A spectrophotometric atlas of Narrow-Line Seyfert 1 galaxies

¹ Observatoire de Haute Provence, CNRS, 04870 Saint-Michel l'Observatoire, France e-mail: mira@obs-hp.fr
² European Southern Observatory (ESO), Karl Schwarzschild Strasse 2, 85748 Garching bei München, Germany e-mail: adarbon@eso.org

Corresponding author: P. Véron, veron@obs-hp.fr

Received: 8 February 2001
Accepted: 20 March 2001

Abstract

We have compiled a list of 83 objects classified as Narrow-Line Seyfert 1 galaxies (NLS1s) or known to have a broad Balmer component narrower than 2 000 km s^-1. Of these, 19 turned out to have been spectroscopically misidentified in previous studies; only 64 of the selected objects are genuine NLS1s. We have spectroscopically observed 59 of them and tried to characterize their Narrow and Broad-Line Regions (NLR and BLR) by fitting the emission-lines with Gaussian and/or Lorentzian profiles. In most cases, the broad Balmer components are well fitted by a single Lorentzian profile, confirming previous claims that Lorentzian rather than Gaussian profiles are better suited to reproduce the shape of the NLS1s broad emission lines. This has consequences concerning their FWHMs and line ratios: when the broad Balmer components are fitted with a Lorentzian, most narrow line regions have line ratios typical of Seyfert 2s while, when a Gaussian profile is used for fitting the broad Balmer components, the line ratios are widely scattered in the usual diagnostic diagrams (Veilleux & Osterbrock [CITE]); moreover, the FWHM of the best fitting Lorentzian is systematically smaller than the FWHM of the Gaussian. We find that, in general, the [O III] lines have a relatively narrow Gaussian profile (~200-500 km s^-1 FWHM) with often, in addition, a second broad (~500-1 800 km s^-1 FWHM), blueshifted Gaussian component. We do not confirm that the [O III] lines are weak in NLS1s. As previously suggested, there is a continuous transition of all properties between NLS1s and classical Broad-Line Seyfert 1 Galaxies (BLS1s) and the limit of 2000 km s^-1 used to separate the two species is arbitrary; R₄₅₇₀, the ratio of the Fe II to the Hβ fluxes, could be a physically more meaningful parameter to distinguish them.