So I had a curious thought while re-reading about the IAU's definitions of planetary bodies.

A planet is:

• Is in orbit around the Sun,
• Has sufficient mass to assume hydrostatic equilibrium (a nearly round shape), and
• Has "cleared the neighborhood" around its orbit.

Now consider a celestial body that is unequivocally large enough to be a planet - Titan for instance. Picture a simple 3 body problem. Jupiter + Titan + Sun. But instead of Titan orbiting just Jupiter, have it orbiting both the Sun and Jupiter. There are actually a couple of orbits that allow for this, whether it be similar to a hohman transfer or something similar to the apollo era lunar transfer with a free return trajectory type orbit.

Given the prevalence of binary star systems out there I can see the possibility of planetary bodies developing and orbiting both. (Although, two stars would definitely starve the system of potential planetary building blocks)

Thoughts? What about other scenarios or special conditions that might not quite fit the IAU definition (ignore the "sun" requirement, I take that to mean star and it's piss poor language choice by the IAU)

Titan is an interesting celestial body and might possibly be the most habitable place outside of earth in the long term.

Although I love discussing astronomy (as a novice) I'm sorry for not contributing to your discussion. This is primarily a classification issue and I don't find it particularly compelling how Titan is labeled. Although I would like to know more about Titan.

Also, even though Pluto was downgraded to dwarf planet status, I was thrilled to see photos. It certainly surpassed expectations.

I'll take a stab at this
Titan is gravitationally bound to Saturn (not Jupiter as stated in your post) and therefore orbits Saturn. Saturn is gravitationally bound to the Sun and orbits the Sun. While Titan goes along for the ride, its orbit is clearly around Saturn and not the sun itself.
Now planets around binary stars, well I am not going even try to tackle that one

ER757 wrote:

I'll take a stab at this
Titan is gravitationally bound to Saturn (not Jupiter as stated in your post) and therefore orbits Saturn. Saturn is gravitationally bound to the Sun and orbits the Sun. While Titan goes along for the ride, its orbit is clearly around Saturn and not the sun itself.
Now planets around binary stars, well I am not going even try to tackle that one

I suppose I wasn't clear about what I was asking. I'm not asking if Titan itself is a moon or planet. It's certainly a satellite (to use a better term). What I'm wondering is if you had a satellite the size of Titan orbiting both a star and a planet, say Jupiter, or Saturn, or Earth, whichever is not important. So this quasi moon/planet would be orbiting both a star and a planet at the same time. What should it's classification be?

I would expect it to still be a moon as the other body of planet size would likely have a larger impact on its orbit than the sun. But I am also not an astronomer.

T18 wrote:

I would expect it to still be a moon as the other body of planet size would likely have a larger impact on its orbit than the sun. But I am also not an astronomer.

That's kind of what i'm leaning towards but the IAU guidelines don't necessarily make it crystal clear either. What if it's orbit to clear is between planet and star?

The 'moon' however is unlikely to have cleared its neighbourhood and it's ultimate fate is liable to be a fiery death in the star as such orbits are inherently unstable. I'd go with moon / stolen moon.

I suppose they'll work it out if and when they discover such a system.

trpmb6 wrote:

ER757 wrote:

I'll take a stab at this
Titan is gravitationally bound to Saturn (not Jupiter as stated in your post) and therefore orbits Saturn. Saturn is gravitationally bound to the Sun and orbits the Sun. While Titan goes along for the ride, its orbit is clearly around Saturn and not the sun itself.
Now planets around binary stars, well I am not going even try to tackle that one

I suppose I wasn't clear about what I was asking. I'm not asking if Titan itself is a moon or planet. It's certainly a satellite (to use a better term). What I'm wondering is if you had a satellite the size of Titan orbiting both a star and a planet, say Jupiter, or Saturn, or Earth, whichever is not important. So this quasi moon/planet would be orbiting both a star and a planet at the same time. What should it's classification be?

I think it would still be a moon - it's in a circular or elliptical orbit around the planet - the fact that the planet goes around the star doesn't really translate to the other body being considered as in orbit around the star. If not for the planet, the other body would either be drawn in towards the star and vaporized or flung into interstellar space. (or end up in an asteroid belt or Oort Cloud). Just my thoughts on the matter - where's Neil DeGrasse Tyson when we need him?

mbmbos wrote:

Titan is an interesting celestial body and might possibly be the most habitable place outside of earth in the long term.

Although I love discussing astronomy (as a novice) I'm sorry for not contributing to your discussion. This is primarily a classification issue and I don't find it particularly compelling how Titan is labeled. Although I would like to know more about Titan.

Also, even though Pluto was downgraded to dwarf planet status, I was thrilled to see photos. It certainly surpassed expectations.

I think Europa or Enceladeus are more likely candidates to host life in their sub-surface water oceans than Titan is. It's just so darned cold on Titan - molecules are just not very energetic at 200 degrees below zero. Hard to imagine anything that could drive metabolic processes at those temps

trpmb6 wrote:

Thoughts? What about other scenarios or special conditions that might not quite fit the IAU definition (ignore the "sun" requirement, I take that to mean star and it's piss poor language choice by the IAU)

It's a deliberate choice of words, as the IAU definition is only for for planets in the solar system.

Best regards
Thomas

ER757 wrote:

Now planets around binary stars, well I am not going even try to tackle that one

While it's very tricky (and boring admittevely) to unly crunch that in numbers, it's great fun to visualize it. I can recommend "Universe Sandbox", a program which lets you simulate pretty much everything from two body setups all the way to entire galaxies. It's great fun to fool around with it, for example create an accretion disk around a young star and watch what happens over time. Or give a large planet multiple close moons and watch the carnage unfold
The physics are very accurate and you need a very powerful computer to run it on, I have a 24 core dual Xeon system with 128GB RAM, and even that struggles when there are something like 50K fragments hurling around a star/planet. It even simulates climates on the planets, when you have multiple planets with atmospheres things really slow down as you need some serious computing power.

For binary star systems with planets, they're only really stable if you have certain configurations: The stars need to be quite far apar, at least 100AU, and one of the stars needs to be very massive, at least 5 times the mass of its companion. The planets will then form around the smaller star in rather close orbits.

That's my currently most stable binary system:







As you can see that system is HUGE, but the orbits of the planet around the smaller star are remarkably stable, only the four small inner planets have some eccentricity as well as the two outer gas giants. The seven planets in between are in orbital resonance with each other and have almost perfectly circular orbits (less than 0.02 eccentricity). They also have very stable climates, more stable than earth actually, around 3K seasonal variation.

Here's one of the planets about to experience a "solar solar eclipse"



And one currently "in between" the two stars, with only a very small stretch of night on its surface:



Ah forgot to mention, the one planet you see around the larger star is a gas giant that formed around the smaller star initially too, but was ejected from its orbit when it had an ancounter with another gas giant. One was captured by the gravity well of the larger star and eventually settled into that orbit, the other one reached escape velocity in the encounter and was ejected from the entire system. The same happened to many smaller planets as well, some were flung out with 500km/s+

Zeppi wrote:

For binary star systems with planets, they're only really stable if you have certain configurations: The stars need to be quite far apar, at least 100AU, and one of the stars needs to be very massive, at least 5 times the mass of its companion. The planets will then form around the smaller star in rather close orbits.

Out of curiosity... ever done a scenario like this:

System with fairly massive sun and a stable planetary system captures a rogue white dwarf into an highly elliptical orbit, ~90° to the systems plane with a perihelion close to its Roche limit?

I think the carnage would be fun to watch

Best regards
Thomas

Zeppi wrote:

""

This is awesome Zeppi! Thanks for sharing, definitely going to check it out. Took graduate level orbital mechanics in College and while I never have use to apply that knowledge it's fun to play with. Definitely will be sharing with a few of my friends. Just a little afraid about the computing requirements.

tommy1808 wrote:

Out of curiosity... ever done a scenario like this:

System with fairly massive sun and a stable planetary system captures a rogue white dwarf into an highly elliptical orbit, ~90° to the systems plane with a perihelion close to its Roche limit?

I think the carnage would be fun to watch

Best regards
Thomas

Thomas, isn't this one of those Planet X theories that some people believe there is a brown dwarf that is in orbit about our system at 90° to the ecliptic plane? Or something along those lines.

tommy1808 wrote:

Out of curiosity... ever done a scenario like this:

System with fairly massive sun and a stable planetary system captures a rogue white dwarf into an highly elliptical orbit, ~90° to the systems plane with a perihelion close to its Roche limit?

I have done a fairly similar setup actually, but instead of a white dwarf I used a pulsar. Also hat it come it with a very high inclination close to 90deg, but quite well clear of the roche limit. But even then, even on the first perihelion pass, the gravitational forces were so enormous that there was matter exchange, also the tidal friction was so strong that the pulsar slowed considerably with each pass, after about 20 orbits the whole thing ended in a supernova.
The planets of the system got stirred around quite a bit, but nothing overly dramatic until the supernova. That basically vaporized them leaving only a couple of iron/nickel cores behind

Well, I know where my paycheck will be going towards this month. Thanks for sharing Zeppi!

Zeppi wrote:

The planets of the system got stirred around quite a bit, but nothing overly dramatic until the supernova. That basically vaporized them leaving only a couple of iron/nickel cores behind

Awesome! What will kill you if you witness a supernova from 1 AU? Surprisingly it would appear to be the neutrinos, as that event produces such a high flux even that ghostly particle can kill you. And probably heat those core remnants quite a bit... as if they need heating D

Do you happen to have that simulation on hand a way that we can see it?

best regards
Thomas

Didn't want to start a new thread for this.

It is suspected that Kepler 1625b (a super giant gas planet orbiting at about 1AU) also has a "moon" that is the size of Neptune.

https://www.nationalgeographic.com/scie ... cts-space/

About 8,000 light-years away, a giant planet circles an aging star, marching once around its sun in a single Earth-year. But that planet, called Kepler 1625b, might not be traveling completely alone.

Scientists now suspect the planet’s skies are filled by an orbiting mega-moon, a stunningly large world the size of Neptune that may be the first moon spotted outside our solar system.

Should be noted, if this really does come out to be a moon, it shouldn't be surprising that it was the first type of exo-moon discovered. By the very nature of how they search for exo-planets and now exo-moons you will almost certainly find the biggest ones first. Whichever ones create the largest perturbations.

tommy1808 wrote:

Zeppi wrote:

Do you happen to have that simulation on hand a way that we can see it?

I don't think I have it saved, but just for the fun of it I can rerun a similar setup with a white dwarf when I get home towards the end of the week and capture some screenshots of the action.

Zeppi wrote:

tommy1808 wrote:

Zeppi wrote:

Do you happen to have that simulation on hand a way that we can see it?

I don't think I have it saved, but just for the fun of it I can rerun a similar setup with a white dwarf when I get home towards the end of the week and capture some screenshots of the action.

Awesome, thank you!

best regards
Thomas

One Man's Moon is another Man's Planet . . .

Free morning - carnage time!

Basic setup: Yellow main sequence star, 3.6Msun, 6.2 million km radius, 5 rocky inner planets, 5 gas giants.



And here comes the intruder! White carbon dwarf, 800 jupiter masses, 11km radius. inclination at 87deg, perihelion at 20 million km, just skimming the roche limit of the star.



First few orbits rather uneventful, some matter exchange and reduction of eccentricity, while throwing some of the inner planets into ever more increasing inclination:



The dwarf gets pretty hot on every perihelion pass, tidal heating at 1.37E+20W. Average temp 32.000K:



System slowly getting out of shape:



Matter exchange at perihelion increasing, the blue speck on the left is the dwarf:



Due to tidal friction the orbit is getting closer, perihelion at 15 million km well inside the roche limit. Serious matter exchange:



After 182 orbits the perihelion is starting to skim the outer corona of the main star, it has lost 16% of its mass by now, the dwarf has gained 31%:



The dwarf is about to reach its chandrasekhar limit on the next pass, while the main star's stability is also edging close to core collapse:



Timing win! Core collapse of the main star, followed shortly by the dwarf. Double supernova!



Let's go to the closest planet to watch it unfold. It's 1.8 earth masses, 4.28AU from the supernovae and has an iron/nickel core and silicate crust:



Pretty, no? This type of supernova takes around 19 days to reach max luminousity which will peak around 51 billion times the luminousity of our sun. So it's all smooth sailing for now, at least until the shockwave of particles arrives. Shades are certainly recommended on the planet
Shortly before it hits:



Then the particle bombardment begins, after just 30mins the planet looks like this:



And like this after one hour:



After three hours it's just a ball of ionized plasma being completely eroded and becoming part of the plasma cloud of the supernovae:



The luminousity of the supernovae is meanwhile so high that even the furthest gas giant at 91AU is glowing bright, and the particle cloud hasn't even quite reached it yet:



View from a distance of 300AU, the four white specs are the brightly glowing outer gas giants:



View one year later from a distance of 5 lightyears, the planets are all entirely gone and the plasma cloud has a radius of about 0.7 lightyears:

Zeppi, you win the internet today. Many thanks for this. And only fuel my curiosity more of Universe Sandbox.

Zeppi wrote:

Free morning - carnage time!

Awesome. Thank you!
Does the log say how much CPU time went into that?

best regards
Thomas

tommy1808 wrote:

Zeppi wrote:

Free morning - carnage time!

Awesome. Thank you!
Does the log say how much CPU time went into that?

best regards
Thomas

Also interested.

trpmb6 wrote:

Zeppi, you win the internet today. Many thanks for this. And only fuel my curiosity more of Universe Sandbox.

Yes indeed - thank you Zeppi. You've answered a question I have had for many years and never found a definitive answer to on TV shows, magazine articles or even on the interwebz. The question is what fate awaits a planet that orbits a star about the same distance a Earth orbits our sun which goes supernova. I figured it would likely be vaporized but until now never had confirmation.

Can you have a sub-moon, orbiting a moon, orbiting a planet, orbiting a star?

Dieuwer wrote:

Can you have a sub-moon, orbiting a moon, orbiting a planet, orbiting a star?

I saw this today, https://gizmodo.com/astronomers-wonder-can-moons-have-moons-1829639045

TLDR is that it may be possible under certain circumstances.... mainly being having a very large moon with a very large orbit around the main planet... and even then the moon-moon would need to be fairly small.

desertjets wrote:

Dieuwer wrote:

Can you have a sub-moon, orbiting a moon, orbiting a planet, orbiting a star?

I saw this today, https://gizmodo.com/astronomers-wonder-can-moons-have-moons-1829639045

TLDR is that it may be possible under certain circumstances.... mainly being having a very large moon with a very large orbit around the main planet... and even then the moon-moon would need to be fairly small.

This kind of hints at another question that perhaps the IAU should look at (or maybe they have). Is a moon really a moon if it hasn't reached hydrostatic equilibrium? In this case a lot of "moons" (I prefer satellite) would be disqualified. But in order for a moon to have a moon orbiting it the secondary body would almost certainly not have reached hydrostatic equilibrium. Most likely it would just be an asteroid that was captured.

I found this bit from the article enlightening. To me this provides another potential method for classifying planets. (Although you may have to consider large binary planets orbiting a local point too.. hmm..)

Astrophysicist Michele Bannister from the Queen’s University Belfast in Northern Ireland thought the question of whether a moon can its own moon was intriguing. She pointed out that there are already some very strange systems of celestial bodies beyond Neptune that challenge our definition of a “moon.” Charon doesn’t orbit Pluto, for example, but in fact both objects orbit some point located between them, with four other small moons also orbiting that point. Similarly, the 47171 Lempo system contains two minor planets orbited by a third, much smaller satellite.

Suffice to say, I think there are likely many many combinations of planetary systems with all sorts of interesting orbital mechanics out there.

Seems cheap, only $24.99.

http://universesandbox.com/

It's really a gimme for what it can do, yeah. A few years ago such sims were only accessible to labs at immense cost, but now with computing power as high as it is even for the home user it's no biggie any more. Of course it depends a lot on the settings too, you can run it on a high end laptop just fine as well, but the detail will be much less or the simulation speed much slower respectively than on a serious workstation.

tommy1808 wrote:

Does the log say how much CPU time went into that?

Nope, but the whole thing took around an hour on two Xeon 8160s and two Vega56s with the following settings:
No gravity substepping, max positional error 1m/s, max fragments 1M, roche lobe, tidal friction/heating, jeans escape, stellar wind, thermodynamics applied to all bodies and fragments.

Regarding the submoons, it all depends on scale really. A large system with a large gas giant that has a large moon in a wide orbit, that large moon can have smaller moons too. But those are not stable in the long term, the gravitational disturbance from the main planet is just too strong.
Same goes for binary planetary system with moons where all of them orbit a common barycenter, like pluto/charon. It works for a while but eventually the smaller moons will either collide with the primaries or get ejected.

Zeppi wrote:

Nope, but the whole thing took around an hour on two Xeon 8160s and two Vega56s with the following settings:

well, i guess i found a burn-in Test Software for systems beyond 32 Cores....

best regards
Thomas

tommy1808 wrote:

Zeppi wrote:

Nope, but the whole thing took around an hour on two Xeon 8160s and two Vega56s with the following settings:

well, i guess i found a burn-in Test Software for systems beyond 32 Cores....

best regards
Thomas

No kidding. Now I'm curious how my workstation will hold up.

trpmb6 wrote:

tommy1808 wrote:

Zeppi wrote:

Nope, but the whole thing took around an hour on two Xeon 8160s and two Vega56s with the following settings:

well, i guess i found a burn-in Test Software for systems beyond 32 Cores....

best regards
Thomas

No kidding. Now I'm curious how my workstation will hold up.

...and as if one of my customers reads this Forum, he just inquired a system config with 240 Cores.....

best regards
Thomas