The first galaxies: assembly, cooling and the onset of turbulence

We investigate the properties of the first galaxies at   z ≳ 10  with highly resolved numerical simulations, starting from cosmological initial conditions and taking into account all relevant primordial chemistry and cooling. A first galaxy is characterized by the onset of atomic hydrogen cooling, once the virial temperature exceeds  ≃10⁴ K  , and its ability to retain photoheated gas. We follow the complex accretion and star formation history of a  ≃5 × 10⁷ M_⊙  system by means of a detailed merger tree and derive an upper limit on the number of Population III (Pop III) stars formed prior to its assembly. We investigate the thermal and chemical evolution of infalling gas and find that partial ionization at temperatures  ≳10⁴ K  catalyses the formation of  H₂  and hydrogen deuteride, allowing the gas to cool to the temperature of the cosmic microwave background. Depending on the strength of radiative and chemical feedback, primordial star formation might be dominated by intermediate-mass Pop III stars formed during the assembly of the first galaxies. Accretion on to the nascent galaxy begins with hot accretion, where gas is accreted directly from the intergalactic medium and shock heated to the virial temperature, but is quickly accompanied by a phase of cold accretion, where the gas cools in filaments before flowing into the parent halo with high velocities. The latter drives supersonic turbulence at the centre of the galaxy and could lead to very efficient chemical mixing. The onset of turbulence in the first galaxies thus likely marks the transition to Pop II star formation.     

Impact of cosmic rays on Population III star formation

We explore the implications of a possible cosmic-ray (CR) background generated during the first supernova explosions that end the brief lives of massive Population III stars. We show that such a CR background could have significantly influenced the cooling and collapse of primordial gas clouds in minihaloes around redshifts of   z ∼ 15–20  , provided the CR flux was sufficient to yield an ionization rate greater than about 10⁻¹⁹ s⁻¹ near the centre of the minihalo. The presence of CRs with energies  ≲10⁷  eV would indirectly enhance the molecular cooling in these regions, and we estimate that the resulting lower temperatures in these minihaloes would yield a characteristic stellar mass as low as  ∼10 M_⊙  . CRs have a less-pronounced effect on the cooling and collapse of primordial gas clouds inside more massive dark matter haloes with virial masses  ≳10⁸ M_⊙  at the later stages of cosmological structure formation around   z ∼ 10–15  . In these clouds, even without CR flux the molecular abundance is already sufficient to allow cooling to the floor set by the temperature of the cosmic microwave background.     

The cooling of shock-compressed primordial gas

We find that at redshifts   z ≳ 10, HD  line cooling allows strongly shocked primordial gas to cool to the temperature of the cosmic microwave background (CMB). This temperature is the minimum value attainable via radiative cooling. Provided that the abundance of HD, normalized to the total number density, exceeds a critical level of  ∼10⁻⁸  , the CMB temperature floor is reached in a time which is short in comparison to the Hubble time. We estimate the characteristic masses of stars formed out of shocked primordial gas in the wake of the first supernovae, and resulting from the virialization of dark matter haloes during hierarchical structure formation to be  ∼10 M_⊙  . In addition, we show that cooling by HD enables the primordial gas in relic H  ii regions to cool to temperatures considerably lower than those reached via H₂ cooling alone. We confirm that HD cooling is unimportant in cases where the primordial gas does not go through an ionized phase, as in the formation process of the very first stars in   z ≳ 20  minihaloes of mass  ∼10⁶ M_⊙  .     

Two populations of metal-free stars in the early Universe

We construct star formation histories at redshifts z ≳ 5 for two physically distinct populations of primordial, metal-free stars, motivated by theoretical and observational arguments that have hinted towards the existence of an intermediate stellar generation between Population III and Population I/II. Taking into account the cosmological parameters as recently revised by the Wilkinson Microwave Anisotropy Probe after three years of operation, we determine self-consistent reionization histories and discuss the resulting chemical enrichment from these early stellar generations. We find that the bulk of ionizing photons and heavy elements produced at high redshifts must have originated in Population II.5 stars, which formed out of primordial gas in haloes with virial temperatures ≳10⁴ K, and had typical masses ≳10 M_⊙. Classical Population III stars, formed in minihaloes and having masses ≳100 M_⊙, on the other hand, had only a minor impact on reionization and early metal enrichment. Specifically, we conclude that only ≃10 per cent by mass of metal-free star formation went into Population III.     

Baryonic conversion tree: the global assembly of stars and dark matter in galaxies from the Sloan Digital Sky Survey

Using the spectroscopic sample of the Sloan Digital Sky Survey Data Release 1 (SDSS DR1), we measure how gas was transformed into stars as a function of time and stellar mass: the baryonic conversion tree (BCT). There is a clear correlation between early star formation activity and present-day stellar mass: the more massive galaxies have formed approximately 80 per cent of their stars at   z > 1  , while for the less massive ones the value is only approximately 20 per cent. By comparing the BCT with the dark matter merger tree, we find indications that star formation efficiency at   z > 1  had to be approximately a factor of two higher than today (∼10 per cent) in galaxies with present-day stellar mass larger than  2 × 10¹¹ M_⊙  , if this early star formation occurred in the main progenitor. Therefore, the λ cold dark matter (LCDM) paradigm can accommodate a large number of red objects. On the other hand, in galaxies with present-day stellar mass less than  10¹¹ M_⊙  , efficient star formation seems to have been triggered at   z ∼ 0.2  . We show that there is a characteristic mass  ( M _*∼ 10¹⁰ M_⊙)  for feedback efficiency (or lack of star formation). For galaxies with masses lower than this, feedback (or star formation suppression) is very efficient while for higher masses it is not. The BCT, determined here for the first time, should be an important observable with which to confront theoretical models of galaxy formation.     

Large-scale galactic turbulence: can self-gravity drive the observed H i velocity dispersions?

Observations of turbulent velocity dispersions in the H  i component of galactic discs show a characteristic floor in galaxies with low star formation rates and within individual galaxies the dispersion profiles decline with radius. We carry out several high-resolution adaptive mesh simulations of gaseous discs embedded within dark matter haloes to explore the roles of cooling, star formation, feedback, shearing motions and baryon fraction in driving turbulent motions. In all simulations the disc slowly cools until gravitational and thermal instabilities give rise to a multiphase medium in which a large population of dense self-gravitating cold clouds are embedded within a warm gaseous phase that forms through shock heating. The diffuse gas is highly turbulent and is an outcome of large-scale driving of global non-axisymmetric modes as well as cloud–cloud tidal interactions and merging. At low star formation rates these processes alone can explain the observed H  i velocity dispersion profiles and the characteristic value of  ∼10 km s⁻¹  observed within a wide range of disc galaxies. Supernovae feedback creates a significant hot gaseous phase and is an important driver of turbulence in galaxies with a star formation rate per unit area  ≳10⁻³ M_⊙ yr⁻¹ kpc⁻²  .     

Population III stars: hidden or disappeared?

A Population III/Population II transition from massive to normal stars is predicted to occur when the metallicity of the star-forming gas crosses the critical range   Z _cr= 10^−5±1 Z_⊙  . To investigate the cosmic implications of such a process, we use numerical simulations which follow the evolution, metal enrichment and energy deposition of both Population II and Population III stars. We find that: (i) due to inefficient heavy element transport by outflows and slow 'genetic' transmission during hierarchical growth, large fluctuations around the average metallicity arise; as a result, Population III star formation continues down to   z = 2.5  , but at a low peak rate of  10⁻⁵ M_⊙ yr⁻¹ Mpc⁻³  occurring at   z ≈ 6  (about 10⁻⁴ of the Population II one); and (ii) Population III star formation proceeds in an 'inside–out' mode in which formation sites are progressively confined to the periphery of collapsed structures, where the low gas density and correspondingly long free-fall time-scales result in a very inefficient astration. These conclusions strongly encourage deep searches for pristine star formation sites at moderate  (2 < z < 5)  redshifts where metal-free stars are likely to be hidden.     

The local star formation rate and radio luminosity density

We present a new determination of the local volume-averaged star formation rate from the 1.4-GHz luminosity function of star forming galaxies. Our sample, taken from the   B ≤12  Revised Shapley–Ames catalogue (231 normal spiral galaxies over an effective area of 7.1 sr) has ≃100 per cent complete radio detections and is insensitive to dust obscuration and cirrus contamination. After removal of known active galaxies, the best-fitting Schechter function has a faint-end slope of  −1.27±0.07  in agreement with the local H α luminosity function, characteristic luminosity   L ∗=(2.6±0.7)×10²² W Hz⁻¹  and density   φ ∗=(4.8±1.1)×10⁻⁴ Mpc⁻³.  The inferred local radio luminosity density of  (1.73±0.37±0.03)×10¹⁹ W Hz⁻¹ Mpc⁻³  (Poisson noise, large-scale structure fluctuations) implies a volume-averaged star formation rate ∼2 times larger than the Gallego et al. H α estimate, i.e.   ρ _1.4 GHz=(2.10±0.45±0.04)×10⁻² M_⊙ yr⁻¹ Mpc⁻³  for a Salpeter initial mass function from  0.1–125 M_⊙  and Hubble constant of 50 km s⁻¹ Mpc⁻¹. We demonstrate that the Balmer decrement is a highly unreliable extinction estimator, and argue that optical–ultraviolet (UV) star formation rates (SFRs) are easily underestimated, particularly at high redshift.     

The census of nuclear activity of late-type galaxies in the Virgo cluster

The first spectroscopic census of active galactic nuclei (AGNs) associated with late-type galaxies in the Virgo cluster was carried out by observing 213 out of a complete set of 237 galaxies more massive than   M _dyn > 10^8.5 M_⊙  . Among them, 77 are classified as AGNs [including 21 transition objects, 47 low-ionization nuclear emission regions (LINERs) and nine Seyferts] and comprise 32 per cent of the late-type galaxies in Virgo. Due to spectroscopic incompleteness, at most 21 AGNs are missed in the survey, so that the fraction would increase up to 41 per cent. Using corollary near-infrared observations that enable us to estimate galaxy dynamical masses, it is found that AGNs are hosted exclusively in massive galaxies, i.e.   M _dyn≳ 10¹⁰ M_⊙  . Their frequency increases steeply with the dynamical mass from zero at   M _dyn≈ 10^9.5 M_⊙  to virtually 1 at   M _dyn > 10^11.5 M_⊙  . These frequencies are consistent with those of low-luminosity AGNs found in the general field by the Sloan Digital Sky Survey. Massive galaxies that harbour AGNs commonly show conspicuous r -band star-like nuclear enhancements. Conversely, they often, but not necessarily, contain massive bulges. A few well-known AGNs (e.g. M61, M100, NGC 4535) are found in massive Sc galaxies with little or no bulge. The AGN fraction seems to be only marginally sensitive to galaxy environment. We infer the black hole masses using the known scaling relations of quiescent black holes. No black holes lighter than  ∼10⁶ M_⊙  are found active in our sample.     

Dry mergers: a crucial test for galaxy formation

We investigate the role that dry mergers play in the build-up of massive galaxies within the cold dark matter paradigm. Implementing an empirical shut-off mass scale for star formation, we find a nearly constant dry merger rate of  ∼6 × 10⁻⁵ Mpc⁻³ Gyr⁻¹  at   z ≤ 1  and a steep decline at larger z . Less than half of these mergers are between two galaxies that are morphologically classified as early-types, and the other half is mostly between an early- and late-type galaxy. Latter are prime candidates for the origin of tidal features around red elliptical galaxies. The introduction of a transition mass scale for star formation has a strong impact on the evolution of galaxies, allowing them to grow above a characteristic mass scale of   M _{*, c} ∼ 6.3 × 10¹⁰ M_⊙  by mergers only. As a consequence of this transition, we find that around   M _{*, c}   , the fraction of 1:1 mergers is enhanced with respect to unequal mass major mergers. This suggests that it is possible to detect the existence of a transition mass scale by measuring the relative contribution of equal mass mergers to unequal mass mergers as a function of galaxy mass. The evolution of the high-mass end of the luminosity function is mainly driven by dry mergers at low z . We however find that only 10–20 per cent of galaxies more massive than   M _{*, c}   experience dry major mergers within their last Gyr at any given redshift   z ≤ 1  .     

On the formation of massive galaxies: a simultaneous study of number density, size and intrinsic colour evolution in GOODS

The evolution of number density, size and intrinsic colour is determined for a volume-limited sample of visually classified early-type galaxies selected from the Hubble Space Telescope /Advanced Camera for Surveys images of the Great Observatories Origins Deep Survey (GOODS) North and South fields (version 2). The sample comprises 457 galaxies over 320 arcmin² with stellar masses above  3 × 10¹⁰ M_⊙  in the redshift range  0.4 < z < 1.2  . Our data allow a simultaneous study of number density, intrinsic colour distribution and size. We find that the most massive systems  (≳3 × 10¹¹ M_⊙)  do not show any appreciable change in comoving number density or size in our data. Furthermore, when including the results from 2dF galaxy redshift survey, we find that the number density of massive early-type galaxies is consistent with no evolution between   z = 1.2  and 0, i.e. over an epoch spanning more than half of the current age of the Universe. We find large discrepancies between the predictions of semi-analytic models. Massive galaxies show very homogeneous intrinsic colour distributions, with nearly flat radial colour gradients, but with a significant negative correlation between stellar mass and colour gradient, such that red cores appear predominantly in massive galaxies. The distribution of half-light radii – when compared to   z ∼ 0  and   z > 1  samples – is compatible with the predictions of semi-analytic models relating size evolution to the amount of dissipation during major mergers.     

Recurrent gas accretion by massive star clusters, multiple stellar populations and mass thresholds for spheroidal stellar systems

We explore the gravitational influence of pressure-supported stellar systems on the internal density distribution of a gaseous environment. We conclude that compact massive star clusters with masses  ≳10⁶ M_⊙  act as cloud condensation nuclei and are able to accrete gas recurrently from a warm interstellar medium which may cause further star formation events and account for multiple stellar populations in the most massive globular and nuclear star clusters. The same analytical arguments can be used to decide whether an arbitrary spherical stellar system is able to keep warm or hot interstellar material or not. These mass thresholds coincide with transition masses between pressure supported galaxies of different morphological types.     

Quantifying the coexistence of massive black holes and dense nuclear star clusters

In large spheroidal stellar systems, such as elliptical galaxies, one invariably finds a  10⁶–10⁹ M_⊙  supermassive black hole at their centre. In contrast, within dwarf elliptical galaxies one predominantly observes a  10⁵–10⁷ M_⊙  nuclear star cluster. To date, few galaxies have been found with both types of nuclei coexisting and even less have had the masses determined for both central components. Here, we identify one dozen galaxies housing nuclear star clusters and supermassive black holes whose masses have been measured. This doubles the known number of such hermaphrodite nuclei – which are expected to be fruitful sources of gravitational radiation. Over the host spheroid (stellar) mass range  10⁸–10¹¹ M_⊙  , we find that a galaxy's nucleus-to-spheroid (baryon) mass ratio is not a constant value but decreases from a few per cent to ∼0.3 per cent such that  log[( M _BH+ M _NC)/ M _sph]=−(0.39 ± 0.07) log[ M _sph/10¹⁰ M_⊙]− (2.18 ± 0.07)  . Once dry merging commences and the nuclear star clusters disappear, this ratio is expected to become a constant value.
As a byproduct of our investigation, we have found that the projected flux from resolved nuclear star clusters is well approximated with Sérsic functions having a range of indices from ∼0.5 to ∼3, the latter index describing the Milky Way's nuclear star cluster.     

Hyperluminous infrared galaxies

39 galaxies are now known, from follow-up of faint IRAS sources and from submillimetre observations of high-redshift AGN, with far-infrared luminosities >10¹³ L_⊙. 13 of these, which have been found in 60- or 850-μm surveys, form an important unbiased subsample. 12 have been found by comparison of 60-μm surveys with quasar or radio galaxy catalogues, or from infrared surveys with colour selection biased towards AGN, while a further 14 have been found through submillimetre observations of known high-redshift AGN. In this paper I argue, on the basis of detailed modelling of the spectral energy distributions of hyperluminous galaxies with accurate radiative transfer models, and from evidence of high gas mass in several cases, that the bulk of the emission from these galaxies at rest frame wavelengths ≥50 μm is caused by star formation. Even after correction for the effects of lensing, hyperluminous galaxies with emission peaking at rest frame wavelengths ≥50 μm are therefore undergoing star formation at rates >10³ M_⊙ yr⁻¹ and are strong candidates for being primeval galaxies, in the process of a major episode of star formation.     

Ly+ and ultraviolet emission from high-redshift gamma-ray burst hosts: to what extent do gamma-ray bursts trace star formation?

We report the result of a search for Lyα emission from the host galaxies of the gamma-ray bursts  (GRBs) 030226 ( z = 1.986), 021004 ( z = 2.335)  and  020124 ( z = 3.198)  . We find that the host galaxy of GRB 021004 is an extended (around 8 kpc) strong Lyα emitter with a rest-frame equivalent width (EW) of 68⁺¹²₋₁₁Å, and a star formation rate of  10.6 ± 2.0 M_⊙ yr⁻¹  . We do not detect the hosts of GRB 030226 and GRB 020124, but the upper limits on their Lyα fluxes do not rule out large rest-frame EWs. In the fields of GRB 021004 and GRB 030226 we find seven and five other galaxies, respectively, with excess emission in the narrow-band filter. These galaxies are candidate Lyα-emitting galaxies in the environment of the host galaxies. We have also compiled a list of all   z ≳ 2  GRB hosts, and demonstrate that a scenario where they trace star formation in an unbiased way is compatible with current observational constraints. Fitting the   z = 3  luminosity function (LF) under this assumption results in a characteristic luminosity of   R *= 24.6  and a faint-end slope of  α=−1.55  , consistent with the LF measured for Lyman-break galaxies.     

On the origin of high-velocity runaway stars

We explore the hypothesis that some high-velocity runaway stars attain their peculiar velocities in the course of exchange encounters between hard massive binaries and a very massive star (either an ordinary  50–100 M_⊙  star or a more massive one, formed through runaway mergers of ordinary stars in the core of a young massive star cluster). In this process, one of the binary components becomes gravitationally bound to the very massive star, while the second one is ejected, sometimes with a high speed. We performed three-body scattering experiments and found that early B-type stars (the progenitors of the majority of neutron stars) can be ejected with velocities of  ≳200–400 km s⁻¹  (typical of pulsars), while  3–4 M_⊙  stars can attain velocities of  ≳300–400 km s⁻¹  (typical of the bound population of halo late B-type stars). We also found that the ejected stars can occasionally attain velocities exceeding the Milky Ways's escape velocity.     

Simulating the formation of a protocluster at z∼ 2

We present results from two high-resolution hydrodynamical simulations of protocluster regions at   z ≃ 2.1  . The simulations have been compared to observational results for the so-called Spiderweb galaxy system, the core of a putative protocluster region at   z = 2.16  , found around a radio galaxy. The simulated regions have been chosen so as to form a poor cluster with   M ₂₀₀≃ 10¹⁴  h ⁻¹ M_⊙  (C1) and a rich cluster with   M ₂₀₀≃ 2 × 10¹⁵  h ⁻¹ M_⊙  (C2) at   z = 0  . The simulated protoclusters show evidence of ongoing assembly of a dominating central galaxy. The stellar mass of the brightest cluster galaxy of the C2 system is in excess with respect to observational estimates for the Spiderweb galaxy, with a total star formation rate which is also larger than indicated by observations. We find that the projected velocities of galaxies in the C2 cluster are consistent with observations, while those measured for the poorer cluster C1 are too low compared with the observed velocities. We argue that the Spiderweb complex resembles the high-redshift progenitor of a rich galaxy cluster. Our results indicate that the included supernovae feedback is not enough to suppress star formation in these systems, supporting the need of introducing active galactic nuclei feedback. According to our simulations, a diffuse atmosphere of hot gas in hydrostatic equilibrium should already be present at this redshift, and enriched at a level comparable to that of nearby galaxy clusters. The presence of this gas should be detectable with future deep X-ray observations.     

Faint Lyman-break galaxies as a crucial test for galaxy formation models

It has recently been shown that galaxy formation models within the Λ cold dark matter cosmology predict that, compared to the observed population, small galaxies (with stellar masses  <10¹¹ M_⊙  ) form too early, are too passive since   z ∼ 3  and host too old stellar populations at   z = 0  . We then expect an overproduction of small galaxies at   z ≳ 4  that should be visible as an excess of faint Lyman-break galaxies. To check whether this excess is present, we use the morgana galaxy formation model and grasil spectrophotometric  +  radiative transfer code to generate mock catalogues of deep fields observed with Hubble Space Telescope Advanced Camera for Surveys. We add observational noise and the effect of Lyman α emission, and perform colour–colour selections to identify Lyman-break galaxies. The resulting mock candidates have plausible properties that closely resemble those of observed galaxies. We are able to reproduce the evolution of the bright tail of the luminosity function of Lyman-break galaxies (with a possible underestimate of the number of the brightest i -dropouts), but uncertainties and degeneracies in dust absorption parameters do not allow to give strong constraints to the model. Besides, our model shows a clear excess with respect to observations of faint Lyman-break galaxies, especially of   z ₈₅₀∼ 27 V   -dropouts at   z ∼ 5  . We quantify the properties of these 'excess' galaxies and discuss the implications: these galaxies are hosted in dark matter haloes with circular velocities in excess of 100 km s⁻¹, and their suppression may require a deep rethinking of stellar feedback processes taking place in galaxy formation.     

The SCUBA HAlf Degree Extragalactic Survey – VI. 350-μm mapping of submillimetre galaxies

A follow-up survey using the Submillimetre High-Angular Resolution Camera (SHARC-II) at 350 μm has been carried out to map the regions around several 850-μm-selected sources from the Submillimetre HAlf Degree Extragalactic Survey (SHADES). These observations probe the infrared (IR) luminosities and hence star formation rates in the largest existing, most robust sample of submillimetre galaxies (SMGs). We measure 350-μm flux densities for 24 850-μm sources, seven of which are detected at ≥2.5σ within a 10 arcsec search radius of the 850-μm positions. When results from the literature are included the total number of 350-μm flux density constraints of SHADES SMGs is 31, with 15 detections. We fit a modified blackbody to the far-IR (FIR) photometry of each SMG, and confirm that typical SMGs are dust-rich  ( M _dust≃ 9 × 10⁸ M_⊙)  , luminous  ( L _FIR≃ 2 × 10¹² L_⊙)  star-forming galaxies with intrinsic dust temperatures of ≃35 K and star formation rates of  ≃400 M_⊙ yr⁻¹  . We have measured the temperature distribution of SMGs and find that the underlying distribution is slightly broader than implied by the error bars, and that most SMGs are at 28 K with a few hotter. We also place new constraints on the 350-μm source counts, N ₃₅₀(>25 mJy) ∼ 200–500 deg⁻².     

The SCUBA Half Degree Extragalactic Survey (SHADES) – IX. The environment, mass and redshift dependence of star formation

We present a comparison between the SCUBA (Submillimetre Common User Bolometer Array) Half Degree Extragalactic Survey (SHADES) at 450 and  850 μm  in the Lockman Hole East with a deep Spitzer Space Telescope survey at  3.6–24 μm  conducted in guaranteed time. Using stacking analyses we demonstrate a striking correspondence between the galaxies contributing the submm extragalactic background light, with those likely to dominate the backgrounds at Spitzer wavelengths. Using a combination BRIzK plus Spitzer photometric redshifts, we show that at least a third of the Spitzer -identified submm galaxies at  1 < z < 1.5  appear to reside in overdensities when the density field is smoothed at 0.5–2 Mpc comoving diameters, supporting the high-redshift reversal of the local star formation–galaxy density relation. We derive the dust-shrouded cosmic star formation history of galaxies as a function of assembled stellar masses. For model stellar masses  <10¹¹ M_⊙  , this peaks at lower redshifts than the ostensible   z ∼ 2.2  maximum for submm point sources, adding to the growing consensus for 'downsizing' in star formation. Our surveys are also consistent with 'downsizing' in mass assembly. Both the mean star formation rates  〈d M _*/d t 〉  and specific star formation rates  〈(1/ M _*) d M _*/d t 〉  are in striking disagreement with some semi-analytic predictions from the Millenium Simulation. The discrepancy could either be resolved with a top-heavy initial mass function, or a significant component of the submm flux heated by the interstellar radiation field.