a rule of thumb I consider orbital relationships based on the barycenter of the system. For a two body system the one with the barycenter inside of it is the planet, and the other is the moon. If the barycenter is external of both objects then barring extreme physical differences they are binary (dwarf) planets, as in the case of Pluto and Charon. Assuming of course that they are both objects that can be considered at most planets (no stars or other things).

Under this, the Moon is definitely a moon, while Charon is technically a paired dwarf planet but is traditionally a moon.

[Starglider]

I thought the definition of moon was that the center of gravity of the orbital system was within the larger body (the planet). This is true for the Earth-Moon system and untrue for the Pluto-Charon system, which is why the later is eligible to be a double (dwarf) planet and the former is not.

EDIT: sorry, missed that last post above

[Jub]

I thought the definition of moon was that the center of gravity of the orbital system was within the larger body (the planet). This is true for the Earth-Moon system and untrue for the Pluto-Charon system, which is why the later is eligible to be a double (dwarf) planet and the former is not.

I'd say it is different. I'm not an astronomer or astrophysicist, but as a rule of thumb I consider orbital relationships based on the barycenter of the system. For a two body system the one with the barycenter inside of it is the planet, and the other is the moon. If the barycenter is external of both objects then barring extreme physical differences they are binary (dwarf) planets, as in the case of Pluto and Charon. Assuming of course that they are both objects that can be considered at most planets (no stars or other things).

Under this, the Moon is definitely a moon, while Charon is technically a paired dwarf planet but is traditionally a moon.

That system works, but it doesn't account for things like a planet that gets swept into a gas giants orbit and ends up as a satellite. This is speculation, but given the distances at which we spot exoplanets, we may also have trouble figuring out the barycenter of a binary pair of planets and thus this definition fits better with what we can observe further from home. If that last bit isn't correct, then I think the best thing to do would be to designate between binary planets, planet-scale satellites, and non-planet scale satellites.

[Imperial528]

In the case of captured objects it's probably best to base the initial classification on the object's orbital characteristics at the time of discovery. Of course as with Pluto the classification can be refined as more data is incorporated but I don't see a reason to call a gas giant's captured planet a planet if it no longer has the orbital properties of one.

[SpottedKitty]

Eventually, the Earth will eject the Moon, it probably won't settle into a stable orbit within our solar system but if it did would it then suddenly become a planet in your eyes?

FWIW, this probably won't happen. The reason the Moon is receding from the Earth is that the Earth/Moon system is transferring angular momentum around within itself due to tidal drag — the Moon has long since stopped rotating relative to the Earth, and the Earth is rotating much more slowly than it was when it formed. This angular momentum has to go somewhere, and it goes into increasing the Earth-Moon distance.

The important thing to remember about this tidal drag is that it depends on difference — at some point in the far future, the Earth and Moon will be tidally locked to each other. No more relative rotation means no more tidal drag; the tidal bumps will still be there, of course, but they won't be moving any more. (There will still be a small effect from solar tides, and because the Moon's orbit isn't exactly circular or exactly on the ecliptic, but this will be much smaller than the rotation effect.)

There are some uncertainties in the final result. The mutual tidal locking will probably happen before Earth totally loses its grip on the Moon, and it might or might not happen before the Sun expands into its red giant phase, rendering the whole question moot.

Incidentally, this means that perhaps as soon as a few hundred million years from now, the Moon will be too far away to cause a total solar eclipse. Watch them before they're all gone.

[Jub]

FWIW, this probably won't happen. The reason the Moon is receding from the Earth is that the Earth/Moon system is transferring angular momentum around within itself due to tidal drag — the Moon has long since stopped rotating relative to the Earth, and the Earth is rotating much more slowly than it was when it formed. This angular momentum has to go somewhere, and it goes into increasing the Earth-Moon distance.

The important thing to remember about this tidal drag is that it depends on difference — at some point in the far future, the Earth and Moon will be tidally locked to each other. No more relative rotation means no more tidal drag; the tidal bumps will still be there, of course, but they won't be moving any more. (There will still be a small effect from solar tides, and because the Moon's orbit isn't exactly circular or exactly on the ecliptic, but this will be much smaller than the rotation effect.)

There are some uncertainties in the final result. The mutual tidal locking will probably happen before Earth totally loses its grip on the Moon, and it might or might not happen before the Sun expands into its red giant phase, rendering the whole question moot.

Incidentally, this means that perhaps as soon as a few hundred million years from now, the Moon will be too far away to cause a total solar eclipse. Watch them before they're all gone.

Most sources I've seen have said that the Moon will probably be ejected shortly before tidal lock happens, but it's not something that will happen in a time frame that makes it worth arguing over. Hell, with Earth spinning that slowly and the sun heating up we probably couldn't live here anyway. Still, it raises the question of what we'd call such an ejected satellite if we could track its ejection and stabilization within its own system.

[jwl]

This is speculation, but given the distances at which we spot exoplanets, we may also have trouble figuring out the barycenter of a binary pair of planets and thus this definition fits better with what we can observe further from home. If that last bit isn't correct, then I think the best thing to do would be to designate between binary planets, planet-scale satellites, and non-planet scale satellites.

I think this is exactly what he was getting at when he mentioned double planets in his paper: the main idea of the thing is creating a definition of planetary status which allows you to easily work out whether you are looking at an exoplanet or exoplanetiod. When writing his paper, however, I don't think that he had thought of that the moon-earth system might be thought of as a double planet under his definition; he only realised this when the new scientist brought it up during interview.

I think purple might have a valid point here: there's no need to jump down his throat. The fact is, people are not used to thinking of the moon, titan, and galiliean moons as part of a double planet system. So making sure the definition of a double planet is such that the moon isn't one might be something to take into consideration.

Another thing to consider in this is that the paper seems to show a gap in the critical mass ratio for the planets and planetiods, but the moons don't respect that gap. This may work as an argument to consider moons and planets/planetoids as fundamentally different objects.

[Jub]

I think this is exactly what he was getting at when he mentioned double planets in his paper: the main idea of the thing is creating a definition of planetary status which allows you to easily work out whether you are looking at an exoplanet or exoplanetiod. When writing his paper, however, I don't think that he had thought of that the moon-earth system might be thought of as a double planet under his definition; he only realised this when the new scientist brought it up during interview.

This is the Pluto situation all over again. If a new definition fits the data better across more cases we should roll with it instead of clinging to traditional ideas of what constitutes a planet or a moon.

I think purple might have a valid point here: there's no need to jump down his throat. The fact is, people are not used to thinking of the moon, titan, and galiliean moons as part of a double planet system. So making sure the definition of a double planet is such that the moon isn't one might be something to take into consideration.

That kind of rigid thought isn't what I'd expect from anybody with a science background. The idea that we could think of our moon as another world isn't so far fetched either.

Another thing to consider in this is that the paper seems to show a gap in the critical mass ratio for the planets and planetiods, but the moons don't respect that gap. This may work as an argument to consider moons and planets/planetoids as fundamentally different objects.

Yeah, but that again brings up the fact that we simply can't reach the level of detail with exo planets.

[Terralthra]

By the "barycenter inside the larger object" definition, we're in a binary system between a main sequence star (Sol) and a failed brown dwarf (Jupiter), since the Sun/Jupiter barycenter is outside Sol.

[Jub]

By the "barycenter inside the larger object" definition, we're in a binary system between a main sequence star (Sol) and a failed brown dwarf (Jupiter), since the Sun/Jupiter barycenter is outside Sol.

If people are still mad over Pluto and crying about the Moon as a planet I think that would cause some heads to explode.

[Gaidin]

Let them cry. As I understand this they're literally classifying based on number crunching the data? And as noted, no waiting your turn for those hard to get images.

How the hell does it get more scientific based on observations of the system in question?