The Size, Structure, and Ionization of the Broad-Line Region in NGC 3227

Abstract

Hubble Space Telescope spectroscopy of the Seyfert 1.5 galaxy, NGC 3227, confirms previous reports that the broad Hα emission line flux is time variable, decreasing by a modest ∼11% between 1999 and 2000 in response to a corresponding ∼37% decrease in the underlying continuum. Modeling the gas distribution responsible for the broad Hα, Hβ, and Hγ emission lines favors a spherically symmetric inflow as opposed to a thin disk. Adopting a central black hole mass of 7.6 × 10 6 M , determined from prior reverberation mapping, leads to the following dimensions for the size of the region emitting the broad Hα line: an outer radius ∼90 lt-days and an inner radius ∼3 lt-days. Thus, the previously determined reverberation size for the broad-line region (BLR) consistently coincides with the inner radius of a much larger volume of ionized gas. However, the perceived size of the BLR is an illusion, a consequence of the fact that the emitting region is ionization bounded at the outer radius and diminished by Doppler broadening at the inner radius. The actual dimensions of the inflow remain to be determined. Nevertheless, the steady-state mass inflow rate is estimated to be ∼10 −2 M yr −1 which is sufficient to explain the X-ray luminosity of the active galactic nucleus (AGN) in terms of radiatively inefficient accretion. Collectively, the results challenge many preconceived notions concerning the nature of BLRs in AGNs.

A New Method to Obtain the Broad Line Region Size of High Redshift Quasars

“…As a consequence, the BLR physical conditions cannot be described by a single value of density and ionization parameter. Even the time lag from reverberation should be interpreted with some care, since it is a single number that is a rather abstract representation of the BLR distance from the central continuum source and may not have a well-defined structural counterpart (Devereux 2013). Nonetheless, a correction based on the Aliiiλ1860/Ciii]λ1909 ratio can lower the product n H U increasing the derived r BLR and minimize the bias in the photoionization estimates with respect to reverberation lag (i.e.…”

REVERBERATION AND PHOTOIONIZATION ESTIMATES OF THE BROAD-LINE REGION RADIUS IN LOW-zQUASARS

“…The Aliiiλ1860, Siiii]λ1892, Civλ1549 lines are all emitted in the fully ionized zone of emitting gas clouds or slabs (Paper I), so that they are sensitive to the ionizing photon flux which is exactly the product n H U . A more realistic approach would be to consider a model that allows gas density n H , column density N c , and ionization parameter U , to be functions of r (Devereux 2013;Devereux & Heaton 2013). As pointed out by Devereux (2013) the r BLR derived from reverberation mapping is an abstraction that may not have a clear structural counterpart, and the RM r BLR can indeed be lower than an emissivity weighted average if the geometry of the BLR is thick (Netzer 1990).…”

“…A more realistic approach would be to consider a model that allows gas density n H , column density N c , and ionization parameter U , to be functions of r (Devereux 2013;Devereux & Heaton 2013). As pointed out by Devereux (2013) the r BLR derived from reverberation mapping is an abstraction that may not have a clear structural counterpart, and the RM r BLR can indeed be lower than an emissivity weighted average if the geometry of the BLR is thick (Netzer 1990). Nonetheless, RM r BLR has been considered as the best approximation available for the "virial radius" of the BLR and widely employed as such (e.g.…”

Quasars and their emission lines as cosmological probes

“…The radius r BLR is obtained by measuring the peak or centroid displacement of the crosscorrelation function between the light curves of the continuum and a strong line. While the meaning of this measure is not fully clear (Devereux, 2013), r BLR is measured in a way that is redshift independent. If it were possible to measure the angular size of the BLR then a redshift independent value of the angular distance d A would follow: d A (H 0 , Ω M , Ω Λ ) = cτ /θ (Elvis and Karovska, 2002).…”

“…It is known since long (e.g., Netzer, 1990) that r BLR derived with reverberation mapping can differ from what is ideally expected to measure i.e., an emissivity weighted distance from the central continuum source. The issue is certainly not settled (Devereux, 2013). Unclear aspects involve the uncertain effect of radiation forces and the physical conditions that are likely to be different from object to objects, as pointed out by Watson et al (2011).…”