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

Linear vs non linear matter power spectrum

  1. Apr 28, 2013 #1


    User Avatar


    I have troubles understanding the difference between linear and non linear matter power spectrum. These words are commonly used in the litterature, but I have found no definitions yet.

    My understanding is that there is one definition of the power spectrum for matter distribution, basically as the Fourier transform of the 2-point correlation function of the density field.

    It covers both very large scales where linear approximation for the dynamics should be accurate, while at small scales the non linear gravitational collapse heavily impacts the dynamics.
    Still, we have only one power spectrum.

    So, what does linear or nonlinear power spectrum refers to? (in terms of maths, I mean)

    Thanks for your help!
  2. jcsd
  3. Apr 28, 2013 #2
    If you graph a linear relation the graph will show a straight line whether or not its flat, diagonal or vertical. This is described as value a is directly or indirectly proportional to value b.
    A non linear relation will be a curve, examples include bell curves, sinusoidal wave forms
    etc. Hence non linear.
    value a is exponentially proportional to b. This type of relation will give a bell curve
  4. Apr 28, 2013 #3


    User Avatar

    Yes, I'am aware of this, but I'm rather talking about mathematical description of large scale structure formation! I have found that details can be found in peeble's book, 1980. But I don't have it. Any other ref is welcome! :)
  5. Apr 28, 2013 #4
    Ah ok wasn't sure on what you were unclear on. I may have an article that will help but I'll have to dig around in my archives to find it.
  6. Apr 29, 2013 #5


    User Avatar
    Science Advisor

    Well, I think the general idea is to start using the non-linear power spectrum at the scales in which there is a noticeable difference between the linear power spectrum and the non-linear approximations. I doubt there is a firm limit on precisely where this occurs, but the best way to investigate it is probably to start investigating some nonlinear power spectrum approximations.
  7. Apr 29, 2013 #6
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook