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

Speed of light effected by expansion of space?

  1. Aug 27, 2009 #1
    I once learned (way back when) that nothing can move faster than light in a vacuum. But this now seems simplistic. For if there is an object A whose light we see as it was released 13.7 billion years ago and object A is now 45 billion light years away, then that object A has moved further from its starting point in 13.7 billion years than its own emitted light. Was object A once moving faster than light and now is moving much slower than it? Does this imply that objects are effected by inflation and expansion but light isn't? Thanks
  2. jcsd
  3. Aug 27, 2009 #2


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Rasp, in the case of ordinary motion--the kind we are used to--there is always somewhere you are going. Some destination which at least temporarily you are approaching.

    The simple expansion of distances that you get in General Relativity is not like that. Distances between all the widely separated galaxies are increasing without any galaxy going anywhere. Think of it as changing geometry, not motion. GR is a dynamic geometry theory, distances can change. Indeed a lot of distances are changing at a faster rate than c----and this is not breaking any speed rule. It is not ordinary motion of the sort governed by special rel.

    Individual galaxies in clusters do orbit their neighbors, they have diddly little motions in their local neighborhoods---typically a few tens of km/s or something on the order of 100 km/s. Those are expanding distance rates. The distances that are expanding are the largescale distances between widely separated galaxies, outside of any gravitationally bound cluster.

    Try the uni.edu link in my signature and see if you can make it work for you. You need to prime it by putting in 0.25 for matter, 0.75 for cosmo constant "Lambda", and 74 for Hubble constant.

    Or any three numbers approximately like that. Then you put in the redshift of some galaxy, like z = 8, and it will tell you how far away the thing is now and how fast the present distance to it is increasing now.

    It will also tell you distance then (when the light was emitted that is now reaching us) and recession rate then.

    Try it. If you have any trouble, ask and one of us will talk you thru it.
    Last edited: Aug 27, 2009
  4. Aug 27, 2009 #3
    Rasp, while what marcus is telling you is correct for the currently used mainstream model, it is not necessarily proved fact. There are alternative explanations that produce the same results without such 'faster than the speed of light' behavior.

    Much of the current 'mainstream' model has many public attention getting features -- which often seem to be as important as the supporting data.

    At this point, it is not clear what model will actually be found to be best. This is not intended to start a heated discussion. But is only a statement to clarify what is really fact from what MIGHT be fiction.
  5. Aug 27, 2009 #4
    Marcus, I was originally prepared to disagree with you on a finer point with the following:

    "In the spacetime of general relativity, it makes sense to say that the velocity of an object is only meaningfully defined in an inertial frame local to that object. We can setup a local system of syncronized clocks and gridwork of distances to define space time coordinates. We can measure or extrapolate velocities with respect to these coordinates. But at any great distance out, the coordinates don't describle space and time."

    I was preparing to say, that our local coordinates, mapped over all of spacetime are non-physical, but one must define 'non-physical'. In this context, it would be that vlight ≠ c. There doesn't seem to be any good reason to call these coordinate systems non-physical, but only nonlocal.
    Last edited: Aug 27, 2009
  6. Aug 27, 2009 #5
    Good point. That is in essence the difference in what I was saying. In other models this becomes more important -- while often the Standard Model tends to talk in terms of its nonlocal interpretation.
  7. Aug 28, 2009 #6


    User Avatar
    Science Advisor

    Well, the correct statement is that differences in velocity are only well-defined locally. When comparing the velocities of two objects far away, General Relativity provides no answer. We can, of course, write down some coordinates and come up with some measure of velocity. But whatever measure we write down will be arbitrary.

    This is, fundamentally, why there is no problem with far-away objects moving "faster than light" compared to us: If I compare the velocities between far-away objects, I can come up with any velocity I choose just by picking different coordinates. Obviously there can't be a limit if I can make the velocity anything I please!

    The actual speed of light limitation in General Relativity, then, is purely a local limit. It states that any person that is observing the speed of a photon passing by them will always measure it to be moving at the speed of light. This also means that if we look at a far-away object, no matter how it is moving compared to us, it will always be seen to move more slowly than the photons moving past it.
  8. Aug 28, 2009 #7


    User Avatar
    Science Advisor

    It's proven about as much as any fact in science.

    No, there aren't. The other alternative 'explanations' fall flat when compared against the full gamut of cosmological evidence.
  9. Aug 28, 2009 #8
    No they don't. What appears to be falling FLAT is the Standard Model use of General Relativity on the cosmological scale. Note basic physics:

    IF the universe is FLAT as current measurements indicate, then

    1) Space is Euclidean -- meaning NO 'space/time warp' at the cosmological scale
    2) Therefore, General Relativity does NOT apply.
    3) Special Relativity DOES apply.
    4) The well established and verified relation for redshift is the Special Relativity Doppler Redshift.
    5) The Doppler redshift relation NEVER exceeds the speed of light for any value of redshift.

    NOTE: General Relativity is still a valid mathematical relation for gravitational fields but ONLY within the universe -- not when applied to its entirety.
  10. Aug 28, 2009 #9


    User Avatar
    Science Advisor

    Our universe is nearly spatially flat. There is space-time curvature. So the rest of your argument fails.

    Edit: One other slightly pedantic point is that General Relativity is quite valid in flat space-time. It just reduces to special relativity. This doesn't describe our universe on large scales, however.

    No deviations from General Relativity have yet been found, though physicists are trying very hard to do so.
  11. Aug 28, 2009 #10


    User Avatar
    Science Advisor
    Gold Member

    In a sense, expansion of space 'speeds up' photons by stretching them out [aka redshift]. In a mythical, absolute, fixed reference frame, photons would appear to travel across space at variable speeds. But, to any observer embedded in the physical universe, all photons appear to travel at exactly 'c'. So even though really distant stuff, like the surface of last scattering, appear to be receding faster than 'c', their photons still reach us.
  12. Aug 28, 2009 #11
    Yes, key word is 'appear'. Note, our current interpretation of gravity MAY only be valid after the time of last scattering -- IF gravity like other forces -- is found to be due to a 'transport particle'. This explanation is appealing in that it would explain the flatness that is found -- assuming the 'particles' cannot travel backward in time.

    Note, if force cannot be transmitted through the time coordinate, then there is no possibility left for a 'warp-age' -- and the argument holds.

    Actually, while General Relativity is 'valid' conceptually in a flat space time (reducing to Special Relativity) there is a problem with the assumed metric for a non-zero Lambda. Also, other problems in the structure of the secondary relations. The Standard Model does not address this correctly in a perfectly flat scenario. It wasn't designed to -- it assumes it is warped. This is likely the cause for the apparent 'hump' that has been interpreted as 'dark energy'.
  13. Aug 28, 2009 #12


    User Avatar
    Science Advisor

    Gravity most certainly didn't change at the surface of last scattering. That was only around 3000K, after all, which is quite a bit cooler than our own sun. And our current knowledge of gravity works quite well for the behavior of stars, as well as in the lab when we perform tests at these and much higher temperatures.

    What problem?

    I can't make heads or tails of what you're trying to say here.
  14. Aug 28, 2009 #13
    IF gravity is due to a 'transport particle', then presumably this particle was being scattered as well. It is the cessation of scattering that could cause the loss of repulsion we see as gravity. So in this possible concept, gravity as we now know it only came into existence AFTER the time of last scattering. This is all related to the idea of a 'transport particle'. Agreed, not proved -- but also not falsified. Temperature has nothing to do with this (any relation is an end result not a causative).

    Question for you:

    Is a General Relativity Lambda term needed in any solution other than a cosmological one?

    Also note in a flat (non-warped) universe, the normalization used is likely incorrect. By that I mean, matter will NOT be expanding away from the expansion origin at the speed of light (which is effectively what is assumed with the FLRW metric -- when flat). Expansion is 'two-tiered' -- photons at the speed of light and matter at some velocity less (this is actually what is meant by the surface of last scattering). ALSO, there is no density term involved (no critical density is possible). AND no time dependent scaling factor needed. Likely more -- this is what comes to mind at the moment.
  15. Aug 28, 2009 #14
    No. The stretched photons have lesser frequencies
  16. Aug 28, 2009 #15


    User Avatar
    Science Advisor

    The surface of last scattering is an electromagnetic phenomenon. There is no reason whatsoever to believe that it would have similar behavior upon other forces. In fact, there are very strong reasons to believe this is not the case.

    The biggest reason, in this case, is that our models of structure formation make use of gravitational dynamics before the surface of last scattering. And they manage to predict very accurately what the surface of last scattering looks like. This interpretation is confirmed by comparing the distribution of the CMB to the distribution of closer galaxies (through Baryon Acoustic Oscillation observations).

    Another, more theoretical reason is that the unification scale of gravity with the other forces is expected to be at the planck scale, which is about 10^19 GeV, or very roughly 10^33K, around 10^30 times higher in temperature than the surface of last scattering. There is some possibility that this might be brought down all the way to the TeV scale (about 10^26K), but experiment completely rules out anything below this.

    If it turns out that the cosmological constant is the explanation for the observed acceleration, then its value is far, far too small to be detectable in any other context at this time.

    Once again you're making no sense. This paragraph smacks of a fundamental lack of understanding of general relativity as well as the FLRW metric and its implications.
  17. Aug 28, 2009 #16


    User Avatar
    Science Advisor

    Well, that too. But the effect Chronos described is a valid one, from a certain point of view.

    First, to talk about this sort of thing, we have to define some sort of global reference frame. This choice is somewhat arbitrary, but a particularly nice choice is a reference frame in which the universe looks approximately homogeneous to all observers, and observers at different spots in the universe see the same average density at the same time.

    One might think of this as defining this global reference frame by the cosmological constant: "now" is comprised of all observers who see the CMB at the temperature we see it today. As our universe expands, the CMB will cool, and then the global "now" will be comprised of all observers who see it at that temperature.

    Now, if we define this global reference frame (which is arbitrary, but convenient), then we can talk about what a far-off object is doing right now. If we take a photon that I send off into space, for instance, we can say how much space it has traveled in a given time. In this global reference frame, the photon will always be traveling at speed c with respect to the local matter. But the universe is expanding, so its total velocity in this reference frame will be:

    v = c + Hd

    So we will see this photon as receding from us faster than the speed of light. It will be sort of "carried away" by the expansion.

    Please bear in mind, as I said in my earlier post, that this is an artifact of the fact that we can define velocities however we choose when comparing objects that are separated by some distance. This choice is just a convenient one.
    Last edited: Aug 28, 2009
  18. Aug 28, 2009 #17
    Let me ask you this:Where is the origin of this universal reference frame?
    No answer, and hence no such thing.

    No observer can measure the speed of light different from c (in vacuum, of couse)
    You fail to prove that the speed of light varies because you cannot define this unique frame
    All the real frames are based on an illusion of still space, that's why you say v=Hd in the first place
    Last edited: Aug 28, 2009
  19. Aug 28, 2009 #18


    User Avatar
    Science Advisor

    The physics of the production of our region of the universe acted in such a way that for some observers, the universe appears nearly homogeneous and isotropic. We can take advantage of this fact to define a universal reference frame.

    Arbitrary does not equate to "no such thing".

    No observer can measure the speed of light locally to be different from c. But it is perfectly possible to look at light (or even non-relativistic objects) traveling between far-away points at speeds that appear to exceed that of light.
  20. Aug 28, 2009 #19
    That's because the velocity of a "nothing" between two photons reaching a point from two different points is >c
    Last edited: Aug 28, 2009
  21. Aug 28, 2009 #20


    User Avatar
    Science Advisor

    Well, that's one situation. We're talking about a different one, though, due to the expansion of the universe.
  22. Aug 28, 2009 #21
    Yes, the surface of last scattering is a electromagnetic phenomenon -- or better termed a 'photon' phenomenon -- and so may be the gravitational one.

    I'm not referring to our models before and during the surface of last scattering -- have no disagreement with them as far as I know.

    Your statement "unification scale of gravity with the other forces is expected to be at the planck scale" is simply an opinion -- not justified.

    I see we agree on the cosmological constant.

    The problem is with the FLRW metric and its application to cosmology -- agreed it makes no sense.
  23. Aug 28, 2009 #22


    User Avatar
    Science Advisor

    Nope. Gravity is a different force.

    What is not opinion, however, is that the unification scale is no lower than the TeV scale. If it were, we would have produced microscopic black holes in particle accelerators by now. We haven't, so it isn't.

    However, the theories where the unification scale is as low as the TeV scale are extremely speculative, requiring the "true" strength of gravity to be many orders of magnitude larger than what we measure, because it is being diluted into extra dimensions. If the true strength of gravity is given by the G we measure, then the unification scale basically has to be somewhere near the planck scale.

    No, the problem is with your understanding of it. There are no self consistency issues with the FLRW metric.
  24. Aug 28, 2009 #23
    "The simple expansion of distances that you get in General Relativity is not like that. Distances between all the widely separated galaxies are increasing without any galaxy going anywhere. Think of it as changing geometry, not motion. GR is a dynamic geometry theory, distances can change. Indeed a lot of distances are changing at a faster rate than c----and this is not breaking any speed rule. It is not ordinary motion of the sort governed by special rel."

    Marcus, thanks. I can understand that the galaxies are not moving per se at velocities greater than c, but only in relation to each other as "the "backdrop of dynamic geometry" expands. Can we call this backdrop the balloon? Well, then what of the balloon itself? Is it moving? If so, what is moving? Is the space of the balloon a membrane of sorts, carrying within it's structure the objects of galaxies, photons and such? Or are you equating the balloon to empty space; a pure vacuum which only separates and defines distance between objects?
    I know you are sparing me the technical details (and I appreciate that!) but to call the expansion of the Universe "a changing geometry not motion" seems like we are substituting the mathematical model for the reality. Thanks!
  25. Aug 28, 2009 #24
    Your opinion.

    Nothing to do with the subject.

    Again nothing to do with the subject.

    Again, the point was FLRW does NOT apply -- so any 'self consistency' issues are moot.
  26. Aug 28, 2009 #25


    User Avatar
    Science Advisor

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook