The High-Redshift Universe

1
The High-Redshift Universe

• Alberto Fernández-Soto
• Universitat de València

2
Plan of this talk

• Definition and evolution of high-redshift
• Objects at high redshift
• Selection techniques for high-z objects
• Observations
• Towards galaxy evolution

3
How high is high?

• Start with the discovery and identification of
  QSOs in the 60s (3C273 z0.16)
• For 30 years monsters dominated (z 4)
• Radiogalaxies -- Quasars
• Colour selection drove towards normal galaxies
• LBGs _at_ z 3 -- Dropouts _at_ z 3 5
• QSOs again pushed in (SDSS z 6)
• GRBs stand a chance (GRB050904 _at_ z6.29)
• but not as high as z1000!

4
(No Transcript)
5

• Data compiled by X. Fan

6
Search for high-z objects

• Identification of peculiar objects
• X-ray sources ? AGNs ? Clusters of
  galaxies ? Peculiar stars (or)
• Radio sources ? Radio-loud QSOs
• Important selection effects
• Plagued QSO catalogues for years
• Non-detectability of normal galaxies
• By definition!

7
Colour selection

• Use of hydrogen-imprinted features
• Peebles 1967, Partridge (1974), BVR selection

(Piskunov y Kupka 2001)
8
Intergalactic Absorption
quasar
Charlton Churchill (2000)
z 1.3
Charlton Churchill (2000)
z 3.6
z 6.3
Becker et al. (2001)
9
The perfect QSO spectrum
10
(Steidel 1999)
11
NOAODWFS(Jannuzzi et al)
12
The importance of it all

• Neutral hydrogen is ubiquitous
• The features imprinted by HI are the same, for
  all types of objects, at a given redshift
• ? Possibility of unbiased selection

13
Colour-selected normal galaxiesin the HDF

• (Lanzetta et al 1996)

14

• (Giavalisco 2001)

15
Expected colors of high z Lyman break galaxies
are well defined.
(Steidel et al 1999)
16
(Steidel et al 1999)
17
(Le Fevre et al 2005)
18
(Le Fevre et al 2005)
19
Quasars re-enter the stage

• Colour selection can also be applied to QSOs
• High-redshift QSOs much more luminous than
  galaxies, but much more scarce
• Need for photometric, large-area surveys
• SDSS ? ugriz colour images ? 10 000 square
  degrees ? selection of QSOs up to z gt 6

20

17,000 Quasars from the SDSS Data Release One
5
Ly a
3
2
CIV
CIII
1
MgII
OIII
Ha
0
4000 A
9000 A
21
The Highest Redshift Quasars Today(SDSS, Fan
et al)

• zgt4 700 known
• zgt5 30
• zgt6 7
• SDSS i-dropout Survey
• By Spring 2004 6000 deg2 at zABlt20
• Fourteen luminous quasars at zgt5.7
• 20 40 at z6 expected in the whole survey

Total Discoveries
SDSS Discoveries
22
(No Transcript)
23
Photometric redshifts

• Originally in Baum (1963)
• Koo (1985) Poor mans redshift machine
• Loh Spillar (1986) Cosmology
• Impulse with Hubble Deep Fields
• Deep, high quality, multi-colour images
• High spatial resolution
• Good spectral sample for calibration
• Nowadays common tool of the trade

24
Applications

• Poor mans redshift machine
• Applications in (particularly)
• Large volumes of data
• Faint galaxy samples
• Crowded fields
• Proven accuracy Dz/(1z) 0.05 (rms)
• Catastrophic error rate lt 5
• Already tested out to z 6

25
Sbc z0.66
26
Sbc z1.29
27
Irr z2.51
28
SB1 z4.36
29
SB1 z7.96
30
Z(phot) vs Z(spec)
(Fernandez-Soto et al 2001)
31
Colour selection grism spectra

• z5.83
• Selected inside HUDF via BVizJH photometry
• (Malhotra et al 2005)

32
Emission-line searches

• Emission line searches for high redshift galaxies
  (tuned to Lyman-a) existed for many years, with
  very low success rates
• Only over the last few years narrow-band searches
  tuned at z6 have offered results
• Particular mention to
• LALA
• Keck
• Subaru

33
Subaru emission-line searches

• NB711 _at_ z4.86 (Masami et al 2003)
• NB816 _at_ z5.75 (Ajiki et al 2003)
• NB921 _at_ z6.55 (Kodaira et al 2003)

34
Subaru emission-line search

• Narrow-band filter tuned to Lyman a _at_ z6.55
• (Kodaira et al 2004)

35
Keck emission-line survey(Hu et al 2005)
36
The Large Area Lyman Alpha Survey
z O Volume Sensitivity Candidates, Spectroscopic Success rate
4.5 1.4x106 Mpc3 (1/2 in Bootes) 1.7x10-17 ergs/s/cm2 350 gt 70
5.7 6 x105 Mpc3 (1/3 in Bootes) 1x10-17 ergs/s/cm2 50 70
6.6 1.5x105 Mpc3 (all in Bootes) 2x10-17 ergs/s/cm2 3 1 of 3 confirmed.
37
Candidate z6.5 LALA galaxies
Ic (NDWFS) zSDSS NB918
All data from NOAO 4m telescopes NB918 stack is
24 hours integration.
38
z 9180
A

• Gemini image of z6.535 galaxy
• (Rhoads et al. 2004)

39
LALA J142442.24353400.2 _at_ z6.535

• Gemini spectrum shows asymmetric line, no
  continuum.
• (Rhoads et al. 2004)

40
Observing the Distant Universe with clusters
Gravitational Telescopes Lensed Galaxies are
much brighter
Use a BIG telescope! 1021m primary with an 8m
secondary! The Cosmic Telescope! NOTE magnifica
tion x25 ? from z6 to z1.5 ? from z10 to
z2.5
1021m (M1014Mo)
41
A2218 (zcl0.17)
A2218 3 spectroscopic confirmed multiply
imaged systems.
42
A z10 galaxy

• (Pello et al 2004)

43
Some large ongoing surveys
44
DEEP2

• DEEP
• Keck LRIS
• 1000 galaxies to I24.5
• DEEP2
• Keck Deimos, R5000
• 50000 galaxies 0.7 lt z lt 1.4 (BRI selected)

45
VVDS

• VLT VIMOS
• Imaging with CFHT (BVRI), NTT (K), ESO2.2 (U)
• Shallow
• 100 000 galaxies to AB(I)22.5, R250
• Subsample observed at R2500
• Deep
• 50 000 galaxies to AB(I)24
• Ultra-Deep
• 1000 galaxies to AB(I)26 observed with IFU

46
GOODS

• 300 sq. arcmins., HDFN and CDFS
• Spitzer (Dickinson)
• Ultradeep IRAC, deep IRAC and MIPS
• Hubble (Giavalisco)
• BViz deep images
• Extra data from
• KPCTIO (U),
• VLT (NIR),
• VLT GeminiKeck (spectroscopy),
• XMMChandra (X rays)

47
ALHAMBRA

• Aim 8 x ½ square degree fields
• Calar Alto 3.5m, LAICA W2000
• AB(BVRI) 25, AB(JHK) 22
• Designed with photometric redshifts in mind
• Optical coverage from 3500 to 9500 A
• 20 non-overlapping 300 A-wide filters
• Redshift accuracy Dz / (1z) 0.02
• 800 000 galaxies to AB(I)24
• 2000 galaxies at zgt5

48
Filter system and survey depth

• (Moles et al 2006)
• (Benitez et al 2006)

49
(No Transcript)
50

• What have we learned about the
• high-redshift
• universe?

51
VVDS Redshift distribution
(Le Fevre et al 2005) First results from the
Deep survey, over 9000 galaxies with 17.5 lt AB(I)
lt 24 1000 galaxies _at_ z gt 1.4
52
Luminosity functions of quasars and galaxies

• (Fan et al 2004)
• (Iwata et al 2005)

53
Extremely red objects explained?

• Using GOODSIRAC observations of the UDF
• Old population mixed with younger one at redshift
  2.9
• (Yan et al 2004)

54
(No Transcript)
55
but still

• z6.5 galaxy
• or dusty z3 object?
• (Mobasher et al 2004)

56
Double populations at z6 too

• gt1010 Msun galaxies observed at z6
• zform 7 20
• Consistent with solar metallicity and low dust
  content
• (Yan et al 2005)

57
Distant red galaxies

• 153 galaxies in GOODS field, 1ltzlt3.5
• 10 pure old
• 40 mixed
• 50 dusty
• (Papovich et al 2006)

58
and HEROes
(Maihara et al 2001)

• Observed in the Subaru Deep Field
• Not passive Ellipticals
• z10 LBGs?
• Possibly dusty
• z2.5 starbursts (protoellipticals?)
• (Totani et al 2001)
• But not SCUBA sources
• (Coppin et al 2004)

59
HDF-S objects (Yahata et al 2000)
Selection of very red objects in the HDF-S Most
undetected in WFPC2 UBVI images, detected in
NICMOS JHK reaching AB26
60
HDFS objects (Yahata et al 2000)

• Redshift estimates reach to
• z 15, if (JHK) colours are
• takenface value as Ly-break

61
Bouwens search in HST deeps

• No good z10 candidates have been found in a
  search of (J-H) red objects
• (or J-dropouts) in all available deep NIR HST
  images
• (Bouwens et al 2005)

62
Quasar Density at z6

• Based on nine zgt5.7 quasars
• Density declines by a factor of 20 from z3
• Emergence of earliest supermassive BHs
• Cosmological implication
• MBH109-10 Msun
• Mhalo 1013 Msun
• How to form such massive galaxies and BHs in
  less than 1Gyr??
• The rarest and most biased systems at early times
• The initial assembly of the system must start at
  zgtgt10
• ? co-formation and co-evolution of the earliest
  SBH and galaxies

(Fan et al. 2004)
63
The Star Formation Rate

• or why I do not really like this quantity

64
Measurements of the SFD (1)

• Madau et al. 1996
• (MNRAS, 283, 1388)

65
Measurements of the SFD (2)

• Connolly et al. 1997
• (ApJ, 486, L11)

66
Measurements of the SFD (3)

• Pascarelle et al. 1998
• (ApJ, 508, L1)

67
Measurements of the SFD (4)

• Barger et al. 2000
• (AJ, 119, 2092)

68
Measurements of the SFD (5)

• Lanzetta et al. 2002
• (ApJ, 570, 492)

69
Measurements of the SFD (6)

• Rowan-Robinson 2003
• (MNRAS, 345, 819)

70
Measurements of the SFD (7)

• Appenzeller et al. 2004
• Msgr, 116, 18

71
Measurements of the SFD (8)

• Bouwens et al. 2005
• ApJ, 624, L5

72
Measurements of the SFD (9)

• Taniguchi et al. 2005
• PASJ 57, 165

73
(little) Conclusions about the SFR

• Perhaps many effects are not well understood
  yet
• Dust effect, distribution and abundance
• Calibration of different estimators
• Selection effects on emitters
• Observational effects (dimming?)

74
Gunn-Peterson troughs at zgt6 ?
75
Strong Evolution ofGunn-Peterson Optical Depth
Transition at z6?
(Fan et al. 2003)
76
But(Songaila 2004)

• a measurement of the Lya optical depth in a
  large sample of QSOs does not show any
  discontinuity in the highest-redshift QSOs

77
Are we really reaching out to the reionisation
epoch?

• Latest SDSS QSO data point towards that
• BUT for WMAP, and
• Reionisation is a patchy process
• Possible inhomogeneity (Miralda-Escude et al
  2000)
• Proximity effect of QSOs over environment
• They reside inside large overdensities
• Infall processes plus redshift
• Ionisation effect (Barkana Loeb 2003)
• and WHO is doing the reionisation?

78
(Razoumov et al 2002)

• Numerical simulations of the reionisation epoch

79
Could we see behind zreion?

• Loeb, Barkana, Hernquist (2005)
• To observe an object behind reionisation it
  needs
• either large-scale ionising sources around it, or
• large numbers of massive (gt100 Msun) stars in it
• in order to ionise a sphere around the object
  and permit Lya photons to escape absorption

80
Quasars are boring
The Lack of Evolution in Quasar Intrinsic
Spectral Properties

Ly a
NV
OI
Ly a forest
SiIV
81
so boring

• No metallicity evolution detected out to z6
• (Maiolino et al 2004)

82
Chemical Enrichment at zgtgt6?

• Strong metal emission ? consistent with
  supersolar metallicity
• NV emission ? multiple generation of star
  formation
• Fe II emission ? might be from metal-free Pop III

(Barth et al. 2003)
(Fan et al. 2001)
83
First generation of stars

• (Madau, Ferrara, Rees 2001)
• A first generation of pre-galactic SN could
• enrich medium (approx 20 filling factor) to
• the metallicity level seen in the Lya forest at
• z 3 without perturbing galaxy formation

84
A Double Reionization?(Cen 2003)

• The universe reionised first at z 16 via Pop
  III stars, and then became neutral again when
  those disappeared
• The universe reionised again at z 6 when AGNs
  became important and galaxies formed normal
  stars
• Furlanetto Loeb (2005) Extended period more
  physically plausible than double-peaked history

85
GRBs, the new probe
86
GRB050505 _at_ z 5.3

• (Jakobsson et al 2006)

87
GRB050904 6.3
(Tagliaferri et al 2005)
(Kawai et al 2006)
88
GRB030329 sn2003dh (Ic)(Stanek et al 2003)
89
First Hi-res spectra VLTUVES

• MgII systems _at_ z1.4
• en GRB021004
• (z2.328)
• MgII systems _at_ z1.25
• en GRB020813
• (z1.255)
• (Fiore et al 2005)

90
GRB Echelle spectroscopy _at_ z 4

• DLA detected at GRB host redshift, similar to
  other known z4 DLAs
• Low metallicity,
• low dust content,
• very high N(HI)
• (Chen et al 2005)

91
GRB050505 _at_ z4.275

• (Berger et al 2006)

92
Probing a different medium
93

• GRB DLAs probe higher metallicity environments
  than regular QSO DLAs
• Linked to higher N(HI) or environment dependent?

94
Let me finish with a twist
95
Measurements of UV luminosity and slope
96
Galaxy evolution in a different plot
97
So

• Many probes of the high-z universe agree
• Star formation histories (?)
• Metallicity evolution
• Evolution of the gas content
• Galaxy formation and evolution
• But
• The scatter is too large in many of them
• Time for forming BHs, GRBs, and galaxies is
  running short
• The link between WMAP and the optical is not
  clear

98
Conclusions

• Its the beginning of a great adventure
• and, indeed we live interesting times