4. The Yohkoh/SXT data: morphology, pressure and filling factor
Two SXT images, close in time and made in two different filters, allow us to obtain maps of temperature and emission measure. The temperature is obtained directly from the ratio of the brightness in the two different filters. The emission measure is obtained from the ratio:
where and are the pixel luminosity and the expected luminosity per unit emission measure in the k-th filter.
The SXT data are taken in the Al.1 and AlMg filters (Tsuneta et al. 1991). We have combined couples of short and long exposure images (78 and 2668 ms in the Al.1 filter, 168 and 5338 ms in the AlMg filters) to obtain a higher dynamic range. The SXT and NIXT images have been co- aligned in the same way as we did with the magnetograms. In Figs. 2 and 3 we show the selected regions as observed with the SXT. Region A contains a larger and more complex structure than that detected by NIXT, presumably a bundle of adjacent loops, whereas NIXT seems to detect a single structure. The same path as marked on the NIXT image does not seem to mark so clearly a single loop as, instead, it does on the NIXT image. Regions B , C , and E are characterized by smaller bright structures, whose size is of the order of one pixel. Region D does not show any clearly identifiable structure on the SXT images.
We have applied our analysis on all five regions, even on regions A and D which show no clear correspondence with loop-like structures. The same five paths as in the NIXT image have been marked on the SXT images. Since the SXT pixel is 4 times larger than the NIXT pixel; it has been possible to consider strips as thin as done in the NIXT image only for regions A and D ; for the other smaller loops we have considered a minimal strip width of 2 pixels. The strips of loops A and D has also been divided into sectors, so as to have an idea of the temperature and emission measure distribution within the loop. A single sector has been instead considered for the other loops. The detected emission has been integrated in each sector.
The resulting temperature and emission measure distributions for regions A and D are practically uniform along the selected paths within the uncertainties, but, since no loop is clearly marked, we do not expect to obtain any well-defined trend, comparable to those typical from loop models.
In order to evaluate the loop pressure from the SXT data, we assume the temperature averaged on the marked strips, as the loop average temperature (of the plasma emitting in the SXT band). We then consider the same grid of model loops as those used for the NIXT analysis (Sect. 3) and compute their average temperature from the ratio of the total loop emissions synthesized in the two selected SXT filters.
Then we take as loop pressure the base pressure of the loop model which yields . It is worth noting that the average temperature considered here is not the same as the temperature at the loop apex, the same used in the scaling laws for hydrostatic loops, but instead it is systematically lower, being an average including plasma at lower temperatures.
From the ratio of the total detected loop emission to the total model emission (in one filter) multiplied by a cross-section diameter, we then derive the volume filling factor of the plasma emitting in the SXT band. The cross-section diameter has been determined as equal to the SXT FWHM at one of the footpoints, as done in the analysis of NIXT data, for the loops well-defined in SXT (B , C , E ), and directly as equal to diameter as determined in the NIXT analysis for loops A and D .
Table 2 shows the temperature, the emission measure per pixel, the cross-section diameter (in pixels), the loop pressure and the filling factor as derived from the method above.
Table 2. Loops as imaged by Yohkoh/SXT Notes: Loop average temperature Average emission measure per pixel Loop cross-section diameter (in pixels) Pressure obtained from SXT data on the basis of loop models Volume filling factor of the loop plasma observed in the SXT band The analysis has been carried out (on the same strip as marked in the NIXT image) even if the loop is not clearly identified in the SXT image Not a proper plasma loop filling factor.
(where fully ionized gas has been assumed), by making an assumption on the emitting volume V.
We have verified that, even assuming conservatively high volumes, the pressure values derived from Eq. (2) invariably lead to unreasonably high SXT filling factors, i.e. much larger than one, for all our selected loops. Indeed this approach is not self-consistent since the assumptions on volume do not take into account the loop filamentation.
As is clear from Table 2, all the pressure values derived from SXT are systematically higher (a factor 3 to 9) than those derived from NIXT. This is essentially due to the different passbands of the two instruments which select plasma at different temperatures (and therefore pressure). Even for the large region A the pressure is more similar to that of an active region loop. As occurs for NIXT, the compact loops have higher pressure, typical of core active regions. The filling factors are instead all considerably higher that those obtained from NIXT. The values for the compact structures B , C , E are comprised in the range 0.1-1. Such values, not far from unity, indicate loops filled up with plasma emitting in the SXT band. Indeed, the real filling factor may be closer to unity than the obtained one, because the relatively limited resolution of the SXT image forces us probably, especially for small loops, to consider volumes outside the loop.
The value for region D is compatible with unity. Since this structure is not well-defined in the SXT image, we will not discuss it any further. Instead the very high value obtained for region A deserves some comments. As discussed previously, the boundaries which clearly identify a single loop structure on the NIXT image do not identify it as clearly on the SXT image. In the SXT image the region appears relatively less bright than the immediately surrounding region. At a first glance, one may think that the different pass-bands make the loop footpoints appear brighter in the NIXT band and the apex in the SXT band. If this were the case, we would still have obtained a realistic value of the filling factor. Indeed, it is possible to obtain filling factor values comparable to, or smaller than, unity only if we consider a loop or bundle of loops considerably smaller/shorter than that identified in the NIXT image (as shown in Fig. 2). These loops therefore are not the same loops as the NIXT loop: SXT is detecting a structure or structures different and distinct from that visible in the NIXT image and falling in the same field of view .
© European Southern Observatory (ESO) 1999
Online publication: February 22, 1999