Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 359, 998-1010 (2000)

Previous Section Next Section Title Page Table of Contents

1. Introduction

Cataclysmic variables are known to be interacting binary stars in which a Roche-lobe filling late type star (the secondary) looses mass to a white dwarf primary. In most of these systems the dominant light source at optical and ultraviolet wavelengths is an accretion disk formed around the white dwarf by matter transferred from the secondary star.

One of the most characteristic observational properties of all cataclysmic variables are stochastic brightness variations known as flickering, occurring on time scales of seconds and minutes with amplitudes ranging from a few dozen millimagnitudes up to more than an entire magnitude, depending on the subtype of the CV, the photometric state, and the individual system. Flickering is observed in almost all photometric states of CVs, except during outbursts of classical and recurrent novae when only the optically thick expanding envelope around the system is visible, and during extreme low states of some stars when mass transfer from the secondary star is probably suspended and the accretion disk has vanished.

Although flickering is so typical for CVs it is not restricted to this kind of stars. It may rather be regarded as a fingerprint of systems with mass accretion via an accretion disk onto a central body. For example, several symbiotic stars (albeit only a minority!) exhibit flickering with characteristics very similar to those observed in CVs (Dobrzycka et al. 1996). Flickering is also doubtlessly present in low-mass x-ray binaries, although due to a lack of optical high speed photometry it is by far not as well documented in these systems as in CVs. An exception is Sco X-1 (see e.g. Hiltner & Mook 1967, Sandage et al. 1969 Robinson & Warner 1972, and Augusteijn et al. 1992). Herbst & Shevchenko (1999) have recently explained the irregular variations observed in UX Ori stars (i.e. early type T Tau stars with irregular variations on comparatively long time scales; for a more complete definition, see Herbst et al. 1994) as due to luminosity variations caused by unsteady accretion of matter onto the central star out of the surrounding disk. The longer time scales involved in these cases are naturally explained by the larger dimensions of the central bodies. A study of the irregular variations of AGNs along these guidelines may also be rewarding.

Whereas the assumption that flickering is intimately related to mass transfer in binary systems or mass accretion onto a central body (be it a single object or a component of a binary) is certainly well justified the nature of the underlying physical mechanism is by no means clear. It is even doubtful that a single mechanism can be pinpointed. The fact that flickering occurs in all CVs independent of whether they have well developed accretion disks as the non-magnetic systems, truncated disks as the intermediate polars, or no disks at all as the AM Her stars immediately suggests that more than one mechanism must be involved. This is supported by the differences of the statistical properties of the flickering in magnetic and non-magnetic CVs found in a comprehensive wavelet analysis of the flickering in numerous systems performed by Fritz & Bruch (1998). Moreover, Bruch (1996) and Bruch et al. (2000) have shown that in the dwarf novae Z Cha and V893 Sco, respectively, a part of the short time scale flickering originates in the hot spot while the rest comes from the central parts of the accretion disk and/or the white dwarf. Localizing the flickering within a CV obviously constrains possible mechanisms. Therefore, in this study the location of the flickering light source is determined for several more systems.

This can best be done in eclipsing CVs by quantifying the random variations due to flickering in an ensemble of light curves as a function of the orbital phase (i.e. by measuring the scatter curve). During phases when the flickering light source is best visible the scatter of the brightness measurements in high-speed photometry within a small phase interval or with respect to the light curve averaged over many cycles should on the mean be enhanced while it should become small when the flickering light source is invisible (e.g. eclipsed).

This technique was first applied to the nova-like variable RW Tri by Horne & Stiening (1985). Later, Bruch (1996) developed a somewhat different technique and applied it to Z Cha. Similar studies were performed for HT Cas (Welsh & Wood 1995, Welsh et al. 1996), V893 Sco (Bruch et al. 2000), and again for RW Tri (Bennie et al. 1996). In this contribution the location of the flickering light source will be studied in four systems: HT Cas (a SU UMa type dwarf nova), V2051 Oph (a dwarf nova of no well established sub-type), IP Peg (a U Gem type dwarf nova) and UX UMa (a nova-like variable). With the exception of HT Cas none of them has been subjected to a corresponding study before.

Concerning the details to calculate the scatter curve two schools exist, called `single' and `ensemble' by Welsh et al. (1996). They will be compared to each other in Sect. 2 before the observational material on which the present study is based is presented in Sect. 3 and the `single' method is applied to the data in Sect. 4. Finally, Sect. 5 contains a summary of the principal conclusions of this study.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: July 13, 2000