The detection of stellar velocity dispersion drops in the central regions of five isolated Seyfert spirals ^*

¹ Instituto de Astrofísica de Andalucía (CSIC), Apartado 3004, 18080 Granada, Spain
² Institut d'Astrophysique de Paris, CNRS, 98bis Bld Arago, 75014 Paris, France
³ Departamento de Astronomía, Universidad de Chile, Casilla 36D, Santiago, Chile
⁴ Observatories of the Carnegie Institution of Washington, 813 Barbara Street, Pasadena, CA91101, USA

Corresponding author: I. Márquez, isabel@iaa.es

Received: 7 March 2003
Accepted: 25 June 2003

Abstract

We analyze the kinematics of the central regions of five isolated Seyfert spiral galaxies from the DEGAS sample (four with new data presented in this paper, IC 184, UGC 3223, NGC 2639, NGC 6814, and NGC 6951 from our previous data), by using long slit spectroscopy in the CaII triplet range (at ≈8600 Å) obtained with a 4 m-class telescope. A drop of the velocity dispersions in the innermost ±(1–3) arcsec is observed in four of them, and hinted at in the remaining galaxy (NGC 6814). The available HST images for our sample together with another nine galaxies with reported velocity dispersion drops are also used to investigate the presence of morphological inner structures at the scales of the kinematical drops. Evidence for disk-like shapes is found in 12 out of the 14 cases. The only exceptions are NGC 6814 and NGC 6951. Existing N-body simulations including stars, gas and star formation predict that such a drop is most probably due to a young stellar population born from dynamically cold gas accreted in a circumnuclear disk formed during an episode of central gas accretion driven by a bar. The equivalent widths of the calcium triplet lines for our 5 galaxies have been measured. Even if the profiles could be formally consistent with constant EW(CaT) values, they seem to indicate the presence of a local maximum in the regions corresponding spatially to the drops; if confirmed, this would imply the presence of a different stellar population, whose properties could help constrain the models.