After the Dark Ages the Evolution of Luminous Sources at z5

a r X i v :a s t r o -p h /9812087v 2 4 D e c 1998After the dark ages:the evolution of

luminous sources at z <5

Piero Madau Space Telescope Science Institute,Baltimore,USA.Abstract I review recent observational and theoretical progress in our understanding of the cosmic evolution of luminous sources.Through a combination of deep HST imaging,Keck spectroscopy,and COBE background measurements,important constraints have emerged on the emission history of the galaxy population as a whole.A simple stellar evolution model,de?ned by a star-formation density that rises from z =0to z ≈1.5,a universal Salpeter IMF,and a moderate amount of dust with A V =0.23mag (A 1500=1.2mag),is able to account for most of the optical-FIR extragalactic background light,and reproduces the global ultraviolet,optical,and near-IR photometric properties of the universe.By contrast,a star-formation density that stayed roughly constant at all epochs appears to overproduce the local K -band luminosity density.While the bulk of the stars present today formed relatively recently,the existence of a decline in the star-formation density above z ≈2remains uncertain.If stellar sources are responsible for photoionizing the intergalactic medium at z ≈5,the rate of star formation at this epoch must be comparable or greater than the one inferred from optical observations of galaxies at z ≈3.A population of dusty AGNs at z ～<2could make a signi?cant contribution to the FIR

background if the accretion e?ciency is ～10%.

1Introduction

In the last few years,the remarkable progress in our understanding of faint galaxy data made possible by the combination of HST deep imaging [58]and ground-based spectroscopy [31],[11],

[52]has permitted to shed some light on the evolution of the stellar birthrate in the universe,to tentatively identify the epoch 1～

[2],[19].The explosion in the quantity of information available on the high-redshift universe at optical wavelengths has been complemented by the detection of the far-IR/sub-mm background by DIRBE and FIRAS [24],[12],that has revelead the optically ‘hidden’side of galaxy forma-tion,and shown that a signi?cant fraction of the energy released by stellar nucleosynthesis is re-emitted as thermal radiation by dust.The underlying goal of all these e?orts is to under-stand the growth of cosmic structures and the mechanisms that shaped the Hubble sequence,and ultimately to map the transition from the cosmic ‘dark age’[45]to a ionized universe pop-ulated with luminous sources.While one of the important questions recently emerged is the

Figure1:Spectrum of the extragalactic background light as derived from a compilation of ground-based and space-based galaxy counts in the U,B,V,I,and K-bands(?lled dots),together with the FIRAS125–5000μm(solid and dashed lines)and DIRBE140and240μm(?lled squares)detections. nature(starbursts or AGNs?)and redshift distribution of the ultraluminous sub-mm sources discovered by SCUBA[27],[1],[32],of perhaps equal interest is the possible existence of a large population of faint galaxies still undetected at high-z,as the color-selected ground-based and Hubble Deep Field(HDF)samples include only the brightest and bluest star-forming objects.In hierarchical clustering cosmogonies,high-z dwarfs and/or mini-quasars(i.e.an early generation of stars and accreting black holes in dark matter halos with circular velocities v c～50km s?1) may actually be one of the main source of UV photons and heavy elements at early epochs[38], [21],[22].In this talk I will focus on some of the open issues and controversies surrounding our present understanding of the history of the conversion of cold gas into stars within galaxies, and of the evolution with cosmic time of the space density of luminous sources.An Einstein-de Sitter universe(q0=0.5)with H0=50h50km s?1Mpc?1will be adopted in the following.

2Extragalactic background light

The extragalactic background light(EBL)is an indicator of the total luminosity of the uni-verse.It provides unique information on the evolution of cosmic structures at all epochs,as the cumulative emission from galactic systems and AGNs is expected to be recorded in this background.Figure1shows the optical EBL from known galaxies together with the recent COBE results.The value derived by integrating the galaxy counts[44]down to very faint magnitude levels[because of the?attening at faint magnitudes of the N(m)di?erential counts most of the contribution to the optical EBL comes from relatively bright galaxies]implies a lower limit to the EBL intensity in the0.3–2.2μm interval of I opt≈12nW m?2sr?1.When combined with the FIRAS and DIRBE measurements(I FIR≈16nW m?2sr?1in the125–5000

Figure2:Left:Mean comoving density of star formation as a function of cosmic time.The data points with error bars have been inferred from the UV-continuum luminosity densities of[31](?lled dots),[7] (?lled squares),[35](?lled pentagons),[56](empty dot),and[54](empty square).The dotted line shows the?ducial rate, ˙ρ? =0.054M⊙yr?1Mpc?3,required to generate the observed EBL.Right:dust corrected values(A1500=1.2mag,SMC-type dust in a foreground screen).The Hαdeterminations of [15],[55],and[17](?lled triangles),together with the SCUBA lower limit[27](empty pentagon)have been added for comparison.

μm range),this gives an observed EBL intensity in excess of28nW m?2sr?1.The correction factor needed to account for the residual emission in the2.2to125μm region is probably～<2 [10].We shall see below how a population of dusty AGNs could make a signi?cant contribution to the FIR background.In this talk I shall adopt a conservative reference value for the total EBL intensity associated with star formation activity over the entire history of the universe of I EBL=40I40nW m?2sr?1.

3Star formation history

It has become familiar to interpret recent observations of high-redshift sources via the comoving volume-averaged history of star formation.This is the mean over cosmic time of the stochastic, possibly short-lived star formation episodes of individual galaxies,and follows a relatively simple dependence on redshift.Its latest version,uncorrected for dust extinction,is plotted in Figure 2(left panel).The measurements are based upon the rest-frame UV luminosity function(at 1500and2800?A),assumed to be from young stellar populations[33].The prescription for a ‘correct’de-reddening of these values has been the subject of an ongoing debate.Dust may play a role in obscuring the UV continuum of Canada-France Reshift Survey(CFRS,0.32).As noted already by[33]and[35],a consequence of such large extinction values is the possible overproduction of metals and red light at low redshifts.Most recently,the evidence for more moderate extinction corrections has included measurements of star-formation rates(SFR)from Balmer lines by[55](×2at z=0.2),[17](×3.1±0.4at z=1),and[42](×2?6at z=3).ISO follow-up of CFRS?elds[13]has shown that the star-formation density derived by FIR?uxes

(×2.3±0.7at0≤z≤1)is about3.5times lower than in[46].Figure2(right panel)depicts an extinction-corrected(with A1500=1.2mag,0.4mag higher than in[35])version of the same plot.The best-?t cosmic star formation history(shown by the dashed-line)produces a total EBL of37nW m?2sr?1.About60%of this is radiated in the UV+optical+near-IR between 0.1and5μm;the total amount of starlight that is absorbed by dust and reprocessed in the far-IR is13nW m?2sr?1.Because of the uncertainties associated with the incompleteness of the data sets,photometric redshift technique,dust reddening,and UV-to-SFR conversion,these numbers are only meant to be indicative.On the other hand,the model is not in obvious disagreement with any of the observations,and is able,in particular,to provide a reasonable estimate of the near-IR luminosity density in the range0～

4The stellar mass density today

With the help of some simple stellar population synthesis tools it is possible at this stage to make an estimate of the integrated stellar mass density today.The total bolometric luminosity of a simple stellar population(a single generation of coeval stars)having mass M can be well approximated by a power-law with time for all ages t～>100Myr,

L(t)=1.3L⊙M

1Gyr

?0.8

(1)

(cf.[5]),where we have assumed solar metallicity and a Salpeter IMF truncated at0.1and125 M⊙.In a stellar system with arbitrary star-formation rate per unit cosmological volume,˙ρ?, the comoving bolometric emissivity at time t is given by the convolution integral

ρbol(t)= t0L(τ)˙ρ?(t?τ)dτ.(2) The total background light observed at Earth(t=t H)is

I EBL=c

1+z

dt,(3)

where the factor(1+z)at the denominator is lost to cosmic expansion when converting from observed to radiated luminosity density.From the above equations it is easy to derive

I EBL=740nW m?2sr?1 ˙ρ?

13Gyr

1.87

.(4)

The observations shown in Figure1therefore imply a“?ducial”mean star formation density of ˙ρ? =0.054I40M⊙yr?1Mpc?3.In the instantaneous recycling approximation,the total stellar mass density observed today is

ρ?(t H)=(1?R) t H0˙ρ?(t)dt≈5×108I40M⊙Mpc?3,(5) (corresponding to??=0.007I40)where R is the mass fraction of a generation of stars that is returned to the interstellar medium,R≈0.3for a Salpeter IMF.The optical/COBE background therefore requires that about10%of the nucleosynthetic baryons(?b h50=0.08[4])are in the forms of stars and their remnants.The predicted stellar mass-to-blue light ratio is M/L B ≈5. Note that these values are quite sensitive to the lower-mass cuto?of the IMF,as very-low mass stars can contribute signi?cantly to the mass but not to the integrated light of the whole stellar population.A lower cuto?of0.5M⊙instead of the0.1M⊙adopted would decrease the mass-to-light ratio(and??)by a factor of1.9for a Salpeter function.

Figure3:Left:Evolution of the near-IR luminosity density at rest-frame wavelengths of1.0μm(long-dashed line)and2.2μm(short-dashed line).The data points are taken from[31](?lled dots)and[16] (?lled square).The model assumes a constant star-formation rate of˙ρ?=0.054M⊙yr?1Mpc?3 (Salpeter IMF).Right:Same but with the star-formation history depicted in the right panel of Fig. 2.

5A constant star-formation density?

Based on the agreement between the z≈3and z≈4luminosity functions at the bright end, it has been recently argued by[54]that the decline in the luminosity density of faint HDF Lyman-break galaxies observed in the same redshift interval[33]may not be real,but simply due to sample variance in the HDF.When extinction corrections are applied,the emissivity per unit comoving volume due to star formation may then remain essentially?at for all redshift z～>1(see Fig.2).While this has obvious implications for hierarchical models of structure formation,the epoch of?rst light,and the reionization of the intergalactic medium(IGM),it is also interesting to speculate on the possibility of a constant star-formation density at all epochs 0≤z≤5,as recently advocated by[41].Figure3(left panel)shows the time evolution of the near-IR rest-frame luminosity density of a stellar population characterized by a Salpeter IMF, solar metallicity,and a(constant)star-formation rate of˙ρ?=0.054M⊙yr?1Mpc?3(needed to produce the observed EBL).1The predicted evolution appears to be a poor match to the observations:it overpredicts the local K-band luminosity density[16]and undepredicts the 1μm emissivity at z≈1from the CFRS survey[31].

6A population of hidden AGNs?

Recent dynamical evidence indicates that supermassive black holes reside at the center of most nearby galaxies.The available data(about30objects)show a strong correlation(but with a large scatter)between bulge and black hole mass[36],with M bh=0.006M bulge as a best-?t. The total mass density in spheroids today is?bulge=0.0036+0.0024

[14],implying a mean mass

?0.0017

density of dead quasars

ρbh=1.34+0.9

?0.6

×106M⊙Mpc?3.(6) Noting that the observed energy density from all quasars is equal to the emitted energy divided by the average quasar redshift[59],the total contribution to the EBL from accretion onto black

holes is

I bh=c3

1+z

≈18nW m?2sr?1η0.1 1+z ?1,(7)

whereη0.1is the e?ciency for transforming accreted rest-mass energy into radiation(in units of 10%).A population of AGNs at(say)z～1could then make a signi?cant contribution to the FIR background if dust-obscured accretion onto supermassive black holes is an e?cient process [20].

7Reionization of the IGM

The history of the transition from a neutral universe to one that is almost fully ionized can reveal the character of cosmological ionizing sources and constrain the star formation activity at high redshifts.The existence of a?lamentary,low-density intergalactic medium(IGM), which contains the bulk of the hydrogen and helium in the universe,is predicted as a product of primordial nucleosynthesis[8]and of hierarchical models of gravitational instability with “cold dark matter”(CDM)[6],[26].The application of the Gunn-Peterson constraint on the amount of smoothly distributed neutral material along the line of sight to distant objects requires the hydrogen component of the di?use IGM to have been highly ionized by z≈5[49], and the helium component by z≈2.5[9].From QSO absorption studies we also know that neutral hydrogen accounts for only a small fraction,～10%,of the nucleosynthetic baryons at early epochs[29].It thus appears that substantial sources of ultraviolet photons were present at z～>5,perhaps low-luminosity quasars[22]or a?rst generation of stars in virialized dark matter halos with T vir～104?105K[40],[21],[38].The existence of a decline in the space density of bright quasars at redshifts beyond～3was?rst suggested by[39],and has been since then the subject of a long-standing debate.In recent years,several optical surveys have consistently provided new evidence for a turnover in the QSO counts[23],[57],[48],[28].The interpretation of the drop-o?observed in optically selected samples is equivocal,however,because of the possible bias introduced by dust obscuration arising from intervening systems.Radio emission, on the other hand,is una?ected by dust,and it has recently been shown[50]that the space density of radio-loud quasars also decreases strongly for z>3.This argues that the turnover is indeed real and that dust along the line of sight has a minimal e?ect on optically-selected QSOs (Figure4,left panel).The QSO emission rate of hydrogen ionizing photons per unit comoving volume is shown in Figure4(right panel)[34].It is important to notice that the procedure adopted to derive this quantity implies a large correction for incompleteness at high-z.With a?t to the quasar luminosity function(LF)which goes asφ(L)∝L?1.64at the faint end[43], the contribution to the emissivity converges rather slowly,as L0.36.At z=4,for example,the blue magnitude at the break of the LF is M?≈?25.4,comparable or slightly fainter than the limits of current high-z QSO surveys.A large fraction,about90%at z=4and even higher at earlier epochs,of the ionizing quasar emissivity is therefore produced by sources that have not been actually observed,and are assumed to be present based on an extrapolation from lower redshifts.

Galaxies with ongoing star-formation are another obvious source of Lyman continuum pho-tons.Since the rest-frame UV continuum at1500?A(redshifted into the visible band for a

Figure4:Left:comoving space density of bright QSOs as a function of redshift.The data points with error bars are taken from[23](?lled dots),[57](?lled squares),[48](crosses),and[28](?lled pentagon).The empty triangles show the space density of the Parkes?at-spectrum radio-loud quasars with P>7.2×1026W Hz?1sr?1[25].Right:comoving emission rate of hydrogen Lyman-continuum photons(solid line)from QSOs,compared with the minimum rate(dashed line)which is needed to fully ionize a fast recombining(with gas clumping factor C=30)Einstein–de Sitter universe with ?b h250=0.08.Models based on photoionization by quasar sources appear to fall short at z=5. The data point shows the estimated contribution of star-forming galaxies at z≈3,assuming that the fraction of Lyman continuum photons which escapes the galaxy H I layers into the intergalactic medium is f esc=0.5(see[34]for details).

source at z≈3)is dominated by the same short-lived,massive stars which are responsible for the emission of photons shortward of the Lyman edge,the needed conversion factor,about one ionizing photon every10photons at1500?A,is fairly insensitive to the assumed IMF and is inde-pendent of the galaxy history for t?107yr.Figure4shows the estimated Lyman-continuum luminosity density of galaxies at z≈3.2The data point assumes a value of f esc=0.5for the unknown fraction of ionizing photons which escapes the galaxy H I layers into the intergalactic medium.A substantial population of dwarf galaxies below the detection threshold,i.e.having star-formation rates<0.3M⊙yr?1,and with a space density in excess of that predicted by extrapolating to faint magnitudes theα=1.38best-?t Schechter function,may be expected to form at early times in hierarchical clustering models,and has been recently proposed by[38] and[34]as a possible candidate for photoionizing the IGM at these epochs.One should note that,while highly reddened galaxies at high redshifts would be missed by the dropout color technique(which isolates sources that have blue colors in the optical and a sharp drop in the rest-frame UV),it seems unlikely that very dusty objects(with f esc?1)would contribute in any signi?cant manner to the ionizing metagalactic?ux.

As the hydrogen mean recombination timescale,ˉt rec,at high redshifts is much smaller than the then Hubble time[34],it is possible to compute at any given epoch a critical value for the photon emission rate per unit cosmological comoving volume,

˙N ion (z)=

ˉn H(0)

6 3

?b h250

6 3M⊙yr?1Mpc?3.(9)

(The conversion factor assumes a Salpeter IMF with solar metallicity).The star-formation density given in equation(9)is comparable with the value directly“observed”(i.e.,uncorrected for dust reddening)at z≈3[35].

References

[1]Barger,A.J.,et al.1998,Nature,394,248

[2]Baugh,C.M.,Cole,S.,Frenk,C.S.,&Lacey,C.G.1998,ApJ,498,504

[3]Bruzual,A.C.,&Charlot,S.1998,in preparation

[4]Burles,S.,&Tytler,D.1998,astro-ph/9803071

[5]Buzzoni,A.1995,ApJS,98,69

[6]Cen,R.,Miralda-Escud′e,J.,Ostriker,J.P.,&Rauch,M.1994,ApJ,437,L9

[7]Connolly,A.J.,et al.1997,ApJ,486,L11

[8]Copi,C.J.,Schramm,D.N.,&Turner,M.S.1994,Science,267,192

[9]Davidsen,A.F.,Kriss,G.A.,&Zheng,W.1996,Nature,380,47

[10]Dwek,E.,et al.1998,ApJ,508,106

[11]Ellis,R.S.,et al.1996,MNRAS,280,235

[12]Fixsen,D.J.,et al.1998,ApJ,508,123

[13]Flores,H.,et al.1998,ApJ,in press(astro-ph/9811202)

[14]Fukugita,M.,Hogan,C.J.,&Peebles,P.J.E.1998,ApJ,503,518

[15]Gallego,J.,Zamorano,J.,Arag′o n-Salamanca,A.,&Rego,M.1995,ApJ,455,L1

[16]Gardner,J.P.,Sharples,R.M.,Frenk,C.S.,&Carrasco,B.E.1997,ApJ,480,L99

[17]Glazebrook,K.,et al.1998,MNRAS,submitted(astro-ph/9808276)

[18]Gnedin,N.Y.,&Ostriker,J.P.1997,ApJ,486,581

[19]Guiderdoni,B.,et al.1997,Nature,390,257

[20]Haehnelt,M.G.,Natarajan,P.,&Rees,M.J.1998,MNRAS,300,817

[21]Haiman,Z.,&Loeb,A.1997,ApJ,483,21

[22]Haiman,Z.,&Loeb,A.1998,ApJ,in press

[23]Hartwick,F.D.A.,&Schade,D.1990,ARA&A,28,437

[24]Hauser,M.G.,et al.1998,ApJ,508,25

[25]Hook,I.M.,Shaver,P.A.,&McMahon,R.G.1998,in The Young Universe:Galaxy Formation

and Evolution at Intermediate and High Redshift,ed.S.D’Odorico,A.Fontana,&E.Giallongo (San Francisco:ASP),in press

[26]Hernquist,L.,Katz,N.,Weinberg,D.H.,&Miralda-Escud′e,J.1996,ApJ,457,L51

[27]Hughes,D.,et al.1998,Nature,398,241

[28]Kenne?ck,J.D.,Djorgovski,S.G.,&de Carvalho,R.R.1995,AJ,110,2553

[29]Lanzetta,K.M.,Wolfe,A.M.,&Turnshek,D.A.1995,ApJ,440,435

[30]Leitherer,C.,Ferguson,H.C.,Heckman,T.M.,&Lowenthal,J.D.1995,ApJ,454,L19

[31]Lilly,S.J.,Le F′e vre,O.,Hammer,F.,&Crampton,D.,1996,ApJ,460,L1

[32]Lilly,S.J.,et al.1998,astro-ph/9807261

[33]Madau,P.,et al.1996,MNRAS,283,1388

[34]Madau,P.,Haardt,F.,&Rees,M.J.1998,ApJ,in press

[35]Madau,P.,Pozzetti,L.,&Dickinson,M.E.1998,ApJ,498,106

[36]Magorrian,G.,et al.1998,AJ,115,2285

[37]Meurer,G.R.,et al.1997,AJ,114,54

[38]Miralda-Escud′e,J.,&Rees,M.J.1998,ApJ,497,21

[39]Osmer,P.S.1982,ApJ,253,280

[40]Ostriker,J.P.,&Gnedin,N.Y.1996,ApJ,472,L63

[41]Pascarelle,S.M.,Lanzetta,K.M.,&Fernandez-Soto,A.1998,ApJ,in press(astro-ph/9810060)

[42]Pettini,M.,et al.1998,ApJ,in press(astro-ph/9806219)

[43]Pei,Y.C.1995,ApJ,438,623

[44]Pozzetti,L.,et al.1998,MNRAS,298,1133

[45]Rees,M.J.1996,astro-ph/9608196

[46]Rowan-Robinson,M.,et al.1997,MNRAS,289,490

[47]Sawicki,M.,&Yee,H.K.C.1998,AJ,115,1329

[48]Schmidt,M.,Schneider,D.P.,&Gunn,J.E.1995,AJ,110,68

[49]Schneider,D.P.,Schmidt,M.,&Gunn,J.E.1991,AJ,101,2004

[50]Shaver,P.A.,et al.1996,Nature,384,439

[51]Songaila,A.1997,ApJ,490,L1

[52]Steidel,C.C.,et al.1996,ApJ,462,L17

[53]Steidel,C.C.,et al.1998,ApJ,492,428

[54]Steidel,C.C.,et al.1998,ApJ,submitted(astro-ph/9811399)

[55]Tresse,L.,&Maddox,S.J.1998,ApJ,495,691

[56]Treyer,M.A.,et al.1998,MNRAS,in press(astro-ph/9806056)

[57]Warren,S.J.,Hewett,P.C.,&Osmer,P.S.1994,ApJ,421,412

[58]Williams,R.E.,et al.1996,AJ,112,1335

[59]Zoltan,A.1982,MNRAS,200,115