We present an analysis of $sim$1500 H160-selected photometric galaxies detected to a limiting magnitude of 27.8 in the HUDF, using imaging from the HST WFC3/IR camera in combination with archival UV, optical, and NIR imaging. We fit photometric redsh
ifts and stellar population estimates for all galaxies with well-determined Spitzer IRAC fluxes, allowing for the determination of the cumulative mass function within the range $1<z<6$. By selecting samples of galaxies at a constant cumulative number density, we explore the co-evolution of stellar masses and star formation rates from z$sim$6. We find a steady increase in the SFRs of galaxies at constant number density from z$sim$6 to z$sim$3. The peak epoch of star formation is found to shift to later times for galaxies with increasing number densities, in agreement with the expectations from cosmic downsizing. The observed SFRs can fully account for the mass growth to z$sim$2 amongst galaxies with cumulative number densities greater than 10$^{-3.5}$ Mpc$^{-3}$. For galaxies with a lower constant number density we find the observed stellar masses are $sim$3 times greater than that which may be accounted for by the observed star formation alone at late times, implying that growth from mergers plays an important role at $z<2$. We additionally observe a decreasing sSFR, equivalent to approximately one order of magnitude, from z$sim$6 to z$sim$2 amongst galaxies with number densities less than 10$^{-3.5}$ Mpc$^{-3}$ along with significant evidence that at any redshift the sSFR is higher for galaxies at higher number density. The combination of these findings can qualitatively explain the previous findings of a sSFR plateau at high redshift. Tracing the evolution of the fraction of quiescent galaxies for samples matched in cumulative number density over this redshift range, we find no unambiguous examples of quiescent galaxies at $z>4$.
3D-HST is a 248-orbit Treasury program to provide WFC3 and ACS grism spectroscopy over four extra-galactic fields (AEGIS, COSMOS, GOODS-South, and UDS), augmented with previously obtained data in GOODS-North. We present a new data release of the 3D-H
ST survey, version v3.0. This release follows the initial v0.5 release that accompanied the survey description paper (Brammer et al. 2012). The new v3.0 release includes the deepest near-IR HST grism spectra currently in existence, extracted from the 8-17 orbit depth observations in the Hubble Ultra Deep Field. Contamination-corrected 2D and 1D spectra, as well as derived redshifts, are made available for >250 objects in this 2x2field. The spectra are of extraordinary quality, and show emission features in many galaxies as faint as F140W=26-27, absorption features in quiescent galaxies at z~2, and several active galactic nuclei. In addition to these extremely deep grism data we provide reduced WFC3 F125W, F140W, and F160W image mosaics of all five 3D-HST/CANDELS fields.
The assembly of galaxies can be described by the distribution of their star formation as a function of cosmic time. Thanks to the WFC3 grism on HST it is now possible to measure this beyond the local Universe. Here we present the spatial distribution
of Halpha emission for a sample of 54 strongly star-forming galaxies at z~1 in the 3D-HST Treasury survey. By stacking the Halpha emission we find that star formation occurred in approximately exponential distributions at z~1, with median Sersic index of n=1.0+-0.2. The stacks are elongated with median axis ratios of b/a=0.58+-0.09 in Halpha, consistent with (possibly thick) disks at random orientation angles. Keck spectra obtained for a subset of eight of the galaxies show clear evidence for rotation, with inclination-corrected velocities of 90 to 330 km/s. The most straightforward interpretation of our results is that star formation in strongly star-forming galaxies at z~1 generally occurred in disks. The disks appear to be scaled-up
We present deep Keck/LRIS spectroscopy and HST/WFC3 imaging in the rest-frame optical for a sample of eight galaxies at z~1.5 with high photometrically-determined stellar masses. The data are combined with VLT/XShooter spectra of five galaxies from v
an de Sande et al. (2011, 2012 to be submitted). We find that these thirteen galaxies have high velocity dispersions, with a median of sigma=301 km s^{-1}. This high value is consistent with their relatively high stellar masses and compact sizes. We study their stellar populations using the strength of Balmer absorption lines, which are not sensitive to dust absorption. We find a large range in Balmer absorption strength, with many galaxies showing very strong lines indicating young ages. The median Hdelta_A equivalent width, determined directly or inferred from the H10 line, is 5.4 Angstroms, indicating a luminosity-weighted age of ~1 Gyr. Although this value may be biased towards higher values because of selection effects,high-dispersion galaxies with such young ages are extremely rare in the local Universe. Interestingly we do not find a simple correlation with rest-frame U-V color: some of the reddest galaxies have very strong Balmer absorption lines. These results demonstrate that many high-dispersion galaxies at z~1.5 were quenched recently. This implies that there must be a population of star-forming progenitors at z~2 with high velocity dispersions or linewidths, which are notoriously absent from CO/Halpha selected surveys.
We measure stellar masses and structural parameters for 5,500 quiescent and 20,000 star-forming galaxies at 0.3<zleq1.5 in the Newfirm Medium Band Survey COSMOS and UKIDSS UDS fields. We combine these measurements to infer velocity dispersions and de
termine how the number density of galaxies at fixed inferred dispersion, or the Velocity Dispersion Function (VDF), evolves with time for each population. We show that the number of galaxies with high velocity dispersions appears to be surprisingly stable with time, regardless of their star formation history. Furthermore, the overall VDF for star-forming galaxies is constant with redshift, extending down to the lowest velocity dispersions probed by this study. The only galaxy population showing strong evolution are quiescent galaxies with low inferred dispersions, whose number density increases by a factor of ~4 since z=1.5. This build-up leads to an evolution in the quiescent fraction of galaxies such that the threshold dispersion above which quiescent galaxies dominate the counts moves to lower velocity dispersion with time. We show that our results are qualitatively consistent with a simple model in which star-forming galaxies quench and are added to the quiescent population. In order to compensate for the migration into the quiescent population, the velocity dispersions of star-forming galaxies must increase, with a rate that increases with dispersion.
We present evidence of large-scale outflows from three low-mass (log(M/M_sun)~9.75) star-forming (SFR >4 M_sun/yr) galaxies observed at z=1.24, z=1.35 and z=1.75 in the 3D-HST Survey. Each of these galaxies is located within a projected physical dist
ance of 60 kpc around the sight line to the quasar SDSS J123622.93+621526.6, which exhibits well-separated strong (W_r>0.8A) Mg II absorption systems matching precisely to the redshifts of the three galaxies. We derive the star formation surface densities from the H-alpha emission in the WFC3 G141 grism observations for the galaxies and find that in each case the star formation surface density well-exceeds 0.1 M_sun/yr/kpc^2, the typical threshold for starburst galaxies in the local Universe. From a small but complete parallel census of the 0.65<z<2.6 galaxies with H_140<24 proximate to the quasar sight line, we detect Mg II absorption associated with galaxies extending to physical distances of 130 kpc. We determine that the W_r>0.8A Mg II covering fraction of star-forming galaxies at 1<z<2 may be as large as unity on scales extending to at least 60 kpc, providing early constraints on the typical extent of starburst-driven winds around galaxies at this redshift. Our observations additionally suggest that the azimuthal distribution of W_r>0.4A Mg II absorbing gas around star-forming galaxies may evolve from z~2 to the present, consistent with recent observations of an increasing collimation of star formation-driven outflows with time from z~3.
Recent studies have identified a population of compact quiescent galaxies at zsim2. These galaxies are very rare today and establishing the existence of a nearby analog could allow us to study its structure in greater detail than is possible at high
redshift. Here we present such a local analog, NGC 5845, which has a dynamical mass of M_dyn = 4.3pm0.6times10^10 M_sun and an effective radius of only r_e = 0.45pm0.05kpc. We study the structure and kinematics with HST/WFPC2 data and previously published spatially resolved kinematics. We find that NGC 5845 is similar to compact quiescent galaxies at zsim2 in terms of size versus dynamical mass (r_e-M_dyn), effective velocity dispersion versus size (sigma_e-r_e), and effective velocity dispersion versus dynamical mass (sigma_e-M_dyn). The galaxy has a prominent rotating disk evident in both the photometry and the kinematics: it extends to well beyond geq1/3 effective radius and contribute to geq1/4 of the total light of the galaxy. Our results lend support to the idea that a fraction of zsim2 compact galaxies have prominent disks and positive mass-to-light ratio gradients, although we caution that NGC 5845 may have had a different formation history than the more massive compact quiescent galaxies at zsim2.
We study the properties of massive galaxies at an average redshift of z~0.34 through stacking more than 42000 images of Luminous Red Galaxies from the Sloan Digital Sky Survey. This is the largest dataset ever used for such an analysis and it allows
us to explore the outskirts of massive red galaxies at unprecedented physical scales. Our image stacks extend farther than 400 kpc, where the r-band profile surface brightness reaches 30 mag arcsec-2. This analysis confirms that the stellar bodies of luminous red galaxies follow a simple Sersic profile out to 100 kpc. At larger radii the profiles deviate from the best-fit Sersic models and exhibit extra light in the g, r, i and z-band stacks. This excess light can probably be attributed to unresolved intragroup or intracluster light or a change in the light profile itself. We further show that standard analyses of SDSS-depth images typically miss 20% of the total stellar light and underestimate the size of LRGs by 10% compared to our best fit r-band Sersic model of n=5.5 and r_e=13.1 kpc. If the excess light at r>100 kpc is considered to be part of the galaxy, the best fit r-band Sersic parameters are n=5.8 and r_e=13.6 kpc. In addition we study the radially dependent stack ellipticity and find an increase with radius from e=0.25 at r=10 kpc to e=0.3 at r=100 kpc. This provides support that the stellar light that we trace out to at least 100 kpc is physically associated with the galaxies themselves and may confirm that the halos of individual LRGs have higher ellipticities than their central parts. Lastly we show that the broadband color gradients of the stacked images are flat beyond roughly 40 kpc, suggesting that the stellar populations do not vary significantly with radius in the outer parts of massive ellipticals.
Using a sample of nine massive compact galaxies at z ~ 2.3 with rest-frame optical spectroscopy and comprehensive U through 8um photometry we investigate how assumptions in SED modeling change the stellar mass estimates of these galaxies, and how thi
s affects our interpretation of their size evolution. The SEDs are fit to Tau-models with a range of metallicities, dust laws, as well as different stellar population synthesis codes. These models indicate masses equal to, or slightly smaller than our default masses. The maximum difference is 0.16 dex for each parameter considered, and only 0.18 dex for the most extreme combination of parameters. Two-component populations with a maximally old stellar population superposed with a young component provide reasonable fits to these SEDs using the models of Bruzual & Charlot (2003); however, using models with updated treatment of TP-AGB stars the fits are poorer. The two-component models predict masses that are 0.08 to 0.22 dex larger than the Tau-models. We also test the effect of a bottom-light IMF and find that it would reduce the masses of these galaxies by 0.3 dex. Considering the range of allowable masses from the Tau-models, two-component fits, and IMF, we conclude that on average these galaxies lie below the mass-size relation of galaxies in the local universe by a factor of 3-9, depending on the SED models used.
Using deep near-infrared spectroscopy Kriek et al. (2006) found that ~45% of massive galaxies at z~2.3 have evolved stellar populations and little or no ongoing star formation. Here we determine the sizes of these quiescent galaxies using deep, high-
resolution images obtained with HST/NIC2 and laser guide star-assisted Keck/AO. Considering that their median stellar mass is 1.7x10^11 Solar masses the galaxies are remarkably small, with a median effective radius of 0.9 kpc. Galaxies of similar mass in the nearby Universe have sizes of ~5 kpc and average stellar densities which are two orders of magnitude lower than the z~2.3 galaxies. These results extend earlier work at z~1.5 and confirm previous studies at z>2 which lacked spectroscopic redshifts and imaging of sufficient resolution to resolve the galaxies. Our findings demonstrate that fully assembled early-type galaxies make up at most ~10% of the population of K-selected quiescent galaxies at z~2.3, effectively ruling out simple monolithic models for their formation. The galaxies must evolve significantly after z~2.3, through dry mergers or other processes, consistent with predictions from hierarchical models.
