Extending the Calibration of CIV-Based Single-Epoch Black Hole Mass Estimators for Active Galactic Nuclei

Abstract

We provide an updated calibration of CIV $\lambda1549$ broad emission line-based single-epoch (SE) black hole (BH) mass estimators for active galactic nuclei (AGNs) using new data for six reverberation-mapped AGNs at redshift $z=0.005-0.028$ with BH masses (bolometric luminosities) in the range $10^{6.5}-10^{7.5}$ $M_{\odot}$ ($10^{41.7}-10^{43.8}$ erg s$^{\rm -1}$). New rest-frame UV-to-optical spectra covering 1150-5700 \AA\ for the six AGNs were obtained with the Hubble Space Telescope (HST). Multi-component spectral decompositions of the HST spectra were used to measure SE emission-line widths for the CIV, MgII, and H$\beta$ lines as well as continuum luminosities in the spectral region around each line. We combine the new data with similar measurements for a previous archival sample of 25 AGNs to derive the most consistent and accurate calibrations of the CIV-based SE BH mass estimators against the H$\beta$ reverberation-based masses, using three different measures of broad-line width: full-width at half maximum (FWHM), line dispersion ($\sigma_{\rm line}$) and mean absolute deviation (MAD). The newly expanded sample at redshift $z=0.005-0.234$ covers a dynamic range in BH mass (bolometric luminosity) of $\log\ M_{\rm BH}/M_{\odot} = 6.5-9.1$ ($\log\ L_{\rm bol}/$erg s$^{\rm -1}=41.7-46.9$), and we derive the new CIV-based mass estimators using a Bayesian linear regression analysis over this range. We generally recommend the use of $\sigma_{\rm line}$ or MAD rather than FWHM to obtain a less biased velocity measurement of the CIV emission line, because its narrow-line component contribution is difficult to decompose from the broad-line profile.