Astron. Astrophys. 363, 863-868 (2000)

6. Discussion

While the Narrow Line Seyfert 1 Galaxy Mrk 766 shares with other members of its class the large amplitude and short time-scale variability, it does not share (at least during the observation described here) another characteristic which is common in Narrow Line Seyfert 1 galaxies, i.e. a prominent soft X-ray emission. The power law component is, actually, somewhat steeper than the typical value for classical Seyfert 1 galaxies, but the soft excess is modest, if present at all. The warm absorber is also typical of classical, `broad lines' Seyfert 1s, being dominated by oxygen and neon edges. No 1 keV feature (besides those related to the warm absorber and cured by the inclusion of this component) is present, differently from what observed in other sources of this class (e.g. Leighly et al. 1997; Fiore et al. 1998; Leighly 1999; Vaughan et al. 1999a). This is not surprising, as these features have been observed so far only in very steep spectrum sources, a fact naturally explained if these features are due to a blend of resonant absorption lines, mainly from iron L-shell (Nicastro et al. 1999a). The column density and ionized parameter of the warm absorber appear to have been changed between the two halves of the observations, in agreement with the fact that the variability is the largest around 1 keV. A so dramatic change of the column density, however, is unplausible, and it may be an artifact of having used pure photoionization equilibrium, single-zone models, while in reality it is possible that: the absorber is out of equilibrium most of the time (not surprisingly, given the flux variability; see Nicastro et al. 1999b for non equilibrium models); there is some contribution from collisional ionization; the absorbing medium is geometrically or physically complex.

Similar to other Narrow Line Seyfert 1s (TON S 180: Comastri et al. 1998; Turner et al. 1998; Ark 564: Vaughan et al. 1999b) is, instead, the possible presence of reprocessing from ionized matter. Ionized accretion discs are expected when is high (Ross & Fabian 1993; Matt et al. 1993), and accretion rates close to the Eddington one have been indeed invoked to explain the NLS1 phenomenon (e.g. Pounds et al. 1995;). The (mild) ionization of the reprecessor may also account for the lack of any observed iron line (Matt et al. 1993, 1996), as the line photons may be resonantly trapped and then destroyed by the Auger effect.

Reflection from a mildly ionized disc may also explain at least part of the O VII emission line observed in the first half of the observation (in the second half only an upper limit is obtained, which however is consistent, within the error, with the value measured in the first half). In this case, the line is expected to be broadened by relativistic effects; the quality of our data is not good enough to distinguish between a narrow and a relativistic line, provided that the inclination angle of the disc is low. An oxygen line possibly from an ionized disc was observed by Piro et al. (1997) in the ASCA data of E1615+061, while the same line observed by BeppoSAX in the spectrum of NGC 5548 (Nicastro et al. 2000) clearly originates from outflowing material, as demonstrated by the Chandra/LETG observation (Kaastra et al. 2000). In Mrk766, it is possible that both the accretion disc and the warm absorber contribute to the observed emission, as the best fit ionization structure for both materials includes a significant fraction of O VII . High energy resolution observations are needed to definitely settle this issue.

© European Southern Observatory (ESO) 2000

Online publication: December 5, 2000
helpdesk.link@springer.de