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

S^3 sphere metric-unit vector

  1. Jun 22, 2014 #1

    ChrisVer

    User Avatar
    Gold Member

    We have the (I think FRW) metric in the coordinates
    [tex] y^{0}=t,~~y^{1}=\psi,~~y^{2}=\theta,~~y^{3}=\varphi [/tex]

    [tex] g_{00}=1,~~g_{00}= - \frac{f^{2}(t)}{\alpha} ,~~ g_{00}= - \frac{f^{2}(t)}{\alpha} \sin^{2}\psi ,~~g_{00}= - \frac{f^{2}(t)}{\alpha} \sin^{2}\psi sin^{2}\theta [/tex]

    Suppose we have define a unit vector [itex]n \in \mathbb{R}^{4} [/itex] such that:

    [tex] n= ( \cos\psi , \sin\psi \sin\theta\cos\varphi, \sin\psi \sin\theta \sin\varphi, \sin\psi \cos\theta ) [/tex]

    So far I was able to show that (by doing the derivative calculations- is there any faster way one can work?)

    [tex] g_{ij} = - \frac{f^{2}(t)}{\alpha} \sum_{A=1}^{4} \frac{\partial n^{A}}{∂y^{i}}\frac{\partial n^{A}}{\partial y^{j}} [/tex]

    So I would like to interpret this result... I need some confirmation of how I interpreted it :)
    Suppose you have the vector [itex]n[/itex]. The metric is then by the equation above, defined by how the [itex]n[/itex] vector changes [itex]\partial n[/itex] along the change of the i-th coordinate [itex]∂y^{i}[/itex]. As I wrote it, by the module of the velocity of [itex]n[/itex] wrt [itex]y^{i}[/itex]. Also tried to do a grid diagram which I think is correct for [itex]S^{2}[/itex] of coordinates [itex](\theta,\varphi)[/itex], just imagining the generalization of it with a 3rd coordinate ##\psi##.
    Finally the metric is scaling by the flow of time (or [itex]y^{0}[/itex]-coord) so it's more like, as time passes, we get different images of a 3-sphere, each having its "grid" rescaled by some factor.

    Is that correct? Do you think I'm missing something important?
     
    Last edited by a moderator: Jun 22, 2014
  2. jcsd
  3. Jun 22, 2014 #2

    micromass

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    LaTeX hint: use \sin and \cos.
     
  4. Jun 22, 2014 #3

    ChrisVer

    User Avatar
    Gold Member

    Better now? Well the form of it doesn't make the big difference since my maths were correct and the interpretation has to do with the final equation, but thanks :) that way other people might understand it better.
     
  5. Jun 22, 2014 #4

    ChrisVer

    User Avatar
    Gold Member

    Another question... I did it for my own fun... because I said about that thing with "images" of sphere, and I wanted to see how , by the flow of "time", the metric would change.

    Suppose I have at time [itex]y^{0}=t [/itex] that:

    [itex] g_{ij}(t, y^{a})= - \frac{f^{2}(t)}{\alpha} |n_{,i} \cdot n_{,j}| [/itex]

    And I let time flow, to [itex] t' = t + \delta t [/itex]

    Then:

    [itex] g_{ij}'=g_{ij}(t+\delta t, y^{a})= g_{ij}(t, y^{a}) + \delta t \frac{\partial g_{ij}(t, y^{a})}{\partial t}= g_{ij}(t, y^{a}) -\frac{2}{\alpha} \delta t f(t) \dot{f}(t) |n_{,i} \cdot n_{,j}| = g_{ij}(t, y^{a}) -\frac{2 \dot{f}(t)}{\alpha f(t)} \delta t f^{2}(t) |n_{,i} \cdot n_{,j}|[/itex]

    Inserting the Hubble's parameter [itex] H= \frac{\dot{f}}{f}[/itex]

    [itex] g_{ij}'=g_{ij}(t, y^{a}) + \delta t 2 H(t) g_{ij}(t, y^{a})[/itex]

    Or

    [itex]g_{ij}(t+\delta t, y^{a})=(1+ 2H(t) \delta t) g_{ij}(t, y^{a}) [/itex]

    Could I write with that:

    [itex] \delta_{t} g_{ij} = 2 H(t) \delta t g_{ij} [/itex]

    Or equivalently:
    [itex]g_{ij}(t', y^{a})= e^{2 H(t) (t'-t)} g_{ij} (t, y^{a}) [/itex]
    ?

    Meaning that the Hubble's parameter is somewhat related to the generator of the translation of time for the spatial components of the metric?
    However that's not true for the [itex]g_{00}=1[/itex] because it's constant.
     
  6. Jun 23, 2014 #5

    Matterwave

    User Avatar
    Science Advisor
    Gold Member

    You posted four different ##g_{00}##, so I think you've not specified the metric correctly.

    Also, when you use the Latin indices ##i,j## do they go from 1-3 or from 0-3?
     
  7. Jun 24, 2014 #6

    ChrisVer

    User Avatar
    Gold Member

    latin indices= spatial indices= 1,2,3...
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: S^3 sphere metric-unit vector
  1. Twin Paradox in S^3. (Replies: 117)

Loading...