[30] One can calculate the temperature and emission measure of the emitting flare solar region on the basis of the registered X-ray emission at the Earth's orbit in the scope of the "quasi-thermal'' model of the energy spectrum, that is, assuming the bremsstrahlung mechanism of the X-ray emission from the optically thin thermal plasma. In this case the intensity of the thermal X-ray emission of the plasma from the flare region of the solar atmosphere at a distance of one astronomical unit is determined by the expression [Crannell et al., 1978]
|
where I(E) is the intensity of the X-ray emission in the units photon cm-2s-1keV-1, E and T are the energy and temperature, respectively in the keV units, EM= ne2V is the emission measure of the emitting electrons in the units 1045cm-3, ne is the electron density, V is the volume of the emitting region, and gff(E, T) is the "Gaunt'' factor.
[31] An approximate value of the Gaunt factor with the accuracy up to
2% is given by the expression
gff(E, T) 
0.90(T/E)a, where
a=0.37(30 keV/T)0.15 [Matzler et al., 1978]. Therefore we have
fairly accurate (with the accuracy up to 2%) expression for the
intensity of the X-ray emission:
[32] The measured count rate
mj for the time interval
Dt for each out
of 5 energy channels [E
j,
Ej+1] ( j =1,2,...,5) of the IRIS
spectrometer is presented by the integral expression as a function
of the intensity of the source emitting within the energy range
[E1, E2] and response function of the device
F(E,E), forming the
redundant system of nonlinear equation relative the
T and EM
parameters [Hoyng and Stevens, 1974]:
 |
[33] Further, the equations of this system are linearized, and the system is solved by the sequent approximations method relative unknowns T and EM. As an initial approximation, the evaluation of these parameters by more simple method and less accurate expression for the emission intensity is used [Crannell et al., 1978]:
 |
Here Gaunt factor is approximated by the expression gff(E,T)=(T/E)0.4 with the accuracy of 20%.
|
| Figure 4 |
[35] Figure 4 visually shows that at the growth stage of the flare there occurs an intense heating (Figure 4c, an increase in the temperature values) in the relatively compact region (Figure 4b, the values of the emission measure of the flare at the growth stage do not increase). Then, in the moment of the event maximum, the temperature evaluations reach their maximum under relatively unchanged values of the emission measure. The latter manifests a maximum heating of the emission region without a significant change in its volume. Finally, at the concluding stage of the event decay, there occurs cooling of the emission region with its gradual expansion or with evaporation process (see, e.g., Liu et al., [2006], Figure 4c). The values of the temperature gradually decrease and the values of the emission measure increase (Figure 4b).
[36] The modified spectral analysis method was applied to the calculated values of the emission measure and temperature of this flare in order to find a periodical structure (Figures 4e and 4f) and to compare it with the similar structure of the count rate (Figure 4d).
[37] In the temperature series, there presents the most powerful component 19.25 s which is not observed in the count rate spectrum and is much weaker in the spectrum of the emission measure (19 s). In the spectrum of the count rate the 9.5 s harmonic dominates which is much weaker in the emission measure spectrum (8.75 s) and is absent in the temperature spectrum.
[38] One may suppose that for the discussed flare pulsed "injection'' of particle beam into the region of radiation with the period 9.5 s during the flare manifests itself as a quasiperiodic "heating'', or "increase-decrease'' of the emitting region as a 19-s oscillation.
Powered by TeXWeb (Win32, v.2.0).