Virial Black Hole Mass Estimates for 280,000 AGNs from the SDSS Broad-Band Photometry and Single Epoch Spectra [GA]

We use the Sloan Digital Sky Survey (SDSS) Quasar Data Release 12 (DR12Q), containing nearly 300,000 AGNs, to calculate the monochromatic luminosities at 5100\AA, 3000\AA, and 1350\AA, derived from the broad-band extinction-corrected SDSS magnitudes. After matching them to their counterparts based on spectra and published in the SDSS Quasar Data Release 7 (DR7Q), we find perfect correlations with minute mean offsets ($\sim$0.01 dex) and dispersions of differences of 0.11, 0.10, 0.12 dex, respectively, across a 2.5 dex luminosity range. We then estimate the active galactic nuclei (AGNs) black hole masses using the broad line region radius–luminosity relations and the FWHM of the MgII and CIV emission lines, to provide a catalog of 283,032 virial black hole mass estimates (132,451 for MgII, 213,068 for CIV, and 62,487 for both) along with the bolometric luminosity and the Eddington ratio estimates for $0.1<z<5.5$ and for roughly a quarter of the sky covered by SDSS. The black hole mass estimates from MgII turned out to be closely matched to the ones from DR7Q with the dispersion of differences of 0.34 dex across a $\sim$2 dex BH mass range. We uncovered a bias in the derived CIV FWHMs from DR12Q as compared to DR7Q, that we correct empirically. The CIV BH mass estimates should be used with caution because the CIV line is known to cause problems in the BH mass estimation from single-epoch spectra. Finally, after the FWHM correction, the AGN black hole mass estimates from CIV closely match the DR7Q ones (with the dispersion of 0.28 dex), and more importantly the MgII and CIV BH masses agree internally with the mean offset of 0.07 dex and the dispersion of 0.39 dex.

Read this paper on arXiv…

S. Kozlowski
Fri, 30 Sep 16

Comments: 5 pages, 4 figures, one table; the data file can be downloaded from this ftp URL