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

Light Years Away

  1. Aug 17, 2012 #1
    If something is supposedly 1 billion light years away from us, how are we able to calculate this..? Wouldn't we have to send a light out and/or x/-ray out to that point and have it bounce back, calculating the time it takes to bounce back to us..? If so, it will take 1 billion years to get there and 1 billion years for the light and/or x/-ray to return, therefore rendering this process too long for us to even calculate.

    Having said that, if we can truely even see something that far away, how big must that object be compared to our sun..? From the estimations I've read about, that the closest star, Alpha Centuri, which is supposedly 4.3 light years away, only has the size of it as about the same as our sun. That seems to be grossly erroneous, imo..!!

    If we can see a star and it's that far away, my guestamation is that Alpha Centuri has to be X(umteen) times bigger than our sun.

    Your thoughts...
     
  2. jcsd
  3. Aug 17, 2012 #2
    There are various ways of calculating astronomical distances. For stars within a few parsecs, photos taken at opposite sides of the earth's orbit (six months apart) show how much the position of the star has shifted compared to the background stars much farther away.

    Farther than that, Cepheid variable stars are used. These are variable stars whose brightness varies in a regular cycle and the period of that cycle is closely related to the stars brightness. Knowing the absolute brightness of the star allows the astronomers to calculate its distance. Cepheid variables have been seen in nearby galaxies.

    Using these two methods, astronomers (Hubble) noticed an approximately linear relationship between distance and red shift. Beyond the distance where Cepheid variables can be observed, red shift is used.

    Recently a relationship between a type of supernova and its brightness was observed allowing the distance between those supernovas to be calculated by that method also. It was discovered the red shift versus distance was not as linear as believed, leading to the conclusion that the expansion of the universe is accelerating.
     
  4. Aug 17, 2012 #3

    tiny-tim

    User Avatar
    Science Advisor
    Homework Helper

    welcome to pf!

    hi buzzdiamond! welcome to pf! :smile:

    we see alpha centauri as a dot, not a disc

    we know its distance from using parallax

    we measure the difference in its position at two opposite points on the earth's orbit (ie, 6 months apart), and use a bit of trig :wink:

    for galaxies, we can't use parallax because the difference in position is too small

    instead we can use "standard candles" …

    particular types of star which always have exactly the same brightness​
     
  5. Aug 17, 2012 #4
    The size of a star isn't as important as its brightness. Generally the size of a star can only be determined with an interferometer. Perhaps the Hubble can resolve the disks of closer stars.

    How much more is the apparent brightness of the sun than Alpha Centauri? If you were to take the square root of that ratio and multiply it by the distance from the earth to the sun, you would have the distance to Alpha Centauri.
     
  6. Aug 17, 2012 #5
    Thanks Tiny, glad to be here. I am very perplexed by the universe and have some doubts about how things are being calculated and/or what we're really seeing. Feel free to chime in refuting my questions and/or thoughts, as I will be trying to tear apart what is currently accepted. Hope you don't mind hearing my position and theories..?
    If we're trying to get a grasp on the size of the universe, the surrounding planets and such, then yes, the size of the star would be important or at least very interesting to know.

    I'm willing to bet that it's not possible to differentiate between a star that's less bright or one that's farther away, as a brighter star farther away will look the same as a closer star that's less bright. Therefore, we can't accurately calculate the distance of a star. Correct..?
     
    Last edited: Aug 17, 2012
  7. Aug 18, 2012 #6

    tiny-tim

    User Avatar
    Science Advisor
    Homework Helper

    hi buzzdiamond! :smile:

    (just got up :zzz:)
    if you want to try to "tear apart what is currently accepted" here, you need to provide links to what you want to criticise, with a brief quotation of something you disagree with

    start with http://en.wikipedia.org/wiki/Parallax and http://en.wikipedia.org/wiki/Cosmic_distance_ladder and
    300px-Extragalactic_distance_ladder.JPG
    for most stars, that's exactly correct

    we can only tell the distance of most stars by finding another star in the same galaxy whose intrinsic brightness, or actual size, we do know
     
  8. Aug 18, 2012 #7
    Sounds good, will do. I'm headed out for the weekend so have a good one yourself.
     
  9. Aug 18, 2012 #8

    mfb

    User Avatar
    2017 Award

    Staff: Mentor

    If you have a method to find the size of the object (with a model, or with a comparison to other objects with known size), there is a direct way to measure its distance: The total (thermal) light emission of the object just depends on the surface and its temperature. You can measure the temperature with spectroscopy, and compare the total power output with the intensity here on earth to calculate the distance.

    If you have a double-star system, it is even better, as you get an additional constraint based on the orbital period and the angle between the position of the stars.

    And for stars nearby, the parallax is useful, too.


    By the way: A star with the diameter of sun in a distance of 4 light years appears with a visual angle of ~10 milliarcseconds, the best resolution of VLT is ~1 milliarcsecond. In theory, it should be possible to get a two-dimensional image of a star.
     
  10. Aug 18, 2012 #9

    phinds

    User Avatar
    Gold Member

    You need to study the concept of red shift. In the meanwhile, don't put actually money on that bet.
     
  11. Aug 18, 2012 #10

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Just by looking at it, or even measuring its brightness? No. But astronomers do more than that.

    Incorrect. As many people have already explained, there are various techniques for measuring distances in astronomy, each of which is applicable out to different maximum range. One that hasn't been mentioned (EDIT: mfb mentioned it in post #8) is one that can be done for individual stars, even if they are too far away for a parallax to be determined (although I should mention that the Gaia mission, a European satellite scheduled for launch next year, will be able to measure parallax angles as small as 0.000000006 degrees, allowing us to compute distances for stars all the way out to the edge of the Galactic disc). But I digress. As I was saying, even if a star is too far away to measure its parallax currently, another technique involves measuring the spectrum of that star. The thing you have to realize is that we understand a lot about the physics of stars, especially during the longest portion of their lifetime, when they are happily fusing hydrogen into helium in their cores. We call this portion of a stellar lifetime the "Main Sequence", because if you make a graph of luminosity vs. surface temperature for stars in this portion of their lifetime, they will all lie along a line or "sequence" on the diagram. In other words, there is a well defined relation between luminosity and surface temperature for stars when they are in this hydrogen-fusing portion of their lifetime. I should define some terms. The diagram I mentioned above is called a Hertzsprung-Russell or H-R diagram. The luminosity of a star is its power output: how much light energy it outputs every second. So you can think of luminosity as a measure of the intrinsic brightness of a star (how much brighter or dimmer than other stars it would look if it were right next to them) as opposed to the apparent brightness, which is how bright that star appears to us. As you've correctly pointed out, the apparent brightness of a star depends not only on luminosity, but also on distance.

    What if you could figure out how luminous a star was? I.e. what if you could figure out its intrinsic brightness? Well then, you could determine its distance by comparing the luminosity to the apparent brightness. This works because of the inverse-square law for dimming: the amount of light you receive from an object varies inversely with the square of the distance to that object. So if you take a given object, and double the distance to it, you'll receive 1/4 the light, and if you triple the distance, then the brightness will go down by a factor of 9, etc. So, by comparing the luminosity to the apparent brightness, you can determine distance.

    How do you get the luminosity? From the Main Sequence: as I mentioned before, stars on the Main Sequence have a well-defined relationship between their surface temperature and their luminosity. So if you can determine the surface temperature of a star, you can figure out how luminous it is. How do you determine the surface temperature of a star? Spectroscopy. We categorize stars by spectral type (which is determined from the measured properties of their spectra), and the spectral type depends on surface temperature. Roughly speaking, stars at different temperatures will be different colours, meaning that their emission will peak at different wavelengths. The hottest stars are blue or blueish-white, and then we go down a sequence to white, yellow, orange, and red. It's more than that: stars of different spectral types will have different absorption lines in their spectra, because the chemical compositions of stellar atmospheres vary with temperature. Anyway, the Main Sequence tells us that this sequence in spectral type/colour is also a sequence in luminosity: the hot blue stars are much more luminous than the cool red stars. If you can measure the spectrum of a star accurately enough to determine its spectral type, you can determine its luminosity, and hence the distance to it: This technique is known as Spectroscopic Parallax: http://en.wikipedia.org/wiki/Spectroscopic_parallax

    (The "parallax" part is a misnomer.) The only limit to this technique is that you need a lot of light to get an accurate spectrum, and at about 10,000 parsecs, things start becoming too faint for this method to be useful.

    So: Q. can you tell the difference between a close and moderately bright star and a distant and extremely luminous star? A. Just by looking at them, no. By actually analyzing the light from them scientifically? Yes.
     
    Last edited: Aug 18, 2012
  12. Aug 18, 2012 #11

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Phinds, can you give some indication of how exactly you think redshift would be applicable to measuring the distances to individual stars?
     
  13. Aug 18, 2012 #12
    Cepheid, great response..!! My argument.to that.would go back to what skeptic said, that size isn't important. If you have a large star with the same surface temperature as a small star, it will be more luminus than the smaller star. Therefore, it will seem closer, but in actuality, it may not be. Having said this, how accurate are we when distances are being given for stars and/or the size and dimensions of galaxies?
     
  14. Aug 18, 2012 #13

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Before I answer your question, let me briefly describe spectral types. We assign letters to spectral types, and these are O B A F G K and M. O-type stars are the hottest and bluest (surface temperatures of 30,000 to 50,000 kelvins), and M-type stars are the coolest and reddest (surface temps of 3000 K). Our sun is a G-type star (yellowish-white and surface temp of 6000 kelvins). For more info, just Google stellar spectral types.

    On to your question:

    There are many stellar properties that vary in a progression across the main sequence. I already mentioned surface temperature and luminosity. Another is stellar radius. Where a star lies on the main sequence entirely determines its radius. So, two main sequence stars of the same spectral type cannot have significantly different radii. Take M-type main sequence stars as an example. These are often called M-dwarfs or red dwarfs. What about a red giant? It's cool enough for it to be red. In fact it has the same spectral type (it's an M-giant). However, it is significantly larger and therefore it will be significantly more luminous. So much so, that it won't lie on the main sequence of the HR diagram. Indeed, the red giant stage of stellar evolution is the stage that many stars enter after their main sequence lifetime is over. (They have fused all of the hydrogen in their cores, and the absence of an internal energy source causes them to evolve off the main sequence). Check out this HR diagram to see where stars in different phases of stellar evolution end up: http://en.m.wikipedia.org/wiki/File:HRDiagram.png

    The method of spectroscopic parallax only applies to main sequence stars, which have well defined relations amongst their various stellar properties. Some other method of distance determination would have to be used for a more evolved star like a giant.

    You might be wondering why the placement of a star on the main sequence determines so many of its properties uniquely. It's because the fundamental stellar parameter that determines everything else is mass. Where a star will end up on the main sequence is determined by the mass that it has when it forms. A star forms from a cloud of interstellar gas that collapses under its own gravity. But as it collapses, it heats up. This heat (especially after fusion ignites in the core), creates an outward pressure that fights against the inward force of gravity. A star is said to have formed when a balance is achieved between these two forces, and the star is stable (a condition called hydrostatic equilibrium). The more mass that is present, the higher the internal temperature that will be reached before equilibrium. The higher the core temp, the higher the rate of nuclear fusion, and the more luminous the star, and the hotter its surface temp will be. And of course the radius of the star at hydrostatic equilibrium is determined by its mass as well. In this sense, the main sequence can really be thought of as a sequence from high-mass to low-mass stars.
     
    Last edited: Aug 18, 2012
  15. Aug 18, 2012 #14

    mfb

    User Avatar
    2017 Award

    Staff: Mentor

    Well, you do not have this in the main sequence.

    It depends on the star. For stars nearby, parallax measurements can give an uncertainty of less than 1%. For other methods (mainly outside the galaxy), the uncertainty is a bit larger (something like ~10%, but depending on the method), you can find some numbers at wikipedia, for example.

    If you have a value for the distance of a galaxy, the size is easy to measure.
     
  16. Aug 18, 2012 #15

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Your mention of other galaxies here makes me think you havent understood what others have said in this thread. Every distance measurement technique has a maximum distance out to which it works. The method I described is only relevant for distance measurement of stars within our own Galaxy and has no relevance to extragalactic distance measurements. For all but the most nearby galaxies, we can't even resolve individual stars in them. For these, we have to use "standard candles" like Cepheid variables or Type Ia supernovae. I think other people have described these methods for you.
     
  17. Aug 20, 2012 #16
    What im getting at is, because measuring distance of so vast a scale is physically impossible, are we really outside our galaxy when we're supposedly seeing other stars and/or galaxies?
     
  18. Aug 20, 2012 #17

    tiny-tim

    User Avatar
    Science Advisor
    Homework Helper

    hi buzzdiamond! :smile:

    look at all the galaxies in the zoomable "deep field image" taken by the hubble telescope between December 18 and 28, 1995


    anyway, even the andromeda galaxy is outside our galaxy, and that's close enough to measure its distance by a number of means
     
  19. Aug 20, 2012 #18
    OK Tim, thanks for the link. But, I've checked some images out and deduced that the Hubble space images are not real.

    They show pictures of galaxies billions of light years away, outside of our galaxy, but there are no pictures that focus on a star in our galaxy. Why is that..? If Hubble can supposedly show us a beautiful color photo of some distant galaxy, bazillions of light years outside of our galaxy, why can't it produce a photo of a star in our galaxy to appear as close as the photo's we have of our sun?

    Can someone please explain to me why the hubble pictures of the planets in our solar system http://hubblesite.org/gallery/album/solar_system/ don't show any stars in the background..? The pictures of the planets really look fake.

    Here's a link http://www.nasa.gov/mission_pages/LRO/news/apollo-sites.html showing pictures of the moon, with so-called lunar rover tracks. Would someone please estimate the distance of those tracks for me..?

    OK people, have at it...
     
  20. Aug 20, 2012 #19

    Chronos

    User Avatar
    Science Advisor
    Gold Member

    Buzz, you've deduced your way into the twilight zone of conspiracy theories. Planets are vastly brighter than background stars, which also explains the lack of stars in pictures taken on the moon by astronauts.
     
  21. Aug 20, 2012 #20
    I'm sorry Chronos, but Mars does not give off light or is burning at extreme temperatures. Planets reflect light and are cold. Therefore, Mars and the other planets in our solar system are not brighter than stars..!!

    Nice try. Do you wish to comment on my other two paragraphs..?

    Looking for others input. Thanks.
     
  22. Aug 20, 2012 #21

    Drakkith

    User Avatar
    Staff Emeritus
    Science Advisor

    Because the diameter of a star is so much smaller than a galaxy we run into the problem of not being able to get enough resolution to see the full extent of the star. When viewed in a picture a star is referred to as a "point source" of light. To understand what this means you need to learn a few things about optics.

    First, when light is focused down to a point, called an airy disc, that point has a finite size that directly depends on the diameter of your optical aperture and the wavelength of the light. Small telescopes cannot focus the light onto as small a point as a larger telescope if the focal length is the same for both scopes. When we increase the focal length of a telescope we increase the magnification of the image. When we do this the airy disc itself also gets larger, which keeps us from just zooming in until we can see detail on the star. Our "angular resolution" simply isn't high enough to see almost all stars. When the size of the airy disk is much larger than the apparent diameter of the object, that object is referred to as a "point source" because you can treat it in almost all aspects as a point "infinitesimal in size" that emits light. (These don't really exist, but at a certain point the difference between a "true" point source and a real object is small enough that it simply doesn't matter)

    Now, stars are VERY small compared to galaxies. In optical terms we refer to how "big" something looks as it's angular diameter. Closer objects look bigger than objects that are further away, such as watching as a car shrinks in apparent size as it moves away from you.

    So a star, being so very far away compared to its physical size has a very small angular diameter. Stars further away look even smaller than closer ones do. They are generally so small that we actually don't have telescopes capable of seeing detail on any stars but a very select few, such a Betelgeuse. (Which Hubble has actually taken a picture of, because the star is close enough and large enough to be within the Hubble's capability. See here: http://en.wikipedia.org/wiki/File:Betelgeuse_star_%28Hubble%29.jpg [Broken])

    Galaxies, especially closer ones, are so big that we can see plenty of detail in them even though we cannot see individual stars in our own galaxy. Realize that a galaxy is about 100,000 LIGHT YEARS in diameter, while a star is only around 200,000 - 1 billion kilometers across, which is FAR less than even 1% of one light year. (The sun is about 650,000 km in diameter for comparison. The largest stars are about 1,000 times larger)

    Stars are very far away, so the intensity of the light has dropped off substantially by the time it gets here. Put simply, the amount of light reflecting off a planet into a camera is FAR more than any single star. Because of this the exposure time has to be short enough to keep the image of the planet from saturating the sensor, which means that there isn't enough light entering from the background stars in that time to register in the image.

    If we take the planets and put them at the same distance that the stars are at then of course they aren't brighter than the star for the reasons you stated. However, stars are light-years away and the apparent brightness of an object falls off with the inverse square of the distance.
     
    Last edited by a moderator: May 6, 2017
  23. Aug 21, 2012 #22

    cepheid

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    LOL! Good on you for your powers of deduction! Too bad they have no basis whatsoever and demonstrate a clear misunderstanding of basic physics. :rolleyes:

    A telescope has a certain limit to its ability to resolve two distinct objects as being separate. Suppose two objects are very close together on the sky. This does not necessarily mean that they are close to each other physically in space, it just means that the direction of your line of sight to one object is not very different from the direction of your line of sight to the other object. In other words, there is a very small angle between these two lines of sight. This is how we measure the separation of objects on the sky: by the their angular separation. If two objects have 0 angular separation (i.e. the line of sight to one is the same as to the other), then they'll appear superimposed on top of each other. Now, as a fundamental limitation of how telescopes work (imposed by the laws of physics), every point of light in the scene being imaged is not mapped to a perfect point of light in the image. Instead, the light from that point is smeared out a little over a finite area. The light is spread out into a circular disc, because of a phenomenon called diffraction. Now, the size of this diffraction disc, in other words, the angle over which this light gets smeared out, depends on two things: 1. the diameter of your telescope, and 2. the wavelength of the light you're observing. The bigger the telescope diameter, the smaller the angular size of this diffraction disc. You want this disc to be as small as possible (i.e. the light from a point source should not get spread out too much). Think about it: this diffraction disc places a fundamental limit on your ability to resolve two objects as being separate. If the angular separation of those two objects is smaller than the angular size of the diffraction disc of each one, then the diffraction discs of those two sources will overlap. In other words, the light from one will overlap with the light from the other, in your image, and you'll not be able to tell that there are two separate objects there. So your ability to resolve fine detail is limited by this.

    The diffraction-limited angular resolution of the Hubble space telescope is 0.05 arcseconds. There are 60 arcminutes in a degree (of angle), and 60 arcseconds in an arcminute. That means that 1 arcsecond = 1/3600 of a degree. So Hubble can see two distinct objects as being distinct, even if they are separated on the sky by an angle of less than 0.000014 degrees (I just converted the 0.05 arcseconds to degrees).

    Now, we should ask ourselves, what's the apparent (angular) size of even the closest star? In other words, by what angle does the line of sight to one end of the object differ from the line of sight to the other end of it? What angle does it span, or how much of my field of view does it take up? To measure the angular size of something, you just take its physical size, and divide that by the distance to it. (This gives you the angle in radians, which you can then convert to degrees).

    The closest star is Alpha Centauri at a distance of 4.366 light years, and having a radius of 1.227 times the radius of the sun, or about 853,000 km. When I divide the second number by the first, I get an angular size for this star of about 0.00000119195 degrees or about 0.004 arcseconds. (EDIT: multiply these numbers by 2 since I used the radius of the star and not the diameter) That is smaller than the size of Hubble's diffraction disc, by a factor of 10 (EDIT: a factor of 5, actually). So the image of this star (and any other star) just looks like whatever the shape of the diffraction disc is. All of the light from this star gets smeared out into an area much larger than the actual size of the star's disc itself. So it is not possible to resolve any of the details of the structure of the star itself.

    Now let's compare that to trying to image a galaxy. I think that an image of this type:

    http://hubblesite.org/newscenter/archive/releases/2006/10/image/a/format/large_web/

    is the kind that is troubling you so much. It's an image of the Pinwheel galaxy, also known as M101. It has a diameter of something like 170,000 light years, and its distance is given as 21 million light years away. When I divide the physical size by the distance to get the angular size, I get a result of 0.46 degrees or 1670 arcseconds. The thing is 1670 arcseconds across, and the Hubble can resolve details even if they are as closely spaced as 0.05 arcseconds. So, Hubble has more than enough angular resolution to make out fine details of the structure of this galaxy.

    As you can see, it's the angular size of an object that determines whether a telescope will be able to produce a resolved and detailed image of it, and since stars are all so small (relative to how far away they are), their angular sizes are all so small that they will just appear as points of light (i.e. their images will be diffraction disks or worse) in even the best telescopes. In contrast, galaxies are much larger (relative to how far away they are) causing them to have larger angular sizes, large enough that we can resolve them as being extended objects.



    Chronos's explanation is correct, that a planet, (which is really close by, and therefore reflects a lot of sunlight towards us), appears much much brighter than a star, which although it is much more luminous, appears much fainter due to its extreme distance away from us. In fact, all you have to do is go outside and look up at the night sky to confirm for yourself that most of the planets in our solar system appear much brighter than the brightest stars in the sky. I could run the numbers in as much detail as I did above to show you why this is, but I'm too tired now.

    If you understand anything about photography, then Chronos's explanation should make sense to you. Hubble, in order to take a nice photograph of Mars, needs to take an exposure for a certain amount of time. Since Mars is so much brighter than the background stars, an exposure time that exposes Mars nicely is not long enough of an exposure for the faint stars to show up in it. Conversely, if Hubble took a long enough exposure that the background stars became visible, Mars would then be horribly overexposed.

    The key take home message of my post is this: just because something seems "obvious" to you doesn't mean that it is correct. You have to support the claims that you make quantitatively, you can't just make a bunch of assertions with no evidence to back them up other than "it's obviously true." The things you said in your previous post are neither obvious nor true. If you're here to learn, that's great, but if you're just here to spout off a bunch of conspiratorial nonsense, then I'd advise you not to waste everyone's time, and your threads will get locked.
     
    Last edited: Aug 21, 2012
  24. Aug 21, 2012 #23
    I'm not here to spout off, I'm here to learn and sorry that I have some different thoughts. The knowledge everyone here has (besides me) is really amazing and I appreciate the interaction. I've learned that with computers, even though numbers are the same, you can have two distinct differences in the outcome, and/or reactions, so please excuse my "conspiratorial nonsense", so to speak. =:^D

    Maybe I can open a new window of thought for people, which in turn, will lead people to a new conclusion of our universe. Having said that, I'm gonig to start a new thread, called My Theory of the Solar System. Please feel free to analyze it with your thoughts and comments.
     
  25. Aug 21, 2012 #24

    Ryan_m_b

    User Avatar
    Staff Emeritus
    Science Advisor

    If you keep with an attitude better suited to learning you will find it far more rewarding. Asking what scientists have discovered and dismissing what doesn't match your personal feelings is no way to learn or develop.
    My advice would be to get rid of any ideas regarding your own theory. If you haven't learnt what scientists have so far discovered then the chances that your theory will hold anything of merit are virtually zero. Instead start from the ground up by learning what is currently known.
     
  26. Aug 21, 2012 #25
    I understand your point Ryan. Sometimes you come up with an idea first, then the physics and numbers put it to the test afterwards. I see things in the future, then try to explain it coming backwards. It's a different way of thinking, and with all the knowledge here, I thought this was the place to be to get some answers to my thoughts. I guess you can't have an open mind and/or your own thoughts here. Sorry to intrude and please accept my sincere apology.

    I don't want to offend anyone, so if anyone cares to read and analyze my theory outside this forum, feel free to pm me. Thanks.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook