21 cm tomography and the epoch of reionization

1. 21 cm tomography and the epoch of reionization Jonathan Pritchard (CfA) Collaborators: Steve Furlanetto (UCLA) Avi Loeb (Harvard) Mario Santos (CENTRA - IST) Alex Amblard (UCI) Hy Trac (CfA) Renyue Cen (Princeton) Asantha Cooray (UCI)

2. Overview • 21 cm as probe of high-z radiation backgrounds • Fluctuations in Lya andX-ray backgrounds lead to 21 cm fluctuations • What might 21 cm observations tell usabout first sources? • Reionization • Experimental prospects z~300 z~30 z~12 z~7

3. Ionization history • Gunn-Peterson Trough Becker et al. 2005 • Universe ionized below z~6, some neutral HIat higher z • black is black • WMAP measurement of optical depth Page et al. 2006 WMAP1 WMAP3 WMAP5 Zreion=10.8±1.4 (instant) Zreion~9 ±1.4 (extended)

4. Thermal history • Lya forest Zaldarriaga, Hui, & Tegmark 2001 Hui & Haiman 2003 ?? • IGM retains short term memory of reionization - suggests zR<10 • Photoionization heating erases memory of thermal history beforereionization • CMB temperature • Knowing TCMB=2.726 K and assuming thermal coupling byCompton scattering followed by adiabatic expansion allows informed guess of high z temperature evolution

5. TS Tb Tg HI TK 21 cm basics • Use CMB backlight to probe 21cm transition • HI hyperfine structure n1 11S1/2 l=21cm 10S1/2 n0 z=0 z=13 n1/n0=3 exp(-hn21cm/kTs) fobs=100 MHz f21cm=1.4 GHz • 3D mapping of HI possible - angles + frequency • 21 cm brightness temperature • 21 cm spin temperature Coupling mechanisms: Radiative transitions (CMB) Collisions Wouthuysen-Field

6. Wouthysen-Field effect Hyperfine structure of HI 22P1/2 21P1/2 Effective for Ja>10-21erg/s/cm2/Hz/sr Ts~Ta~Tk 21P1/2 20P1/2 W-F recoils Field 1959 nFLJ Lymana 11S1/2 Selection rules: DF= 0,1 (Not F=0F=0) 10S1/2

7. Experimental efforts MWA: Australia Freq: 80-300 MHz Na= 500 Atot= 0.007 km2 21CMA: China Freq: 70-200 MHz Na= 20 Atot= 0.008 km2 LOFAR: Netherlands Freq: 120-240 MHz Na= 64 Atot= 0.042 km2 SKA: S Africa/Australia Freq: 60 MHz-35 GHz Na= 5000 Atot= 0.6 km2 (f21cm=1.4 GHz)

8. Pixel Scales MWA Z=9 Z=8.5 Z=8 • Better frequency resolution than angular resolution • Bandwidth limited by cosmic evolution • FT data cube to get 3D power spectrum

9. Foregrounds • Many foregrounds • Galactic synchrotron (especially polarized component) • Radio Frequency Interference (RFI) e.g. radio, cell phones, digital radio • Radio recombination lines • Radio point sources • Foregrounds dwarf signal: foregrounds ~1000s K vs 10s mK signal • Strong frequency dependence Tskyn-2.6 • Foreground removal exploits smoothness in frequency and spatial symmetries • Cross-correlation with galaxies also important

10. Global history Furlanetto 2006 Adiabaticexpansion X-rayheating + + Comptonheating Heating expansion UV ionization + recombination HII regions X-ray ionization + recombination IGM Lya flux continuum injected • Sources: Pop. II & Pop. III stars (UV+Lya) Starburst galaxies, SNR, mini-quasar (X-ray) • Source luminosity tracks star formation rate • Many model uncertainties

11. Ionization history zR~7t~0.07 Xi>0.1 • Models differ by factor ~10 in X-ray/Lya per ionizing photon • Reionization well underway at z<12 (Pop II=later generation stars, Pop. III = massive metal free stars)

12. Thermal history

13. Constraining the global signal EDGES • reionization at zR=8 Bowman, Rogers, & Hewitt 2008

14. 21 cm fluctuations W-FCoupling Velocitygradient BaryonDensity Neutralfraction Gas Temperature Brightnesstemperature b Cosmology Reionization X-raysources Lyasources Cosmology Amount ofneutral hydrogen Spin temperature Velocity(Mass)

15. Angular separation? W-FCoupling Velocitygradient BaryonDensity Neutralfraction Gas Temperature b • In linear theory, peculiar velocities correlate with overdensities Bharadwaj & Ali 2004 • Anisotropy of velocity gradient term allows angular separation Barkana & Loeb 2005 • Initial observations will average over angle to improve S/N

16. Temporal separation? W-FCoupling Velocitygradient BaryonDensity Neutralfraction Gas Temperature Brightnesstemperature b Cosmology Reionization X-raysources Lyasources Cosmology Twilight Dark Ages

17. 21 cm fluctuations: Lya Neutralfraction Gas Temperature W-FCoupling Velocitygradient Density b -negligible heating of IGM-tracks density IGM still mostlyneutral Lya flux varies • Lya fluctuations unimportant after coupling saturates (xa>>1) • Three contributions to Lya flux: • Stellar photons redshifting into Lya resonance • Stellar photons redshifting into higher Lyman resonances • X-ray photoelectron excitation of HI Chen & Miralda-Escude 2004, 2006 Chuzhoy & Shapiro 2006

18. Higher Lyman series • Two possible contributions • Direct pumping: Analogy of the W-F effect • Cascade: Excited state decays through cascade to generate Lya • Direct pumping is suppressed by the possibility of conversion into lower energy photons • Ly a scatters ~106 times before redshifting through resonance • Ly n scatters ~1/Pabs~10 times before converting • Direct pumping is not significant • Cascades end through generation of Ly a or through a two photon decay • Use basic atomic physics to calculate fraction recycled into Ly a • Discuss this process in the next few slides… Pritchard & Furlanetto 2006 Hirata 2006

19. Lyman b A3p,2s=0.22108s-1 A3p,1s=1.64108s-1 gg • Optically thick to Lyman series • Regenerate direct transitions to ground state • Two photon decay from 2S state • Decoupled from Lyman a • frecycle,b=0 Agg=8.2s-1

20. Lyman g • Cascade via 3S and 3D levelsallows production of Lyman a • frecycle,g=0.26 • Higher transitions frecycle,n~ 0.3 gg

21. Lyman alpha flux • Stellar contribution continuum injected • Including line width of Ly resonance causes photons to scatter nearersource making Ly flux more inhomogeneous Chuzhoy & Zheng 2007

22. X-rays and Lya production spiE-3 HI HII photoionization e- X-ray collisionalionization e- Lya excitation (fa0.8) HI Chen & Miralda-Escude 2006 Shull & van Steenberg 1985 heating

23. d dV Fluctuations from the first stars • Fluctuations in flux from source clustering, 1/r2 law, optical depth,… • Relate Lya fluctuations to overdensities Barkana & Loeb 2005 • W(k) is a weighted average

24. Determining the first sources da dominates source properties density bias Chuzhoy,Alvarez, & Shapiro 2006 Sources Ja,* vs Ja,X Pritchard & Furlanetto 2007 Spectra aS z=20 D=[k3P(k)/2p]1/2

25. 21cm fluctuations: TK Neutralfraction Gas Temperature W-FCoupling Velocitygradient Density b couplingsaturated IGM still mostlyneutral density + x-rays • In contrast to the other coefficients bT can be negative • Sign of bT constrains IGM temperature Pritchard & Furlanetto 2007

26. Temperature fluctuations TS~TK<Tg Tb<0 (absorption)Hotter region = weaker absorption bT<0 TS~TK~Tg Tb~021cm signal dominated by temperature fluctuations TS~TK>Tg Tb>0 (emission) Hotter region = stronger emission bT>0

27. d dV X-ray heating • X-rays provide dominant heating source in early universe(shocks possibly important very early on) • X-ray heating often assumed to be uniform as X-rays have long mean free path • Simplistic, fluctuations may lead to observable 21cm signal • X-ray flux -> heating rate -> temperature Mpc adiabatic index -1 Barkana & Loeb 2005

28. Growth of fluctuations expansion X-rays Compton temperature fluctuations Heating fluctuations Fractional heating per Hubble time at z dT/ d=

29. TK fluctuations • Fluctuations in gas temperature can be substantial • Amplitude of fluctuations contains information about IGM thermal history

30. TK<Tg Angle averaged power spectrum TK>Tg m2 part of power spectrum Indications of TK dT dominates • Constrain heating transition • Dm2 <0 on largescales indicates TK<Tg(but can have Pdx<0)

31. dx dT da ? X-ray background? • X-ray background at high z is poorly constrained Extrapolating low-z X-ray:IR correlation gives: Glover & Brand 2003 • 1st Experiments might see TK fluctuations if heating late

32. Cosmology? Neutralfraction Gas Temperature W-FCoupling Velocitygradient Density b couplingsaturated IGM still mostlyneutral IGM hotTK>>Tg • Probe smaller scales than CMB • Unless pristine very difficult toimprove on Planck level Kleban, Sigurdson & Swanson 2007

33. Density power spectrum McQuinn et al. 2006 • Moderate improvements - cross-check • Improve ns, S and Wnmost • See baryonic oscillation at few sigma

34. Reionization Neutralfraction Gas Temperature W-FCoupling Velocitygradient Density b Lya coupling saturated IGM hotTK>>Tg HII regions large Z=12.1 Z=9.2 Z=8.3 Z=7.6 Furlanetto, Sokasian, Hernquist 2003

35. Growth of HII Regions Trac & Cen 2006 • (100 h-1 Mpc)3 cube N-body + UV photon ray tracing • Details of heating and coupling are neglected

36. Theory vs Simulation Furlanetto, Zaldarriaga, Hernquist 2004 • Bubbles imprint “knee” on large scales • Limited agreement at end of reionizationon large scales Santos, Amblard, JRP, Trac, Cen, Cooray 2008

37. Temporal evolution Post-reionization Reionization Dark ages Signal from denseneutral clumps at low z Pritchard & Loeb 2008

38.  x  T SKA ? MWA Temporal evolution

39. High-z Observations poor angular resolution foregrounds • Need SKA to probe these brightnessfluctuations • Observe scalesk=0.025-2 Mpc-1 • Easily distinguishtwo models • Optimistic foregroundremoval • B~12MHz ->Dz~1.5

40. Low-z Observations LOFAR MWA SKA • Low k cutofffrom antennaesize • Sensitivity ofLOFAR and MWA drops off by z=12 McQuinn et al. 2006

41. MWA SKA LARC Angular Separation Astrophysics Cosmology Pritchard & Loeb 2008

42. Redshift slices: Lya z=19-20 • Pure Lya fluctuations

43. Redshift slices: Lya/T z=17-18 • Growing T fluctuationslead first to dip inDTb then to double peak structure • Double peak requiresT and Lya fluctuationsto have different scaledependence

44. Redshift slices: T z=15-16 • T fluctuations dominate over Lya • Clear peak-trough structure visible • Dm2 <0 on largescales indicates TK<Tg

45. Redshift slices: T/d z=13-14 • After TK>Tg , thetrough disappears • As heating continuesT fluctuations die out

46. Redshift slices: d/x z=8-12 • Bubbles imprint featureon power spectra once xi>0.2 • Non-linearities in density fieldbecome important (TK fluc. not included)

47. Conclusions • 21 cm fluctuations potentially contain much information about the first sources • Bias and spectrum of sources • X-ray background • IGM temperature evolution • Topology of reionization • Lya and X-ray backgrounds may be probed by future 21 cm observations • Foregrounds pose a challenging problem at high z • Gastrophysics easier to do than cosmology • SKA needed to observe the fluctuations described here • Early days for 21 cm and still unclear what will and will not be possible - foregrounds will be determining factor

49. The first sources 1000 Mpc Hard X-rays Lya 330 Mpc Soft X-rays HII 5 Mpc 0.2 Mpc z=15

50. Spectral Distortion of Lya • Recoil sources featureas photons diffuse in frequency • Colour temperature flux frequency Chen & Miralde-Escude 2004