Introduction

Solar X-ray jets (transient X-ray bursts in the form of collimated flows of high-temperature plasma) were first observed by Yohkoh; see any of several Yohkoh Science nuggets such as the one on the triple jet. Jets occur in regions of emerging magnetic field, X-ray bright points and active regions, and it is supposed that magnetic reconnection is responsible for the acceleration of particles and heating of the plasma to temperatures of a few million degrees during these events. The emerging-flux model (Heyvaerts - Shibata) fits well. It describes the interaction of emerging photospheric field with pre-existing overlying coronal magnetic field. As the new emerging flux rises, it is pushed up against and reconnects with the overlying coronal magnetic field lines, and the reconfiguration allows the the previously highly stressed field lines of the emerging flux to act as a slingshot as they straighten out, resulting in the ejection of plasma that had been previously supported in the emerging flux tube. At the same time the surrounding plasma is heated, and this can be seen as the jet X-ray emission. In the process we would expect the acceleration of electrons, as in a solar flare, and indeed jet-related electrons have been detected directly in space (see our earlier Nugget) as well as via the radio signatures (type III bursts).

Of course, as we know from these RHESSI Nuggets, the major diagnostic of fast electrons is bremsstrahlung X-ray emission, but until now we have never seen any evidence of hard X-ray emission directly from the jet in the corona. This could be because it is rare to find a coronal jet dense enough to provide a bremsstrahlung target for the electrons, or hot enough to generate high energy thermal emission. In this Nugget we report what we believe to be the first observation of hard X-ray emission formed in a coronal jet.

Event overview

In this nugget we present a jet which occurred on the 22nd August 2002 preceding a GOES M5.4 flare. Its evolution was observed by a number of instruments. In particular we concentrate on the pre-impulsive and impulsive phases of the flare. Figure 1 shows time profiles from GOES, RHESSI, TRACE and the Nobeyama Radioheliograph (NoRH). The dashed vertical lines on these plots show time intervals over which RHESSI images were obtained.

From the movie of TRACE 195 observations a series of collimated ejections can be recognized. TRACE images covering the main ejection of material can be seen in the "confusogram" of Figure 2. This diagram displays time slices top to bottom, and different RHESSI energy ranges left to right. At the time of the main ejection, ~01:50:30 (row 2), RHESSI emission to as high as the 30-50 keV energy band can be seen in the region of the ejected material. This suggests the presence of nonthermal hard X-ray emission in the jet. As the event continues, footpoints can be seen to form; we interpret these in the standard model as resulting from the precipitation of electrons accelerated in the corona and guided into the lower atmosphere on newly-connected field lines. RHESSI spectroscopy for this event shows temperatures of around 20MK, with a power-law tail. Without imaging spectroscopy, it's not possible to be sure how much the jet emission is contributing to the power law, but there is strong evidence from radio (below) that non-thermal electrons are present in the jet.

As seen in Figure 1, this event coincided with radio observations from the Nobeyama Solar Radio Observatory in Japan. This observatory makes images at 17 and 34 GHz that are extremely useful for diagnosing the flaring plasma, especially in conjunction with RHESSI. At these frequencies we can assume that the microwave emission is gyrosynchrotron radiation from particles at even higher energies than those observed via RHESSI hard X-rays, and hence definitely non-thermal in nature. For the main time interval of interest (row 2 at 01:50:30 UT) the radio images also show radio emission in the jetting region, confirming the presence of non-thermal electrons.

Figure 3 (left) shows the Nobeyama 17 GHz, 34 GHz and RHESSI 30-50 keV contours overlaid on the corresponding TRACE image. This suggests evidence of gyrosynchrotron emission hence supporting the case for nonthermal particles present in the jet at this time. From the flux at 17 GHz and 34 GHz it is possible to determine a radio spectral index α from F_ν ~ ν^α. Figure 3 (right) then shows a contour map of this spectral index alpha with contours of positive alpha, ranging from 0 to 1.5 in steps of 0.25, corresponding to an optically thin plasma in the jetting region through which the fast moving electrons propagate. As expected, type III radio bursts could be seen throughout this interval (see, for example, this RHESSI Browser excerpt).

Conclusions

All of the evidence is consistent with the presence of non-thermal electrons in the jet region just at the onset of the impulsive phase of the flare. The novelty of this event is the presence of a clear 30-50 keV hard X-ray signature and its confirmation as non-thermal via the Nobeyama spectral-index mapping. Further work can be carried combining these observations by Nobeyama and RHESSI to estimate the magnetic field strength and direction in the jetting region. The electron distribution of the energetic electrons producing emission at these wavelengths can also be determined. We note that observations of this type tend to be ambiguous and model-dependent, and we hope that other examples will be found.