arXiv subjects

Renyue Cen

Publications and source records attributed to Renyue Cen.

At least 199 records · Page 11Linked to original sources

arXiv2003

Cosmological Mestel Disks and the Rossby Vortex Instability: The Origin of Supermassive Black Holes

A scenario is put forth for the formation of supermassive black holes at the centers of galaxies. It depends upon the formation of a Mestel disk with a flat rotation curve, where the mass is proportional to r and thickness to 1/r. Such disks should form from the collapse of uniformly rotating, isolated, gaseous clouds, either proto-galactic, galaxy-mass damped Lyman alpha clouds or the gas that survives galaxy mergers. We propose that in any case the disk will be unstable to the Rossby vortex instability (RVI). This instability grows from any large, steep pressure gradient in an optically thick disk. Such pressure gradients either occur adjacent to compact objects or could be triggered by individual supernovae in and around the disk. Upon excitation, the RVI transports angular momentum far more efficiently than the turbulence of a viscosity-based Shakura-Sunyaev disk. The critical column density condition in a L* Mestel disk predicts that the black hole mass is ~ 10^8 Msolar and is also proportional to the 1/4 power of the velocity dispersion, in accordance with observations.

astro-ph↗

arXiv2003

Galaxies Inside Stromgren Spheres of Luminous Quasars at z>6: Detection of The First Galaxies

The intrinsic Lyman-alpha emission lines of normal galaxies before reionization are much absorbed by the damping wing of the Gunn-Peterson trough, rendering their direct detection nearly impossible, if their intrinsic line widths are less than ~100km/s. High redshift luminous quasars prior to the completion of cosmological reionization at z~6, on the other hand, are capable of producing large HII regions around them (Stromgren spheres) to allow their intrinsic Lyman-alpha emission lines to be transmitted without overwhelming absorption (Cen & Haiman 2000). We suggest that targeted observations at the Stromgren spheres of known luminous quasars at z >= 6 would be able to detect Lyman-alpha emission lines of galaxies inside the Stromgren spheres largely unattenuated. A tunable, very narrowband filter of Δλ\over λ~ 0.1% or a narrowband filter of Δλ\over λ~1% with follow-up spectroscopic identifications will be required. Such observations could directly observe the sources of cosmological reionization including possibly the Pop III galaxies at z=6-20 by JWST. Possible applications include determinations of the ionization state of the intergalactic medium, the sizes of the Stromgren spheres, the ages of the quasars, the luminosity function of high redshift galaxies and its evolution, the spatial distribution of galaxies and its evolution, the biased distribution of galaxies around quasars and the anisotropy of quasar emission. Observations using Keck-class telescopes may already be made to enable a differentiation between a fully neutral and a 10% neutral intergalactic medium at z>6.

astro-ph↗

arXiv2003

Metallicity Evolution of Damped Lyman Alpha Systems In Lambda CDM Cosmology

Utilizing a new, high mass resolution hydrodynamic simulation we compute the metallicity evolution of damped Lyman alpha systems (DLAs) and find a reasonable agreement with observations. In particular, the observed slow evolution of the DLA metallicity occurs naturally in the simulation due to the combined effects of physical and observational selection. The slow metallicity evolution is caused by the steady transformation, with increasing time, of the highest metallicity systems to "galaxies", thus depleting this category, while all the lower metallicity systems show, individually, an increase in metallicity. Although the trend of DLA metallicity with redshift is in good agreement with observations, it appears that the average metallicity of simulated DLAs is higher than observed by 0.3-0.5 dex in the probed redshift range (z=0-5). Our study indicates that this difference may be attributed to observational selection effects due to dust obscuration. If we allow for a dust obscuration effect, our model reproduces the observed metallicity evolution in both amplitude and slope. We find that DLAs are not a simple population but probe a range of different systems and the mix changes with redshift. About 50% of all metals in the gaseous phase is in DLAs at all times from z=5 to z=1, making a rapid downturn at z\le 1 to ~20% by z=0, as metals are swept into the hotter components of the IGM as well as locked up in stars. While not the primary focus of this study, we find that the model provides good matches to observations with respect to column density distribution and evolution of neutral gas content, if the same dust obscuration is taken into account. We find Omega_{DLA,comp}=(1-3)E-3, depending on the effect of dust obscuration.

astro-ph↗

arXiv2003

The Probability Distribution Function of Light in the Universe: Results from Hydrodynamic Simulations

While second and higher order correlations of the light distribution have received extensive study, the lowest order probability distribution function (PDF) -- the probability that a unit volume of space will emit a given amount of light -- has received very little attention. We estimate this function with the aid of hydrodynamic simulations of the Lambda CDM model, finding it significantly different from the mass density PDF, and not simply related to it by linear bias or any of the other prescriptions commonly adopted. If the optical light PDF is, in reality, similar to what we find in the simulations, then some measures of Omega_M based on mass-to-light ratio and the cosmic virial theorem will have significantly underestimated Omega_M. Basically, the problem is one of selection bias, with galaxy forming regions being unrepresentative of the dark matter distribution in a way not described by linear bias. Knowledge of the optical PDF and the plausible assumption of a log-normal distribution for the matter PDF will allow one to correct for these selection biases. We find that this correction (which amounts to 20-30%) brings the values of Omega_M estimated by using the mass-to-light ratio and the cosmic virial theorem to the range 0.2-0.3, in better agreement with the WMAP result than the uncorrected estimates. In addition, the relation between mass and light PDFs gives us insight concerning the nature of the void phenomenon. In particular our simulation indicates that 20% of mass is distributed in voids, which occupy 85% of volume in the universe.

astro-ph↗

arXiv2003

Implications of WMAP Observations On the Population III Star Formation Processes

In an earlier paper (Cen 2003) we pointed out the strong likelihood for the universal reionization to occur twice, giving rise to a larger Thomson optical depth. Here we perform a more focused analysis of the Thomson optical depth in light of the WMAP observations. While the current uncertainties on the observed Thomson optical depth are still relatively large, with tau_e=0.17+-0.04 (68%) (Kogut et al 2003), important implications on Pop III star formation processes at high redshift can already be inferred. We are able to draw four conclusions: (1) in the absence of a top-heavy initial stellar mass function (IMF) for Pop III metal-free stars and without a dramatic upturn in the star formation efficiency and ionizing photon escape fraction at high redshift (z>6), we find tau_e =< 0.09; (2) with a top-heavy IMF for the Pop III metal-free stars and plausible star formation efficiency and ionizing photon escape fraction, it is expected that tau_e =< 0.12; (3) it is possible to reach tau_e = 0.15, if the metal enrichment efficiency of the intergalactic medium by Pop III stars is very low thus Pop III era is prolonged; (4) to reach tau_e >= 0.17 requires either of the following two conditions: the cosmological model power index n is positively tilted to n >= 1.03, Pop III star formation in minihalos with molecular hydrogen cooling has an efficiency c_*(H_2,III)>0.01 (with ionizing photon escape fraction greater than 30%). If the current observed value of Thomson optical depth withstands future data, we will have strong observational evidence that Pop III stars are massive and their formation efficiency may be much higher than current theoretical works suggest. Alternatively, there may be unknown, non-stellar ionizing sources at very high redshift.

astro-ph↗

arXiv2003

The Universe Was Reionized Twice

We show the universe was reionized twice, first at z~15-16 and second at z~6. Such an outcome appears inevitable, when normalizing to two well determined observational measurements, namely, the epoch of the final cosmological reionization at z~6 and the density fluctuations at z~6, which in turn are tight ly constrained by Lyman alpha forest observations at z~3. These two observations most importantly fix the product of star formation efficiency and ionizing photon escape fraction from galaxies at high redshift. To the extent that the relative star formation efficiencies in gaseous minihalos with H2 cooling and large halos with atomic cooling at high redshift are still unknown, the primary source for the first reionization could be Pop III stars either in minihalos or in large halos. We show that gas in minihalos can be cooled efficiently by H2 molecules and star formation can continue to take place largely unimpeded throughout the first reionization period, thanks to two new mechanisms for generating a high X-ray background during the Pop III era, put forth here. Moreover, an important process for producing a large number of H2 molecules in relic HII regions of Pop III galaxies, first pointed out by Ricotti, Gnedin, & Shull, is quantified here. It is shown that the Lyman-Werner background may never build up during the Pop III era. The long cosmological reionization and reheating history is complex. We discuss a wide range of implications and possible tests for this new reionization picture. In particular, Thomson scattering optical depth is increased to 0.10 +- 0.03, compared to 0.027 for the case of only one rapid reionization at z=6. Upcoming Microwave Anisotropy Probe observation of the polarization of the cosmic microwave background should be able to distinguish between these two scenarios.

astro-ph↗

arXiv2002

A Constraint on the Gravitational Lensing Magnification and Age of the Redshift z=6.28 Quasar SDSS 1030+0524

The recent discovery of bright quasars around redshift z=6 suggests that black holes (BHs) with masses in excess of 10^9 Msun have already assembled at a very early stage in the evolution of the universe. An alternative interpretation is that these quasars are powered by less massive BHs, but their fluxes are strongly magnified through gravitational lensing by intervening galaxies. Here we analyze the flux distribution of the Ly alpha emission of the quasar with the highest known redshift, SDSS 1030+0524, at z=6.28. We show that this object could not have been magnified by lensing by more than a factor of five. The constraint arises from the large observed size, 30 (comoving) Mpc, of the ionized region around this quasar, and relies crucially only on the assumption that the quasar is embedded in a largely neutral IGM. Based on the line/continuum ratio of SDSS 1030+0524, we argue further that this quasar also cannot be beamed by a significant factor. We conclude that the minimum mass for its resident BH is 4 x 10^8 Msun (for magnification by a factor of five); if the mass is this low, then the quasars had to switch on prior to redshift z=9. From the size of the ionized region, we are also able to place an absolute lower bound on the age of this quasar at t > 2 x 10^7 years.

astro-ph↗

arXiv2001

Large-scale Correlation of Mass and Galaxies with the Lyman-alpha Forest Transmitted Flux

We present predictions of the correlation between the Lyman-alpha forest absorption in quasar spectra and the mass within \sim 5 Mpc/h (comoving) of the line of sight, using fully hydrodynamic and hydro-PM numerical simulations of the cold dark matter model supported by present observations. The observed correlation based on galaxies and the Lya forest can be directly compared to our theoretical results, assuming that galaxies are linearly biased on large scales. Specifically, we predict the average value of the mass fluctuation, , conditioned to a fixed value of the Lya forest transmitted flux delta_F, after they have been smoothed over a 10 Mpc/h cube and line of sight interval, respectively. We find that /sigma_m as a function of delta_F/sigma_F has a slope of 0.6 at this smoothing scale, where sigma_m and sigma_F are the rms dispersions (this slope should decrease with the smoothing scale). We show that this value is largely insensitive to the cosmological model and other Lya forest parameters. Comparison of our predictions to observations should provide a fundamental test of our ideas on the nature of the Lya forest and the distribution of galaxies, and can yield a measurement of the bias factor of any type of galaxies that are observed in the vicinity of Lya forest lines of sight.

astro-ph↗

arXiv2001

A Fast, Accurate and Robust Algorithm For Transferring Radiation in Three-Dimensional Space

We have developed an algorithm for transferring radiation in three-dimensional space. The algorithm computes radiation source and sink terms using the Fast Fourier Transform (FFT) method, based on a formulation in which the integral of any quantity (such as emissivity or opacity) over any volume may be written in the classic convolution form. The algorithm is fast with the computational time scaling as N (log N)^2, where N is the number of grid points of a simulation box, independent of the number of radiation sources. Furthermore, in this formulation one can naturally account for both local radiation sources and diffuse background as well as any extra external sources, all in a self-consistent fashion. Finally, the algorithm is completely stable and robust. While the algorithm is generally applicable, we test it on a set of problems that encompass a wide range of situations in cosmological applications, demonstrating that the algorithm is accurate. These tests show that the algorithm produces results that are in excellent agreement with analytic expectations in all cases. In particular, radiation flux is guaranteed to propagate in the right direction, with the ionization fronts traveling at the correct speed with an error no larger than one cell for all the cases tested. The total number of photons is conserved in the worst case at 10% level and typically at 1-5% level over hundreds of time steps. As an added advantage, the accuracy of the results depends weakly on the size of the time step, with a typical cosmological hydrodynamic time step being sufficient.

astro-ph↗

arXiv2001

Evolution of the Ionizing Radiation Background and Star Formation in the Aftermath of Cosmological Reionization

The temporal evolution of the ionizing UV background radiation field at high redshift provides a probe of the evolution of the early star formation rate. By comparing the observed levels of absorption in the highest redshift quasar spectra to the predictions of a hydrodynamic simulation, we determine the evolution of the photoionization rate Gamma for neutral hydrogen in the intergalactic medium, over the redshift range 4.9 < z < 6.1. After accounting for sampling variance, we infer a sharp increase in Gamma from z~6.1 to z~5.8, probably implying reionization at this redshift. We find a decrease in Gamma from z~5.6 to 5.2, at 3sigma significance. This feature is a generic signature in the aftermath of reionization, entirely consistent with the cosmological reionization process being completed at z~6.1. This generic feature is a result of a significant change in the star formation rate subsequent to the cosmological reionization. There is an abrupt rise of the temperature of the intergalactic medium due to photo-heating, when it is reionized. This translates to a correspondingly sudden jump in the Jeans mass and a dramatic suppression of gas accretion onto the most abundant (sub-galactic) halos at the epochs of interest. The star formation rate suffers a temporary setback in the aftermath of reionization, resulting in a temporary decrease in the amplitude of the ionizing radiation field.

astro-ph↗

arXiv2001

Mass-Temperature Relation of Galaxy Clusters: A Theoretical Study

Combining conservation of energy throughout nearly-spherical collapse of galaxy clusters with the virial theorem, we derive the mass-temperature relation for X-ray clusters of galaxies $T=CM^{2/3}$. The normalization factor $C$ and the scatter of the relation are determined from first principles with the additional assumption of initial Gaussian random field. We are also able to reproduce the recently observed break in the M-T relation at $T \sim 3 \keV$, based on the scatter in the underlying density field for a low density $Λ$CDM cosmology. Finally, by combining observational data of high redshift clusters with our theoretical formalism, we find a semi-empirical temperature-mass relation which is expected to hold at redshifts up to unity with less than 20% error.

astro-ph↗

arXiv2001

Evolution of the Luminosity Function and Colors of Galaxies in a Lambda-CDM Universe

The luminosity function of galaxies is derived from a cosmological hydrodynamic simulation of a Lambda cold dark matter (CDM) universe with the aid of a stellar population synthesis model. At z=0, the resulting B band luminosity function has a flat faint end slope of α\approx -1.15 with the characteristic luminosity and the normalization in a fair agreement with observations, while the dark matter halo mass function is steep with a slope of α\approx -2. The colour distribution of galaxies also agrees well with local observations. We also discuss the evolution of the luminosity function, and the colour distribution of galaxies from z=0 to 5. A large evolution of the characteristic mass in the stellar mass function due to number evolution is compensated by luminosity evolution; the characteristic luminosity increases only by 0.8 mag from z=0 to 2, and then declines towards higher redshift, while the B band luminosity density continues to increase from z=0 to 5 (but only slowly at z>3).

astro-ph↗

arXiv2001

Revealing the Warm-Hot Intergalactic Medium with OVI Absorption

Hydrodynamic simulations of growth of cosmic structure suggest that 30-50% of the total baryons at z=0 may be in a warm-hot intergalactic medium (WHIM) with temperatures ~10^5-10^7K. The O VI λλ1032, 1038 absorption line doublet in the FUV portion of QSO spectra provides an important probe of this gas. Utilizing recent hydrodynamic simulations, it is found that there should be ~5 O VI absorption lines per unit redshift with equivalent widths >= 35 mA, decreasing rapidly to ~0.5 per unit redshift at >= 350 mA. About 10% of the total baryonic matter or 20-30% of the WHIM is expected to be in the O VI absorption line systems with equivalent width >= 20 mA; the remaining WHIM gas may be too hot or have too low metallicity to be detected in O VI. We find that the simulation results agree well with observations with regard to the line abundance and total mass contained in these systems. Some of the O VI systems are collisionally ionized and some are photoionized, but most of the mass is in the collisionally ionized systems. We show that the gas that produces the O VI absorption lines does not reside in virialized regions such as galaxies, groups, or clusters of galaxies, but rather has an overdensity of 10-40 times the average density. These regions form a somewhat connected network of filaments. The typical metallicity of these regions is 0.1-0.3Zsun.

astro-ph↗

arXiv2001

Synchronized Formation of Sub-Galactic Systems at Cosmological Reionization: Origin of Halo Globular Clusters

Gas rich sub-galactic halos with mass Mt <= 10^7.5 Msun, while incapable of forming stars due to lack of adequate coolants, contain a large fraction of baryonic mass at cosmological reionization. We show that the reionization of the universe at z=10-20 has an interesting physical effect on these halos. The external radiation field causes a synchronous inward propagation of an ionization front towards each halo, resulting in an inward, convergent shock. The resident gas of mass Mb~10^4-10^7 Msun in low spin (initial dimensionless spin parameter lambda <= 0.01) halos with a velocity dispersion sigmav <= 11km/s would be compressed by a factor of ~100 in radius and form self -gravitating baryonic systems. Under the assumption that such compressed gaseous systems fragment to form stars, the final stellar systems will have a size 2-40pc, velocity dispersion 1-10km/s and a total stellar mass of M* 10^3-10^6 Msun. The characteristics of these proposed systems seem to match the observed properties of halo globular clusters. The expected number density is consistent with the observed number density of halo globular clusters. The observed mass function of slope ~-2 at the high mass end is predicted by the model. Strong correlation between velocity dispersion and luminosity (or surface brightness) and lack of correlation between velocity dispersion and size, in agreement with observations, are expected. Metallicity is, on average, expected to be low and should not correlate with any other quantities of globular clusters, except that a larger dispersion of metallicity among globular clusters is expected for larger galaxies. The observed trend of specific frequency with galaxy type may be produced in the model. We suggest that these stellar systems are seen as halo globular clusters today.

astro-ph↗

arXiv2001

Star Formation History and Stellar Metallicity Distribution in a Cold Dark Matter Universe

We study star formation history and stellar metallicity distribution in galaxies in a Lambda cold dark matter universe using a hydrodynamic cosmological simulation. Our model predicts star formation rate declining in time exponentially from an early epoch to the present with the time-scale of 6 Gyr, which is consistent with the empirical Madau plot with modest dust obscuration. Star formation in L^* galaxies continues intermittently to the present also with an exponentially declining rate of a similar time-scale, whereas in small galaxies star formation ceases at an early epoch. The mean age of the extant stars decreases only slowly with increasing redshift, and exceeds 1 Gyr at z=3. Normal galaxies contain stars with a wide range of metallicity and age: stars formed at z<1 have metallicity of 0.1-1.0 Zsun, while old stars take a wide range of values from 10^{-6} Zsun to 3.0 Zsun. The mean metallicity of normal galaxies is in the range 0.1-1.0 Zsun. Dwarf galaxies that contain only old stars have a wide range of mean metallicity (10^{-4}-1.0 Zsun), but on average they are metal deficient compared with normal galaxies.

astro-ph↗

arXiv2001

Cosmic Mach Number as a Function of Overdensity and Galaxy Age

We carry out an extensive study of the cosmic Mach number (\mach) on scales of R=5, 10 and 20h^-1Mpc using an LCDM hydrodynamical simulation. We particularly put emphasis on the environmental dependence of \mach on overdensity, galaxy mass, and galaxy age. We start by discussing the difference in the resulting \mach according to different definitions of \mach and different methods of calculation. The simulated Mach numbers are slightly lower than the linear theory predictions even when a non-linear power spectrum was used in the calculation, reflecting the non-linear evolution in the simulation. We find that the observed \mach is higher than the simulated mean by more than 2-standard deviations, which suggests either that the Local Group is in a relatively low-density region or that the true value of Ω_m is ~ 0.2, significantly lower than the simulated value of 0.37. We show from our simulation that the Mach number is a weakly decreasing function of overdensity. We also investigate the correlations between galaxy age, overdensity and \mach for two different samples of galaxies --- DWARFs and GIANTs. Older systems cluster in higher density regions with lower \mach, while younger ones tend to reside in lower density regions with larger \mach, as expected from the hierarchical structure formation scenario. However, for DWARFs, the correlation is weakened by the fact that some of the oldest DWARFs are left over in low-density regions during the structure formation history. For giant systems, one expects blue-selected samples to have higher \mach than red-selected ones. We briefly comment on the effect of the warm dark matter on the expected Mach number.

astro-ph↗

arXiv2000

Why Are There Dwarf Spheroidal Galaxies?

There appears to exist a conflict between the standard structure formation theory and the existence of the faint dwarf spheroidal galaxies in the Local Group. Theoretical considerations suggest that a cold dark matter universe has been a very hostile place for the formation of small galaxies. In particular, gas would not have been able to cool and condense to form stars inside dark matter halos with a velocity dispersion <10km/s since before the cosmological reionization epoch. In other words, one should not expect to see any major star formation activities in dwarf galaxies with a velocity dispersion <10km/s during the past 12Gyr, according to the current structure formation theory. The real universe, on the other hand, shows that all dwarf spheroidal galaxies in the Local Group have velocity dispersions <10km/s and extended and recent star formation activities are quite common in them. This apparent conflict between cold dark matter models and local observations can be resolved, if one allows the cold dark matter particles to decay relatively recently. The resolution comes about in that the dwarf spheroidal galaxies with a velocity dispersion of \~10km/s seen today were born, in this picture, in a small fraction of halos with a velocity dispersion of ~20km/s, where gas can be retained and is able to cool and contract to form stars. The presently observed lower velocity dispersion ~10km/s of these dwarf spheroidal galaxies is a consequence of the decay of one half of the dark matter in and subsequent expansion of the halos since redshift z~2.

astro-ph↗

arXiv2000

Metal Enrichment and Temperature of the Intergalactic Medium

Hydrodynamic simulations of Lyman alpha clouds based on ab inito cosmological models have produced results that are in broad agreement with observations. However, further analyses have revealed that, with progressively higher numerical resolution, the median or cutoff line width of the simulated Lyman alpha clouds (i.e. the Doppler parameter) appears to converge to a value significantly below what is observed at z~3 (by about a factor of 1.5). These convergence test simulations do not include feedback from star formation. Given the observed metallicity in the Lyman alpha clouds we suggest that supernovae, which presumably polluted the IGM with metals, may have deposited a sufficient amount of energy in the IGM to reconcile the theory with observations. Simple arguments immediately narrow the redshift range of pollution down to 4 =0.01Zsun. Finally, assuming the specific supernova heating energy is proportional to the metallicity of a gas, we note that the picture proposed here would be consistent with supernovae being the apparently needed heating source for the intra-cluster gas, if the required heating of the intra-cluster gas is no greater than 1 keV per particle.

astro-ph↗