arXiv subjects

Rachel S. Somerville

Publications and source records attributed to Rachel S. Somerville.

At least 307 records · Page 17Linked to original sources

arXiv2004

Modeling the Multi-wavelength Universe: the assembly of massive galaxies

Pan-chromatic galaxy surveys are providing tightening constraints on the global mass assembly history, and high-resolution imaging of large fields is telling us when and where stars were formed. How well are state-of-the-art hierarchical galaxy formation models currently doing at reproducing these observations? I present results here that suggest that hierarchical models are doing quite well at reproducing the global star formation and stellar mass assembly history obtained from galaxies selected in the optical and Near IR. However, the same models fail to reproduce two very important populations at high redshift: quiescent red spheroids and vigorously star-forming, dust-enshrouded starbursts. This mismatch carries important lessons about how star formation is triggered and regulated in early galaxies, and may force us to consider new ideas about the formation of massive spheroids.

astro-ph↗

arXiv2004

GEMS: Galaxy Evolution from Morphologies and SEDs

GEMS, Galaxy Evolution from Morphologies and SEDs, is a large-area (800 arcmin2) two-color (F606W and F850LP) imaging survey with the Advanced Camera for Surveys on HST. Centered on the Chandra Deep Field South, it covers an area of ~28'x28', or about 120 Hubble Deep Field areas, to a depth of m_AB(F606W)=28.3 (5sigma and m_AB(F850LP)=27.1 (5sigma) for compact sources. In its central ~1/4, GEMS incorporates ACS imaging from the GOODS project. Focusing on the redshift range 0.2<=z<=1.1, GEMS provides morphologies and structural parameters for nearly 10,000 galaxies where redshift estimates, luminosities and SEDs exist from COMBO-17. At the same time, GEMS contains detectable host galaxy images for several hundred faint AGN. This paper provides an overview of the science goals, the experiment design, the data reduction and the science analysis plan for GEMS.

astro-ph↗

arXiv2003

Evolution in the Colors of Lyman-Break Galaxies from Z~4 to Z~3

The integrated colors of distant galaxies provide a means for interpreting the properties of their stellar content. Here, we use rest-frame UV-to-optical colors to constrain the spectral-energy distributions and stellar populations of color-selected, B-dropout galaxies at z ~ 4 in the Great Observatories Origins Deep Survey. We combine the ACS data with ground-based near-infrared images, which extend the coverage of galaxies at z ~ 4 to the rest-frame B-band. We observe a color-magnitude trend in the rest-frame m(UV) - B versus B diagram for the z ~ 4 galaxies that has a fairly well-defined "blue-envelope", and is strikingly similar to that of color-selected, U-dropout galaxies at z ~ 3. We also find that although the co-moving luminosity density at rest-frame UV wavelengths (1600 Angstroms) is roughly comparable at z ~ 3 and z ~ 4, the luminosity density at rest-frame optical wavelengths increases by about one-third from z ~ 4 to z ~ 3. Although the star-formation histories of individual galaxies may involve complex and stochastic events, the evolution in the global luminosity density of the UV-bright galaxy population corresponds to an average star-formation history with a star-formation rate that is constant or increasing over these redshifts. This suggests that the evolution in the luminosity density corresponds to an increase in the stellar-mass density of more than 33%.

astro-ph↗

arXiv2003

Morphologies and Spectral Energy Distributions of Extremely Red Galaxies in the GOODS-South Field

Using U'- through Ks-band imaging data in the GOODS-South field, we construct a large, complete sample of 275 ``extremely red objects'' (EROs; K_s<22.0, R-K_s>3.35; AB), all with deep HST/ACS imaging in B_435, V_606, i_775, and z_850, and well-calibrated photometric redshifts. Quantitative concentration and asymmetry measurements fail to separate EROs into distinct morphological classes. We therefore visually classify the morphologies of all EROs into four broad types: ``Early'' (elliptical-like), ``Late'' (disk galaxies), ``Irregular'' and ``Other'' (chain galaxies and low surface brightness galaxies), and calculate their relative fractions and comoving space densities. For a broad range of limiting magnitudes and color thresholds, the relative number of early-type EROs is approximately constant at 33-44%, and the comoving space densities of Early- and Late-type EROs are comparable. Mean rest-frame spectral energy distributions (SEDs) at wavelengths between 0.1 and 1.2 um are constructed for all EROs. The SEDs are extremely similar in their range of shapes, independent of morphological type. The implication is that any differences between the broad-band SEDs of Early-type EROs and the other types are relatively subtle, and there is no robust way of photometrically distinguishing between different morphological types with usual optical/near-infrared photometry.

astro-ph↗

arXiv2003

The Size Evolution of High Redshift Galaxies

Hubble Space Telescope images of high-redshift galaxies selected via color and photometric redshifts are used to examine the size and axial-ratio distribution of galaxies as a function of redshift at lookback times t > 8 Gyr. These parameters are measured at rest-frame UV wavelengths (1200-2000 Angstroms) on images with a rest-frame resolution of less than 0.8 kpc. Galaxy radii are found to scale with redshift approximately as the inverse of the Hubble parameter H(z). This is in accord with the theoretical expectation that the typical sizes of the luminous parts of galaxies should track the expected evolution in the virial radius of dark-matter halos. The mean ratio of semi-major to semi-minor axis for a bright well-resolved sample of galaxies at z ~ 4 is b/a = 0.65, suggesting that these Lyman break galaxies are not drawn from a spheroidal population. However the median concentration index of this sample is C = 3.5, which is closer to the typical concentration indices, C ~ 4, of nearby elliptical galaxies than to the values, C<2 for local disk galaxies of type Sb and later.

astro-ph↗

arXiv2003

Cosmic Variance in the Great Observatories Origins Deep Survey

Cosmic variance is the uncertainty in observational estimates of the volume density of extragalactic objects such as galaxies or quasars arising from the underlying large-scale density fluctuations. This is often a significant source of uncertainty, especially in deep galaxy surveys, which tend to cover relatively small areas. We present estimates of the relative cosmic variance for one-point statistics (i.e. number densities) for typical scales and volumes sampled by the Great Observatories Origins Deep Survey (GOODS). We use two approaches: for objects with a known two-point correlation function that is well-approximated by a power law, one can use the standard analytic formalism to calculate the cosmic variance (in excess of shot noise). We use this approach to estimate the cosmic variance for several populations that are being studied in the GOODS program: Extremely Red Objects (ERO) at z~1, and Lyman Break Galaxies (LBG) at z~3 and z~4, using clustering information for similar populations in the literature. For populations with unknown clustering, one can use predictions from Cold Dark Matter theory to obtain a rough estimate of the variance as a function of number density. We present a convenient plot which allows one to use this approach to read off the cosmic variance for a population with a known mean redshift and estimated number density. We conclude that for the volumes sampled by GOODS, cosmic variance is a significant source of uncertainty for strongly clustered objects (~40-60% for EROs) and less serious for less clustered objects, ~10-20% for LBGs.

astro-ph↗

arXiv2003

The Redshift Distribution of Near-IR Selected Galaxies in the Great Observatories Origins Deep Survey as a Test of Galaxy Formation Scenarios

The redshift distribution of near-IR selected galaxies is often used to attempt to discriminate between the classical view of galaxy formation, in which present-day luminous galaxies were assembled at early times and evolve due to the passive aging of their stellar populations, and that of hierarchical structure formation, in which galaxies were assembled more recently via the merging of smaller objects. We carry out such a test here, by computing the distribution of photometric redshifts of K_AB<22 galaxies in the Great Observatories Origins Deep Field Survey GOODS Southern Field, and comparing the results with predictions from a semi-analytic model based on hierarchical structure formation, and a classical `passive evolution' model. We find that the redshift distributions at z <~ 1.5 of both the hierarchical and passive models are very similar to the observed one. At z>~1.5, the hierarchical model shows a deficit of galaxies, while the passive model predicts an excess. We investigate the nature of the observed galaxies in the redshift range where the models diverge, and find that the majority have highly disturbed morphologies, suggesting that they may be merger-induced starbursts. While the hierarchical model used here does not produce these objects in great enough numbers, the appearance of this population is clearly in better qualitative agreement with the hierarchical picture than with the classical passive evolution scenario. We conclude that the observations support the general framework of hierarchical formation, but suggest the need for new or modified physics in the models.

astro-ph↗

arXiv2003

The Epoch of Reionization in Models with Reduced Small Scale Power

Reducing the power on small scales relative to the `standard' LCDM model alleviates a number of possible discrepancies with observations, and is favored by the recent analysis of WMAP plus galaxy and Lyman-alpha forest data. Here, we investigate the epoch of reionization in several models normalized to WMAP on large scales, and with sufficiently reduced power on small scales to solve the halo concentration and substructure problems. These include a tilted model, the WMAP running-index model, and a Warm Dark Matter model. We assume that the Universe was reionized by stellar sources comprised of a combination of supermassive (~200 M_sun) Pop III stars and Pop II stars with a `normal' IMF. We find that in all of these models, structure formation and hence reionization occurs late, certainly at redshifts below ten, and more probably at z < 6. This is inconsistent (at 2 sigma) with the determination of z_reion ~ 17 from the WMAP temperature-polarization data and is only marginally consistent with SDSS quasar observations. The tension between the galactic-scale observations, which favor low-power models, and the early reionization favored by WMAP can only be resolved if the efficiency of Pop III star formation is dramatically higher than any current estimate, or if there is an exotic population of ionizing sources such as mini-quasars. Otherwise, we may have to live with the standard LCDM power spectrum, and solve the small-scale problems in some other way.

astro-ph↗

arXiv2003

Star Formation at the Twilight of the Dark Ages: Which Stars Reionized the Universe?

We calculate the global star formation rate density (SFRD) from z ~ 30-3 using a semi-analytic model incorporating the hierarchical assembly of dark matter halos, gas cooling via atomic hydrogen, star formation, supernova feedback, and suppression of gas collapse in small halos due to the presence of a photoionizing background. We compare the results with the predictions of simpler models based on the rate of dark matter halo growth and a fixed ratio of stellar-to-dark mass, and with observational constraints on the SFRD at 3 < z < 6. We also estimate the star formation rate due to very massive, metal-free Pop III stars using a simple model based on the halo formation rate, calibrated against detailed hydrodynamic simulations of Pop III star formation. We find that the total production rate of hydrogen-ionizing photons during the probable epoch of reionization (15 < z < 20) is approximately equally divided between Pop II and Pop III stars, and that if reionization is late (less than about 15, close to the lower limit of the range allowed by the WMAP results), then Pop II stars alone may be able to reionize the Universe.

astro-ph↗

arXiv2002

Damped Lyman alpha systems and galaxy formation models - II. High ions and Lyman limit systems

We investigate a model for the high-ionization state gas associated with observed damped Lyman-alpha systems, based on a semi-analytic model of galaxy formation set within the paradigm of hierarchical structure formation. In our model, the hot gas in halos and sub-halos gives rise to CIV absorption, while the low-ionization state gas is associated with the cold gas in galaxies. The model matches the distribution of CIV column densities and leads naturally to kinematic properties that are in good agreement with the data. We examine the contribution of both hot and cold gas to sub-damped systems and suggest that the properties of these systems can be used as an important test of the model. We expect that sub-DLA systems will generally be composed of a single gas disk and thus predict that they should have markedly different kinematics than the damped systems. Finally, we find that hot halo gas produces less than one third of Lyman limit systems at redshift three. We model the contribution of mini-halos (halos with virial velocities < 35 km/s) to Lyman limit systems and find that they may contain as much gas as is observed in these systems. However, if we adopt realistic models of the gas density distribution we find that these systems are not a significant source of Lyman limit absorption. Instead we suggest that uncollapsed gas outside of virialized halos is responsible for most of the Lyman limit systems at high redshift.

astro-ph↗

arXiv2002

Estimating photometric redshifts with artificial neural networks

A new approach to estimating photometric redshifts - using Artificial Neural Networks (ANNs) - is investigated. Unlike the standard template-fitting photometric redshift technique, a large spectroscopically-identified training set is required but, where one is available, ANNs produce photometric redshift accuracies at least as good as and often better than the template-fitting method. The Bayesian priors on the underlying redshift distribution are automatically taken into account. Furthermore, inputs other than galaxy colours - such as morphology, angular size and surface brightness - may be easily incorporated, and their utility assessed. Different ANN architectures are tested on a semi-analytic model galaxy catalogue and the results are compared with the template-fitting method. Finally the method is tested on a sample of ~ 20000 galaxies from the Sloan Digital Sky Survey. The r.m.s. redshift error in the range z < 0.35 is ~ 0.021.

astro-ph↗

arXiv2002

Evaluating the disk formation paradigm in the LCDM framework: constraints from the Milky Way

We investigate the properties of a galaxy similar to the Milky Way within the context of standard disk formation theory in a LCDM universe. Using the standard assumption that baryons conserve specific angular momentum when they collapse, we conclude that the mean properties of the model galaxies are in good agreement with the Milky Way and other similar spiral galaxies, but the predicted scatter in disk scale lengths may be too large. A model in which half of the initial specific angular momentum is transfered to the dark matter may produce a smaller scatter, if very compact disks are unstable and evolve into spheroids or early type galaxies.

astro-ph↗

arXiv2002

Studying Galaxy Formation with Hubble's Successor

In this paper, I discuss the capabilities and limitations of an 8-10 meter ultraviolet/optical telescope in space, the proposed successor to the Hubble Space Telescope, in the context of galaxy studies. The exquisite spatial resolution and excellent sensitivity of such a facility would open up new possibilities for the study of nearby dwarf galaxies ($z \la 0.5$), and for studying the internal structure and kinematics of more luminous galaxies at high redshift ($z \ga 2$). These applications are of particular importance because they would address areas in which the popular Cold Dark Matter theory is in potential conflict with observations.

astro-ph↗

arXiv2002

The Masses, Ancestors and Descendents of Extremely Red Objects: Constraints from Spatial Clustering

Wide field near-infrared (IR) surveys have revealed a population of galaxies with very red optical$-$IR colors, which have been termed ``Extremely Red Objects'' (EROs). Modeling suggests that such red colors (R-K > 5) could be produced by galaxies at z>~1 with either very old stellar populations or very high dust extinction. Recently it has been discovered that EROs are strongly clustered. Are these objects the high-redshift progenitors of present day giant ellipticals (gEs)? Are they already massive at this epoch? Are they the descendents of the $z\sim3$ Lyman Break Galaxies (LBG), which have also been identified as possible high redshift progenitors of giant ellipticals? We address these questions within the framework of the Cold Dark Matter paradigm using an analytic model that connects the number density and clustering or bias of an observed population with the halo occupation function (the number of observed galaxies per halo of a given mass). We find that EROs reside in massive dark matter halos, with average mass > 1E13/h100 Msun. The occupation function that we derive for EROs is very similar to the one we derive for z=0, L>L* early type galaxies, whereas the occupation function for LBGs is skewed towards much smaller host halo masses ( ~ 1E11 to 1E12/h100 Msun. We then use the derived occupation function parameters to explore the possible evolutionary connections between these three populations.

astro-ph↗

arXiv2002

Squelched Galaxies and Dark Halos

There is accumulating evidence that the faint end of the galaxy luminosity function might be very different in different locations. The luminosity function might be rising in rich clusters and flat or declining in regions of low density. If galaxies form according to the model of hierarchical clustering then there should be many small halos compared to the number of big halos. If this theory is valid then there must be a mechanism that eliminates at least the visible component of galaxies in low density regions. A plausible mechanism is photoionization of the intergalactic medium at a time before the epoch that most dwarf galaxies form in low density regions but after the epoch of formation for similar systems that ultimately end up in rich clusters. The dynamical timescales are found to accommodate this hypothesis in a flat universe with Omega_m < 0.4. If small halos exist but simply cannot be located because they have never become the sites of significant star formation, they still might have dynamical manifestations. These manifestations are hard to identify in normal groups of galaxies because small halos do not make a significant contribution to the global mass budget. However, it could be entertained that there are clusters of halos where there are only small systems, clusters that are at the low mass end of the hierarchical tree. There may be places where only a few small galaxies managed to form, enough for us to identify and use as test probes of the potential. It turns out that such environments might be common. Four probable groups of dwarfs are identified within 5 Mpc and the assumption they are gravitationally bound suggests M/L_B ~ 300 - 1200 M_sun/L_sun, 6 +/- factor 2 times higher than typical values for groups with luminous galaxies.

astro-ph↗

arXiv2001

The Masses of Lyman Break Galaxies

Data on galaxies at high redshift, identified by the Lyman-break photometric technique, can teach us about how galaxies form and evolve. The stellar masses and other properties of such Lyman break galaxies (LBGs) depend sensitively on the details of star formation. In this paper we consider three different star formation prescriptions, and use semi-analytic methods applied to the now-standard $Λ$CDM theory of hierarchical structure formation to show how these assumptions about star formation affect the predicted masses of the stars in these galaxies and the masses of the dark matter halos that host them. We find that, within the rather large uncertainties, recent estimates of the stellar masses of LBGs from multi-color photometry are consistent with the predictions of all three models. However, the estimated stellar masses are more consistent with the predictions of two of the models in which star formation is accelerated at high redshifts $z\gsim3$, and of these models the one in which many of the LBGs are merger-driven starbursts is also more consistent with indications that many high redshift galaxies are gas rich. The clustering properties of LBGs have put some constraints on the masses of their host halos, but due to similarities in the halo occupation of the three models we consider and degeneracies between model parameters, current constraints are not yet sufficient to distinguish between realistic models.

astro-ph↗

arXiv2001

LCDM-based models for the Milky Way and M31 I: Dynamical Models

We apply standard disk formation theory with adiabatic contraction within cuspy halo models predicted by the standard LCDM cosmology. The resulting models score remarkably well when confronted with the broad range of observational data available for the Milky Way and M31 galaxies, giving a Milky Way virial mass of 1-2x10^12Msun and concentration C=12-17. We consider two types of models, in which: (A) baryons conserve angular momentum and (B) some of the angular momentum of the baryons is transferred to the dark matter. Type-A models produce good agreement with observed rotation curves and obey constraints in the solar neighborhood, but may have too much dark matter in the center to allow a fast rotating bar. The type-B models with angular momentum transport have a slightly more massive disk and less dark matter in the central part, allowing a fast rotating bar to persist. Both classes of models probably have sufficient baryonic mass in the central 3.5kpc to reproduce recent observational values of the optical depth to microlensing events towards the Galactic center. All models require that about 1/2 of all baryons expected inside the virial radius must not be in the disk or bulge. We investigate whether the range of virial masses allowed by our dynamical models is compatible with constraints from the galaxy luminosity function, and find a range of parameter space that is allowed by this constraint. We conclude that rotation curves and dynamical properties of ``normal'' high surface brightness spiral galaxies appear to be consistent with standard LCDM.

astro-ph↗

arXiv2001

Can Photoionization Squelching Resolve the Sub-structure Crisis?

Cold Dark Matter theory predicts that the Local Group should contain many more dwarf-sized objects than the observed number of dwarf galaxies --- the so-called sub-structure problem. We investigate whether the suppression of star formation in these small objects due to the presence of a photoionizing background can resolve the problem. We make use of results from recent hydrodynamic simulations to build a recipe for the suppression of gas infall into semi-analytic galaxy formation models, and use these to predict the luminosity function of dwarf galaxies in the Local Group. In the models without photoionization ``squelching'', we predict a large excess of faint dwarf galaxies compared with the observed number in the Local Group --- thus, the usual recipe for supernovae feedback used in semi-analytic models does not solve the sub-structure problem on its own. When we include photoionization squelching, we find good agreement with the observations. We have neglected tidal destruction, which probably further reduces the number of dwarf galaxies. We conclude that photoionizing squelching easily solves the sub-structure problem. In fact, it is likely that once this effect is taken into account, models with reduced small-scale power (e.g. Warm Dark Matter) would underproduce dwarf galaxies.

astro-ph↗