next up previous
Next: Sample selection Up: Ultraviolet-Optical observations of the Previous: Ultraviolet-Optical observations of the

Introduction

According to the unified scheme of AGN, the two types of Seyfert galaxies are the same phenomenon, but they appear as Seyfert 1 or Seyfert 2 depending on their orientation relative to the line of sight (e.g., Lawrence 1991). During the last decade, sufficiently convincing observational support has been obtained for this scenario. One of the strongest pieces of observational evidence for this model was the discovery of broad recombination and FeII lines in the optical spectra in polarized light of some Seyfert 2 nuclei (Antonucci & Miller 1985; Miller & Goodrich 1990; Miller, Goodrich, & Mathews 1991; Tran 1995). Furthermore, the high excitation gas observed extending out from the nucleus with conical or biconical morphology (Wilson, Ward, & Haniff 1988; Tadhunter & Tsvetanov 1989; Pogge 1989; Pérez et al 1989; Storchi-Bergmann, Wilson, & Baldwin 1992; Cappetti, Axon, & Machetto 1997) is taken as additional observational evidence in favor of the unified scheme. However, some problems are still left unresolved by the unified scheme.

One of the most outstanding of such issues is the nature of the optical and UV continuum in type 2 Seyfert nuclei. In most such nuclei, the primary contributor to the optical continuum is the light from an old population of stars. This light is however diluted by an additional component in which the normal spectral features of old stars is weak or absent. This component has traditionally been called the `featureless continuum', on the assumption that is produced by the AGN in some way. As we will show in this paper, this continuum is not featureless at all! It typically contributes 10 to 30% of the nuclear optical continuum in type 2 Seyfert nuclei, but is much more conspicuous in the UV.

What is this light? After the contribution of the old stars is removed, the remaining optical continuum has a significantly lower fractional polarization than the broad emission-lines, implying that the majority can not be due to scattered light from a hidden type 1 Seyfert nucleus ( Miller & Goodrich 1990; Kay 1994; Tran 1995). In a related vein, Terlevich, Díaz, & Terlevich (1990) proposed a stellar origin for `featureless continuum' in Seyfert 2 galaxies based in the strength of the stellar features observed in the near-infrared. If the `featureless continuum' is radiation scattered into our line of sight from the obscured Seyfert 1 nucleus then it is difficult to self-consistently explain why the stellar CaII triplet feature (8498, 9542, 8662 Å) has an equivalent width similar to normal galaxies. Cid Fernandes & Terlevich (1995) have also proposed that a population of young stars in the vicinity of the nucleus is the source of the featureless continuum. Heckman et al. (1995) have constructed an ultraviolet spectral template using IUE spectra of 20 of the brightest Seyfert 2 nuclei. They find that only 20% of the UV continuum emission can be attributed to a hidden Seyfert 1 nucleus. They propose that most of the ultraviolet continuum in Seyferts 2 may be produced by a reddened circumnuclear starburst, which is producing most of the far-infrared continuum detected in Seyfert 2 galaxies. Direct observational evidence that only a small fraction of the total UV light detected in Seyfert 2 galaxies is emitted by a hidden nucleus comes from UV HST images of Seyfert 2 galaxies (Heckman et al 1997; Colina et al 1997a).

The existence of a population of young stars very close to the active nucleus has been required by several AGN models. In the `Starburst Model' for AGN proposed by Terlevich and collaborators, the activity is produced by a centrally concentrated burst of star formation (Terlevich & Melnick 1985; Terlevich et al. 1992). In the so called hybrid models, a very compact (r~10 pc or even smaller) circumnuclear starburst coexists with a massive central black hole, which is responsible for generating the ionizing continuum (Norman & Scoville 1988; Scoville & Norman 1988; Perry & Dyson 1985; Perry & Williams 1994).

Observational evidence for a possible connection between nuclear activity and circumnuclear starbursts in Seyfert 2 galaxies comes from the strength of the far-infrared continuum (Rodríguez-Espinosa, Rudy, & Jones 1986; Dultzin-Hacyan, Moles, & Masegosa 1990; Pier & Krolik 1993), the CO 115 GHz emission (Heckman et al 1989), the strength of the CaII triplet in absorption (Terlevich et al, 1990), the departure of the galaxy or bulge blue luminosity from the Tully-Fisher and Faber-Jackson relationship (Whittle 1992a,b; Nelson & Whittle 1995), the mid-infrared 10 micron emission (Maiolino et al 1995), the ratio of the luminosity in the H band and mass of the circumnuclear region in Seyfert 2 (Origlia, Moorwood, & Oliva, 1997), the diffuse radio emission around some Seyfert nuclei (Wilson 1988) and the fraction of circumnuclear starbursts in Seyfert 2 galaxies (González Delgado et al 1997).

To understand the origin of the featureless continuum in Seyfert 2 nuclei and its relationship with circumnuclear starbursts, we have undertaken a program that comprises high resolution UV images and UV spectra with the HST, and ground based spectra from near UV to the near infrared. The first results, for Mrk 477, have been presented by Heckman et al (1997). The data provide direct evidence that the continuum (from UV to near infrared) is dominated by a circumnuclear dusty starburst in Mrk 477. Here, we present the results for three other Seyfert 2 galaxies (NGC 7130, NGC 5135 and IC 3639). Together with Mrk 477, these nuclei are among the 20 brightest Seyfert 2 nuclei. Our goal is to present general conclusions about the AGN-Starburst connection and the origin of the featureless continuum in Seyfert 2 nuclei. The paper is organized as follows: In Section 2 we present the criteria used to select the objects. Section 3 presents the observations and data reduction. Sections 4-7 deal with the results and interpretation of the HST images, GHRS spectra, and ground-based optical spectra. We discuss our results and their implications in Section 8, and summarize our conclusions in Section 9.


next up previous
Next: Sample selection Up: Ultraviolet-Optical observations of the Previous: Ultraviolet-Optical observations of the 
Rosa Gonzalez

1998-06-20