Radio versus X-ray Variations

In figure 3 we show the radio and x-ray flux density and spectral properties plotted against radio phase. The error bars shown for the spectral data are plus and minus one standard deviation. The upper panel shows the 4.9 GHz radio flux density and the x-ray flux. For x-ray data we have adopted the average value for the three model spectral distributions. Both data sets have been normalized to a peak value of unity. Two facts are readily apparent: one, the radio and and x-ray emission exhibit a similar fractional change in flux of about a factor of 10, and two, the timing of the peak intensity at radio and x-ray differ by approximately 0.5 in phase. The exact phase difference between the radio and x-ray curves is difficult to estimate, both because of the coarse sampling of the data and the ambiguity about whether the x-ray emission leads or lags the radio outburst.

The central panel shows the radio spectral index, , defined by assuming a power dependence, . There is a clear evolution from positive to negative spectral index during the rise and decay of the first radio outburst. The spectral index crosses zero close to the time of peak radio flux density, and decays to a constant value of -0.3. The average spectral index following the decay is -0.330.04. The spectral index appears to remain at this value into the rise of the second outburst. However, the time sampling of the radio light curve is poor, and it is possible that the radio peak occurred between the last two radio measurements which are widely separated in phase - 0.55 and 0.74. This is the more common range of phase of peak flux density (Taylor & Gregory 1984; Paredes et al. 1990). In that case, the last measurement of spectral index would be during radio decay.

To examine the possibility of a change in the spectral shape of the x-ray emission, we calculated a hardness ratio based on the observed spectra. The hardness ratio was defined as the ratio of photon counts in spectral channels defining the energy range 1.0 to 2.48 keV to the counts in the energy range 0.07 to 1.0 keV. The hardness ratio appears to be roughly constant for most of the observations. For the first and last observations which were obtained at very close to the same radio phase the spectral shapes are identical to within the limits of the data. However, there is a suggestion that the spectrum hardens for the observations at peak x-ray flux at phase 0.48. This occurs at the onset of the second radio outburst.