Galaxy Formation and Evolution _in Clusters_ _3

Document Sample
Galaxy Formation and Evolution _in Clusters_ _3 Powered By Docstoc
					Large-Scale Structure -- From the Moon?
Alice Shapley (Princeton) November 29th, 2006

(credit: Springel et al. 2005)

Overview and Motivation
• • • • Importance of Large Scale Structure studies Review of state-of-the-art galaxy LSS results from z~0 to high redshift Planned ground-based LSS studies How does this relate to planned lunar observatories (emphasis on zenith-pointing liquid mirror telescope)

Large Scale Structure of Galaxies
• Use large-scale spatial distribution of galaxies to constrain cosmological parameters (mh, bh2, k, ns, w0, w1*), also combine with complementary constraints (CMB, SNe, weak lensing) Use large-scale spatial distribution of galaxies to learn about galaxy and structure formation and evolution, relationship between galaxies and dark matter halos


*parametrize equation of state of the universe, w=p/ as w(z) = w0 + w1z

Local Redshift Surveys
• Sloan Digital Sky Survey (SDSS): u’g’r’i’z’ photometric survey of ~10,000 deg2, redshifts for ~106 galaxies with r mag < 17.77 (Apache Point observatory, 2.5 m telescope)


2 Degree Field Galaxy Redshift Survey (2dFGRS), photometric survey of ~1500 deg2, redshifts for ~2x105 galaxies with bj mag < 19.45, Anglo Australian Observatory, 4 meter telescope)

Equatorial slice from SDSS, ~67,000 gals out to z=0.2

Local Redshift Surveys: P(k)
• LSS results on cosmological parameters • Based on 200,000 SDSS galaxies • P(k), when combined with other data, can place constraints on mh, 8 of galaxies, bh2, ns (Tegmark et al. 2004)

Local Redshift Surveys: BAO

• Pattern of acoustic oscillations of coupled matter and radiation imprinted on the surface of last scattering, as observed in the CMB radiation, ~300,000 years after the Big bang, at z=1089. • Preferred scale at ~150 comoving Mpc, sound horizon at recombination.

Local Redshift Surveys: BAO
(Eisenstein et al. 1998)

• Preferred scale also imprinted on late-time clustering of galaxies. • Initial dark matter perturbation grows in place, while baryonic perturbation carried outward in expanding spherical shell, which expands to 150 Mpc radius by epoch of recombination. Manifest as spike in galaxy correlation function, or wiggles in P(k).

Local Redshift Surveys: BAO

(SDSS, Eisenstein et al. 2005)

(2dFGRS, Cole et al. 2005)

• Baryon acoustic oscillations (BAO) have now been observed in both SDSS and 2dF low-redshift galaxy samples • SDSS used real space correlation function of Luminous Red Galaxies (subsample with larger volume); 2dF used P(k)

Local Redshift Surveys: BAO

(SDSS, Eisenstein et al. 2005)

(2dFGRS, Cole et al. 2005)

• SDSS: 46,748 galaxies over 3816 deg2, 0.72h-3 Gpc3 • position of peak in correlation function interpreted as a standard ruler, yields 5% distance measurement to z=0.35, 4% measurement on ratio of D(z=0.35)/D(z=1089), mh2(8%), k (1%) (assuming Dark Energy is cosmological constant)

Local Redshift Surveys: HOD

(Zehavi et al. 2005)

• SDSS: Galaxy correlation function on smaller scales (0.1-10 h-1 Mpc), use sample of 2x105 galaxies over 2500 deg 2 • Investigate luminosity- and color- dependence of clustering (assume cosmology): stronger clustering for brighter and redder galaxies, inflection at rp~1-3 h-1 Mpc

Local Redshift Surveys: HOD

(Zehavi et al. 2005)

• Halo Occupation Distribution (HOD) -- relates galaxies of given luminosity and color to dark matter halos of given mass • HOD model naturally explains inflection (1 halo->2halo regime), and color-dependence of clustering in terms of central and satellite galaxies of halos as a function of halo mass

z~1: DEEP2 Redshift Survey
• University of California project to survey ~40,000 galaxies at z~1 (8 billion light years away) using the Keck 10-meter telescope in Hawaii, 4 square degrees on the sky, galaxies down to R mag =24.1, spectra for ~40-50% of z~1 candidates. • Goal is to look at large scale structure at z~1 (how has the large-scale structure evolved over 8 billion years), as well as galaxy evolution, though survey volume doesn’t contain any rich clusters (~107 Mpc3). Main results about galaxy evolution.
Map of galaxies in DEEP2 field (courtesy: A. Coil)

z~1: DEEP2 Redshift Survey

(Coil et al. 2006)

• Investigate clustering of galaxies at z=0.7-1.4 vs. color and luminosity (Coil et al. 2005, 2006): more luminous galaxies and redder galaxies are more strongly clustered; small-scale rise in correlation function (1-halo term). z~0 patterns already in place. • Interpret results in terms of minimum dark matter halo mass for hosting DEEP2 galaxies (~1012M).

z>1 (UV selection and others)
• In this regime (z~1.5-5.0), samples are also currently not large enough to constrain cosmological parameters

• Focus has been on relation between galaxies and dark matter, linking galaxy populations at different redshifts, galaxies in different environments (protoclusters)
• Main limitation for most high-z clustering samples: dependence on photometric redshifts -- even for determining n(z) for inverting w(); exception is UV-selected samples (Lyman Break Galaxies (LBGs) and analogous selection at z~2.3 and z~1.7) • Danger: contamination, incorrect redshift distribution, which can lead to biased estimates of physical correlation length

Clustering of LBGs: z~1.5-3.0
• 28,500 galaxies, 1.4<z<3.5 • 21 fields, 0.8 degree2 •~1600 spectroscopic redshifts • RAB=23.5-25.5 • Correlation lengths: z=2.9 (LBG), r0=4.00.6 h-1 Mpc z=2.2 (BX), r0=4.20.5 h-1 Mpc z=1.7 (BM), r0=4.50.6 h-1 Mpc •Implied halo masses: ~1011.5 M (LBG) ~1012M (BX/BM) (from comparison with GIFLCDM numerical simulation, DM Halos with same clustering)

(Adelberger et al. 2005, based on w())

Evolution of Clustering to z~1, 0
Follow evolution of DM halo clustering in simulation
• Matches early-type absorption line DEEP2 galaxies at z~1 (Coil et al. 2003)
• Matches SDSS ellipticals at z~0.2 (Budavari et al. 2003) • Typical UV-selected galaxy at z=2-3 will evolve into an elliptical by z=0 • Quadri et al. (2006): K<21, Kselected galaxies and red galaxies at 2<z<3.5 are hosted by halos whose descendents end up in groups and clusters

(Adelberger et al. 2005)

Clustering of LBGs: z>3

• Lee et al. (2005): HOD modeling of LBGs at z~4 (2463 gals) and z~5 (878) in the GOODS fields (2 10’x17’ fields), deviation from power-law correlation function is clear at small scales, evidence for substructure in DM halos; also find brighter galaxies are more strongly clustered; DM halos of z~5 galaxies less massive; mostly based on photo-z’s • Ouchi et al. uses Subaru samples at z~4-5, finds similar results

Future LSS Projects
• • WFMOS (Wide-Field Multi-object Spectrograph) DES (Dark Energy Survey)


SKA (Square Kilometer Array)
(International collaborations)

Future LSS Projects: WFMOS
Parameters of WFMOS: • Prime focus of Subaru

• 1.5 degree FOV
• 0.39-1.0 m •~4500 1” fibers, 3000 of which go to R~2000-4000 spectrographs •Timeline: Dark Energy Science Survey 2013-2016 (assumes pre-imaging available) (Glazebrook et al. 2005)

Future LSS Projects: WFMOS

• Planned surveys target z~1 and z~3, probe BAO (Dark Energy) in 3 different redshift ranges z<1, 1<z<3, 3<z<1000 • Get constraints on both H(z) and DA(z) •**Volumes are > Gpc3**, get ~106 galaxies, not especially faint

Future LSS Projects: WFMOS
• Observables in spectroscopic galaxy redshift survey: angular positions () and redshifts (z)

• Use BAO method to observe standard ruler (horizon scale at recombination); know true size, use radial and transverse clustering to infer cosmological parameters H(z) and DA(z)

Crucial fact: In order to distinguish between w=-1 (cosmological constant) and w=-0.9, need to measure H(z) and DA(z) to 1%. This required precision determines volumes that need to be probed.

• Two sources of statistical noise: sample variance and shot noise • Sample variance: limit on number of independent wavelengths of fluctuations that can fit into finite volume of survey

Future Plans: Statistics

• Shot noise: imperfect sampling of underlying fluctuations due to finite number of galaxies (not important for nP>>1)
• Above critical number density, uncertainty on power spectrum (and cosmological parameters), scales as inverse square-root of survey volume: VOLUME LIMITED!!!

Future LSS Projects: WFMOS
(WFMOS Feas. Study) • To yield 1-2% errors on distances (and 8% on constant w, and 25% on rate of change of w), need to survey 2000 deg2 at z~1 and 300 deg2 at z~3. This represents the goal for groundbased studies that will be completed by the time lunar operations would be taking place.

Future LSS Projects: DES
• Deep optical/near-IR survey of 5000 deg2 to 24th mag in griz (10s of Gpc3) • Use new instrument DECam, 3 deg2 FOV, mounted on CTIO 4-m telescope • Complete observations by 2014 (DES White Paper, 2005)

Future LSS Projects: DES
Use photometric dataset for 4 independent probes of cosmological parameters out to z~1-1.5:

1. Galaxy cluster mass function and spatial distribution (probes volume element and growth of structure)
2. Weak lensing tomography (probes geometry and growth of structure) 3. Galaxy angular clustering (BAO, geometric test of parameters) 4. Supernova luminosity distances (geometric test of parameters) Obtain comparable errors on w to that of WFMOS survey

Future LSS Projects: SKA
• Square Kilometer Array (operational by 2020?), 0.1-25 GHz. • Proposed experiment, assuming 10 deg2 FOV: All hemisphere (20,000 deg2) HI survey to z~1.5; 109 galaxies with clustering statistics limited by survey volume (compare with WFMOS volume); constraints on cosmological parameters improve accordingly (1% on w).

(Rawlings et al. 2005, Blake et al. 2005)

Future LSS Projects: The Moon?
• One proposed design for IR lunar telescope: 20 m, fixedaxis, spinning liquid mirror, operating at 1-10 m, located at/near lunar pole, 15’ field • Study rest-frame optical/near-IR properties of gals at 1<z<6, UV/optical properites of gals at z>6

• Obtain ultra-deep, and high-spatial resolution images and spectra of distant galaxies, the first stars and globular clusters. Deeper than JWST, higher angular resolution.
• What about large-scale structure, and clustering of these objects?

Future LSS Projects: The Moon?
• Critical point: volume probed by proposed design. • With telescope at pole, survey 3.1 deg2 annulus in 18 years. For comparable redshift ranges to, e.g., WFMOS project, volumes are 0.1%, 1%, with correspondingly worse errors on cosmological parameters. • Survey area is comparable to that of DEEP2 at z~1, and to existing higher-z samples. Need more area for both cosmological (from and galaxy Angel formation 2006) studies.

• Many ground-based LSS projects planned, that will happen by the time lunar observatory would be in place (space-based also: SNAP, ADEPT) • These projects will cover volumes > (Gpc)3. To constrain cosmological parameters at the level of these projects, comparable volumes are necessary (sample variance is limiting factor). • For galaxy formation studies, also need representative volumes (contain range of environments) • Planned lunar observatory (at least optical/IR) would only cover few to 10s of degrees -- just not enough volume to provide competitive constraints relative to other planned projects.

• Planned lunar observatory (zenith-pointing, liquid mirror telescope) will cover few to 10s of degrees. • This is just not enough volume to provide competitive constraints on cosmological parameters or even distribution of galaxies in a full range of cosmic environments. • Will tell us small-scale clustering of galaxies down to incredible faint magnitudes (and as a function of restframe optical morphology). • Strengths of planned lunar observatory complementary to goals of most current and future LSS studies.

z>1.5 Rest-UV Color Selection

• z~3 UGR Lyman Break criteria, adjusted for z~2 (Adelberger et al. 2004) • Spectroscopic follow-up with optimized UV-sensitive setup (Keck I/LRIS-B) • ~1000 galaxies at z~3, >750 galaxies with spectroscopic redshifts at z=1.4-2.5, in what was previously called the Redshift Desert

z>1.5 Rest-UV Color Selection


• z~3 UGR Lyman Break criteria, adjusted for z~2 (Adelberger et al. 2004) • Spectroscopic follow-up with optimized UV-sensitive setup (Keck I/LRIS-B) • ~1000 galaxies at z~3, >750 galaxies with spectroscopic redshifts at z=1.4-2.5, in what was previously called the Redshift Desert

z>1.5 FIRES/DRG selection (~20 zs)

• J-K>2.3 criteria meant to select massive evolved galaxies with significant
Balmer/4000 Å breaks at z>2; turns out selection also yields massive dusty starbursts •~25% appear to contain AGN (much higher than fraction of UV-selected population) (Papovich et al. 2005) • Only limited number of spectroscopic redshifts

(Reddy et al. 2005)

(Franx et al. 2003)

Finding High-z Protoclusters: UV
• Multiple examples of galaxy spikes discovered in UV surveys by Steidel et al. Benefits of UV-selected protoclusters.

SSA22a: z=3.09 (first evidence LBGs were strongly clustered)

Q1700: z=2.30 (age/mass-density relation?)

Q1549: z=2.85 (at the same redshift as bright QSO)

Finding High-z Protoclusters
• Highest redshift X-ray-detected cluster at z=1.45 (Stanford et al. 2006). BUT: large-scale overdensities of galaxies detected at z>2, may evolve into >1014Mclusters by z~0

• Significant galaxy overdensities serendipitously discovered during UV-selected survey (z=3.09, 2.30, 2.85) (more later)
• Over-densities of Ly emitters discovered around targeted radio galaxies at 2<z<5
 HiZRG may have extreme stellar masses (1012M), based on K-band mags, strong clustering at z~1  extended clumpy morphology, like simulations of BCG formation  extreme radio rotation measures, indicative of dense hot gas environments; large Ly halos, regions consistent w/ originating from a cooling flow

(work by Miley, Venemans, Kurk, Pentericci, Overzier et al.)

Hierarchical Stucture Formation
• Whereas monolithic collapse works backwards from present using understanding of stellar evolution and stellar dynamics (what’s the cosmological model?), hierarchical structure formation works from within the CDM cosmological framework, provides ab initio model for galaxy formation, motivated by CMB and large-scale structure • Galaxy formation and evolution is a natural consequence of the growth of the power spectrum of fluctuations by gravitational instability, in a universe dominated by dark matter

• Model predicts evolution of dark matter halo mass function through merging and accretion
(Springel et al. 2005)