Search NASA⌕ Search

Engineering topics

Gebhardt, Karl

Publications and source records attributed to Gebhardt, Karl.

At least 37 records · Page 2

Ancient light from young cosmic cities: Physical and observational signatures of galaxy proto-clusters

A growing number of galaxy clusters at z = 1-2 is being discovered as part of deep optical, IR, X-ray, and Sunyaev-Zel'dovich effect surveys. For a complete picture of cluster formation, however, it is important that we also start probing the much earlier epoch, between redshifts of about 2 and 7, during which these clusters and their galaxies first began to form. Because the study of these so-called proto-clusters is currently quite limited by small number statistics, widely varying selection techniques, and many assumptions, we have performed a large systematic study of cluster formation utilizing cosmological simulations. We use the Millennium Simulations to track the evolution of dark matter and galaxies in about 3000 clusters from the earliest times to z = 0. We define an effective radius R{sub e} for proto-clusters and characterize their growth in size and mass with cosmic time. We show that the progenitor regions of galaxy clusters (ranging in mass from ∼10{sup 14} to a few times 10{sup 15} M {sub ☉}) can already be identified in galaxy surveys at very early times (at least up to z ∼ 5), provided that the galaxy overdensities are measured on a sufficiently large scale (R{sub e} ∼ 5-10 Mpc comoving) and with sufficient statistics. We present the overdensities in matter, dark matter halos, and galaxies as functions of present-day cluster mass, redshift, bias, and window size that can be used to interpret the wide range of structures found in real surveys. We also derive the probability that a structure having a galaxy overdensity δ{sub gal}, defined by a set of observational selection criteria, is indeed a proto-cluster, and we show how their z = 0 masses can already be estimated long before virialization. We present overdensity profiles as a function of radius, and we further show how the projected surface overdensities of proto-clusters decrease as the uncertainties in redshift measurements increase. We provide a table of proto-cluster candidates selected from the literature and discuss their properties in light of our simulation predictions. This paper provides the general framework that will allow us to extend the study of cluster formation out to much higher redshifts using the large number of proto-clusters that are expected to be discovered in, e.g., the upcoming HETDEX and Hyper Suprime-Cam surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

Angular momenta, dynamical masses, and mergers of brightest cluster galaxies

Using the VIMOS integral field unit (IFU) spectrograph on the Very Large Telescope, we have spatially mapped the kinematic properties of 10 nearby brightest cluster galaxies (BCGs) and 4 BCG companion galaxies located within a redshift of z = 0.1. In the hierarchical formation model, these massive galaxies (10{sup 10.5} M {sub ☉} < M {sub dyn} < 10{sup 11.9} M {sub ☉}) are expected to undergo more mergers than lower mass galaxies, and simulations show that dry minor mergers can remove angular momentum. We test whether BCGs have low angular momenta by using the λ {sub Re} parameter developed by the SAURON and ATLAS{sup 3D} teams and combine our kinematics with Sloan Digital Sky Survey photometry to analyze the BCGs' merger status. We find that 30% (3/10) of the BCGs and 100% of the companion galaxies (4/4) are fast rotators as defined by the ATLAS{sup 3D} criteria. Our fastest rotating BCG has a λ {sub Re} = 0.35 ± 0.05. We increase the number of BCGs analyzed from 1 in the combined SAURON and ATLAS{sup 3D} surveys to 11 BCGs total and find that above M {sub dyn} ∼ 11.5 M {sub ☉}, virtually all galaxies, regardless of environment, are slow rotators. To search for signs of recent merging, we analyze the photometry of each system and use the G – M {sub 20} selection criteria to identify mergers. We find that 40% ± 20% of our BCGs are currently undergoing or have recently undergone a merger (within 0.2 Gyr). Surprisingly, we find no correlation between galaxies with high angular momentum and morphological signatures of merging.

79 ASTRONOMY AND ASTROPHYSICS↗

Galaxy redshift surveys with sparse sampling

Survey observations of the three-dimensional locations of galaxies are a powerful approach to measure the distribution of matter in the universe, which can be used to learn about the nature of dark energy, physics of inflation, neutrino masses, etc. A competitive survey, however, requires a large volume (e.g., V{sub survey} ∼ 10Gpc{sup 3}) to be covered, and thus tends to be expensive. A ''sparse sampling'' method offers a more affordable solution to this problem: within a survey footprint covering a given survey volume, V{sub survey}, we observe only a fraction of the volume. The distribution of observed regions should be chosen such that their separation is smaller than the length scale corresponding to the wavenumber of interest. Then one can recover the power spectrum of galaxies with precision expected for a survey covering a volume of V{sub survey} (rather than the volume of the sum of observed regions) with the number density of galaxies given by the total number of observed galaxies divided by V{sub survey} (rather than the number density of galaxies within an observed region). We find that regularly-spaced sampling yields an unbiased power spectrum with no window function effect, and deviations from regularly-spaced sampling, which are unavoidable in realistic surveys, introduce calculable window function effects and increase the uncertainties of the recovered power spectrum. On the other hand, we show that the two-point correlation function (pair counting) is not affected by sparse sampling. While we discuss the sparse sampling method within the context of the forthcoming Hobby-Eberly Telescope Dark Energy Experiment, the method is general and can be applied to other galaxy surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

THE STELLAR HALOS OF MASSIVE ELLIPTICAL GALAXIES. II. DETAILED ABUNDANCE RATIOS AT LARGE RADIUS

We study the radial dependence in stellar populations of 33 nearby early-type galaxies with central stellar velocity dispersions σ{sub *} ∼> 150 km s{sup –1}. We measure stellar population properties in composite spectra, and use ratios of these composites to highlight the largest spectral changes as a function of radius. Based on stellar population modeling, the typical star at 2R{sub e} is old (∼10 Gyr), relatively metal-poor ([Fe/H] ≈ –0.5), and α-enhanced ([Mg/Fe] ≈ 0.3). The stars were made rapidly at z ≈ 1.5-2 in shallow potential wells. Declining radial gradients in [C/Fe], which follow [Fe/H], also arise from rapid star formation timescales due to declining carbon yields from low-metallicity massive stars. In contrast, [N/Fe] remains high at large radius. Stars at large radius have different abundance ratio patterns from stars in the center of any present-day galaxy, but are similar to average Milky Way thick disk stars. Our observations are thus consistent with a picture in which the stellar outskirts are built up through minor mergers with disky galaxies whose star formation is truncated early (z ≈ 1.5-2)

79 ASTRONOMY AND ASTROPHYSICS↗

THE SPECTRALLY RESOLVED Lyα EMISSION OF THREE Lyα-SELECTED FIELD GALAXIES AT z ∼ 2.4 FROM THE HETDEX PILOT SURVEY

We present new results on the spectrally resolved Lyα emission of three Lyα-emitting field galaxies at z ∼ 2.4 with high Lyα equivalent width (>100 Å) and Lyα luminosity (∼10{sup 43} erg s{sup –1}). At 120 km s{sup –1} (FWHM) spectral resolution, the prominent double-peaked Lyα profile straddles the systemic velocity, where the velocity zero point is determined from spectroscopy of the galaxies' rest-frame optical nebular emission lines. The average velocity offset from systemic of the stronger redshifted emission component for our sample is 176 km s{sup –1} while the average total separation between the redshifted and main blueshifted emission components is 380 km s{sup –1}. These measurements are a factor of ∼2 smaller than for UV-continuum-selected galaxies that show Lyα in emission with lower Lyα equivalent widths. We compare our Lyα spectra to the predicted line profiles of a spherical 'expanding shell' Lyα radiative transfer grid that models large-scale galaxy outflows. Specifically, blueward of the systemic velocity where two galaxies show a weak, highly blueshifted (by ∼1000 km s{sup –1}) tertiary emission peak, the model line profiles are a relatively poor representation of the observed spectra. Since the neutral gas column density has a dominant influence over the shape of the Lyα line profile, we caution against equating the observed Lyα velocity offset with a physical outflow velocity, especially at lower spectral resolution where the unresolved Lyα velocity offset is a convoluted function of several degenerate parameters. Referring to rest-frame ultraviolet and optical Hubble Space Telescope imaging, we find that galaxy-galaxy interactions may play an important role in inducing a starburst that results in copious Lyα emission as well as perturbing the gas distribution and velocity field, both of which have strong influence over the Lyα emission line profile.

79 ASTRONOMY AND ASTROPHYSICS↗

VARIATIONS IN A UNIVERSAL DARK MATTER PROFILE FOR DWARF SPHEROIDALS

Using a newly developed modeling technique, we present orbit-based dynamical models of the Carina, Draco, Fornax, Sculptor, and Sextans dwarf spheroidal (dSph) galaxies. These models calculate the dark matter profiles non-parametrically without requiring any assumptions to be made about their profile shapes. By lifting this restriction, we discover a host of dark matter profiles in the dSphs that are different from the typical profiles suggested by both theorists and observers. However, when we scale these profiles appropriately and plot them on a common axis, they appear to follow an approximate r {sup –1} power law with considerable scatter.

79 ASTRONOMY AND ASTROPHYSICS↗

THE VIRUS-P EXPLORATION OF NEARBY GALAXIES (VENGA): SURVEY DESIGN, DATA PROCESSING, AND SPECTRAL ANALYSIS METHODS

We present the survey design, data reduction, and spectral fitting pipeline for the VIRUS-P Exploration of Nearby Galaxies (VENGA). VENGA is an integral field spectroscopic survey, which maps the disks of 30 nearby spiral galaxies. Targets span a wide range in Hubble type, star formation activity, morphology, and inclination. The VENGA data cubes have 5.''6 FWHM spatial resolution, {approx}5 A FWHM spectral resolution, sample the 3600 A-6800 A range, and cover large areas typically sampling galaxies out to {approx}0.7R{sub 25}. These data cubes can be used to produce two-dimensional maps of the star formation rate, dust extinction, electron density, stellar population parameters, the kinematics and chemical abundances of both stars and ionized gas, and other physical quantities derived from the fitting of the stellar spectrum and the measurement of nebular emission lines. To exemplify our methods and the quality of the data, we present the VENGA data cube on the face-on Sc galaxy NGC 628 (a.k.a. M 74). The VENGA observations of NGC 628 are described, as well as the construction of the data cube, our spectral fitting method, and the fitting of the stellar and ionized gas velocity fields. We also propose a new method to measure the inclination of nearly face-on systems based on the matching of the stellar and gas rotation curves using asymmetric drift corrections. VENGA will measure relevant physical parameters across different environments within these galaxies, allowing a series of studies on star formation, structure assembly, stellar populations, chemical evolution, galactic feedback, nuclear activity, and the properties of the interstellar medium in massive disk galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

THE EFFECT OF SPATIAL GRADIENTS IN STELLAR MASS-TO-LIGHT RATIO ON BLACK HOLE MASS MEASUREMENTS

We have tested the effect of spatial gradients in stellar mass-to-light ratio (Y) on measurements of black hole masses (M{sub .}) derived from stellar orbit superposition models. Such models construct a static gravitational potential for a galaxy and its central black hole, but typically assume spatially uniform Y. We have modeled three giant elliptical galaxies with gradients {alpha} {identical_to} dlog (Y)/dlog (r) from -0.2 to +0.1. Color and line strength gradients suggest mildly negative {alpha} in these galaxies. Introducing a negative (positive) gradient in Y increases (decreases) the enclosed stellar mass near the center of the galaxy and leads to systematically smaller (larger) M{sub .} measurements. For models with {alpha} = -0.2, the best-fit values of M{sub .} are 28%, 27%, and 17% lower than the constant-Y case, in NGC 3842, NGC 6086, and NGC 7768, respectively. For {alpha} = +0.1, M{sub .} are 14%, 22%, and 17% higher than the constant-Y case for the three respective galaxies. For NGC 3842 and NGC 6086, this bias is comparable to the statistical errors from individual modeling trials. At larger radii, negative (positive) gradients in Y cause the total stellar mass to decrease (increase) and the dark matter fraction within one effective radius to increase (decrease).

79 ASTRONOMY AND ASTROPHYSICS↗

THE VIRUS-P EXPLORATION OF NEARBY GALAXIES (VENGA): THE X {sub CO} GRADIENT IN NGC 628

We measure the radial profile of the {sup 12}CO(1-0) to H{sub 2} conversion factor (X {sub CO}) in NGC 628. The H{alpha} emission from the VENGA integral field spectroscopy is used to map the star formation rate (SFR) surface density ({Sigma}{sub SFR}). We estimate the molecular gas surface density ({Sigma}{sub H2}) from {Sigma}{sub SFR} by inverting the molecular star formation law (SFL), and compare it to the CO intensity to measure X {sub CO}. We study the impact of systematic uncertainties by changing the slope of the SFL, using different SFR tracers (H{alpha} versus far-UV plus 24 {mu}m), and CO maps from different telescopes (single-dish and interferometers). The observed X {sub CO} profile is robust against these systematics, drops by a factor of two from R {approx} 7 kpc to the center of the galaxy, and is well fit by a gradient {Delta}log(X {sub CO}) = 0.06 {+-} 0.02 dex kpc{sup -1}. We study how changes in X {sub CO} follow changes in metallicity, gas density, and ionization parameter. Theoretical models show that the gradient in X {sub CO} can be explained by a combination of decreasing metallicity, and decreasing {Sigma}{sub H2} with radius. Photoelectric heating from the local UV radiation field appears to contribute to the decrease of X {sub CO} in higher density regions. Our results show that galactic environment plays an important role at setting the physical conditions in star-forming regions, in particular the chemistry of carbon in molecular complexes, and the radiative transfer of CO emission. We caution against adopting a single X {sub CO} value when large changes in gas surface density or metallicity are present.

79 ASTRONOMY AND ASTROPHYSICS↗

MEASURING DARK MATTER PROFILES NON-PARAMETRICALLY IN DWARF SPHEROIDALS: AN APPLICATION TO DRACO

We introduce a novel implementation of orbit-based (or Schwarzschild) modeling that allows dark matter density profiles to be calculated non-parametrically in nearby galaxies. Our models require no assumptions to be made about velocity anisotropy or the dark matter profile. The technique can be applied to any dispersion-supported stellar system, and we demonstrate its use by studying the Local Group dwarf spheroidal galaxy (dSph) Draco. We use existing kinematic data at larger radii and also present 12 new radial velocities within the central 13 pc obtained with the VIRUS-W integral field spectrograph on the 2.7 m telescope at McDonald Observatory. Our non-parametric Schwarzschild models find strong evidence that the dark matter profile in Draco is cuspy for 20 {<=} r {<=} 700 pc. The profile for r {>=} 20 pc is well fit by a power law with slope {alpha} = -1.0 {+-} 0.2, consistent with predictions from cold dark matter simulations. Our models confirm that, despite its low baryon content relative to other dSphs, Draco lives in a massive halo.

79 ASTRONOMY AND ASTROPHYSICS↗

DYNAMICAL MEASUREMENTS OF BLACK HOLE MASSES IN FOUR BRIGHTEST CLUSTER GALAXIES AT 100 Mpc

We present stellar kinematics and orbit superposition models for the central regions of four brightest cluster galaxies, based upon integral-field spectroscopy at Gemini, Keck, and McDonald Observatories. Our integral-field data span radii from <100 pc to tens of kiloparsecs, comparable to the effective radius of each galaxy. We report black hole masses, M{sub .}, of 2.1{sup +1.6}{sub -1.6} Multiplication-Sign 10{sup 10} M{sub Sun} for NGC 4889, 9.7{sup +3.0}{sub -2.5} Multiplication-Sign 10{sup 9} M{sub Sun} for NGC 3842, and 1.3{sup +0.5}{sub -0.4} Multiplication-Sign 10{sup 9} M{sub Sun} for NGC 7768, with errors representing 68% confidence limits. For NGC 2832, we report an upper limit of M{sub .} < 9.0 Multiplication-Sign 10{sup 9} M{sub Sun }. Our models of each galaxy include a dark matter halo, and we have tested the dependence of M{sub .} on the model dark matter profile. Stellar orbits near the center of each galaxy are tangentially biased, on comparable spatial scales to the galaxies' photometric cores. We find possible photometric and kinematic evidence for an eccentric torus of stars in NGC 4889, with a radius of nearly 1 kpc. We compare our measurements of M{sub .} to the predicted black hole masses from various fits to the relations between M{sub .} and stellar velocity dispersion ({sigma}), luminosity (L), or stellar mass (M{sub *}). Still, the black holes in NGC 4889 and NGC 3842 are significantly more massive than all {sigma}-based predictions and most L-based predictions. The black hole in NGC 7768 is consistent with a broader range of predictions.

79 ASTRONOMY AND ASTROPHYSICS↗

THE STELLAR HALOS OF MASSIVE ELLIPTICAL GALAXIES

We use the Mitchell Spectrograph (formerly VIRUS-P) on the McDonald Observatory 2.7 m Harlan J. Smith Telescope to search for the chemical signatures of massive elliptical galaxy assembly. The Mitchell Spectrograph is an integral-field spectrograph with a uniquely wide field of view (107'' Multiplication-Sign 107''), allowing us to achieve remarkably high signal-to-noise ratios of {approx}20-70 pixel{sup -1} in radial bins of 2-2.5 times the effective radii of the eight galaxies in our sample. Focusing on a sample of massive elliptical galaxies with stellar velocity dispersions {sigma}{sub *} > 150 km s{sup -1}, we study the radial dependence in the equivalent widths (EW) of key metal absorption lines. By twice the effective radius, the Mgb EWs have dropped by {approx}50%, and only a weak correlation between {sigma}{sub *} and Mgb EW remains. The Mgb EWs at large radii are comparable to those seen in the centers of elliptical galaxies that are {approx} an order of magnitude less massive. We find that the well-known metallicity gradients often observed within an effective radius continue smoothly to 2.5 R{sub e} , while the abundance ratio gradients remain flat. Much like the halo of the Milky Way, the stellar halos of our galaxies have low metallicities and high {alpha}-abundance ratios, as expected for very old stars formed in small stellar systems. Our observations support a picture in which the outer parts of massive elliptical galaxies are built by the accretion of much smaller systems whose star formation history was truncated at early times.

79 ASTRONOMY AND ASTROPHYSICS↗

THE DARK MATTER DENSITY PROFILE OF THE FORNAX DWARF

We construct axisymmetric Schwarzschild models to measure the mass profile of the Local Group dwarf galaxy Fornax. These models require no assumptions to be made about the orbital anisotropy of the stars, as is the case for commonly used Jeans models. We test a variety of parameterizations of dark matter density profiles and find cored models with uniform density {rho}{sub c} = (1.6 {+-} 0.1) Multiplication-Sign 10{sup -2} M{sub Sun} pc{sup -3} fit significantly better than the cuspy halos predicted by cold dark matter simulations. We also construct models with an intermediate-mass black hole, but are unable to make a detection. We place a 1{sigma} upper limit on the mass of a potential intermediate-mass black hole at M{sub .} {<=} 3.2 Multiplication-Sign 10{sup 4} M{sub Sun }.

79 ASTRONOMY AND ASTROPHYSICS↗

THE CENTRAL DARK MATTER DISTRIBUTION OF NGC 2976

We study the mass distribution in the late-type dwarf galaxy NGC 2976 through stellar kinematics obtained with the Visible Integral Field Replicable Unit Spectrograph Prototype and anisotropic Jeans models as a test of cosmological simulations and baryonic processes that putatively alter small-scale structure. Previous measurements of the H{alpha} emission-line kinematics have determined that the dark matter halo of NGC 2976 is most consistent with a cored density profile. We find that the stellar kinematics are best fit with a cuspy halo. Cored dark matter halo fits are only consistent with the stellar kinematics if the stellar mass-to-light ratio is significantly larger than that derived from stellar population synthesis, while the best-fitting cuspy model has no such conflict. The inferred mass distribution from a harmonic decomposition of the gaseous kinematics is inconsistent with that of the stellar kinematics. This difference is likely due to the gas disk not meeting the assumptions that underlie the analysis such as no pressure support, a constant kinematic axis, and planar orbits. By relaxing some of these assumptions, in particular the form of the kinematic axis with radius, the gas-derived solution can be made consistent with the stellar kinematic models. A strong kinematic twist in the gas of NGC 2976's center suggests caution, and we advance the mass model based on the stellar kinematics as more reliable. The analysis of this first galaxy shows promising evidence that dark matter halos in late-type dwarfs may in fact be more consistent with cuspy dark matter distributions than earlier work has claimed.

79 ASTRONOMY AND ASTROPHYSICS↗

IS THERE A BLACK HOLE IN NGC 4382?

We present Hubble Space Telescope (HST) Space Telescope Imaging Spectrograph observations of the galaxy NGC 4382 (M85) and axisymmetric models of the galaxy to determine mass-to-light ratio (Y{sub V}) and central black hole mass (M{sub BH}). We find Y{sub V} = 3.74 {+-} 0.1 M{sub sun}/L{sub sun} and M{sub BH} = 1.3{sup +5.2}{sub -1.2} Multiplication-Sign 10{sup 7} M{sub sun} at an assumed distance of 17.9 Mpc, consistent with no black hole. The upper limit, M{sub BH} < 9.6 Multiplication-Sign 10{sup 7} M{sub sun}(2{sigma}) or M{sub BH} < 1.4 Multiplication-Sign 10{sup 8}(3{sigma}), is consistent with the current M-{sigma} relation, which predicts M{sub BH} = 8.8 Multiplication-Sign 10{sup 7} M{sub sun} at {sigma}{sub e} = 182 km s{sup -1}, but low for the current M - L relation, which predicts M{sub BH} = 7.8 Multiplication-Sign 10{sup 8} M{sub sun} at L{sub V} = 8.9 Multiplication-Sign 10{sup 10} L{sub sun,V}. HST images show the nucleus to be double, suggesting the presence of a nuclear eccentric stellar disk, analogous to the Tremaine disk in M31. This conclusion is supported by the HST velocity dispersion profile. Despite the presence of this non-axisymmetric feature and evidence of a recent merger, we conclude that the reliability of our black hole mass determination is not hindered. The inferred low black hole mass may explain the lack of nuclear activity.

79 ASTRONOMY AND ASTROPHYSICS↗

ORBIT-BASED DYNAMICAL MODELS OF THE SOMBRERO GALAXY (NGC 4594)

We present axisymmetric, orbit-based models to study the central black hole (BH), stellar mass-to-light ratio (M/L), and dark matter (DM) halo of NGC 4594 (M104, the Sombrero Galaxy). For stellar kinematics, we use published high-resolution kinematics of the central region taken with the Hubble Space Telescope, newly obtained Gemini long-slit spectra of the major axis, and integral field kinematics from the Spectroscopic Areal Unit for Research on Optical Nebulae instrument. At large radii, we use globular cluster kinematics to trace the mass profile and apply extra leverage to recovering the DM halo parameters. We find a BH of mass M{sub .} = (6.6 {+-} 0.4) x 10{sup 8} M{sub sun} and determine the stellar M/L{sub I} = 3.4 {+-} 0.05 (uncertainties are the 68% confidence band marginalized over the other parameters). Our best-fit DM halo is a cored logarithmic model with asymptotic circular speed V{sub c} = 376 {+-} 12 km s{sup -1} and core radius r{sub c} = 4.7 {+-} 0.6 kpc. The fraction of dark to total mass contained within the half-light radius is 0.52. Taking the bulge and disk components into account in our calculation of {sigma}{sub e} puts NGC 4594 squarely on the M-{sigma} relation. We also determine that NGC 4594 lies directly on the M-L relation.

79 ASTRONOMY AND ASTROPHYSICS↗

THE HETDEX PILOT SURVEY. II. THE EVOLUTION OF THE Ly{alpha} ESCAPE FRACTION FROM THE ULTRAVIOLET SLOPE AND LUMINOSITY FUNCTION OF 1.9 < z < 3.8 LAEs

We study the escape of Ly{alpha} photons from Ly{alpha} emitting galaxies (LAEs) and the overall galaxy population using a sample of 99 LAEs at 1.9 < z < 3.8 detected through integral-field spectroscopy of blank fields by The Hobby Eberly Telescope Dark Energy Experiment Pilot Survey. For 89 LAEs with broadband counterparts we measure ultraviolet (UV) luminosities and UV slopes, and estimate E(B - V) under the assumption of a constant intrinsic UV slope for LAEs. These quantities are used to estimate dust-corrected star formation rates (SFRs). Comparison between the observed Ly{alpha} luminosity and that predicted by the dust-corrected SFR yields the Ly{alpha} escape fraction. We also measure the Ly{alpha} luminosity function and luminosity density ({rho}{sub Ly{alpha}}) at 2 < z < 4. Using this and other measurements from the literature at 0.3 < z < 7.7 we trace the redshift evolution of {rho}{sub Ly{alpha}.} We compare it to the expectations from the star formation history of the universe and characterize the evolution of the Ly{alpha} escape fraction of galaxies. LAEs at 2 < z < 4 selected down to a luminosity limit of L(Ly{alpha}) > (3-6) x 10{sup 42} erg s{sup -1} (0.25-0.5 L*), have a mean (E(B - V)) = 0.13 {+-} 0.01, implying an attenuation of {approx}70% in the UV. They show a median UV uncorrected SFR = 11 M{sub sun} yr{sup -1}, dust-corrected SFR = 34 M{sub sun} yr{sup -1}, and Ly{alpha} equivalent widths (EWs) which are consistent with normal stellar populations. We measure a median Ly{alpha} escape fraction of 29%, with a large scatter and values ranging from a few percent to 100%. The Ly{alpha} escape fraction in LAEs correlates with E(B - V) in a way that is expected if Ly{alpha} photons suffer from similar amounts of dust extinction as UV continuum photons. This result implies that a strong enhancement of the Ly{alpha} EW with dust, due to a clumpy multi-phase interstellar medium (ISM), is not a common process in LAEs at these redshifts. It also suggests that while in other galaxies Ly{alpha} can be preferentially quenched by dust due to its scattering nature, this is not the case in LAEs. We find no evolution in the average dust content and Ly{alpha} escape fraction of LAEs from z {approx} 4 to 2. We see hints of a drop in the number density of LAEs from z {approx} 4 to 2 in the redshift distribution and the Ly{alpha} luminosity function, although larger samples are required to confirm this. The mean Ly{alpha} escape fraction of the overall galaxy population decreases significantly from z {approx} 6 to z {approx} 2, in agreement with recent results. Our results point toward a scenario in which star-forming galaxies build up significant amounts of dust in their ISM between z {approx} 6 and 2, reducing their Ly{alpha} escape fraction, with LAE selection preferentially detecting galaxies which have the highest escape fractions given their dust content. The fact that a large escape of Ly{alpha} photons is reached by z {approx} 6 implies that better constraints on this quantity at higher redshifts might detect re-ionization in a way that is uncoupled from the effects of dust.

79 ASTRONOMY AND ASTROPHYSICS↗