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

I Do gravitational waves really propagate?

  1. Aug 22, 2016 #1
    I understand gravitational waves as ripples in spacetime. As PeterDonis said in an earlier post,
    " The 4-d spacetime geometry does not have to "propagate" anything; it just is."
    On the other hand, this is block-universe speak. In this language, verbs that imply change are forbidden. We might just as well say that radiation is present within the light cone of a supernova. In this language light does not propagate either.
    When we detect gravitational waves coming from some event, we can say that our wordline intersects with a region of spacetime that contains ripples which are there because of some extreme event in another point of spacetime (a bit like experiencing bumps on the road).
    Alternatively, we can express ourselves in a more natural way and just say that there is some passing wave that makes *space* (not spacetime) contract and expand for a few moments and then passes. Is this an acceptable way of expressing it?
    I guess the problem of this way of talking is that prompts the question "a wave of what?". Which puts me in a substantivalist frame of mind...
  2. jcsd
  3. Aug 22, 2016 #2


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    It's the wave of the gravitational field, and in my opinion that't the more physical view than the spacetime point of view since physics is about what's observable, and for something to be observable you have to define a measurement device, and this measurement device defines a local reference frame and thus a local split of spacetime into space and time, i.e., a specific local "time slicing" of spacetime, and that tells you within GR what's measured by local devices, e.g., the LIGO Michelson-like interferometers used successfully to detect gravitational waves exactly in the way you describe it in your poting: "here is some passing wave that makes *space* (not spacetime) contract and expand for a few moments and then passes." What's "waving" is the gravitational field in the same sense as light is a electromagnetic wave, where the electromagnetic field is what's "waving".
  4. Aug 22, 2016 #3


    User Avatar
    Staff Emeritus
    Science Advisor

    The propagation models for gravitational waves I've seen have all been interperable as ripples in space, which can then be regarded as propagating through space with time. That is to say, one can find coordinates such that ##h_{tt} = h_{tx} = h_{ty} = h_{tz}=0##, where the ##h_{ii}## are the metric pertubations that are the mathematical models of the "ripple".

    It's not quite clear to me if we can say that all GW's can be modelled in this manner. Technically speaking, I know that we can always put the GW solution into the Lorentz gauge, but I believe the olutions with the above properties of having only spatial components are a further specialization of the Lorentz gauge, the transverse-traceless (TT) gauge. My suspicion is that TT waves exist in a flat space-time, but it's much less clear if TT waves still exist when one imagines GW's propagating through a curved space-time, for instance one with a Schwarzschild background metric rather than a flat background metric.

    It's possible to view GW's in other coordinate systems, in which case the simple propagation model above might not directly apply. For instance, coordinates exist in which GW's don't even satisfy the wave equation. But one can argue that the basic propagation model is good as long as some coordinates exist in which it works, I think.

    It's certainly also possible to view a GW space-time as something that "just is", for example to use the block universe interpretation.
  5. Aug 22, 2016 #4


    User Avatar
    Gold Member

    I wonder what Peter has to say. Is any of this weirdness expected to change once we have a quantum theory of gravity?
  6. Aug 22, 2016 #5


    Staff: Mentor

    Both ways of talking (the "block universe, nothing changes" way and the "things propagating and changing in space with time" way) are ways of trying to express the physics using vague, imprecise, ordinary language. We should not expect either one (or any ordinary language description) to properly capture all aspects of the physics. If you really want to see the most precise description we have of the physics, you have to use math.

    So to me the title question of this thread is not well posed.
  7. Aug 22, 2016 #6


    Staff: Mentor

    Only in the sense that it is likely to get weirder still. :wink:
  8. Aug 22, 2016 #7
    Be very careful, one can think of waves "moving" free test masses as shown in most web explanations which can perhaps make you think of them being glued onto the old "rubber sheet" model used for explaining gravity. However, you can also think of the masses as just hanging, unmoving, in space and then what changes is the coordinate speed of light between them, the waves become a "ripple" of a change of a property of the vacuum between the masses and the illusion of substantivalism evaporates. The difference between those views is nothing more than your choice of coordinates.
  9. Aug 22, 2016 #8
    I guess everyone's different. I first need to grasp it intuitively and possibly visualize it geometrically and then I feel ready to tackle the math. Not the other way around. When I come full circle, and the math and my intuition match, then I feel I have a full understanding.
  10. Aug 22, 2016 #9


    Staff: Mentor

    You seem to be assuming that your intuition won't have to change. It probably will; you are trying to understand a domain that our intuitions simply did not evolve to deal with. So you can't assume that your intuition as it currently stands will even have any way to grasp and visualize what is going on.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted