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

Measuring redshift from type 1a supernova

  1. Apr 14, 2015 #1
    I know that you can measure redshift from stars by looking at the shift in there balmer lines or more generally there line absorption spectra and seeing how far these lines have shifted from actual balmer lines or line absorption spectra of the same elements on earth.

    But how would you calculate the red shift of type 1a supernova when you don't have any absorption lines because they are not surrounded by hot gasses of known elements?
     
  2. jcsd
  3. Apr 14, 2015 #2
    The red shift of known elements Is easy to determine.
    I think it unlikely that supernovea would emit quantities of unheard of and unpredicted exotic matter which greatly distorts the picture
     
  4. Apr 14, 2015 #3

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    This 1998 Perlmutter et al slide show:
    http://arxiv.org/pdf/astro-ph/9812473v1.pdf
    shows spectra and light curves.
    Might be interesting.

    I think the lightcurve is important, the overall luminosity as it changes over several weeks.
    Type!A SNe have a particular schedule or shape of lightcurve and this schedule itself is stretched out according to the same z+1 stretch factor that the waves are elongated by.
    It appears that the Perlmutter et al team watched their sample of SNe over several weeks so they were able to study the lightcurves, that may be one key to how they determined redshifts precisely.
     
  5. Apr 14, 2015 #4

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Woah! Look at this!
    http://arxiv.org/pdf/astro-ph/9804065v1.pdf
    This may well have superseded what I had in mind a few minutes ago:
    Snapshot Distances to Type Ia Supernovae -- All in ``One'' Night's Work
    Adam G. Riess, Peter Nugent, Alexei V. Filippenko, Robert P. Kirshner, Saul Perlmutter
    (Submitted on 6 Apr 1998)
    We present an empirical method which measures the distance to a Type Ia supernova (SN Ia) with a precision of ~ 10% from a single night's data. This method measures the supernova's age and luminosity/light-curve parameter from a spectrum, and the extinction and distance from an apparent magnitude and color. We are able to verify the precision of this method from error propagation calculations, Monte Carlo simulations of well-sampled SNe Ia, and the Hubble diagram of scarcely observed supernovae. With the reduction in telescope time needed, this method is three to four times more efficient for measuring cosmological parameters than conventional light-curve based distance estimates.
    32 pages, 5 figures, Accepted Astrophysical Journal
    I've sat in on several semesters of lectures by one of these guys and I have the highest respect for them as a group. Riess, Perlmutter, Filippenko are known top people in cosmology.
    Since I don't know the detailed history of the 1998 discovery I can't swear that this method was used successfully, but it's worth trying to understand it. they claim it supersedes the earlier "light-curve" method, which required observation over a longer span of time.

    The slide presentation abstract http://arxiv.org/abs/astro-ph/9812473 suggests looking at this longer,more technical paper, for details:
    http://arxiv.org/abs/astro-ph/9812133
     
    Last edited: Apr 14, 2015
  6. Apr 15, 2015 #5

    wabbit

    User Avatar
    Gold Member

    An example is this article about high-z supernovas, which discusses details of their identification of the spectrum from observations at several points on the lightcurve. If I read them correctly, they model the rest frame spectrum and it's change over time, factor in the absorption spectrum from the host galaxy, then translates this to a series of expected observer frame spectra from time dilation, as a function of z. After that they do a best fit of observed curves, allowing (a) identifying the SN type and selecting the relevant ones (b) measuring their redshift.

    I think each study uses slightly different methods for this as well as for the luminosity estimate factoring in extinction from various intervening things.

    http://arxiv.org/abs/1205.3494
    Precision Measurement of The Most Distant Spectroscopically Confirmed Supernova Ia with the Hubble Space Telescope
     
  7. Apr 15, 2015 #6

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Nice find, Wabbit! they say redshift 1.71 so wave stretch factor 2.71
    Let me estimate the distance using this integrand in NEDI
    17.3*(((17.3/14.4)^2 - 1)*s^3 + 1)^(-1/2)
    It says 15.817
    that is 15.8 billion light years.
    this is a kind of test to see how good the eventual Hubble time number 17.3 is.
    now I have to compare 15.8 billion ly with whatever distance they might have inferred from the brightness of the supernova
     
    Last edited: Apr 15, 2015
  8. Apr 15, 2015 #7

    wabbit

    User Avatar
    Gold Member

  9. Apr 15, 2015 #8

    marcus

    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    In their conclusions on page 8, I recall their saying distance modulus 45.60
    My computer got disconnected from internet just as I was about to edit post #6 to that effect.
    Too sleepy now, after midnight. I still want to compare that 15.8 billion lightyears with whatever they have for distance to the supernova, to see if the 15.8 is way off or not.
     
  10. Apr 15, 2015 #9
    Guys thanks for the help but i'm only in highschool physics please explain in simpler terms
     
  11. Apr 15, 2015 #10

    wabbit

    User Avatar
    Gold Member

    The spectrum of a supernova is characterized by the process producing it, not by it's surroundings.
    It is actually better if there is no or little intervening absorption lines (from host galaxy or gas clouds in our line or sight, etc.) since you are then looking at a "pristine" spectrum that you can directly match with the model. Part of the work in analysing actual SN spectra consisys presisely of removing the distortions induced by such absorption.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Measuring redshift from type 1a supernova
Loading...