From kozuka Mon Feb 13 10:31:01 1995 Received: by ewsste4.stelab.nagoya-u.ac.jp (4.1/6.4J.6-R5.2) id AA17259; Mon, 13 Feb 95 10:30:53 JST Date: Mon, 13 Feb 95 10:30:53 JST From: kozuka (Yukio Kozuka) Message-Id: <9502130130.AA17259@ewsste4.stelab.nagoya-u.ac.jp> To: watanabe Subject: May 7 paper Dear Watanabe sensei: I send you our manuscrpt which was revised after your suggestion. Added sentences are enclosed by the mark '****'. Please read the manuscript and reply as soon as possible. Sincerely yours, Yukio Kozuka ------------------------------------------------------------------ The Dynamical Characteristics of a Disappearing-Filament Associated Interplanetary Disturbance Observed in Early May, 1992 Yukio KOZUKA^1, Takashi WATANABE^2, Masayoshi KOJIMA^1, Masamitsu OHYAMA^1, Saku TSUNETA^3, Josef I. KHAN^4, and Shin-ichi WATARI^5 1 Solar-Terrestrial Environment Laboratory, Nagoya University, Toyokawa 442 2 Department of Earth Science, Ibaraki University, Mito 310 3 Institute of Astronomy, The University of Tokyo, Mitaka, Tokyo 181 4 Mullard Space Science Laboratory, University College London, RH5 6NT, U. K. 5 Communications Research Laboratory, Koganei, Tokyo 184 Abstract Dynamical properties of an interplanetary disturbance relating to a sudden commencement of a geomagnetic storm at 15h41m UT on May 9, 1992 are discussed using solar wind data obtained with the interplanetary scintillation technique and soft X-ray images taken with the Soft X-ray Telescope on board {\it Yohkoh}. It is suggested here that the sudden commencement was associated with a disappearance of a quiescent filament which took place in the south-east quadrant of the solar disk at about 07h UT on May 7, 1992. An associated shock wave propagated with an approximately constant speed of about 1000 km/s up to about 0.3 AU from the Sun, then showed a blast-wave like deceleration. If this is the case then the duration of the "driven-like phase" of the shock was about 12 hours. According to Yohkoh soft X-ray images, a transient coronal hole was formed near the disappearing filament. The lifetime of this coronal hole, about 17 hours, is comparable to the duration of the driven phase of the shock wave. A close connection between the dynamical characteristics of the shock wave and the formation of the transient coronal hole is suggested. Key words: Sun: disappearing filament --- Sun: solar wind --- Sun: corona --- Sun: X-rays 1. Introduction Disappearances of quiescent filaments have been recognized to be the solar sources of a considerable number of geomagnetic storms (Joselyn, McIntosh 1981). Previous statistical studies of shock waves (Cane et al. 1986; Woo 1988) have shown a variety of dynamical characteristics, e.g., some of them propagated with a nearly constant speed as far as 1 AU, and there were several examples having an apparently driven phase, in which the shock wave propagated with a nearly constant speed, followed by a deceleration phase. The mechanism which controls the dynamical characteristics of the shock waves is not known. It is important to observe large-scale changes in coronal magnetic-field structures associated with a filament eruption to find the agent to drive the shock wave. For this purpose, soft X-ray observations of the solar corona will be suitable because we may study temporal and spatial variations of the coronal structures taking place on the solar disk, with fairly good time resolution. The first attempt to study coronal changes in association with disappearing filaments was made using Skylab soft X-ray images. For example, Webb et al. (1976) found that an eruption of a quiescent filament was frequently accompanied by an enhancement of the soft X-ray brightness, which suggests the formation of a bright arcade structure due to reconnection of the coronal magnetic field under the erupting filament. On the other hand, a decrease of coronal soft X-ray intensity was frequently observed in association with filament eruptions. According to Solodyna et al. (1977), small coronal holes were formed adjacent to the sites of a filament disappearance. Harvey and Sheeley (1979) also pointed out that the boundaries of established coronal holes observed in the He I 1083.0 nm line were altered in association with filament eruptions, and moreover that short-lived coronal holes were sometimes observed during the courses of quiescent filament eruptions or flares. Rust (1983) discussed transient X-ray coronal holes observed during Skylab and speculated that transient holes might cause transient increases of the solar wind speed. No attempt, however, has been made to find solar-wind consequences of the transient changes of the coronal-hole geometry associated with disappearing filaments. Watanabe (1991) proposed that a transient coronal hole, which was occasionally formed adjacent to the disappearing filament, had a close connection to the presence of the driven phase of the shock wave. Since the solar coronal images taken with the Soft X-ray Telescope (SXT; Tsuneta et al. 1991) on board {\it Yohkoh} are useful to examine transient changes in coronal geometry, the possible influence of the transient coronal change on the dynamical properties of interplanetary shock waves may be clarified by using soft X-ray images of the Sun. In the present paper, we report the propagation characteristics of an interplanetary shock apparently caused by a disappearing dark filament on May 7, 1992 by using {\it Yohkoh} soft X-ray images and solar wind data. The source of the solar wind data is multi-station observations (Toyokawa, Fuji, Sugadaira) of IPS (interplanetary scintillation; Kojima, Kakinuma 1987). 2. The Interplanetary Disturbances Which Caused Two Geomagnetic Sudden Commencements on May 9 Two geomagnetic sudden commencements (SCs) occurred consecutively at 15h41m and 19h56m UT on May 9, 1992. Two shock waves corresponding to these SCs were also observed by IMP-J (Solar Geophysical Data). Two Type II radio bursts caused by independent solar phenomena were observed before the SCs (Solar Geophysical Data). The first one was observed at 06h43m UT on May 7 in association with a disappearance of a large quiescent filament initially located in the south-east quadrant (center; S23E48) of the solar disk. An associated 2F/C3.4 flare was observed at 06h40m UT. The second Type II burst was observed at 15h28m UT on May 8. The relevant solar event for the second one is suggested to be the 4B/M7.4 solar flare which took place at S26E08 in the NOAA region 7154 at 15h12m UT on May 8. The initial shock speeds in the corona derived from Type II radio observations are about 1300 km/s for the first Type II burst and about 2000 km/s for the second one (Manoharan et al. 1994). There were no other significant solar events around this time. Assuming that these two coronal shocks propagated separately into interplanetary space and caused the two SCs mentioned above, the mean propagation speeds of the shocks which are estimated from the time intervals between the Type II bursts and the related SCs, are calculated to be 731 km/s for the first shock wave and 1464 km/s for the second one, respectively. Since the mean shock speeds, which may be treated as approximate shock speeds at the mid-point between the Sun and the Earth (namely, 0.5 AU), are slower than those of the coronal shock speeds derived from type II radio observations, it is concluded that these shock waves made decelerating propagations. It is also suggested that, since the shock speed of the following one was much higher than the preceding one, the latter might have been overtaken by the following one shortly after their arrivals at the Earth. These shock waves in question were detected by IPS observations as transient increases in the solar wind speed and the level of the electron density fluctuations. The lines-of-sight geometry of observed radio sources on May 8, 1992 is shown in Figure 1 as the projection onto the solar equatorial plane. A tick on each line-of-sight indicates the location of the monitoring point (the maximum of the scattering weighting function). The longitude of the disappearing quiescent filament in question is also shown in the figure. A disturbed solar wind with a speed of 637 km/s, about 85 km/s higher than that of the previous day, was observed at a distance of approximately 0.53 AU from the Sun with the IPS observation of 3C119 at 04h37m UT on May 8. The IPS of 3C138 at 05h10m UT on May 8, monitoring the solar wind at a heliocentric distance of about 0.55 AU, did not show any enhancement in the flow speed and the density fluctuation level. Since these two radio sources were observed before the 4B flare occurred, it is reasonable to conclude that the high-speed solar wind observed by the IPS of 3C119 on May 8 was related to the disappearing-filament associated shock wave. The upper and the lower limits of the mean shock speed between the Sun and the lines-of-sight of the radio sources can be determined from IPS observations of these two radio sources mentioned above. The mean shock speed between the Sun and the line-of-sight of 3C119 is $>$1010 km/s. This gives the lower limit of the mean shock speed at 0.265 AU (a half of the heliocentric distance of the monitoring point on the line-of-sight of 3C119). The upper limit of the mean shock speed at 0.275 AU is estimated from the IPS observation of 3C138 to be $<$1042 km/s. A disturbed solar wind with a speed of 648 km/s, which is 280 km/s higher than the speed of the previous day, was observed by the IPS of 3C138 at 05h05m UT on May 9. The observation of 3C147 at 05h31m UT, which showed a speed of 590 km/s at about 0.72 AU from the Sun, also showed the presence of the shock wave. The IPS observation of 3C161 at 06h16m UT, by which the solar wind at 0.83 AU from the Sun was observed, indicates that the mean shock speed at 0.415 AU from the Sun is $<$727 km/s. These observations indicate also the presence of the shock wave near the Earth's orbit on May 9, shortly before the SC at 15h41m UT on May 9, 1992. Estimated shock speeds associated with the disappearing filament on May 7 are summarized in Table 1. The IPS observations of radio sources used in the present analysis provided us with the solar wind speed in interplanetary space near the meridian of the disappearing quiescent filament, as seen in Figure 1. Anisotropic propagation characteristics of the shock wave may be neglected because we used the solar wind speeds obtained in a narrow region of the interplanetary space around the meridian of the disappearing filament. Detailed discussion on the propagation properties of the shock wave based on the table will be given in the next section. On May 10, most of the observed radio sources showed an enhancement in the solar wind speed caused by an interplanetary shock wave associated with the 4B flare on May 8. This indicates the presence of a large-scale shock wave expanding in a wide range of the interplanetary space, both in the eastern and the western hemispheres centered at the flare site. Detailed discussion on this event will be given elsewhere. 3. Propagation Characteristics of the Shock Wave Associated with the Disappearing Filament on May 7, 1992 The mean propagation speeds of the shock wave related to the May 7 event at various heliocentric distances are estimated from the onset time of the SC and the observational time of each IPS source. The propagation properties of the shock wave are inferred from the mean shock speeds and the {\it in situ} shock speeds derived from the observed flow speed under a strong-shock approximation (Watanabe, Kakinuma 1984). Figure 2 shows the relationship between the shock speed (V) and the heliocentric distance (R) in association with the disappearing solar filament on May 7, 1992. The speed of the relevant Type II burst is also shown in this figure. It is seen from this graph that the shock wave propagated outward with a nearly constant speed of 1000---1300 km/s from the Sun as far as about 0.3 AU, and then experienced a blast-wave like deceleration (the shock speed is proportional to R$^{-0.5}$; e.g. Dryer et al. 1975). The duration of the constant speed propagation (driven-like phase) was about 12 hours. Similar combination of the driven-like and the decelerated phases of a shock wave was reported by Watanabe and Schwenn (1989). In the case of the interplanetary shock wave associated with a disappearance of a quiescent filament on April 22, 1979, the shock wave propagated with a constant speed of about 1000 km/s (or with a very weak deceleration) from the Sun up to, at least, 0.41 AU, i.e., for about 17 hours. The blast-wave like deceleration of the shock speed took place after the initially nearly constant speed propagation. 4. Soft X-Ray Observations {\it Yohkoh} SXT images were examined to find changes in the solar coronal structures associated with the disappearing filament on May 7, 1992. Figure 3 shows a section of a soft X-ray synoptic chart covering the interval from 5 to 14 May, 1992 and superposing the H-alpha synoptic chart reported in the Preliminary Report and Forecast of Solar Geophysical Data (SESC PRE 873). ***************************************************************** The soft X-ray synoptic chart was made from daily SXT images. We selected an image for each day and extracted a "lune", whose width was 15 degrees in longitude, at the central meridian. The extracted images were stretched in the latitudinal direction and placed at the corresponding longitude in the chart. ***************************************************************** The relevant quiescent filament which disappeared was indicated by the arrow. Since the synoptic chart is constructed from soft X-ray images around the central meridian of the Sun, the coronal structures about four days after the filament disappearance on May 7 are shown. An enhanced X-ray coronal region is seen around the original place of the disappearing quiescent filament. Figure 4 shows selected SXT images taken in the interval from 06h05m UT on May 7 to 13h08m UT on May 8. A dynamically active arcade (Khan et al. 1994, in preparation) was seen in association with the disappearing filament. Instead of the structureless coronal enhancement surrounding the site of the disappearing filament shown in Figure 3, it is found that a small coronal hole appeared to the east of the arcade in the image taken at 10h42m UT on May 7. This transient coronal hole expanded continuously until 14h32m UT on May 7. The maximum size of the coronal hole is about 15 degrees in the longitudinal and the latitudinal directions. The hole can be clearly seen until 03h43m UT on May 8 (see Figure 4). Afterwards, the hole faded out gradually, and then completely disappeared in images taken on May 9. Transient depressions of the coronal brightness may be caused by stretched magnetic field lines due to the filament escaping from the Sun. It will not be, however, the case for the transient coronal hole observed on May 7, 1992 because the coronal hole appeared at the place apart from the arcade. In Figure 4, We may see a depression of the coronal brightness along the southern edge of the arcade. The presence of the northern counterpart is uncertain, but it is suggested that this part was geometrically hidden by the arcade. It is suggested that, consequently, the transient coronal hole in question was not caused by the motion of the filament. The location of the counterpart of the transient coronal hole is, however, not clear in this stage. 5. A Driving Mechanism of the Shock Wave We have seen in the previous section that a transient coronal hole appeared after the dark quiescent filament erupted. The lifetime of the transient coronal hole was approximately 17 hours. As shown in Section 3, the shock wave associated with the disappearing filament of May 7, 1992 had a driven phase lasting for about 12 hours. This means that we need a mechanism to drive the shock wave up to the heliocentric distance of 0.3 AU. Since the lifetime of the transient coronal hole (17 hours) is close to the duration of the driven phase of the shock wave (12 hours), it is suggested that the transient change of the coronal hole geometry including the formation of the transient coronal hole may be an indication of an additional energy release to drive the shock wave. Generation of a transient high-speed stream will be a candidate of the manifestation of energy input. Since the solar wind condition after the appearance of the interplanetary disturbance in IPS data was disturbed by the flare-associated shock wave related to the 4B solar flare of May 8, 1992, the presence of the high-speed stream is uncertain. *************************************************************** Similar event (April 22--24, 1979) was reported by Watanabe and Schwenn (1989) and Watanabe (1991). *************************************************************** Thermal energy input from the bright arcade structure formed after the filament eruption is another candidate of the energy source to drive the shock wave. The duration of the enhancement of soft X-ray intensity due to the formation of a bright arcade structure, however, was shorter than four hours. We conclude that the thermal energy input was not an important energy source to drive the shock wave. *************************************************************** Kinetic energy in a transient high speed solar wind is suggested to be the source for the driven-like expansion of the shock wave. *************************************************************** 6. Concluding Remarks The propagation characteristics of the interplanetary disturbances related to two sudden commencements (SCs) on May 9, 1992 were investigated using IPS observations and soft X-ray images from {\it Yohkoh}. Our conclusions are summarized as follows: (1) The SC observed at 15h41m UT on May 9, 1992 was related to a disappearance of a quiescent filament on May 7, 1992. The SC observed at 19h56m UT on May 9, 1992 was related to a 4B/M7.4 flare on May 8, 1992. (2) The first shock wave propagated outward with a nearly constant speed of 1000---1300 km/s from the Sun as far as about 0.3 AU. The shock experienced a blast-wave like deceleration after the driven-wave like propagation. The duration of the driven time was about 12 hours. (3) {\it Yohkoh} SXT images show a transient change of the coronal hole geometry including a transient coronal hole which was formed shortly after the disappearance of the filament. The hole was not caused by the motion of the eruptive filament. The lifetime of the transient coronal hole was about 17 hours. Since this lifetime is close to the duration of the driven-like phase, the shock is suggested to have been driven by an additional driving force related to the change of the large-scale magnetic configuration around the site of the disappearing filament. The lifetime of the bright arcade structure was about four hours. It is suggested that the thermal energy input caused by magnetic reconnection taking place below the erupting filament was not sufficient to drive the shock wave. It is shown that the duration of the driven-like phase of the shock wave was close to the lifetime of the transient coronal hole which was formed in association with the filament disappearance on May 7, 1992. Since the thermal energy input will not be the principal energy source to drive the shock wave for 12 hours, a large-scale coronal mass ejection (CME) generated by an extended change in the coronal magnetic geometry, which was indicated by the erupting filament itself and the transient coronal hole, is suggested to be the probable energy source to drive the shock wave: the propagation speed of the CME slowed down after the closed magnetic configuration in the solar corona was reestablished. It is required further observational and theoretical works to study the driving mechanism of interplanetary shock waves associated with filament eruptions. **************************************************************** On the flare event on May 8, 1992, a transient coronal hole was not observed with {\it Yohkoh}. The shock speeds estimated from the Type II and the start time of the SC suggest that the shock was decelerated for this event. Harvey and Sheeley (1979), however, pointed out that there were some cases that a transient change of the coronal hole geometry was caused by a flare. The driving mechanism should be also investigated in the case of a flare-associated event by using {\it Yohkoh} soft X-ray data. *************************************************************** Acknowledgments The authors would like to express their sincere thanks to all of the members of the {\it Yohkoh} team for their efforts in running this highly successful mission. References Cane, H. V., Kahler, S. W., Sheeley, N. R., Jr. 1986, J. Geophys. Res. 91, 13321 Dryer, M., Eviatar, A., Frohlich, A., Jacobs, A., Joseph, J. H., Weber, E. J. 1975, J. Geophys. Res. 80, 2001 Harvey, J. W., Sheeley, N. R., Jr. 1979, Space Sci. Revs. 23, 139 Kojima, M., Kakinuma, T. 1987, J. Geophys. Res. 92, 7269 Manoharan, P. K., Ananthakrishnan, S., Detman, T. R., Dryer, M., Leinbach, H., Kojima, M., Watanabe, T., Khan, J. I. 1994, Proc. of Kofu Symposium, New look at the Sun with emphasis on advanced observations of coronal dynamics and flares, ed S. Enome, T. Hirayama, NRO Report No. 360, p109 Joselyn, J. A., McIntosh, P. S. 1981, J. Geophys. Res. 86, 4555 Rust, D. M. 1983, Space Sci. Revs. 34, 21 Solodyna, C. V., Krieger, A. S., Nolte, J. T. 1977, Solar Phys. 54, 123 Tsuneta, S., Acton, L., Bruner, M., Lemen, J., Brown, W., Caravalho, R., Catura, R., Freeland, S. et al. 1991, Solar Phys. 136, 37 Watanabe, T. 1991, in Flare Physics in Solar Activity Maximum 22, Lecture Notes in Physics 387, ed Y. Uchida, R. C. Canfield, T. Watanabe, E. Hiei, (Springer-Verlag, Berlin), p353 Watanabe, T., Kakinuma, T. 1984, Adv. Space Res. 4, No. 7, 331 Watanabe, T., Schwenn, R. 1989, Space Sci. Revs. 51, 147 Webb, D. F., Krieger, A. S., Rust, D. M. 1976, Solar Phys. 48, 159 Woo, R. 1988, J. Geophys. Res. 93, 3919 Figure Captions Figure 1. Line-of-sight geometry of IPS radio sources on May 8, 1992, projected onto the solar equatorial plane. A tick on each line-of-sight indicates the location of the monitoring point (the maximum of the scattering weighting function). The location of the disappearing filament is also shown. Figure 2. Relationship between the shock speed and the heliocentric distance associated with the filament disappearance of May 7, 1992. Figure 3. Soft X-ray synoptic chart from 5 to 14 May, 1992 constructed from daily images of the {\it Yohkoh} Soft X-ray Telescope with a superposed H-alpha synoptic chart. The location of the disappearing solar filament is indicated by the arrow. Figure 4. Selected soft X-ray images from 06h05m UT on May 7 to 13h08m UT on May 8. A transient coronal hole, which is indicated by the arrows, is seen to the east of the active arcade associated with the filament disappearance. ----------------------------------------------------- end of file