Miocrowave spectral imaging, H-alpha and hard X-ray observations of a solar limb flare

We compare the microwave, H-alpha, and hard X-ray observations for a west limb C7.3 flare that occurred at 17:10 UT, 1992 June 26. H-alpha movies were obtained at Big Bear Solar Observatory. Before the onset of the flare, overexposed H-alpha images show the complicated flux loop structure above the limb. Material was observed to descend along the loops toward the site where the flare occurred hours later. Using the five-antenna solar array at Owens Valley Radio Observatory, we obtain two-dimensional maps of flare emission from 1.4 to 14 GHz. In all three temporal peaks of the microwave bursts, the maps show the same characteristics. The peak low-frequency emission comes from the top of one bundle of the H-alpha loops and gradually shifts to the foot-point of the loops (the location of H-alpha flare) as the frequency increases. The location of the emission peak shifts 88 sec between 1 and 14 GHz. Seventy percent of the shift occurs between 1 and 5 GHz. The locus of the shift of the emission peak follows the shape of an H-alpha surge that occurred after the flare. For each point along the locus, we create the microwave brightness temperature spectrum and compare the radio-derived electron distribution with that derived from the high-resolution hard X-ray spectra measured with Burst and Transient Source Experiment (BATSE) on board the Compton Gamma Ray Observatory (CGRO). We find that the peak frequency changes from approximately 3 GHz at the loop top to approximately 7 GHz at the footprint, presumably due to the increase of the magnetic field from approximately 160 GHz at the loop top to approximately 300 G at the footpoint. The high-frequency slope of the microwave power-law spectrum decreases from approximately 10 at the loop top to approximately 5 at the footprint due to a change in the energy distribution of the dominant electrons. The microwave brightness temperature spectral index predicted by the BATSE power-law hard X-ray spectra agrees with the measured value only at the footpoint. At the loop top, the emission may be thermal gyrosynchrotron with a temperature of 3.5 x 10(exp 7) K, which is likely to correspond to the superhot component seen in the hard X-ray emission.