Weekly Notes from the Yohkoh Soft X-Ray Telescope

(Week 14, 2002)

Science Nugget: April 5, 2002


WBS flare statistics

Introduction

A solar flare emits huge amount of energy (typically 1027 erg/sec in total), in a wide range of wavelengths; from radio, visible, EUV, X-rays, and sometimes reaches up to the gamma-ray region. The amount of energy carried by the energetic particles (both electrons and ions) that produce the hard X-ray and gamma-ray bursts is known to be comparable with the total energy of a flare. This is why understanding the formation mechanism of energetic particles is important for understanding the whole flare mechanism. This is generally the case for flares of all classes. Although the first gamma-ray detection in solar flares was reported in the early 1970's by the OSO-7 satellite , full-fledged studies only began in the 1980's with the Solar Maximum Mission (SMM) , which observed nearly 150 flares with gamma-ray emission. Gamma-ray radiation in solar flares is thought to come from several different mechanisms. Since there is no space to say much about them, the following just lists the major components found in flare spectra.

According to the SMM results, the continuum component due to Bremsstrahlung is dominant in limb flares, whereas the flares with strong neutron capture line emission often occurred on the disk. The former results from the fact that the Bremsstrahlung mechanism, in the gamma-ray region (meaning MeV photons and relativistic electrons), radiates the mainly in the forward direction. This would be from corona toward the solar surface, for coronal acceleration; this biases against detection at disk center. On the other hand the neutron capture line results from energetic particles penetrating into the deeper and denser layer of the photosphere. The radiation from these deeper layers, conversely, is better observed on the disk where the line-of-sight depth is shorter than at the limb.

WBS, the most minor instrument of Yohkoh -- ?

Although they have seldom been mentioned in our science nuggets, Yohkoh carries two spectrometers for high-energy X-rays and gamma-rays (BCS and WBS), in addition to the two well-known imaging telescopes (SXT and HXT). In contrast with the BCS, which observes detailed spectra in narrow wavelength ranges around four emission lines appropriate for solar flares, the WBS (Wide Band Spectrometer) observes wider ranges of spectra from 2 keV all the way to 10 MeV. This wide range of energy detected with WBS is covered by three types of spectrometers, as follows.

One can't tell much from these pictures, of course, but the GRS consisted of large BGO (bismuth germanate) scintillation counters, just about the densest and best type of detector for high-energy radiation. In this respect (stopping power for these difficult photons) the GRS instrument rivals the more modern RHESSI.

For more details about the WBS instrument, see here.

Although one can come across some descriptions and displays of the BCS data in previous nuggets, (e.g. the 16-Jun-2000 edition), those of the WBS are harder to find. In fact, there haven't been any yet! For one reason, spectral data are generally less visually appealing than image data and thus many authors may not want to use them. But the main reason must be the fact that, thanks to Yohkoh's long life, a good number of flare data have been newly accumulated during the rising through maximum phase of this (23rd) solar cycle. The nugget writers simply have not noticed these good data, for which almost no analysis has yet been done!

Recently an able data-analyzer, Y. Matsumoto of the University of Tokyo, carried through the troublesome work of WBS/GRS data calibration and investigated properties of the gamma-ray flares observed with Yohkoh. In her thesis work in 2001, she at first made statistical studies of the whole data set. She then picked two specific flares (18-Aug-1998 and 27-Oct-1991) and made a closer analysis to derive some interesting results; she plans to report these results by herself in a future nugget. This nugget is a brief review of the first part of her work (the statistical study), and it turns out to be the first memorable nugget featuring the WBS!

Identification of the Yohkoh gamma-ray flares

To discuss the gamma-ray flares observed with Yohkoh/GRS the first thing to do is to select flare events from the contiguous stream of GRS data. This is not a trivial task, and she proceeded as follows. Out of 2,359 flares (events that triggered flare mode) detected with Yohkoh/HXT from 1991 October to 2001 June, she first selected 157 events in which the initial phase of the flare was observed with sufficient intensities (counts > 100). For these 157 well-observed (ie, imaged in hard X-rays) flares, she looked for the corresponding GRS counts and found substantial amounts of gamma-ray emission in 38 events (see the huge table transcribed by the undersigned). Further, among these 38 events, she selected 14 events which have enough GRS counts and were well-observed enough to get spectral fits. To save space, the huge table only shows the photon index due to the power-law fit for the continuum, and the equivalent width of the neutron-capture line (2.223 Mev).

Statistical properties

The figure below, taken from Matsumoto's thesis, contains a map and two histograms showing the location of the 38 flares detected with GRS (red crosses and columns) together with 144 flares well-observed with HXT (purple diamonds and columns).

From the above table and figures, it turns out that the gamma-ray flares tend to be observed towards the limb region. This supports a result obtained with SMM. The highlighted cells in the huge table show the locations of the flares which occurred within ranges of absolute heliographic longitude, 60 through 70 degrees (yellow), and 70 through 90 degrees (pink). One can see that nearly half of the events (18 out of 38) came from the longitude larger than 50 degrees, in spite of the fact that the area of those regions is only one third of the whole hemisphere. Is this statistically significant?

Turning to the flares showing the 2.2 MeV neutron-capture line, Matsumoto found that the Yohkoh gamma-ray flares show the same tendency as that seen in the SMM data: the line is less well detected in the limb flares. In the case of the Yohkoh flares, there were three events which have strong 2.2 Mev line emission (with equivalent width larger than 1000 keV), and they were all observed onn the disk at absolute longitudes less than 20 degrees. The relation between the equivalent width and the absolute longitudes is shown in the plot below, which is also taken from her thesis.


Conclusions

Here we have introduced the first statistical study of the Yohkoh gamma-ray flares, as reported by Y. Matsumoto in her PhD thesis for the University of Tokyo. Although the results essentially support those of SMM rather than being radically new findings, we think that her analysis is a brilliant one since nobody else had tackled with the WBS data so intensively during Yohkoh's ten-year history. The equivalent-width result in the figure above may be new, actually. Is it statistically significant? Finally, we repeat that the main part of Matsumoto's analysis is not the statistical study reported here, but the detailed study of a few particular events. The latter contains some interesting results, for which we should look forward to the future opportunity of Dr. Matsumoto's maiden Yohkoh science nugget.

[Topical index] -o- [Chronological index]

April 5, 2002

A. Takeda (takeda@isass1.solar.isas.ac.jp) with thanks to Y. Matsumoto and H. Hudson