Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 335, 545-548 (1998)

Previous Section Next Section Title Page Table of Contents

3. Interpretation

Among the features which need interpretation, we may mention the lack of exact periodicity in the U variations, the apparently different nature of the variations in B and V, as well as the presence of an additional period to the orbital one of Meinunger, detected in B and V and perhaps also in the interval between the two light peaks seen during activity.

If we investigate the cause of variations in U, we can firstly note that the contribution of the radiation due to the hot compact component of the binary should be not more than about 10 % in the U band (0.1 magnitudes) during quiescence according to the temperatures and radii indicated, by the results of Mikolajewska et al. (1995) and of Greiner et al. (1997). These results assume that the hot component radiates as a black body. Mikolajewska et al. 1995 reached their conclusions by analyzing the ultraviolet spectrum observed with IUE, while Greiner et al. 1997 used X-ray observations made by the ROSAT satellite. It seems from these results and from the period analysis that the quiescence variations in U can be explained by varying visibility of an ionized gaseous region, which is in fact the wind from the giant ionized by the hot radiation. This is also indicated by the presence of strong Balmer continuum radiation.

The orbital modulations in the U band are almost certainly not produced by Rayleigh scattering. The log cross section of one hydrogen atom is near -25.9 in the U band according to Nussbaumer et al. 1989. For a cool component mass loss rate of [FORMULA] at a velocity of 30 [FORMULA] (distance 2.5 kpc) given by Mikolajewska et al. 1995, the amount of Rayleigh scattering of the wind is negligible in U.

The changes in the orbital variation in U are probably due to a lack of constancy of the wind from the cool component at least in quiescence, as there is no clear evidence for significant temperature changes of the compact component at such times (Greiner et al. 1997). In any case the ROSAT PSPC quiescence intensities observed by these authors at times near that of our photometry were almost constant. Proga et al. 1998 suggest the acceleration of ionized regions of the wind from the cool components of symbiotic binaries by thermal pressure of the plasma, which can moreover vary with illumination and small changes in the stellar mass loss rate. In addition wind acretion by the compact component is also unstable in such situations (Ruffert 1996 and Benensohn et al. 1997). The changes in the phases and shapes of U maxima, also suggest varying deviations from axial symmetry of the ionized parts of the cool star wind.

The small quiescence variations in B and V are different than those in U. This could suggest the presence of an additional mechanism, perhaps connected with the cool giant, whose radiation becomes more important at longer wavelengths. One possibility is the pulsation of the cool giant, but this is still far from sure. In any case the near equality between this period and the separation of maxima during activity, might suggest a relation between the physical processes of the cool giant and the modulation of activity in active phases.

It clearly turned out that the data must be analyzed separately for the quiescent and the active stages. In any case, we find a striking indication that the 350 days periodicity is connected to the maxima of outbursts during the activity. On the other hand, the 550 days orbital period is very well detected in the overall long-term data with removed outbursts' peaks in all three colours. Such analysis is beyond the scope of this paper and is going to be the subject of our next study.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: June 18, 1998