6. Summary and conclusions
We have studied an eruptive, two-ribbon flare, developed around 15:43 UT on Feb. 4, 1995 in a region near disk center. We concentrate mainly on the "decay-phase", i.e. on the flare evolution after the maximum emission both in H and soft X-rays. We analyse coronal and chromospheric signatures, obtained with the cluster of instruments onboard Yohkoh and at the VTT of NSO/SP. In this later phase of the flare, we witness new episodes of coronal energy release, as proved by observations of coronal and chromospheric heating. We find convincing evidence that such energy releases are due to magnetic reconnection.
At chromospheric level we observe the development of two new flaring ribbons, in different position with respect to the ones developed in the impulsive phase. The ribbons separate from each other at an average velocity of few tens of km s-1. This kind of motion, often observed, is considered one of the signatures of magnetic reconnection in the solar atmosphere.
At coronal levels we see SXR emission overlaying the whole flaring region. The analysis of the Lo-HXT and Fe XXV light curves points to a thermal origin of the HXR emission, from a source of Te =22 MK, EM=5. cm-3. This source is visible in the only HXT image that we could recover, as a point-like source positioned between the two new ribbons.
Within the new ribbons, several small kernels (4- in size) brighten at different times; analyzing their spatially resolved H 1.5 Å light curves we find a close temporal correspondence (within 2.5 s) between two pairs of kernels. We interpret this temporal coincidence assuming that each kernel pair represents the chromospheric footpoints of a coronal loop (or small system of loops) interested by the energy release, as in the case studied by Kurokawa et al. (1988). In both footpoints of the second pair of flaring kernels (maximum at 15:47:30 UT) we see He D3 emission, but no enhancement of the continuum, a signature often used to infer the site of non-thermal electrons precipitation (Cauzzi et al., 1995; Qiu et al., 1997). This fact, and the presence of a hot thermal source somewhere near the loop top at the same time, suggest that the chromospheric footpoints are heated by a conduction front. Other evidence lead us to believe that conduction from a coronal source is the mechanism responsible also for the brightening of the other kernels observed in this later phase of the flare.
The presence of a conduction front, directly related to magnetic reconnection episodes, is also consistent with an interesting, observational result of this study: from CaII K and H spectra acquired in several positions across the flare, we find that the most intense chromospheric downflows are present at the outer edge of the ribbons. These flows have an amplitude of tens of km s-1, and are observed whenever the slit crosses the outer edge of a flaring kernel or ribbon, including those observed in the flare impulsive phase. Other flaring structures show only a moderate redshift of the spectral lines, corresponding to flows of few km s-1. The downward flows seem to diminish, whenever the kernels (or ribbons) slow their horizontal motion on the solar surface and fade away.
We believe that these observations provide a clear evidence of magnetic reconnection. As suggested in the classical two-ribbon flare model, and explicitily shown in recent models that consider heat conduction from the reconnecting site, the conduction front directly maps the outer boundary magnetic field lines of the reconnecting loops. Its effect on the chromospheric layers would be to compress the plasma, triggering downflows only on the outer edge of the ribbons, consistently with our observations. We are monitoring the chromospheric counterpart of the same physical effect that causes coronal (soft X-ray) cusp-like structures, and hotter coronal loops at the outer boundary of a flaring system, usually assumed as direct evidence of magnetic reconnection (see e.g. Tsuneta, 1996). Finally, our observations could provide quantitative constraints to models of coronal magnetic reconnection that take into consideration the effects of chromospheric ablation and compression from a conduction front.
© European Southern Observatory (ESO) 1997
Online publication: March 24, 1998