Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Featured B First Interstellar Asteroid Found

  1. Nov 2, 2017 #21

    OmCheeto

    User Avatar
    Gold Member

    In my universe, there are apparently only 30 seconds per minute; "seconds/year ≠ 365.2422×24×60×30"

    Final travel time from Vega-ish distance answer: 291,097 years

    I was scanning the Lyra constellation area on Google Earth the other day and found a "most likely" candidate: NGC 6745

    These galaxies did not merely interact gravitationally as they passed one another; they actually collided. [ref]​

    And per wiki;
    NGC 6745 (also known as UGC 11391) is an irregular galaxy about 206 million light-years (63.5 mega-parsecs) away in the constellation Lyra. It is actually a trio of galaxies in the process of colliding.

    The three galaxies have been colliding for hundreds of millions of years.

    The only problem with this is:

    NGC 6745: 1.9E+24 meters distance to Earth
    travel time: 2,398,637,430,644 years
    age of the universe: 13,820,000,000 years
    complicating factor: 173.6 (=travel time/age of universe)
    conclusion: either the universe is older than we think, or this object is not from NGC5745​

    So, are there closer candidate "colliding" galaxies?
    (google google google)

    Wow!

    WHAT IS THE CLOSEST GALAXY TO THE MILKY WAY?

    Closest Galaxy:
    At present, the closet known galaxy to the Milky Way is the Canis Major Dwarf Galaxy – aka. the Canis Major Overdensity. This stellar formation is about 42,000 light years from the galactic center, and a mere 25,000 light years from our Solar System. This puts it closer to us than the center of our own galaxy, which is 30,000 light years away from the Solar System.
    ...

    The Milky Way became the size it is now by eating up other galaxies like Canis Major, and it continues to do so today. And since stars from the Canis Major Dwarf Galaxy are technically already part of the Milky Way, it is by definition the nearest galaxy to us.

    At only 25,000 light years, and from my interpretation, that "the Milky way is currently colliding with the Canis Major dwarf galaxy", my guess is that we will never know the origin of this asteroid.

    Kind of reminds me of what Zed said in Men in Black; "We're not hosting an intergalactic kegger down here."

    Oh yes we are. The galaxies are all drunk, crashing into each other, and sending debris flying every which way.

    Other globular clusters that orbit the center of our Milky Way as a satellite – i.e. NGC 1851, NGC 1904, NGC 2298 and NGC 2808 – are thought to have been part of the Canis Major Dwarf Galaxy before its accretion.
     
  2. Nov 2, 2017 #22
    From supernovae or neutron star collisions million or even billion years before the asteroid has been formed. Maybe my question was not precise enough. What I want to know is this:

     
  3. Nov 2, 2017 #23
    What kind of measurements of the composition of the asteroid are possible? It would be very interesting to be able to get an idea if the ratios of heavy elements contained in it are consistent with what we find in other known asteroids.
     
  4. Nov 2, 2017 #24
    https://io9.gizmodo.com/astronomers-say-theyve-found-a-rogue-planet-with-no-su-1443571329
     
  5. Nov 2, 2017 #25

    stefan r

    User Avatar
    Gold Member

    An object can leave a star at painfully slow speed. Barely nudged out of orbit. As an object's orbit goes further it becomes easier and more likely that the object gets a nudge.

    The sun is moving at around 20 km/s relative to the average velocity of nearby stars. So the parent star could, for example, be moving at 5.999 km/s relative to average stars in our neighborhood and the asteroid could leave orbit at 0.001 km/s. Then we could add the velocities to get 26 km/hr. More likely there are some angles involved and the host parent star had a higher velocity relative to the sun's neighborhood and/or the object exited a little faster. No need for a high velocity exit.
     
  6. Nov 3, 2017 #26
    Obviously you could.
    Provided the massive object itself is in Oort cloud, but NOT part of it.
     
  7. Nov 3, 2017 #27

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Thanks for the replies. I find the quantitative velocity argument the only compelling one.

    • The composition argument (rocky vs. icy) is, at best, probabilistic. There exist a small number of tail-less comets.
    • The "easier to eject from an inner solar system" argument is not true kinematically (sqrt(2) * circular orbit speed will eject an object). It's easier only in the sense that the density of objects is higher.
    • As I said before, I would believe the directional argument if I had a statistical ensemble of observations, but not for a single event.
    • Getting to 26 km/s from the outer Solar System is hard. An interaction with Jupiter would do it, but the trajectory doesn't allow it. However, this is still puzzling when coupled with the direction argument. That means the object was almost at rest (~8 km/s, about 40% of the sun's velocity) with respect to the Local Standard of Rest, Not impossible, just puzzling.
     
  8. Nov 3, 2017 #28

    tony873004

    User Avatar
    Science Advisor
    Gold Member

    You don't need an ensemble of observations.
    You don't need to see it rain to know that if it does rain, a rain drop is more likely to hit the windshield than the side windows.
     
  9. Nov 3, 2017 #29

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    But if you have a single drop of water on your windshield, will you conclude that it must be rain? (As opposed to from a sprinkler, or a garden hose, or something else)
     
  10. Nov 3, 2017 #30

    tony873004

    User Avatar
    Science Advisor
    Gold Member

    If there was a single drop on my windshield, I would only conclude that the drop of water likely originated from outside the car. Sprinkler and garden hoses are external to the car. Water drops from them are also more likely to strike a moving car's windshield than its side windows.
     
  11. Nov 4, 2017 #31
    What is the local standard of rest?
    Is Sun, at 20 km/s, an unusually fast-moving star moving through stars whose speeds relative to each other/local standard of rest are slower than 20 km/s, an unusually slow-moving star sitting among stars which are moving at high speed at all directions but slightly more in one direction, or an average star?
     
  12. Nov 4, 2017 #32

    OmCheeto

    User Avatar
    Gold Member

    My guess is, that there isn't one. [see below]
    As might be common knowledge, Barnard's star is moving pretty fast relative to us.

    per wiki's entry on Barnard's Star;
    The radial velocity of Barnard's Star towards the Sun is measured from its blue shift to be 110 km/s. Combined with its proper motion, this gives a space velocity (actual velocity relative to the Sun) of 142.6 ± 0.2 km/s. Barnard's Star will make its closest approach to the Sun around AD 11,800, when it will approach to within about 3.75 light-years.

    I did some rough calculations, based on an image on that page:

    2017.11.04.high.relative.motion.stars.png

    , and came up with similar numbers for Barnard's star. So my confidence level is high that I've gotten the maths correct this time, and have included the relative velocities of other local stars:

    Code (Text):
    _closest__  rel vel
      AD   ly    km/s     star(s)
    30000  3.2     25     Proxima & Alpha Centauri
    12000  3.7    130     Barnard's star
    22000  4.7     82     Lalande 21185
    39000  3.0     76     Ross 248
    48000  3.5     27     Ross 128
    76000  5.2    106     Gliese 445
    Now, I just have to figure out where they are/were, and how Oort cloud "outer borders" are determined.

    ps. I'm really bad at maths, so everyone is welcome to jump in here. :angel:
     
  13. Nov 4, 2017 #33

    Bandersnatch

    User Avatar
    Science Advisor
    Gold Member

    LSR is an astronomical term for the mean motion of galactic matter in the solar neighbourhood. It's what you'd think about if you wanted to treat galaxies as composed of matter following idealised regular orbits, where all stars and gas at a given distance have mostly identical velocities governed by the dynamics of the entire galaxy, without whatever motion local interactions induce.
    It's what peculiar motions are measured against.

    Here's an example paper discussing recent attempts at measuring Sun's peculiar velocity w/r to the LSR, which may shine some light on what's being done and how:
    https://arxiv.org/abs/1411.3572
    edit:
    this lecture presentation is probably more accessible:
    http://astroweb.case.edu/ssm/ASTR421/lecture11.pdf
    /edit

    Regarding that, though...
    Why is this puzzling any more than any other number? I really don't get the argument here. Would 15 km/s be not puzzling? 1 km/s? A thousand?
     
    Last edited: Nov 4, 2017
  14. Nov 4, 2017 #34

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    What does one expect? I'd expect ejected bodies to be going a little faster (relative to the LSR) than the bodies they were ejected from. So maybe 30 km/s would be most probable. Now, 8 km/s is not an impossible number, but it is unusual - phase space considerations (an idealization - maybe even an oversimplification - to be sure) would suggest maybe 1 in 10 or so such objects would be going that fast or slower.

    So 1km/s would be more puzzling. Not impossible, but slower than ~99.9% of the expected objects. 15 km/s would be less puzzling - maybe a quarter of the objects would be that slow or slower. 1000 km/s would be very surprising, as it is faster than the galaxy's escape velocity.
     
  15. Nov 4, 2017 #35
    Vanadium explained why 1000 would be puzzling, and 1 likewise. Arguments which I thought of myself.
    250 would NOT be puzzling, because that happens to be the peculiar speed of Kapteyn´s star, for example.
    The lecture presentation gives velocity dispersion of 9 km/s for A stars, compared to 20 of Sun.
    Does this suggest that the asteroid is likely to be that of some A star, such as Sirius, Altair, Vega or Fomalhaut, and not of a G star, such as Rigil Kentaurus, tau Ceti, sigma Draconis or eta Cassiopeiae?
     
  16. Nov 5, 2017 #36
  17. Nov 6, 2017 #37

    Bandersnatch

    User Avatar
    Science Advisor
    Gold Member

    First, there seems to be a matter of language use: when you say puzzling, it implies that the cause is somehow unknown, perplexing, completely unexpected. As in, 'how in the world did it get to have that velocity?' kind of puzzling.
    But from your post it looks like what you meant is that it is just an outlier in some distribution. I was objecting to the former (all of those velocities can be produced by known interactions, hence they're not puzzling), not to the latter.

    But let's look at whether 8 km/s w/r to LSR is really an outlier. There are two things to take into consideration here:
    1.
    Ejection events are more likely to be caused by gradual (over many orbits) changes to orbital angular momentum of an asteroid whose orbit is coupled to some massive planet, rather than a one-off close encounter resulting in large delta V - simply because there will always be more asteroids affected at long range than in close encounters.
    Since the most likely ejection is by incremental boosting of orbital momentum, the velocities of ejecta should be clustered around the escape velocity, i.e. one would expect rogue asteroids to have velocities close to the peculiar velocity of their parent star, with similar distribution.

    2.
    Peculiar velocities of stars w/r to LSR are, by definition, directed every which way. An asteroid ejected with some velocity in a random direction w/r to its parent system will then have its velocity w/r to LSR be a nett result of the two.
    E.g. even in a fantasy scenario where all stars have peculiar velocities equal to 20 km/s, and all asteroids are ejected with 20 km/s over escape velocity, the expected LSR velocity of an asteroid would be anywhere in the range of [0, 40] km/s.
    That is, one doesn't even need a slow-moving star, nor slow ejection, to get a slow-moving rogue asteroid.

    This is the velocity distribution in the galactic plane of some 20 thousand stars in the solar neighbourhood:
    Sol Neighbors V distribution.PNG
    Where the first dark contour contains approx 50% of stars. Vx is radial, Vy is in the direction or galactic rotation. The triangle is LSR.
    (Taken from: https://arxiv.org/abs/0912.3262, fig 1. This earlier paper contains contours for plane-normal direction as well: https://arxiv.org/abs/0905.2980 - they're narrower, but similarly clustered around 0 km/s w/r to LSR)

    So, taking into account both points mentioned above, as well as the velocity distribution of nearby stars, I think the 8 km/s is not only within the expected range (i.e. not-puzzling), but also in the most likely range.

    The dispersion you mentioned is for plane-normal velocities only. It is not the nett peculiar velocity which you compare it with. The plane-normal velocity of the Sun is approx. +7 km/s.

    Same thing as with Vanadium's post, if by 'not puzzling' you mean 'likely', then 250 km/s would be unlikely, as velocities like that of Kapteyn's star are strong outliers.
     
  18. Nov 6, 2017 #38

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Yes, by "puzzling" I mean "an outlyer". An it's not a yes or no thing. A 1 in 2 outlyer doesn't bother me at all. A 1 in 10 bothers me some, 1 in 100 more, and so on.

    Yes, but in this fantasy I would expect the distribution to peak at ~30 km/s (20 and 20 in quadrature). 8 would be relatively unlikely, 4 even more so, and so on.
     
  19. Nov 6, 2017 #39

    Bandersnatch

    User Avatar
    Science Advisor
    Gold Member

    But that's a fantasy scenario, meant to illustrate a specific point. In reality, stellar velocities cluster around 0 km/s w/r to LSR, as shown on the graph above. Couple that with ejection velocities predominantly being expected to be close to escape velocity, and 8 km/s is within 1 sigma for the population. As is 1 km/s, or 0.
     
  20. Nov 6, 2017 #40

    stefan r

    User Avatar
    Gold Member

    Wikipedia says oxygen has a critical ionization velocity of 12 km/s. Atoms in interstellar space frequently have thermal velocities of several km/s. Silicon and iron have lower ionization energies and more atomic mass. Higher velocity objects will build up more surface charges. Ionizing radiation and plasma can levitate dust. The effect has been observed on the moon. Charged particles interact with plasma in a wider radius than the particle radius. Asteroids with velocities 30 km/s would build up a lot more charge.

    The milky way averages around 5 x 105 molecules/m3 of interstellar gas. That is not enough to drag down the velocity of planets or stars. Suppose we guestimate the asteroid to have mass 3 x 106 kg and 2 x 104 m2 cross section. If it moves at 104m/s it will hit 1014 molecules per second. Assuming inelastic collisions (probably wrong), 1.7 x 10-13 kg/s at 104 m/s should decelerate a 3 x 106 kg asteroid at 5.5 x 10-16 ms-2. In one billion years that becomes 17 m/s.

    A one micro gram piece of dust orbiting (or following) the asteroid 7500 meters (100x radius) would have gravitational acceleration F=G(m1m2)r-2= 6.674 x 10-11 x 3 x 105 x 1 x 10-6 x 1.8 x 10-8 = 3.6 x 10-19 newtons.
    The piece of dust could have diameter 10-4 m, cross sectional area 8 x 10-9 m2. The dust hits about 40 molecules per second or 6.6 x 10-26 kg/s. Traveling at 104 m/s (simplified) drag force would be around 6.6 x 10-22 newtons. The force of interstellar gas on an orbiting grain of dust is 500x less than the force of the asteroid's gravity. So the grain could stay in orbit. A comet's dust tail could have much higher drag force than the comet's body.
    If the dust tail has 1000x the total surface area then it could drop velocity from 10 km/s to 8 km/s in 120 million years. I am not sure if a dust trail would be detectable. It also would not need to be currently present.

    I have not seen any evidence that A/2017 U1 has a dusty regolith. Just throwing out the possibility that high velocity meteors could slow down.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted