A comparison of stellar and gaseous kinematics in the nuclei of active galaxies

To investigate the relationship between black holes and their host galaxies, many groups have used the width of the [O (III)] lambda 5007 line as a substitute for the stellar velocity dispersion (sigma(*)) of galaxy bulges. We directly test this assumption with a large and homogeneous sample of narrow-line active galactic nuclei from the Sloan Digital Sky Survey. We consider multiple transitions ([O II] lambda 3727, [O III] lambda 5007, and [S II] lambda lambda 6716, 6731) and various techniques for quantifying the line width in order to obtain a calibration between the gas velocity dispersion, sigma(g), and sigma(*). We find that sigma(g) of the low-ionization lines traces sigma(*), as does sigma(g) for the core of [O iii] after its asymmetric blue wing is properly removed, although in all cases the correlation between sigma(g) and sigma(*) has considerable scatter. While the gas kinematics of the narrow- line region of active galaxies are primarily governed by the gravitational potential of the stars, the accretion rate, as traced by the Eddington luminosity ratio, seems to play an important secondary role. Departures from virial motions correlate systematically with accretion rate. We discuss the implications of these results for previous studies that use [O iii] line widths to infer stellar velocity dispersions in quasars and narrow- line Seyfert 1 galaxies.