Luis C. Ho (1), Paolo Goldoni (2,3), Xiao-Bo Dong (1,4), jenny E. Greene (5), and Gabriele Ponti (2,6) ((1) The Observatories of the Carnegie Institution for Science, 813 Santa Barbara Street, Pasadena, CA 91101, USA; (2) Laboratoire Astroparticule et Cosmologie, 10 rue A. Domon et L. Duquet, 75205 Paris Cedex 13, France; (3) DSM/IRFU/Service d’Astrophysique, CEA/Saclay, 91191 Gif-sur-Yvette, France; (4) Key laboratory for Research in Galaxies and Cosmology, The University of Sciences and Technology of China, Chinese Academy of Sciences, Hefei, Anhui 230026, China; (5) Department of Astrophysical Sciences, Princeton University, Peyton Hall, Ivy Lane, Princeton, NJ 08544, USA; (6) School of Physics and Astronomy, University of Southampton, Highfield, SO17 1BJ, UK)
The X-shooter instrument on the VLT was used to obtain spectra of seven moderate-redshift quasars simultaneously covering the spectral range 3000 Ang to 2.5 microns. At z ~ 1.5, most of the prominent broad emission lines in the ultraviolet to optical region are captured in their rest frame. We use this unique dataset, which mitigates complications from source variability, to intercompare the line profiles of C IV 1549, C III] 1909, Mg II 2800, and Hα and evaluate their implications for black hole mass estimation. We confirm that Mg II and the Balmer lines share similar kinematics and that they deliver mutually consistent black hole mass estimates with minimal internal scatter (< 0.1 dex) using the latest virial mass estimators. Although no virial mass formalism has yet been calibrated for C III], this line does not appear promising for such an application because of the large spread of its velocity width compared to lines of both higher and lower ionization; part of the discrepancy may be due to the difficulty of deblending C III] from its neighboring lines. The situation for C IV is complex and, because of the limited statistics of our small sample, inconclusive. On the one hand, slightly more than half of our sample (4/7) have C IV line widths that correlate reasonably well with Hα line widths, and their respective black hole mass estimates agree to within ~0.15 dex. The rest, on the other hand, exhibit exceptionally broad C IV profiles that overestimate virial masses by factors of 2-5 compared to Hα. As C IV is widely used to study black hole demographics at high redshifts, we urgently need to revisit our analysis with a larger sample.
Complete preprint ===> http://arxiv.org/abs/1205.3224