The Clowes-Campusano Large Quasar Group Survey G Williger

The Clowes-Campusano Large Quasar Group Survey G. Williger (UL, USA) L. Haberzettl (UL, USA) J. T. Lauroesch (UL, USA) M. Graham (Caltech, USA) R. Davé (Steward Obs. , USA) A. Koekemoer (STSc. I, USA) L. Campusano (Univ. de Chile, CL) R. Clowes (Univ. Lancashire, UK) I. Söchting (Oxford Univ. , UK) K. Harris (Univ. Lancashire, UK) C. Haines (Univ. Birmingham, UK) J. Loveday (Univ. Sussex, UK) D. Valls-Gabaud (Obs. de Paris, F) M. Lehnert (Obs. De Paris, F) N. Nesvadba (Univ. Paris-Sud, F)

The Clowes-Campusano LQG Survey Outline Background - Why important? - How to find them? - LQG "zoo" The CCLQG Survey Lyman Break Galaxies in the CCLQG Future

Why Large Quasar Groups Important Quasars Signatures of physical mechanisms - supermassive black holes+high accretion - massive haloes - feeding by gas-rich (major? ) mergers? - - associated with high star formation - Quenching in high density regions (gas stripping etc. ) - feedback mechanisms - strong winds - high ionization Quasars thrive in somewhat overdense but not too overdense regions

Quasars: a stage of galaxy formation duration of quasar phase << timescale for quiescent evolution of AGN and age of Universe LIFETIMES: ~10 -30 (1 -100? ) Myr scales at any time only small number of galaxies in quasar phase

Large Quasar Groups: Efficient Sites of Quasar-Galaxy Relations Galaxies bear signatures of merger rates Galaxies give star formation histories give clue to recent merger activity in region clue to past merger/galaxy formation in region Galaxies give measure of halo masses Deep, wide galaxy surveys with lots of quasars efficient laboratories for studying physical mechanisms in both galaxy and quasar evolution

Background: LQGs DISCOVERY: Webster 1982 →close triplet + one more distant QSO at z~0. 37 scales ~75 h-1 Mpc soon after: two other LQGs: Crampton et al. (87, 89; 23 QSOs) at z~1. 1 Clowes & Campusano (1991; 18 QSOs) z~1. 3 large irregular shaped, filamentary structures on scales of 50 -200 Mpc with concentrations of 5 -20 QSO's too large to be virialised, probable relics of large scale fluctuations CURRENT VIEW: rare (4σ) structures, ~1/3 space density of galaxy super-clusters (Pilipenko 2007)

How to find Large Quasar Groups Assumption: quasars randomly distributed among galaxies with sufficient gas accretion compare real catalogues QSO distributions to random usual spatial correlation functions not efficient for finding filamentary structures Alternatives: minimal spanning tree, skeletons, spine of cosmic web

How to Find Structures Barrow et al. 1985 2 Zwicky Galaxy Catalog 2 Random Sample 1091 galaxies in the North Galactic Cap 1091 galaxies over the same sky area with Pmag ≤ 14 mag and δ≥ 0 and b≥ 40°

Minimal Spanning Tree (e. g. Barrow et al. 1985) generalization of the nearest-neighbour or friend-of-friend method connect points with unique path distribution in tree length: 1 D: w 1(l) = 1/l 0 exp(-l/l 0) with = l 0 2 D: w 2(l) = 2 l/l 02 exp(-l 2/l 02) with = √(2π)/2 l 0 ⇩ minimal tree if sum of length of segments is minimal MST can be used to identify under- and over-dense regions

Minimal Spanning tree define thresholds lt for maximum (over-dense) minimum (under-dense) length of branches lt and minimum number of objects in a domain M 3 D: density of clusters with l ≥ lt is higher than threshold density if ρt h <ρ>/lt 3 (strongly depends on the choice of lt, M, determines statistical significance)

Finding Structures in MST two reduction methods to find structures a) prune: strip branches of level k (≥ 3 connections) of dead-end connections b) Prune + separate: also remove edges above a cutoff length

The Clowes-Campusano LQG Survey MST pruned to level 10 (branches k≤ 10 removed) random mean edge length: = 0. 0215 rad (1. 232°) and = 0. 0267 (1. 530°)

MST: pruned and separated random cut-offs: 2 and 1. 6 0. 043 rad (3. 142˚) separation in both cases

MST length distributions random length frequency distribution of MST(Zwicky) shows excess of large and small lzwicky length frequency distribution of MST (random) follows Gauss distribution centered on

Pilipenko (2007) LQG survey search in 2 d. F+SDSS QSO catalogs (> 100, 000 QSOs) 18 new LQG identified by MST + 2 LQG confirmed contain 6 – 16 QSOs on scales ~40 – 155 h-1 Mpc LQG TYPES: "Regular": 14 LQGs 6 – 8 members, scales ~60 spatial overdensity ≈ 10 "Jumbo": 6 LQGs 15 – 19 members, scales spatial overdensity ≈ 4 h-1 Mpc, ~140 h-1 Mpc, space density: ~ 7 h 3 Gpc 3 ⇨ ~500 – 1000 Jumbo LQGs morphologies: walls+blobs rather than filaments

Clowes-Campusano LQG automated search on UKST objective-prism plate (~25. 3 deg 2) ESO/SERC field 927 (1045+05 J 2000) 18 (up to 23+ depending on selection) quasars with 1. 2≤z≤ 1. 4, Bj<20. 4 (BRIGHT!) = 1. 27 selection effect by objectiveprism -- Ly-α emission shifted out of selection band at z>1. 8 cover 2. 5° x 5° on the sky banana like structure

Subset of CCLQG (2. 5 x 5 deg total, ~20 QSOs) CCLQG z~1. 3 3 x Mg. II-absorber overdensity Williger et al. 2002 LQG z~0. 8 2 x Mg. II-absorber overdensity discovery of 2 nd foreground LQG

Clowes-Campusano LQG Survey Galaxy populations in LQGs Lyman red Break Galaxies (LBGs) galaxy population red sequence/blue cloud at z~1 in dense environment (colour-density inversion at z~1? ) Quasar-galaxy mechanism correlations/feedback

Early Results 0. 5º subfield: Red Galaxies Haines et al. 2004 CTIO BTC 4 m V, I data ~0. 25 deg 2 subfield 31 x 27 h-2 Mpc 2 at z ~ 1. 2 2 x overdensity in red galaxies 3 x overdensity in red galaxies dashed contours 1. 65 gal. arcmin-2

The Clowes-Campusano LQG Survey Existing Data Set: 2 x 1. 2˚ GALEX FUV+NUV mlim~24. 0 mag SDSS u, g, r, i, z ~1. 6˚ Bok g mlim ~ 26 mag 2 x 1˚CFHT r+z mlim ~ 26 mag mlim ~ 24 mag ~1. 2˚KPNO 2. 1 m FLAMINGOS NIR J+Ks ~1° UKIRT Ks-band ~600 Magellan IMACS spectra 2 GALEX Medium Imaging Survey (used for Wiggle. Z bright LBGs)

The Clowes-Campusano LQG Survey Summary

Efficient search for z~1 galaxies Lyman Break Galaxies (LBGs) search for LBGs at z~1 using FUV-dropout technique Lyman Break 912 Å at z~1 FUV-dropout examples: FUV NUV SDSS

Lyman Break Galaxies: Star-Formers LBGs found over 0. 5

The Clowes-Campusano LQG Survey LBG search + selection criteria GALEX NUV selected sample: ~15, 800 objects SDSS DR 5 cross-correlation: ~13, 800 primary counterparts LBG selection criteria - Burgarella et al. (2006): m. NUV ≤ 24. 5 mag + FUV – NUV ≥ 2 additional selection criteria: resolved by Sloan Survey (SDSS) - extended source = galaxy resulting sample ~1000 LBG candidates

The Clowes-Campusano LQG Survey photometric redshift distribution dz = 0. 05 -0. 10 LBG candidates LQG@z~0. 8 CCLQG@z~1. 3

The Clowes-Campusano LQG Survey selecting redshift+luminosity limited subsample 2 redshift bins in front of the LQGs LQG@z~0. 8 CCLQG@z~1. 3 bright: MNUV ≤ M*NUV faint: MNUV > M*NUV Arnouts et al. (2005)

LBGs: old population component SFHs of stacked/averaged LBG SEDs χ2 -fit of averaged and normalized LBG SEDs to library of PÉGASE models LQG@z~0. 8 3 Gyr ≤ tbest ≤ 7 Gyr although fits with 250 ≤ tyoungest ≤ 800 are acceptable CCLQG@z~1. 3 3 Gyr ≤ tbest ≤ 7 Gyr

The Clowes-Campusano LQG Survey ~500 Myr results for best fitting ages show significant older tbest than Burgarella et al. 2007 (250≤tbest≤ 500 Myr) Burgarella sample include fainter LBGs younger star bursts ~250 Myr younger averaged SEDs? Our survey shallower, biased toward higher mass LBGs?

Results Affected by Confusion? GALEX point spread function ~5 -6 arcsec Depth of NUV~23. 5 -24 begins to be affected by confusion Compare with deep, 1 arcsec resolution CFHT r -band images ~20% of LBGs have >1 r-selected counterpart Confusion effect is small compared to scatter (factor of few to 10 -100) in stacked spectral energy distributions

QSOs on LBG Cluster Outskirts? LBG concentrations + filaments LBGs in proximity to QSOs QSO feedback mechanism?

z~0. 8 Quasar-LBG correlations P=0. 0027 1'~0. 5 local Mpc P=0. 0051, 0. 0056 117 LBGs, 17 quasars, 10000 random quasar placements

QSOs on (red) cluster outskirts? Need better photometric redshifts (near-IR) to bin in redshift red galaxies tend to avoid QSOs formation of large filaments

Summary/Conclusions 1) 2) 3) 4) Large quasar groups ~ quasar superclusters, useful laboratories for studying quasar-galaxy relations in large structure contexts Clowes-Campusano LQG field being explored in 2 deg 2 multi-wavelength survey z~1 LBGs several Gyr old, older than only other study – due to shallower survey, more massive LBGs? LBG-quasar correlation suggested to eye but does not show robust statistical signal (yet? ) – need more data

The Clowes-Campusano LQG Survey Future Plans future: - Bok+90 prime observations g in southern field medium band imaging (N, S) (done Mar 2008) - AAT spectra - CFHT queue observation r+z extend existing field ~5 deg 2 -- GALEX PI team Medium Deep Survey Extension: 6 fields -- Science: individual galaxy SEDs, color-density relation, AGN, … -- Study LQG analogues in Millennium Simulation -- Mark Younger discovered 3 z~2 LQGs – theory v. data