PDA

View Full Version : Galaxy collision/merging questions



GKMasterson
2014-Aug-17, 12:23 AM
Hello,

I hope this is the right forum for these questions (and that I'm not completely stupid for asking them). I've been re-watching Cosmos and just finished the 3rd episode where it was mentioned that the Milky Way and Andromeda galaxies will collide or merge (whatever the proper term for it is) in a few billion years. I know that there will be few (if any) star collisions during that. However, my questions are:


What will happen to the supermassive black holes at the center of both galaxies and how might their movement through the other galaxy during collision impact local (to them at that point) stars and planets?
The simulation makes it look like both galaxies are spraying out loads and loads of stars. Will stars be ejected from each galaxy during this collision?
Can a star with a planetary exist by itself in interstellar space if it is ejected from a galaxy?


I'm not a scientist (though I love the subject) and I'm no good with any mathematics beyond algebra so if you know the answer, dumb it down for me, please. :)

Thanks!

-- G.K.

antoniseb
2014-Aug-17, 10:59 AM
...


What will happen to the supermassive black holes at the center of both galaxies and how might their movement through the other galaxy during collision impact local (to them at that point) stars and planets?
The simulation makes it look like both galaxies are spraying out loads and loads of stars. Will stars be ejected from each galaxy during this collision?
Can a star with a planetary exist by itself in interstellar space if it is ejected from a galaxy?

...

Let me answer #2 & #3 first. Yes, stars will be sprayed out, and Yes a star with planets can be in intergalactic space with no special harm to the planets. Do an image search for "Tadpole Galaxy" or "Antennae Galaxy" to see nice pictures of streams of stars getting flung out by a merger.

OK, Your question #1 was more interesting, since I haven't seen a write-up about that yet.

There are stars (red supergiants) which have bigger diameters than Sgr A* (the Milky Way's SMBH). The SMBH in M31 has about 25 times the diameter of ours. For the purposes of your question we'd like to look at a bigger diameter though, such that anything passing closer would get torn apart. In some cases, planetary systems passing even further way than that would have some or all of their planets separated from the host star. Anyway, find the disruption phenomenon you are interested in, and get the radius that make that happen and calculate the volume of the cylinder of that radius times the distance within the other galaxy each of the SMBHs will follow as the galaxies merge. Compare that volume to the average space per star in the galaxy and you'll get a sense of how many stars will be affected.

As an example (I don't have the numbers, so I'm making guesses).
- for disrupting stars, the radius of the M31 SMBH would be (let's say) 6 billion kilometers (250 million km for Sgr A*). And that the path through the Milky Way to the Merger may involve several loops, but lets call it 200,000 light years, which is 2x1018 kilometers. So the volume is Pi r-squared times the length, so I get about 2x1038 cubic kilometers as the disrupting stars volume of the bigger SMBH, which is smaller than one cubic light year (1039 cubic km). So, even that is unlikely to happen often. Now another issue is that these SMBHs will passe through regions of dense dust and gas, which should result in quasar like activity, and that might cause a bit of radiation that might sterilize planets at a much wider radius, but that's another set of guesses.

GKMasterson
2014-Aug-17, 08:08 PM
Anyway, find the disruption phenomenon you are interested in, and get the radius that make that happen and calculate the volume of the cylinder of that radius times the distance within the other galaxy each of the SMBHs will follow as the galaxies merge.

I'm glad I asked an interesting question. :) How do I get the radius of what will make the disruption happen and how do I calculate the cylinder of a radius? I tried looking this up on Google and a few math sites and books I have but I'm afraid that I'm really no great shakes with math. I kind of remember the πrē thing you're talking about -- it's got something to do with circles, right? Is it how you find the area of a circle? (Yes, I am this stupid when it comes to numbers. Sorry).

Also, some new questions: will the supermassive black holes in both galaxies merge themselves and become the size of the largest of them as if (or because) the largest one "eats" the others? Or would they average their size (which seems impossible to me but then what do I know)? Lastly, is this how all elliptical galaxies are formed? I can kind of wrap my head around how spiral galaxies and irregular galaxies might form

I got the clues of it after reading Contact and thinking about how a spiral galaxy would revolve around a supermassive black hole the way the Earth revolves around the Sun and how that revolution would result in the "gear teeth" effect to create the spiral arms because it's closer to what happens if you twirl around a wet paint brush. I figured that irregulars form when there isn't a supermassive black hole to cause rotation and that they're really just a bunch of big star forming regions where the average gravitational pull evens out over the galaxy just enough to keep the stars from wandering off too far but not enough to start the galaxy spinning at the kinds of speeds that would create the spiral arms.

Or at least, that's what I've reasoned out on my own since I can't parse the math in physics books well enough to see if I'm on the right path. :)

Thanks!

-- G.K.

Hornblower
2014-Aug-18, 12:06 AM
First things first. If I am not mistaken, Contact is fiction, and I would not count on fiction to get the mechanics of a galaxy right. Even the largest central black holes are only a small fraction of the total mass of the dense central core of a typical galaxy, and their gravitational contribution to the stellar orbits out in the disks is no different from that of the rest of the core, which is mostly ordinary stars. My rudimentary understanding is that the spiral arms in a disk galaxy are wave patterns that have nothing to do with the presence or absence of a black hole in the core. I will have to leave it to others to explain how these wave patterns got started in the first place.

The event horizon radius of a black hole is proportional to its mass. Thus when two black holes merge, the new radius will equal the sum of the radii of the original ones. Actually it will be slightly less because some of the total mass will be carried off as gravitational wave energy, if I am not mistaken. The two black holes in this merger are expected to spiral in and merge, first from losing orbital energy as they eject nearby stars, and then from gravitational radiation as they get really close.

I cannot answer your first question any better than antoniseb did, which he did to my satisfaction. With your acknowledged shortage of advanced math skills (no disgrace, we all have to start somewhere), I think you would be better off sitting down with someone who knows his stuff rather than having us try to teach you in a forum like this. Find an astronomy club in your area and try picking their brains.

neilzero
2014-Aug-18, 01:22 AM
As far as I know, those are excellent answers, but here are some thoughts: I have not seen any discussion that Andromeda may miss the Milky Way by a few thousand light years, but a glancing blow is possible, in which case the central SMBHs will be little affected. Not much will happen the first million years, except many of the stars in both galaxies will be perturbed to new proper motion, at least slightly different than their earlier path. After a few million years quite a few stars (possibly billions) will be out side the two galaxies, but with a vector that is less than escape velocity, so they are in more elliptical orbits, but will return to one of the galaxies in a billion years or so. If only the outer edges intersect, the average relative speed of both galaxies will be reduced, but it might be a trillion years before an other collision of these two galaxies occurs. If this is incorrect, please someone put me straight.

Hornblower
2014-Aug-18, 02:06 AM
Astronomer John Dubinski of the University of Toronto has done computer simulations of hypothetical glancing encounters such as you have described. He finds that the complex interactions of these extended multibody systems will cause large losses of orbital energy, with the result that a second pass will occur within about 3 billion years and result in a complete merger. Unlike the case of a two body problem, the calculations are exceedingly complex, and it appears that the results are counterintuitive to many of us. He wrote an article for Sky and Telescope eight years ago in which he presented a summary of his findings.

Jeff Root
2014-Aug-18, 02:22 AM
I'd be interested to read a description of how orbital energy can
be lost such that galaxies merge. I'd expect them to mostly pass
through each other and zoom apart again.

-- Jeff, in Minneapolis

Hornblower
2014-Aug-18, 02:41 AM
I'd be interested to read a description of how orbital energy can
be lost such that galaxies merge. I'd expect them to mostly pass
through each other and zoom apart again.

-- Jeff, in Minneapolis
Once again, the complex gravitational interactions result in ejection of some stars at high velocities, with the energy they acquired being at the expense of the remaining stars. I think the details are too complicated for a short answer, but it has to do with tidal distortion of the galaxies as they get close together.

profloater
2014-Aug-18, 03:44 AM
I read that the moon is still hot and soft inside due to tidal energy. So a huge amount of orbital energy must be lost to heat in all close encounters of massive objects with changing gravity forces.

Jeff Root
2014-Aug-18, 03:13 PM
Once again, the complex gravitational interactions result in
ejection of some stars at high velocities, with the energy
they acquired being at the expense of the remaining stars.
Oh, well, that's pretty straightforward. I suppose the ejected
stars don't even have to escape, but could just go into very
long orbits, like asteroids being thown out by Jupiter to form
the Oort cloud, simultaneously causing Jupiter to migrate
inward.



I think the details are too complicated for a short answer,
but it has to do with tidal distortion of the galaxies as they
get close together.
I can imagine the details are still unfathomable even if the
general mechanism is pretty simple.

-- Jeff, in Minneapolis

antoniseb
2014-Aug-18, 06:26 PM
Once again, the complex gravitational interactions result in ejection of some stars at high velocities, with the energy they acquired being at the expense of the remaining stars. I think the details are too complicated for a short answer, but it has to do with tidal distortion of the galaxies as they get close together.
Also keep in mind that it is not two objects orbiting, but rather that 80+ % of the mass will be in two huge overlapping oblate spheroids of dark matter, every particle of which is on its own trajectory, and will follow what ever evolving gravitational influences are driving it. This is a big reason that galaxies merge as fast as they do.

Jeff Root
2014-Aug-18, 08:11 PM
If dark matter doesn't interact with other dark matter, either
(as it appears not to, aside from gravitationally, of course),
then I'd think the dark matter would make mergers more
difficult and slower, not faster. The stronger the interaction,
the more rapidly the matter should virialize, I would expect.
Non-interacting dark matter should pass through everything,
and visible matter would tend to go along with it.

-- Jeff, in Minneapolis

antoniseb
2014-Aug-18, 08:29 PM
If dark matter doesn't interact with other dark matter, either
(as it appears not to, aside from gravitationally, of course),
then I'd think the dark matter would make mergers more
difficult and slower, not faster. ...
If it were just two concentrated objects, I'd agree with you. It isn't. It is two dark matter halos perhaps 250,000 light years across. The path of each individual particle is not quite the same as if the whole halo were one ballistic projectile. You can perhaps see some of that in the famous Bullet Cluster image, where the dark matter halos have clearly been deformed.