Spectral evolution of multiply impulsive solar bursts

Abstract

Hard X-ray and microwave observations of multiply-impulsive solar bursts, identified in the OSO-S data, have been analyzed in a search for the physical basis of the multiplicity of spikes within these events. Spectra in both frequency ranges have been used to determine whether or not the source p roperties change from peak to peak within individual bursts. Two categories of microwave spectral behaviour have been identified: those events during which the microwave turnover frequency and spectral shape remain the same from peak to peak, and those during which the turnover f.ret.uei,cy and spectral shape change significantly.1'liese categories correspond to two classes of multiply-impulsive bursts: those fc;r which the emission can be characterized by a constant magnetic field and therefore a single source region, in which case the multiplicity may be due to modulation of the emission process; and those in which groups of component spikes appear to originate in regions of different magnetic-field strengths, corresponding to separate source regions which flare sequentially. Examples of the latter type of events are presented, with a detailed analysis of the discrete flaring regions and estimates of their spatial separations.

The role of betatron acceleration in complex solar bursts

“…Although a least-squares fitting routine (as was used in the analysis described in Karpen et al, 1979) would provide the most accurate determinations of the spectral index, y, and the normalization constant, A, this procedure is far too time-consuming to be practical for analysis of thousands of spectra. A less precise but adequate technique was implemented, wherein 3' is the weighted mean of the spectral indices determined for each pair of energy channels showing significant flux above the noise level; a pair consists of Channel 2 and one of the higher-energy channels (see Crannell et al, 1978, Table 1).…”

“…These events were chosen from the larger set of multiply-impulsive bursts, initially selected according to the criteria outlined by Karpen et al (1979), by requiring sufficient X-ray flux to allow determination of the thick-target parameters F(25) and 3' throughout most, if not all, of the event. Table I lists the bursts and the time periods over which they were tested for the presence of betatron acceleration.…”

“…The multiply-impulsive bursts, introduced in Section 4, were selected solely on the basis of the appearance of the hard X-ray time-intensity profiles (cf. Karpen et al, 1979), without consulting the coincident coverage at other wavelengths. A closer look at all available coverage of the April 26 event, however, strongly supports the identification of this burst as a two-stage event: The initial phase of X-ray emission is associated with type III bursts, while the second stage of X-ray activity is closely associated with a type II burst (see Figure 4).…”

“…Frost, 1969;Frost and Dennis, 1971;Hoyng, 1975). As shown by Karpen et al (1979), the multiplicity of certain impulsive bursts can be attributed to separate flaring sites within the main region, each contributing a group of emission spikes to the entire event. An additional explanation is needed, however, for the existence of multiple peaks in those events whose emission originates in a single source location.…”

“…In the present work, two pertinent facets of the betatron model are investigated: The possibility that the multiplicity characteristic of complex impulsive bursts is due to the betatron process; and the possibility that some or all of the second-stage emission during two-stage bursts can be attributed to betatron acceleration. To test for the pattern of X-ray spectral behavior predicted by the betatron model, a number of multiply-impulsive events (cf., Karpen et al, 1979) and two-stage bursts (cf., Frost and Dennis, 1971) were selected from the OSO-5 hard X-ray spectrometer data for in-depth analysis. The purely impulsive emissions show no signs of the effects of betatron action, thus eliminating this process as a potential source of impulsive-phase multiplicity.…”