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

Lost in the Milky Way scenario

  1. Nov 5, 2015 #1
    Let's say a spaceship from Earth (via a wormhole or something) suddenly found itself in a random place in the Milky Way galaxy. Assuming the crew possesses all of our current knowledge of the galaxy, and time was not a factor, would they be able to locate where our Sun is, and travel back to it?
     
  2. jcsd
  3. Nov 5, 2015 #2
    Yes, in fact, Voyager contains a map that tells any extraterrestrial that finds it where we came from. If you can detect pulsars near our Sun, you could use simple trig to figure out where we are.
     
  4. Nov 5, 2015 #3
    Our current knowledge of our galaxy is by no means complete though, albeit rapidly improving.
    Yes the crew might be able to identify a known pulsar or some other unique object and work things out from there.
    They may however find themselves in a region that we have no detailed information about at all, such as the side opposite of the galaxy from where the solar system is, from where easily identified markers cannot be seen.
    Then again, since the ship is able to leap through wormholes, they could always try other ones, and probably stand a good chance of getting somewhere recognisable eventually.
     
  5. Nov 5, 2015 #4

    Janus

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    If it was me, I probably do something like this. First locate the direction to the Center of the Galaxy. This should be fairly easy by looking for the Halo of globular clusters around it, or assuming you have the equipment to do so, finding Sagittarius A. By noting the apparent angular size of the aforementioned Halo, you can get a good estimate of your distance from the center. Now look for some and easily recognizable extra-galactic bodies like the Magellanic clouds or Andromeda, And note their direction and orientation with respect to the center of the galaxy. This will give you how many degrees around the galactic disk you are from home. This should be enough to get you close enough to the neighborhood of Sol to be able to get the rest of the way by other means such as the already mentioned pulsar map.
     
  6. Nov 5, 2015 #5
    You don´t need to find the Halo to note the direction. See the panorama:
    http://emanuelscirlet.com/tehnici/universe/milky.html [Broken]
    The direction to centre is shown simply by maximum brightness and dark rifts.
    But can you spot the halo here?
     
    Last edited by a moderator: May 7, 2017
  7. Nov 5, 2015 #6

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    I would start from the same place as Janus - aligning to extragalactic objects. From there I would go to an extraordinarily bright star, like Alnilam (Epsilon Orionis). That star would be really hard to miss - it's one of the very brightest in the galaxy and has a well-known spectrum. From there I should be able to spot Canopus. From Canopus, I should be able to spot Arcturus, from Arcturus Vega, from Vega Sirius, and from Sirius Sol.
     
    Last edited by a moderator: Nov 10, 2015
  8. Nov 5, 2015 #7
    After aligning to extragalactic objects, my next move would be try to spot objects which are inside Milky Way, but far from disc. Meaning globular clusters.
    From Sun and naked eye, there are about two - Omega Centauri and 47 Tucanae. And then the marginal ones like M22, M13.

    The proximity of 47 Tucanae to Small Magellanic Cloud gives a convenient pointer to solar neighbourhood!

    First, you need a spectroscope. Then you need to distinguish the spectrum of Alnilam from other bright stars, nearby or far away. Finally you need to be sure it has not changed in these lightyears.
    I would look for star clusters. Up to magnitude 6,0 from Sun there are about 6000 stars. Nebulae and star clusters are, I think, few tens:
    4 galaxies (2 Magellanic clouds, Andromeda and Triangulum)
    about 4 globular clusters (discussed above)
    a couple of gas nebulae
    And a few tens of resolved and unresolved open clusters.
    These should be easier to identify.
     
  9. Nov 5, 2015 #8
    Have a look at Messier and Caldwell. 109 objects each. Caldwell specifically targeted any deserving objects omitted by Messier, whether Messier did not bother to include them or they were south.
    My inclusion threshold: brighter than 6,0.
    So:
    M4 5,9 7200ly globular
    M6 4,2 1600ly open (Butterfly Cluster)
    M7 3,3 1000ly open (Ptolemy Cluster)
    M13 5,8 22000ly globular
    M22 5,1 10000ly globular
    M24 4,6 10000ly star cloud
    M25 4,6 2000 ly open
    M31 3,4 2500000ly galaxy (Andromeda Nebula)
    M33 5,7 2500000ly galaxy (Triangulum)
    M34 5,5 1500ly open
    M35 5,3 2800ly open
    M39 5,5 800ly open
    M41 4,5 2300ly open
    M42 4,0 1300ly nebula (Orion)
    M44 3,7 600ly open (Beehive)
    M45 1,6 400ly open (Pleiades)
    M47 4,2 1600ly open
    M48 5,5 1500ly open
    M50 5,9 3200ly open
    M52 5,0 5000ly open
    20 Messier objects in total
    C14 3,7 7500ly open (Double Cluster)
    C20 4 1600ly nebula (North America Nebula)
    C28 5,7 1300ly open
    C37 5,7 2000ly open
    C41 0,5 150ly open (Hyades)
    C50 4,2 5200ly open
    C64 4,1 5000ly open
    C71 5,8 3600ly open
    C76 2,6 6000ly open
    C80 3,7 17000ly globular (Omega Centauri)
    C82 5,2 4000ly open
    C85 2,5 500ly open
    C86 5,7 7500ly globular
    C89 5,4 3500ly open
    C91 3 1600ly Wishing Well Cluster
    C92 3 7500ly nebula (Carina)
    C93 5,4 13000ly globular
    C94 4,2 6000ly open (Jewel Box)
    C95 5,1 2500ly open
    C96 3,8 1300ly open
    C97 5,3 5500ly open
    C100 4,5 6000ly open
    C102 1,9 500ly open (Theta Carinae)
    C106 4,9 17000ly globular (47 Tucanae)
    24 Caldwell objects. So total 44 Messier and Caldwell objects out of 218 are naked eye objects - compared to about 6000 stars.
     
    Last edited: Nov 5, 2015
  10. Nov 5, 2015 #9

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    Hey, if the rules allow a starship, surely they allow a spectroscope. But I picked these stars because they are the brightest in the neighborhood. You don't need much more than "the really bright blue one over on the left" to identify them.


    The problem with distances to known objects is that we know these distances only to within a few percent. If your cluster is 5000 ly away, you know where yo are to maybe 200 ly.
     
  11. Nov 5, 2015 #10
    You do. Because there are a lot of bright blue stars in Milky Way. You do not know which one you are looking at. The three stars of Orion´s belt look much the same, so look at the belt from a different direction, and it´s hard to be sure of identification.
    Yes. Which is why the angular direction from Sun is much better known one.
    The line between Small Magellanic Cloud and 47 Tucanae nicely points near the Sun. From Sun, the angular distance between centres of SMC and 47 is under 3 degrees - 47 is in front of outskirts of SMC.
    47 is about 17 000 ly from Sun. Therefore, it is findable and identifiable in the general region of Magellanic Clouds for thousands of ly around Sun.
    The said angular distance of 3 degrees is about 6 times the size of Moon. Suppose you are able to measure the distance with error of moon´s size - whether is´s 5 or 7 times the size of Moon. That 30 minutes should be improvable with tools, even without magnification.
    Ascertaining that 47 Tucanae is displaced 30 minutes from its normal position would on the assumption that the distance is 17 000 ly tell you that you are 150 ly from Sun. Suppose your assumed 17 000 ly is in error by 1000 ly. This then means that you are making 10 ly error in your position - which you also would be making by 2 arc minute error in measuring the 3 degree distance between two hazy objects.
    So, once you are within a few hundred ly of Sun, you can start picking up nearer objects. Like open clusters.
    Making the detection threshold harsher, like magnitude 5,0, by my count leaves 10 Messier and 15 Caldwell objects, total 25.
    Of these 25, 1 is a star cloud (M24), 1 a galaxy (M31 Andromeda), 3 nebulae (Orion, North America, Carina), 2 globulars (C80 Omega Centauri, C106 47 Tucanae). Remaining 18 are open clusters.
    Noting the distances (and leaving the 3 nebulae in):
    M6 4,2 1600ly open (Butterfly Cluster)
    M7 3,3 1000ly open (Ptolemy Cluster)
    M25 4,6 2000 ly open
    M41 4,5 2300ly open
    M42 4,0 1300ly nebula (Orion)
    M44 3,7 600ly open (Beehive)
    M45 1,6 400ly open (Pleiades)
    M47 4,2 1600ly open
    8 Messier objects in total, 1 of them nebula, 400 to 2300 ly
    C14 3,7 7500ly open (Double Cluster)
    C20 4 1600ly nebula (North America Nebula)
    C41 0,5 150ly open (Hyades)
    C50 4,2 5200ly open
    C64 4,1 5000ly open
    C76 2,6 6000ly open
    C85 2,5 500ly open
    C91 3 1600ly Wishing Well Cluster
    C92 3 7500ly nebula (Carina)
    C94 4,2 6000ly open (Jewel Box)
    C96 3,8 1300ly open
    C100 4,5 6000ly open
    C102 1,9 500ly open (Theta Carinae)

    13 Caldwell objects. So total 21 Messier and Caldwell objects, of which 3 are nebulae and 18 open clusters. 5 of the 18 are within 1000 ly of Sun (M44, M45, C41, C85, C102), the remaining 13 are up to 7500 ly away (shared by Double Cluster and Carina Nebula)
    What is the angular distance between Carina Nebula and Theta Carinae cluster?
     
  12. Nov 7, 2015 #11
    Thanks for all the replies everyone! A very interesting discussion.
     
  13. Nov 7, 2015 #12

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    The pioneer gold album used pulsars. If its good enough for NASA its good enough for me. Since pulsars have a unique signature and are detectable at long distances, they should be adequate for locating just about any position relative to the sun within the MW galaxy
     
  14. Nov 7, 2015 #13
    I´m not particularly confident of either uniqueness or detectability.

    Can you spot the logic of using nebular alignments?
    A constellation is a chance alignment of objects which may be far from each other, but happen to be near a line going through Sun.
    We on Earth have problems measuring long distances precisely, but angles can be measured pretty well both by mapmakers on Earth and by the shipboard observers.
    The only Earth based information which is needed to navigate by the alignment of 47 Tucanae to Small Magellanic Cloud is a correct guess that 1) 47 Tucanae is physically far from SMC and 2) correctly identifying 47 Tucanae as the foreground object. After that, the line to Sun is easily found. If your guess as to the distance between Sun and 47 Tucanae is wrong then your mistake is only in terms of distance to Sun - you still get the correct direction. And since 47 Tucanae is at high galactic latitude, the correct position along the line through 47 Tucanae can be sought by reference to Milky Way plane.
    Then the next thing would be to find another, nearer constellation. And Carina looks like a good catch: background pointer Carina Nebula at 7500 ly, foreground pointer Theta Carinae at 500 ly.
     
  15. Nov 7, 2015 #14

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    See http://arxiv.org/abs/0905.4121, Using pulsars to define space-time coordinates, for discussion. It is claimed millisecond pulsars could be used to locate any object in the galaxy within a meter.
     
  16. Nov 7, 2015 #15
    Break off the recording for a while, and you have no idea which specific millisecond cycle you are observing.
    Unlike stars and star clusters, pulsars are anisotropic, for example. Travel a distance from Sun, and the pulsars known on Earth will not be observable (because you are out of the path of the beam) whereas pulsars undiscoverable from Earth will appear.
     
    Last edited by a moderator: May 7, 2017
  17. Nov 7, 2015 #16

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    Given the OP presumed the 'lost' craft originated from earth, it reasonable to assume the pilot would have access to at least the same pulsar data encoded onto the pioneer plaque. Given it includes data for 14 different pulsars spanning several thousand light years, it is reasonable to assume at least a few would be visible from almost any location. Since pulsar periods are unique the pilot need only find a few corresponding periodicity matches to plot his course back to earth. If unique pulsar identification is as unreliable as you suggest, then the whole NASA pulsar navigation idea was a colossal screw up. See http://46blyz.com/tag/pioneer-plaque/ for further discussion.
     
  18. Nov 7, 2015 #17
    Travel just a thousand lightyears away from Taurus, and Crab pulsar does not exist, because you are looking at times before the Crab supernova.
     
  19. Nov 7, 2015 #18

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    ???
     
  20. Nov 7, 2015 #19

    Janus

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    What he means is this: The supernova that formed the Crab Nebula and its pulsar was seen on Earth in 1054 AD or 961 years ago. So, if you were 1000 light years from the Earth in the opposite direction from the Crab nebula, the light from that supernova would not have arrived yet. In other words, neither the nebula nor the pulsar would be visible yet in the sky for an observer in this part of the galaxy, just the star that would eventually become the pulsar.
     
  21. Nov 7, 2015 #20

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    So how do you get there from earth before that light arrives - wormhole?
     
  22. Nov 7, 2015 #21

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    There are several problems with using pulsars. One is that they are directional - you might not see the ones you want. Another is that they are variable - apart from not even existing when you get far enough away, their periods change over time, and do so as a mix of a continual acceleration and sudden accelerations. Over thousands of years (which equals thousands of light years) their periods become an unreliable means of identification. Finally, to get good angular resolution requires a very large dish or set of dishes.

    So you probably need to find some way to get in the rough (1000 ly?) ballpark before this technque even starts to work. Also, my inclination would be not to use the angles at all - because it's unlikely I brought aboard some giant dishes. Instead I'd just look at periods: I'd measure the pulsar periods, compare with the known periods, and move in one direction until the chi-squared is minimized, Then I'd go on a perpendicular trajectory and minimize the Chi-squared again. And so on and so on until I was home.

    Which gave me an idea. Look at the LMC. Do you see the SN1987a progenitor or not? Move radially with respect to the LMC until you have found the spot where you see the supernova. You know the Earth is on a sphere centered on the LMC, but 28 light years closer. If you wish, you can improve your knowledge of the location of the sphere by repeating this at widely separated places. Now, there are maybe a dozen or two supernovae where we know the progenitor stars. Repeat this with them. One supernova gets you the surface of a sphere. Two gets you a circle. Three gives you two points (most likely only one is in the galactic plane). Four gives you a point and a chi-squared. After about a half dozen, you will know the position of Earth to within about a light-day.

    And my star-hopping plan worked in Celestia. It would have been easier with more steps, but it worked.
     
  23. Nov 8, 2015 #22
    My inclination would be to not use radio band. In visible light, decent angles can be measured by using eye lens.
    I see this as a bad approach. Can you, first, identify them?
    On average, you are not seeing any supernova most of time. As for supernova progenitors, they are relatively inconspicuous.
    Seeing a supernova gives you a sphere. Seeing an angle between two identified objects gives you a circle.
    You should be watching Magellanic Clouds anyway, because they are the brightest and most identifiable objects outside Milky Way disc. The alignment of 47 Tucanae with SMC gives a circle that intersects the disc at a large angle, so a good point around Sun. Then your error is mostly your measurement error of the 47 Tucanae angle.
    Spot SMC supernova, and you know you are exactly 28 ly too far - but this does not improve your angle measurement error along the sphere.
    So, the time is to find pointers nearer to you than the 17000 ly of 47 Tucanae. My suggestion is Carina. Any alternatives?
     
  24. Nov 8, 2015 #23

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    I firmly disagree with the implication the NASA choice of using quasars as locating beacons was naive. I believe I have provided credible references in tension with such a view. A similar courtesy would be appreciated.
     
  25. Nov 8, 2015 #24
    Yeah the OP postulates this at the beginning, although I don't think he is assuming that traveling through wormholes realistically is possible.
     
  26. Nov 8, 2015 #25

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2017 Award

    As I said, there are a dozen or two that are known. You don't have to say "Hmmm...is that a progenitor or not?" You can ask "Do I see this particular star or not?" And remember, every known progenitor was found in a survey. (For obvious reasons)
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook




Loading...