Yes, a planet around an old star should raise the odds of old, hence, metal-poor planets, but how much? Old stars have plenty of time to do other things to acquire planets, such as stealing them or creating them while passing through metal-rich nebulas. Can we directly measure the composition of this planet?
In the particular case I linked to, there are two planets around the same star. It is extremely unlikely to pick up multiple planets from floating rogues. As to metal rich nebulas, my understanding is that they aren’t that dense so wouldn’t do much. And at least if that had occurred, we’d likely see the star having higher metal content as well. In this case the iron content of the star is slightly under a tenth as common as it for the sun, and many other metals have more extreme ratios.
It is extremely unlikely to pick up multiple planets from floating rogues.
Do you have any support for that statement? (I’m not arguing, just curious how does one go about estimating the frequency of planetary capture given what I thought to be very little data.)
Planetary capture is low. For multiple capture events the probabilities are independent (this isn’t quite true, there are some complicating factors but this is very close to true), so the probability of capturing two is roughly the square of capturing a single one, which is estimated as around 3-6% under generous conditions(rogue planet numbers at least equal to the number of stars). So no more than around 1 in every 288 planets should have a double capture event, and the likely number is much lower than that. With around 700 known planetary systems, the chance that a given one is in this category is low, but that number isn’t as important since what needs to be asked is whether it is more likely that they’ve formed around the star or that multiple captures occurred. Note also if one assumes 3% rather than 6% then one gets around 1 in every 1100 planets which means we shouldn’t have even seen any examples. And if one thinks that the rogue planet percentage is lower than 1-1, all of this drops quite quickly.
A related issue is that if the first capture is in a somewhat stable orbit elliptical orbit, the introduction of another somewhat similar size body in orbit can destabilize the first making it get flung out of system, so once the second capture event occurs there’s a chance one will lose the first planet.
I don’t have a reference off-hand, but this is pretty standard logic to the point where multiple capture events as an explanation for strange systems is generally not often considered compared to deciding that it indicates our models are wrong.
ETA: Some of this logic is severely off. I just remembered that more recent estimates drastically increase the number of rogue planets floating around. See e.g. here so the 6% may in fact be an underestimate, in which case multiple capture planets becomes a much more plausible explanation. I don’t know then why it isn’t considered frequently other than the last issue of one capture possibly kicking another planet out of system but that shouldn’t change things by more than a small factor.
Hi Joshua, thanks for answering. Quick follow-up question: how come only “rogue” planets are mentioned in these arguments? (Well, it makes sense for studies about rogue planets, but it seems to happen even in discussions explicitly about captured planets, your comment being an example.) Can’t planets be “exchanged directly” between closely-passing stars? (I mean, without the exchanged planet spending a long time unbound to a solar system, in a sort of larger-scale analogue of close binaries exchanging envelope matter.)
I imagine close encounters are rare in general, but given the large number of binary and multiple-star systems that we seem to see everywhere, and my (admittedly vague) recollections of some rather tight clusters of stars with complicated/chaotic dynamics, it seems like it should be feasible (even common) for stars to exchange planets early (while they’re still part of a young cluster or complex multi-star system, and they interact closely) and then separate taking with them “stolen” planets (my understanding was that a significant fraction of stars in young clusters acquire high velocities and are “evaporated” away from their birth cluster, especially in “tight” clusters). Are the interaction time-frames incompatible with that kind of scenario or something?
how come only “rogue” planets are mentioned in these arguments? (Well, it makes sense for studies about rogue planets, but it seems to happen even in discussions explicitly about captured planets, your comment being an example.) Can’t planets be “exchanged directly” between closely-passing stars? (I mean, without the exchanged planet spending a long time unbound to a solar system, in a sort of larger-scale analogue of close binaries exchanging envelope matter.)
Yes, they can happen. But my understanding is that exchange isn’t a likely result of system interaction whereas losing a planet (that is a planet getting sent out of orbit) is much more likely a result than exchange. But this is pushing the limits of my knowledge base in this area.
Yes, a planet around an old star should raise the odds of old, hence, metal-poor planets, but how much? Old stars have plenty of time to do other things to acquire planets, such as stealing them or creating them while passing through metal-rich nebulas. Can we directly measure the composition of this planet?
In the particular case I linked to, there are two planets around the same star. It is extremely unlikely to pick up multiple planets from floating rogues. As to metal rich nebulas, my understanding is that they aren’t that dense so wouldn’t do much. And at least if that had occurred, we’d likely see the star having higher metal content as well. In this case the iron content of the star is slightly under a tenth as common as it for the sun, and many other metals have more extreme ratios.
Do you have any support for that statement? (I’m not arguing, just curious how does one go about estimating the frequency of planetary capture given what I thought to be very little data.)
Planetary capture is low. For multiple capture events the probabilities are independent (this isn’t quite true, there are some complicating factors but this is very close to true), so the probability of capturing two is roughly the square of capturing a single one, which is estimated as around 3-6% under generous conditions(rogue planet numbers at least equal to the number of stars). So no more than around 1 in every 288 planets should have a double capture event, and the likely number is much lower than that. With around 700 known planetary systems, the chance that a given one is in this category is low, but that number isn’t as important since what needs to be asked is whether it is more likely that they’ve formed around the star or that multiple captures occurred. Note also if one assumes 3% rather than 6% then one gets around 1 in every 1100 planets which means we shouldn’t have even seen any examples. And if one thinks that the rogue planet percentage is lower than 1-1, all of this drops quite quickly.
A related issue is that if the first capture is in a somewhat stable orbit elliptical orbit, the introduction of another somewhat similar size body in orbit can destabilize the first making it get flung out of system, so once the second capture event occurs there’s a chance one will lose the first planet.
I don’t have a reference off-hand, but this is pretty standard logic to the point where multiple capture events as an explanation for strange systems is generally not often considered compared to deciding that it indicates our models are wrong.
ETA: Some of this logic is severely off. I just remembered that more recent estimates drastically increase the number of rogue planets floating around. See e.g. here so the 6% may in fact be an underestimate, in which case multiple capture planets becomes a much more plausible explanation. I don’t know then why it isn’t considered frequently other than the last issue of one capture possibly kicking another planet out of system but that shouldn’t change things by more than a small factor.
Hi Joshua, thanks for answering. Quick follow-up question: how come only “rogue” planets are mentioned in these arguments? (Well, it makes sense for studies about rogue planets, but it seems to happen even in discussions explicitly about captured planets, your comment being an example.) Can’t planets be “exchanged directly” between closely-passing stars? (I mean, without the exchanged planet spending a long time unbound to a solar system, in a sort of larger-scale analogue of close binaries exchanging envelope matter.)
I imagine close encounters are rare in general, but given the large number of binary and multiple-star systems that we seem to see everywhere, and my (admittedly vague) recollections of some rather tight clusters of stars with complicated/chaotic dynamics, it seems like it should be feasible (even common) for stars to exchange planets early (while they’re still part of a young cluster or complex multi-star system, and they interact closely) and then separate taking with them “stolen” planets (my understanding was that a significant fraction of stars in young clusters acquire high velocities and are “evaporated” away from their birth cluster, especially in “tight” clusters). Are the interaction time-frames incompatible with that kind of scenario or something?
Yes, they can happen. But my understanding is that exchange isn’t a likely result of system interaction whereas losing a planet (that is a planet getting sent out of orbit) is much more likely a result than exchange. But this is pushing the limits of my knowledge base in this area.