Introduction

The "ribbon" structures of solar flares have long been observed at H-alpha and UV wavelengths. In the standard flare model, the ribbon structure in the lower atmosphere represents a series of footpoints of coronal arcade loops that result from magnetic reconnection at a coronal X-point, and the well-observed separation motion of the flare ribbon is interpreted as successive reconnections that takes place higher and higher above the arcade. This is essentially a 2D picture and we do not know yet how well it translates to 3D, but it certainly appears to organize the data observed during the gradual phase of a flare.

On the other hand, flare hard X-ray emissions often exhibits point-like compact sources, often two though sometimes only one, which lie within the H-alpha/UV ribbons. This difference between the spatial distributions of ribbons and the hard X-ray sources is widely recognized by experts and is an active subject of research. One explanation was proposed by Asai et al. (2002), who interpreted the more confined hard X-ray kernels as regions of more intense energy release which might be expected in the strongest-field parts of the ribbons.

Prior to RHESSI there were few reports of ribbon-like images in hard X-rays. The first was the Yohkoh/HXT observations reported by S. Masuda et al. These were the ribbons of the celebrated Bastille Day 2000 flare, a complicated X5.7 event. The ribbons could be seen even in the HXT High channel, 53-93 keV. Such observations suggest that electrons are accelerated in the whole system of a coronal arcade, and not merely in a particular dominant loop.

RHESSI has now brought new insight into the distinction between the H-alpha-inspired ribbons and the hard X-ray footpoints. This nugget reports on observations from a flare of 2005 May 13, whose date incidentally echoes the date in 1981 of a memorable Hinotori flare.

The 2005 May 13 flare

We have found ribbon-like hard X-ray sources during this M8.0 disk event, which was associated with a fast halo CME and an intense geomagnetic storm. We chose a six one-minute time intervals from the rising to decay phase of the hard X-ray light curve and present the RHESSI images in Figure 1.

Figure 1: A time sequence of RHESSI 25-50 keV hard X-ray images across the flare impulsive phase. Each RHESSI image was reconstructed with the CLEAN algorithm using grids 1-9 with the natural weighting scheme. The green contours show flux at levels of 0.1, 0.115, and 0.13 photons/cm ²/s/arcsec². Panel f also shows RHESSI 6-12 keV image with yellow contours at levels of 50%, 70%, and 90% of its maximum flux. The white contours outline the TRACE 1600 Å ribbons taken near the center of each RHESSI time interval.

We discuss the results as follows. First, hard X-ray emissions appear as compact sources until the flare maximum (intervals a, b, and c). These are located within the flare ribbons. Four hard X-ray kernels are seen at flare maximum. As suggested by Asai (2002), we find that the average field strength of the hard X-ray parts of the ribbons is about two times larger than that of the other parts of the ribbons (those without hard X-ray emissions). Second, after flare maximum (intervals d, e, and f) the hard X-ray sources become elongated and appear to form a true ribbon structure. This footpoint-to-ribbon evolution of hard X-ray emissions is more evident for the much stronger eastern hard X-ray sources. Several kernels can be seen within the ribbon during time interval e. At the time interval f, significant hard X-ray emission (although with a much lower flux level compared with peak time) is found along the entire length of each ribbon. The same trend is found at higher energies (50-100 keV), although at lower energies (6-12 keV) the X-ray sources lie between the ribbons. This presumably shows high-temperature thermal sources at the tops of the loops joining the ribbon sources.

We want to understand why the hard X-ray ribbon structure appears so prominently in this specific event. Actually, this eruption is also interesting in other respects. The flaring active region, NOAA 10759, appeared in a conspicuous sigmoid shape in the TRACE 171 Å channel. Sigmoid shapes are well-known as special magnetic structures prone to eruption. Following the event, the visible structure changed from a sigmoid to an arcade (see Figure 2).

Figure 2: Pre- and postflare images from TRACE 171 Å channel showing the sigmoid-to-arcade evolution of the coronal magnetic field. The MDI longitudinal magnetic field is superimposed as red and green contours representing positive and negative fields, respectively.

We consider a scenario as follows: as in the standard model for eruptive bipoles proposed especially by Ron Moore and termed tether-cutting. Reconnection begins between the two elbows in the middle of the sigmoid (see Figure 3a), and progressively less sheared field lines from the outer sigmoid core reconnect. The hard X-ray sources should be largely footpoint-like at this stage. At the flare maximum, the envelope is blown out with the twisted flux rope inside it, after which the opened legs of the envelope will continue to reconnect and this accelerates electrons in the whole system of arcade or fan out throughout the arcade thus leading to the ribbon-like hard X-ray emissions (see Figure 3b).

Figure 3: Schematic picture interpreting our observations, based on the eruptive model proposed by Moore and LaBonte (among others).

Conclusions

Combined with a rich collection of radio emission features, this beautiful event shows an picture of the flare/CME process that is quite similar to the model for sigmoidal bipoles elaborated by Moore and LaBonte. We speculate that the footpoint-to-ribbon transformation of the hard X-ray source morphology is a natural outcome of the sigmoid-to-arcade evolution of the magnetic field configuration. Therefore, this might suggest that only the events that exhibit a special magnetic field configuration can have truly ribbon-like hard X-ray emissions.

---

Biographical note: Chang Liu is a PhD student at the Center for Solar-Terrestrial Research at New Jersey Institute of Technology. CL is grateful to Dr. Brian R. Dennis for valuable discussions. This work is part of a manuscript presenting a multiwavelength study of this event.