Astron. Astrophys. 364, 557-562 (2000)

5. Discussion and conclusions

The availability of an estimate of the radius of the primary of AG Dra gives us the chance to calculate some other parameters too. The abundance analysis of this star performed by Smith et al. (1996) provided an estimate of the surface gravity of . Using this estimate and a radius of 28-32 we derive a mass of 1.1-1.5 which is in agreement with the result of the analysis of this component by Mikolajewska et al. (1995). Having the radius and the effective temperature of the star, we found its bolometric luminosity, which turned out to be in the interval 242316 .

The values for the distance to the system lead to the reduction of some of the parameters of its hot companion obtained by Greiner et al. (1997) which are based on a distance of 2.5 kpc. Since a very hot star was observed, these authors proposed that it is in a steady state hydrogen burning near its surface allowing its luminosity to be due only to nuclear processes. A steady state burning can be realized when the burning rate is equal to the accretion rate. Greiner et al. found that if the hot companion of the AG Dra system had such an accretion rate, it will be in the range of steady burning at the surface of a white dwarf with a mass of 0.3 . This mass was calculated using the core mass-luminosity relation (Yungelson et al. 1996). Making corrections of their parameters we obtained that the bolometric luminosity and the accretion rate are in the intervals 9671302 and 1.21.7 10^-8 yr^-1. The mass of the companion was calculated to be about 0.3 , which leads to a minimum accretion rate for steady burning of 1.8 10^-9 yr^-1. So it can be concluded that these new values are also in agreement with the model of a steady state burning at the surface of a white dwarf.

We summarize the main results of our analysis as follows:

• The variations of the U flux between the different orbital maxima are due to variations of the amount of the emitting gas in the circumbinary nebula. This is possible only if the hot companion has an excess of luminosity, which makes it able to ionize always the whole nebula (but some region occulted by the cool star) when the giant changes its mass-loss rate.

• The shape of the U light curve indicates that the flux equal to the orbital amplitude is emitted by a small high-density region located most probably around the hemisphere of the giant which faces the hot companion.

• The cool giant's mass-loss rate was estimated from the H profile to be in the range 2.02.5 10^-7 yr^-1 on the basis of the theoretical models of Schwank et al. (1997), and from the similarity with the H profile of AG Peg.

• Using the mass-loss rate and the U fluxes of the different components of the system we calculated its distance and the radius of its cool giant to be in the ranges 15601810 pc and 2832 , respectively.

© European Southern Observatory (ESO) 2000

Online publication: January 29, 2001
helpdesk.link@springer.de