Mkn 110 is one of the very few Seyfert galaxies with spectral variability coverage over a time interval of ten years. Different continuum ranges show different variability amplitudes; this holds for different optical emission lines, too. But the mean fluxes of the continuum and of all emission lines remain nearly constant integrated over time scales of a few years (see Fig. 4). There are considerable variations over time scales of days to years. The strongest variability amplitudes in the continuum shows the blue spectral range (see Figs. 2 to 4). There are intensity variations of a factor of . The strongest amplitudes in the blue spectral range might be explained by a greater share of the non-thermal continuum with respect to the underlying galaxy continuum.
The optical line variations of H are very strong in comparison to other Seyfert galaxies (e.g. Peterson et al. 1998a). The HeII4686 line shows the strongest variations of nearly a factor of 8 within two years. On the other hand the H and the continuum(5100) vary only by a factor of 1.7 and 3.0 respectively within the same time interval. Apart from the variation of the HeII4686 line in NGC 5548 in 1984 (Peterson & Ferland 1986) these are the strongest optical line variations within such a time interval. In the case of Mkn 110 we can show that the appearance of the very broad HeII4686 and H component (see Figs. 3 and 13) is not a unique event in the accretion rate. It is connected to a very strong ionizing continuum flux as can be seen from the light curves. The intensity ratio HeII4686/H comes to a value of 1.3 (see Fig. 13) in the very broad line region. Such a line ratio is still in correspondence with photoionization of broad emission-line clouds in quasars (Korista et al. 1997).
The very broad line region (VBLR) originates close to the central ionizing source at a distance of about 9 light days. It is not connected to the "normal" BLR. As can be seen from the line profiles there exists no continuous transition region between these BLRs. The center of the VBLR line profiles is shifted by 400100 km s-1 with respect to the "normal" BLR profiles (Figs. 10, 11, 13).
Apart from this VBLR component we could show that the line profiles of the Balmer and HeI lines are similar but not identical. The H line profile is narrower than the H profile. Besides the cross-correlation results this is an independent indication that these two lines do not originate in exactly the same region.
The observed Ly/H ratio comes to a value of 11.0 in Mkn 110. This is about a factor of two higher than the mean observed Ly/H ratio in Seyfert galaxies (Wu et al. 1983). Photoionization models of Kwan & Krolik (1981) result in Ly/H without the presence of dust. Therefore, dust may not play an important role in the BLR of Mkn 110. The Balmer decrement in Mkn 110 varies as a function of the ionizing continuum flux. This might be explained by radiative transfer effects rather than by variation of dust extinction.
The profiles of the broad emission lines in Mkn 110 are neither symmetric nor smooth (Figs. 10, 11). This is a further indication that the broad-line regions in AGN are structured as e.g. in NGC 4593 (Kollatschny & Dietrich 1997). In Fig. 12 it is shown that during the first half of our campaign a red line component was present in the H spectra. This component was not visible during the second half of the campaign.
The size of the H line emitting region (r = 40 ld corresponding to 1.0 1017 cm) and the optical continuum luminosity is compared to those of other Seyfert galaxies. The continuum luminosity amounts to erg s-1 Å- 1. In this case we used km s-1 Mpc-1 in order to compare directly the radius and luminosity of Mkn 110 with those of other Seyfert galaxies compiled by Carone et al. (1996). The values of Mkn 110 fit nicely into the general radius-luminosity relationship for the broad-line regions in Seyfert galaxies. The Balmer line emitting region as well as the luminosity of Mkn 110 are arranged in the upper region of the radius-luminosity plane close to the galaxies Mkn 590 and Mkn 335.
Table 7. Virial mass estimations
The central mass in Mkn 110 can be estimated from the width of the broad emission line profiles (FWHM) under the assumption that the gas dynamics are dominated by the central massive object. Furthermore, one needs the distance of the dominant emission line clouds to the ionizing central source (e.g. Koratkar & Gaskell 1991, Kollatschny & Dietrich 1997). We presume that the characteristic velocity of the emission line region is given by the FWHM of the rms profile and the characteristic distance R is given by the centroid of the corresponding cross-correlation function:
In Table 7 we list our virial mass estimations of the central massive object in Mkn 110. Altogether we determine a central mass of:
We can independently estimate an upper limit of the central mass if we interpret the observed redshift of the very broad HeII component () as gravitational redshift (e.g. Zheng & Sulentic 1990):
Again we presume that this line component originates at a distance of 9 ld from the central ionizing source. We derive an upper limit of the central mass of
This second independent method confirms the former mass estimation.
© European Southern Observatory (ESO) 1999
Online publication: April 12, 1999