Published in The Astrophysical Journal, Volume 747, Issue 1, March 1, 2012, pages 1-18.
The definitive version is available at https://doi.org/10.1088/0004-637X/747/1/30.
We investigate the calibration and uncertainties of black hole (BH) mass estimates based on the single-epoch (SE) method, using homogeneous and high-quality multi-epoch spectra obtained by the Lick Active Galactic Nucleus (AGN) Monitoring Project for nine local Seyfert 1 galaxies with BH masses <108 M ☉. By decomposing the spectra into their AGNs and stellar components, we study the variability of the SE Hβ line width (full width at half-maximum intensity, FWHMHβ or dispersion, σHβ) and of the AGN continuum luminosity at 5100 Å (L 5100). From the distribution of the "virial products" ( FWHMHβ 2 L 0.5 5100 or σHβ 2 L 0.5 5100) measured from SE spectra, we estimate the uncertainty due to the combined variability as ~0.05 dex (12%). This is subdominant with respect to the total uncertainty in SE mass estimates, which is dominated by uncertainties in the size-luminosity relation and virial coefficient, and is estimated to be ~0.46 dex (factor of ~3). By comparing the Hβ line profile of the SE, mean, and root-mean-square (rms) spectra, we find that the Hβ line is broader in the mean (and SE) spectra than in the rms spectra by ~0.1 dex (25%) for our sample with FWHMHβ s-1. This result is at variance with larger mass BHs where the difference is typically found to be much less than 0.1 dex. To correct for this systematic difference of the Hβ line profile, we introduce a line-width dependent virial factor, resulting in a recalibration of SE BH mass estimators for low-mass AGNs.
2012 IOP Publishing.