From RHESSI Wiki
|1st Author:||Säm Krucker|
|2nd Author:||Hugh Hudson|
|Published:||5 January 2007|
|Next Nugget:||Bouncing Thick-Target Coronal X-Ray Sources|
|Previous Nugget:||The RHESSI fifth anniversary: X-flares persist!|
Solar flares invariably generate hard X-rays in a spectrum very much more energetic than that expected from the coronal plasma. As discussed in the RHESSI science nugget by Paolo Grigis, this hard X-ray spectrum shows a particular kind of time evolution in most solar flares. The spectrum flattens, then steepens after the peak. We call this the soft-hard-soft (SHS) spectral evolution. However, certain flares exhibit a HXR spectrum that progressively hardens throughout successive peaks, and in events like this the time duration is usually much longer. The prototype event in this category was observed by Frost and Dennis in 1969 with a quite primitive instrument. This alternative pattern, referred to as soft-hard-harder (SHH) or "progressive hardening," is more frequently found in flares with a gradual rather than an impulsive time profile and occurs much less often than the above-mentioned SHS behavior. One theory is that it could be due to trapping of energetic electrons in the natural "magnetic mirrors" formed by the coronal magnetic field. In a 1995 study, Alan Kiplinger found that flares with the SHH pattern are closely associated with high-energy solar proton events observed in interplanetary space. While the physical association between the progressively hardening X-ray spectrum and the particles is not understood at present, the Kiplinger study obtained a 96% success rate in predicting large proton events. If one saw the SHH pattern, a proton event was almost sure to happen. This association strongly suggests a physical connection between the X-ray-producing electrons in the flare (on the closed flare loops that create the magnetic mirror geometry) and the escaping energetic protons on open field lines. This correlation is rather puzzling, especially since most people believe that an interplanetary shock front, remote from the flare itself, is the main proton accelerator. Clearly we need to study this relationship if we are to understand the mysteries of particle acceleration by the Sun.
In this Nugget, we investigate the relationship between the five X-class flares and the solar particle events that occurred during the January 2005 solar storm.
Flare and SEP observations during the January 2005 storm
During the week between January 14 to January 21, 2005, six large flares with GOES class above M8.5 occurred in the same solar active region (Figure 1). From those six flares, four are clearly associated with a temporally correlated increase in the GOES >10 MeV proton time profiles, while one flare (January 15, 00:30 UT) shows no clearly related enhancement. For the January 19th flare, the pre-event GOES proton flux was still enhanced from the previous SEP event, especially at 10 MeV, making it difficult to determine the existence of a new injection. The >10 MeV proton flux shows a small increase after the January 19 event,, about 10 (cm2 s sr)-1; this could be the related SEP event. In-situ ion observations at lower energy from ACE clearly show a new increase at lower energies. Therefore, the January 19 flare likely produced an SEP event as well.
Figure 1: Several days of data from the GOES solar X-ray and particle monitors, showing the events of January 2005. All but one of the major X-ray events (upper plot) showed clear SHH spectral development. Click for a clearer view.
Hard X-ray spectra
RHESSI has good coverage for the five X-class flares, but missed the M8.6 flare (January 15, 6:00 UT) event. Hence, the sample of events to study the spectral evolution are four flares with SEP events, and one without. The January 15th (00:30 UT) event, the only X-class flare in January 2005 that did not produce an SEP event, clearly shows SHS behavior as the HXR flux and the spectral index are anti-correlated. The other event that only shows a single behavior is the flare of January 20 with a clear SHH behavior seen in every peak (Figure 2). The other three flares show first SHS and later SHH. Hence, all events with SEP production also show progressive spectral hardening, while the event without SEP shows SHS behavior, directly confirming Kiplinger's results within the active lifetime of this single highly energetic active region.
Figure 2: Illustration of the SHH pattern for the major event of January 20. Upper panel, the soft (red) and hard (blue) X-ray fluxes; lower panel, the hard X-ray spectral ratio in the 100-200 keV range. Note that the very hardest moment (lowest spectral index, 06:52 UT), does not coincide with a flux peak as in the SHS pattern. Click for a clearer view.
RHESSI images made during the progressively hardening phase of the flares indicate that hard X-rays originate from the footpoints rather than the corona (Figure 3). This was previously noted by Qiu et al., using Yohkoh/HXT observations of a flare with similar progressive hardening. Hence, if coronal trapping is responsible for the gradual hardening, the hard X-ray emission is not produced by electrons in the corona, but by electrons leaving the trap and precipitating to the footpoints. Therefore, the trapping time in the corona must be shorter than the collisional loss time, at least at energies (<100 keV) considered here (but see also an earlier Nugget).
Figure 3: X-ray images from the early (left) and late (right) phases of the January 20 event. Note the clear presence of compact footpoints even in the late phase, even when we think the sources show long-duration particle trapping.
Spectral analysis of five X-class flares observed in January 2005 by RHESSI clearly reinforce the relationship between progressively hardening seen in the flare hard X-ray emission and the solar energetic particle production first noted by Kiplinger. The nature of their association, however, is unknown, and it is not clear how the spectral hardening of flare-accelerted electrons in closed magnetic field structures is related to high energy protons that are able to escape the Sun and travel along open field lines towards the Earth.
This nugget provides first results from some of the most prominent flares with this progressive hardening (SHH) behavior. A statistical study containing all SEP events with good RHESSI coverage is being planned. Furthermore, no quantitative analysis of the spectral fit results have not yet been done. We look forward to really pinning down the models for electron trapping and injection by following the details of the temporal variations of the spectral parameters, including the break energy seen in some events. Using imaging spectroscopy we hope to map out the trap efficiency and the time variations of the actual velocity distribution functions of the trapped particles. Biographical note: Sa"m Krucker and Hugh Hudson are senior RHESSI team members in Berkeley.