Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 363, 279-288 (2000)

Previous Section Next Section Title Page Table of Contents

3. Results and discussions

The results of our measurements are presented in Table 2. The important results of this study are that for all 3 brightness levels


Table 2. Results of the measurements of INBPs and IN magnetic elements. No. of cells: 106

  1. about 60% of the INBPs with strong emission spatially coincide with IN magnetic elements of flux density [FORMULA] 4 Mx cm-2,

  2. about 25% of the INBPs with strong emission lie over areas where opposite polarity IN magnetic elements with flux densities [FORMULA] 4 Mx cm-2 meet, merge and cancel resulting in fields with values [FORMULA] 4 Mx cm-2, and

  3. the rest, 15% of the INBPs lie over areas with fields [FORMULA] 4 Mx cm-2, which is the noise level set by us for the magnetic scans used in the present analysis.

Our study shows that if those magnetic elements of flux density [FORMULA] 4 Mx cm-2 alone are used, the coincidences are 60% which may not appear as a striking correlation. In the case of bipoles, although the fields are [FORMULA] 4 Mx cm-2, the INBPs associated with them show strong emission. Nindos & Zirin (1998) have many instances in their time series where magnetic elements of opposite polarity move towards each other, merge and form bipoles accompanied by significant increase in emission in the associated INBP. Although such an increase in emission is a magnetic-field-driven phenomenon, since the measured value of the field at the site of the merging bipole is inevitably low (and hence will not be recorded by magnetographs however precise the measurements be) such cases would be left out in the search for coincidences of excess emission with excess magnetic fields. It is our opinion that the cases of excess emission at the sites of the bipoles as well as those at the sites of fields [FORMULA] 4 Mx cm-2 are both instances of magnetic-field-related emission although they might possibly differ in the details of the precise role of the magnetic fields in producing the emissions. Hence if the cases of bipoles are not taken into account as coincidences, then the correlation will drop down to that extent. Lites et al. (1999) looked for the spatial correlation using their H[FORMULA] brightness modulation and magnetic flux density space - time charts and conclude that there is no obvious spatial correlation between the bright H[FORMULA] locations and magnetic field enhancements. Their analysis does not take into account the cases of emission associated with bipoles and hence they find "no obvious association". According to our analysis the coincidences of INBPs with magnetic elements without counting the bipole cases are only 60% which is not a high figure. What we have done in addition (and what they have missed) is to add the cases of bipole associated K2V emission and show this increases the percentage of coincidences to 85%. Thus our present analysis adds a new dimension in finding a solution to this problem by emphasising the need to include the cases of bipole associated emissions while looking for coincidences (or otherwise). We have estimated the percentage of chance coincidence by looking for coincidences of INBPs in 24 cells with IN magnetic elements in 24 cells in a different region on the sun. We find that the mean chance coincidence is about 28%, which is less than half the percentage of true coincidences.

In Fig. 2 we present the scatter plot of the brightness of INBPs vs magnetic field density, of all INBPs that show coincidences with IN magnetic elements. Similar plots for INBPs [FORMULA] 1.5 [FORMULA] and INBPs [FORMULA] 1.7 [FORMULA] (not shown here) appear very similar to Fig. 2 except that the density of data points is correspondingly less. A large part of the scatter is due to the 3-min oscillations, which we have not filtered as our data do not form a time sequence suitable for filtering. This scatter does not permit us to obtain a correlation between the brightness and magnetic fields even if it exists on the sun. The vertical line in Fig. 2 represents the 4 Mx cm-2 limit, which is the noise level set by us for the magnetic scans. The INBPs to the left of this line contain those associated with bipoles as well as the unresolved emission features with low fields. The population density of the INBPs drastically falls above the 10 Mx cm-2 level, whereas Sivaraman & Livingston (1982) noticed an association with magnetic elements in the range of 10-20 Mx cm-2, the maximum fields reaching 70-80 Mx cm-2. Whether the fields of IN magnetic elements are intrinsically low during the solar minimum and climb to higher values during solar maximum may be a point of interest for a future investigation.

[FIGURE] Fig. 2. Scatter plot of the maximum brightness of the INBPs vs the maximum absolute value of the magnetic field of the cospatial IN magnetic elements over an area of 2 [FORMULA] 2 pixels. The vertical line represents the 4 Mx cm-2 limit. The data points to the left of this line contain both the INBPs that coincide with bipoles (with fields [FORMULA] 4 Mx cm-2) as well as those with weak unresolved fields [FORMULA] 4 Mx cm-2 (see text for details).

We now proceed to provide additional evidence in support of our results from a variety of observations reported in the literature by other workers, particularly those from the studies on innernetwork magnetic fields such as flux distribution, motion patterns, lifetimes and finally on their relation to INBPs, in recent years from Big Bear Solar Observatory using high-quality deep magnetograms and co-temporal K-line filtergrams.

  1. The mean number of IN magnetic elements ([FORMULA] 4 Mx cm-2 level) in many cells in our magnetic images is 17. It is encouraging to note that the mean number of IN magnetic elements per cell (above the noise level) from the BBSO deep magnetograms is 20 (Wang et al., 1995). This equality of the numbers of INBPs and IN magnetic elements has been used by Kalkofen (1996) for postulating on the possible spatial correspondence between the two structures. The K-line observations by Sivaraman & Livingston (1982) used a band pass of 1.1 Å. This wide passband itself will impose a high threshold and cut off a substantial number of INBPs. Thus, there would be an excess of IN magnetic elements (as no threshold was applied to the magnetic scan images) within each cell over the INBPs and, so, while looking for coincidences, it is to be expected that many IN magnetic elements would be found that do not have corresponding INBPs. This is what they noticed.

  2. Nindos & Zirin (1998) find that the INBPs associated with IN magnetic fields possess horizontal velocities of the order of 1 km s-1, while Zhang et al. (1998a) find from their deep magnetograms that the IN magnetic elements move within the network with horizontal velocities of almost of the same order.

  3. Damé (1984), from an analysis of a 52-min long time sequence of 1.2 Å band pass K-filtergrams (obtained at the DST, Sacramento Peak), concluded that the INBPs recur mostly at the same location within the cell for this duration and they do not occur at random. He interprets this as suggesting their association with an underlying structure which possibly could be magnetic flux tubes. Since the IN magnetic elements have much longer lifetimes (Zhang et al., 1998c) the possibility of this association cannot be ruled out from lifetime considerations.

  4. According to Wang et al. (1995) most IN magnetic elements emerge as clusters of mixed polarities from an emergence centre within the network. Zhang et al. (1998b) have studied the evolution of the IN magnetic elements using a 10-hour long deep magnetogram time sequence at BBSO and arrive at a similar conclusion. We have made a detailed examination of the excellent K2V spectroheliogram near the limb obtained by Bruce Gillespie on September 17, 1975 with the spectroheliograph set up at the East Auxiliary facility that existed at that time at the then McMath telescope. We find instances of the cluster formation (like the rosette in [FORMULA]) within the network cells in Gillespie's spectroheliogram, that are similar in appearance to the clusters in the IN magnetograms described by Wang et al. (1995) and Zhang et al. (1998b). In Fig. 3 we show one example (frame 6) of these clusters from Gillespie's spectroheliogram.

  5. Nindos & Zirin (1998) and Zhang et al. (1998b) have seen bipoles within the network and according to Martin (1988) the intranetwork fields might consist of several small loops. We have identified in Gillespie's spectroheliogram instances of short fibril structures or small loops running from one INBP to its neighbour. In Fig. 3 (frames 1 to 5) we show five examples of such short fibrils (or loops). The fibril structure or the loop provides a strong evidence for the INBP - IN magnetic field association. Such fibril structure might be a common feature with all opposite polarity elements that are close to each other, but only a spectroheliogram near the limb (curved geometry), and of exceptional quality, can show these with such clarity as Gillespie's.

[FIGURE] Fig. 3. Reproduction of six regions from the K-line spectroheliogram by Bruce Gillespie. Examples of short fibrils (or loops) connecting adjacent INBPs are indicated by arrows in frames 1, 2, 3, 4 and 5.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: December 5, 2000