Probing the Cosmic Evolution with the Most Distant Quasars

1
Probing the Cosmic Evolution with the Most
Distant Quasars

• Xiaohui Fan
• Steward Observatory
• The University of Arizona

2
Introduction

• The Study of Highest-Redshift Quasars Probes
• The epoch of first generation of galaxies/quasars
• Models of black hole formation
• Role of quasar/BH activity in galaxy evolution
• State of intergalactic medium
• Ionizing background at high-z
• History of reionization ? probing the end of
  cosmic dark ages

3
SDSS High-redshift Quasar Survey

• Zgt4 300 from SDSS
• Zgt5 21 from SDSS
• Zgt6 5
  

Total Discoveries
SDSS Discoveries
4
Outline

• SDSS Quasar Survey
• Search for the First Quasars
• Co-formation of First Quasars and Galaxies
• Lyman Absorption of z6 quasars
• Discovery of complete Gunn-Peterson trough
• Implication on the epoch of reionization
• Collaborators Strauss, Schneider, Becker,
  White, Richards, Penterricci, Rix, Narayanan et
  al.

5
SDSS Overview

• Primary Telescope 2.5m wide-field (2.5 deg)
• Imaging Survey (wide-field 54 CCD imager)
• Main Survey 10000 deg2
• Five bands, 3000 10000 Å
• rlim 22.5, zlim 20.5
• Spectroscopic Survey
• 106 galaxies (rlt17.8)
• 105 quasars ( 0 lt z lt 6.5)
• Interesting stars, radio/x-ray sources etc.

6
SDSS Quasar Survey

• Color selected, flux-limited sample of 100,000
  quasars over 10,000 deg²
• Fully automated pipeline selection up to z5.5
• z band (9000 Å) allows detection of quasars up to
  redshift of 6.5
• Progress 30,000 quasars discovered from SDSS
  data

Stellar locus
Z3
Z4
quasar
Z5

Fan, Richards, Newberg, Strauss
7

8
SDSS at Your Service Data Release 1
9
SDSS Data Release 1

• Websitewww.sdss.org/dr1
• Photometry
• 2100 sq. deg
• 53 million objects in u,g,r,i,z
• 2.3TB data
• Limiting mag 22.2 in r, 20.5 in z
• Spectroscopy
• 1550 sq. deg
• 186,259 objects
• 134,000 galaxies
• 18,600 quasars
• R1800, 3800-9200 A, 30km/s redshift accuracy

10
Evolution of Quasar Luminosity Function
SFR of Normal Gal
Exponential decline of quasar density at
high redshift, different from normal galaxies


11
Evolution of Quasar Luminosity Function
SFR of Normal Gal
Exponential decline of quasar density at
high redshift, different from normal galaxies


12
Clustering of Quasars

• What does quasar clustering tell us?
• Bias factor of quasars ? average DM halo mass
• A biased large scale power spectrum at high-z
• Combining with quasar density ? quasar lifetime

Zlt2.5 SDSS result Van den Berk et al.
13
Evolution of Quasar Clustering
Fan et al. in preparation
14
Evolution of Quasar Clustering

• At zlt2 r0 7 Mpc/h
• Slow evolution with redshift
• Might be function of luminosity
• At z4
• Quasars strongly clustered (r0 20 Mpc/h)
• Stronger than the clustering of low-z quasars or
  high-z LBGs
• Luminous high-z quasars are in massive system and
  represent very rare peaks of the density field
• Quasar correlation consistent with a quasar halo
  mass of 1012 1013 Msun, and a short life-time
  of 107-8 years.

15
Whats Next A Faint High-redshift Quasar Survey?

• Limitations of current generation quasar surveys
• Only probing the most luminous quasars ? most
  massive BHs and galaxies majority of high-z
  quasars have not been detected!
• Faint end evolution of quasars unknown total
  contribution to UV background uncertain
• Very sparsely sampled ? few pairs, difficult for
  large scale structure studies
• Faint quasar survey
• Probing normal BHs in average galaxies
• Understand quasar contribution to the ionizing
  background
• Quasar power spectrum at zgt3
• Providing close pairs and multiples to map out
  high-redshift IGM and use for cosmological test?
• Possible with new generation wide-field imaging
  and multi-object spectroscopy on 4-8 m class
  telescopes

16
Search for the First Quasars
Fan, Narayanan, Lupton, Strauss et al.

• Color selection of i-drop
  out quasars
• At zgt5.5, Lya enters z-band ? quasars have only
  red i-z measurement faint objects z-band only
  detections
• Technical Challenges
• Rare objects ? contaminant elimination
• Elimination of false z-band only detections ?
  improved cosmic ray rejection
• Reliability of faint z photometry ? follow-up
  high S/N z photometry
• Major contaminants are L and T type Brown Dwarfs
  ? additional IR photometry

17
Search for the First Quasars
Fan, Narayanan, Lutpon, Strauss et al.

• Separating z6 quasars and BDs
• Follow-up IR photometry
• For quasar z-J 1
• For late-L to T
• z-J gt 2

Zgt5.7 quasar
18
Z6 quasars from SDSS

• Survey area 3000 deg2 zABlt20 i-dropout object
  search for both brown dwarfs and high-z quasars
• Large number of L dwarfs and 20 T dwarfs
  discovered
• Establish L/T spectral and temperature sequence
• Complete sample of brown dwarfs LF and mass
  density
• Ten luminous quasars at zgt5.7
• Discovered at APO 3.5m and Magellan
• High S/N, high resolution spectroscopy at
  Keck/VLT/HET/MMT
• Four new objects in 2002 z 5.85, 6.05, 6.2 and
  6.4 (current redshift holder)
• Two new ones this season so far at z5.92 and
  z6.07
• More to come, stay tuned

19
New zgt6 Quasars from the SDSS
z6.1
z6.2
z6.4
20


21
Quasar Density at z6

• Based on seven zgt5.7 quasars
• Density declines by a factor of 20 from z3
• Number density implies that quasars are unlikely
  to provide enough UV background if LF is similar
  to that at low-z
• Cosmological implication
• MBH109-10 Msun
• Mhalo 1013 Msun
• How to produce such massive BH in less than
  1Gyr??
• They are the highest peak of density field and on
  the tail of mass function, might provide
  sensitive test to structure formation and BH
  evolution
• Need to Find Faint Ones

Fan et al. 2001
22
Metals at z6.3

• Strong metal lines detected ? strength comparable
  to low-redshift quasars
• NV/CIV ratio metallicity
• NV/CIV 0.7 ? Supersolar metallicity for z6.28
  quasar (t0.8Gyr) ?
• Early production of heavy element and
  multi-generation of stat formation

Keck ESI/NIRSPEC
Keck ESI/ NIRSPEC
Fan, Narayanan, Lupton, Strauss et al.
23
Iron at zgt6

• Fe is mostly made out of Type Ia SN, 1Gyr delay
• Existence of Fe in high-z quasar would indicate
  star-formation started hundred Myrs ago ? relics
  of the first stars???

Freduling et al. (2003) NICMOS spectra of
three z6 SDSS Quasars
24

X-ray Observation of the Highest-redshift
Quasars

• First exploration of X-ray universe at zgt5
• 30 SDSS quasars at zgt4.5 observed with
  Chandra/XMM (5-10ks, Brandt et al.)
• most detected
  
  
• 10 SDSS objects detected at zgt5 (out of 12 total
  in literature)
• opt/X flux ratio weakly dependent on redshift

Tentative evidence that
quasars at zgt4 X-ray fainter than at z2
? Presence of large amount of gas??
Average slope at z2
Average slope at z4.6
Average slope at z4.5

25
X-ray Observations of high-redshift Quasars

• X-ray emission appears to be a universal property
  of luminous AGNs even at z6
• At zgt4, optical-to-x ray energy distribution is
  slightly steeper than at zlt2 luminosity effect
  or presence of gas???

26
Sub-mm and Radio Observationof High-z Quasars

• A 250 GHz and 1.4GHz survey of 41 quasars at
  zgt3.6 (Carilli, Rupen, Fan, Strauss et al.)
• 16/41 detected in 250GHz (rest-frame sub-mm)
  brighter than 1mJy
• Combination of cm and submm ? submm radiation in
  radio-quiet quasars come from thermal dust with
  mass 108 Msun
• If dust heating came from starburst ? star
  forming rate of 500 2000 Msun/year ?Quasars are
  likely sites of intensive star formation

Arp 220
27
Detection of zgt6 quasars in submm

• Two zgt6 quasars detected in IRAM recently z6.4
  (5mJy) and z6.2 (3mJy)
• Implied FIR luminosity 1013 solar luminosities
• Implied star formation rate 103 solar masses per
  year
• Next line (CO, CII) and SIRTF (FIR SED) searches

IRAM observation of z6.4 quasar 5 mJy at 1.2mm
Bertoldi et al. 2003
28
Co-evolution of early galaxies and supermassive
BHs

• Presence of 109-10 solar mass BH at zgt6 ?
  it has to begin the assemble at zgt10
• High metallicity in the quasar environment ?
  recent star formation and chemical enrichment
• Presence of heated dust (submm) and gas (at three
  out of eight zgt5.7 quasars seem to show strong
  associated absorption) ? possibly on-going star
  formation with rate of 1000 solar mass/year

29
Searching for Gunn-Peterson Trough

• Gunn and Peterson (1965)
• It is observed that the continuum of the source
  continues to the blue of Ly-a ( in quasar 3C9,
  z2.01)
• only about one part of 5x106 of the total mass
  at that time could have been in the form of
  intergalactic neutral hydrogen
• Absence of G-P trough ? the universe is still
  highly ionized

30
reionization
From Avi Loeb
31
A brief cosmic history

• recombination

• Cosmic Dark Age no light
• no star, no quasar IGM HI

• First light the first galaxies
• and quasars in the universe

• Epoch of reionization radiation from the first
  object lit up and ionize IGM HI ? HII

? reionization completed, the universe is
transpartent and the dark ages ended
? today
32
The end of dark ages Movie
33
Cold gas
Light background
Gnedin 2000
Gas density
Gas temperature
34
Cold gas
Light background
Gas density
Gas temperature
Gnedin 2000
35
Life as a Hydrogen atom at the end of cosmic dark
ages
36
5
Increasing Lya absorption with redshift
zabs fobs/fcon -------------------------
-- 5.5 0.10 5.7
0.05 6.0 lt0.006 Zero flux over
300Å immediately blueward of Lya emission
in z6.28 quasar ? Detection of complete
Gunn-Peterson Trough tgtgt1 over large
region of IGM
Becker, Fan, White et al.
37
VLT/FOS2
Detection of A Complete Gunn-Peterson Trough
VLT observation
T-0.0010.003
Pentericci et al.
Pentericci, Fan, Rix et al. 2001
38
Keck/ESI 30min exposure ?
Gunn-Peterson Trough in z6.28 Quasar
Keck/ESI 10 hour exposure ?
39
Keck/ESI spectrum
G-P Trough
G-P Trough

• 10 hour Keck exposure
• shows very deep G-P trough
• Dectection of G-P troughs
• for Lyß, ?
• Need the second quasar
• to confirm!!
• what does this mean to
• reionziation ???

40
Gunn-Peterson troughs confirmed by new zgt6 quasars
41
Z6.4 quasar (6 hours on Keck)

• Weak, but significant flux inside the trough ?
  highly ionized bubble by intervening galaxy or
  local void?
• Z6.37 by fitting weak metal lines ? presence of
  neutral gas in the quasar vicinity?
  

42
Strong Evolution ofGunn-Peterson Optical Depth
Transition at z5.5?
White et al. 2003
43
Implications of Complete Gunn-Peterson Trough

• G-P optical depth at z6
• Small neutral fraction needed for complete G-P
  trough
• By itself not indication that the object is
  beyond the reionization epoch
• For uniform IGM
• Measurement of optical depth can be used to
  constrain ionizing background
• IGM is highly non-uniform
• regions with different density have different
  Lya transmission
• to constrain ionization state have to take
  into account the density distributions of the IGM

44
Evolution of Ionizing Background

• Ionizing background estimated by comparing with
  cosmological simulations of Lyman absorption in a
  LCDM model
• Stronger constraint from the Lyß and Ly?
  Gunn-Peterson trough
• Ionizing background declines by a factor of gt25
  from z3 to z6
• Indication of a sudden change at z6?

mass-averaged
volume-averaged
Fan, Narayanan, Strauss et al. 2002
45
Constraining the Reionization Epoch

• Neutral hydrogen fraction
• Volume-averaged HI fraction increased by gt100
  from z3 to z6
• Mass-averaged HI fraction gt 1
• Gunn-Peterson test only sensitive to small
  neutral fraction and saturates at large neutral
  fraction
• Mean free path of ionizing photons
• Also decreases sharply towards high-z
• At z6
• Last remaining neutral regions are being ionized
• The universe is 1 neutral
• Marks the end of reionization epoch??

Mass-averaged
Volume-averaged
Fan et al. 2002
46
Constraining Reionization Epoch

• Critical overdensity ?i
• Reionization progresses from voids to overdense
  regions
• Regions with ? gt ?i remain neutral
• At z6 moderately dense regions in the IGM
  remained neutral
• Mean free path of ionizing photons
• Also decreases sharply towards high-z
• At z6
• Last remaining neutral regions are being ionized
• The universe is 1 neutral
• Marks the end of reionization epoch??

47
Three stages
Pre-overlap
Overlap
Post-overlap
From Haiman Loeb
48
Mass-averaged
postoverlap
Reionization epoch
Pre- overlap
Volume-averaged
49
Mass-averaged
postoverlap
Reionization epoch
Pre- overlap
Volume-averageed
50
Reionization History Combining GP test with CMB

• G-P test shows at z6, the IGM is about 1
  neutral ? the tail end of the reionization
  process
• Discovery of three G-P troughs in the three
  highest redshift quasars known ? end of
  reionization at z6 with small dispersion among
  different lines of sight
• CMB polarization shows substantial ionization by
  z17
• sensitive to the beginning and mid-point of
  reionization
• These two measurements are not contradictory
  combining GP with CMB ? reionization history
• First star formation happens very early?
• Reionization last from 20 to 6? (600 million
  years)
• Reionization is not a phase transition
• Reionization seems to be more complicated by the
  simplest theory

51
Did the Universe Reionize Twice?

• Challenge
• How can first star formation happen so early
• Why does reionzation epoch last so long?
• Cen (2002), Wyithe and Loeb (2003) propose twice
  reionization
• At z20 population III stars formed by H2
  reionized the universe
• IGM heats up and raises Jean mass ? star
  formation stops and HII recombines to form the
  second dark age
• z6 pop II forming with increasing global star
  formation rate resulting in the second
  reionization

52
Probing the Reionization Epochbefore JWST

• More lines of sight from high-redshift quasars
• About 20 40 quasars 6 lt z lt 6.6 from SDSS
• Spectroscopy in far-optical and near-IR with
  large telescopes
• G-P test probing non-uniformity of the end of
  reionization
• Metal absorption lines at zgt6
• metals from early objects
• Imprints of the earliest star formation (Pop III)
• Looking for objects at zgt7 first sources of
  light
• z6.6 is the limit of optical survey ? IR
  searches
• Search for high-z galaxies lensed by massive
  clusters in near-IR
• High-redshift GRBs
• 21cm mapping of reionization
• Understanding the topology of the reionization
  from multiple lines of sights
• Searching for tests sensitive to large neutral
  fraction

53
Need high S/N, high resolution spectroscopy of
new zgt6 quasars
54
Probing the first Metals in the IGM?
10 hour Keck/ESI

• 5 10 hour exposure in the far-optical and
  
• near-IR of z6 quasars on 6 10m class telescope
  
• High S/N to detect intergalactic CIV and SiIV
  absorption
• To probe the metal density of the IGM at zgt5
  

55
Probing the first metals?
56
Summary

• High-redshift quasars evolve strongly with
  redshift
• Density declines by 20 from z3 to z6
• Evolution much faster than normal galaxies
• High-redshift quasars are sites of spectacular
  star formation
• Sub-mm detections and possible X-ray weakness
• Possible supersolar metallicity at zgt6 in quasar
  environment
• High-redshift quasars probe the end of
  reionization epoch
• Lya absorption increases dramatically at zgt5.7
• Consistent detections of complete Gunn-Peterson
  troughs in the highest-redshift objects
• At z6 ionizing background much lower, neutral
  fraction gt1, moderately overdense regions still
  neutral
• ? it marks the end of the reionization
  epoch when the last
  remaining HI in the IGM is being
  ionized
• ? combining with CMB results
  revealing the reionization history