Astron. Astrophys. 356, 975-988 (2000)

Astron. Astrophys. 356, 975-988 (2000)

9. Conclusions

We developed a continuous jet model for the radio outbursts of galactic microquasars. The model naturally explains the observed rather flat decaying lightcurves of these bursts as the signature of synchrotron radiation of relativistic particles accelerated by internal shocks in the conical jets. The comparatively long duration of the bursts implies that this model is a `time-resolved' version of the internal shock model proposed for GRB (Rees & Meszaros 1994), though the synchrotron emission is produced at much lower frequencies. The gradual steepening of the radio spectrum is explained by a superposition of the radiation of different populations of relativistic particles with different ages. This spectrum of ages results from the shock traveling along the jet with older populations of accelerated particles left behind.

We find that only a roughly constant rate of acceleration of relativistic particles followed by an exponential decay can explain the observed light curves for the strong outburst of GRS 1915+105 in 1994. We interpret this behaviour as the signature of two colliding shells of jet material, as in the internal shock model for GRB. A consequence of this is the continued steepening of the lightcurves of a given outburst coupled with a decreasing flux ratio of the emission observed from the approaching to that from the receding jet side.

The energy requirements of the continuous jet model for producing radio outbursts are similar to those of the plasmon model. However, much of the energy underlying the outbursts may be stored in the continuous jet while the passage of the internal shock only `lights up' the jet. This implies that the rate at which the energy of the outburst is supplied by the central engine to the jet is much lower than in the plasmon model.

The occurrence of mini-bursts in microquasars with flat spectra up to infrared frequencies (Mirabel et al. 1998) and the observation of K-band emission in the jet a considerable distance away from the core (Sams et al. 1996) suggest different modes of jet production: (i) A stable `mini-burst' mode with relative little variation in the bulk jet speed and therefore also only weak internal shocks. (ii) A more variable outburst mode with strong variations in the jet speed and strong internal shocks (see also Fender 1999). The weaker flavour internal shocks seem to produce flatter relativistic particle spectra extending to higher energy compared to the strong shocks. However, the total number of accelerated particles must be much larger in the strongly variable phase.

We show that the properties of the continuous jets of microquasars should lead to strong shocks at their ends where they are in contact with the surrounding ISM. This is consistent with the recent observations of the decelerating radio emission region of XTE J1748-288 (Hjellming et al. 1999) and its persistence for 15 months after the start of the original outburst (Rupen, private communication). The shocked jet material may subsequently inflate a low density cavity around the jets similar to the radio lobes in extragalactic jet sources of type FRII. This is observed in SS433 (Dubner et al. 1998) and may be in a few other microquasars. The absence of such shocks and radio lobes in GRS 1915+105 may indicate that the jets in this source are young and/or that they recollimate because of the pressure of their environment. If this is the case then the detection of a non-thermal emission region in the more extended environment of GRS 1915+105 (Rodríguez & Mirabel 1998) may imply a recurrence time of the jet activity scale of [FORMULA] years.

Many of the predictions of this model for microquasars can be tested observationally. However, to clearly distinguish between this model of continuous jets and the plasmon model it would be necessary to spatially resolve the superluminal emission regions during outbursts, preferentially at more than one radio frequency. Additional support for the scenario of continuous jets may come from further high resolution observations of the cores of microquasars during quiescence. These should show at least some spatial extension of the radio emission along the jet axis as observed in Cygnus X-1 (Fender et al. 2000). These observations can potentially provide us with valuable information on the properties of the jets which otherwise we can only study during strong outbursts when strong shocks pass through them.

Online publication: April 17, 2000
helpdesk.link@springer.de