The nature of the 380-day periodicity cannot be derived from broad-band photometry. The main purpose of the present paper is thus to initiate relevant additional observations before the end of the present outburst phase of AG Dra.
Accretion events have often been suggested as the cause of outbursts in symbiotic systems (e.g. by Kafatos et al., 1993). An obvious scenario for the cause of the periodicity would be low-amplitude pulsations of the giant component, leading to corresponding modulations in the density of the material flowing towards the white dwarf. Such fluctuations could be expected regardless of the precise mechanism leading to the flow (be it the giant wind alone, or direct Roche overflow). This scenario can be tested by more extensive and more precise radial-velocity measurements for the cool giant component, and probably by precise photometry in the red or near-IR spectral range. In passing, it should be noted that the scatter of the individual outburst maxima around the linear ephemeris is well in the range of the irregularities of pulsating giants.
One could also imagine an unstable self-excited oscillation of the mass flow due to the back-heating of the giant's surface by the radiation from the burning white dwarf. A chance increase of the heat flux from the white dwarf could cause an expansion of the giant's atmosphere, thus producing increased mass flow and - in turn - increased energy output from the white dwarf.
A third possibility are independent oscillations of the hydrogen-burning envelope of the white dwarf, unrelated to the giant component, as can often be seen in novae after the initial eruption.
The two latter scenarios can be tested by detailed observations of the white dwarf in the blue/UV spectral range. Whatever the mechanism leading to the periodicity in the outburst maxima may be, its investigation will provide new constraints on astrophysical models for this symbiotic system.
In addition to potentially being interesting astrophysically, the
discovery of the 380-day periodicity is a good example for the
possible uses of visual data that still exist today. Visual
observations of variable stars clearly have big disadvantages compared
to photoelectric measurements. The random error of the individual
observation is large (0.1 to 0.2 mag), the absolute calibration is
uncertain, and the photometric passband is broad and ill-defined.
Nevertheless, for the purpose of studying light variations on long
time scales, visual observations can offer a most suitable data base.
The main reasons for this were very well summarized by Richman et al.
(1994): `This is primarily
because amateur astronomers outnumber professionals by factors of 100-1000. This ensures a steady supply of data over many years, unlike the typical research programs of the professional astronomer, which produce high-quality data for a while, but are eventually scuttled by disinterest, death, or disappearance of observing capability. These hazards do not afflict visual observations of amateur astronomers.... Finally, the response curve of the average human eye can be expected to remain constant over any relevant time scales, unlike scientific instruments. ... The large random error can be overcome simply by amassing a very large amount of data.'
© European Southern Observatory (ESO) 1998
Online publication: December 16, 1997