Spectral Classification and Ionized Gas Outflows in z ∼ 2 WISE-selected Hot Dust-obscured Galaxies

Abstract

We present VLT/XSHOOTER rest-frame UV-optical spectra of 10 Hot Dust-Obscured Galaxies (Hot DOGs) at z ∼ 2 to investigate AGN diagnostics and to assess the presence and effect of ionized gas outflows. Most Hot DOGs in this sample are narrow-line dominated AGN (type 1.8 or higher), and have higher Balmer decrements than typical type 2 quasars. Almost all (8/9) sources show evidence for ionized gas outflows in the form of broad and blueshifted [O III] profiles, and some sources have such profiles in Hα (5/7) or [O II] (3/6). Combined with the literature, these results support additional sources of obscuration beyond the simple torus invoked by AGN unification models. Outflow rates derived from the broad [O III] line ( 10 3 M ⊙ yr −1 ) are greater than the black hole accretion and star formation rates, with feedback efficiencies (∼ 0.1 − 1%) consistent with negative feedback to the host galaxy's star formation in merger-driven quasar activity scenarios. We find the broad emission lines in luminous, obscured quasars are often better explained by outflows within the narrow line region, and caution that black hole mass estimates for such sources in the literature may have substantial uncertainty. Regardless, we find lower bounds on the Eddington ratio for Hot DOGs near unity.

Fast Outflows in Hot Dust-obscured Galaxies Detected with Keck/NIRES

“…While W1838+3429 has a previously published redshift, the NIRES spectrum does not clearly detect any emission lines, despite the predicted presence of [O III]λ 5007 Å at 2.1 μm based on the redshift from Tsai et al (2015). Similarly, the NIRES spectrum cannot confirm the published redshift of 2.61 for W2016−0041 from Jun et al (2020)…”

“…All targets except W1835+4355 and W2238+2653 have narrow Balmer decrements substantially above the extinctionfree value of 3.1 (Kim et al 2006), indicative of significant extinction at optical wavelengths. Large optical extinctions in Hot DOGs were also reported in Assef et al (2015) and Jun et al (2020). Applying the Balmer decrement/extinction relation from Domínguez et al (2013) with an intrinsic decrement of 3.1 to the measured Hα/Hβ flux ratios gives a mean A V,narrow of 2.4 mag and median of 3 mag, disregarding W1835+4355.…”

“…), and typical energy injection rates on the order of . W2216 +0723 was also observed in Jun et al (2020), allowing a direct comparison of the analyses. Despite similar outflow velocities, we report a dramatically smaller mass outflow rate.…”

“…Subsequent observations of the Hα emission in five Hot DOGs detected broad emission in all targets, with Eddington ratios close to unity (Wu et al 2018). Full rest-frame optical spectroscopy of 12 targets was first published in Jun et al (2020); those authors found ionized outflows in optical emission lines showing extreme kinematics, with typical [O III]λ5007 Å blueshifts of approximately 1100 km s −1 and full widths at half maximum (FWHMs) of approximately 2600 km s −1 .…”

Hot Dust-obscured Galaxies with Excess Blue Light

“…Indeed, Díaz-Santos et al (2016) presented a study of the [C II] 157.7 μm emission line in the highest luminosity Hot DOG, and possibly the most luminous galaxy known, WISEA J224607.56-052634.9 (W2246-0526; Tsai et al 2015), and determined based on the emission-line kinematics that the central gas of the host galaxy is likely undergoing an isotropic outflow event. Additional ionized gas outflow signatures have been observed in the optical narrow emission lines of some other Hot DOGs (Wu et al 2018;Jun et al 2020), supporting the presence of strong AGN feedback in the interstellar medium (ISM) of these targets. Assef et al (2015, also see Eisenhardt et al 2012Tsai et al 2015Tsai et al , 2018 showed that the UV through mid-IR SED of the majority of Hot DOGs (specifically "W12-drops" with z>1) can be well modeled as a combination of a star-forming galaxy that dominates the optical/UV emission, and a luminous, obscured AGN that dominates the mid-IR SED and the bolometric luminosity of the system.…”