We present a full data analysis of the pure-parallel Hubble Space Telescope (HST) imaging observations in the Brightest of Reionizing Galaxies Survey (BoRG[z9]) in Cycle 22. The medium-deep exposures with five HST/WFC3IR+UVIS filter bands from 79 independent sightlines (∼ 370 arcmin 2 ) provide the least biased determination of number density for z ∼ > 9 bright galaxies against cosmic variance. After a strict twostep selection for candidate galaxies, including dropout color and photometric redshift analyses, and revision of previous BoRG candidates, we identify one source at z ∼ 10 and two sources at z ∼ 9. The z ∼ 10 candidate shows evidence of line-of-sight lens magnification (µ ∼ 1.5), yet it appears surprisingly luminous (M UV ∼ −22.6 ± 0.3 mag), making it one of the brightest candidates at z > 8 known (∼ 0.3 mag brighter than the z = 8.68 galaxy EGSY8p7, spectroscopically confirmed by Zitrin and collaborators). For z ∼ 9 candidates, we include previous data points at fainter magnitudes and find that the data are well fitted by a Schechter luminosity function with α = −2.1 +0.3 −0.3 , M * UV = −21.0 +0.7 −1.4 mag, and log φ * = −4.2 +0.6 −0.9 Mpc −3 mag −1 , for the first time without fixing any parameters. The inferred cosmic star formation rate density is consistent with unaccelerated evolution from lower redshift.

We use 317,000 emission-line galaxies from the Sloan Digital Sky Survey to investigate line-ratio selection of active galactic nuclei (AGNs). In particular, we demonstrate that "star formation dilution" by Hii regions causes a significant bias against AGN selection in low-mass, blue, star-forming, diskdominated galaxies. This bias is responsible for the observed preference of AGNs among high-mass, green, moderately star-forming, bulge-dominated hosts. We account for the bias and simulate the intrinsic population of emission-line AGNs using a physically-motivated Eddington ratio distribution, intrinsic AGN narrow line region line ratios, a luminosity-dependent L bol /L[O iii] bolometric correction, and the observed M BH − σ relation. These simulations indicate that, in massive (log(M * /M ⊙ ) 10) galaxies, AGN accretion is correlated with specific star formation rate but is otherwise uniform with stellar mass. There is some hint of lower black hole occupation in low-mass (log(M * /M ⊙ ) 10) hosts, although our modeling is limited by uncertainties in measuring and interpreting the velocity dispersions of low-mass galaxies. The presence of star formation dilution means that AGNs contribute little to the observed strong optical emission lines (e.g., [O iii] and Hα) in low-mass and star-forming hosts. However the AGN population recovered by our modeling indicates that feedback by typical (low-to moderate-accretion) low-redshift AGNs has nearly uniform efficiency at all stellar masses, star formation rates, and morphologies. Taken together, our characterization of the observational bias and resultant AGN occupation function suggest that AGNs are unlikely to be the dominant source of star formation quenching in galaxies, but instead are fueled by the same gas which drives star formation activity.

We present the results from a VLT/SINFONI and Keck/NIRSPEC near-infrared spectroscopic survey of 16 Lyman-alpha emitters (LAEs) at z = 2.1 -2.5 in the COSMOS and GOODS-N fields discovered from the HETDEX Pilot Survey. We detect rest-frame optical nebular lines (Hα and/or [O iii]λ5007) for 10 of the LAEs and measure physical properties, including the star formation rate (SFR), gas-phase metallicity, gas-mass fraction, and Lyα velocity offset. We find that LAEs may lie below the mass-metallicity relation for continuum-selected star-forming galaxies at the same redshift. The LAEs all show velocity shifts of Lyα relative to the systemic redshift ranging between +85 and +296 km s −1 with a mean of +180 km s −1 . This value is smaller than measured for continuum-selected star-forming galaxies at similar redshifts. The Lyα velocity offsets show a moderate correlation with the measured star formation rate (2.5σ), but no significant correlations are seen with the SFR surface density, specific SFR, stellar mass, or dynamical mass ( 1.5σ). Exploring the role of dust, kinematics of the interstellar medium (ISM), and geometry on the escape of Lyα photons, we find no signature of selective quenching of resonantly scattered Lyα photons. However, we also find no evidence that a clumpy ISM is enhancing the Lyα equivalent width. Our results suggest that the low metallicity in LAEs may be responsible for yielding an environment with a low neutral hydrogen column density as well as less dust, easing the escape of Lyα photons over that in continuum-selected star-forming galaxies.

We use broadband photometry extending from the rest-frame UV to the near-IR to fit the individual spectral energy distributions (SEDs) of 63 bright (L(Lyα) > 10 43 ergs s −1 ) Lyα emitting galaxies (LAEs) in the redshift range 1.9 < z < 3.6. We find that these LAEs are quite heterogeneous, with stellar masses that span over three orders of magnitude, from 7.5 < log M/M < 10.5. Moreover, although most LAEs have small amounts of extinction, some high-mass objects have stellar reddenings as large as E(B − V ) ∼ 0.4. Interestingly, in dusty objects the optical depths for Lyα and the UV continuum are always similar, indicating that Lyα photons are not undergoing many scatters before escaping their galaxy. In contrast, the ratio of optical depths in low-reddening systems can vary widely, illustrating the diverse nature of the systems. Finally, we show that in the star formation rate (SFR)-log mass diagram, our LAEs fall above the "main-sequence" defined by z ∼ 3 continuum selected star-forming galaxies. In this respect, they are similar to sub-mm-selected galaxies, although most LAEs have much lower mass.

We compare the physical and morphological properties of z ∼ 2 Lyα emitting galaxies (LAEs) identified in the HETDEX Pilot Survey and narrow band studies with those of z ∼ 2 optical emission line galaxies (oELGs) identified via HST WFC3 infrared grism spectroscopy. Both sets of galaxies extend over the same range in stellar mass (7.5 < log M/M < 10.5), size (0.5 < R < 3.0 kpc), and star-formation rate (∼ 1 < SFR < 100 M yr −1 ). Remarkably, a comparison of the most commonly used physical and morphological parameters -stellar mass, half-light radius, UV slope, star formation rate, ellipticity, nearest neighbor distance, star formation surface density, specific star formation rate, [O III] luminosity, and [O III] equivalent width -reveals no statistically significant differences between the populations. This suggests that the processes and conditions which regulate the escape of Lyα from a z ∼ 2 star-forming galaxy do not depend on these quantities. In particular, the lack of dependence on the UV slope suggests that Lyα emission is not being significantly modulated by diffuse dust in the interstellar medium. We develop a simple model of Lyα emission that connects LAEs to all high-redshift star forming galaxies where the escape of Lyα depends on the sightline through the galaxy. Using this model, we find that mean solid angle for Lyα escape is Ω Lyα = 2.4 ± 0.8 steradians; this value is consistent with those calculated from other studies.

We use deep Hubble Space Telescope spectroscopy to constrain the metallicities and (light-weighted) ages of massive (log M * /M 10) galaxies selected to have quiescent stellar populations at 1.0 < z < 1.8. The data include 12-orbit depth coverage with the WFC3/G102 grism covering ∼ 8, 000 < λ < 11, 500 Å at a spectral resolution of R ∼ 210 taken as part of the CANDELS Lyman-α Emission at Reionization (CLEAR) survey. At 1.0 < z < 1.8, the spectra cover important stellar population features in the rest-frame optical. We simulate a suite of stellar population models at the grism resolution, fit these to the data for each galaxy, and derive posterior likelihood distributions for metallicity and age. We stack the posteriors for subgroups of galaxies in different redshift ranges that include different combinations of stellar absorption features. Our results give light-weighted ages of t z∼1.1 = 3.2 ± 0.7 Gyr, t z∼1.2 = 2.2 ± 0.6 Gyr, t z∼1.3 = 3.1 ± 0.6 Gyr, and t z∼1.6 = 2.0 ± 0.6 Gyr, for galaxies at z ∼ 1.1, 1.2, 1.3, and 1.6. This implies that most of the massive quiescent galaxies at 1 < z < 1.8 had formed > 68% of their stellar mass by a redshift of z > 2. The posteriors give metallicities of Z z∼1.1 = 1.16 ± 0.29 Z , Z z∼1.2 = 1.05 ± 0.34 Z , Z z∼1.3 = 1.00 ± 0.31 Z , and Z z∼1.6 = 0.95 ± 0.39 Z . This is evidence that massive galaxies had enriched rapidly to approximately Solar metallicities as early as z ∼ 3.

We assemble an unbiased sample of 29 galaxies with [O II] λ3727 and/or [O III] λ5007 detections at z < 0.15 from the Hobby-Eberly Telescope Dark Energy Experiment (HETDEX) Pilot Survey (HPS). HPS finds galaxies without pre-selection based on their detected emission lines via integral field spectroscopy. Sixteen of these objects were followed up with the second-generation, low resolution spectrograph (LRS2) on the upgraded Hobby-Eberly Telescope. Oxygen abundances were then derived via strong emission lines using a Bayesian approach. We find most of the galaxies fall along the massmetallicity relation derived from photometrically selected star forming galaxies in the Sloan Digital Sky Survey (SDSS). However, two of these galaxies have low metallicity (similar to the very rare green pea galaxies in mass-metallicity space). The star formation rates of this sample fall in an intermediate space between the SDSS star forming main sequence and the extreme green pea galaxies. We conclude that spectroscopic selection fills part of the mass-metallicity-SFR phase space that is missed in photometric surveys with pre-selection like SDSS, i.e., we find galaxies that are actively forming stars but are faint in

We present a Bayesian approach to the redshift classification of emission-line galaxies when only a single emission line is detected spectroscopically. We consider the case of surveys for high-redshift Lyα-emitting galaxies (LAEs), which have traditionally been classified via an inferred rest-frame equivalent width (W Lyα ) greater than 20 Å. Our Bayesian method relies on known prior probabilities in measured emission-line luminosity functions and equivalent width distributions for the galaxy populations, and returns the probability that an object in question is an LAE given the characteristics observed. This approach will be directly relevant for the Hobby-Eberly Telescope Dark Energy Experiment (HETDEX), which seeks to classify ∼ 10 6 emission-line galaxies into LAEs and low-redshift [O ii] emitters. For a simulated HETDEX catalog with realistic measurement noise, our Bayesian method recovers 86 % of LAEs missed by the traditional W Lyα > 20 Å cutoff over 2 < z < 3, outperforming the equivalent width (EW) cut in both contamination and incompleteness. This is due to the method's ability to trade off between the two types of binary classification error by adjusting the stringency of the probability requirement for classifying an observed object as an LAE. In our simulations of HETDEX, this method reduces the uncertainty in cosmological distance measurements by 14 % with respect to the EW cut, equivalent to recovering 29 % more cosmological information. Rather than using binary object labels, this method enables the use of classification probabilities in large-scale structure analyses. It can be applied to narrowband emission-line surveys as well as upcoming large spectroscopic surveys including Euclid and WFIRST.