Cosmology from Large-Scale Galaxy Surveys Matthew Colless, - PowerPoint PPT Presentation

Cosmology from Large-Scale Galaxy Surveys Matthew Colless, Anglo-Australian Observatory A Century of Cosmology, Venice, 27-31 August 2007

A brief history of galaxy surveys AAO � Before 1980: Pre-history - first galaxy redshift surveys � Hubble, the expanding universe, and all that FOCAP FOC � 1985-1995: Age of Discovery - large-scale structures � The CfA redshift survey and the iconic stick-figure LDSS LDSS � The development of multi-object spectrographs � Large (10 4 ) z-surveys - LDSS, Autofib, PSCz, LCRS… Autofib utofib � 1995-2005: Industrial Revolution - ‘precision cosmology’ 2d 2dF � Massive (10 5 ) z-surveys - 2dFGRS, SDSS, 6dFGS, 2MRS… � Deep z-surveys - DEEP, VVDS, zCOSMOS… 6dF 6d � Present & future: Post-Modernism - death or glory? AA � � Dark energy - holy grail? much ado about nothing? � BAO (10 6 ) surveys - WiggleZ, FASTsound, WFMOS… WFM FMOS OS � Galaxy surveys with ELTs - baryon tomography…

State of the art ~1997 CfA fA Survey ~15000 z z’s Las Campanas Redshift Survey ~25000 z’s

Evolution of Redshift Surveys SDSS 2dFGRS

Surveys with the AAT & UKST UK Schmidt Telescope Anglo-Australian (1.2m) Telescope (3.9m)

The 2-degree Field Spectrograph Multi-ob ulti-object ject fi fibre re sp spectrog ectrogra raph w h with ith rob robotic otic positioner ositioner Spectra Sp ectra for 400 for 400 gala laxies ov xies over 2º er 2º 2dF ena 2d F enabled led a a massiv ssive e z -surv urvey ey

The 2dFGRS map of 221000 galaxies

Large-Scale Structure in Model Universes � The large-scale structure of the galaxy distribution, on scales from millions to billions of light-years, depends on… � the amounts of the various constituents of the universe (baryonic matter, dark 100 million light-years matter, dark energy etc.) � the recipe for how galaxies are formed (when, where, and with what bias relative to the dark matter) � The rich structure of the galaxy distribution encodes much physics and many parameters

Surveys versus simulations � Comparis ison of slic lices fr from CfA fA2, 2dFGRS and SDSS surveys on diffe ifferent scale les, all ll compared to equiv ivale lent slic lices, sele lected fo for their ir sim imila ilarit ity, fr from the Mille illenniu ium Sim imula latio ion (Sprin pringe gel, Fre renk k & W White ite 2 2006)

2dFGRS structure & cosmology results

2dFGRS structure & cosmology results � The large-scale structure of the galaxy distribution is reliable and precisely determined on size scales from 1 Mpc to 200 Mpc � The properties of the galaxy distribution confirm that the large-scale structure grows by gravitational instability… �� � consistent with quantum fluctuations emerging from Big Bang being amplified by gravity into galaxies/clusters/superclusters � The total density of all types of matter in the universe is � M = 0.23 �� � there is only 23% of critical density needed for a flat universe � The total density in ordinary matter is � B = 0.04 �� � baryons are 19% and CDM is 81% of all matter in the universe � Neutrinos make up less than 13% of all the matter in the universe �� � the total mass of the 3 neutrino species is less than 0.7 eV � First detection of baryon acoustic oscillations (simultaneous with SDSS)

Composition of the cosmos Baryons 19% 19% SNe CMB 77% 77%

The 6dF Galaxy Survey � A redshift & peculiar velocity survey of galaxies in the local universe � Covers southern sky with |b|>10º � Primary galaxy sample selected from 2MASS with K tot <12.75 � Also H<13.0, J<13.75 (2MASS) and r<15.6, b<16.75 (SuperCosmos) � 11 other samples: radio, X-ray, IRAS � Peculiar velocity survey uses FP for 15,000 bright early-type galaxies � Observations obtained May 2001 to Jan 2006 using 6dF spectrograph on the UK Schmidt Telescope � Database:137k spectra, 124k galaxy redshifts over 80% of southern sky � Final data release Aug 2007

Sky coverage & redshift completeness Obse bserve rved 1 d 1464/1 /1598 fi fields; 9 lds; 92% o of th the so south thern rn sky sky w with ith |b|> |b|>10° Mean an re redsh dshif ift c t comple plete teness ss for th r the K-ban K-band prim d primary ary sam sample ple is 8 is 88%

Comparison with other z -surveys � Comparing the 6dFGS to the 2dFGRS and SDSS… � 6dFGS galaxy sample is NIR selected rather than optically-selected � 6dFGS surveys < z >=0.05 local universe (cf. 2dFGRS/SDSS < z >=0.1) � 6dFGS volume is comparable to the 2dFGRS, but 3x smaller than SDSS � Large fibres, so aperture effects are less for 6dFGS than 2dFGRS/SDSS

6dFGS redshift space maps z < 0.1

The 6dFGS view of the local universe

6dFGS power spectra and constraints Galaxy power spectrum z -only ly z z + v Velocity power spectrum 1 � conto tours rs on pairs o pairs of param paramete ters rs

2MASS Redshift Survey � All-sky z -survey to K s =11.25 (limit is 1.5 mag brighter than 6dFGS) � 93% complete with � cz � =6000 km/s z’s from Arecibo, Green Bank and FLWO 1.5m in North and 6dFGS and CTIO in South � 2MRS contains 23,000 galaxies and is densest all-sky z -survey to date Erdogdu et al., ., 2006, , MNRAS, , 368, , 151

Local density field reconstruction 2MRS de density sity fi field re ld reconstru structio tion by by F Fourie rier-Be r-Besse ssel de l decompo positio sition an and W d Wie iener fi r filte lterin ring Erdogdu et al., ., 2006, , MNRAS, , 373, , 45

Predicted local velocity field � The linea The linear v r velocity field elocity field in the in the Sup uperg ergala lactic ctic Pla lane a ne as p s pred redicted icted from from the reconstructed the reconstructed 2M 2MRS d RS density field ensity field (a (assum ssuming ing � =0.5) 0.5) Erdogdu et al., ., 2006, , MNRAS, , 373, , 45

Dark energy from geometry � The g The geom eometry of the univ etry of the universe erse ca can b n be m e mea easured sured a at v t very ea ery early rly tim times from es from C CMB, a , and nd a at la t later ter tim times b es by a y a v variety of m riety of method ethods: s: � ‘Sta tand ndard rd ca cand ndle le’ prov rovid ided ed by sup y supernov ernovae � ‘Sta tand ndard rd sca scale le’ from from gra ravita itationa tional l lensing ensing � Grow Growth of d th of density ensity perturb erturbations in clusters tions in clusters � ‘Sta tand ndard rd ruler ruler’ of b of baryon ryon acoustic oscilla coustic oscillations (B tions (BAO) O) in the g in the gala laxy d xy distrib istribution ution � The geometry of f the univ iverse is is governed by the equatio ion of f state, knowle ledge of whic ich reveals ls the nature of f the dark energy

The WiggleZ BAO survey � Science goals… � Primary: first z -survey measure of BAO scale at z ~1; aim for ~2% precision to constrain w to ~10% and test for evolution � Secondary: study properties of star-forming galaxies at z ~1 � Observing plan… � Targets: 400,000 b~1 galaxies (200,000 with z>0.5) over 1000 deg 2 � Sample: from GALEX + SDSS/RCS; FUV-NUV colour selects Lyman break at z >0.7; NUV-r selects emission line galaxies � Observations: using AAOmega spectrograph with 2dF on AAT � Status: ~80,000 redshifts to date; aim to complete in 2009 � WiggleZ team… � Australia: Drinkwater, Jurek, Pimbblet (UQ); Blake, Couch, Forbes, Glazebrook, Brough, Jones, Barnes (Swinburne); Woods (UNSW); Croom (Sydney); Pracy (ANU); Colless, Sharp (AAO); USA: Forster, Madore, Martin, Small (GALEX); Canada: Gilbank, Gladders, Yee (RCS2) Glazebrook et al. , astro-ph/0701876

BAO precision and constraints on w � Sim Simula ulations of tions of Wig iggleZ leZ BAO m O mea easurem surement com ent compared red to m to mod odel el pow ower sp er spectrum ectrum � exp expect ~2% p ect ~2% precision on sta recision on stand ndard rd ruler sca ruler scale le � Mea easure sure H(z (z) and nd D A (z) at t z ~0.7, us 0.7, use p priors on riors on � M and nd H 0 (CMB es � M h 2 ), reconstruct d giv ives , reconstruct density field ensity field � 2.7% on .7% on H , 1.8% on 1.8% on D A � Eq Equa uation of sta tion of state p te precision recision � 10% on 0% on w 0 (sim (simila ilar to r to SNe e surv urveys b eys but ut com complem lementa entary in ha ry in having ing a a d different m ifferent method ethod & d & different ifferent deg egenera eneracies cies) � Monte Carlo average data — Model power spectrum