A KinkUnstable Flux Rope Eruption in Region NOAA

A Kink-Unstable Flux Rope Eruption in Region NOAA 10696 David R. Williams 1, Tibor Török 1, Pascal Démoulin 2, Lidia van Driel-Gesztelyi 1, 2, 3, Bernhard Kliem 4 Mullard Space Science Laboratory, University College London, Holmbury St Mary, Surrey, RH 5 6 NT, U. K. 2 Laboratoire d’Etudes Spatiales et d’Instrumentation en Astrophysique, Observatoire de Paris, F-92195, France 3 Konkoly Observatory, Pf. 67, H-1525 Budapest, Hungary 4 Astrophysical Insititute Potsdam, An der Sternwarte 16, D-14482 Potsdam, Germany 1 Abstract On 2005 November 10, TRACE observed an X 2. 5 flare in NOAA Active Region 10696. The observations were mostly made in the ultraviolet 1600 Å channel at very short cadence (~3. 7 s). The flare is accompanied by a filament eruption, and we investigate its initiation using both TRACE and SOHO/MDI images. We find that there are signatures of three types of mechanism at work in this eruption. However, the evolution of the height and shape of the filament is consistent with that of a flux rope experiencing the MHD kink instability. Tether Weakening by Flux Emergence In the first SOHO/MDI magnetogram taken after the eruption, we see evidence of the emergence of a small bipole (Fig. 3 b). This bipole may locally reconnect (Figs. 3 & 4 a), weakening the field overlying the filament (Fig. 5 a; Moore & Roumeliotis 1992) so that its twist suddenly exceeds the threshold for the kink instability to set in (Török & Kliem 2005). Tether Weakening by “Lateral” Break-out However, there is likely to be shear loaded into the region due to continual westward motion of the positive -polarity elements near this new bipole. This could expand the overlying arcade (Fig. 5 b; Aulanier et al. 2000), also lowering the tension so that the kink instability drives the eruption. This is distinct from central break-out (Antiochos et al. 1999) in that the overlying tension is not removed (explosively or otherwise); rather, it is just weakened. Figure 1 (above): (a) EIT 195 Å image showing the filament f in its pre-eruption position between; the white box shows the area covered by the adjacent inset. The inset shows the filament traced by a solid white line, with the apparent footpoints of the filament marked by black plus signs. (b, c) TRACE 1600 Å images showing the evolution of shape of the bright kinked filament. The black in the inset traces the kinked shape of the erupting structure. Note the reversed J shape of the ribbons between the filament’s apparent feet in (b), similar to the expected shape of quasi-separatrix layers (QSLs) intersecting the photosphere. They are a signature of reconnection in these layers in the corona. (d–f) Magnetic field lines outlining the core of the kink-unstable flux rope at t = 0, 21, and 25 Alfvén times in the numerical simulation of Török & Kliem (2005). The grey-scale levels in the plane indicate positive (white) and negative (black) concentrations of the magnetic field strength component normal to the plane (cf. photospheric magnetic field distribution in Fig. 4 b) Figure 3 (below): Pixel-averaged light curves, during the impulsive and pre-impulsive stages of the flare, for the four boxed regions shown in Fig. 1 a, along with the full-Sun GOES soft X-ray flux. Hatched zone shows time range covered by data in Fig. 2. Figure 2 (above): Upper panel shows the height of the apex of the filament as a function of time, on a logarithmic axis. The lower panel shows the residuals to the exponential fit used. Evolution of the Filament’s Shape. Fig. 1 a shows the filament (in absorption) almost one hour before eruption. Once the filament begins to erupt, it takes on the form of a kinking flux rope. The morphology of the erupting filament shown in Figs. 1 b – 1 c is seen to be qualitatively similar to that of a curved, highly twisted flux rope in numerical simulations (Török & Kliem 2005) of the helical kink instability, seen in Figs. 1 d – 1 f. The profile of the real filament’s apex height with time (Fig. 2) is very close to that of the simulated flux rope. We fit the height of the apex by the function and find excellent agreement. Conclusions We see three effects at work in this filament eruption: 1. Shear is gradually loaded into the photosphere beneath/around the filament 2. More immediately, an emerging bipole may weaken the overlying tethers due to reconnection 3. Based on the morphology and rise profile, the kink instability seems to be the driver of the eruption proper. Figure 5 (above): 2 D approximation of the magnetic topology. Dashed lines show connectivities transformed by reconnection. Dotted lines are separatrices. Thick orange lines are new connectivities after reconnection. Solid black lines indicate unchanged connectivities. Figure 4 (above): (a) TRACE 1600 image showing brightening, assumed to be due to reconnection between the field above the filament and the emerging bipole, circled in panel (b). (c) Four ribbons are seen just before the filament brightens in TRACE’s 1600 A channel. These are assumed to be due to the larger-scale quadrupolar reconnection described in Fig. 5 b. The emerging bipole EF reconnects with the pre-existing field lines to lengthen (and thereby reduce tension in) the field between A and B. The reconnection is seen in Figure 1 as a bump in intensity at ~01: 51 UT. This reduction in tension allows the flux rope bearing the filament f to begin rising. References Antichos, de Vore & Klimchuk (1999) Ap. J 510 485 Aulanier et al. (2000) Ap. J 540 1126 Démoulin, Priest & Lonie (1996) JGR 101 7631 Moore & Roumeliotis (1992) in ‘Eruptive Solar Flares’ Török & Kliem (2005) Ap. J, accepted.