We present initial results of a deep near-IR spectroscopic survey covering the 15 fields of the Keck Baryonic Structure Survey (KBSS) using MOSFIRE on the Keck 1 telescope, focusing on a sample of 251 galaxies with redshifts 2.0 < z < 2.6, star-formation rates 2 < ∼ SFR < ∼ 200 M yr −1 , and stellar masses 8.6 < log(M * /M ) < 11.4, with high-quality spectra in both H-and K-band atmospheric windows. We show unambiguously that the locus of z ∼ 2.3 galaxies in the "BPT" nebular diagnostic diagram exhibits a disjoint, yet similarly tight, relationship between the ratios [NII]λ6585/Hα and [OIII]/Hβ as compared to local galaxies. Using photoionization models, we argue that the offset of the z ∼ 2.3 locus relative to z ∼ 0 is explained by a combination of harder ionizing radiation field, higher ionization parameter, and higher N/O at a given O/H than applies to most local galaxies, and that the position of a galaxy along the z ∼ 2.3 star-forming BPT locus is surprisingly insensitive to gas-phase oxygen abundance. The observed nebular emission line ratios are most easily reproduced by models in which the net ionizing radiation field resembles a blackbody with effective temperature T eff = 50000 − 60000 K and N/O close to the solar value at all O/H. We critically assess the applicability of commonly-used strong line indices for estimating gas-phase metallicities, and consider the implications of the small intrinsic scatter in the empirical relationship between excitation-sensitive line indices and M * (i.e., the "mass-metallicity" relation), at z 2.3.

We study the Lyα profiles of 36 spectroscopically detected Lyα-emitters (LAEs) at z ∼ 2-3, using Keck MOSFIRE to measure systemic redshifts and velocity dispersions from rest-frame optical nebular emission lines. The sample has a median optical magnitude R = 26.0, and ranges from R 23 to R > 27, corresponding to rest-frame UV absolute magnitudes M UV −22 to M UV > −18.2. Dynamical masses range from M dyn < 1.3 × 10 8 M to M dyn = 6.8 × 10 9 M , with a median value of M dyn = 6.3 × 10 8 M . Thirty of the 36 Lyα emission lines are redshifted with respect to the systemic velocity with at least 1σ significance, and the velocity offset with respect to systemic Δv Lyα is correlated with the R-band magnitude, M UV , and the velocity dispersion measured from nebular emission lines with >3σ significance: brighter galaxies with larger velocity dispersions tend to have larger values of Δv Lyα . We also make use of a comparison sample of 122 UV-color-selected R < 25.5 galaxies at z ∼ 2, all with Lyα emission and systemic redshifts measured from nebular emission lines. Using the combined LAE and comparison samples for a total of 158 individual galaxies, we find that Δv Lyα is anti-correlated with the Lyα equivalent width with 7σ significance. Our results are consistent with a scenario in which the Lyα profile is determined primarily by the properties of the gas near the systemic redshift; in such a scenario, the opacity to Lyα photons in lower mass galaxies may be reduced if large gaseous disks have not yet developed and if the gas is ionized by the harder spectrum of young, low metallicity stars.

We present new results on the Lyα emission-line kinematics of 18 z ∼ 2-3 star-forming galaxies with multiplepeaked Lyα profiles. With our large spectroscopic database of UV-selected star-forming galaxies at these redshifts, we have determined that ∼30% of such objects with detectable Lyα emission display multiple-peaked emission profiles. These profiles provide additional constraints on the escape of Lyα photons due to the rich velocity structure in the emergent line. Despite recent advances in modeling the escape of Lyα from star-forming galaxies at high redshifts, comparisons between models and data are often missing crucial observational information. Using Keck II NIRSPEC spectra of Hα (z ∼ 2) and [O iii]λ5007 (z ∼ 3), we have measured accurate systemic redshifts, rest-frame optical nebular velocity dispersions, and emission-line fluxes for the objects in the sample. In addition, rest-frame UV luminosities and colors provide estimates of star formation rates and the degree of dust extinction. In concert with the profile sub-structure, these measurements provide critical constraints on the geometry and kinematics of interstellar gas in high-redshift galaxies. Accurate systemic redshifts allow us to translate the multiple-peaked Lyα profiles into velocity space, revealing that the majority (11/18) display double-peaked emission straddling the velocity-field zero point with stronger red-side emission. Interstellar absorption-line kinematics suggest the presence of large-scale outflows for the majority of objects in our sample, with an average measured interstellar absorption velocity offset of Δv abs = −230 km s −1 . A comparison of the interstellar absorption kinematics for objects with multiple-and single-peaked Lyα profiles indicate that the multiple-peaked objects are characterized by significantly narrower absorption line widths. We compare our data with the predictions of simple models for outflowing and infalling gas distributions around high-redshift galaxies. While popular "shell" models provide a qualitative match with many of the observations of Lyα emission, we find that in detail there are important discrepancies between the models and data, as well as problems with applying the framework of an expanding thin shell of gas to explain high-redshift galaxy spectra. Our data highlight these inconsistencies, as well as illuminating critical elements for success in future models of outflow and infall in high-redshift galaxies.

Narrowband imaging of the rest-frame Lyman continuum (LyC) of galaxies at z 3.1 has produced a large number of candidate LyC-emitting galaxies. These samples are contaminated by galaxies at lower redshift. To better understand LyC escape, we need an uncontaminated sample of galaxies that emit strongly in the LyC. Here we present deep Hubble Space Telescope imaging of five bright galaxies at z 3.1 that had previously been identified as candidate LyC emitters with ground-based images. The WFC3 F336W images probe the LyC of galaxies at > z 3.06 and provide an order-of-magnitude increase in spatial resolution over ground-based imaging. The non-ionizing UV images often show multiple galaxies (or components) within ~ 1 of the candidate LyC emission seen from the ground. In each case, only one of the components is emitting light in the F336W filter, which would indicate LyC escape if that component is at > z 3.06. We use Keck/NIRSPEC near-IR spectroscopy to measure redshifts of these components to distinguish LyC emitters from foreground contamination. We find that two candidates are low-redshift contaminants, one candidate had a previously misidentified redshift, and the other two cannot be confirmed as LyC emitters. The level of contamination is consistent with previous estimates. For the galaxies with > z 3.06, we derive strong s 1 limits on the relative escape fraction between 0.07 and 0.09. We still do not have a sample of definitive LyC emitters, and a much larger study of low-luminosity galaxies is required. The combination of high-resolution imaging and deep spectroscopy is critical for distinguishing LyC emitters from foreground contaminants.