Abstract. I review some recent progress made in our understanding of galaxy evolution and the cosmic history of star formation. Like bookends, the results obtained from deep ground-based spectroscopy and from the Hubble Deep Field imaging survey put brackets around the intermediate redshift interval, 1 < z < 2, where starbirth probably peaked at a rate 10 times higher than today. The steady decline observed since z ∼ 1 is largely associated with late-type galaxies. At z > ∼ 2.5, the Lyman-break selected objects may represent the precursors of present-day spheroids, but appear, on average, quite underluminous relative to the expectations of the standard early-and-rapidly forming picture for spheroidal systems. The observed ultraviolet light density accounts for the bulk of the metals seen today in "normal" massive galaxies.

The Lyman decrement associated with the cumulative effect of H I in QSO absorption systems along the line of sight provides a distinctive feature for identifying galaxies at z ∼ > 2.5. Color criteria, which are sensitive to the presence of a Lyman-continuum break superposed on an otherwise flat UV spectrum, have been shown, through Keck spectroscopy, to successfully identify a substantial population of star-forming galaxies at 3 ∼ < z ∼ < 3.5 (Steidel et al. 1996a). Such objects have proven surprisingly elusive in fieldgalaxy redshift surveys; quantifying their surface density and morphology is crucial for determining how and when galaxies formed. The Hubble Deep Field (HDF) observations offer the opportunity to exploit the ubiquitous effect of intergalactic absorption and obtain useful statistical constraints on the redshift distribution of galaxies considerably fainter than current spectroscopic limits. We model the H I cosmic opacity as a function of redshift, including scattering in resonant lines of the Lyman series and Lymancontinuum absorption, and use stellar population synthesis models with a wide variety of ages, metallicities, dust contents, and redshifts, to derive color selection criteria that provide a robust separation between high redshift and low redshift galaxies. From the HDF images we construct a sample of star-forming galaxies at 2 ∼ < z ∼ < 4.5. While none of the ∼ 60 objects in the HDF having known Keck/LRIS spectroscopic redshifts in the range 0 ∼ < z ∼ < 1.4 is found to contaminate our high-redshift sample, our color criteria are able to efficiently select the 2.6 ∼ < z ∼ < 3.2 galaxies identified by Steidel et al. (1996b).The ultraviolet (and blue) dropout technique opens up the possibility of investigating cosmic star and element formation in the early universe. We set a lower-limit to the ejection rate of heavy elements per unit comoving volume from Type II supernovae at z = 2.75 of ≈ 3.6 × 10 −4 M ⊙ yr −1 Mpc −3 (for q 0 = 0.5 and H 0 = 50 km s −1 Mpc −1 ), which is 3 times higher than the local value, but still 4 times lower than the rate observed at z ≈ 1. At z = 4, our lower limit to the cosmic metal ejection rate is ≈ 3 times lower than the z = 2.75 value. We discuss the implications of these results on models of galaxy formation, and on the chemical enrichment and ionization history of the intergalactic medium.

The integrated ultraviolet ux arising from QSOs and/or hot, massive stars in metal-producing young galaxies is likely responsible for maintaining the intergalactic di use gas and the Ly forest clouds in a highly ionized state. The spectrum and intensity of such UV background have generally been obtained by modeling the reprocessing due to intervening material as a pure photoelectric absorption process. However, owing to the emission from radiative recombinations within the absorbing clouds, a photoionized clumpy medium could contribute substantially to the metagalactic ux. In other words, QSO absorption-line systems are sources not just sinks of ionizing photons. We present a detailed calculation of the propagation of AGN-like ionizing radiation through the intergalactic space. We model the ionization state of absorbing clouds, and show that the universe will be more opaque above 4 Ryd than previously estimated. Singly ionized helium in Ly forest clouds and Lyman-limit systems is found to be very e cient in reprocessing soft X-ray, helium-ionizing photons into ultraviolet, hydrogen-ionizing ones. We demonstrate that a signi cant fraction of the absorbed primary photons (emitted, e.g., by quasar sources) will be reradiated by the photoionized gas through Ly line emission, two-photon continuum, and recombination continuum radiation.In the light of new data and recent studies, we also reassess the contribution of the QSOs observed in optical surveys to the UV extragalactic background, and nd that the stochastic reprocessing of quasar Lyman continuum radiation by hydrogen and helium along the line of sight will signi cantly a ect the amplitude, spectral shape, and uctuations properties of the metagalactic ux. In a scenario in which QSOs are the primary source of ionizing photons in the universe, the integrated H I Ly emission at z = 0 from photoionized Ly clouds and Lymanlimit systems is found to be at a level of less than 5% of current observational limits on the far-UV extragalactic radiation ux. We show that J 912 increases from 10 23 ergs cm 2 s 1 Hz 1 sr 1 at the present epoch to 5 10 22 ergs cm 2 s 1 Hz 1 sr 1 at z = 2:5. The attenuated direct quasar emission plus recombination radiation from intergalactic gas appears to provide enough hydrogen-ionizing photons to satisfy the proximity e ect at large redshift. The He II /H I ratio in the di use intergalactic medium and the Ly clouds increases from 25 at z = 0 to 45 at z = 2:5, to decrease again below 30 for z > 4:5. The spectrum of the ionizing background at high redshift is shown to have a hump at energies below 40.8 eV due to redshift-smeared He II Ly line and two-photon continuum emission. We propose that observations of low-ionization species such { 3 {

We present improved synthesis models of the evolving spectrum of the UV/X-ray diffuse background, updating and extending our previous results. Five new main components are added to our radiative transfer code CUBA: (1) the sawtooth modulation of the background intensity from resonant line absorption in the Lyman series of cosmic hydrogen and helium; (2) the X-ray emission from the obscured and unobscured quasars that gives origin to the X-ray background; (3) a piecewise parameterization of the distribution in redshift and column density of intergalactic absorbers that fits recent measurements of the mean free path of 1 ryd photons; (4) an accurate treatment of the photoionization structure of absorbers, which enters in the calculation of the helium continuum opacity and recombination emissivity; and (5) the UV emission from star-forming galaxies at all redshifts. We provide tables of the predicted H i and He ii photoionization and photoheating rates for use, e.g., in cosmological hydrodynamics simulations of the Lyα forest and a new metallicity-dependent calibration to the UV luminosity density-star formation rate density relation. A "minimal cosmic reionization model" is also presented in which the galaxy UV emissivity traces recent determinations of the cosmic history of star formation, the luminosity-weighted escape fraction of hydrogen-ionizing radiation increases rapidly with look-back time, the clumping factor of the highredshift intergalactic medium evolves following the results of hydrodynamic simulations, and Population III stars and miniquasars make a negligible contribution to the metagalactic flux. The model provides a good fit to the hydrogenionization rates inferred from flux decrement and proximity effect measurements, predicts that cosmological H ii (He iii) regions overlap at redshift 6.7 (2.8), and yields an optical depth to Thomson scattering, τ es = 0.084 that is in agreement with Wilkinson Microwave Anisotropy Probe results. Our new background intensities and spectra are sensitive to a number of poorly determined input parameters and suffer from various degeneracies. Their predictive power should be constantly tested against new observations. We are therefore making our redshift-dependent UV/X emissivities and CUBA outputs freely available for public use at http://www.ucolick.org/∼pmadau/CUBA.

We present two new non-parametric methods for quantifying galaxy morphology: the relative distribution of the galaxy pixel flux values (the Gini coefficient or G) and the second-order moment of the brightest 20% of the galaxy's flux (M 20 ). We test the robustness of G and M 20 to decreasing signal-to-noise and spatial resolution, and find that both measures are reliable to within 10% for images with average signal-to-noise per pixel greater than 2 and resolutions better than 1000 pc and 500 pc, respectively. We have measured G and M 20 , as well as concentration (C), asymmetry (A), and clumpiness (S) in the rest-frame near-ultraviolet/optical wavelengths for 148 bright local "normal" Hubble type galaxies (E-Sd) galaxies, 22 dwarf irregulars, and 73 0.05 < z < 0.25 ultraluminous infrared galaxies (ULIRGs). We find that most local galaxies follow a tight sequence in G − M 20 − C, where early-types have high G and C and low M 20 and latetype spirals have lower G and C and higher M 20 . The majority of ULIRGs lie above the normal galaxy G − M 20 sequence, due to their high G and M 20 values. Their high Gini coefficients arise from very bright nuclei, while the high second-order moments are produced by multiple nuclei and bright tidal tails. All of these features are signatures of recent and on-going mergers and interactions. We also find that in combination with A and S, G is more effective than C at distinguishing ULIRGs from the "normal" Hubble-types. Finally, we measure the morphologies of 49 1.7 < z < 3.8 galaxies from HST NICMOS observations of the Hubble Deep Field North. We find that many of the z ∼ 2 galaxies possess G and A higher than expected from degraded images of local elliptical and spiral galaxies, and have morphologies more like low-redshift ULIRGs.

The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS) is designed to document the first third of galactic evolution, over the approximate redshift (z) range 8-1.5. It will image >250,000 distant galaxies using three separate cameras on the Hubble Space Telescope, from the mid-ultraviolet to the near-infrared, and will find and measure Type Ia supernovae at z > 1.5 to test their accuracy as standardizable candles for cosmology. Five premier multi-wavelength sky regions are selected, each with extensive ancillary data. The use of five widely separated fields mitigates cosmic variance and yields statistically robust and complete samples of galaxies down to a stellar mass of 10 9 M to z ≈ 2, reaching the knee of the ultraviolet luminosity function of galaxies to z ≈ 8. The survey covers approximately 800 arcmin 2 and is divided into two parts. The CANDELS/Deep survey (5σ point-source limit H = 27.7 mag) covers ∼125 arcmin 2 within Great Observatories Origins Deep Survey (GOODS)-N and GOODS-S. The CANDELS/Wide survey includes GOODS and three additional fields (Extended Groth Strip, COSMOS, and Ultra-deep Survey) and covers the full area to a 5σ pointsource limit of H 27.0 mag. Together with the Hubble Ultra Deep Fields, the strategy creates a three-tiered "wedding-cake" approach that has proven efficient for extragalactic surveys. Data from the survey are nonproprietary and are useful for a wide variety of science investigations. In this paper, we describe the basic motivations for the survey, the CANDELS team science goals and the resulting observational requirements, the field selection and geometry, and the observing design. The Hubble data processing and products are described in a companion paper.

We develop a method for interpreting faint galaxy data which focuses on the integrated light radiated from the galaxy population as a whole. The emission history of the universe at ultraviolet, optical, and near-infrared wavelengths is modeled from the present epoch to z~4 by tracing the evolution with cosmic time of the galaxy luminosity density, as determined from several deep spectroscopic samples and the Hubble Deep Field (HDF) imaging survey. The global spectrophotometric properties of field galaxies can be well fit by a simple stellar evolution model, defined by a time-dependent star formation rate (SFR) per unit comoving volume and a universal IMF extending from 0.1 to 125 M_sun. In the best-fit models, the global SFR rises sharply, by about an order of magnitude, from a redshift of zero to a peak value at z~1.5, to fall again at higher redshifts. The models are able to account for the entire background light recorded in the galaxy counts down to the very faint magnitude levels probed by the HDF. Since only 20% of the current stellar content of galaxies is produced at z>2, a rather low cosmic metallicity is expected at these early times, in good agreement with the observed enrichment history of the damped Lyman-\alpha systems. A ``monolithic collapse'' model, where half of the present-day stars formed at z>2.5 and were shrouded by dust, can be made consistent with the global history of light, but overpredicts the metal mass density at high redshifts as sampled by QSO absorbers.Comment: LaTeX, 33 pages, 9 figures, uses aasms4 style, submitted to Ap

For almost two decades the properties of 'dwarf' galaxies have challenged the cold dark matter (CDM) model of galaxy formation. Most observed dwarf galaxies consist of a rotating stellar disk embedded in a massive dark-matter halo with a near-constant-density core. Models based on the dominance of CDM, however, invariably form galaxies with dense spheroidal stellar bulges and steep central dark-matter profiles, because low-angular-momentum baryons and dark matter sink to the centres of galaxies through accretion and repeated mergers. Processes that decrease the central density of CDM halos have been identified, but have not yet reconciled theory with observations of present-day dwarfs. This failure is potentially catastrophic for the CDM model, possibly requiring a different dark-matter particle candidate. Here we report hydrodynamical simulations (in a framework assuming the presence of CDM and a cosmological constant) in which the inhomogeneous interstellar medium is resolved. Strong outflows from supernovae remove low-angular-momentum gas, which inhibits the formation of bulges and decreases the dark-matter density to less than half of what it would otherwise be within the central kiloparsec. The analogues of dwarf galaxies-bulgeless and with shallow central dark-matter profiles-arise naturally in these simulations.