bowman_hubble_2010

							The Dawn of 21 cm Cosmology
        with EDGES




          Judd D. Bowman
              Caltech
          Alan E. E. Rogers
        Haystack Observatory
          21 cm cosmology




75% Hydrogen          WFC3/HUDF09, z = 7-8

  (by mass)




                                          Illingworth


                             Image: Scientific American 2006
         21 cm hyperfine line of hydrogen




•   For a cloud of hydrogen gas between us and a radiation source,
    21 cm scattering changes the observed brightness temperature:




           standard model
                                          local
             cosmology
                                       over-density
                                                         spin
                             neutral                  temperature
                            fraction

•   Universal radiation source: Cosmic Microwave Background
21 cm Background




                             “Science with the MWA”
               Greenhill, Bowman, et al. (2010, in prep)
                             Figure by Matt McQuinn
Science, Vol. 325. no. 5948, pp. 1617 – 1619, 25 September 2009
    CMB                              Spin, TS
                          Kinetic



                  Ionized fraction
                     xi = 1 - xHI




Mean brightness
 temperature



                                         Pritchard & Loeb 2008
                     Why global 21 cm?

• Straightforward probe of mean neutral fraction and HI gas
  temperatures (spin + kinetic)
   •   Star formation history, galaxy evolution, early feedback
       mechanisms, etc.
   •   Direct constraint on redshift and duration of reionization

• “Simpler” than imaging/power spectrum
   –   Average over large solid angle
   –   Signal fills aperture of any antenna – a single dipole is sufficient
   –   Ignore ionospheric distortions
   –   Polarized foregrounds reduced

• The only feasible probe of the Dark Ages (z>15) IGM for at
  least the next decade
                 Foregrounds for 21 cm cosmology

                                                   z
      Milky Way – synchrotron emission
         100-1000 K @ 200 MHz            Foregrounds




                                               |21 cm|




                       TF()                   [MHz]




Haslam et al. (1982)
     Experiment to Detect the
Global Epoch of Reionization Signature
              (EDGES)
                          EDGES
Spectral band:         90-205 MHz
Spectral resolution:   13 kHz
FOV:                   ~80 deg (FWHM)
Dynamic range:         >106




    Antenna                             RFI trailer
EDGES block diagram
Bowman & Rogers (in prep)
      Comparison-switched spectrometer


• 3-position switch to measure (cycle every 10s):




• Solve for antenna temperature:




               (Tcal > TL  300 K, TA  250 K, TR  20 K)


• Limitations:
   – Total power differences between TL and TA produce residuals
   – Temporal variations: comparing measurements distinct different times
Comparison-switched spectrometer




                                          Noise source (p1)

                                          Internal load (p0)
                                          Antenna (p2)
        “Calibrated” sky spectrum         p1– p0
           RFI filtering sky spectrum
        w/“Calibrated”and integration
            T_A ~ (p2 – p0) / (p1 – p0)   p2 – p0
EDGES Latest Results
Measured spectrum

         Murchison Radio-Astronomy
         Observatory (MRO)

         Aug 20 – Oct 20, 2009

         1440 wall-clock hours on sky
         ~500 hours after RFI filtering
          ~50 hours actual integration
Total power in band vs. time (Aug 23, 2009)




     Average antenna temperature 90-205 MHz
              Integration… rms vs. time
                w/ baseband removal


                                              20 Oct 2009
RMS [K]




          Systematic free performance level:
          RMS = 30 mK in 13 kHz channels (thermal)
                  ~5 mK in 2 MHz bins


                       Integration time [hours]
                     Model fitting

                            11
• Polynomial term:    p   an  n         to account for impedance
                                           mismatch + galactic spectrum
                            n 0


                                                T21
• Simple step model of reionization:         m      erfc   0 
                                                 2
                                     “instantaneous” reionization
                     T21




        3 science parameters:           T21, , and 0
        12 nuisance parameters:         an    (ACKK!!)
Was reionization instantaneous?

          Fit for T21
       with fixed  = ∞
           Test all 0
Confidence intervals on T21 with fixed =


                              October 2009



                                                reionization
                                                   barrier




                              yellow: 68%
                              gray: 95%




                                      Bowman & Rogers (in prep)
Pace of progress


                September 2008
                 February
                   August
              Bowman et al. 2009




                                   reionization
                                      barrier




                   yellow: 68%
                   gray: 95%
How long was reionization?

          Fit for 
 with fixed T21 = CDM
         Test all 0
Confidence intervals on dx/dz with fixed T21




      Δz> 0.4                   Δz> 0.2




                                     Bowman & Rogers (in prep)
21 cm derivative: constraints and forecasts
     z=6                            z=13                 z=25

                Feb 2009                        NOT
                                           reionization…

                                            absorption
                Oct 2009

           anticipated systematic
             limit (no/low RFI)




          Integrate +
       improve bandpass
                            Summary

• Concluded 3 month deployment in MRO
   – Deepest broadband spectrum ever acquired: 5 mK rms
   – Instantaneous reionization ruled out: 21 cm rapid step
     constrained to <30 mK between 6<z<13
   – First direct dxi/dz constraints on diffuse IGM


• The next step:
   – Upgrade digital backend; exploring Berkeley CASPER open
     architecture boards for high throughput
   – Redesign antenna to improve impedance match (use lower
     order polynomial for continuum removal)
   – Attempt detection of z>15-25 absorption feature to “set clock”
     for interpreting reionization
The end