Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 351, L15-L18 (1999)

Previous Section Next Section Title Page Table of Contents

3. Results

Two typical examples of GRS 1915+105 X-ray flux histories in the [FORMULA] keV band are presented in Fig. 1 along with hardness ratios ([FORMULA] keV to 2-13 keV) and dynamic PDS integrated over 4 s time intervals. The source behavior in the flaring state is characterized by frequently occurring episodes of hard energy spectrum lasting typically [FORMULA] s. The prominent, relatively narrow variable-frequency [FORMULA] Hz QPO peak (dark `U'-shaped band in the lower panels in Fig. 1) is a generic feature of the source power density spectrum during hard episodes (Markwardt et al. 1999; Muno et al. 1999).

[FIGURE] Fig. 1. Part of the light curves (upper panels ), corresponding hardness ratios ([FORMULA] keV)/([FORMULA] keV) (middle panels ) and dynamic PDS (lower panels ) of GRS 1915+105 for the Oct. 7, 1996 (left panels ) and Oct. 25, 1996 (right panels ) observations ([FORMULA] keV energy band, PCA data). The QPO peak appears as `U'-shaped dark band in the lower panels . The light curves were not corrected for the instrument dead time which was [FORMULA]. The overall count rate corresponds to 5 Proportional Counter Units of the PCA detector.

Using the data from our sample of flaring state observations (Table 1), we studied the relation between the duration of a hard episode, [FORMULA] (determined using the hardness ratio) and the lowest value of the QPO frequency reached (determined from the fitting of the PDS to the analytical model). In Fig. 2 the duration of a hard episode is plotted as a function of the minimal QPO frequency. For all observations used in the analysis there is a strong anticorrelation between these two quantities. The dependence has a nearly power law form for durations of a hard episode shorter than [FORMULA] s and probably flattens for longer events. According to the value of the slope [FORMULA] of this power law dependence the observations can be subdivided into two groups: [FORMULA] for the 06/04/96, 07/10/96, 18/06/97, 07/07/97 and 17/12/97 observations (the first group, left panel in Fig. 2) and [FORMULA] for the 26/05/96, 25/10/96, 07/11/96 and 18/05/97 observations (the second group, right panel in Fig. 2). The analysis of the flux histories showed that these two groups have qualitatively different lightcurves: for the first group the transition from `hard' to `soft' state is characterized by the growth of the [FORMULA] keV flux (Fig. 1, upper left panel ), while for the second group this transition corresponds to the decrease of the X-ray flux (Fig. 1,upper right panel ). In the subsequent paper (Trudolyubov et al. 1999b) it will be shown that these two groups are also distinguished by the properties of their energy spectra: for the first group the soft thermal component makes a significant contribution to the overall luminosity in the [FORMULA] keV energy range, for the second group the energy spectrum is dominated by the hard component (the contribution of the soft thermal component to the total luminosity is [FORMULA]).

[FIGURE] Fig. 2. The relation between the duration of a hard episode and a corresponding minimal QPO frequency for the flaring state observations listed in Table 1 (first group - left panel ; second group - right panel ). The dependences [FORMULA] of the viscous time scale upon the Keplerian frequency at the inner edge of the radiation pressure dominated disk with the mass accretion rate [FORMULA] and [FORMULA] are shown in the left and right panels by long-dashed and dotted lines respectively (see text).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: November 3, 1999