Comparison of the 1998 April 29 M6.8 and 1998 November 5 M8.4 flares

We combined, and analyzed in detail, the H alpha and magnetograph data from Big Bear Solar Observatory (BBSO), full-disk magnetograms from the Michels on Doppler Imager (MDI) on board Solar and Heliospheric Observatory (SOHO), coronagraph data from the Large Angle Spectrometric Coronagraph (LASCO) of SOHO, Fe XII 195 Angstrom data from the Extreme ultraviolet Imaging Telesc ope (EIT) of SOHO, and Yohkoh soft X-ray telescope (SXT) data of the M6.8 f lare of 1998 April 29 in National Oceanic and Atmospheric Administration (N OAA) region 8375 and the M8.4 flare of 1998 November 5 in NOAA region 8384. These two flares have remarkable similarities: 1. Partial halo coronal mass ejections (CMEs) were observed for both events . For the 1998 April 29 event, even though the flare occurred in the southe ast of the disk center, the ejected material moved predominantly across the equator, and the central part of the CME occurred in the northeast limb. T he direction in which the cusp points in the postflare SXT images determine s the dominant direction of the CMEs. 2. Coronal dimming was clearly observed in EIT Fe XII 195 Angstrom for both but was not observed in Yohkoh SXT for either event. Dimming started 2 hr before the onset of the flares, indicating large-scale coronal restructurin g before both flares. 3. No global or local photospheric magnetic held chan ge was detected from either event; in particular, no magnetic field change was found in the dimming areas. 4. Both events lasted several hours and, thus, could be classified as long duration events (LDEs). However, they are different in the following important aspects. For the 199 8 April 29 event, the flare and the CME are associated with an erupting fil ament in which the two initial ribbons were well connected and then gradual ly separated. SXT preflare images show the classical S-shape sheared config uration (sigmoid structure). For the 1998 November 5 event, two initial rib bons were well separated, and the SXT preflare image shows the interaction of at least two loops. In addition, no filament eruption was observed. We c onclude that even though these two events resulted in similar coronal conse quences, they are due to two distinct physical processes: eruption of shear ed loops and interaction of two loops.