I review how AGN black hole masses are calculated from emission-line reverberation-mapping data, with particular attention to both assumptions and caveats. I discuss the empirical relationship between AGN luminosity and broad-line region radius that underpins the indirect methods by which most AGN masses are estimated. I also discuss how line widths are characterized in this method and illustrate how different ways of measuring the line-widths can lead to systematic errors in the mass scale. I discuss specific implications for NLS1 galaxies and consider whether the NLS1 phenomenon is better explained by source inclination or by Eddington rate, and conclude that there is evidence that both of these effects are contributing factors and that at least the high-Eddington rate NLS1s are physically similar to some high-luminosity quasars.
Black hole masses for samples of active galactic nuclei (AGN) are currently estimated from single-epoch optical spectra. In particular, the size of the broad-line emitting region needed to compute the black hole mass is derived from the optical or ultraviolet continuum luminosity. Here we consider the relationship between the broad-line region size, R, and the near-infrared (near-IR) AGN continuum luminosity, L, as the near-IR continuum suffers less dust extinction than at shorter wavelengths and the prospects for separating the AGN continuum from host-galaxy starlight are better in the near-IR than in the optical. For a relationship of the form R propto L^alpha, we obtain for a sample of 14 reverberation-mapped AGN a best-fit slope of alpha=0.5+/-0.1, which is consistent with the slope of the relationship in the optical band and with the value of 0.5 naively expected from photoionisation theory. Black hole masses can then be estimated from the near-IR virial product, which is calculated using the strong and unblended Paschen broad emission lines (Pa alpha or Pa beta).
We present HST/ACS observations of ten galaxies that host narrow-line Seyfert 1 (NLS1) nuclei, believed to contain relatively smaller mass black holes accreting at high Eddington ratios. We deconvolve each ACS image into a nuclear point source (AGN), a bulge, and a disk, and fitted the bulge with a Sersic profile and the disk with an exponential profile. We find that at least five galaxies can be classified as having pseudobulges. All ten galaxies lie below the mbh--L$_{bulge}$ relation, confirming earlier results. Their locus is similar to that occupied by pseudobulges. This leads us to conclude that the growth of BHs in NLS1s is governed by secular processes rather than merger-driven. Active galaxies in pseudobulges point to an alternative track of black hole--galaxy co-evolution. Because of the intrinsic scatter in black hole mass--bulge properties scaling relations caused by a combination of factors such as the galaxy morphology, orientation, and redshift evolution, application of scaling relations to determine BH masses may not be as straightforward as has been hoped.
We review briefly direct and indirect methods of measuring the masses of black holes in galactic nuclei, and then focus attention on supermassive black holes in active nuclei, with special attention to results from reverberation mapping and their limitations. We find that the intrinsic scatter in the relationship between the AGN luminosity and the broad-line region size is very small, ~0.11 dex, comparable to the uncertainties in the better reverberation measurements. We also find that the relationship between reverberation-based black hole masses and host-galaxy bulge luminosities also seems to have surprisingly little intrinsic scatter, ~0.17 dex. We note, however, that there are still potential systematics that could affect the overall mass calibration at the level of a factor of a few.
We present high-resolution HST images of all 35 AGNs with optical reverberation-mapping results, which we have modeled to create a nucleus-free image of each AGN host galaxy. From the nucleus-free images, we determine the host-galaxy contribution to ground-based spectroscopic luminosity measurements at 5100A. After correcting the luminosities of the AGNs for the contribution from starlight, we re-examine the Hbeta R-L relationship. Our best fit for the relationship gives a powerlaw slope of 0.52 with a range of 0.45 - 0.59 allowed by the uncertainties. This is consistent with our previous findings, and thus still consistent with the naive assumption that all AGNs are simply luminosity-scal
We investigate the relationship between black hole mass and bulge luminosity for AGNs with reverberation-based black hole mass measurements and bulge luminosities from two-dimensional decompositions of Hubble Space Telescope host galaxy images. We find that the slope of the relationship for AGNs is 0.76-0.85 with an uncertainty of ~0.1, somewhat shallower than the M_BH propto L^{1.0+/-0.1} relationship that has been fit to nearby quiescent galaxies with dynamical black hole mass measurements. This is somewhat perplexing, as the AGN black hole masses include an overall scaling factor that brings the AGN M_BH-sigma relationship into agreement with that of quiescent galaxies. We discuss biases that may be inherent to the AGN and quiescent galaxy samples and could cause the apparent inconsistency in the forms of their M_BH-L_bulge relationships.
We present a reverberation analysis of the strong, variable optical Fe II emission bands in the spectrum of Akn 120, a low-redshift AGN which is one of the best candidates for such a study. On time scales of several years the Fe II line strengths follow the variations in the continuum strength. However, we are unable to measure a clear reverberation lag time for these Fe II lines on any time scale. This is due to the very broad and flat-topped nature of the Fe II cross correlation functions, as compared to the H-beta response which is much more sharply localized in time. Although there is some suggestion in the light curve of a 300-day response time, our statistical analysis does not pick up such a feature. We conclude that the optical Fe II emission does not come from a photoionization-powered region similar in size to the H-beta emitting region, but we cannot say for sure where it does come from. Our results are generally consistent either with emission from a photoionized region several times larger than the H-beta zone, or with emission from gas heated by some other means, perhaps responding only indirectly to the continuum variations.
We present a stellar dynamical estimate of the black hole (BH) mass in the Seyfert 1 galaxy, NGC 4151. We analyze ground-based spectroscopy as well as imaging data from the ground and space, and we construct 3-integral axisymmetric models in order to constrain the BH mass and mass-to-light ratio. The dynamical models depend on the assumed inclination of the kinematic symmetry axis of the stellar bulge. In the case where the bulge is assumed to be viewed edge-on, the kinematical data give only an upper limit to the mass of the BH of ~4e7 M_sun (1 sigma). If the bulge kinematic axis is assumed to have the same inclination as the symmetry axis of the large-scale galaxy disk (i.e., 23 degrees relative to the line of sight), a best-fit dynamical mass between 4-5e7 M_sun is obtained. However, because of the poor quality of the fit when the bulge is assumed to be inclined (as determined by the noisiness of the chi^2 surface and its minimum value), and because we lack spectroscopic data that clearly resolves the BH sphere of influence, we consider our measurements to be tentative estimates of the dynamical BH mass. With this preliminary result, NGC 4151 is now among the small sample of galaxies in which the BH mass has been constrained from two independent techniques, and the mass values we find for both bulge inclinations are in reasonable agreement with the recent estimate from reverberation mapping (4.57[+0.57/-0.47]e7 M_sun) published by Bentz et al.
