ARE OUTFLOWS BIASING SINGLE-EPOCH C IV BLACK HOLE MASS ESTIMATES?

Abstract

We use a combination of reverberation mapping data and single-epoch spectra of the CIV emission line in a sample of both low and high-redshift active galactic nuclei (AGNs) to investigate sources of the discrepancies between CIV- and Hbeta-based single-epoch black hole mass estimates. We find that for all reverberation mapped sources, there is a component of the line profile that does not reverberate, and the velocity characteristics of this component vary from object-to-object. The differing strength and properties of this non-variable component are responsible for much of the scatter in CIV-based black hole masses compared to Hbeta masses. The CIV mass bias introduced by this non-variable component is correlated with the shape of the CIV line, allowing us to make an empirical correction to the black hole mass estimates. Using this correction and accounting for other sources of scatter such as poor data quality and data inhomogeneity reduces the scatter between the CIV and Hbeta masses in our sample by a factor of ~2, to only ~0.2 dex. We discuss the possibility that this non-variable CIV component originates in an orientation-dependent outflow from either the proposed broad line region (BLR) disk-wind or the intermediate line region (ILR), a high-velocity inner extension of the narrow line region (NLR).