A Spectral Analysis of the Masuda Flare Using Yohkoh Hard X-ray Telescope Pixon Reconstruction

MasudaÏs discovery of a compact hard X-ray impulsive source at the apex of a Ñaring coronal loop has received a great deal of recent attention in the solar physics community. The Masuda Ñare, which occurred on 1992 January 13, exhibited evidence of energy deposition in a compact region some distance above the soft X-ray loop, suggesting, to some authors, a Ñare process similar to the classical model for two-ribbon Ñares proposed by Shibata et al. These conclusions were made on the basis of a maximum entropy method (MEM) reconstruction of the Yohkoh Hard X-Ray Telescope (HXT) observations. Recently, a new approach has been developed for reconstructing the spatial information from the HXT: that of pixon reconstruction, proposed by Metcalf et al. In this paper, we apply the pixon reconstruction technique to the event of 1992 January 13 and determine the temporal and spectral characteristics of the loop-top source. While our emphasis here is on the spectral properties of the Masuda Ñare, we also provide a brief comparison between the pixon reconstruction and that of MEM for the hard X-ray loop top. In carrying out the comparison between the methods, we have applied recent improvements to the instrument response functions and reconstruction algorithms. We have also identiÐed a previously unknown e†ect of weak source suppression that was inherent in previous analyses and that signiÐcantly compromised the ability to study weak sources of hard X-ray emission in the presence of strong sources. The improved response functions and the better Ñux estimation used in this paper reduce (but do not eliminate) the e†ects of this suppression, and consequently, it should be noted that the MEM analysis presented in this paper is quite distinct from any that have been carried out previously. Our conclusions are that (a) a compact loop-top hard X-ray source exists with an impulsive temporal proÐle spanning the peak of the Ñare ; (b) the loop-top source is nonthermal in nature at the peak of the Ñare ; (c) there is a distinct dearth of HXT LO channel emission, relative to the higher energy channels, from the loop-top region, indicating either a very hard spectrum or the presence of a low-energy cuto† in the energetic electron spectrum; (d) the footpoint and loop-top emission during the impulsive phase of the Ñare are produced by two distinct particle populations ; (e) following the main phase of this Ñare, the loop top is clearly thermal in nature with a peak temperature of D40 MK that decreases with time as the event proceeds ; and ( f ) the disparity between the present pixon results and previous MEM results is primarily due to the intrinsically better photometry achieved by the pixon method and the avoidance of suppression e†ects in the present analysis. These conclusions therefore support, in part, those made in previous works, conÐrming the existence of an impulsive source of hard X-rays in the corona above a Ñaring loop. Our analysis does, however, allow for a more comprehensive understanding of the temporal and spectral development of this event in the context of an alternative reconstruction technique. Subject headings : Sun: corona È Sun: Ñares È Sun: X-rays, gamma rays