Docstoc

lecture16

Document Sample
lecture16 Powered By Docstoc
					      Distant Universe

            Seb Oliver
Lecture 16: Gunn-Peterson Effect &
   The others search for the first
             galaxies
               Main Topics
•   Standard Hot Big Bang Model
•   Classical Observational Cosmology
•   Galaxy Evolution
•   The Hunt for the First Galaxies
•   Background Light
•   Structure Formation
 The Hunt for the First Galaxies
• The Search for Ly a Galaxies
• Photometric Redshifts and the UV-drop-out
  Technique
• Distant Absorbers
• Dusty Galaxies
Lyman a Forest
     The Gunn-Peterson Effect
• Observation of quasar absorption systems
  demonstrate that the Universe at high redshift is
  ionised
• Absorption by Ly a clouds and indeed any neutral
  hydrogen eats away at quasar continuum
• However although we see individual clouds
  absorbing some continuum we do not see an
  overall drop in the continuum
• The quantification of the optical depth that would
  be expected if gas in the Universe were neutral is
  known as the Gunn-Peterson effect, which has yet
  to be observed
The Gunn-Peterson Effect



  Observed continuum
Expected continuum if
gas in Universe near
quasar were neutral
         The Gunn-Peterson Effect
                                        dP  dN   n0 1  z 
                                                                c    dz
                                                                H 0 1  z
  Absorption by Lya clouds eats away at the quasar continuum

    The total optical depth       N.B. nhi is proper number density


                                              1  z  0 nHI z 
                   l                    zQ

          n dl    
                               c
                                         
                                                                         dz
                   0
                               H0        0            1  z 2         1  z



    10-5.16
                   dp 2
                      m       c
                                   zQ   10   5.16   dp
                                                        1  z  0 nHI z  dz
                   d      
                              H0   
                                   0
                                                     dn
                                                       1  z 2               1  z
          The Gunn-Peterson Effect

    10-5.16
                   dp 2
                      m       c
                                    zQ   10   5.16   dp
                                                         1  z  0 nHI z  dz
                   d      
                              H0    
                                    0
                                                      dn
                                                        1  z 2               1  z


          dp                   dp
         d d  1            d
                                          Is a d function

       Problem sheet 4
                                nHI  z                                0
        0    4.62 3
                      1
                                                            1 z 
                10     hm 1  z 2 1  z                           121.6nm

                                                              0
                                                                       1  zQ
                                                          121.6nm
                Gunn-Peterson Effect
    But (apart from the forest lines) we do not see a general
    drop in the continuum short of (1+zQ) 121.6 nm

    Say  ~ 0.1 for =1 this implies
    a co-moving density of n0 < 10-6.8 h m-3 at z = 5

    When we take into account the clouds this changes a bit

     HI  10 7.8 h 1   Neutral Hydrogen in uniform gas

     HI  10 5.5 h 1   Neutral Hydrogen in clouds

     B h 2 ~ 0.015       Total Baryons from Nucleosynthesis
Not enough neutral gas to make up all the baryons  Universe Ionised
 The Hunt for the First Galaxies
• The Search for Ly a Galaxies
• Photometric Redshifts and the UV-drop-out
  Technique
• Distant Absorbers
• Dusty Galaxies
               The Search For Lyman-a
                      Galaxies
      • Assume a galaxy forms in one monolithic
        episode of star-formation  a “primeval
        galaxy”
      • What would a “primeval galaxy” look like
      • Since we know SFR  LUV
6.3 10 22 kg s 1  1M  yr 1 produces   L  2.2  10 9 L  8.4  10 35 W


                       1 kg s 1  1.3 1013 W
       The Search For Lyman-a
              Galaxies
• At high z a reasonable lower limit to
  luminosity would be if all the stars had
  formed over the age of the Universe
                         2                2
     M M          t         
            t          3H  z  3H 0 1  z  1  z
•   This implies that a 1011 M galaxy at z = 3 would
    have a luminosity L=1.8×1011L
•   Roughly 10 times its present day luminosity!
•   Stellar synthesis predicts ~10% of this would be
    emitted in Lyman alpha line!
•   Lyman -a is n=21 transition @ 121.6 nm
The problem
              Too many galaxies

              Even if high-z
              galaxies are
              luminous they
              will still be faint
              (low flux)

              How can we tell
              them apart from
              the low-z
              galaxies?
      The Search for Lyman-a
• Instead look for the Lyman-a galaxies by
  imaging a field in a narrow filter that only
  let through light at

0  1  z 121 .6 nm             Where z is the redshift
                                   they had chosen to search

      e.g. z = 3
     This would highlight all the primeval galaxies
     While light from other galaxies would be
     negligible
       The Search for Lyman-a
•    Many such Lyman-a surveys were carried
     out no success.
•    Possible explanations…
    1. Primeval do not exist, galaxies form from
       aggregation of smaller objects. The SFR in
       these small objects on their own would not be
       sufficient to be detected in Ly-a searches
    2. Dust in primeval galaxies obscures the sites of
       SFR
    3. Looking at the wrong z
 Photometric z and UV drop-out
• The prospect of finding high-z galaxies
  looked bleak
• In mid 1990s a new technique for finding
  galaxies was developed.
• This turned out to be very successful
• Now 1000s of examples of galaxies z>2.5
                     Photometric z
• The brute force approach to determining a
  galaxies z is to take high resolution spectra
• This is because an accurate redshift requires
  “narrow” features
                                                         0
                                              1  z  
                                                         e
  f




                     e           0
                          
                0
1  z                  But high res. Spectra are expensive
             121.6 nm
             Photometric z
• These features are usually no use at very
  low resolution because they get washed out
 f




             e           0


 • However broader features can be used …
Blackbodies                                            40K
• It is quite easy to                                  10K
  determine a blackbody                                 2K
  temperature even from
  broad band colours
         0.0029                0.0051
T K                  T K
         Bpeak m             Bpeak m

• If we know the
  intrinsic or emission
  temperature Te we can                        e.g. CMB
  estimate the redshift                          T0 ~ 3K
       Te            1      1         Te       Te ~ 3000K
T0             T0               
     1  z         0 e 1  z  1  z     z ~ 1000
       Lyman-Break galaxies
• If a galaxy has sufficient neutral hydrogen
  (column N>1024 m-2) then almost all Ly-
  limit photons l < 91.2nm will be absorbed
 f




                   e=91.2nm
                      
 • Known as Lyman-break galaxies, Lyman-
   limit systems, UV-Drop-outs
      The Drop out Technique
• As a Lyman-break galaxy moves to higher
  redshift the break moves from band to band
 f




                     
             UV-Drop-out
• At z~2.7 the Lyman Break passes out of the
  U-band (and into the B-band) so a galaxy
  will have negligible flux in the U-band
  while similar fluxes in the B and e.g. V -
  band
  FU/FB




                     z
         Photometric Redshifts
• drop-out techniques and estimating blackbody
  temperatures are examples of photometric redshift
  techniques
• I.e. using photometry (rather than spectroscopy) to
  measure redshift
• In general such techniques take a “template”
  spectrum and fit this to the observed photometric
  data with z as a free parameter
• Some techniques use neural nets or similar to
  avoid having to explicitly define the “templates”

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:4
posted:2/24/2012
language:
pages:26