Search for: All records

We present 65 extremely dust-obscured galaxies from the UltraVISTA DR3 survey of the COSMOS field at 1 < z < 4. In contrast to other studies of dusty galaxies, we select our sample based on dust attenuation measured by UV–MIR spectral energy distribution (SED) modelling that allows for extreme attenuation levels. We construct our sample by making cuts at 1 ≤ z ≤ 4, AV ≥ 3, and log(M*/M⊙) ≥ 10.5. This method reliably selects galaxies exhibiting independent indicators of significant dust content, including far-infrared detection rates. We perform panchromatic SED modelling with matched Herschel photometry and find stellar and dust properties that differ from typical submillimetre galaxy (SMG) samples as well as Herschel sources matched in redshift and stellar mass. Our sources have lower star formation rates and higher AV than SMGs, but comparable total infrared luminosities. Most of our sample falls on or near the star-forming main sequence for this redshift range. Finally, we perform a morphological analysis with galfit using the KS-band images and Hubble F814W and F160W imaging when available. Typical axial ratios of ∼0.4 suggest disc-like morphology for the majority of our sources, and we note only three apparent merging systems. Our sample generally agrees with the size–mass relation for star-forming galaxies, with a tail extending to smaller sizes. We conclude that the most heavily obscured galaxies in this redshift range share many characteristics with typical star-forming galaxies, forming a population of dusty galaxies that overlaps, but is not encompassed by, those selected through dust emission.

Broad-line regions (BLRs) in high-redshift quasars provide crucial information on chemical enrichment in the early universe. Here we present a study of BLR metallicities in 33 quasars at redshift 5.7 <z< 6.4. Using the near-IR spectra of the quasars obtained from the Gemini telescope, we measure their rest-frame UV emission-line flux and calculate flux ratios. We then estimate BLR metallicities with empirical calibrations based on photoionization models. The inferred median metallicity of our sample is a few times the solar value, indicating that the BLR gas had been highly metal enriched atz∼ 6. We compare our sample with a low-redshift quasar sample with similar luminosities and find no evidence of redshift evolution in quasar BLR metallicities. This is consistent with previous studies. The Feii/Mgiiflux ratio, a proxy for the Fe/αelement abundance ratio, shows no redshift evolution as well, further supporting rapid nuclear star formation atz∼ 6. We also find that the black hole mass–BLR metallicity relation atz∼ 6 is consistent with the relation measured at 2 <z< 5, suggesting that our results are not biased by a selection effect due to this relation.