Session 1 - Chromosphere, Corona, Flares.
Display session, Friday, June 27
Ballroom B, Chair: Charles Kankelborg

## [1.70] Hard X-ray Footpoints and Soft X-ray Loops in Solar Flares

S. Masuda (STE Laboratory, Nagoya University), T. Sakao, T. Kosugi, J. Sato (National Astronomical Observatory)

Yohkoh Hard X-ray Telescope (HXT) found, in the impulsive phase of solar flares, a compact hard X-ray (HXR) source located above the corresponding soft X-ray (SXR) loop apex observed with Yohkoh Soft X-ray Telescope (SXT). This observation suggests that the flare primary energy release takes place above (outside of) the bright SXR loop, and that magnetic reconnection is responsible for the energy release. One possible interpretation is that electrons are energized in or near the above-the-loop-top'' HXR source where reconnection outflow impinges the underlying loop and forms a fast MHD shock. The energized electrons stream down along magnetic field lines to the footpoints (double footpoint HXR sources'') of a newly reconnected loop. This implies that the footpoint HXR sources must be located further from the loop center than the two ends of the loop found via SXR because reconnection starts from innermost magnetic fields reaching the outermost ones later and also because it takes a few minutes for the newly reconnected loop to be filled with hot plasma and become bright in SXRs.

Under such a motivation, we have compared HXR images in the HXT M1-band a (23-33 keV) with SXR images taken simultaneously with the SXT Be filter. In this analysis, the center of each HXR footpoint source represents the position of that source. For each SXR loop, first we remove the low-intensity part (< 1/10 of the maximum brightness) then the two ends of the remaining loop emission are taken to represent the positions of the SXR footpoints. In seven out of the eleven cases analyzed, we find that the footpoint HXR sources are located outside the corresponding SXR loop footpoints. The distance between the HXR sources is a little longer than that between the SXR loop footpoints, on average, by \sim 0.9 \pm 1.5 arcsec. Although this result seems to be consistent with the scenario mentioned above, this difference is not significantly large. A further study is in progress.