Magnetic reconnection as a source of jets from a penumbral intrusion into a sunspot umbra
aa r X i v : . [ a s t r o - ph . S R ] S e p Received ........; Accepted .........
Magnetic reconnection as a source of jets from a penumbral intrusion into asunspot umbra
L. Bharti , S. K. Solanki , and J. Hirzberger
1. Max-Planck-Institute für sonnensystemforschung, Justus-von-Liebig-Weg 3, 37077, Göttingen,Germany2. School of Space Research, Kyung Hee University, Yongin, Gyeonggi Do, 446-701, Korea [email protected]
ABSTRACT
We present the results of high resolution co-temporal and co-spatial photosphericand chromospheric observations of sunspot penumbral intrusions. The data was takenwith the Swedish Solar Telescope (SST) on the Canary Islands. Time series of Ca IIH images show a series of transient jets extending roughly 3000 km above a penum-bral intrusion into the umbra. For most of the time series jets were seen along thewhole length of the intruding bright filament. Some of these jets develop a clear λ -shaped structure, with a small loop appearing at their footpoint and lasting for arounda minute. In the framework of earlier studies, the observed transient λ shape of thesejets strongly suggests that they are caused by magnetic reconnection between a curvedarcade-like or flux-rope like field in the lower part of the penumbral intrusion andthe more vertical umbral magnetic field forming a cusp-shaped structure above thepenumbral intrusion. Subject headings:
Sun: convection – sunspots – Sun: photosphere — chromosphere
1. Introduction
High resolution observations have revealed that the chromosphere above sunspots is verydynamic at small scales. Jet-like phenomena observed above sunspot penumbrae (penumbral microjets, Katsukawa et al. 2007) and above umbrae (umbral microjets, Bharti et al. 2013) in Ca II Hare examples of such dynamic events. The penumbral micro jets are associated with transient 2 –brightenings and are observed at the boundaries of penumbral filaments. Their lifetime is about1 minute and the apparent speed projected on the image plane is about 100 km/s. The length ofthese jets is approximately 1000-4000 km and their width is 400 km. Jurˇcák & Katsukawa 2008found that they are aligned with the background magnetic field . Umbral microjets (Bharti et al.2013) are likely also aligned with the background magnetic field. The typical length and width ofumbral microjets is less than 1 ′′ and 0 . ′′
3, respectively. They last around one minute. Reardon etal. (2013) found that penumbral micro jets show extended emission in the wings of spectral lines ,thus giving a similar signature to Ellerman bombs (Ellerman 1917) outside sunspot.Other interesting dynamical phenomena in the sunspot chromosphere are brightenings andsurge activity above light bridges(Roy 1973, Asai et al. 2001, Berger & Berdyugina 2003, Bhartiet al. 2007, Louis et al. 2008, 2009, Shimizu et al. 2009, Shimizu 2011). In particular, Shimizuet al. (2009) and Shimizu (2011) clearly detected jet-like structure in light bridges. The observedbright plasma ejection was intermittent and recurrent for more than a day. The length of the brightejections was reported to be 1500-3000 km, i.e. significantly less than previously reported length10000 km of dark surges in H α images by Roy (1973) and Asai et al. (2001). Shimizu et al.(2009) occasionally observed fan-shaped ejection as well as a chain of ejections from one endthe of the light bridge to the other. Louis et al. (2008) reported arch-like structure above a lightbridgeand brightness enhancement along lightbridge. Bharti (2015) finds jets above a light bridgewhich reach up to coronal heights and the leading edge of jets are hotter in the transition region andcorona. The jets show a coordinated behaviour i.e. neighbouring jets move up and down together(Yang et al. 2015). Kleint & Sainz Dalda (2013) reported on brightenings above several unusualfilamentary structures (umbral filaments) in a sunspot. Coronal images show end points of brightcoronal loops end in these unusual filaments.Sakai & Smith (2008) and Magara (2010) proposed models for penumbral microjets thatbuild on magnetic reconnection taking place between more inclined magnetic field in penumbralfilaments and a more vertical background field. A similar interpretation was proposed by Shimizuet al. (2009) for the (bidirectional) jets emanating from a light bridge.Chromospheric jets and surges are also found outside active regions (Shibata et al. 2007).In particular, Shibata et al. (2007) found λ -shaped jets in the quiet chromosphere, which wasinterpreted to support the notion of reconnection between an emerging magnetic bipole and apreexisting uniform vertical field (Yokoyama & Shibata 1995). Such a loop configuration givesindirect evidence of small- scale reconnection in the solar atmosphere (Singh at al. 2012a, Yanet al. 2015). Particularly, Yan et al. (2015) found evidence for intermittent reconnection in a λ -shaped jet driven by loop advection and followed by an outflow that excites waves and jets. Theauthors interpret these phenomena as causes and consequences of reconnection ,respectively. Adetailed analysis of quiet sun jets is presented by Nishizuka et al. 2011. A study of anemone jets 3 –by Morita et al. (2010) suggests jets generated in the lower chromosphere. On the other handa multiwavelength study of a jet show simultaneous appearance in the lower and in the upperatmosphere (Nishizuka et al. 2008). The appearance of jets in the atmosphere depends on the sizeof the bipole (Shibata et al. 2007, Singh et al. 2011).The present study is based on the high resolution G-continuum, Ca II H and FeI 6302 Å ob-servations taken with the Swedish Solar Telescope to detect λ -shaped jets above a penumbralintrusion into the umbra. This letter is organized as follows: in Sec. 2 we present the observations,in Sec. 3 we describe the λ –shaped jets and their association with the underlying photosphericumbral features, we then present our conclusions in Sec. 4.
2. Observations
The observations were carried out in various wavelength bands at the Swedish Solar Telescope(SST, Scharmer et al. 2003), La Palma, Canary Islands on 2006 August 13. The sunspot belongedto the active region NOAA 10904 and positioned at the heliocentric angle θ = 40 . ◦ ( µ = 0 . x = - ′′ and y = - ′′ on the solar disk.The sunlight was divided into a blue and a red channels. In the blue beam various interferencefilters were used to obtain images in G-band (4305 Å ), G-continuum (4363 Å ), Ca II H (3968.5 Å) line core and in the ling wing (0.06 Å away from line center). The image scale in the blue beamcorresponded to 0.041"/pixel.The red beam was fed to the Solar Optical Universal Polarimeter (SOUP, see Title & Rosen-berg 1981) filter to scan the FeI 6302.5 Å line at 6 wavelength positions. The width of the filter was75 mÅ. Full Stokes polarimetry was performed to measure the magnetic field vector at each pixelof the field of view. In addition, continuum (broad band) images at 6302 Å were recorded. Theplate scale in the red beam corresponded to 0. 065"/pixel. More detail on the data acquisition andreconstruction are described in Hirzberger et al. (2009). Measurement of instrumental polarizationeffects of the SST optical setup were done by inserting calibration optics, consisting of a rotatingpolarizer and a quarter wave plate into the beam. A code developed by Selbing (2005) was used todetermine the demodulation matrix (see Hirzberger et al. (2009) for more details.)Speckle interferometric techniques (see, Hirzberger et al. 2009) were used to reconstructedthe G-continuum, the Ca II H core and Ca II H wing time series and SOUP polarization data. In ad-dition, the HeLIx inversion code (Lagg et al. 2004) was used to invert the Stokes vector data assum-ing a simple one-component-plus-straylight Milne-Eddington atmosphere. The obtained magneticparameters were transformed to the local coordinate system and a code developed by Georgoulis(2005) was used to resolve the 180 ◦ ambiguity. 4 –Fig. 1.— Left panel: contrast-enhanced G-continuum image of the inner part of the observedsunspot at 9:11:31 UT. The arrow labelled ’DC’ points towards disk center. Right panel: cotem-poral and cospatial contrast-enhanced Ca II H line core image. Locations marked A, B and C inthe left panel indicate penumbral intrusions where jet-like events are seen. White boxes outline thefield-of-view of events discussed in more details in the text. Blow-ups of these boxes are shown inFig. 2 and in the movies available as on-line material. 5 –Fig. 2.— Evolution of the jets and λ -shaped jets above penumbral intrusion C. Upper row: G-continuum images; lower row: simultaneous and cospatial images in the Ca II H line core. Theplotted scene corresponds to the white squares in Fig. 1. Time ∆ t=0 corresponds to 9:11:12 UT(see movie-II for more details which is available as on-line material). 6 –Fig. 3.— Continuum image and plasma parameters from inversions of the Stokes profiles at around6302 Å . a: continuum at 6302 Å. b: line-of-site velocity, c: magnetic field strength, d: inclinationof magnetic vector, e: azimuth. The overplotted contours outline the umbral-penumbral boundaryin the continuum images. 7 –
3. Analysis and result
Contrast enhanced co-temporal and co-spatial images of the sunspot in the G-continuum andin the Ca II H line core are displayed in Fig. 1. A broad light bridge(LB), a larger and a smallerumbrae is visible in both the images. In the G-continuum image both the umbra show peripheralumbral dots and several dark patches (nuclei). However, in the larger umbra central umbral dotsand many penumbral intrusions are also recognised (Bharti et al. 2013). These intrusions are alsovisible with lower contrast in the Ca II H core image. Several jet-like transient bright structures arevisible above penumbral filaments (penumbral microjets, cf. Katsukawa et al. 2007), LB (Shimizuet al. 2009 and Shimizu 2011) and penumbral intrusions in the Ca II H core image (best seen inthe mpeg movie-I which is provided as online material).Occasionally bright jets similar to penumbral microjets are visible above penumbral intrusionsA and B as can be gleaned from Movie-I. One can notice around 9:11 UT the jets above intrusionB have a longer life time and a larger width than jets above intrusion A. Most striking is penumbralintrusion C, however. There jets are both brighter and more extended than those from the otherintrusions and occur without interruption over the entire observing span of about 47 min. The jetsabove the intrusion follow the orientation of the nearby penumbral microjets. In the following, weconcentrate on the activity above penumbral intrusion C.In the beginning at 8:28:27 UT the jets occur only along the tip of this penumbral intrusion.Later on jets appear somewhat closer to the umbral-penumbral boundary.The jets’ length and brightness is higher around the tip of the filament. Around 9:02:01 UT abunch of jets appears near the umbral-penumbral boundary with comparable length and brightnessas around the tip of the filament. This strong jet activity then rapidly migrates along the intrusioninto the umbra. Apart from migration, neighbouring jets merge with each other. Such migrationand merging of jets has been also reported above a light bridge in the transition region by Bharti(2015).In Fig. 2, from 9:11:12 UT on, one can clearly see λ -shaped jets, i.e. jets coming out of thetops of (small)loops whose footpoint lie in or next to the penumbral intrusion. Later, after 9:12:28UT, again only elongated jets are visible in subsequent frames, although a weak λ -like structure isstill hinted at.In the G-continuum images the penumbral intrusion shows a dynamical behaviour reminiscentof twisting motions and migrations of bright grains (see mpeg Movie-II). A clear association ofbright penumbral grain migration towards the umbra and the jets is visible around the tip of thepenumbral intrusion from 8:42:23 UT to 8:45:52 UT. Similarly, from 9:11:12 UT onwards (i.e.starting with the appearance of λ -shaped jets), migration of bright penumbral grains towards theumbra in G-continuum images close to the footpoints of λ -shaped jets can be recognised. 8 –We applied an intensity threshold (c.f. Bharti et al. 2013 for more details) to determinevarious parameters of the jets. The projected lengths of jets are found to be 1700-3000 km. The λ -shaped jets belong to the shorter ones with lengths of 1700-1900 km. The separation betweenthe footpoints of λ -shaped jet is 450-600 km and the loop at the bottom of the λ structure reachesa height of around 600 km above the loop base. These jets have shorter length than penumbralmicrojets (Katsukawa et al. 2007) but longer length than umbral microjets (Bharti et al. 2013). Thelifetime of these jets (2-3 min) is comparable with penumbral microjets (Katsukawa et al. 2007)and umbral microjets (Bharti et al. 2013) as well as with jets reported in umbra by Yurchyschyn etal. (2014).Panel ’a’ of Fig. 3 display the continuum intensity (broad-band at 6302 Å) at 8:44:31 UT.The penumbral intrusion appears similar to that in the G-continuum (see Movie-II), but, due to thesomewhat lower spatial resolution and contrast (caused by the longer wavelength), the fine struc-tures are less clear. The line-of-sight velocity is depicted in panel ’b’. The part of the intrusionclose to the penumbra-umbra boundary shows upflows, similar to the ones typical of the innerpenumbra while the part away from the penumbra-umbra boundary shows upflows in the sectiontowards disk center and downflows away from disk center. From x = 1 . ′′ towards the tip of the in-trusion only downflows are visible. The magnetic field strength is plotted in panel ’c’. It is weakerin the intrusion compared to the surroundings, but always above 1 kG (in the mid-photosphere),again similar to filaments embedded in the penumbra. Panel ’d’ suggests that the field in the in-trusion is more transverse, where the direction perpendicular to the solar surface refers to zerodegrees. Interestingly, there is an opposite polarity patch, around x = 2 . ′′ and y = 1 . ′′ , at the edgeof the intrusion, close to the penumbra-umbra boundary. The azimuth (panel ’e’) also shows about90 ◦ rotation compared to the rest of the field surrounding the opposite polarity patch. The positivey axis is taken to correspond to a field azimuth of zero. The arrows overplotted on the intensityimage in Fig. 4 show the horizontal field. Note that there may be smaller amounts of oppositepolarity flux also along other parts of the filament hidden by the dominant umbral field. Althoughwe observe patches of opposite polarity, we are aware of the fact that with 6 wavelength points andhighly redshifted profiles, this interpretation is not straightforward. Since there are observationalhints (Esteban Pozuelo et al. 2015) and well-established results from simulations (Rempel 2012)for the existence of opposite polarity field at the edges of penumbral filaments as well as in lightbridges (Bharti et al. 2007, Lagg et al. 2014, Louis et al. 2014) the observed opposite polaritypatches might be real. The presence of jets beyond the opposite polarity patch support this notion.The left panel of Fig. 4 displays a Ca II H wing image of the penumbral intrusion at 08:44:55UT. The umbral-penumbral boundary, and the penumbral intrusion as visible in the continuumimage (6302 Å), are marked by white contours. The orange contour outlines the opposite polaritypatch. In the Ca II H wing image the penumbral intrusion shows a filamentary structure with acentral dark lane along its whole length. In continuum radiation, formed somewhat deeper than 9 –the line wing, the intrusion is much narrower and only the bright part adjacent to the central darklane toward disk center is clearly visible. The other part, on the limbward side of the central darklane, is hardly seen. Note that the sunspot is located away from the disc center at θ = 40 . ◦ , andthe intrusion corresponds to an elevated structure (Lites et al. 2004) due to the raised optical depthunity level(Cheung et al. 2010). It is unclear if this 3-D structure of the iso- τ surface is responsiblefor the difference in visibility and structure of the intrusion at different wavelengths. The velocitiesmay also give an indication why the intrusion looks so different in the lower photosphere than inthe middle and upper photosphere. The weak upflow in the disc center side (i.e. in the continuumintrusion) and the strong downflows in the part close to the limb (visible only in the Ca wing) canbe described as in Fig. 5. We propose that these flows have different physical causes: a) convection(narrow arrows in Fig. 5), with upflows on the disc center side of the arcade and downflows on theother side. These convective flows heat the gas in the lower photosphere, but only at the location ofthe upflows, producing a narrow intrusion in lower photosphere. b) in the upper photosphere andthe chromosphere, the intrusion is heated more by the reconnection, which accelerates the gas inboth directions (thick arrows in Fig. 5) and thus brightens both parts of the intrusion located at thefootpoints of the emerging arcade. The fact that the reconnection takes place closer to the limbwardfootpoint of the arcade may explain why the intrusion is brighter in Ca wing on the limbward side atmany places. Due to increasing gas density with depth, the reconnection does not affect the lowerphotosphere. The width of the penumbral intrusion in the 6302 Å continuum images is 400-550km, which is comparable to, but somewhat narrower than the footpoint separation of the λ -shapedjets in the Ca II H line core images (see Fig. 2). The width of the intrusion, as seen in the wing ofCa II H, is sufficient, however, to easily host both legs of the λ -shaped jets. The right panel of Fig.4 depicts a Ca II H line core image with overplotted contours of the continuum (6302 Å) and theopposite polarity patch as in the left panel. At x = 2 . ′′ and y = 1 . ′′ , a jet can be seen which seemsto emerge directly from the opposite polarity. Close inspection of Movie-I shows that the locationof this jet activity corresponds to x = 8 . ′′ and y = 6 ′′ in Ca II H that peaked at 08:44:55 UT. It isevident from Fig. 3 and 4 that there is also a downflow and opposite polarity patch surrounding x = 2 . ′′ and y = 1 . ′′ . This location corresponds to the footpoints of the jets seen in Ca II H linecore. Louis et al. (2014) found small scale jets above light bridges in Ca II H line core imagesfrom Hinode (SOT) observations which are associated with localized patches of opposite polarityin the photosphere. The jets are triangular-shaped and exhibit a spike-like structure. The width ofthe filter used in the present study is narrower (1.1 Å) than the one used in Hinode (SOT/BFI). Inaddition, the spatial resolution of the present observations is nearly a factor of two higher, whichmight enable us to unravel the sub-structure ( λ -shape) of the jets. 10 –
4. Discussion and conclusion
We have for the first time clearly detected λ -shaped jets in a sunspot. These lie above apenumbral intrusion into the umbra. With a length of 1800 km they are at the small end of thequiet sun λ -shaped jets (lengths of 2000-5000 km) reported by Shibata et al. (2007), based onHinode/SOT Ca II H observations. This shape of a jet is considered to be a signal of reconnec-tion between the loop connecting an emerging magnetic bipole and preexisting more vertical field(Yokoyama and Shibata 1995, Shibata et al. 1992, 2007). In our case a whole arcade of emergingflux is required, as we see jets all along the intrusion, although not all of them display a λ -typestructure possibly due to overlap between jets or insufficient spatial resolution. As can be deducedfrom the movie-II, the group of jets to which the λ -shaped jets belong moves rapidly along thepenumbral intrusion, starting at the umbral-penumbral boundary and ending at the tip of the intru-sion. This implies an emerging arcade and suggests that the arcade first started to emerge near theumbra-penumbra boundary and later towards the tip of the intrusion. This notion is supported bythe fact that we also see a migration of bright penumbral grains (in the G-continuum) toward thetip of the intrusion and close to the λ -shaped jets. Instead of an arcade, a rising and heavily twistedflux rope (and possibly rotating) could also be the cause the jets, or a strong elongated convectiveupwelling, carrying opposite polarity magnetic flux to its edges. Singh et al. (2012) found system-atic motions of λ -shaped jets outside a sunspot in Ca II H observations from Hinode in terms ofmigration from one end of the footpoint to the other end of an arcade that, finally, leads to mergingof jets similar to the merging of jets in the peumbral intrusion presented here. Such a morphologyis illustrative of the emergence of twisted flux rope (see Fig. 4 of Singh et al. 2014). The presenceof the λ -shaped loops, however, requires a relatively ordered loop-like field, as illustrated in Fig. 5.This scenario is reproduced by a rising arcade of loops connected with convective upwelling. Theconvective upflow heats the gas in the lower photosphere thus loops connected with convectiveupwelling brightens the disk side penumbral intrusion in the continuum. The appearance of thebright λ -shaped loop in the upper photosphere is due to the heating caused by the reconnection,which accelerates the gas in both directions, filling both the loop with hot and bright gas as well asthe jet above it.Magara (2010) performed a MHD simulation to explain penumbral microjets. In his modela strongly twisted flux rope (penumbral filament) is placed within the more vertical backgroundfield (umbral field). Magnetic reconnection takes place on the side of the flux rope with field lineshaving opposite polarity to the background field. Bharti et al. (2010) found reverse polarities atedges of larger UDs, in the simulations of Schüssler & Vögler (2006), caused by strong convec-tive downflows. The presence of opposite polarity along the lateral edges of penumbral filamentshas been reported recently by Ruiz Cobo & Asensio Remos (2013), Scharmer et al. (2013), Ti-wari et al. (2013) and Esteban Pozuelo et al. (2015). These opposite polarities are thought to 11 –be caused by convective downflows (Joshi et al. 2011; Scharmer et al. 2011). The observed upand downflows pattern, their correlation with lower photospheric brightness (bright upflows, darkdownflows) opposite polarity and (possibly to a lesser extent) apparent twisting motions in thepenumbral intrusion similar to penumbral filaments (Ichimoto at al 2007, Zakharov et al. 2008,Bharti et al. 2010, 2012) suggests that the intrusion is of convective nature similar to penumbralfilaments, umbral dots and light bridges (Cheung et al. (2010). This supports the reconnectionscenario of Magara (2010), although the full structure of the magnetic field may be different fromwhat he proposed to produce a jet-like structure (see Tiwari et al. 2013). The approximate align-ment of penumbral microjets with the background field found by Jurˇcák, & Katsukawa (2008) wasexplained by Nakamura et al. (2012) as caused by reconnection between a weaker more horizon-tal and a stronger background field. Shimizu et al. (2009) proposed a model to interpret plasmaejections above a LB. In their model the current-carrying and highly twisted LB field is trappedbelow a cusp-shaped magnetic structure formed by the background field. This geometry is thenproposed to lead to reconnection on the side of the LB at which the field is of opposite polarity tothe umbral field. In a laboratory experiment Nishizuka et al. (2012) succeeded in producing jetsthat are qualitatively similar to penumbral jets as a result of magnetic reconnection in a roughlysimilar magnetic configuration. The general magnetic configuration presented by Magara (2010)and Shimizu et al. (2009) is in general agreement with our findings for the penumbral intrusion.However, different sources of the magnetic flux forming the arcade are possible. It may be anemerging arcade of loops along the intrusion, or an emerging or twisting flux rope. Finally, fieldstwisted and dragged down by magnetoconvection are also candidates. Since the λ jets are seen inchromospheric radiation and the loop at the base of the jet is around 600 km high, we expect thatthe reconnection is taking place in the upper photosphere.We thank an anonymous referee for helpful comments that improved this paper. The Swedish1-m Solar telescope is operated on the island of La Palma by the Institute for Solar Physics of theRoyal Swedish Academy of sciences in the Spanish Observatorio del Roque de los Muchachosof the Instituto de Astrofísica de Canarias. L.B. is grateful to the Inter University Centre forAstronomy and Astrophysics (IUCAA) Reference Center at the Department of Physics, MohanlalSukhadia University, Udaipur, India, for providing computational facilities. This work has beenpartly supported by the BKZI plus program through the NRF funded by the Korean Ministry ofEducation. REFERENCES
Asai, A., Ishii, T., Kurokawa, H. 2001, ApJ, 555, L65 12 –Berger, T. E., Berdyugina, S. V. 2001, ApJ, 589, L117Bharti, L. 2015, MNRAS, 452, L16Bharti, L., Hirzberger, J., Solanki, S. K. 2013, A&A, 552, L1Bharti, L., Cameron, R. H., Rempel, M., Hirzberger, J., Solanki, S. K. 2012, ApJ, 752, 128Bharti, L., Beek, B., Schüssler, M. 2010, A&A, 510, 12Bharti, L., Solanki, S. K., Hirzberger, J. 2010, ApJ, 722, L194Bharti, L., Rimmele, T., Jain, R., Jaaffrey, S. N. A., Smartt, R. N. 2007, MNRAS, 376, 1291Cheung, M. C. M., Rempel, M., Title, A. M.; Schüssler, M. 2010, ApJ, 720, 233Ellerman, F. 1917, ApJ, 46, 298Esteban Pozuelo, S., Bellot Rubio, L. R., de la Cruz Rodríguez, J. 2015, ApJ, 803, 93Georgoulis, M. K. 2005, ApJ, 629, L69Hirzberger, J., Riethmüller, T., Lagg, A., Solanki, S. K., Kobel, P. 2009, A&A, 505, 771Ichimoto et al. 2007, Science, 318, 1597Joshi, Jayant, Pietarila, A., Hirzberger, J., Solanki, S. K., Aznar Cuadrado, R., Merenda, L. 2011,ApJ, 734L, 18Jurˇcák, Katsukawa, Y. 2008, A&A, 488, L33Katsukawa, Y. et al. 2007, Science, 318, 1594Kleint, L. & Sainz Dalda, A. 2013, 770, 74Lagg, A., Woch, J., Krupp, N.; Solanki, S. K. 2004, A&A, 414, 1109Lites, B. W., Scharmer, G. B., Berger, T. E., Title, A. M. 2004, Sol. Phy., 221, 65Louis, R. E., Beck, C., Ichimoto, K. 2014, A&A, in pressLouis, R. E., Bellot Rubio, L. R., Mathew, S. K., Venkatakrishnan, P. 2009, ApJ, 704, L29Louis, R. E., Bayanna, A. R., Mathew, S. K., Venkatakrishnan, P. 2008, Sol. Phy., 252, 43Magara, T. 2010, ApJ, 715, L40 13 –Morita, S, Shibata, K., Ueno, S., et al. 2010, PASJ, 62, 901Nakamura, N., Shibata, K., Isobe, H. 2012, ApJ, 761, 87Nishizuka, N., Hayashi, Y., Tanabe, H., Kuwahata, A., Kaminou, Y., Ono, Y., Inomoto, M.,Shimizu, T. 2012, ApJ, 756,152Nishizuka, N., Nakamura, T., Kawate, T., Singh, K. P. A., & Shibata, K. 2011, ApJ, 731, 43Nishizuka, N., Shimizu, M., Nakamura, T. et al. 2008, ApJL, 683, L83Reardon, K., Tritschler, A., Katsukawa, Y. 2013, ApJ, 779, 143Roy, J. 1973, Sol. Phy., 28,95Ruiz Cobo, B., Asensio Ramos, A. 2013, A&A, 549, L4Sakai, J. I., Smith, P. D. 2008, ApJ, 687, L127Scharmer, G. B., Henriques, V. M. J., Kiselman, D., de la Cruz Rodríguez, J. 2011, Science, 333,316Scharmer, G. B., de la Cruz Rodriguez, J., Sütterlin, P., Henriques, V. M. J. 2013, A&A, 553, 63Scharmer, G. B., Bjelksjo, K., Korhonen, T. K., Lindberg, B., & Petterson, B. 2003, Proc.SPIE, 4853, 341Schüssler, M. & Vögler, A. 2006, ApJ, 641, L73Selbing, J., 2005, Master thesis, Stockholm ObservatoryShibata, K., Nakamura, T. et al. 2007, Science, 318, 1591Shibata et al. 1992, PASJ, 44, 123Shimizu, T. 2011, ApJ, 738, 83Shimizu et al. 2009, ApJ, 696, 66Singh, K. A. P., Isobe, H., Nishida, K., Shibata, K. 2012a, ApJ, 760, 28Singh, K. A. P., Isobe, H., Nishizuka, N., Nishida, K., Shibata, K. 2012b, ApJ, 759, 33Tiwari, S. K., van Noort, M.; Lagg, A., Solanki, S. K. 2013, A&A, in pressTitle, A. M. & Rosenberg, W. J. 1981, Opt. Eng., 20, 815Yan, L, He, J, Xia, L., Jiao, F. 2015, ApJ, 804, 69 14 –Yokoyama, T., Shibata, K., 1995, Nature, 375, 42Yurchyshyn, V., Abramenko, V., Kosovichev, A., Goode, P., 2014, ApJ, 787,58Zakharov, V., Hirzberger, J., Riethmüller, T. L., Solanki, S. K., Kobel, P. 2008, A&A, 488, L17
This preprint was prepared with the AAS L A TEX macros v5.2.
15 –Fig. 4.— Vectors showing the horizontal magnetic field structure overplotted on an intensity image. 16 –Fig. 5.— Left: Ca II H wing image. An intensity threshold has been chosen to highlight thepenumbral intrusion. Right : Ca II H core image. The white contours correspond to a continuum(6302 Å) threshold, the orange contour outlines the opposite polarity patch. All images wererecorded around 08:44:55 UT. 17 – L O S Fig. 6.— Schematic picture of λλ