Systematically smaller single-epoch quasar black hole masses using a radius–luminosity relationship corrected for spectral bias

Abstract

ABSTRACT Determining black hole masses and accretion rates with better accuracy and precision is crucial for understanding quasars as a population. These are fundamental physical properties that underpin models of active galactic nuclei. A primary technique to measure the black hole mass employs the reverberation mapping of low-redshift quasars, which is then extended via the radius–luminosity relationship for the broad-line region to estimate masses based on single-epoch spectra. An updated radius–luminosity relationship incorporates the flux ratio of optical Fe ii to H β ($\equiv \mathcal {R}_{\rm Fe}$) to correct for a bias in which more highly accreting systems have smaller line-emitting regions than previously realized. In this work, we demonstrate and quantify the effect of using this Fe-corrected radius-luminosity relationship on mass estimation by employing archival data sets possessing rest-frame optical spectra over a wide range of redshifts. We find that failure to use an Fe-corrected radius predictor results in overestimated single-epoch black hole masses for the most highly accreting quasars. Their accretion rate measures (LBol/LEdd and $\dot{\mathscr{M}}$ ) are similarly underestimated. The strongest Fe-emitting quasars belong to two classes: high-z quasars with rest-frame optical spectra, which, given their extremely high luminosities, require high accretion rates, and their low-z analogues, which, given their low black holes masses, must have high accretion rates to meet survey flux limits. These classes have mass corrections downward of about a factor of two, on average. These results strengthen the association of the dominant Eigenvector 1 parameter $\mathcal {R}_{\rm Fe}$ with the accretion process.