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

Simulating dark matter by assuming stable solution

  1. Feb 16, 2015 #1
    Hey,

    I've a short question. Is there any research attempting to constrain models of dark matter by assuming stable solution for the universe? I'm interested in knowing whether this gives any sensible constraints on the dark matter models.

    I'm still fairly new to studying dark matter but as I understand there are loads of different models floating around; Warm Dark Matter (WDM), Cold Dark Matter (CDM) (with and without collisions), self-annihilating dark matter, fuzzy dark matter (non-relativistic bose condensate), and so forth. The idea has been, as far as I understand, is to try to constrain the dark matter model to fit the observations, and to constrain the "Dark Matter particle" properties.

    The researches I've read about have been done as follows (as far as I understand):
    1. Make an educated guess on the properties of dark matter while letting one or two variables regarding the 'dark matter particle' go unconstrained
    2. Run the simulation from initial stage to final stage (final stage usually being the observable universe)
    3. Analyze the result and compare with observations

    My question is whether there are any attempts at doing the following in any shape or form (and I'd appreciate it if I could read about them):
    1. Make a model with as little constraints as possible
    2. Create an initial stage based on observations and corrections to the results of previous simulations (effectively this means creating the observable universe)
    3. Assume that the system is stable
    4. Constrain the model to only include the solutions that do not shake the stability
    Thanks.
     
    Last edited: Feb 16, 2015
  2. jcsd
  3. Feb 16, 2015 #2

    Chalnoth

    User Avatar
    Science Advisor

    Stable solution? There is no stable solution. The universe is expanding. It is out of equilibrium, and will remain out of equilibrium until there is nothing left but empty space.
     
  4. Feb 17, 2015 #3
    I should clarify; I mean stable solution for galaxies on short time-scales and whichever structures remain stable if left alone (the bare-bone idea).

    In short, I'd like to read about research with heavy emphasis on creating a "minimum constraint" model with constraints directly from observations (and/or previous simulations)

    observations (and/or previous simulations): By this I mean that the starting point of the model is the expected solution.

    Ps. Is there any way to edit the initial question afterwards? It would be helpful to correct it in case some people only read the original post.
     
    Last edited: Feb 17, 2015
  5. Feb 17, 2015 #4

    Vanadium 50

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    On a short enough time scale, any solution is stable, On a long enough time scale, virtually all are not. This condition has no meaning without a better definition of "short", in years,
     
  6. Feb 17, 2015 #5
    One can construct a model which blows up after any finite time-scale (whether this is physical or not is another question).

    I should clarify further; I am not interested in solving this question myself (that means I'm not interested in coming up with the time-scales and constraints from observations; this has already been researched by so many people before, see research on MSP, CCP, catch 22, MFP and dwarf spheroidal galaxies..) I'm interested in the thoughts of other researchers on the subject.
     
    Last edited: Feb 17, 2015
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Simulating dark matter by assuming stable solution
  1. Dark Matter (Replies: 3)

  2. No dark matter (Replies: 59)

Loading...