The Cause of 30 May 1998 MOSS Regions

(Last modified on May 4, 1999. Please send comments and suggestions to Xuepu Zhao)

1. Introduction

The moss feature is a novel emission pattern first identified in high-resolution Fe IX,X 171 A images of solar active regions observed by TRACE on 30 May 1998 (Berger et al., 1999). The feature is characterized by small-scale, inhomogeneous, bright elements surrounding dark inclusions, and appears as a "spongy" or "moss" occurring over certain magnetic plages outside of sunspots. The bright network is confined to layers approximately 1000-3000 km thick at heights between about 2000-4000 km above solar surface. The thin plasma layer resposible for the 171 A emission is at temperatures of about one million degrees.

The moss region that contains moss elements covers about 20,000 km and is relatively static. Based on the one-to-one correspondence of 171 A moss regions with soft X ray bright regions, it is suggested that the energy source of the moss emission is the downward thermal conduction along coronal magnetic field lines to the footpoint of hot loops (Berger et al., 1999).

The aforementioned interpretation for the low-lying emission pattern implicitly assumes the existence of a significant gredient in temperature along magnetic field lines that pass through the moss feature. It is thus expected that the model may be tested by searching for the counterpart of the new emission pattern in other emission lines.

Only some magnetic plages are associated with moss regions. What is the characteristics of the moss-associated magnetic plages? Comparing the difference between the moss-associated and moss-less magnetic plages may provide some clue for understanding the cause of moss regions.

The top panel of Figure 1 shows the moss regions (the bright spongy-like structures) in the full FOV TRACE 171 A image observed at 14:24:34 May 30, 1998. The bottom panel is the part of the image bordered by the dashed square. Overlaid on the panel is the contours (red outlines) for intensity of 250. Moss regions A and C are the same regions as studied by Berger et al (1999). As shown in the MDI magnetic images (see Figure 7), B area is located over a magnetic plage but devoid of 171 A emission. We first searching for in the next section the counterpart of the moss region A in other emission lines formed in the corona, transition region and chromosphere, and then examine the field configuration in A and B in section 3.

2. The counterpart of moss regions in other emission lines

According to the large-scale loop model of the moss, the thin plasma layer resposible for a specific emission line is a segment of the loop, located at a level with a specific value of temperature. If it is the case, similar bright regions are expected to be shown in other emission lines corresponding to other temperatures along the loop. By comparing the size and location between the 171 moss region and its counterpart in other lines, the large-scale loop model may be checked if the time variation may be neglected. The 30 May 1998 moss regions over the NOAA Active Region 8227 has been continuously monitored by the TRACE 171 A bandpass for near ten hours (7:50 -- 17:40). It provides an opportunity to obtain the SOHO EIT images observed at the time nearly simultaneously with the TRACE observations, though no TRACE data are available in other emission lines. Figure 2 displays four EIT full disk images available on 30 May 1998 with the wavelength, emission temperature and observational time showing on the top of each panel. The dashed square in each panel corresponds to the full FOV TRACE image as shown in Figure 1. The EIT image within the area bordered by solid squares and the corresponding TRACE 171 A images are shown in the right and left columns of Figure 3, respectively. It is shown that all four EIT images corresponding to different temperatures in coorna and transition region (see the top of each panel) have bright region with similar shape and size to the moss regions in the TRACE images.

The counterpart of the moss region can also be identified even in the chromosphere. Figure 4 and Figure 5 display the 10830 A and 6563 A solar images observed at Kitt Peak Observatory and Big Bear Observatory, showing the existence of the counterpart of the moss region in the chromosphere. There is a filament between the moss-associated and moss-less magnetic plages. The filament may also be partly identified in the EIT 304 A images (Figure 3)

The fact that the interested moss region can be identified in emission lines having temperatures from chromosphere to corona supports the large-scale loop model for the moss region.

To see the possible time variation of the TRACE 171 A moss and to compare it with its EIT counterpart, the moss in the area bordered by dotted rectangle in Figure 3 are examined. Figure 6 displays images within the dotted-rectangle area with the outline of the moss region at 13:03:27 (in the top-left panel) overlaid on all panels, showing the negligible time varition. It is difficult to find out the systematic difference between the TRACE 171 moss and the EIT 284 A, 195 A and 304 A moss. Using the TRACE data in 284 A and 195 A may help to recognize the difference.

3. The field lines in moss regions

The top panel of Figure 7 displays the full disk magnetic image observed by SOHO/MDI at 14:24:30 on 30 May 1998. The observation time differs from the moss shown in Figure 1 only by 4 seconds.The white and blue areas are magnetic plages with polarities of away from and toward the Sun, respectively. The bottom panel is the part of the image bordered by the dashed square. Overlaid on the panel is the outlines (white curves) of moss regions and other bright regions (see the bottom panel of Figure 1). As shown in the area bordered by the dotted rectangle, there are two kinds of magnetic plages, though their strength distribution is almost identical. One is moss-associated, and the other is moss-less.

Coronal loops are supposed to be the exhibition of magnetic field lines. In the aging and old magnetic plages, the field lines may be approximately computed using the potential field-source surface model (Sakurai, 1989). It is well known that large-scale patterns on the surface are manifest by actual physical connections in the corona, covering immense distances. The way with which the two kinds of magnetic plages connect with other magnetic features surrounding them may show the difference bewteen them.

Figure 8 displays the contours (orange and blue for polarity of away from and toward the Sun) of line-of-sight magnetic field strength, the computed closed field lines (green lines) and the contours of TRACE 171 A intensity of 250 (white outlines). Most, if not all, of the field lines within the white outlines over the blue plages are connect with the red plages with stronger field strength. The computed field lines ending to the quiet region are devoid of bright emission, though they may start from strongest magnetic plages.

Figure 9 shows the computed field lines within the dotted rectangle (see Figure 7). The field lines within the red plage connect with adjacent quiet regions. There is no connection of the red plage with any blue plages. Figure 10 is a part of the Figure 5 that contains the moss regions. The right H_alpha plage over the red magnetic plage indeed exhibit small horizontal lines connected with adjacent regions. There is a filament between the moss-associated and moss-less magnetic plages.

4. Summary

Bright emission regions without loops have previously been noted in the normal-incidence soft X ray (Peres et al., 1994) and SOHO EIT solar images (Moses et al., 1997). By means of hydrostaic models of coronal arches, the emission pattern in X ray images can be explained by the characteristics of high pressure large-scale loops having a thin region of high surface brightness at the base (Peres et al., 1994). However, the unresolved bright region in SOHO EIT solar images are explained as small-scale loops (see Figure 10 of Moses et al., 1997).

The cotemporal counterpart of the moss region observed on 30 May 1998 has been recognized in various emission lines from coronal (soft X ray, 284 A, 195 A, 171 A), transition region (304 A) and chromospheric (10830 A and 6563 A) temperatures. The result strongly supports the large-scale loop models and the assumption that there exist a significant gredient in temperature along the large-scale loops. This may provide a method to obtain the temperatrure distribution along the loops if the shape and size of moss regions can be rather accurately determined from observations. Because of the very large thermal conductivity along the field lines the existence of the significant gredient in temperature may suggest that there are two types of plasma loops. The first type represents regular thin loops formed by single isolated flux tubes. This type of loops may be visible in some of emission lines due to the nearly homogeneous temperature distribution along the tubes. The second type loops are thick and consist of several magnetic flux tubes which are interlaced and woven because of the shuffling and swirling of the footpoints of the field by subphotospheric convection. The thermal conduction in the tube complex may not be as effective as in the thin tube. The segment of the second type is observed as "moss" in the TRACE images.

Moss regions are spatially correlated with certain magnetic plages in the photosphere, though the details of this relationship are complex on the small scale (Berger et al., 1999). If the loop that passes through moss regions is a flux tube complex, as suggested above, the field configuration in the complex must not be potential. However, there are loops that connect some moss-associated magnetic plages in the TRACE movies, and the loops are potential-like. Thus to the first approximation, the large-scale connection of moss-associated magnetic plages with other magnetic features and the shape of the loops may be calculated using the potential field model. The calculation has shown that the moss-associated magnetic plages connect with magnetic plages with opposite polarity, and moss-less magnetic plages connect only with the adjacent "quiet" regions with weak field strength. The result suggestes that moss-less magnetic plages are regions of strong horizontal field component, and moss-associated plages are regions of strong vertical magnetic field component. More realistic model is needed to obtain field lines better match the large-scale loops than the potential field lines and to infer the temperature distribution along loops. The horizontal current-current sheet model (Zhao and Hoeksema, 1992) or the similar model developed recently (Gary and Alexander, 1999) may be good candidates, though the uniqueness and the self-consistence of the solution should be discussed.

The 30 May 1998 moss regions occur over magnetic plages outside of sunspots. It is interested to see if all moss regions occur in aging plages and if there always exist a filament between the moss-associated and moss-less magnetic plages. The anwsers to these questions as well as the difference between the moss-associated and moss-less magnetic plages found here may provide strong constrain for developing detailed physical model of the moss region.

Acknowledgments

I would thank T.E. Berger for sending me his excellent preprint and A. Kosovichev for the enthuiastic discussion. The classification of loops mentioned in the last section is cited from his comment.

References

Berger, T.E., B.De Pontieu, C.J. Schrijver, A.M. Title, preprint, 1999

Gary. G.A. and Alexander, D., preprint, 1999.

Moses, D., et al., 1997, Solar Phys. 175, 571.

Peres, G., Reale, F., & Golub, L., 1994, Astrophys. J., 422, 412.

Sakurai, T., 1982, Solar Phys., 76 301.

Zhao, X.P. and Hoeksema, J.T., 1994, Solar Phys., 151 91.