Abstract

We study the evolution of the broad, double-peaked H? emission-line profile of the LINER/Seyfert 1 nucleus of NGC 1097, using 24 spectra obtained over a time span of 11 years - from 1991 November through 2002 October. While in the first 5 years the main variation was in the relative intensity of the blue and red peaks, in the last years we have also observed an increasing separation between the two peaks, at the same time as the integrated flux in the broad line has decreased. We propose a scenario in which the emission originates in an asymmetric accretion disk around a supermassive black hole, whose source of ionization is getting dimmer, causing the region of maximum emission to come closer to the center (and thus to regions of higher projected velocity). We use the observations to constrain the evolution of the accretion disk emission and to evaluate two models: the elliptical-disk model previously found to reproduce the observations from 1991 to 1996 and a model of a circular disk with a single spiral arm. In both models the peak emissivity of the disk drifts inward with time, while the azimuthal orientation of the elliptical-disk or the spiral pattern varies with time. In the case of the spiral-arm model, the whole set of data is consistent with a monotonic precession of the spiral pattern, which has completed almost two revolutions since 1991. Thus, we favor the spiral-arm model, which, through the precession period, implies a black hole mass that is consistent with the observed stellar velocity dispersion. In contrast, the elliptical-disk model requires a mass that is an order of magnitude lower. Finally, we have found tentative evidence of the emergence of an accretion disk wind, which we hope to explore further with future observations.