CDFS-6664: A Candidate of Lyman-continuum Emission at z ∼ 3.8 Detected by the Hubble Deep UV Legacy Survey

We report the detection of Lyman continuum (LyC) emission from the galaxy, CDFS-6664, at z = 3.797 in a sample of Lyman break galaxies with detected [O iii] emission lines. The LyC emission is detected with a significance ∼5σ in the F336W band of the Hubble Deep UV Legacy Survey, corresponding to the 650–770 Å rest frame. The light centroid of the LyC emission is offset from the galaxy center by about 0.″2 (1.4 pkpc). The Hubble deep images at longer wavelengths show that the emission is unlikely provided by low-redshift interlopers. The photometric and spectroscopic data show that the possible contribution of an active galactic nucleus is quite low. Fitting the spectral energy distribution of this source to stellar population synthesis models, we find that the galaxy is young (∼50 Myr) and actively forming stars with a rate of 52.1 ± 4.9 M ⊙ yr−1. The significant star formation and the spatially offset LyC emission support a scenario where the ionizing photons escape from the low-density cavities in the ISM excavated by massive young stars. From the nebular model, we estimate the escape fraction of LyC photons to be 38% ± 7% and the corresponding intergalactic medium (IGM) transmission to be 60%, which deviates more than 3σ from the average transmission. The unusually high IGM transmission of LyC photons in CDFS-6664 may be related to a foreground type-2 quasar, CDF-202, at z = 3.7, with a projected separation of 1.′2 only. The quasar may have photoevaporated optically thick absorbers and enhance the transmission on the sightline of CDFS-6664.

Massive black holes in high-redshift Lyman Break Galaxies

ABSTRACT Several evidences indicate that Lyman Break Galaxies (LBGs) in the Epoch of Reionization (redshift z > 6) might host massive black holes (MBHs). We address this question by using a merger-tree model combined with tight constraints from the 7 Ms Chandra survey and the known high-z super-MBH population. We find that a typical LBG with MUV = −22 residing in an Mh ≈ 1012 M⊙ halo at z = 6 host an MBH with mass M• ≈ 2 × 108 M⊙. Depending on the fraction, fseed, of early haloes planted with a direct collapse black hole seed (Mseed = 105M⊙), the model suggests two possible scenarios: (i) if fseed = 1, MBHs in LBGs mostly grow by merging and must accrete at a low (λE ≃ 10−3) Eddington ratio not to exceed the experimental X-ray luminosity upper bound $L_\mathrm{ X}^* = 10^{42.5} {\rm erg\, s}^{-1}$; (ii) if fseed = 0.05, accretion dominates (λE ≃ 0.22) and MBH emission in LBGs must be heavily obscured. In both scenarios the UV luminosity function is largely dominated by stellar emission up to very bright mag, $M_{\rm UV} \lower.5ex\hbox{$\,\, \buildrel\gt \over \sim \,\,$}-23$, with BH emission playing a subdominant role. Scenario (i) poses extremely challenging, and possibly unphysical, requirements on DCBH formation. Scenario (ii) entails testable implications on the physical properties of LBGs involving the FIR luminosity, emission lines, and the presence of outflows.

IGM transmission bias for z ≥ 2.9 Lyman continuum detected galaxies

ABSTRACT Understanding the relationship between the underlying escape fraction of Lyman continuum (LyC) photons (fesc) emitted by galaxies and measuring the distribution of observed fesc values at high redshift is fundamental to the interpretation of the reionization process. In this paper, we perform a statistical exploration of the attenuation of LyC photons by neutral hydrogen in the intergalactic medium using ensembles of simulated transmission functions. We show that LyC-detected galaxies are more likely to be found in sightlines with higher than average transmission of LyC photons. This means that adopting a mean transmission at a given redshift leads to an overestimate of the true fesc for LyC-detected galaxies. We note, however, that mean values are appropriate for fesc estimates of larger parent samples that include LyC non-detected galaxies. We quantify this IGM transmission bias for LyC detections in photometric and spectroscopic surveys in the literature and show that the bias is stronger for both shallower observations and for fainter parent samples (i.e. Ly α emitters versus Lyman break galaxies). We also explore the effects of varying the underlying probability distribution function (PDF) of fesc on recovered values, showing that the underlying fesc PDF may depend on sample selection by comparing with observational surveys. This work represents a first step in improved interpretation of LyC detections in the context of understanding fesc from high-redshift galaxies.

MUSE observations towards the lensing cluster A2744: Intersection between the LBG and LAE populations at z ∼ 3–7

We present a study of the intersection between the populations of star forming galaxies selected as either Lyman break galaxies (LBGs) or Lyman-alpha emitters (LAEs) in the redshift range 2.9 − 6.7 and within the same volume of universe sampled by the Multi-Unit Spectroscopic Explorer (MUSE) behind the Hubble Frontier Fields lensing cluster A2744. We define three samples of star-forming galaxies: LBG galaxies with an LAE counterpart (92 galaxies), LBG galaxies without an LAE counterpart (408 galaxies), and LAE galaxies without an LBG counterpart (46 galaxies). All these galaxies are intrinsically faint because of the lensing nature of the sample (M1500 ≥ −20.5). The fraction of LAEs among all selected star-forming galaxies increases with redshift up to z ∼ 6 and decreases for higher redshifts, in agreement with previous findings. The evolution of LAE/LBG populations with UV magnitude and Lyα luminosity shows that the LAE selection is able to identify intrinsically UV faint galaxies with M1500 ≥ −15 that are typically missed in the deepest lensing photometric surveys. The LBG population seems to fairly represent the total population of star-forming galaxies down to M1500 ∼ −15. Galaxies with M1500 < −17 tend to have SFRLyα < SFRuv, whereas the opposite trend is observed within our sample for faint galaxies with M1500 > −17, including galaxies only detected by their Lyα emission, with a large scatter. These trends, previously observed in other samples of star-forming galaxies at high-z, are seen here for very faint M1500 ∼ −15 galaxies; that is, much fainter than in previous studies. The present results show no clear evidence for an intrinsic difference between the properties of the two populations selected as LBG and/or LAE. The observed trends could be explained by a combination of several phenomena, like the existence of different star-formation regimes, the dust content, the relative distribution and morphology of dust and stars, or the stellar populations.

Testing an indirect method for identifying galaxies with high levels of Lyman continuum leakage

ABSTRACT Using a sample of galaxies at z ≈ 3 with detected Lyman Continuum (LyC) leakage in the SSA22 field, we attempt to verify a proposed indirect method for identifying cases with high LyC escape fraction fesc based on measurements of the H β equivalent width (EW) and the β slope of the UV continuum. To this end, we present Keck/MOSFIRE H β emission line flux measurements of LyC galaxies (LCGs) at spectroscopic redshifts zspec ∼ 3.3, Lyman break galaxies (LBGs) at photometric redshifts zphot = 2.7–3.7, and Ly α emitters at zphot = 3.1. We also reconfirm the spectroscopic redshifts and measure the H β emission-line fluxes from two LCGs and six LBGs. For the LCG in our sample with the most extreme fesc, as revealed by the direct detection of LyC photons, we find that the EW(H β)–β method gives a broadly consistent estimate for fesc, although the error bars remain very large. We also discuss how a combination of fesc measurements based on direct and indirect methods can shed light on the LyC escape mechanism and the anisotropy of the leakage.

MUSEQuBES: calibrating the redshifts of Ly α emitters using stacked circumgalactic medium absorption profiles

ABSTRACT Ly α emission lines are typically found to be redshifted with respect to the systemic redshifts of galaxies, likely due to resonant scattering of Ly α photons. Here, we measure the average velocity offset for a sample of 96 z ≈ 3.3 Ly α emitters (LAEs) with a median Ly α flux (luminosity) of ${\approx}10^{-17}~\rm erg~cm^{-2}~s^{-1}$ (${\approx}10^{42}~\rm erg~s^{-1}$) and a median star formation rate (SFR) of ≈1.3 $\rm M_{\odot }~yr^{-1}$ (not corrected for possible dust extinction), detected by the Multi-Unit Spectroscopic Explorer as part of our MUSEQuBES circumgalactic medium (CGM) survey. By postulating that the stacked CGM absorption profiles of these LAEs, probed by eight background quasars, must be centred on the systemic redshift, we measure an average velocity offset, Voffset = 171\pm 8 km s−1, between the Ly α emission peak and the systemic redshift. The observed Voffset is lower by factors of ≈1.4 and ≈2.6 compared to the velocity offsets measured for narrow-band-selected LAEs and Lyman break galaxies, respectively, which probe galaxies with higher masses and SFRs. Consistent with earlier studies based on direct measurements for individual objects, we find that the Voffset is correlated with the full width at half-maximum of the red peak of the Ly α line, and anticorrelated with the rest-frame equivalent width. Moreover, we find that Voffset is correlated with SFR with a sub-linear scaling relation, $V_{\rm offset}\propto \rm SFR^{0.16\pm 0.03}$. Adopting the mass scaling for main-sequence galaxies, such a relation suggests that Voffset scales with the circular velocity of the dark matter haloes hosting the LAEs.

Modelling a bright z = 6 galaxy at the faint end of the AGN luminosity function

ABSTRACT Recent deep surveys have unravelled a population of faint active galactic nuclei (AGNs) in the high-redshift Universe, leading to various discussions on their nature and their role during the Epoch of Reionization. We use cosmological radiation hydrodynamics simulations of a bright galaxy at z ∼ 6 (${M_\star } \gtrsim 10^{10}\, {\rm M}_{\odot }$) hosting an actively growing supermassive black hole to study the properties of these objects. In particular, we study how the black hole and the galaxy coevolve and what is the relative contribution of the AGNs and of the stellar populations to the luminosity budget of the system. We find that the feedback from the AGN has no strong effect on the properties of the galaxy, and does not increase the total ionizing luminosity of the host. The average escape fraction of our galaxy is around $f_{\rm esc} \sim 5{{\ \rm per\ cent}}$. While our galaxy would be selected as an AGN in deep X-ray surveys, most of the ultraviolet (UV) luminosity is originating from stellar populations. This confirms that there is a transition in the galaxy population from star-forming galaxies to quasar hosts, with bright Lyman-break galaxies with MUV around −22 falling in the overlap region. Our results also suggest that faint AGNs do not contribute significantly to reionizing the Universe.