Search

We investigate the characteristic radiative efficiency [?], Eddington ratio l, and duty cycle P 0 of high-redshift active galactic nuclei (AGNs), drawing on measurements of the AGN luminosity function at z = 3-6 and, especially, on recent measurements of quasar clustering at z = 3-4.5 from the Sloan Digital Sky Survey. The free parameters of our models are [?], l, and the normalization, scatter, and redshift evolution of the relation between black hole (BH) mass M BH and halo virial velocity V vir. We compute the luminosity function from the implied growth of the BH mass function and the quasar correlation length from the bias of the host halos. We test our adopted formulae for the halo mass function and halo bias against measurements from the large N-body simulation developed by the MICE collaboration. The strong clustering of AGNs observed at z = 3 and, especially, at z = 4 implies that massive BHs reside in rare, massive dark matter halos. Reproducing the observed luminosity function then requires high efficiency [?] and/or low Eddington ratio l, with a lower limit (based on 2s agreement with the measured z = 4 correlation length) [?] [?] 0.7l/(1 + 0.7l), implying [?] [?] 0.17 for l>0.25. Successful models predict high duty cycles, P 0 [?] 0.2, 0.5, and 0.9 at z = 3.1, 4.5, and 6, respectively, and they require that the fraction of halo baryons locked in the central BH is much larger than the locally observed value. The rapid drop in the abundance of the massive and rare host halos at z > 7 implies a proportionally rapid decline in the number density of luminous quasars, much stronger than simple extrapolations of the z = 3-6 luminosity function would predict. For example, our most successful model predicts that the highest redshift quasar in the sky with true bolometric luminosity L > 1047.5 erg s-1 should be at z [?] 7.5, and that all quasars with higher apparent luminosities would have to be magnified by lensing.

We combine a cosmological reionization simulation with box size of 100 h -1 Mpc on a side and a Monte Carlo Lya radiative transfer code to model Lya Emitters (LAEs) at z [?] 5.7. The model introduces Lya radiative transfer as the single factor for transforming the intrinsic Lya emission properties into the observed ones. Spatial diffusion of Lya photons from radiative transfer results in extended Lya emission and only the central part with high surface brightness can be observed. Because of radiative transfer, the appearance of LAEs depends on density and velocity structures in circumgalactic and intergalactic media as well as the viewing angle, which leads to a broad distribution of apparent (observed) Lya luminosity for a given intrinsic Lya luminosity. Radiative transfer also causes frequency diffusion of Lya photons. The resultant Lya line is asymmetric with a red tail. The peak of the Lya line shifts toward longer wavelength and the shift is anti-correlated with the apparent-to-intrinsic Lya luminosity ratio. The simple radiative transfer model provides a new framework for studying LAEs. It is able to explain an array of observed properties of z [?] 5.7 LAEs in Ouchi et al., producing Lya spectra, morphology, and apparent Lya luminosity function (LF) similar to those seen in observation. The broad distribution of apparent Lya luminosity at fixed UV luminosity provides a natural explanation for the observed UV LF, especially the turnover toward the low luminosity end. The model also reproduces the observed distribution of Lya equivalent width (EW) and explains the deficit of UV bright, high EW sources. Because of the broad distribution of the apparent-to-intrinsic Lya luminosity ratio, the model predicts effective duty cycles and Lya escape fractions for LAEs.

We present predictions for the fluorescent Lya emission signature arising from photoionized, optically thick structures in smoothed particle hydrodynamic cosmological simulations of a LCDM universe using a Monte Carlo Lya radiative transfer code. We calculate the expected Lya image and two-dimensional spectra for gas exposed to a uniform ultraviolet ionizing background as well as gas exposed additionally to the photoionizing radiation from a local quasar, after correcting for the self-shielding of hydrogen. As a test of our numerical methods and for application to current observations, we examine simplified analytic structures that are uniformly or anisotropically illuminated. We compare these results with recent observations. We discuss future observing campaigns on large telescopes and realistic strategies for detecting fluorescence owing to the ambient metagalactic ionization and in regions close to bright quasars. While it will take hundreds of hours on the current generation of telescopes to detect fluorescence caused by the ultraviolet background alone, our calculations suggest that on the order of 10 sources of quasar-induced fluorescent Lya emission should be detectable after a 10 hr exposure in a 10 arcmin2 field around a bright quasar. These observations will help probe the physical conditions in the densest regions of the intergalactic medium as well as the temporal light curves and isotropy of quasar radiation.

We revisit the well known discrepancy between the observed number of Milky Way (MW) dwarf satellite companions and the predicted population of cold dark matter (CDM) subhalos, in light of the dozen new low-luminosity satellites found in imaging data from the Sloan Digital Sky Survey (SDSS) and our recent calibration of the SDSS satellite detection efficiency, which implies a total satellite population far larger than these dozen discoveries. We combine a detailed dynamical model for the CDM subhalo population with simple, physically motivated prescriptions for assigning a stellar content to each subhalo, then apply observational selection effects and compare to the current observational census. Reconciling the observed satellite population with CDM predictions still requires strong mass-dependent suppression of star formation in low-mass subhalos: models in which the stellar mass is a constant fraction F *(O b /O m ) of the subhalo mass M sat at the time it becomes a satellite fail for any choice of F *. However, previously advocated models that invoke suppression of gas accretion after reionization in halos with circular velocity V circ [?] V crit [?] 35 km s-1 can reproduce the observed satellite counts for -15 [?] MV [?] 0. Successful models require F * [?] 10-3 in halos with V circ>V crit and strong suppression of star formation before reionization in halos with V circ [?] 10 km s-1; models without pre-reionization suppression predict far too many satellites with -5 [?] MV [?] 0. In this successful model, the dominant fraction of stars formed after reionization at all luminosities. Models that match the satellite luminosity distribution also match the observed heliocentric radius distribution, and they reproduce the observed characteristic stellar velocity dispersion s* [?] 5-10 km s-1 of the SDSS dwarfs given the observed sizes ([?]50-200 pc) of their stellar distributions. The model satellites have M(<300 pc) [?] 107 M as observed even though their present-day total halo masses span more than two orders of magnitude; the constancy of central masses mainly reflects the profiles of CDM halos. Our modeling shows that natural physical mechanisms acting within the CDM framework can quantitatively explain the properties of the MW satellite population as it is presently known, thus providing a convincing solution to the "missing satellite" problem.

A new model for the formation of Jovian planets is proposed. We consider planets forming at large distances from a protostar ([?]100 AU) through direct fragmentation of a gas cloud, by the same formation mechanism as wide stellar and brown dwarf binaries. We model the gravitational evolution of a system of these distant planets and a second population formed in a disk closer to the star. We compute the typical closest approach of these planets to the star (i.e., smallest pericenter) over the course of their evolution. When the planets reach a pericenter within a gaseous disk surrounding the star, dynamical friction from this disk slows down the planet at each plunge, causing its orbit to be gradually circularized and made coplanar with the disk. After the disk dissipates, a large fraction of these planets may be left at orbits small enough to be detected in present radial velocity surveys. A brief analytic derivation of the rate of orbital energy dissipation during these disk crossings is presented. Observational tests of this model are discussed.

We construct evolutionary models of the populations of active galactic nuclei (AGNs) and supermassive black holes, in which the black hole mass function grows at the rate implied by the observed luminosity function, given assumptions about the radiative efficiency and the luminosity in Eddington units. We draw on a variety of recent X-ray and optical measurements to estimate the bolometric AGN luminosity function and compare to X-ray background data and the independent estimate of Hopkins et al. to assess remaining systematic uncertainties. The integrated AGN emissivity closely tracks the cosmic star-formation history, suggesting that star formation and black hole growth are closely linked at all redshifts. We discuss observational uncertainties in the local black hole mass function, which remain substantial, with estimates of the integrated black hole mass density r* spanning the range 3-5.5 x 105 M Mpc-3. We find good agreement with estimates of the local mass function for a reference model where all active black holes have a fixed efficiency [?] = 0.065 and L bol/L Edd [?] 0.4 (shifting to [?] = 0.09, L bol/L Edd [?] 0.9 for the Hopkins et al. luminosity function). In our reference model, the duty cycle of 109 M black holes declines from 0.07 at z = 3 to 0.004 at z = 1 and 10-4 at z = 0. The decline is shallower for less massive black holes, a signature of "downsizing" evolution in which more massive black holes build their mass earlier. The predicted duty cycles and AGN clustering bias in this model are in reasonable accord with observational estimates. If the typical Eddington ratio declines at z < 2, then the "downsizing" of black hole growth is less pronounced. Models with reduced Eddington ratios at low redshift or black hole mass predict fewer low-mass black holes (M * [?] 108 M ) in the local universe, while models with black hole mergers predict more black holes at M * > 109 M . Matching the integrated AGN emissivity to the local black hole mass density implies [?] = 0.075 x (r*/4.5 x 105 M Mpc-3)-1 for our standard luminosity function estimate, or 25% higher for Hopkins et al.'s estimate. It is difficult to reconcile current observations with a model in which most black holes have the high efficiencies [?] [?] 0.16-0.20 predicted by MHD simulations of disk accretion. We provide electronic tabulations of our bolometric luminosity function and our reference model predictions for black hole mass functions and duty cycles as a function of redshift.

We predict the 21 cm signature of the first metal-free stars. The soft X-rays emitted by these stars penetrate the atomic medium around their host halos, generating Lya photons that couple the spin and kinetic temperatures. These create a region we call the "Lya sphere," visible in 21 cm against the CMB, which is much larger than the H II region produced by the same star. The spin and kinetic temperatures are strongly coupled before the X-rays can substantially heat the medium, implying that a 21 cm absorption signal from the adiabatically cooled gas in Hubble expansion around the star is expected when the medium has not been heated previously. A central region of emission from the gas heated by the soft X-rays is also present, although with a weaker signal than the absorption. The Lya sphere is a universal signature that should be observed around any first star illuminating its vicinity for the first time. The 21 cm radial profile of the Lya sphere can be calculated as a function of the luminosity, spectrum, and age of the star. For a star of a few hundred M and zero metallicity (as expected for the first stars), the physical radius of the Lya sphere can reach tens of kiloparsecs. The first metal-free stars should be strongly clustered because of high cosmic biasing; this implies that the regions producing a 21 cm absorption signal may contain more than one star and will generally be irregular and not spherical, because of the complex distribution of the gas. We discuss the feasibility of detecting these Lya spheres, which would be present at redshifts z [?] 30 in the cold dark matter model. Their observation would represent a direct proof of the detection of a first star.

Recent observations support the idea that nuclear black holes grew by gas accretion while shining as luminous quasars at high redshift, and they establish a relation of the black hole mass with the host galaxy's spheroidal stellar system. We develop an analytic model to calculate the expected impact of mergers on the masses of black holes in massive clusters of galaxies. We use the extended Press-Schechter formalism to generate Monte Carlo merger histories of halos with a mass 1015 h-1 M. We assume that the black hole mass function at z = 2 is similar to that inferred from observations at z = 0 (since quasar activity declines markedly at z < 2), and we assign black holes to the progenitor halos assuming a monotonic relation between halo mass and black hole mass. We follow the dynamical evolution of subhalos within larger halos, allowing for tidal stripping, the loss of orbital energy by dynamical friction, and random orbital perturbations in gravitational encounters with subhalos, and we assume that most black holes will efficiently merge after their host galaxies (represented as the nuclei of subhalos in our model) merge. Our analytic model reproduces numerical estimates of the subhalo mass function. We find that mergers can increase the mass of the most massive black holes in massive clusters typically by a factor ~2, after gas accretion has stopped. In our 10 realizations of 1015 h-1 M clusters, the highest initial (z = 2) black hole masses are 5-7 x 109 M, but four of the clusters contain black holes in the range 1-1.5 x 1010 M at z = 0. Satellite galaxies may host black holes whose mass is comparable to, or even greater than, that of the central galaxy. Thus, black hole mergers can significantly extend the very high end of the black hole mass function.

We present predictions of Lya forest-galaxy correlations at z = 3 from Eulerian simulations that include the effects of galactic winds, driven primarily by supernova explosions. Galactic winds produce expanding bubbles of shock-heated gas within ~0.5 comoving h-1 Mpc of luminous galaxies in the simulation, which have space density similar to that of observed Lyman break galaxies (LBGs). However, most of the low-density intergalactic gas that determines the observed properties of the Lya forest is unaffected by winds. The impact of winds on the Lya optical depth near galaxies is less dramatic than their impact on gas temperature, because winds heat only a small fraction of the gas present in the turnaround regions surrounding galaxies. Hence, Lya absorption from gas outside the wind bubbles is spread out over the same velocity range occupied by the wind-heated gas. In general, Lya absorption is expected to be stronger than average near galaxies because of the high gas density. Winds result in a modest reduction of this expected increase of Lya absorption. Our predictions can be compared to future observations to detect the wind effects and infer their strength, although with the caveat that the results are still dependent on the correspondence of simulated galaxies and observed LBGs. We find that wind effects in our simulations are not strong enough to reproduce the high Lya transmission within 0.5 h-1 Mpc comoving of galaxies that has been suggested by recent observations; powerful galactic explosions or ejecta with hyper-escape velocities would be required, but these are unlikely to be produced by ordinary star formation and supernovae alone.

Precise proper-motion measurements (sm ~ 10 mas yr-1) of the recently discovered hypervelocity star (HVS) SDSS J090745.0+024507 would yield significant constraints on the axis ratios and orientation of a triaxial model for the Galactic halo. Triaxiality of dark matter halos is predicted by cold dark matter models of galaxy formation and may be used to probe the nature of dark matter. However, unless the distance to this star is determined to better than 10%, these constraints suffer from one-dimensional degeneracies, which we quantify. We show how proper-motion measurements of several HVSs could simultaneously resolve the distance degeneracies of all such stars and produce a detailed picture of the triaxial halo. Additional HVSs may be found from radial velocity surveys or from parallax/proper-motion data derived from Gaia. High-precision proper-motion measurements of these stars using the Space Interferometry Mission (SIM PlanetQuest) would substantially tighten the constraints they yield on the Galactic potential.

We use cosmological simulations of reionization to predict the effect of large H II regions around individual high-redshift quasars on the possible signal from the redshifted 21 cm line of neutral hydrogen in the pre-reionization era. We show that these H II regions appear as "spectral dips" in frequency space with equivalent widths in excess of 3 mK MHz or depths in excess of about 1.5 mK, and that they are by far the most prominent cosmological signals in the redshifted H I distribution. These spectral dips are expected to be present in almost every line of sight. If the spectral dips of a large enough sample of H II regions are well resolved in frequency space, then the distribution of line depth and equivalent width in frequency with a known observing beam size can be used to infer the H II region size distribution and the mean difference in neutral hydrogen density between the H II regions (which may contain self-shielded neutral gas clumps) and the surrounding medium, providing a powerful test for models of reionization.

A new theory of quasars is presented in which the matter of thin accretion disks around black holes is supplied by stars that plunge through the disk. Stars in the central part of the host galaxy are randomly perturbed to highly radial orbits, and as they repeatedly cross the disk they lose orbital energy by drag, eventually merging into the disk. Requiring the rate of stellar mass capture to equal the mass accretion rate into the black hole, a relation between the black hole mass and the stellar velocity dispersion is predicted of the form MBH [?] simg1.gif. The normalization depends on various uncertain parameters, such as the disk viscosity, but is consistent with observation for reasonable assumptions. We show that a seed central black hole in a newly formed stellar system can grow at the Eddington rate up to this predicted mass via stellar captures by the accretion disk. Once this mass is reached, star captures are insufficient to maintain an Eddington accretion rate, and the quasar may naturally turn off as the accretion switches to a low-efficiency advection mode. The model provides a mechanism to deliver mass to the accretion disk at small radius, probably solving the problem of gravitational instability to star formation in the disk at large radius. We note that the matter from stars that is incorporated into the disk has an average specific angular momentum that is very small or opposite to that of the disk, and we discuss how a rotating disk may be maintained as it captures this matter if a small fraction of the accreted mass comes from stellar winds that form a disk extending to larger radius. We propose several observational tests and consequences of this theory.

We calculate the reionization history in CDM (cold dark matter) models. The epoch of the end of reionization and the Thomson scattering optical depth to the cosmic microwave background depend on the power spectrum amplitude on small scales and on the ionizing photon emissivity per unit mass in collapsed halos. We calibrate the emissivity to reproduce the measured ionizing background intensity at z = 4. Models in which all CDM halos have either a constant emissivity or a constant energy emitted per Hubble time per unit mass predict that reionization ends near z ~ 6 and that the optical depth is in the range 0.05 < te < 0.09, consistent with Wilkinson Microwave Anisotropy Probe (WMAP) results at the 1-2 s level. If the optical depth is as high as 0.17 (as suggested by WMAP), halos of velocity dispersion ~3-35 km s-1 at z > 15 must have ionizing emissivities per unit mass larger by a factor [?]50 compared with the more massive halos that produce the ionizing emissivity at z = 4. This factor increases to 100 if the CDM power spectrum amplitude is required to agree with the recent Croft et al. measurement from the Lya forest. If te [?] 0.17 is confirmed, a higher ionizing emissivity at z > 15 compared with z = 4 might arise from an enhanced star formation rate or quasar abundance per unit mass and an increased escape fraction for ionizing photons; the end of reionization could have been delayed to z ~ 6 because of the suppression of gas accretion and star formation in low-mass halos as the medium was reionized.

The transverse proximity effect is the expected decrease in the strength of the Lya forest absorption in a QSO spectrum when another QSO lying close to the line of sight enhances the photoionization rate above that due to the average cosmic ionizing background. We select three QSOs from the Early Data Release of the Sloan Digital Sky Survey that have nearby foreground QSOs, with proper line-of-sight tangential separations of 0.50, 0.82, and 1.10 h-1 Mpc. We estimate that the ionizing flux from the foreground QSO should increase the photoionization rate by a factor (94, 13, 13) in these three cases, which would be clearly detectable in the first QSO and marginally so in the other two. We do not detect the transverse proximity effect. Three possible explanations are provided: an increase of the gas density in the vicinity of QSOs, time variability, and anisotropy of the QSO emission. We find that the increase of gas density near QSOs can be important if they are located in the most massive halos present at high redshift, but it is not enough to fully explain the absence of the transverse proximity effect. Anisotropy requires an unrealistically small opening angle of the QSO emission. Variability demands that the luminosity of the QSO with the largest predicted effect was much lower 106 yr ago, whereas the transverse proximity effect observed in the He II Lya absorption in QSO 0302-003 by Jakobsen et al. implies a lifetime longer than 107 yr. A combination of all three effects may better explain the lack of Lya absorption reduction. A larger sample of QSO pairs may be used to diagnose the environment, anisotropy, and lifetime distribution of QSOs.

We investigate the interaction of Lya photons produced by the first stars in the universe with intergalactic hydrogen prior to reionization. The background Lya spectral profile is obtained by solving a Fokker-Planck equation. We present accurate values of the heating and scattering rates and the spin-kinetic temperature coupling coefficient. We show that the heating rate induced by the Lya scattering is much lower than found previously and is basically negligible. The dominant heating source is most likely X-rays from the first ionizing sources, which are able to penetrate into the atomic medium. The scattering of Lya photons couples the hydrogen spin temperature to the kinetic temperature. If the first ionizing sources in the universe did not emit significant X-rays, the spin temperature would be rapidly brought down to the very low gas kinetic temperature, and a 21 cm absorption signal against the cosmic microwave background (CMB) larger than 100 mK would be predicted. However, we argue that sufficient X-rays are likely to have been emitted by the first stellar population, implying that the gas kinetic temperature should rapidly increase, turning a reduced and brief absorption signal into emission, with a smaller amplitude of ~20 mK. The detection of the 21 cm absorption and emission features would be a hallmark in unravelling the history of the "dark age" before reionization.

The average rate of emission of ionizing radiation per unit volume (or emissivity) in the universe can be calculated as the ratio of the intensity of the ionizing background to the mean free path of ionizing photons. The intensity of the background is measured from the mean transmitted flux of the Lya forest, and the mean free path is measured from the abundance of Lyman limit systems, which has been observed so far up to z = 4. This yields an emissivity that is not larger than seven ionizing photons per Hubble time for each atom in the universe at z = 4, which may reasonably arise from QSOs and star-forming galaxies. In order for the reionization to end by z = 6, and assuming that the clumping factor of ionized gas during the reionization epoch is close to unity, this ionizing comoving emissivity cannot decline from z = 4 up to z ~ 9 by more than a factor 1.5. If the clumping factor were much larger than unity, then the emissivity would need to rapidly increase with redshift. Unless the ionizing emissivity increases substantially from z = 4 to z ~ 10-20, the Thomson optical depth to the cosmic microwave background must be in the range 0.045 < t < 0.09.

We present predictions of the correlation between the Lya forest absorption in quasar spectra and the mass within ~5 h-1 Mpc (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, [?]dm[?], conditioned to a fixed value of the Lya forest transmitted flux, dF, after they have been smoothed over a 10 h-1 Mpc cube and line-of-sight interval, respectively. We find that the relation between dF/sF and [?]dm[?]/sm (where sm and sF are the rms dispersions) has a predicted slope of ~0.6 at dF = 0 that 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 galaxy that is observed in the vicinity of Lya forest lines of sight.

A Monte Carlo code to solve the transfer of Lya photons is developed that can predict the Lya image and two-dimensional Lya spectra of a hydrogen cloud with any given geometry, Lya emissivity, neutral hydrogen density distribution, and bulk velocity field. We apply the code to several simple cases of a uniform cloud to show how the Lya image and emitted line spectrum are affected by the column density, internal velocity gradients, and emissivity distribution. We then apply the code to two models for damped Lya absorption systems: a spherical, static, isothermal cloud and a flattened, axially symmetric, rotating cloud. If the emission is due to fluorescence of the external background radiation, the Lya image should have a core corresponding to the region where hydrogen is self-shielded. The emission-line profile has the characteristic double peak with a deep central trough. We show how rotation of the cloud causes the two peaks to shift in wavelength as the slit is perpendicular to the rotation axis and how the relative amplitude of the two peaks is changed. In reality, damped Lya systems are likely to have a clumpy gas distribution with turbulent velocity fields, which should smooth the line emission profile but should still leave the rotation signature of the wavelength shift across the system.

We calculate the H I column density of self-absorption in quasi-stellar objects (QSOs) predicted in a model where the QSOs are located in the same halos that contain the gas in damped Lya absorption systems. The model is parameterized by the probability P0 that any halo has an active QSO. We assume that the QSOs ionize the gas but do not expel or heat it beyond the photoionization effect. When clumping of the gas is ignored, the derived H I column densities produce negligible Lyman limit absorption even in the lowest luminosity QSOs with an optical depth of ~10% for luminosity L = 0.01LB* when P0 = 10-2. We also compute the He II Lyman limit self-absorption, which is slightly higher but still small. The self-absorption is increased if the gas is highly clumped; for P0 = 10-2, a clumping factor of ~10 results in a significantly reduced overall emissivity of H I Lyman limit photons from QSOs and affects the predicted intensity of the ionizing background. A clumping factor of ~5 is sufficient to reduce the He II Lyman limit emissivity and to delay the predicted epoch of He II reionization. The presence of the gas associated with damped absorption systems around QSOs can also be detected from the narrow Lya emission line, which should have an angular extent of 0.''1-1'' in typical high-redshift QSOs.

The recent discovery of a planetary transit in the star HD 209458, and the subsequent highly precise observation of the transit light curve with Hubble Space Telescope, is encouraging to search for any phenomena that might induce small changes in the light curve. Here we consider the effect of the quadrupole moment of the parent star and of a possible second planet perturbing the orbit of the transiting planet. Both of these cause a precession of the orbital plane and of the periastron of the planet, which result in a long-term variation of the duration and the period of the transits. For a transiting planet at 0.05 AU, either a quadrupole moment similar to that of the Sun or the gravitational tug from an Earth-like planet on an orbit of semimajor axis ~0.2 AU and a relative inclination near the optimal 45deg would cause a transit duration time derivative of ~1 s yr-1.

Spergel & Steinhardt proposed the possibility that dark matter particles are self-interacting as a solution to two discrepancies between the predictions of cold dark matter models and observations: First, the observed dark matter distribution in some dwarf galaxies has large, constant-density cores, as opposed to the predicted central cusps; and second, small satellites of normal galaxies are much less abundant than predicted. The dark matter self-interaction would produce isothermal cores in halos and expel the dark matter particles from dwarfs orbiting in large halos. Another consequence of the model is that halos should become spherical once most particles have interacted. Several observations show that the mass distribution in relaxed clusters of galaxies is elliptical. Here, I discuss in particular gravitational lensing in the cluster MS 2137-23, where the ellipticity of the dark matter distribution can be measured to a small radius, r ~ 70 kpc, suggesting that most dark matter particles in clusters outside this radius do not collide during the characteristic age of clusters. If true, this implies that any dark matter self-interaction with a cross section independent of velocity is too weak to have affected the observed density profiles in the dark-matter dominated dwarf galaxies, or to have facilitated the destruction of dwarf satellites in galactic halos. If sx is the cross section and mx the mass of the dark matter particle, then sx/mx < 10-25.5 cm2 GeV-1.

We show that at any given time, the Galactocentric black hole Sgr A* is expected to be microlensing Nlens ~ 1.7 bulge stars if the threshold of detectability of the fainter image is Kthr = 21, and Nlens ~ 8 sources if Kthr = 23. The lensed images then provide a unique way to detect stellar-mass black holes predicted to cluster around Sgr A*. If a black hole passes close to a microlensed image, it will give rise to a short (weeks long) microlensing event. We show that the mass and projected velocity of the stellar-mass black hole can both be measured unambiguously from such an event, provided that either a caustic crossing is observed or the astrometric displacement is measured. For Kthr = 23 and moderate magnifications by Sgr A*, the microlensing event rate from a cluster of 20000 black holes within a radius of 0.7 pc is only 0.06 yr- 1; however, if highly magnified images of a star were found, the rate of events by the stellar black holes would be much higher. In addition, the Nlens sources lensed by Sgr A* provide a unique probe of extinction behind the Galactic center along 2Nlens lines of sight.

The evolution of the temperature in the intergalactic medium is related to the reionization of hydrogen and helium and has important consequences for our understanding of the Lya forest and of galaxy formation. We measure the temperature-density relation of intergalactic gas from Lya forest observations of eight quasar spectra, using a new line fitting technique to obtain a lower cutoff on the distribution of line widths from which the temperature is derived. Using a numerical simulation, we examine the details of this kind of measurement at different densities, finding that the temperature may be difficult to measure for gas with Dg [?] 1 (Dg is the density of the gas in units of the mean density) because the velocities due to expansion always dominate the widths of the corresponding weak lines, and that the temperature measurement is increasingly ambiguous for gas with Dg [?] 5 because the dispersion in temperature at fixed density is high. From our observed spectra, the temperature is most precisely determined at densities slightly above the mean: T* = (20,200 +- 2700, 20,200 +- 1300, 22,600 +- 1900) K (statistical error bars) for gas densities D* = (1.42 +- 0.08, 1.37 +- 0.11, 1.66 +- 0.11) at redshift z = (3.9, 3.0, 2.4). Systematic errors in T* should be less than 2000 K. The power-law index of the temperature-density relation, defined by T = T*(Dg/D*)g-1, is g - 1 = (0.43 +- 0.45, 0.29 +- 0.30, 0.52 +- 0.14) for the same three redshifts. The temperature at fixed overdensity D = 1.4 is T1.4 = (20,100 +- 2800, 20,300 +- 1400, 20,700 +- 1900) K. This unchanging temperature is higher than expected for photoionized gas in ionization equilibrium with a cosmic background. If the heat from the He II reionization is responsible for the high measured temperature, then the temperature should not be constant but should have a maximum at the end of the reionization epoch. We update the lower limit to the baryon density implied by the observed mean flux decrement.

If the stellar population of the bulge contains black holes formed in the final core collapse of ordinary stars with M [?] 30 M, then about 25,000 stellar mass black holes should have migrated by dynamical friction into the central parsec of the Milky Way, forming a black hole cluster around the central supermassive black hole. These black holes can be captured by the central black hole when they randomly reach a highly eccentric orbit due to relaxation, either by direct capture (when their Newtonian peribothron is less than 4 Schwarzschild radii) or after losing orbital energy through gravitational waves. The overall depletion timescale is ~30 Gyr, so most of the 25,000 black holes remain in the central cluster today. The presence of this black hole cluster would have several observable consequences. First, the low-mass, old stellar population should have been expelled from the region occupied by the black hole cluster as a result of relaxation, implying a core in the profile of solar-mass red giants with a radius of ~2 pc (i.e., 1'). The observed central density cusp (which has a core radius of only a few arcseconds) should be composed primarily of young ([?]1 Gyr) stars. Second, flares from stars being captured by supermassive black holes in other galaxies should be rarer than usually expected because the older stars will have been expelled from the central regions by the black hole clusters of those galaxies. Third, the young ([?]2 Gyr) stars found at distances ~3-10 pc from the Galactic center should be preferentially on highly eccentric orbits. Fourth, if future high-resolution K-band images reveal sources microlensed by the Milky Way's central black hole, then the cluster black holes could give rise to secondary ("planet-like") perturbations on the main event.

A sample of eight quasars observed at high resolution and signal-to-noise ratio is used to determine the transmitted flux probability distribution function (TFPDF), and the power spectrum and correlation function of the transmitted flux in the Lya forest, in three redshift bins centered at z = 2.41, 3.00, and 3.89. All the results are presented in tabular form, with full error covariance matrices, to allow for comparisons with any numerical simulations and with other data sets. The observations are compared with a numerical simulation of the Lya forest of a LCDM model with O = 0.4, known to agree with other large-scale structure observational constraints. There is excellent agreement for the TFPDF if the mean transmitted flux is adjusted to match the observations. A small difference between the observed and predicted TFPDF is found at high fluxes and low redshift, which may be due to the uncertain effects of fitting the spectral continuum. Using the numerical simulation, we show how the flux power spectrum can be used to recover the initial power spectrum of density fluctuations. From our sample of eight quasars, we measure the amplitude of the mass power spectrum to correspond to a linear variance per unit ln k of Dimg1.gif(k) = 0.72 +- 0.09 at k = 0.04(km s-1)-1 and z = 3, and the slope of the power spectrum near the same k to be np = -2.55 +- 0.10 (statistical error bars). The results are statistically consistent with those of Croft et al., although our value for the rms fluctuation is lower by a factor of 0.75. For the LCDM model we use, the implied primordial slope is n = 0.93 +- 0.10, and the normalization is s8 = 0.68 + 1.16(0.95 - n) +- 0.04.

A model of the density distribution in the intergalactic medium (IGM), motivated by that found in numerical simulations, is used to demonstrate the effect of a clumpy IGM and discrete sources on the reionization of the universe. In an inhomogeneous universe reionization occurs outside-in, starting in voids and gradually penetrating into overdense regions. Reionization should not be sudden but gradual, with a continuous rise of the photon mean free path over a fair fraction of the Hubble time as the emissivity increases. We show that a hydrogen Gunn-Peterson trough should be present at z [?] 6 unless the emissivity increases with redshift at z > 4. However, the epoch of overlap of cosmological H II regions could have occurred at a higher redshift if sources of low luminosity reionized the IGM; the Gunn-Peterson trough at z ~ 6 would then appear because even the most underdense voids have a large enough neutral fraction in ionization equilibrium to be optically thick to Lya photons. Cosmological H II regions near the epoch of overlap can produce gaps of transmitted flux only if luminous quasars contributed to the reionization, producing large H II regions. Despite the clumpiness of the matter distribution, recombinations do not increase the required emissivity of ionizing photons by a large factor during the reionization of hydrogen because the high-density gas is not ionized until a late time. We show that the He II reionization was most likely delayed relative to the hydrogen reionization but was probably complete by z ~ 3 (the redshift where observations are available). The reported large optical depth fluctuations of He II are not necessarily due to an incomplete He II reionization but can arise from a combination of IGM density fluctuations and variations in the intensity of the He II ionizing background due to luminous QSOs.

A large number of microlensing events have been observed in the direction of the Galactic bulge, with a measured optical depth in the range 2-3x10[?]6. It has been shown that most of these events are due to bulge stars being lensed by other bulge stars or by foreground disk stars. Among the stars observed in the bulge fields there should also be disk stars located behind the bulge; here we consider their effect on the microlensing rates. The optical depth of background disk stars is much higher than that of typical bulge stars, reaching 10-5 at 6 kpc behind the bulge. Thus, although background disk stars are a very small fraction of the stars in Baade's window, we find that ~5%-10% of the optical depth should be due to disk stars more than 3 kpc behind the bulge. This fraction is sensitive to the luminosity function of disk stars at large scale height, to the magnitude cutoff of the survey, and to the amplification bias effect causing large numbers of "blended" events. We consider also the effect of a warp and flare in the disk at large distances behind the bulge; this could increase the optical depth from the background disk to ~20% of the total. Events on background disk stars should on average be longer than other events and could be distinguished as well by measuring the proper motion or distance of the stars that have been microlensed. The number of these events could be an interesting probe to the structure and stellar population of the far side of the Galactic disk.

Gamma-ray bursts are now known to be a cosmological population of objects, which are often accompanied by X-ray and optical afterglows. The total energy emitted in the afterglow can be similar to the energy radiated in the gamma-ray burst itself. If a galaxy containing a large column density of dust is near the line of sight to a gamma-ray burst, small-angle scattering of the X-rays due to diffraction by the dust grains will give rise to an X-ray echo of the afterglow. A measurement of the angular size of the echo at a certain time after the afterglow is observed yields a combination of the angular diameter distances to the scattering galaxy and the gamma-ray burst that can be used to constrain cosmological models in the same way as a time delay in a gravitational lens. The scattering galaxy will generally cause gravitational lensing as well, and this should modify the shape of the X-ray echo from a circular ring. The main difficulty in detecting this phenomenon is the very low flux expected for the echo. The flux can be increased when the gamma-ray burst is highly magnified by gravitational lensing, or when the deflecting galaxy is at low redshift. X-ray echoes of continuous (but variable) sources, such as quasars, may also be detectable with high-resolution instruments and would allow similar measurements.

We use numerical simulations of structure formation in a cold dark matter model to predict the absorption lines in the soft X-rays produced by heavy elements in the shock-heated intergalactic medium at low redshift. The simulation incorporates a model for heavy-element production in galaxies and the subsequent dispersion of the metals to the intergalactic medium. We analyze in particular absorption lines produced by oxygen and calculate the ionization stage, taking into account the observed X-ray background at the present time. We find that oxygen is fully ionized by the X-ray background in low-density voids and is mostly in the form of O VII and O VIII in the sheets and filamentary regions. Strong absorption lines of O VII and O VIII with equivalent widths W ~ 100 km s-1 are produced in filamentary regions of overdensities of ~100 and temperatures of ~106 K, located in the outskirts of groups and clusters of galaxies. The O VII line at E = 574 eV is generally the strongest one in these systems. Our model predicts that any X-ray source (such as a quasar) should typically show about one O VII absorption line with W > 100 km s-1 in the interval from z = 0 to z = 0.3. These lines could be detected with the upcoming generation of X-ray telescopes, and their origin in intervening systems could be confirmed by the association with groups of galaxies and X-ray-emitting halos near the line of sight at the same redshift. The hot intergalactic medium may be one of the main reservoirs of baryons in the present universe, and the heavy-element X-ray absorption lines offer a promising possibility of detecting this new component in the near future.

A model of damped Lya systems is presented based on randomly moving clouds in spherical halos. We use the Press-Schechter model for the abundance of halos and assume that each halo has a similar population of clouds, with total mass and spatial distribution constrained to fit observations of the column density distribution. We show that the kinematics of the multiple absorbing components revealed in absorption profiles of the low-ionization lines, presented by Prochaska & Wolfe, are consistent with our spherical halo model. The presence of multiple absorbing components with a large covering factor, combined with the small impact parameters of the systems predicted in our analytical model and in numerical simulations, implies a high rate of energy dissipation in cloud collisions. We calculate the rate of energy dissipation in our model and show that it is far greater than the rate at which energy can be supplied by gravitational mergers of halos. This poses a possible problem for the model of merging protogalactic clumps of Haehnelt et al., based on numerical simulations. We also present new constraints on the amplitude of the power spectrum in hierarchical theories required to account for the observed velocity dispersion in the absorbers. We find that the linearly extrapolated rms fluctuation at redshift z = 4 on spheres of radius R = 100 km s-1 H-1(z) [where H(z) is the Hubble constant at redshift z] must be greater than 0.75. Although this limit is obtained only for our specific model of the absorbing components, it should not be highly model-dependent because the velocity dispersion of the absorbers is essentially determined by the velocity dispersion of the halos where the gas is moving.

The possibility of measuring the cosmological geometry using the redshift space correlation function of the Lya forest in multiple lines of sight as a function of angular and velocity separation is discussed. The geometric parameter to be measured is f(z)[?]c-1H(z)DA(z), where H(z) is the Hubble constant and DA(z) the angular diameter distance at redshift z. The correlation function is computed in linear theory, assuming that the Lya forest is a result of gravitational instability in a photoionized intergalactic medium. We describe a method to measure the correlation from observations with the Gaussianization procedure of Croft et al. to map the observed Lya forest transmitted flux to an approximation of the linear density field. The effect of peculiar velocities on the shape of the recovered power spectrum is pointed out. We estimate the error in recovering the f(z) factor from observations due to the variance in the Lya absorbers. We show that at least ~25 pairs of quasars (separations <3') are needed to distinguish a flat O0=1 universe from a universe with O0=0.2, OL=0.8. A second parameter that is obtained from the correlation function of the Lya forest is b[?]O(z)0.6/b (affecting the magnitude of the peculiar velocities), where b is a linear theory bias of the Lya forest. In the theory of the Lya forest assumed here, the parameter b can be predicted from numerical simulations; once b is known, the number of quasar pairs needed to constrain f is reduced to about six. On small scales, where the correlation function is higher, f(z) should be measurable with fewer quasars, but nonlinear effects must then be taken into account. The anisotropy of the nonlinear redshift space correlation function as a function of scale should also provide a precise quantitative test of the gravitational instability theory of the Lya forest.

Observations of high-redshift quasars show that the intergalactic medium (IGM) must have been reionized at some redshift z > 5. If a source of radiation could be observed at the rest-frame Lya wavelength, at a sufficiently high redshift where some of the IGM in the line of sight was not yet reionized, the Gunn-Peterson trough should be present. Longward of the Lya wavelength, a damping wing should be observed, caused by the neutral IGM whose absorption profile can be predicted. Measuring the shape of this damping wing would provide irrefutable evidence of the observation of the IGM before reionization and a determination of the density of the neutral IGM. This measurement might be hindered by the possible presence of a dense absorption system associated with the source. Shortward of the Lya wavelength, absorption should be seen from the patchy structure of the IGM in the process of reionization, intersected in the line of sight. We show that a complete Gunn-Peterson trough is most likely to continue to be observed through the epoch where the IGM is partially ionized. The damping wings of the neutral patches around an ionized region should overlap in the spectrum if the proper path length through the ionized region is less than 1 h-1 Mpc; even in larger ionized regions, the characteristic background intensity should be low enough to yield a very high optical depth due to the residual neutral fraction, although occasionally some flux may be transmitted through large, underdense voids within an ionized region. In the case of the He II reionization, the ionization fronts are much thicker than in the case of hydrogen, and the profile of this front determines the shape of the absorption at the edge of a He III region. Analogous to the case of hydrogen, windows of transmitted flux are not likely to be observed until after the low-density IGM has been completely reionized. Therefore, the observation of these transmission windows by Reimers et al. at z [?] 2.85 suggests that the helium reionization was complete by this redshift. The recently discovered afterglows of gamma-ray bursts might soon be observed at the very high redshifts required for these observations. Their featureless continuum spectrum and high luminosities make them ideal sources for studying absorption by the IGM.

We have measured the distribution function of the flux decrement D = 1 - e-t caused by Lya forest absorption from intervening gas in the lines of sight to high-redshift QSOs from a sample of seven high-resolution QSO spectra obtained with the Keck telescope. The observed flux decrement distribution function (FDDF) is compared with the FDDF from two simulations of the Lya forest: a LCDM model (with O = 0.4, L = 0.6), computed with the Eulerian code of Cen & Ostriker, and a standard cold dark matter (SCDM) model (with O = 1), computed with the smoothed particle hydrodynamics code of Hernquist et al. Good agreement is obtained between the shapes of the simulated and observed FDDFs for both simulations after fitting only one free parameter, which controls the mean flux decrement. The difference between the predicted FDDFs from the two simulations is small, and we show that it arises mostly from a different temperature in the low-density gas (caused by different assumptions that were made about the reionization history in the two simulations), rather than differences between the two cosmological models or numerical effects in the two codes, which use very different computational methods. A measurement of the parameter m[?]O 2b h 3/G (where G is the H I ionization rate due to the ionizing background) is obtained by requiring the mean flux decrement in the simulations to agree with the observed one. Estimating the lower limit G > 7 x 10-13 s-1 from the abundance of known QSOs, we derive a lower limit on the baryonic matter density, Obh2 > 0.021 (0.017) for the LCDM (SCDM) model. The difference between the lower limits inferred from the two models is again due to different temperatures in the low-density gas. We give general analytical arguments for why this lower limit is unlikely to be reduced for any other models of structure formation by gravitational collapse that can explain the observed Lya forest. When combined with constraints from big bang nucleosynthesis, the large Ob we infer is inconsistent with some recent D/H determinations (Rugers & Hogan), favoring a low deuterium abundance as reported by Tytler, Fan & Burles. Adopting a fixed Ob, the measurement of m(z) allows a determination of the evolution of the ionizing radiation field with redshift. Our models predict an intensity that is approximately constant with redshift, which is in agreement with the assumption that the ionizing background is produced by known quasars for z < 3, but requires additional sources of ionizing photons at higher redshift given the observed rapid decline of the quasar abundance.

If the universe was reionized by O and B stars in an early population of galaxies, the associated supernovae should have enriched the universe to a mean metallicity Z ${&ucpmathaccent;{Z}{"7016}}$ =10 [?]5(1+nrec $_{{&rm; rec}}$ ), where nrec is the mean number of times that each baryon recombined during the reionization era. This is consistent with recent observations of the metallicity in the Lya forest at z [?] 3. The mean surface brightness observable at present from the galaxies that produced these heavy elements, in the rest-frame wavelengths 1216 A < l [?] 2500 A, should be ~10 [?]6(Z ${&ucpmathaccent;{Z}{"7016}}$ /10 [?]4)(O b h 2/0.02) photons cm-2 s-1 arcsec-2. Most of this radiation should be emitted at z > 5, before reionization was complete. These high-redshift galaxies may be detectable in near-infrared photometric surveys, identifying them via the Gunn-Peterson trough (analogous to the use of the Lyman limit cutoff to search for galaxies at z ~ 3, where the Lya forest blanketing is smaller). Their spectrum may also be characterized by a strong Lya emission line. However, the spectra of galaxies seen behind intervening gas that is still neutral should show the red damping wing of the Gunn-Peterson trough, with a predictable profile that obstructs part of the Lya emission. The low-mass galaxies formed before reionization might constitute a distinctive population; we discuss the signature that this population could have in the faint number counts. Although most of these galaxies should have merged into larger ones, those that survived to the present could be dwarf spheroidals.

We study the clustering properties of z [?] 5.7 Lya emitters (LAEs) in a cosmological reionization simulation with a full Lya radiative transfer calculation. Lya radiative transfer substantially modifies the intrinsic Lya emission properties, compared to observed ones, depending on the density and velocity structure environment around the Lya-emitting galaxy. This environment-dependent Lya selection introduces new features in LAE clustering, suppressing (enhancing) the line-of-sight (transverse) density fluctuations and giving rise to scale-dependent galaxy bias. In real space, the contours of the three-dimensional two-point correlation function of LAEs appear to be prominently elongated along the line of sight on large scales, an effect that is opposite to and much stronger than the linear redshift-space distortion effect. The projected two-point correlation function is greatly enhanced in amplitude by a factor of up to a few, compared to the case without the environment-dependent selection effect. The new features in LAE clustering can be understood with a simple, physically motivated model, where Lya selection depends on matter density, velocity, and their gradients. We discuss the implications and consequences of the effects on galaxy clustering from Lya selection in interpreting clustering measurements and in constraining cosmology and reionization from LAEs.