Transcript History of IGM (C. Carilli)

ionized
History of IGM
C.Carilli (NRAO) Cool Univ
Oct 2004
Neutral
F(HI)=1
Epoch of
Reionization (EoR)
Ionized
F(HI)=1e-5
• bench-mark in cosmic
structure formation
indicating the first
luminous structures
Gunn-Peterson effect
Barkana and Loeb 2001
The Gunn-Peterson Effect
z=5.80
z=5.82
z=5.99
z=6.28
Fast reionization at z=6.3
=> opaque at l_obs<0.9mm
f(HI) > 0.001 at z = 6.3
Neutral IGM evolution (Gnedin 2000): ‘Cosmic Phase transition’ at z=6 to 7
Log (HI
fraction)
Ionizing
intensity
Density
Gas
Temp
8 Mpc
(comoving)
Normalization: GP absorption, LCDM + z=4 LBGs, T_IGM
WMAP Large scale polarization of CMB (Kogut et al.)
20deg
Thompson scattering at EoR
t_e = 0.17 => F(HI) < 0.5 at z=17
Extended period of reionization: z=6 to 15?
Near-edge of reionization: GP Effect
Fairly Fast:
• f(HI) > 1e-3 at z >= 6.4
(0.87Gyr)
• f(HI) < 1e-4 at z <= 5.7
(1.0 Gyr)
Fan et al. 2002
Recombination time vs. Hubble time
z>8: t_rec < t_H
Cen 2002
Stellar fusion produces 7e6eV/H atom, while reionization
requires 13.6eV/H atom =>Need to process only 1e-5 of baryons
through stars to reionize the universe
Complex reionization example: Double reionization? (Cen 2002)
‘normal’ galaxies
(>1e8M_sun)
Pop III stars in
‘mini-halos’
(<1e7 M_sun)
Limitations of current measurements:
CMB polarization:
-- t_e = Ln_es_e = integral measure through universe
=> allows many reionization scenarios
Gunn-Peterson effect:
-- t_Lya >>1 for f(HI)>0.001
-- High z universe is opaque to optical observers
(l_obs<0.9 mm)
The Cool Universe: m/cm/mm probes of the Epoch of
Reionization and the 1st luminous objects
1. CMB large scale polarization
2. Objects within EoR – Molecular gas, dust, star formation
3. Neutral IGM – HI 21cm emission and absorption
Collaborators
USA – Carilli, Walter, Fan, Strauss, Owen, Gnedin
Euro – Bertoldi, Cox, Menten, Omont, Beelen
SKA Key Program science team– Briggs, Carilli, Furlanetto, Rawlings
Science with the Square Kilometer Array (NAR, Carilli & Rawlings)
http://www.aoc.nrao.edu/~ccarilli/CHAPS.shtml
 IRAM 30m + MAMBO: sub-mJy
sens at 250 GHz + wide fields
 IRAM PdBI: sub-mJy sens at 90
and 230 GHz + arcsec resol.
 VLA: uJy sens at 1.4 GHz
 VLA: < 0.1 mJy sens at 20-50 GHz
+ 0.2” resol.
Magic of (sub)mm
L_FIR = 4e12 x S_250(mJy) L_sun for z=0.5 to 8
High redshift QSOs
SDSS + DPOSS:
700 at z > 4
30 at z > 5
9 at z > 6
M_B < -26 =>
L_bol > 1e14 L_sun
M_BH > 1e9 M_sun
York et al 2001; Fan et al
QSO host galaxies – M_BH – s relation
• Most (all?) low z spheroidal galaxies have SMBH
• M_BH = 0.002 M_bulge
‘Causal connection between SMBH and spheroidal galaxy
formation’ (Gebhardt et al. 2002)?
 Luminous high z QSOs have massive host galaxies (1e12 M_sun)
MAMBO surveys of z>2 DPSS+SDSS QSOs
1148+52 z=6.4
1e13L_sun
1048+46 z=6.2
Arp220
• 30% of luminous QSOs have S_250 > 2 mJy, independent of redshift
from z=1.5 to 6.4
• L_FIR =1e13 L_sun = 0.1 x L_bol: Dust heating by starburst or AGN?
L_FIR vs L’(CO)
High-z sources
1e3 M_sun/yr
Index=1
1e11 M_sun
Index=1.7
 M(H_2) = X * L’(CO), X=4 (Milkyway), X=0.8 (ULIRGs)
 Telescope time: t(dust) = 1hr, t(CO) = 10hr
Objects within EoR: QSO 1148+52 at z=6.4
•highest redshift quasar known
•L_bol = 1e14 L_sun
•central black hole: 1-5 x 109 Msun
(Willot etal.)
•clear Gunn Peterson trough (Fan etal.)
1148+52 z=6.42: MAMBO detection
S_250 = 5.0 +/- 0.6 mJy => L_FIR = 1.2e13 L_sun,
M_dust =7e8 M_sun
3’
+
VLA Detection of Molecular Gas at z=6.419
46.6149 GHz
CO 3-2
Off channels
 M(H_2) = 2e10 M_sun
 Size < 1.5” (image)
IRAM Plateau de Bure confirmation
n2
(6-5)
(7-6)
(3-2)
• FWHM = 305 km/s
• z = 6.419 +/- 0.001
• Tkin=100K, nH2=105cm-3
Typical of starburst nucleus
VLA imaging of CO3-2 at 0.4” and 0.15” resolution
rms=50uJy at 47GHz
 CO extended to NW by 1”
(=5.5 kpc) tidal(?) feature
 Separation = 0.3” = 1.7 kpc
 T_B = 20K = T_B (starburst)
 Merging galaxies?
 Or Dissociation by QSO?
1148+5251: radio-FIR SED
S_1.4= 55 +/- 12 uJy
Beelen et al.
T_D = 50 K
1048+46
 Star forming galaxy characteristics: radio-FIR SED, L’_CO/FIR, CO
excitation and T_B => Coeval starburst/AGN: SFR = 1000 M_sun/yr
 Stellar spheroid formation in few e7 yrs = e-folding time for SMBH
=> Coeval formation of galaxy/SMBH at z = 6.4 ?
1148+52: Masses
•M(dust) = 7e8 M_sun
•M(H_2) = 2e10 M_sun
•M_dyn (r=2.5kpc) = 5e10 M_sun
•M_BH = 3e9 M_sun => M_bulge = 1.5e12 M_sun
• Gas/dust = 30, typical of starburst
• Dynamical vs. gas mass => baryon dominated?
• Dynamical vs. ‘bulge’ mass => M – s breaks-down at
high z?
Cosmic (proper) time
1/16 T_univ = 0.87Gyr
1148+52: Timescales
• Age of universe: 8.7e8 yr
• C, O production (3e7 M_sun): 1e8 yr
• Fe production (SNe Ia): few e8 yr (Maiolino, Freudling)
• Dust formation: 1.4e9yr (AGB winds) => dust formed in high
mass stars/SNR (Dunne 03; Maiolino 04)? => silicate grains?
=> Star formation started early (z > 10)?
Cosmic Stromgren Sphere
• Accurate redshift from CO: z=6.419+/0.001
Ly a, high ioniz. Lines: uncertainty >1000km/s (Dz=0.03)
• Proximity effect: photons leaking from 6.32<z<6.419
White et al. 2003
z=6.32
•‘time bounded’ Stromgren sphere: R = 4.7 Mpc
t_qso= 1e5 R^3 f(HI)= 1e7yrs
Richards et al. 2002
SDSS QSOs
Loeb & Rybicki 2000
Constraints on neutral fraction at z=6.4
 GP => f(HI) > 0.001
 If f(HI) = 0.001, then t_qso = 1e4 yrs – implausibly short given
fiducial lifetime, f_lt = 1e7 years?
 Probability arguments suggest: f(HI) > 0.1 at z=6.4 – much
better limit than GP
Wyithe and Loeb 2003
z>6 QSOs with MgII and/or CO redshifts (Walter et
al, Willot et al., Maiolino et al., Iwamuro et al.)
<Dz> = 0.08 => <R> = 4.4 Mpc
Near-edge of reionization: GP + Cosmic Stromgren Spheres
Very Fast?
• f(HI) > 1e-1 at z >= 6.4
(0.87Gyr)
• f(HI) < 1e-4 at z <= 5.7
(1.0 Gyr)
See also Cosmic Stromgren
Surfaces (Mesinger & Haiman
2004)
Gas and dust during the EoR
• FIR luminous galaxy at z=6.42: 1e13 Lsun
observe dust, gas, star formation, AGN
• Merging(?) galaxy: Molecular gas mass =
2x1010 M_sun, M_dyn = 6e10 M_sun
• Early enrichment of heavy elements
and dust produced in the first
stars => star formation commenced at
0.4 Gyr after the big bang
• Coeval formation of SMBH + stars in
earliest galaxies – break-down of M-s at
high z?
• Cosmic Stromgren sphere of 4.7 Mpc =>
‘witnessing process of reionization’
t_qso = 1e7 * f(HI) yrs
‘fast’ reionization: f(HI)>0.1 at z=6.4?
J1048+4637: A second FIR-luminous QSO source at z=6.2
MAMBO 250 GHz
VLA CO 3-2
S_250 = 3.0 +/- 0.4 mJy =>
L_FIR = 7.5e12 L_sun
z(opt)
z(MgII)
S(CO 3-2) = 0.17 +/- 0.09
mJy
EVLA correlator: 8GHz,
16000 channels
VLA detections of HCN 1-0 emission
n(H_2) > 1e5 cm^-3 (vs. CO: n(H_2) > 1e3 cm^-3)
z=4.7
z=6.4
index=1
Solomon et al
z=2.58
70 uJy
PKS 2322+1944 z=4.12: [CI] (492 GHz rest freq; Pety et al.)
VLA CO2-1
PdBI
=> Solar Metalicity
Continuum sensitivity of future arrays: Arp 220 vs z (FIR = 1.6e12 L_sun)
cm: Star formation,
AGN
(sub)mm Dust,
molecular gas
Near-IR: Stars,
ionized gas, AGN
Redshifts for obscured/faint sources: wide band (16 - 32 GHz)
spectrometers on LMT, GBT (Min Yun 04, Harris 04)
L_FIR = 1e13 L_sun
Studying the pristine IGM beyond the EOR: redshifted HI
21cm observations (100 – 200 MHz) with the Square Kilometer Array.
‘Pathfinders’: LOFAR, MWA, PAST, VLA-VHF,…
SKA goal: mJy at 200 MHz
Large scale structure:
density, f(HI), T_spin
Low frequency background – hot, confused sky
Eberg 408 MHz Image (Haslam + 1982)
Coldest regions: T = 100 (n/200 MHz)^2.6 K
Highly ‘confused’: 3 sources/arcmin^2 with S_0.2 > 0.1 mJy
Terrestrial interference
100 MHz
z=13
200 MHz
z=6
Global reionization signature in low frequency HI spectra
(Gnedin & Shaver 2003)
fast
21cm ‘deviations’ at
1e-4 wrt foreground
double
Spectral index
deviations of
0.001
HI 21cm Tomography of IGM
Zaldarriaga + 2003
z=12
9
DT_B(2’) = 10’s mK
SKA rms(100hr) = 4mK
LOFAR rms (1000hr) = 80mK
7.6
Power spectrum analysis
Zaldarriaga + 2003
Z=10
129 MHz
LOFAR
SKA
2deg
1arcmin
Cosmic Web after reionization = Ly alpha forest (d <= 10)
1422+23 z=3.62 Womble et al. 1996
N(HI) = 1e13 -- 1e15 cm^-2, f(HI/HII) = 1e-5 -- 1e-6
=> Before reionization N(HI) =1e18 – 1e21 cm^-2
Cosmic web before reionization: HI 21cm
Forest (Carilli, Gnedin, Owen 2002)
20mJy
Z=10
TCMB 1 + z 1/ 2
t  0.008(
)(
) f HI (1 + d )
TS
10
• Mean optical depth (z = 10) = 1% = ‘Radio GunnPeterson effect’
• Narrow lines (t= few %, few km/s) = HI 21cm
forest (d <= 10), 10/unit z at z=8
• Mini-halos (d >= 100) (Furlanetto & Loeb 2003)
• Primordial disks: low cosmic density=0.001/unit z,
but high opacity=> fainter radio sources -- GRBs?
Radio sources beyond the EOR?
• Radio loud QSO fraction = 10% to z=5.8 (Petric +
2003)
• Models => expect 0.05 to 0.5 deg^-2 at z> 6 with
S_151 > 6 mJy, out of 100 total (Carlli,Jarvis,Haiman)
Z=8
GMRT 228 MHz search for HI21cm abs toward
highest z radio galaxy, 0924-220 z=5.2
8GHz
1”
Van Breugel et al.
z(CO)
230Mhz
Continuum point source = 0.55 Jy;
rms/(40km/s chan) = 5 mJy
‘EoR Pathfinders’: PAST, LOFAR, MWA, VLA-VHF, …
MWA prototype
(MIT/ANU)
LOFAR (NL)
PAST (CMU/China)
VLA-VHF
(CfA/NRAO)
VLA-VHF: 180 – 200 MHz Prime focus X-dipole
(CfA/NRAO – Greenhill et al)
Leverage: existing telescopes,
IF, correlator, operations
First light: Q4, 2005
Main Experiment: Cosmic Stromgren spheres around
z>6 SDSS QSOs (Wyithe & Loeb 2004)
20mK
VLA-VHF
190MHz
250hrs
15’
0.50+/-0.12 mJy
VLA spectral/spatial resolution well matched
to expected signal: 5’, 1000 km/s
Radio astronomy – Probing the EoR
•Study physics of the
first luminous sources
(limited to near-IR to
radio wavelengths)
•Currently limited to
pathological systems
(‘HLIRGs’)
•EVLA, ALMA 10-100x
sensitivity is critical to
study normal galaxies
•Low freq pathfinders:
HI 21cm signatures of
neutral IGM
•SKA imaging of IGM
z6.4
Other Experiments: power spectrum analysis, ‘HI 21cm forest’
Sensitivity per 0.8MHz channel: currently
have 16 channels over 12.5 MHz
Piggy-back on CSS experiment
Centrally condensed uv coverage
Challenges and ‘mitigation’: VLA-VHF CSS

Ionospheric phase errors – higher freq (freq^-2); 4deg FoV; 1km B_max

T_bg – higher freq (freq^-2.75)

Confusion (in-beam) – spectral measurement (eg. Morales & Hewitt 2004);
mJy point source removal w. A array; precise position and redshift

Wide field problems – polarization, sidelobes, bandpass – all chromatic ?

RFI – “interferometric excision” (but D array); consistently ‘clean’ times in
monitor plots (but very insensitive measure) ?
Proposed Cost and Timeline
 100K in parts (CfA) + labor (CfA/NRAO)
 First tests (4 prototypes): Q1, Q2 2005
 First experiments (100-200 hr): D array, Q4 2005
 Large proposal (500 hr): D array, Q1 2007
System/Site characteristics
Work hours
First sidelobe = 15db
Proposed
band
TV
carrier
Ionospheric phase errors: VLA 74MHz (Lane etal.)
 TIDs – ‘fuzz-out’ sources
 ‘Isoplanatic patch’ = few deg = few km
 Phase variation proportional to wavelength^2
SKA timeline
•2004
Science case: “Science with the SKA” Carilli & Rawlings, New Astron. Rev.
•2004-7 demonstrator development
major external review (2006)
submit funding proposals for a 5% demonstrator
•2006
site selection: Australia, USA-SW, South Africa, China
•2008
selection of technical design (may be a combination);
start construction of 5% demonstrator on chosen site
•2009
submit funding proposals for full array
•2012
start construction
•2020
complete construction
Projected cost: 1 G$
(20% for low freq only)
Weak correlation of L_FIR – M_B?
•M_B > -26: 10% detected at 250 GHz mJy sensitivity
•M_B < -26: 30% detected
Thermal State of IGM at high z: Ly a Forest
(Hui and Haiman 2003)
White etal (2002): ‘superluminal’ ionization front
 Stromgren sphere expands at close to speed of light => obs Ly a
photons emitted just after ionizing photons
 “Delay required to allow light to travel from source to the edge of
the sphere is exactly compensated by the ‘speedup’ that results from
that edge being closer to the observer”
 “Expansion law for the observed radius is exactly the same as the
expansion law derived ignoring light-travel effects”
Confusion by free-free emission during EOR
(Oh & Mack 2003)
Difficulty with (LSS) emission observations:
confusion by foreground radio sources (di Matteo 2001)
Beating confusion: exploiting the spectral domain
(Morales & Hewitt 2004)
Foreground: smooth continuum
=> cylindrical symmetry in
Fourier space
EoR HI: noise signal in
frequency => spherical
symmetry in Fourier space
1148+52: starburst+AGN?
S_1.4= 55 +/- 12 uJy
IRAS 2Jy sample (Yun+)
1048+46
1148+52
SFR(>5 M_sun) = 1400 M_sun/year => host spheroid formation in
5e7 yrs at z > 6?
 SMBH formation: n x 2.4e7 yr (Loeb, Wyithe,…)
=> Coeval formation of galaxy/SMBH at z>6?
Gravitational Lensing?
 CO 3-2 double source, 0.3” separation => strong lensing?
 Keck near IR imaging: point source < 0.3” at K (Djorgovski)
 HST/ACS imaging: point source < 0.1” (White et al. 2004 (?))
 Radio continuum: foreground radio sources, but no SDSS cluster z< 0.1
 QSO spectrum => “proto-cluster” at z=5 (White et al. 2002)?
1148+5251
Cosmic Stromgren Sphere
• Accurate redshift from CO: z=6.419+/0.001
Ly a, high ioniz. lines uncertainty >1000 km/s (Dz=0.03)
• Proximity effect: photons leaking from 6.32<z<6.419
•‘time bounded’
Stromgren sphere:
R = 4.7 Mpc
White et al. 2003
z=6.32
t_qso= 1e5 R^3 f(HI)
= 1e7yrs
ALMA/EVLA/GBT redshift coverage for CO
Epoch of Reionization
VLA CO(3-2), PdBI
CO 6-5, 7-6 in
J1148+5251 @ z=6.42
Radio sources beyond the EOR?
• Radio loud QSO fraction = 10% to z=5.8 (Petric + 2003)
• Models => expect 0.05 to 0.5 deg^-2 at z> 6 with S_151 > 6
mJy (out of 100 total)
1.4e5 at z > 6
S_151 > 6mJy
2240 at z > 6
Carilli + 2002
Haiman &
Hui 2004
1148+5251: Starburst + QSO at z=6.42?
 Starburst (double?) nucleus: size = few kpc around QSO
 Star formation rate => stellar spheroid formation in 5e7 yrs?
 SMBH formation: 1/e = few x10^7 yr (Loeb, Wyithe,…)
=> Coeval formation of galaxy/SMBH at z>6?
Temperatures: Spin, CMB, Kinetic and the 21cm signal
Dt = 10mK
z = 11
z=7
Tozzi + 2002
T_s
T_CMB
T_K
•Initially T_S= T_CMB
•T_S = T_CMB => no signal
•T_S couples to T_K via Lya scattering
•T_S = T_K < T_CMB => Absorption
against CMB
•T_K = 0.026 (1+z)^2 (wo. heating)
•T_CMB = 2.73 (1+z)
•T_S > T_CMB => Emission
Phase stability: Fast switching at the VLA
10km baseline rms = 10deg
Structure formation: the Dark Matter
perspective = Press-Schechter Formalism
z
M_2s T_vir
M_sun
K
0
1e14
3e7
5
3e10
3e5
10 6e7
8e3
GMRT 230 MHz 0924-220 z=5.2
channel 20 (229.60MHz)
Cosmic Stromgren Surfaces: damping wing of Ly a line at
sharp edge of sphere leads to apparently smaller sphere size
(Mesinger & Haiman 2004)
 Dark matter: Analytic – “Press-Schechter Formalism”
(Barkana & Loeb 2000 Rev Mod Phys)
 IGM/galaxy formation: Numerical simulations
Structure formation: the Baryons
Minihalos:
• M_’Jeans’ = 1e4 M_sun (z=20)
• H_2 cooling: 1e5 – 1e7 M_sun
=>T_vir = 300 to 1e3 K
• H_2 formation: Near UV
dissociates, but soft Xray catalyze?
• Form 100 M_sun stars (popIII)?
• Totally disrupted by single SNe =>
self-distruct in 1e6 years
Protogalaxies: H line cooling =>
T_vir > 1e4 K
High redshift QSOs
SDSS + DPOSS:
Fan et al 2000
700 at z > 4
30 at z > 5
9 at z > 6
M_B < -26 =>
L_bol > 1e14 L_sun
M_BH > 1e9 M_sun
Luminous “SDSS” QSOs: insufficient to
reionization the universe
z=10 lensed star forming galaxy? (Pello 2004)
L_app= 4e11 L_sun + LBG dust correction (5x) => L_FIR = 2e12L_sun
S_250 = 0.6 mJy => 4s ALMA detection in 1 minute!
Expect 1 – 2 “normal” galaxies ALMA FoV at z>6 per in 10hrs
Ly alpha emitting galaxies within EoR: near-IR observations (Hu,
Taniguchi, Stanway, Bouwens, Yan, Dickinson…)
z=6.56
 1 arcmin^2 (H_AB = 27)
 0.3 L*(z=3)
 SFR = few M_sun/year