A New Method of Identifying 3D Null Points in Solar Vector Magnetic Fields

Employing the Poincare index of isolated null-points in a vector field, we worked out a mathematical method of searching for 3D null-points in coronal magnetic fields. After introducing the relevant differential topology, we test the method by using the analytical model of Brown & Priest. The location of null-point identified by our method coincides precisely with the analytical solution. Finally we apply the method to the 3D coronal magnetic fields reconstructed from an observed MDI magnetogram of a super-active region (NOAA 10488). We find that the 3D null-point seems to be a key element in the magnetic topology associated with flare occurrence.     

Magnetic field gradient and flare: study of a small flare in NOAA 8038

Active region NOAA 8038 was observed from 10 to 13 May, 1997 using the USO solar video magnetograph. During this period, the active region was mostly inactive, and gave rise to only a single notable flare of 1N/C1.3 class on May 12, 1997/04:45 UT. The flare occurred in a weak field location, but new emerging fluxes were observed prior to the flare onset. Horizontal motions of the network photospheric magnetic fluxes were inferred using USO and SOHO magnetograms, and velocities in the range 300–800 m s^–1 were estimated. The initial flare brightening was observed at the flux cancellation site where magnetic field gradients were found to increase. Detailed analyses of flux motions, cancellation and their relation with the flare are presented.     

Diagnosis of coronal magnetic field with data of Nobeyama Radio Heliograph

The expression is derived for the coronal magnetic field strength from the observations of brightness, temperature, peak frequency, spectral index, and polarization degree of solar microwave bursts. One example of solar burst on November 28, 1998 is given for the calculation of coronal magnetic field from the data of Nobeyama Radio Heliograph (NoRH). The results are comparable with the SOHO/MDI magnetogram and the calculation from the Nobeyama Radio Polarimeters (NoRP), as well as the coronal loops in SOHO/EIT and YOHKOH/SXT images. Therefore, it may be the first time that the two-dimensional diagnosis of coronal magnetic field in a microwave burst source from the radio observations has been made.     

Prominent Photospheric Downflows on Magnetic Neutral Line in a Delta-Type Sunspot

It is known that ??-type sunspot groups have a high flare productivity and produce strong flares. In particular, ?|?? type sunspots are the most active type of all. A ?|?? active region NOAA?9957 with frequent flux cancellations but without any marked flare activity during its decay phase was studied in this work. Using SOHO/MDI Dopplergrams and magnetograms, we detected continuous prominent downflow motions of 1500??C?1700?m?s^?1 for several hours on the magnetic neutral line in this region. In the downflow region, penumbral structures were observed to decay. We will interpret and discuss the phenomenon as a case of submergence of the magnetic flux.     

Variations of Photospheric Magnetic Field Associated with Flares and CMEs

Using high-cadence magnetograms from the SOHO/MDI we have investigated variations of the photospheric magnetic field during solar flares and CMEs. In the case of a strong X-class flare of May 2, 1998, we have detected variations of magnetic field in a form of a rapidly propagating magnetic wave. During the impulsive phase of the flare we have observed a sudden decrease of the magnetic energy in the flare region. This provides direct evidence of magnetic energy release in solar flares. We discuss the physics of the magnetic field variations, and their relations to the Moreton Hα waves and the coronal waves observed by the EIT.     

Cme Associated with Transequatorial Loops and a Bald Patch Flare

We study a flare which occurred on 3 November 1997 at 10:31 UT in the vicinity of a parasitic polarity of AR 8100. Using SOHO/EIT 195 Å observations, we identify the brightening of thin transequatorial loops connecting AR 8100 and AR 8102, and dimmings located between the two active regions. Difference images highlight the presence of a loop-like structure rooted near the flare location usually called an EIT wave. The coronal magnetic field derived from potential extrapolations from a SOHO/MDI magnetogram shows that the topology is complex near the parasitic polarity. There, a `bald patch' (defined as the locations where the magnetic field is tangent to the photosphere) is present. We conclude that the flare was a `bald patch flare'. Moreover, the extrapolation confirms that there is a large coronal volume filled with transequatorial field lines interconnecting AR 8100 and AR 8102, and overlaying the bald patch. We show that the dimmings are located at the footpoints of these large field lines, which can be also related to the thin bright loops observed during the flare. As this event was related to a coronal mass ejection (CME) observed by SOHO/LASCO, we propose that the observed dimmings are due to a decrease in plasma density during the opening of the transequatorial loops connecting both ARs. We propose a scenario where these large field lines are in fact pushed up by the opening of low-lying sheared field lines forming the bald patch. We finally discuss how the fast opening of these field lines can produce the brightening near the footpoints of the separatrix, observed as an `EIT wave'.     

Multiwavelength Observations of a Failed Flux Rope in the Eruption and Associated M-Class Flare from NOAA AR 11045

We present the multiwavelength observations of a flux rope that was trying to erupt from NOAA AR 11045 and the associated M-class solar flare on 12 February 2010 using space-based and ground-based observations from TRACE, STEREO, SOHO/MDI, Hinode/XRT, and BBSO. While the flux rope was rising from the active region, an M1.1/2F class flare was triggered near one of its footpoints. We suggest that the flare triggering was due to the reconnection of a rising flux rope with the surrounding low-lying magnetic loops. The flux rope reached a projected height of ≈0.15R _⊙ with a speed of ≈90 km s⁻¹ while the soft X-ray flux enhanced gradually during its rise. The flux rope was suppressed by an overlying field, and the filled plasma moved towards the negative polarity field to the west of its activation site. We found the first observational evidence of the initial suppression of a flux rope due to a remnant filament visible both at chromospheric and coronal temperatures that evolved a couple of days earlier at the same location in the active region. SOHO/MDI magnetograms show the emergence of a bipole ≈12 h prior to the flare initiation. The emerged negative polarity moved towards the flux rope activation site, and flare triggering near the photospheric polarity inversion line (PIL) took place. The motion of the negative polarity region towards the PIL helped in the build-up of magnetic energy at the flare and flux rope activation site. This study provides unique observational evidence of a rising flux rope that failed to erupt due to a remnant filament and overlying magnetic field, as well as associated triggering of an M-class flare.     

Probing the Role of Magnetic-Field Variations in NOAA AR 8038 in Producing a Solar Flare and CME on 12 May 1997

We carried out a multi-wavelength study of a Coronal Mass Ejection (CME) and an associated flare, occurring on 12 May 1997. We present a detailed investigation of magnetic-field variations in NOAA Active Region 8038 which was observed on the Sun during 7??C?16 May 1997. This region was quiet and decaying and produced only a very small flare activity during its disk passage. However, on 12 May 1997 it produced a CME and associated medium-size 1B/C1.3 flare. Detailed analyses of H?? filtergrams and SOHO/MDI magnetograms revealed continual but discrete surge activity, and emergence and cancellation of flux in this active region. The movie of these magnetograms revealed the two important results that the major opposite polarities of pre-existing region as well as in the emerging-flux region were approaching towards each other and moving magnetic features (MMF) were ejected from the major north polarity at a quasi-periodicity of about ten hours during 10??C?13 May 1997. These activities were probably caused by magnetic reconnection in the lower atmosphere driven by photospheric convergence motions, which were evident in magnetograms. The quantitative measurements of magnetic-field variations such as magnetic flux, gradient, and sunspot rotation revealed that in this active region, free energy was slowly being stored in the corona. Slow low-layer magnetic reconnection may be responsible for the storage of magnetic free energy in the corona and the formation of a sigmoidal core field or a flux rope leading to the eventual eruption. The occurrence of EUV brightenings in the sigmoidal core field prior to the rise of a flux rope suggests that the eruption was triggered by the inner tether-cutting reconnection, but not the external breakout reconnection. An impulsive acceleration, revealed from fast separation of the H?? ribbons of the first 150 seconds, suggests that the CME accelerated in the inner corona, which is also consistent with the temporal profile of the reconnection electric field. Based on observations and analysis we propose a qualitative model, and we conclude that the mass ejections, filament eruption, CME, and subsequent flare were connected with one another and should be regarded within the framework of a solar eruption.     

Modeling and Measuring the Flux Reconnected and Ejected by the Two-Ribbon Flare/CME Event on 7 November 2004

Observations of the large two-ribbon flare on 7 November 2004 made using SOHO and TRACE data are interpreted in terms of a three-dimensional magnetic field model. Photospheric flux evolution indicates that ?1.4×10⁴³ Mx² of magnetic helicity was injected into the active region during the 40-hour buildup prior to the flare. The magnetic model places a lower bound of 8×10³¹ ergs on the energy stored by this motion. It predicts that 5×10²¹ Mx of flux would need to be reconnected during the flare to release the stored energy. This total reconnection compares favorably with the flux swept up by the flare ribbons, which we measure using high-time-cadence TRACE images in 1?600 Å. Reconnection in the model must occur in a specific sequence that would produce a twisted flux rope containing significantly less flux and helicity (10²¹ Mx and ?3×10⁴² Mx², respectively) than the active region as a whole. The predicted flux compares favorably with values inferred from the magnetic cloud observed by Wind. This combined analysis yields the first quantitative picture of the flux processed through a two-ribbon flare and coronal mass ejection.     

Ni I 6768 Å line profile variations during a solar flare and their effect on the SOHO/MDI magnetic field measurements

Observational data on the Ni I 6768 Å line profile variations during the impulsive and post-impulsive phases of the July 18, 2002 while light flare (WLF) in the kernel of WLF emission and in other flare kernels are presented. The line profiles at the sites of intense photospheric motions in active regions are also studied. The effect of the observed Ni I 6768 Å line profile variations on the SOHO/MDI magnetic field measurements is estimated. The following conclusions have been reached. (1) The thermodynamic structure of the photo-spheric layers changes significantly during the flare. As a result, the Ni I line profile changes, particularly at the site of WLF emission. At this time, the line depth decreases significantly, but the line does not show any emission reversal. Subsequently, a relatively slow return to the conditions of an undisturbed photosphere is observed. (2) The technique of SOHO/MDI magnetic field measurements is insensitive to such line variations. Therefore, the detected variations during the flare did not result in any noticeable errors in the MDI longitudinal magnetic field measurements. (3) The line profile is broadened, shifted as a whole, and asymmetric at the sites of active regions where intense photospheric motions appear. In the MDI measurements, such changes in the profile lead to an underestimation of the magnetic field by approximately 10% if the line-of-sight velocity of the photo-spheric ejection is about 1.6 km s^?1.     

Slip-Running Reconnection in Quasi-Separatrix Layers

Using time dependent MHD simulations, we study the nature of three-dimensional magnetic reconnection in thin quasi-separatrix layers (QSLs), in the absence of null points. This process is believed to take place in the solar atmosphere, in many solar flares and possibly in coronal heating. We consider magnetic field configurations which have previously been weakly stressed by asymmetric line-tied twisting motions and whose potential fields already possessed thin QSLs. When the line-tied driving is suppressed, magnetic reconnection is solely due to the self-pinching and dissipation of narrow current layers previously formed along the QSLs. A generic property of this reconnection process is the continuous slippage of magnetic field lines along each other, while they pass through the current layers. This is contrary to standard null point reconnection, in which field lines clearly reconnect by pair and abruptly exchange their connectivities. For sufficiently thin QSLs and high resistivities, the field line footpoints slip-run at super-Alfvénic speeds along the intersection of the QSLs with the line-tied boundary, even though the plasma velocity and resistivity are there fixed to zero. The slip-running velocities of a given footpoint have a well-defined maximum when the field line crosses the thinnest regions of the QSLs. QSLs can then physically behave as true separatrices on MHD time scales, since magnetic field lines can change their connections on time scales far shorter than the travel-time of Alfvén waves along them. Since particles accelerated in the diffusive regions travel along the field much faster than the Alfvén speed, slip-running reconnection may also naturally account for the fast motion of hard X-ray sources along chromospheric ribbons, as observed during solar flares. Electronic Supplementary Material Supplementary material is available for this article at     

The Late Gradual Phase of Large Flares: The Case of November 3, 2003

The hard X-ray time profiles of most solar eruptive events begin with an impulsive phase that may be followed by a late gradual phase. In a recent article (Aurass et al. in Astron. Astrophys. 555, A40, 2013), we analyzed the impulsive phase of the solar eruptive event on November 3, 2003 in radio and X-ray emission. We found evidence of magnetic breakout reconnection using the radio diagnostic of the common effect of the flare current sheet and, at heights of ±0.4 R_⊙, of a coronal breakout current sheet (a source site that we called X). In this article we investigate the radio emission during the late gradual phase of this event. The work is based on 40?–?400 MHz dynamic spectra (Radio Spectrograph, Observatorium Tremsdorf, Leibniz Institut für Astrophysik Potsdam, AIP) combined with radio images obtained by the French Nançay Multifrequency Radio Heliograph (NRH) of the Observatoire de Paris-Meudon. Additionally, we use Ramaty High Energy Solar Spectroscopic Imager (RHESSI) hard X-ray (HXR) flux records, and Solar and Heliospheric Observatory (SOHO) Large Angle and Spectrometric Coronagraph (LASCO) and Extreme ultraviolet Imaging Telescope (EIT) images. The analysis shows that the late gradual phase is subdivided into two distinct stages. Stage 1 (lasting five minutes in this case) is restricted to reoccurring radio emission at source site X. We observe plasma emission and an azimuthally moving source (from X toward the NE; speed≈1200 km?s^?1) at levels radially ordered against the undisturbed coronal density gradient. These radio sources mark the lower boundary of an overdense region with a huge azimuthal extent. By the end of its motion, the source decays and reappears at point X. This is the onset of stage 2 traced here during its first 13 minutes. By this time, NRH sources observed at frequencies≤236.6 MHz radially lift off with a speed of ≈?400 km?s^?1 (one third of the front speed of the coronal mass ejection (CME)) as one slowly decaying broadband source. This speed is still observable in SOHO/LASCO C3 difference frames in the wake of the CME four hours later. In stage 2, the radio sources at higher frequencies appear directly above the active region with growing intensity. We interpret the observations as the transit of the lower boundary of the CME body through the height range of the coronal breakout current sheet. The relaxing global coronal field reconnects with the magnetic surroundings of the current sheets that still connect the CME in its wake with the Sun. The accelerated particles locally excite plasma emission, but can also escape toward the active region, the CME, and the large-scale solar magnetic field. The breakout relaxation process may be a source of reconnection- and acceleration rate modulations. In this view, the late gradual phase is a certain stage of the coronal breakout relaxation after the release of the CME. This article is, to our best knowledge, the first observational report of the coronal breakout recovery. Our interpretation of the radio observations agrees with some predictions of magnetic breakout simulations (e.g. Lynch et al. in Astrophys. J. 683, 1192, 2008). Again, combined spectral and imaging radio observations give a unique access to dynamic coronal processes that are invisible in other spectral ranges.     

Simulation of the Unusual Solar Minimum with 3D SIP-CESE MHD Model by Comparison with Multi-Satellite Observations

The observations both near the Sun and in the heliosphere during the activity minimum between solar cycles 23 and 24 exhibit different phenomena from those typical of the previous solar minima. In this paper, we have chosen Carrington rotation 2070 in 2008 to investigate the properties of the background solar wind by using the three-dimensional (3D) Solar?CInterPlanetary Conservation Element/Solution Element Magnetohydrodynamic (MHD) model. We also study the effects of polar magnetic fields on the characteristics of the solar corona and the solar wind by conducting simulations with an axisymmetric polar flux added to the observed magnetic field. The numerical results are compared with the observations from multiple satellites, such as the Solar and Heliospheric Observatory (SOHO), Ulysses, Solar Terrestrial Relations Observatory (STEREO), Wind and the Advanced Composition Explorer (ACE). The comparison demonstrates that the first simulation with the observed magnetic fields reproduces some observed peculiarities near the Sun, such as relatively small polar coronal holes, the presence of mid- and low-latitude holes, a tilted and warped current sheet, and the broad multiple streamers. The numerical results also capture the inconsistency between the locus of the minimum wind speed and the location of the heliospheric current sheet, and predict slightly slower and cooler polar streams with a relatively smaller latitudinal width, broad low-latitude intermediate-speed streams, and globally weak magnetic field and low density in the heliosphere. The second simulation with strengthened polar fields indicates that the weak polar fields in the current minimum play a crucial role in determining the states of the corona and the solar wind.     

Flare-Induced Excitation of Solar p modes

Solar flares release large amounts of energy at different layers of the solar atmosphere, including at the photosphere in the case of exceptionally major events. Therefore, it is expected that large flares would be able to excite acoustic waves on the solar surface, thereby affecting the p-mode oscillation characteristics. We have applied the ring-diagram analysis technique to 3-D power spectra obtained for different flare regions in order to study how flares affect the amplitude, frequency and width of the acoustic modes. Data from the Michelson Doppler Imager (MDI) on board the Solar and Heliospheric Observatory (SOHO) has been used. We have used data obtained for several active regions of the current solar cycle that have produced flares. In most cases, during the period of high flare activity, power in p modes appears to be larger when compared to that in non-flaring regions of similar magnetic field strength.     

Multi-Wavelength Analysis of the 29 September 2002 M2.6/2N Flare

Using TRACE EUV 171 Å line, Hα line, Zürich radio, RHESSI, and HXRS observations the 29 September 2002 flare (M2.6), which occurred in AR NOAA 0134, was analyzed. Flaring structures were compared with a potential magnetic field model (field lines and quasi-separatrix layers) made from SOHO/MDI full-disk magnetogram. Series of high-resolution SOHO/MDI magnetograms and TRACE white-light images were used to find changes in the active region at the photosphere during the flare. The flare began with a rising of a small dark loop followed by the flare brightening observed in 171 Å with TRACE and Hα lines. In radio wavelengths, first type III bursts were observed 5 min prior to the start of hard X-ray emission, indicating a pre-flare coronal activity. The main hard X-ray emission peak (at 06:36 UT) was associated with the second type III burst activity and several slowly negatively drifting features, all starting from one point on the radio spectrum (probably a shock propagating through structures with different plasma parameters). After this time a huge loop formed and three minutes later it became visible in absorption both in Hα and 171 Å EUV lines. The phase of huge dark loop formation was characterized by long-lasting, slowly negatively drifting pulsations and drifting continuum. Finally, considering this huge loop as a surge an evolution of the event under study is discussed.     

Magnetic Properties of Metric Noise Storms Associated with Coronal Mass Ejections

Using Nancay Radioheliograph (NRH) imaging observations, combined with SOHO/Michelson Doppler Imager (MDI) magnetogram observations and coronal magnetic field extrapolation, we studied the magnetic nature of metric noise storms that are associated with coronal mass ejections (CMEs). Four events are selected: the events of 2000 July 14, 2001 April 26, 2002 August 16 and 2001 March 28. The identified noise storm sources cover or partially cover the active regions (ARs), but the centers of storm sources are offset from the ARs. Using extrapolated magnetic field lines, we find that the noise storm sources trace the boundary between the open and closed field lines. We demonstrate that the disappearance of noise storm source is followed by the appearance of the burst source. The burst sources spread on the solar disk and their distributions correspond to the extent of the CME in LASCO C2 field of view. All the SOHO/Extreme Ultraviolet Imaging Telescope (EIT) dimmings associ- ated with noise storm sources are located at the periphery of noise storms where the magnetic lines of force were previously closed and low-lying. When the closed field becomes partially or fully open, the basic configurations of noise storm sources are changed, then the noise storm sources are no longer observed. These observations provide the information that the variations of noise storms manifest the restructuring or reconfiguring of the coronal magnetic field.     

Short-Term Solar Flare Prediction Using a Sequential Supervised Learning Method

Solar flares are powered by the energy stored in magnetic fields, so evolutionary information of the magnetic field is important for short-term prediction of solar flares. However, the existing solar flare prediction models only use the current information of the active region. A sequential supervised learning method is introduced to add the evolutionary information of the active region into a prediction model. The maximum horizontal gradient, the length of the neutral line, and the number of singular points extracted from SOHO/MDI longitudinal magnetograms are used in the model to describe the nonpotentiality and complexity of the photospheric magnetic field. The evolutionary characteristics of the predictors are analyzed by using autocorrelation functions and mutual information functions. The analysis results indicate that a flare is influenced by the 3-day photospheric magnetic field information before flare eruption. A sliding-window method is used to add evolutionary information of the predictors into machine learning algorithms, then C4.5 decision tree and learning vector quantization are employed to predict the flare level within 48 hours. Experimental results indicate that the performance of the short-term solar flare prediction model within the sequential supervised learning framework is significantly improved.     

Comparison of STEREO/EUVI Loops with Potential Magnetic Field Models

The Solar Terrestrial Relations Observatory (STEREO) provides the first opportunity to triangulate the three-dimensional coordinates of active region loops simultaneously from two different vantage points in space. Three-dimensional coordinates of the coronal magnetic field have been calculated with theoretical magnetic field models for decades, but it is only with the recent availability of STEREO data that a rigorous, quantitative comparison between observed loop geometries and theoretical magnetic field models can be performed. Such a comparison provides a valuable opportunity to assess the validity of theoretical magnetic field models. Here we measure the misalignment angles between model magnetic fields and observed coronal loops in three active regions, as observed with the Extreme Ultraviolet Imager (EUVI) on STEREO on 30 April, 9 May, and 19 May 2007. We perform stereoscopic triangulation of some 100?–?200 EUVI loops in each active region and compute extrapolated magnetic field lines using magnetogram information from the Michelson Doppler Imager (MDI) on the Solar and Heliospheric Observatory (SOHO). We examine two different magnetic extrapolation methods: (1) a potential field and (2) a radially stretched potential field that conserves the magnetic divergence. We find considerable disagreement between each theoretical model and the observed loop geometries, with an average misalignment angle on the order of 20°?–?40°. We conclude that there is a need for either more suitable (coronal rather than photospheric) magnetic field measurements or more realistic field extrapolation models.     

Is Pre-Eruptive Null Point Reconnection Required for Triggering Eruptions?

We study the magnetic field evolution and topology of the active region NOAA 10486 before the 3B/X1.2 flare of October 26, 2003, using observational data from the French–Italian THEMIS telescope, the Michelson Doppler Imager (MDI) onboard Solar and Heliospheric Observatory (SOHO), the Solar Magnetic Field Telescope (SMFT) at Huairou Solar Observation Station (HSOS), and the Transition Region and Coronal Explorer (TRACE). Three dimensional (3D) extrapolation of photospheric magnetic field, assuming a potential field configuration, reveals the existence of two magnetic null points in the corona above the active region. We look at their role in the triggering of the main flare, by using the bright patches observed in TRACE 1600 Å images as tracers at the solar surface of energy release associated with magnetic reconnection at the null points. All the bright patches observed before the flare correspond to the low-altitude null point. They have no direct relationship with the X1.2 flare because the related separatrix is located far from the eruptive site. No bright patch corresponds to the high-altitude null point before the flare. We conclude that eruptions can be triggered without pre-eruptive coronal null point reconnection, and the presence of null points is not a sufficient condition for the occurrence of flares. We propose that this eruptive flare results from the loss of equilibrium due to persistent flux emergence, continuous photospheric motion and strong shear along the magnetic neutral line. The opening of the coronal field lines above the active region should be a byproduct of the large 3B/X1.2 flare rather than its trigger.