The timing and location of the type II onset with respect to the X-ray features suggest that the origin of the type II burst is closely related to the dynamics of the plasma on spatial scales of active region loops (size km). The estimated speed of the X-ray blob makes it a plausible driver of a fast shock in the corona. Since the type II source lies above the disrupted X-ray loops and starts in close time coincidence with the disruption, the shock formation seems to be directly related to the disruption of the loop. The soft X-ray signature of the driver persists after the onset of the type II burst. This suggests that intially the type II emission is associated with a driven shock. We cannot decide from the available data if and at which time the energy supply from the driver is turned off so that subsequently the type II shock may become a blast wave.
We conclude a posteriori that a similar case was reported on 27 September 1993 by Klassen et al. (1999; their Fig. 4). In both cases the type II emission starts above an evolving loop, which is seen side-on in the 27 September event, and in projection on the disk on 27 November. Detailed inspection of the Yohkoh data on 27 September 1993 suggests that the loop follows a similar evolution to the event discussed in the present paper (Nitta 1998, pers. comm.). A comparable relative position of a presumed type II source (there was no overlap between the frequencies of the imaging and spectral observations) with respect to a rapidly rising X-ray structure was reported by Gopalswamy et al. (1997).
The X-ray blob is reminiscent of the "plasmoids" analyzed by Ohyama & Shibata (1997; 1998), both by its shape and its early onset with respect to the main hard X-ray peak. But its trajectory is clearly curved, following the pre-existing large-scale loops, and its projected speed is higher (770 km s-1, as compared to 200 to 500 km s-1). No type II bursts were observed by the Tremsdorf spectrograph for the events examined by Ohyama & Shibata.
The alignment of the type II sources at different frequencies demonstrates a strongly non-radial propagation of the type II exciter. Non-radial alignment of type II sources has been discussed before (e.g. Nelson & Robinson 1975; Gergely et al. 1983; Stewart 1984; Aurass et al. 1998; Klassen et al. 1999). Among the explanations invoked were the preferential acceleration of electrons in restricted regions of the shock front (e.g. the quasi-perpendicular regime, Steinolfson 1984) and the refraction of the shock wave into regions of low Alfvén speed (Uchida 1974). The present work suggests that the motion of the driver and the magnetic environment where the type II burst occurs may play a role (cf. also Aurass et al. 1998). In the high-frequency range, i.e. close to the place where the shock is generated, the driver seems to impose the site of the first radiative signature of the shock and the direction of propagation.
© European Southern Observatory (ESO) 1999
Online publication: June 17, 1999