Latitude migration of solar filaments

Transcription

1 Mon. Not. R. Astron. Soc. 5, 6 () doi:./j Latitude migration of solar filaments K. J. Li, National Astronomical Observatories/Yunnan Observatory, CAS, Kunming 65, China Key Laboratory of Solar Activity, National Astronomical Observatories, Chinese Academy of Sciences, China Accepted February 8. Received February ; in original form 9 December INTRODUCTION It is widely believed that the Sun s magnetic field is generated by a magnetic dynamo within the Sun. Solar dynamo models demonstrate that the solar activity cycle displays a recycling of the two main components of the Sun s magnetic field (Parker 955): the poloidal and toroidal components. The solar meridional flow, the flow of material in the meridional plane from the solar equator towards the Sun s poles at the surface and from the poles towards the equator deep inside the Sun to carry the dynamo wave towards the equator, should play an important role in the Sun s magnetic dynamo (Parker 987; Choudhuri, Schussler & Dikpati 995; Durney 995; Hathaway et al. ; Charbonneau 7). Propagation of the toroidal field wave at the base of the solar convection zone is believed to manifest itself on the surface as the migration of the sunspot activity belt toward the equator (the so-called butterfly diagrams: Hathaway et al. ). It is helpful to investigate the latitude migration of sunspot activity for understanding the Sun s magnetic dynamo. Hathaway et al. () eamined the drift of the lkj@mail.ynao.ac.cn ABSTRACT The Carte Synoptique catalogue of solar filaments from 99 March 989 December, corresponding to complete cycles 6, is utilized to show the latitudinal migration of filaments at low latitudes (less than 5 ), and the latitudinal drift of solar filaments in each hemisphere in each cycle of the time interval is compared with the corresponding drift of sunspot groups. The latitudinal drift of filaments obviously differs from that of sunspot groups. At the beginning of a cycle, filaments (sunspot groups) migrate from latitudes of about (8 )withadriftvelocity of about. m s (. m s ) toward the solar equator, reach latitudes of about 5 ( ) yr later at the cycle maimum with a drift velocity of about. m s (. m s ) and halt at about 8 (8 ) at the end of the cycle. When solar activity is programmed into a solar cycle, the difference between the latitudes of appearance of filaments and sunspot groups becomes smaller and smaller. The difference rapidly decreases in the first and last yr of a cycle, but barely decreases at all in the yr after the cycle maimum. For filaments, the latitudinal drift velocity decreases in the first 7.5 yr of a cycle, but increases in the last.5 yr of the cycle. However, for sunspot groups the drift velocity always decreases over the whole cycle. The difference between the latitude drift of filaments in the northern and southern hemispheres is found not to be obvious. The latitudinal drift velocity of filaments differs slightly in the northern and southern hemispheres within a cycle. The physical implication behind the latitudinal drift of filaments is eplored. Key words: methods: data analysis Sun: activity Sun: filaments, prominences. centroid of the sunspot area toward the equator in each hemisphere from 87 and found that the drift rate slows as the centroid approaches the equator, which was also found earlier by Li, Yun & Gu () and Li et al. (a). Hathaway et al. () compared the drift rate at sunspot cycle maimum with the period of each cycle for each hemisphere and found a highly significant anti-correlation: hemispheres with faster drift rates have shorter periods. These observations are consistent with a meridional counterflow deep within the Sun as the primary driver of the migration toward the equator and the period associated with the sunspot cycle (Hathaway et al. ). Solar filaments have been observed since the invention of the spectrohelioscope (Tandberg-Hanssen 995; Heinzel 7). They are prominences projected against the solar disc, displaying themselves as cool and dense clouds in the solar corona (Kiepenheuer 95; Tandberg-Hanssen 995, 998). The shape, fine structure, dynamics and physical properties of solar filaments vary from each other over a wide range of values (Hundhausen, Hansen & Hansen 98; Illing & Hundhausen 986; Heinzel 7; Lin, Martin & Engvold 8). It is always interesting to study the general trend of solar activity by using statistical data such as the Cartes synoptiques of solar filaments (dazambuja 9; Mouradian 998a,b). Downloaded from by guest on September 8 C TheAuthor. Journalcompilation C RAS

2 Latitude migration of solar filaments Figure. Butterfly diagrams of filaments (green dots) and sunspot groups (red dots) from 99 March 989 December. Mein & Ribes (99) used the Cartes synoptiques and found evidence for the meridional circulation and the eistence of convective giant cells by following the filaments used as indicators of magnetic tracers, and it is very important to understand how the solar dynamo works and therefore the eistence of magnetic solar cycles. The phenomenon termed rush to the pole has been confirmed through studying the migration of polar filaments towards the pole at the beginning of solar cycles by many authors (Topka et al. 98; Makarov & Sivaraman 989; Mouradian & Soru-Escaut 99; Shimojo et al. 6; Li et al. 8). Solar filaments are distributed over the whole solar surface, from the solar equator to the poles, and during the whole period of a sunspot cycle (Li et al. 7). Such a distribution is not random, as solar filaments are closely connected with the sites of magnetic fields on the solar disc (Martin 99). The north south asymmetry of filaments in solar cycles 6 is investigated with the use of the Cartes synoptiques (Li et al. ), and filament activity is found to dominate regularly in each of cycles 6 in the same hemisphere as that inferred through sunspot activity (Li, Schmieder & Li 998; Li et al. a). Filaments occur at all heliospheric latitudes and outline the boundary between magnetic fields with different polarities, and this makes them suitable tracers for the large-scale pattern of the weak background magnetic field of the Sun (McIntosh 97; Low 98; Minarovjech, Rybansky & Rusin 998a,b). In addition, study of the occurrence of filaments can help us to better understand the distribution of these fields on the solar disc and their development within a solar cycle and, especially, provide useful insights into the nature of the Sun s magnetic field (Rusin, Rybansky & Minarovjech 998; Rusin, Minarovjech & Rybansky ; Mouradian & Soru-Escaut 99). The study of filaments (prominences) through both individual events and statistical analyses is of importance, and some progress regarding the knowledge of filaments has been achieved up to now, including both morphological aspects and theoretical scenarios of filaments (Anzer ; Engvold ; Heinzel 7; Schmieder, Aulanier & Lopez Ariste 7; Schmieder, Aulanier & Torok 9; Labrosse et al. ; MacKay et al. and references therein). In this paper, we will investigate the latitude migration of solar filaments, focusing on their statistical properties, and further compare their properties with those of sunspots. OBSERVATIONS, ANALYSES AND RESULTS The different solar activity time series analysed in our study are as follows. () The Carte Synoptique solar filaments archive, namely the catalogue of solar filaments, observed at Meudon from 99 March 989 December, corresponding to Carrington solar rotation numbers 876 8, which can be accessed via the NOAA web site. The observations of filaments were mainly shown in maps drawn at Meudon, which were a synthetic representation of solar active regions and filaments (dazambuja 9), and the Meudon maps were complemented with tables of solar active regions and filaments (for datails, see Mouradian 998a,b). The World Data Center A (WDC-A) for Solar Terrestrial Physics has digitized the Carte Synoptiques (Coffey & Hanchett 998). () The observational data of sunspot groups, which come from the augmented Royal Greenwich Observatory (RGO) data set (the RGO data set etends from ; thereafter, the observations are from NOAA) and are available at the NASA Web site. The data set comprises sunspot groups during the period of 87 May 9 March and is updated monthly. The term normal solar activity is usually applied to solar active events with latitudes lower than 5 (Sakurai 998; Li et al. 8), thus here we use filaments with latitudes lower than 5.Fig. shows the positions of both filaments and sunspot groups versus time, namely the familiar butterfly diagrams. The figure indicates how the bands of filaments (sunspot groups) drift toward the equator and how successive cycles overlap during the minimum time of a sunspot cycle. As the figure illustrates, filaments are distributed at ftp://ftp.ngdc.noaa.gov/stp/solar_data/solar_filaments Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6

3 K. J. Li 5 Cycle Cycle 9 Cycle Cycle Cycle 7 Cycle 8 Figure. The monthly mean latitudes (crosses) of filaments in the northern hemisphere in cycles 6 and their corresponding third-order polynomial fits (thick solid lines). The thick dashed lines are the second-order polynomial fits of the monthly mean latitudes of sunspot groups in the northern hemisphere. higher latitudes and within a wider latitude band at a certain time epoch than sunspot groups. It is difficult to divide sunspot groups accurately into solar cycles to which they really belong (Harvey 99) and even more difficult to divide filaments into solar cycles. A criterion for how a solar activity event can be assigned to an active cycle was once proposed by Li et al. (). Here, according to the criterion and with the use of the Carte Synoptique solar filament archive, filaments are divided into individual butterflies. Then monthly mean latitudes of filaments in the northern and southern hemispheres are calculated and plotted in Figs and, respectively. For sunspot activity (sunspot groups or sunspot areas), a second-order polynomial curve could give a satisfactory fit to the monthly mean latitudes of sunspot activity (Li et al. ; Hathaway et al. ), but for filament activity it is a third-order rather than Cycle 6 Cycle 7 Cycle Cycle 9 Cycle Cycle Figure. The monthly mean latitudes (crosses) of filaments in the southern hemisphere in cycles 6 and their corresponding third-order polynomial fits (thick solid lines). The thick dashed lines are the second-order polynomial fits of the monthly mean latitudes of sunspot groups in the southern hemisphere. a second-order polynomial curve that gives a satisfactory fit to the monthly mean latitudes of filaments in the northern and southern hemispheres respectively. The fitting curves are also shown in Figs and. In order to compare the latitude drift of filaments with that of sunspot groups, the best fit to the monthly mean latitudes of sunspot groups in the northern and southern hemispheres respectively in each of cycles 6 is given here in Figs and. As the two figures show, the latitudinal drift of filaments obviously differs from that of sunspot groups: the monthly mean latitude of filaments is higher than the corresponding latitude of sunspot groups in both northern and southern hemispheres at each month of a cycle, but the difference between the two becomes smaller and smaller when the solar activity of a cycle is factored into that cycle. In order to compare the latitude drift of filaments in the northern and southern hemispheres respectively, the unsigned monthly mean latitude of filaments is plotted in Fig.. As the figure shows, the different between the two is not obvious within a cycle for cycles 6, ecept in the case of cycle 9. The unsigned monthly mean latitudes of filaments for all si solar cycles are plotted together in one cycle in Fig. 5 by using the time measured relative to the times of maimum of their corresponding cycles. Also given in the figure is a third-order polynomial fit to these points. As the figure shows, in a statistical view filaments appear at latitudes of about at the beginning of a cycle, move toward the equator, reaching 5 some yr later at the cycle maimum, and then halt at about 8 at the end of the cycle. Similarly, the unsigned monthly mean latitudes of sunspot groups in the si solar cycles are plotted together in one cycle in Fig. 5 using the time measured relative to the times of maimum of their corresponding cycles. A second-order polynomial is used to fit these points, which is also given in the figure. We then know the difference between the two fits as a function of time measured relative to cycle maimum time, which is shown in the bottom panel of the figure as well. Filaments appear at much higher latitudes than sunspot groups do at the beginning of a cycle; when solar activity is programmed into the cycle, the difference between the latitudes where filaments and sunspot groups appear becomes smaller and smaller. The difference decreases rapidly in both the first and last yrof a cycle, but barely decreases in the yr after the maimum of the cycle. Cycle 6 Cycle 7 Cycle Cycle 9 Cycle Cycle Figure. The unsigned monthly mean latitude of filaments in a cycle in the northern (solid lines) and southern (dashed lines) hemispheres respectively for cycles 6. Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6

4 Latitude migration of solar filaments Latitude Difference (degrees) Time (Years from Maimum) 6 8 Time (Years from Maimum) Figure 5. Top panel: latitude of filaments versus time relative to the time of sunspot cycle maimum. The unsigned monthly mean latitude of filaments for each hemisphere and for each cycle is plotted as an individual cross by using the time measured relative to the maimum of that cycle. A third-order polynomial fit to these points is shown by the thick solid line. Similarly, the unsigned monthly mean latitude of sunspot groups for each hemisphere and for each cycle is plotted as an individual cross by using the time measured relative to the maimum of that cycle, and a second-order polynomial (thick dashed line) is used to fit these points. Bottom panel: the difference between the thick solid and dashed lines Figure 6. The latitudinal drift velocity varying with time for filaments (thick solid lines) and sunspot groups (thick dashed lines) respectively in the northern hemisphere in each of cycles 6. The vertical thin dashed and solid lines mark the minimum and maimum times, respectively. In Figs 6 and 7, the variation with time of the latitudinal drift velocity of filaments in a cycle in the northern and southern hemispheres respectively is shown. The latitudinal drift velocity ranges within about.5.5 m s in a cycle, and it differs slightly in the northern and southern hemispheres within a cycle. Also displayed in the two figures is the latitudinal drift velocity of sunspot groups as a function of time in each of cycles 6 and in the northern and southern hemispheres respectively. As the two figures show, the drift velocity decreases linearly with time for sunspot groups in both northern and southern hemispheres in each of cycles Figure 7. The latitudinal drift velocity varying with time for filaments (thick solid lines) and sunspot groups (thick dashed lines) respectively in the southern hemisphere in each of cycles 6. The vertical thin dashed and solid lines mark the minimum and maimum times, respectively. 6, but for filaments it first decreases and then increases within a cycle. Hathaway et al. () investigated the drift of the centroid of the sunspot area toward the equator in each of the northern and southern hemisphere from 87 and found that the drift rate slows as the centroid approaches the equator, ecept for cycle in the northern hemisphere. Fig. 8 shows the general features of the latitudinal drift velocity in a cycle for filaments and sunspot groups respectively, when the unsigned monthly mean latitude of filaments and sunspot groups for each hemisphere and for each cycle is plotted together by using the time measured relative to the maimum of that cycle (see Fig. 5). At the beginning of a cycle, filaments (sunspot groups) migrate from latitudes of about (8 ) with a drift velocity of about. m s (.ms ) toward the solar equator, reach latitudes of about 5 ( ) yr later at the cycle maimum Time (Years from Maimum) Figure 8. The general features of the latitudinal drift velocity in a cycle for filaments (thick solid line) and sunspot groups (thick dashed line) respectively when the unsigned monthly mean latitude of filaments and sunspot groups for each hemisphere and for each cycle is plotted together by using the time measured relative to the maimum time of that cycle (see Fig. 5). Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6

5 K. J. Li Figure 9. Drift velocity varying with latitude for filaments (top panel) and sunspot groups (bottom panel) respectively in each hemisphere in each cycle. The dashed line is for sunspot groups in the northern hemisphere in cycle. with a drift velocity of about. m s (.ms ) and halt at about 8 at the end of the cycle. For filaments, the latitudinal drift velocity decreases in the first 7.5 yr of a cycle, but surprisingly it increases in the last.5 yr of the cycle. The variation of latitudinal drift velocity with latitude is plotted in Fig. 9 for filaments and sunspot groups respectively in each hemisphere in each cycle. For filaments, the drift velocity decreases with latitude within a cycle until they reach latitudes of about, after which it increases with latitude until the end of the cycle. For sunspot groups, drift velocity decreases with latitude in each hemisphere in each of cycles 6 without any eceptions. The variation of latitudinal drift acceleration with time is plotted in Fig. for filaments and sunspot groups respectively; the unsigned monthly mean latitude of filaments and sunspot groups for each hemisphere and for each cycle is plotted together by using the time measured relative to the maimum of that cycle (see Fig. 5). For Acceleration (degrees/year ).. 6 Time (Years from Maimum) Figure. Latitudinal drift acceleration varying with time for filaments (thick solid line) and sunspot groups (thick dashed line) respectively when the unsigned monthly mean latitude of filaments and sunspot groups for each hemisphere and for each cycle is plotted together by using the time measured relative to the maimum time of that cycle (see Fig. 5). sunspot groups, the acceleration is a constant, about. deg s in a cycle, but for filaments the acceleration increases linearly from about. deg s at the beginning of a cycle to about. deg s at the end of the cycle. CONCLUSIONS AND DISCUSSION Shimojo et al. (6) provided a good comparison between prominence activity and the butterfly diagram obtained from synoptic Carrington maps of Kitt Peak magnetograms. The drift of prominences follows the polarward motion of the magnetic flu until the reversal of the polar polarity. This is clearly visible for latitudes 9. The latitudinal migration of filament activity over the full solar disc was qualitatively introduced by Li et al. (8). In the present study, the catalogue of solar filaments from Carrington solar rotation numbers 876 8, corresponding to complete cycles 6, is used to investigate the latitudinal migration of filaments at low latitudes (less than 5 ), and the latitudinal drift of solar filaments in each hemisphere in each cycle of the time interval is compared with that of sunspot groups in the same time interval. The latitudinal drift of filaments obviously differs from that of sunspot groups: for sunspot groups a second-order polynomial can give a satisfactory description of the latitudinal drift in a cycle, but for filaments it is a third-order not second-order polynomial curve that is required to give a satisfactory description. At the beginning of a cycle, filaments migrate from latitudes of about with a drift velocity of about. m s toward the solar equator, reaching latitudes of about 5 yr later at the cycle maimum with a drift velocity of about. m s, and halt at about 8 at the end of the cycle. Compared with filaments, sunspot groups appear at much lower latitudes at the beginning of a cycle; when solar activity is programmed into a cycle, the difference between the latitudes where filaments and sunspot groups appear becomes smaller and smaller. The difference decreases rapidly in the first and last yr of a cycle, but barely decreases during the yr after the maimum of the cycle. Some reasons why filaments appear at higher latitudes statistically than sunspot groups are as follows. () The decay of old sunspots usually diffuses towards the solar poles (the so-called rush to the pole ), and the emergence of new sunspots at a certain latitude acts along with the decay component of old sunspots, which should cause the formation of a filament (or filaments) at higher latitude. () Emergences of the magnetic field usually form sunspots, but sometimes they form ephemeral regions (tiny pores ) rather than sunspots. Statistically, ephemeral regions are located at higher latitudes than sunspots (Harvey 99). Filaments related to these ephemeral regions should thus appear at higher latitudes. () The background poloidal magnetic field should be more easily observed at high latitudes. The toroidal magnetic field of sunspot groups at a particular latitude should act together with the background poloidal magnetic field to form filaments at a higher latitude. For filaments, the drift velocity decreases in the first 7.5 yr of a cycle, but surprisingly it increases in the last.5 yr of the cycle. However, for sunspot groups the drift velocity always decreases over a whole cycle. The increase of the latitudinal drift velocity of filaments at low latitudes could possibly imply the eistence of interaction between the paired wings of a butterfly, such as crosshemispheric coupling (Charbonneau 7). In fact, for filaments the paired wings of a butterfly can scarcely be distinguished from each other near the equator (see Fig. ). Although sunspots are rarely observed near the equator, filaments can be observed to cross Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6

6 the equator, and are then known as transequatorial filaments (Wang ); this also supports the implication. The latitudinal drift acceleration of filaments changes from negative during the first 7.5 yr or so of a cycle to positive during the last.5 yr or so of the cycle, implying that the mechanism driving filaments to drift should be different during these two stages of a cycle. The different between the latitude drift of filaments in the northern and southern hemispheres is found not to be obvious within a cycle for cycles 6, ecept for cycle 9. The latitudinal drift velocity of filaments differs slightly in the northern and southern hemispheres within a cycle. Filaments are large magnetic structures in the corona, but sunspots are the emergence of a strong magnetic field from the Sun s interior to the photosphere and chromosphere. The difference between the latitudinal drift of filaments and sunspots implies that filaments should sometimes be related to a weak background magnetic field. To what etent does the latitudinal drift of sunspots reflect the meridional flow below the solar atmosphere? This is an open issue. In this study, latitudinal drifts are determined from the tabulated heliocentric positions of the centroids of sunspots and filaments. Centroids from sunspot images are a good measure of sunspot position, because the dark spots are compact and clearly visible in whitelight images. In contrast, filaments are long sinuous structures on the disc; their heliocentric positions are thus less accurately determined than sunspot positions. Filaments are known to tend to be aligned more east west on the disc, and this should reduce the difficulty in determining their latitudes. Averages of filament and sunspot latitudes over one month are used in this study, which should further reduce any differences coming from the determination of filament latitudes. In this study, we investigate the latitudinal drift of filaments over a whole solar cycle, not focusing on individual monthly averages of filament latitudes, reducing the effect of differences on our findings yet again. Thus, the influence of such differences in the determinations of filament positions should be very small and can be neglected. One point should be emphasized. We concentrate on the behaviour of a particular class of filament. We study only the active region (AR) or intermediate-class filaments (see the classification by MacKay et al. (). The fact that filament gravity centres are located higher than the AR means that filaments are preferentially located at the periphery of the AR. Therefore, the large drift of filaments located at latitudes around compared with the drift of sunspots at the beginning of the solar cycle could reflect the fact that there is still a miture of polar-crown filaments migrating towards the pole and low-latitude filaments following AR. ACKNOWLEDGMENTS We thank the referee for careful reading of the manuscript and constructive comments, which improved the original version. The data used here are all downloaded from web sites; the authors epress their deep thanks to the staff of these web sites. The work is supported by the NSFC under Grants 58, 9 and 66, the National Key Research Science Foundation (6CB86) and the Chinese Academy of Sciences. REFERENCES Anzer U.,, in Wilson A., ed., Proc. th Solar Physics Meeting, ESA SP-56, Solar variability: from core to outer frontiers. ESA, Noordwijk, p. 89 Latitude migration of solar filaments 5 Charbonneau P., 7, Adv. Space Res.,, 899 Choudhuri A. R., Schussler M., Dikpati M., 995, A&A,, L9 Coffey H. E., Hanchett C. D., 998, in Webb D. F., Schmieder B., Rust D. M., eds, ASP Conf. Ser. Vol. 5, New Perspectives on Solar Prominences. Astron. Soc. Pac., San Francisco, p. 88 D Azambuja L., 9, Comptes Rendus, 76, 95 Durney B. R., 995, Sol. Phys., 6, Engvold O.,, in Stepanov A. V., Benevolenskaya E., eds, Proc. IAU Symp., Multi-Wavelength Investigations of Solar Activity. Cambridge Univ. Press, Cambridge, p. 87 Harvey K. L., 99, in Harvey K. L., ed., ASP Conf. Ser. Vol. 7, The solar cycle. Astron. Soc. Pac., San Francisco, p. 5 Hathaway D. H., Nandy D., Wilson R. M., Reichmann E. J.,, ApJ, 589, 665 Heinzel P., 7, in Heinzel P., Dorotovic I., Rutten R. J., eds, ASP Conf. Ser. Vol. 68, The Physics of Chromospheric Plasmas. Astron. Soc. Pac., San Francisco, p. 7 Hundhausen A. J., Hansen R. T., Hansen S. F., 98, J. Geophys. Res., 86, 79 Illing R. M. E., Hundhausen A. J., 986, J. Geophys. Res., 9, 95 Kiepenheuer K. O., 95, in Kuiper G. P., ed., The Sun. Univ. Chicago Press, Chicago, p. Labrosse N., Heinzel P., Vial J. C., Kucera T., Parenti S., Gunar S., Schmieder B., Kilper G.,, Space Sci. Rev., in press (doi:.7/s ) Li K. J., Schmieder B., Li Q. S., 998, A&AS,, 99 Li K. J., Yun H. S., Gu X. M.,, AJ,, 5 Li K. J., Liang H. F., Yun H. S., Gu X. M., a, Sol. Phys., 5, 6 Li K. J., Wang J. X., Xiong S. Y., Liang H. F., Yun H. S., Gu X. M., b, A&A 8, 68 Li K. J., Li Q. X., Gao P. X., Mu J., Chen H. D., Su T. W., 7, JA&A, 8, 7 Li K. J., Li Q. X., Gao P. X., Shi X. J., 8, J. Geophys. Res.,, A8 (doi:.9/7ja86) Li K. J., Liu X. H., Gao P. X., Zhan L. S.,, New Astron., 5, 6 Lin Y., Martin S. F., Engvold O., 8, in Heinzel P., Dorotovic I., Rutten R. J., eds, ASP Conf. Ser. Vol. 8, The Physics of Chromospheric Plasmas. Astron. Soc. Pac., San Francisco, p. 5 Low B. C., 98, Sol. Phys., 75, 9 McIntosh P. S., 97, Rev. Geophys. Space Sci.,, 87 Mackay D. H., Karpen J. T., Ballester J. L., Schmieder B., Aulanier G.,, Space Sci. Rev., in press, doi:.7/s Makarov V. I., Sivaraman K. R., 989, Sol. Phys.,, 67 Martin S. F., 99, in Stenflo J. O., ed., Proc. IAU Symp. 8, Solar Photosphere: Structure, Convection, and Magnetic Fields. Kluwer, Dordrecht, p. 9 Mein P., Ribes E., 99, A&A, 7, 577 Minarovjech M., Rybansky M., Rusin V., 998a, Sol. Phys., 77, 57 Minarovjech M., Rybansky M., Rusin V., 998b, in Webb D. F., Schmieder B., Rust D. M., eds, ASP Conf. Ser. Vol. 5, New Perspectives on Solar Prominences. Astron. Soc. Pac., San Francisco, p. 8 Mouradian Z., 998a, in Balasubramaniam K. S., Harvey J. W., Rabin D. M., eds, ASP Conf. Ser. Vol., Synoptic Solar Physics. Astron. Soc. Pac., San Francisco, p. 8 Mouradian Z., 998b, in Balasubramaniam K. S., Harvey J. W., Rabin D. M., eds, ASP Conf. Ser. Vol., Synoptic Solar Physics. Astron. Soc. Pac., San Francisco, p. 97 Mouradian Z., Soru-Escaut I., 99, A&A, 9, 79 Parker E. N., 955, ApJ,, 9 Parker E. N., 987, Sol. Phys.,, Rusin V., Rybansky M., Minarovjech M., 998, in Balasubramaniam K. S., Harvey J., Rabin D., eds, ASP Conf. Ser. Vol., Synoptic Solar Physics. Astron. Soc. Pac., San Francisco, p. 5 Rusin V., Minarovjech M., Rybansky M.,, JA&A,, Sakurai T., 998, in Balasubramaniam K. S., Harvey J., Rabin D., eds, ASP Conf. Ser. Vol., Synoptic Solar Physics. Astron. Soc. Pac., San Francisco, p. 8 Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6

7 6 K. J. Li Schmieder B., Aulanier G., Lopez Ariste A., 7, in Shibata K., Nagata S., Sakurai T., eds, ASP Conf. Ser. Vol. 69, New Solar Physics with Solar-B Mission. Astron. Soc. Pac., San Francisco, p. 7 Schmieder B., Aulanier G., Torok T., 9, in Gopalswamy N., Webb D. F., eds, Proc. IAU Symp. 57, Universal Heliophysical Processes. Kluwer, Dordrecht, p. Shimojo M., Yokoyama T., Asai A., Nakajima H., Shibasaki K., 6, PASJ, 58, 85 Tandberg-Hanssen E., 995, The nature of solar prominences. Kluwer, Dordrecht Tandberg-Hanssen E., 998, in Webb D. F., Schmieder B., Rust D. M., eds, ASP Conf. Ser. Vol. 5, New Perspectives on Solar Prominences. Astron. Soc. Pac., San Francisco, p. Topka K., Moore R., La Bonte B. J., Howard R., 98, Sol. Phys., 79, Wang J.,, in Henou J. C., Fang C., Vilmer N., eds, Understanding active phenomena: progress and perspectives nd French Chinese Meeting on Solar Physics. International Scientific Publishers and World Publishing Corporation Press, Beijing, p. 5 This paper has been typeset from a TEX/LATEX file prepared by the author. Downloaded from by guest on September 8 C The Author. Journal compilation C RAS, MNRAS 5, 6