8 on a timescale
of months between the source's high and low states. Priedhorsky and Terrel (1983) found that
this behavior had a characteristic timescale of
120-165 days but found no evidence for periodicity. More recent results from the RXTE All Sky Monitor (ASM) show that Cen X-3's X-ray emission is more variable than this on a timescale of days.
Figure 4 shows the long-term X-ray light curve of Cen X-3 obtained with the ASM. These are
quick-look results provided by the ASM/RXTE team.
Gaps in the light curve are due to the source being too faint to detect.
New results on the variability of Cen X-3's luminosity will be presented in
Chapter 8.
The pulse shape is also variable, changing from an asymmetric single peak in the high state to a rarely observed double peaked structure in the low state (e.g. Long et al. (1975); Nagase et al. (1992); Tuohy (1976)). Relatively short term variability in the pulse profile has also been observed (e.g. Schreier et al. (1976); van der Klis et al. (1980)). Examples of this short-term variability will be shown in later chapters where I propose that they are due to luminosity-induced changes in the structure of the accretion column.
Day and Stevens (1993) have suggested that the source's high and low states are the result of positive feedback between the X-ray emission and an X-ray excited wind. The X-ray emission of the neutron star excites the wind which increases the mass accretion rate and thus the X-ray luminosity until the source becomes buried in an optically thick wind and possibly behind a flared accretion disk and switches off. A prediction of this model is that the column density and iron line equivalent width should be much higher in the low state than in the high state.
