Novel calibrations of virial black hole mass estimators in active galaxies based on X-ray luminosity and optical/near-infrared emission lines

Context. It is currently only possible to accurately weigh, through reverberation mapping (RM), the masses of super massive black holes (BHs) in active galactic nuclei (AGN) for a small group of local and bright broad line AGN. Statistical demographic studies can be carried out considering the empirical scaling relation between the size of the broad line region (BLR) and the AGN optical continuum luminosity. There are still biases, however, against low-luminosity or reddened AGN, in which the rest-frame optical radiation can be severely absorbed or diluted by the host galaxy and the BLR emission lines can be hard to detect. Aims. Our purpose is to widen the applicability of virial-based single-epoch (SE) relations to measure reliably the BH masses for low-luminosity or intermediate and type 2 AGN, which the current methodology misses. We achieve this goal by calibrating virial relations based on unbiased quantities: the hard X-ray luminosities in the 2-10 keV and 14-195 keV bands that are less sensitive to galaxy contamination, and the full width at half maximum (FWHM) of the most important rest-frame near-infrared (NIR) and optical BLR emission lines. Methods. We built a sample of RM AGN with both X-ray luminosity, broad optical and NIR FWHM measurements available to calibrate new virial BH mass estimators. Results. We found that the FWHM of the Hα, Hβ, and NIR lines (i.e. Paα, Paβ, and He iλ10830) all correlate with each other with negligible or small offsets. This result allowed us to derive virial BH mass estimators based on either the 2-10 keV or 14-195 keV luminosity. We also took into account the recent determination of the different virial coefficients, f, for pseudo- and classical bulges. By splitting the sample according to the bulge type and adopting separate f factors, we found that our virial relations predict BH masses of AGN hosted in pseudo-bulges ∼0.5 dex smaller than in classical bulges. Assuming the same average f factor for both populations, a difference of ∼0.2 dex is still found. © ESO, 2017