Pre-flare activity and magnetic reconnection during the evolutionary stages of energy release in a solar eruptive flare
Bhuwan Joshi, Astrid M. Veronig, Jeongwoo Lee, Su-Chan Bong, Sanjiv K. Tiwari, Kyung-Suk Cho
aa r X i v : . [ a s t r o - ph . S R ] S e p PRE-FLARE ACTIVITY AND MAGNETIC RECONNECTIONDURING THE EVOLUTIONARY STAGES OF ENERGYRELEASE IN A SOLAR ERUPTIVE FLARE
BHUWAN JOSHI , ASTRID M. VERONIG , JEONGWOO LEE , SU-CHAN BONG ,SANJIV KUMAR TIWARI , AND KYUNG-SUK CHO ABSTRACT
In this paper, we present a multi-wavelength analysis of an eruptive white-light M3.2 flare which occurred in active region NOAA 10486 on November 1,2003. Excellent set of high resolution observations made by RHESSI and TRACEprovide clear evidence of significant pre-flare activities for ∼ ∼
30 keV and coincided with the beginning of flux emergence at the flaring loca-tion along with early signatures of the eruption. From the RHESSI observations,we conclude that both plasma heating and electron acceleration occurred duringthe precursor phase. The main flare is consistent with the standard flare model.However, after the impulsive phase, intense HXR looptop source was observedwithout significant footpoint emission. More intriguingly, for a brief period thelooptop source exhibited strong HXR emission with energies up to 100 keV andsignificant non-thermal characteristics. The present study indicates a causal rela-tion between the activities in the preflare and main flare. We also conclude thatpre-flare activities, occurred in the form of subtle magnetic reorganization alongwith localized magnetic reconnection, played a crucial role in destabilizing the Udaipur Solar Observatory, Physical Research Laboratory, Udaipur 313 001, India IGAM/Institute of Physics, University of Graz, Universit¨ a tsplatz 5, A-8010 Graz, Austria Physics Department, New Jersey Institute of Technology, 161 Warren Street, Newark, NJ 07102, USA Korea Astronomy and Space Science Institute, Daejeon 305-348, Korea Max-Planck-Institut f¨ur Sonnensystemforschung, Max-Planck-Str. 2, 37191 Katlenburg-Lindau, Ger-many
Subject headings:
Sun: corona — Sun: flares — Sun: X-rays
1. INTRODUCTION
Solar eruptive phenomena correspond to various kinds of transient magnetic activitiesoccurring in the solar atmosphere in the form of flares, eruptive prominences and coronalmass ejections (CMEs). With the availability of multi-wavelength data, especially fromthe space based platforms, it has become apparent that these are different manifestationsof a single physical process implicating the disruption of coronal magnetic fields (see e.g.,Lin et al. 2003). There is also a near-universal consensus that magnetic reconnection playsa key role in the process of disruption of magnetic fields as well as dissipation of storedmagnetic energy in the corona (Lakhina 2000; Priest & Forbes 2002).Flares mostly occur in closed magnetic field configuration associated with active regions.Such closed magnetic structures may embrace one or more neutral lines in the photosphericmagnetic flux. In a simplistic model, we can imagine the structure of a bipolar magneticconfiguration in terms of an inner region, called the core field, and the outer region, calledthe envelope field (Moore et al. 2001). The core fields are rooted close to the neutral linewhile the envelope fields are rooted away from it. Before an eruption, the core fields canusually be traced by a dark filament in the chromosphere when viewed on the solar disk.The core fields are usually strongly nonpotential in the pre-flare phase. In the initial stagesof a large eruptive flare, the core fields containing the prominence erupt stretching out theenvelope fields. With the evolution of the eruption process, the stretched field lines reclosevia magnetic reconnection beneath the erupting filament. Multi-wavelength observationsof solar eruptive flares have revealed several key features of the eruption process: risingarcade of intense (newly formed) soft X-ray loops, HXR and H α emission from the feet ofthe newly formed loops, and X-ray coronal source at their summit. The standard CSHKPmodel of solar flares has been successful to broadly incorporate these multi-wavelength flarecomponents (for a review see Hudson et al. 2004; Benz 2008; Schrijver 2009).Although the standard flare model successfully describes several observational features ofa large eruptive flare, the basic question about the triggering of the eruption remains unclearand debatable. The two representative solar eruption models – tether-cutting and breakout– exploits the role of initial magnetic reconnection in two different ways in order to set up theconditions favorable for the core fields to erupt. The “tether-cutting model” is fundamentally 3 –based on a single, highly sheared magnetic bipole, with the earliest reconnection occurringdeep in the sheared core region (Moore et al. 2001). On the other hand, in the “breakoutmodel” the fundamental topology of the erupting system is multi-polar. Here the eruption isinitiated by reconnection at a neutral point located in the corona, well above the core region(Antiochos et al. 1999). In this manner, the former is built on the concept of “internalreconnection” while the latter is suggestive of an “external reconnection” (Sterling et al.2001).In order to understand the triggering mechanism of solar eruption, it is essential toexamine the pre-eruption phase and probe those features which might have played vital rolein the subsequent processes leading to fast energy release and eruption. Observations of solarflares in soft X-rays clearly indicate an enhancement in the flux before the flare, known asthe X-ray precursor phase (Tappin 1991). There are evidences of active pre-flare structurein soft X–rays co-spatial with the main flare which develops several minutes or more beforethe onset of flare (F´arn´ık et al. 1996; F´arn´ık & Savy 1998; Kim et al. 2008). However, weshould not ignore the fact that significant pre-flare activities may be present even before theX–ray precursor phase in other longer wavelength observations such as H α and EUV/UV. Ithas been suggested that the pre-flare brightening may occur as a result of slow reconnectionand provide a trigger for the subsequent eruption (Moore & Roumeliotis 1992; Chifor et al.2007). Here it is worth mentioning that this pre-eruption reconnection may be very differentfrom the post-eruption coronal reconnection, which is believed to lead a two-ribbon flare(Kim et al. 2001).In this paper, we present a comprehensive multi-wavelength analysis of a well observedM3.2 flare that occurred on November 1, 2003. The motivation of the present investigationis two fold: (1) to study the pre-flare activities and their relation with the eruption process,and (2) to understand the role of magnetic reconnection in the corona in the post-eruptionphase. The study utilizes the excellent data sets from three space missions: Reuven RamatyHigh Energy Solar Spectroscopic Imager (RHESSI), Transition Region and Coronal Explorer(TRACE) and Solar and Heliospheric Observatory (SOHO). These observations were sup-plemented by H α and vector magnetic measurements from ground based stations. Imagesof high temporal and spatial resolution at UV and EUV wavelengths coupled with 1-minutecadence SOHO/MDI magnetograms have enabled us to look for the minute changes thattook place in the pre-flare phase. RHESSI X-ray imaging and spectroscopic analysis wasperformed to understand the thermal and non-thermal characteristics of the flare emission.In section 2, we present the data analysis and describe the multi-wavelength view of theevent. In section 3, we integrate and discuss the observations presented in the previoussection. The conclusions of the present study are summarized in section 4. 4 –
2. OBSERVATIONS AND DATA ANALYSIS2.1. Event overview
The Active region NOAA 10486 (along with 10484 and 10488) produced several pow-erful eruptions during the period of October-November 2003. According to the solar regionsummary reports compiled by the Space Weather Prediction Center (SWPC) AR 10486appeared between 23 October and 5 November, 2003. The observations presented here cor-respond to the flare activity that occurred in AR 10486 on November 1, 2003 at the locationS12 W60 showing GOES SXR intensity of M3.2 and H α class of 1N.According to GOES reports the flare took place between 22:26 and 22:49 UT with a peakat 22:38 UT. In Figure 1, we provide the GOES lightcurves in 0.5–4 and 1–8 ˚A wavelengthbands. Figure 2 provides multi-wavelength aspects of the active region. In Figure 2a weshow a representative H α filtergram on November 1, 2003 obtained from the Udaipur SolarObservatory (USO). We find a filament at the north-west part of the active region (markedby an arrow) which was partially erupted during the flare. A comparison of H α filtergramwith TRACE white light (WL) image (Figure 2b) shows three sunspots in the vicinity ofthe location where the filament erupted, two at the eastern side and one at the western sideof it. It is evident that the two sunspots at the eastern side are more prominent. In Figures2c & d, we plot a TRACE WL image and a SOHO/MDI magnetogram respectively duringthe flare main phase overlaid by RHESSI X-ray images.The event was completely observed by RHESSI spacecraft (Lin et al. 2002). In orderto understand the X-ray emission during the eruption process, we have analyzed the X-raylightcurves and images (see section 2.2) in four energy bands namely 6-12, 12-25, 25-50,and 50-100 keV. The RHESSI light curves, shown in Figure 1, are constructed by takingaverage count rates over front detectors 1, 3–6, 8, and 9 in each energy band. We noteseveral important aspects of variation in X-ray fluxes that indicate important stages of theflare evolution: (1) X-ray count rates at 6–12 and 12–25 keV energy bands show a bumpat ∼ &
25 keV) is still at thebackground level. This bump indicates the precursor phase of the flare and is the mostprominent in the 12–25 keV energy band. RHESSI 12–25 keV light curve clearly describesthe X–ray precursor phase between 22:24 and 22:28 UT. (2) GOES as well as RHESSI timeprofiles show a steady rise in the flux from ∼ ∼ &
25 keV furtherenhances after 22:31 UT and a second HXR burst is observed at ∼ ∼ The RHESSI images have been reconstructed with the CLEAN algorithm with thenatural weighing scheme using front detector segments 3 to 8 (excluding 7) in different energybands, namely, 6–12, 12–25, 25–50, and 50–100 keV (Hurford et al. 2002). We compareRHESSI measurements with TRACE images in 195 ˚A and 1600 ˚A wavelengths. The TRACE195 ˚A filter is mainly sensitive to plasmas at a temperature around 1.5 MK (Fe XII) butduring flares it may also contain significant contributions of plasmas at temperatures around15–20 MK (due to an Fe XXIV line; Handy et al. 1999). The TRACE 1600 ˚A channelis sensitive to plasma in the temperature range between (4–10) × K and represent acombination of UV continuum, C I, and Fe II lines (Handy et al. 1999). Also the brightestand most rapidly varying features in TRACE 1600 ˚A channel are likely to emit in the C IVlines (Handy et al. 1998).It is known that the pointing of TRACE is not very accurate. Therefore, in order tocompare RHESSI images and TRACE images, we need to correct the pointing informationof TRACE images. This is achieved by considering the fact that the pointing informa-tion of RHESSI and SOHO is quite accurate. Therefore, we corrected TRACE pointingby comparing a TRACE WL and a SOHO WL image observed at 22:23:29 and 22:23:33UT, respectively. For the cross correlation, we used the Solar SoftWare (SSW) routine,trace mdi align, developed by T. Metcalf (see also Metcalf et al. 2003).We first describe the observations of TRACE in the 195 ˚A EUV channel along withco-temporal RHESSI X-ray images in Figure 3. TRACE images reveal a dark, elongated,inverted U-shaped structure that already existed in the flaring region which corresponds tothe filament (this filament is marked in H α image shown in Figure 2a). The pre-flare 195 ˚Aimages reveal that the filament was thinner at the middle (cf. Figure 3a). Near the middleof the elongated filamentary structure, we observe very first signatures of the flare in the 6 –form of brightenings at both sides of the filament material at ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ ∼ This flare also produced signatures in white light (WL) observations. We examinedseries of TRACE WL images which are available typically every minute. The WL channel ofTRACE has a very broad response, from 1700 ˚A to 1 µ m making it sensitive to emission intransition region, chromosphere, and photosphere (Handy et al. 1999). We find flare relatedbrightening in WL images from ∼ ∼ ∼ We have studied the evolution of RHESSI X-ray spectra during the flare over consecutive20 s intervals from the precursor to the decline phase (i.e., between 22:25:00 UT and 22:40:00UT). For this analysis we first generated a RHESSI spectrogram with an energy binning of keV from 6–15 keV and 1 keV from 15–80 keV. We only used the front segments ofthe detectors, and excluded detectors 2 and 7 (which have lower energy resolution andhigh threshold energies, respectively). The spectra were deconvolved with the full detectorresponse matrix (i.e., off-diagonal elements were included; Smith et al. 2002).In Figure 9 we show spatially integrated, background subtracted RHESSI spectra de-rived during six time intervals of the flare together with the applied spectral fits. This sixtime intervals are marked in Figure 10d where we have plotted 6–30 keV and 30–80 keVRHESSI light curves. The plot reveals that during the precursor phase emission originatesonly at low X–ray energies (i.e, 6–30 keV). We find rapid increase in X-ray flux in 30–80keV energy band only with the onset of impulsive phase at ∼ γ ), normalization of the power-law, and lowenergy turn-over for the non-thermal component. From these fits, we derive the temperatureand emission measure of the hot flaring plasma as well as the power law index for the non-thermal component. Figure 10 shows the time evolution of these parameters obtained fromfits to the RHESSI spectra integrated over consecutive 20 s intervals.The spectra during the precursor phase suggest that hot thermal emission with tem-perature T >
24 MK already existed at the very start (cf. Figure 10a). The compari-son of RHESSI images with UV and EUV observations indicates that this intense plasmaheating corresponds to localized brightenings at three regions in the form of two X-ray rib-bons/footpoints along with a looptop source. However, the temperature decreases afterwardfor a short period (between 22:26:50 and 22:27:50 UT). The temperature again increases withthe onset of the impulsive phase (after 22:27:50 UT) and peaks ( ∼
30 MK) at 22:28:50 UT.The temperature does not increase any further in the later stages. The emission measureEM shows a gradual increase during the precursor phase (between 22:25:10 and 22:27:50UT) followed by a decrease for a short duration. The EM further shows a gradual rise withthe start of the impulsive phase until the end of the HXR emission. Around the peak of theprecursor phase, we observe significant HXR emission. During this time interval, the spectraat energies ε &
10 keV can be fitted by a power law with photon spectral index γ in therange of ∼ ∼ γ in the range of ∼ γ = 3 .
6. It is interesting to note that there is again a significant increase in the HXR emissionat a later stage of the flare at ∼ γ = 3 . In order to understand the magnetic configuration of the activity site and its role indriving the eruptive phenomenon we analyze SOHO/MDI data (Scherrer et al. 1995). InFigure 11a we show a SOHO white light pre-flare image in which we have marked differentsunspots near the flaring region as S1, S2, and S3. As described in subsection 2.1, S1 andS2 are located toward the eastern side of the filament while S3 is at the western side of thefilament. We have thoroughly examined the 1 minute cadence SOHO/MDI magnetogrammovie before and during the flare between 21:00 UT and 23:00 UT. In Figure 11 (panelsb-d), we show a few representative magnetograms. Regions of different magnetic polaritiesare indicated by arrows in Figure 11b. The comparison of panels (a) and (b) suggests thatS1 and S2 are sunspots of negative polarity while S3 is a positive polarity sunspot. Thesequence of magnetogram images of the activity site reveals the emergence of magnetic fluxin a region close to sunspot S3. This region is marked in Figure 11d as EFR. From theFigure it is apparent that the EFR is of positive polarity. However, we cannot be certain ofthe intrinsic magnetic polarity of the EFR because of its location at W60. In Figure 12, weplot emerging magnetic flux through EFR by selecting a rectangular box of size 28 ′′ × ′′ (cf. Figure 11d). The important observation is that the magnetic flux of the EFR starts torise in one polarity at ∼ o azimuthal ambiguity has been resolved by using minimum energy method (Metcalf 1994;Leka et al. 2009). The transverse vector fields are indicated by green arrows in the figure.The region associated with flaring activity is shown inside the red box in Figure 13. The di-rection of transverse vectors between the negative polarity regions (associated with sunspotsS1 and S2) and positive polarity region (associated with sunspot S3) indicate that magneticfield lines are highly sheared. To quantify the non-potentiality of the field lines in the activ-ity site, we compute the spatially averaged signed shear angle (SASSA; Tiwari et al. 2009)which represents the average deviation of the observed transverse vectors from that of thepotential transverse vectors. Over the region of interest (described by red box in Figure 13)we obtain the value of SASSA as ∼ o . Such a high value of SASSA indicates that the regionwas highly stressed, and capable of driving major eruptions (Tiwari et al. 2010). 12 –
3. RESULTS AND DISCUSSION
We divide the whole flare activity into four distinct evolutionary stages and discussimportant characteristic features of each phase. ∼ The initiation phase is readily visible at UV and EUV wavelengths. UV images indicatethe initial brightenings at two locations, one on each side of the filament. This brighteninglooks much like flare kernels. It is noteworthy that main flare occurred at this location only.EUV images reveal rapid changes in the configuration of short loops embedding the filament.We observed interactions among short EUV loops which resulted in the opening of field lines.Vector magnetogram and EUV images suggest that the magnetic field is highly sheared atthis site. We further notice that the discrete, localized brightenings can be identified in bothUV and EUV images.It is evident that the initiation phase represents the initial energy release at distinctlocations in a region of of highly sheared magnetic fields close to the filament, i.e., the “core”of the erupting region. It is likely that during this phase a small volume of plasma is heatedup at different locations insufficient to produce detectable level of X-ray emission. ∼ The precursor phase shows significant X-ray emissions below 30 keV, while the countsrates at higher energies ( ≥
30 keV) are still at the background level. We find an emergingflux region (EFR) within the core region, and very first signatures of the eruption in the formof a bright arc-like feature in UV/EUV images. Since an EFR may destabilize the shearedmagnetic structures leading to solar eruption (Choudhary et al. 1998), it is likely that theonset of eruption is intimately connected to the emergence of magnetic flux. Further, theinitial eruption took place at the location where EUV loops interacted during the initiationphase.The plasma temperature was very high at the beginning of the precursor phase, butthe emission measure was still low and increased gradually (cf. Figure 10). This indicatesthat the low energy X-rays emission at this stage is originated from discrete volumes of hotplasma. This fact is further confirmed with EUV/UV images which still display localizedbrightenings which are cospatial with the X-ray sources and corresponds to emission from 13 –X-ray ribbons/footpoints and looptop. We find that around the peak of the precursor phaseHXR emission follows power law which provides evidence for electron acceleration.To synthesize the initiation and precursor phases, the initial energy release took placein the form of localized brightenings in the highly sheared core region and is associated withthe early signatures of eruption. It has been suggested that the X-ray precursor phase, withdistinct, localized brightenings can be understood in terms of localized magnetic reconnectionwhich acts as a common trigger for both flare emission and filament eruption (Chifor et al.2007). ∼ The impulsive phase is represented by the onset of high energy ( ≥
30 keV) HXR emis-sion at 22:28 UT indicating the impulsive release of large amount of energy. The plasmatemperature also rises impulsively and attains a maximum value of ∼
30 MK at 22:28:50 UT.The temperature slowly decreases in the later stages throughout the flare while emissionmeasure gradually increases. It implies that now the flare involves a larger volume with thefilling of hot plasma in the loop system (Uddin et al. 2003).At the time of maximum plasma temperature, the EUV images show a major re-organization in the structure of the flaring region as indicated in Figure 4. The multi-wavelength signatures observed at HXR, UV, and WL measurements at this stage showconsistency with the standard flare model (see, e.g., Joshi et al. 2007, 2009). Spatial cor-relation between HXR FP sources and WL emitting regions suggests that WL emission isclosely connected with the flare energy deposition by non-thermal particles in the chromo-sphere (Hudson 1972; Metcalf et al. 2003).It is important to note that the converging motion of HXR FPs at the beginning ofthe impulsive phase is marked by rapidly evolving EUV loops into a simplified structure,indicating a possible connection between the two processes. We interpret this as an evidencefor the relaxation of highly sheared magnetic loops (Ji et al. 2007; Joshi et al. 2009).The X-ray spectra exhibit a significant non-thermal component throughout the impul-sive phase with the hardest X-ray emission at the time of the second HXR burst. In general,we note a distinct anti-correlation between the evolution of HXR flux and photon spectralindex (cf. Figure 10c & d). Such a behavior indicates that each non-thermal emission peakrepresents a distinct acceleration event of the electrons in the flare (Grigis & Benz 2004). 14 – ∼ During this interval GOES soft X-rays attain the maximum phase. We found a HXRsource at 25–50 keV at the top of the EUV flare loop system throughout the decline phasewhich shows continuous upward motion, consistent with the standard flare model. However,it is noteworthy that the HXR looptop source is observed without footpoint component.High energy HXR LT source is believed to be closely associated with the site of electronacceleration in the corona (Krucker et al. 2007, 2008a, 2010). Moreover, we observed a HXRburst in this late phase, at ∼ γ = 3.9. At this time a single HXR source was detected for a brief periodwhich is cospatial with the 25–50 keV LT source and shows movement in the same direction.Further it is located away from the HXR FP sources detected earlier during the impulsivephase (cf. Figure 6). Therefore, we interpret that HXR emission at 50–100 keV at thistime originates from the looptop. Coronal HXR emission has been reported in some of therecent RHESSI observations (Lin et al. 2003; Veronig & Brown 2004; Veronig et al. 2005;Krucker et al. 2008b; Krucker & Lin 2008). However, the physical mechanism for such astrong non-thermal source in the tenuous corona is still not clearly understood. Here it isvery interesting to see that the strong HXR emission from the LT source between 22:34 and22:46 UT is temporally associated with the steep rise in the magnetic flux emergence (cf.Figure 12). Further we find that the LT source seems to be spatially located within the EFRregion (Figure 11). Therefore it is likely that the new magnetic flux was continuously fedto the magnetic reconnection site in the corona causing the prolonged non-thermal looptopemission.
4. CONCLUSIONS
The availability of excellent high-cadence multi-wavelength data has enabled us to makea detailed investigation of the physical processes that led to the M3.2 flare on November 1,2003 and the associated eruption. The main emphasis of this study lies in understanding thepre-flare activity which manifested for ∼ ∼
28 MK, and corresponding EUV/UV images showed enhanced brightening along withplasma eruption. More importantly, we find HXR non-thermal emission which suggests thatelectron acceleration occurred during the precursor phase. We therefore conclude that pre-flare brightenings correspond to events of localized magnetic reconnection in the core region,i.e., close to neutral line where filament lies. It is likely that the interactions among shortEUV loops, overlying the filament, followed by the flux emergence played a crucial role indriving the eruption and successive large-scale magnetic reconnection that resulted the mainflare.The impulsive phase of the flare is mostly consistent with the standard flare scenario.However, a HXR looptop source is observed during the impulsive as well as decay phase. Itis noteworthy that HXR LT source became stronger in the decay phase and showed non-thermal emission. Further, the HXR LT sources at the decay phase is rather unusual in thatthere is no significant footpoint emission.The present study indicates a causal relation between pre-flare activity and main flare. Italso follows that the signatures of magnetic reconnection during the initiation and precursorphase occur in the form of localized instances of energy release. In this manner, one candifferentiate pre-eruption reconnection from the post-eruption coronal reconnection whichis generally understood in the framework of standard flare model. Our understanding ofthe pre-eruption reconnection is still limited because of observational constraints. However,we should keep in mind that some times the earliest pre-flare activities can be anticipatedwith EUV/UV measurements, well before the X-ray precursors. The new data sets fromSolar Dynamic Observatory (SDO), with superior resolution, would be very useful for suchinvestigations.We sincerely acknowledge the anonymous referee for critical comments that provide anew direction to discuss the observational results and significantly improved the quality ofthe manuscript. We acknowledge RHESSI, TRACE, SOHO, and GOES for their open datapolicy. RHESSI and TRACE are NASA’s small explorer missions. SOHO is a joint project ofinternational cooperation between the ESA and NASA. We acknowledge the NASA/MSFCdata archive for providing magnetogram data. AV gratefully acknowledges support by theEuropean Community Framework Programme 7, “High Energy Solar Physics Data in Europe(HESPE)”, grant agreement no.: 263086. J. L. was supported by NSF grant AST-0908344.This work was partially supported by the “Development of Korean Space Weather Center” 16 –of KASI and the KASI basic research funds. We thank M. Karlick´y for useful discussions.
REFERENCES
Antiochos, S. K., DeVore, C. R., & Klimchuk, J. A. 1999, ApJ, 510, 485Benz, A. O. 2008, Living Reviews in Solar Physics, 5, 1Brown, J. C. 1971, Sol. Phys., 18, 489Chifor, C., Tripathi, D., Mason, H. E., & Dennis, B. R. 2007, A&A, 472, 967Choudhary, D. P., Ambastha, A., & Ai, G. 1998, Sol. Phys., 179, 133Fang, C., & Ding, M. D. 1995, A&AS, 110, 99F´arn´ık, F., Hudson, H., & Watanabe, T. 1996, Sol. Phys., 165, 169F´arn´ık, F., & Savy, S. K. 1998, Sol. Phys., 183, 339Grigis, P. C., & Benz, A. O. 2004, A&A, 426, 1093Hagyard, M. J., Cumings, N. P., West, E. A., & Smith, J. E. 1982, Sol. Phys., 80, 33Handy, B. N., et al. 1999, Sol. Phys., 187, 229Handy, B. N., Bruner, M. E., Tarbell, T. D., Title, A. M., Wolfson, C. J., Laforge, M. J., &Oliver, J. J. 1998, Sol. Phys., 183, 29Hudson, H., Fletcher, L., Khan, J. I., & Kosugi, T. 2004, in Astrophysics and Space ScienceLibrary, Vol. 314, Solar and Space Weather Radiophysics, ed. D. E. Gary & C. Keller,153Hudson, H. S. 1972, Sol. Phys., 24, 414Hurford, G. J., et al. 2002, Sol. Phys., 210, 61Ji, H., Huang, G., & Wang, H. 2007, ApJ, 660, 893Joshi, B., Manoharan, P. K., Veronig, A. M., Pant, P., & Pandey, K. 2007, Sol. Phys., 242,143Joshi, B., et al. 2009, ApJ, 706, 1438 17 –Kim, J., Yun, H. S., Lee, S., Chae, J., Goode, P. R., & Wang, H. 2001, ApJ, 547, L85Kim, S., Moon, Y., Kim, Y., Park, Y., Kim, K., Choe, G. S., & Kim, K. 2008, ApJ, 683, 510Kontar, E. P., Hannah, I. G., Jeffrey, N. L. S., & Battaglia, M. 2010, ApJ, 717, 250Krucker, S., et al. 2008a, A&A Rev., 16, 155Krucker, S., Hannah, I. G., & Lin, R. P. 2007, ApJ, 671, L193Krucker, S., Hudson, H. S., Glesener, L., White, S. M., Masuda, S., Wuelser, J., & Lin, R. P.2010, ApJ, 714, 1108Krucker, S., Hurford, G. J., MacKinnon, A. L., Shih, A. Y., & Lin, R. P. 2008b, ApJ, 678,L63Krucker, S., & Lin, R. P. 2008, ApJ, 673, 1181Lakhina, G. S. 2000, Bulletin of the Astronomical Society of India, 28, 593Leka, K. D., Barnes, G., & Crouch, A. 2009, in Astronomical Society of the Pacific ConferenceSeries, Vol. 415, Astronomical Society of the Pacific Conference Series, ed. B. Lites,M. Cheung, T. Magara, J. Mariska, & K. Reeves, 365Lin, J., Soon, W., & Baliunas, S. L. 2003, New A Rev., 47, 53Lin, R. P., et al. 2002, Sol. Phys., 210, 3Lin, R. P., et al. 2003, ApJ, 595, L69Metcalf, T. R. 1994, Sol. Phys., 155, 235Metcalf, T. R., Alexander, D., Hudson, H. S., & Longcope, D. W. 2003, ApJ, 595, 483Moore, R. L., & Roumeliotis, G. 1992, in Lecture Notes in Physics, Berlin Springer Verlag,Vol. 399, IAU Colloq. 133: Eruptive Solar Flares, ed. Z. Svestka, B. V. Jackson, &M. E. Machado, 69Moore, R. L., Sterling, A. C., Hudson, H. S., & Lemen, J. R. 2001, ApJ, 552, 833Priest, E. R., & Forbes, T. G. 2002, A&A Rev., 10, 313Scherrer, P. H., et al. 1995, Sol. Phys., 162, 129Schrijver, C. J. 2009, Advances in Space Research, 43, 739 18 –Smith, D. M., et al. 2002, Sol. Phys., 210, 33Sterling, A. C., Moore, R. L., Qiu, J., & Wang, H. 2001, ApJ, 561, 1116Tappin, S. J. 1991, A&AS, 87, 277Tiwari, S. K., Venkatakrishnan, P., & Gosain, S. 2010, ApJ, 721, 622Tiwari, S. K., Venkatakrishnan, P., & Sankarasubramanian, K. 2009, ApJ, 702, L133Uddin, W., Joshi, B., Chandra, R., & Joshi, A. 2003, Bulletin of the Astronomical Societyof India, 31, 303Veronig, A. M., & Brown, J. C. 2004, ApJ, 603, L117Veronig, A. M., Brown, J. C., & Bone, L. 2005, Advances in Space Research, 35, 1683
This preprint was prepared with the AAS L A TEX macros v5.2.
19 –Table 1: Summary of different phases of the flare evolutionPhases Start time–End time (UT) Observing wavelengthInitiation phase 22:19–22:24 EUV and UVPrecursor phase 22:24–22:28 X-ray ( .
30 keV), EUV, and UVImpulsive Phase 22:28–22:34 X-ray (upto ∼
100 keV), EUV, and UVDecay phase 22:34–22:49 X-ray (upto ∼
100 keV), EUV, and UVFig. 1.— RHESSI and GOES lightcurves of the flare with a time cadence of 4 s and 3 srespectively. In order to present different RHESSI light curves with clarity, the RHESSIcount rates are scaled by factors of 1, 1/4, 1/5, and 1/10 for the energy bands 6–12, 12–25,25–50, and 50–100 keV, respectively. 20 –Fig. 2.— Multi-wavelength view of the active region during pre-flare and flare timings.(a) H α filtergram taken from Udaipur Solar Observatory at about ∼
13 hours before theevent. The filament that erupted during the flare is marked by an arrow. (b) White lightobservation of the active region by TRACE taken at the time of H α filtergram shown inpanel (a). We find that the filament lie in the north-west part of the active region. (c) &(d) TRACE white light image and SOHO/MDI magnetogram during the impulsive phaseof the flare. Red and blue contours represent co-temporal X-ray sources at 50–100 keV and6–12 keV energy bands respectively and denote the region where the X-ray intensity is 60%of its peak value. 21 –Fig. 3.— Sequence of TRACE 195 ˚A images from pre-flare to post-flare stages. Panels c, ande–i show co-temporal RHESSI X-ray images in 6–12 keV (blue), 12–25 keV (white) and 50–100 keV (red) energy bands overlaid on TRACE images. RHESSI images are reconstructedwith the CLEAN algorithm using grids 3–8 and natural weighing scheme. The integrationtime for RHESSI images is 30 s. The contour levels for RHESSI images are 60%, 75%, 85%,and 95% of the peak flux in each image. 22 –Fig. 4.— TRACE 195 ˚A images just before and after the first HXR burst. 23 –Fig. 5.— Sequence of TRACE 1600 ˚A images from pre-flare to post-flare stages. Panelsb-c, e–g, i, and k show co-temporal RHESSI X-ray images in 6–12 keV (green), 12–25 keV(white) and 50–100 keV (blue) energy bands overlaid on TRACE images. The contour levelsfor RHESSI images are 60%, 75%, 85%, and 95% of the peak flux in each image. RHESSIimage parameters are the same as in Figure 3. 24 –Fig. 6.— Temporal evolution of X-ray sources in the 6–12 keV (blue), 12–25 keV (green), 25–50 keV (red) and 50–100 keV (black) energy bands. The contour levels for RHESSI imagesare 60%, 75%, 85%, and 95% of the peak flux in each image. RHESSI image parameters arethe same as in Figure 3. 25 –Fig. 7.— Evolution of the altitude of the RHESSI looptop source observed in the 6–12,12–25, 25–50, and 50–100 keV energy bands. Note the 50–100 keV LT source appeared onlyfor a short period (three data points at ∼ ′′ × ′′ shown in panel (d) define EFR . The temporal evolution of emergingmagnetic flux through this region is shown in Figure 12. 30 –Fig. 12.— Temporal evolution of emerging magnetic flux through EFR which is defined inFigure 11d by a rectangular box of size 28 ′′ × ′′ . 31 –Fig. 13.— The line of sight magnetogram of NOAA AR 10486 on November 1, 2003 at 14:08UT overlaid with the transverse vectors (indicated by green arrows). The red/blue contoursrepresent the positive/negative polarity. The contour levels are ± ± ±±