fl069.nakajima01 Posted: 03-Sep-92 Updated: 31-Jul-93, 11-Feb-94, 29-Oct-94, 17-Jul-95, 08-Apr-96 Events specified: Flare on 28-Jun-92 at 04:45 UT
Collaboration: H.Nakajima, Y.Hanaoka, M.Nishio, K.Shibasaki, and Nobeyama Radioheliograph Group
Motivation and Method: This event is the strongest event that was observed with Nobeyama Radioheliograph. We want to synthesis a comprehensive image about this event using radio, x-ray, and optical data. Main phase of this event was not observed with Yohkoh satellite, and only decay phase was observed. We want to use any data from SXT.
Update 08-Apr-96
Recently, we were able to get stable HXT images by applying the pixon method to the June 28 event and also to the 1992 November 2 event. Now, I am writing a paper regarding microwave and X-ray observations for both events. So, I have already reached the goals and want to close the project.
Hard X-Ray Cusp and Microwave Flare-Loop Sources in Solar Flares
Hiroshi Nakajima, Thomas R. Metcalf, Ken'ichi Fujiki, and Sharad R. Kane
Abstract
We study two intense, extended microwave and hard X-ray flares which occurred slightly behind the west limb on 1992 June 28 and November 2, using data from the Nobeyama Radioheliograph at 17 GHz and the Hard X- Ray and Soft X-Ray Telescopes onboard the Yohkoh. Our analyses are focussed on phases which were simultaneously observed with both the Nobeyama Radioheliograph and the Yohkoh Hard X-Ray Telescope, i.e., the initial phase before the main phase for the June 28 event and the main phase just after the peak time for the November 2 event. We show that: (1) For both events, the major hard X-ray source at 23-33 keV and 33-53 keV is located in the cusp region, considerably higher than the corresponding soft X-ray flare-loop, while the 17 GHz microwave source, where electrons at energy > several hundreds keV mainly contribute to the 17 GHz emission, nearly coincide with the soft X-ray flare-loop in extent. (2) For both events, the electron spectrum, which is derived from the microwave spectrum, is significantly harder than the electron spectrum at energy < 100 keV, which is derived from the hard X-Ray spectrum. (3) The time profile of total intensity at 17 GHz is delayed by about 25 s with respect to that of the hard X-rays for the June 28 event. The above observational results give evidence that
lower-energy (< 100 keV) electrons with a softer spectrum are accelerated at the cusp region, while higher-energy (>> 100 keV) electrons with a harder spectrum are accelerated at the top of the soft X-rayloop and trapped. Probably, the lower-energy electrons accelerated in the cusp region are transferred by a fast reconnection outflow to the top of the soft X-ray loop and further accelerated to higher energy at the shock located at the soft X-ray loop top where the fast reconnection flow impinges on it.
Update 17-Jul-95
We have encountered a difficult problem, how to get stable HXT images from this event with large sources. We could not get stable images by the present MEM method. Recently, however, Dr. T. Metcalf proposed the pixon method as one of HXT imaging methods. He and I tried to apply the pixon method to the 1992 June 28 event, and are reaching a conclusion that the pixon method can give us more stable images for this event which consists of extended sources. I need a few months for confirmation of usefulness of the pixon method for this event. Then, I can reach the goal to write a paper. The content of the paper was already given in the last updated version of this proposal.
Update 29-Oct-94
Collaboration: H. Nakajima and Radioheliograph Group, T. Sakao, T. Kosugi, T. Takakura, SXT Group, and Flare Telescope Group
Present Status of Analysis : This event is still in analysis phase. We encounter a difficult problem, how to get stable hard X-ray images from this event which have large sources with nearly 2 arcmin in size. Up to now, we have the following results. (1) Pre-flare heating, which was observed since more than 2 hours before the onset of the flare, increases with time, and finally the flare occurs.
(2) The onset time of soft X-ray emissions ( before 0440 UT for GOES and 0441 UT for HXT/L ) is earlier than that of the radio emission ( 0443.4 UT for 17 GHz total flux ).(3) In the pre-flash phase: Radioheliograph: The flare starts at the position near and lower than the pre-flare heating source, and expands to the north along the limb, resulting in formation of an elongated source. HXT/L: The flare started in the northern part of the microwave elongated source and expands to the larger region including the elongated microwave source. Overlays of the HXT/L and M1 sources with the corresponding microwave source : Some intense part of HXT/L source and the major intense part of the HXT/M1 source seem to be located above the intense part of the elongated microwave source in the later time of the pre-flash phase. Image quality of the HXT/M1 source is not so good (this is a problem). Overlays of the HXT/L and M1 source with a SXT source around the beginning of the post-burst increase phase: Some intense part of the HXT/L source and the major intense part of the HXT/M1 in the later time of the pre-flash phase are located above the intense part of a SXT source at the beginning of the post-burst increase phase. Overlay of the microwave source in the later time of
the pre-flash phase with a SXT source in the beginning of the post-burst increase phase: Extension of both sources is almost the same if comparison is done at half power level of the peak intensity.(4) In the flash phase: The intense microwave source coincides in position with the polarized component of the microwave source at the onset of the pre-flash phase. (5) In the post-burst increase phase: Evolution of post-flare loops is observed for SXT and H-alpha images. The corresponding microwave source is also observed. The brighter region in the SXT image has the larger emission measure and lower electron temperature. The region above the brighter region in the SXT image has the higher electron temperature. (6) Microwave spectra: The spectra are characterized by a narrow bandwidth, low turn-over frequency around 5 GHz, a soft spectrum index of 4.1 in the pre-flash phase, while characterized by a broader bandwidth extending well in the millimeterwave frequency, a higher turn-over frequency around 9 GHz, a harder spectrum index of 2.1 in the flash phase. Both spectra can be explained by gyrosynchrotron emission due to nonthermal electrons.
Some of important conclusions derived from this event are as follows. (1) In the pre-flash phase, the HXT/L source is located in the higher and larger region than the SXT post-flare loops which is identified with the brighter region in the SXT image in the post-burst increase phase, and HXT/M1 source is located above it. The microwave image (nonthermal component) in the pre-flash phase has a structure similar to the SXT (microwave) post-flare loops, that is, the brighter regions surrounded by broad, weak emissions roughly coincide with each other. These facts suggest that energy release and particle acceleration occur in the larger and higher region than the SXT (microwave) post-flare loops and that the structure similar to that in the post-burst increase phase, that is a cusp-type structure, is set up in the pre-flash phase. (2) The brightest position of the microwave emission at 17GHz in the flash phase comes from the same location as the polarized source at the onset of the pre-flash phase. (3) The flare starts with heating which is followed by electron acceleration. (4) Pre-flare heating must have some relationship with trigger of the flare. What mechanism works is not known from the present analysis. Farther study is needed.
Update 11-Feb-94
Collaboration: H. Nakajima and Radioheliograph Group, T. Sakao, T. Kosugi, and HXT Group, SXT Group, and Flare Telescope Group
Present Status of Analysis : Recently, some new findings were added to the old findings in this event from overlays of the HXT/L,M1 source in the pre-flash phase ( weak enhancement phase before the flash phase) with the SXT source around the begginning of the post-burst increase phase. ( Note that the flash phase is lack of the Yohkoh observations. ) They show that some intense part of the HXT/L source and the major part of the HXT/M1 source extends well above the intense part of the SXT source at the begginning of the post-burst increase phase. This fact suggests that particle acceleration occur in the region higher and larger than the SXT post-flare loops. Now, we speculate that a large-scale magnetic reconnection structure, that is a cusp-type structure, has been already set up in the pre-flash phase of such an intense long-duration event and that particle acceleration occurs in this structure. So, we are still analysing this event extensively and are preparing to write a paper.
Update 31-Jul-93
Collaboration: H.Nakajima, Nobeyama Radioheliograph Group, and others
Analysis and Result: A long-duration X-class flare on June 28, which was accopanied with intense microwave and millimeterwave emissions, was well observed with the Nobeyama Radioheliograph and for a part of the event, with the YOHKOH. Anaslysis of the event was almost completed. We are now accomplishing to write a paper. Main results are as follows. The magnetic field configuration in the flare region consists of a big coronal loop and an arcade adjacent to one of the footpoints of the big coronal loop. Pre-flare heating is observed high in the corona, along the big coronal loop. A weak non-thermal emission is initiated at the intersection of one of the footpoints of the big coronal loop and a part of the adjacent arcade, and expand into the whole arcade region. The intense non-thermal emission originates from a part of the arcade.