1: \documentclass{emulateapj}
2: \usepackage{graphicx}
3: \usepackage{apjfonts}
4:
5:
6: \slugcomment{To appear in ApJ Letters}
7: \shorttitle{Spectropolarimetry of dark-cored penumbral filaments}
8: \shortauthors{Bellot Rubio et al.}
9: \begin{document}
10:
11: \title{Vector spectropolarimetry of dark-cored penumbral filaments with
12: {\em Hinode}}
13:
14: \author{L.R.\ Bellot Rubio\altaffilmark{1}, S.\ Tsuneta\altaffilmark{2}, K.\
15: Ichimoto\altaffilmark{2}, Y.\ Katsukawa\altaffilmark{2}, B.W.\
16: Lites\altaffilmark{3}, \\ S.\ Nagata\altaffilmark{4}, T.\
17: Shimizu\altaffilmark{5}, R.A.\ Shine\altaffilmark{6}, Y.\
18: Suematsu\altaffilmark{2}, T.D.\ Tarbell\altaffilmark{6}, \\ A.M.\
19: Title\altaffilmark{6}, and J.C.\ del Toro Iniesta\altaffilmark{1} }
20:
21: \altaffiltext{1}{Instituto de Astrof\'{\i}sica de
22: Andaluc\'{\i}a (CSIC), Apdo.\ de Correos 3004, 18080 Granada, Spain}
23:
24: \altaffiltext{2}{National Astronomical Observatory of Japan,
25: 2-21-1 Osawa, Mitaka, Tokyo 181-8588, Japan}
26:
27: \altaffiltext{3}{High Altitude
28: Observatory, NCAR, 3080 Center Green Dr.\ CG-1, Boulder, CO 80301, USA}
29:
30: \altaffiltext{4}{Hida Observatory, Kyoto University, Takayama, Gifu 506-1314,
31: Japan}
32:
33: \altaffiltext{5}{Institute of Space and Astronautical Science, JAXA,
34: Sagamihara, Kanagawa 229-8510, Japan}
35:
36: \altaffiltext{6}{Lockheed Martin Solar
37: and Astrophysics Laboratory, Bldg.\ 252, 3251 Hanover St., Palo Alto, CA
38: 94304, USA}
39:
40: \begin{abstract}
41: We present spectropolarimetric measurements of dark-cored penumbral filaments
42: taken with Hinode at a resolution of 0\farcs3. Our observations demonstrate
43: that dark-cored filaments are more prominent in polarized light than in
44: continuum intensity. Far from disk center, the Stokes profiles emerging from
45: these structures are very asymmetric and show evidence for magnetic fields of
46: different inclinations along the line of sight, together with strong Evershed
47: flows of at least 6-7 km~s$^{-1}$. In sunspots closer to disk center,
48: dark-cored penumbral filaments exhibit regular Stokes profiles with little
49: asymmetries due to the vanishing line-of-sight component of the
50: horizontal Evershed flow. An inversion of the observed spectra indicates that
51: the magnetic field is weaker and more inclined in the dark cores as compared
52: with the surrounding bright structures. This is compatible with the idea that
53: dark-cored filaments are the manifestation of flux tubes carrying hot Evershed
54: flows.
55: \end{abstract}
56:
57: \keywords{sunspots -- Sun: magnetic fields -- Sun: photosphere
58: -- polarization}
59:
60: \section{Introduction}
61: \label{sec:intro}
62:
63: Since their discovery by Scharmer et al.\ (2002), dark-cored penumbral
64: filaments have attracted much interest because they could reveal the primary
65: mode of magnetoconvection taking place in sunspot penumbrae. The dark cores
66: of penumbral filaments are narrow ($\leq 200$~km) lanes running along the
67: filament axes, surrounded by two lateral brightenings. It has been proposed
68: that dark-cored penumbral filaments are the manifestation of magnetic flux
69: tubes (S\"utterlin et al.\ 2004; Bellot Rubio et al.\ 2005; Borrero 2007) or
70: field-free gaps in the penumbra (Spruit \& Scharmer 2006; Scharmer \& Spruit
71: 2006), but no consensus has been reached yet.
72:
73: The morphology and temporal evolution of these structures are well known from
74: high-resolution (0\farcs1--0\farcs2) filtergrams. Imaging observations show
75: the two lateral brightenings to move with the same speed and direction as a
76: single entity. Spectroscopic measurements at 0\farcs2 resolution (Bellot Rubio
77: et al.\ 2005; Rimmele \& Marino 2006) have revealed that the Evershed flow is
78: concentrated preferentially in the dark cores, where it seems to be an upflow
79: near the umbra/penumbra boundary. They also provide some evidence that the
80: dark cores possess weaker magnetic fields than their lateral brightenings.
81: One-wavelength longitudinal magnetograms at 0\farcs2 resolution (Langhans et
82: al.\ 2005; Langhans et al.\ 2007) show lower circular polarization signals in
83: the dark cores as compared with the bright edges, supporting the conclusions
84: derived from spectroscopy. However, high-precision vector magnetic field
85: measurements of these structures have never been made because of the
86: relatively modest angular resolution attained by ground-based
87: spectropolarimeters.
88:
89: In this Letter, we report on the first spectropolarimetric measurements of
90: dark-cored penumbral filaments. The observations were taken with Hinode's
91: Solar Optical Telescope at a resolution of 0\farcs3. Dark-cored filaments turn
92: out to be prominent structures in polarized light, showing anomalous, very
93: asymmetric Stokes $V$ profiles in sunspots outside the disk center. In
94: addition, we find that the dark cores possess weaker and more horizontal
95: fields than their surroundings.
96:
97: \section{Observations}
98: \label{sec:obs}
99:
100: We used the spectropolarimeter (SP; Lites et al.\ 2001) of the Solar Optical
101: Telescope (SOT) aboard Hinode (Kosugi et al.\ 2007) to observe the \ion{Fe}{1}
102: 630.2~nm lines emerging from AR 10923 on November 10 and 14, 2006. On those
103: days the spot was located at heliocentric angles of $50^\circ$ and $8^\circ$,
104: respectively. The four Stokes profiles of the two \ion{Fe}{1} lines were
105: measured by scanning the spectrograph slit across the spot in steps of
106: 0\farcs148, with a wavelength sampling of 2.153~pm~pixel$^{-1}$. The SP slit
107: width and the CCD pixel size are 0\farcs16. The integration time per slit
108: position was 4.8~s, resulting in a noise level of $10^{-3} I_{\rm c}$ as
109: measured in the continuum of Stokes $Q$, $U$ and $V$. Image stabilization down
110: to 0\farcs01 rms was provided by the SOT tip-tilt system (Shimizu et al.\
111: 2007). These measurements reach a spatial resolution of about 0\farcs32, while
112: the diffraction limit of the SOT is $\lambda/D = 0\farcs25$ at 630.2~nm.
113:
114: The data have been corrected for dark current, flat field, and instrumental
115: polarization. We estimate that residual instrumental crosstalk is not larger
116: than a few $10^{-4} I_{\rm c}$. The calibrated Stokes profiles have been used
117: to create maps of continuum intensity, $I_{\rm c}$, total circular
118: polarization, ${\rm TCP} = \int |V| \, {\rm d}\lambda$, total linear
119: polarization, ${\rm TLP} = \int (Q^2+U^2)^{1/2} \, {\rm d}\lambda$, total
120: polarization, ${\rm TP} = \int (Q^2+U^2+V^2)^{1/2} \, {\rm d}\lambda$, and
121: area asymmetry, ${\delta A} = \int V \, {\rm d}\lambda/ \int |V| \, {\rm
122: d}\lambda$, where the integrals extend over a wavelength range of 0.1098~nm to
123: encompass the 630.25~nm line.
124:
125:
126: \section{General appearance and Stokes spectra}
127: Figure~\ref{fig1} shows two $12\farcs8 \times 12\arcsec$ regions in the
128: limb-side and center-side penumbra of AR 10923 at a heliocentric angle of
129: 50$^\circ$. The small white arrows mark the positions of six dark-cored
130: penumbral filaments. As noticed by S\"utterlin et al.\ (2004), the dark cores
131: show higher continuum intensity contrasts in the center-side penumbra (right
132: panels).
133:
134: We find the dark-cored filaments to be much more prominent in polarized light
135: than in continuum intensity (compare, e.g., the first and fourth panels of
136: Figure~\ref{fig1}). This is also the case for the November 14 observations,
137: when the spot was at only $8^\circ$ from disk center. The fact that
138: dark-cored filaments show up more prominently in polarized light implies that
139: there is no one-to-one correspondence between the underlying magnetic fields
140: and their continuum intensity signatures. Thus, a complete characterization
141: of the penumbra cannot be based solely on intensity measurements.
142:
143: The dark cores exhibit smaller circular and linear polarization signals than
144: the bright structures surrounding them, which is highly suggestive of weaker
145: fields. Contrary to one-wavelength magnetograms, the TCP, TLP and TP signals
146: are not affected by velocities, but no unambiguous conclusion on field
147: strengths can be drawn from simple inspections of these parameters because
148: they still depend on temperature.
149:
150: An interesting fact is that the dark-cored filaments are detected as
151: structures of enhanced Stokes $V$ area asymmetries in the limb-side penumbra
152: of spots away from disk center (bottom left panel of Fig.~\ref{fig1}). On the
153: center side they are not distinct features in area asymmetry. Very likely,
154: this is the result of projection effects that shift the optical depth scale
155: toward higher geometrical heights and thus place the Evershed flow deeper in
156: the line forming region or even outside of it. The enhanced area asymmetries
157: imply the existence of gradients/discontinuities of velocity (and possibly
158: other atmospheric parameters) along the line of sight (LOS) which are
159: stronger in the filaments than in the surroundings.
160:
161:
162: \begin{figure}
163: \centering
164: \resizebox{1.01\hsize}{!}{\includegraphics{./f1.eps} }
165: \caption{Dark-cored penumbral filaments observed on the limb side (left) and
166: center side (right) of AR 10923 on November 10, 2006, between 16:01 and 17:26
167: UT. The heliocentric angle of the spot was $50^\circ$. North is up and West to
168: the right. Minor tick marks are separated by 0\farcs5. The black arrows point
169: to disk center. From top to bottom: maps of continuum intensity, total
170: circular polarization, total linear polarization, total polarization, and area
171: asymmetry for the two regions. White arrows indicate the position of
172: dark-cored filaments.}
173: \label{fig1}
174: \end{figure}
175:
176: \begin{figure*}[t]
177: \centering \resizebox{.19\hsize}{!}{\includegraphics[bb=107 352 326
178: 540]{./f2a.ps}}
179: \resizebox{.8\hsize}{!}{\includegraphics[bb=60 365 702
180: 515]{./f2b.ps} }
181: \caption{\ion{Fe}{1}~630.15 and 630.25~nm Stokes $V$ profiles emerging from a
182: dark-cored filament in the limb-side penumbra and the surrounding umbra as a
183: function of distance to the filament head. The distance increases in steps of
184: 0\farcs48 from right to left (total polarization map) and from left to right
185: (Stokes spectra panels). The umbra is sampled 0\farcs64 away from the dark
186: core; the corresponding Stokes profiles are shifted upward by 2\% for better
187: visibility. The dark-cored filament makes an angle of $7^\circ$ to the line of
188: symmetry. The polarity of the spot is negative.}
189: \label{fig2}
190: \end{figure*}
191:
192: \begin{figure}
193: \centering
194: \resizebox{.93\hsize}{!}{\includegraphics[bb=20 60 450 690]
195: {./f3.eps}}
196: \caption{Observed (dots) and best-fit (solid lines) profiles from a
197: one-component inversion with SIR, for a pixel inside a dark core (left) and
198: its lateral brightening (right). Both structures were observed in the
199: limb-side penumbra of AR 10923 at $8^\circ$ from disk center
200: (Fig.~\ref{fig4}).}
201: \label{fig3}
202: \end{figure}
203:
204:
205: This conclusion is supported by the anomalous Stokes $V$ spectra emerging from
206: dark-cored filaments. Figure~\ref{fig2} displays the circular polarization
207: profiles observed along the dark core of a limb-side filament near the inner
208: penumbral boundary. For comparison, we also plot the spectra from the
209: surrounding umbra at the same radial distances (cf.\ the dots in the left
210: panel of Fig.~\ref{fig2}). The bright penumbral grain that forms the head of
211: the filament exhibits regular Stokes $V$ profiles with the polarity of the
212: spot. Further outward along the dark core, the circular polarization profiles
213: show three lobes indicating the existence of opposite polarity fields along
214: the LOS, {\em one of which is strongly Doppler-shifted to the red}. Its
215: LOS velocity can roughly be estimated as $6-7$ km~s$^{-1}$ from the wavelength
216: separation between the redmost Stokes $V$ peaks. This magnetic field component
217: gradually takes over, giving rise to one-lobed Stokes $V$ spectra at a distance
218: of $d \sim 1\farcs92$ and two-lobed profiles of opposite polarity farther than
219: $d = 2\farcs4$. In contrast, the circular polarization profiles emerging from
220: the adjacent background umbra keep the polarity of the spot at all radial
221: distances, and do not exhibit anomalous shapes.
222:
223: The multi-lobed Stokes $V$ spectra observed in limb-side filaments at 0\farcs3
224: resemble those measured from the ground at much lower angular resolution. In
225: our data, we can for the first time distinguish between the dark cores and the
226: lateral brightenings, each of which may differ in their magnetic properties.
227: Thus, these measurements constrain horizontal variations of the physical
228: parameters relative to those along the LOS much more tightly than earlier
229: observations. Both Fig.~2 and the observed Stokes $V$ asymmetries indicate a
230: complex magnetic topology in the dark-cored filaments, with discontinuities or
231: gradients along the LOS, and possibly also lateral gradients. To explain the
232: observations, the height variations/discontinuities of atmospheric parameters
233: must be present well within the Stokes $V$ formation region (i.e., significantly
234: above the $\tau = 1$ level; Cabrera Solana et al.\ 2005), as otherwise they would
235: be unable to produce the multi-lobed profiles depicted in Fig.~\ref{fig2}.
236: Finally, we note that one-wavelength longitudinal magnetograms of complex
237: spectra such as the ones discussed here may even fail in retrieving the
238: correct polarity of the field, depending on the exact wavelength of the
239: observations.
240:
241:
242: \section{Magnetic fields of dark-cored filaments}
243: We estimate the vector magnetic field of dark-cored penumbral filaments and
244: their surroundings by inverting the observed spectra with the SIR code (Ruiz
245: Cobo \& del Toro Iniesta 1992). We restrict our analysis to the November 14
246: data set, since on that day the Stokes profiles show negligible asymmetries in
247: the absence of strong LOS velocities (due to the proximity of the spot to disk
248: center). The spectra can be fitted remarkably well in terms of one-component
249: model atmospheres featuring constant (height-independent) vector magnetic
250: fields, LOS velocities, and microturbulent velocities (see Fig.~\ref{fig3}).
251: The temperature stratification is determined by changing the slope and
252: $T(\tau=1)$ value of the penumbral model of Bellot Rubio et al.\ (2006). In
253: total, the inversion returns the values of 8 free parameters. We assume zero
254: stray light contamination and unity magnetic filling factors.
255:
256: \begin{figure*}[t]
257: \centering
258: \resizebox{.96\hsize}{!}{
259: \includegraphics[bb=105 395 517 620]{./f4a.ps}
260: \includegraphics[bb=105 395 517 620]{./f4b.ps}}
261: \resizebox{.96\hsize}{!}{
262: \includegraphics[bb=104 395 516 620]{./f4c.ps}
263: \includegraphics[bb=103 395 515 620]{./f4d.ps}}
264: \vspace*{1.5em}
265: \caption{{\em Top:} Continuum intensity (left) and total polarization
266: (right) maps of the limb-side penumbra of AR 10923 as observed by
267: Hinode/SP on November 14, 2006 between 16:30 and 17:15 UT. Four
268: dark-cored filaments are indicated with small horizontal arrows. The
269: long arrows point to disk center. {\em Bottom:} Maps of
270: magnetic field strength and inclination in the LRF, as
271: deduced from an inversion of the spectra. }
272: \vspace*{.5em}
273: \label{fig4}
274: \end{figure*}
275:
276:
277: Figure~\ref{fig4} displays continuum intensity and total polarization maps of
278: a small region in the limb-side penumbra of AR 10923 on November 14. At least
279: four dark-cored filaments can easily be recognized, especially in the
280: polarization map. Also shown in Fig.~\ref{fig4} are the field strengths and
281: inclinations resulting from the inversion, expressed in the local reference
282: frame (LRF). The dark cores stand out clearly as long, narrow streaks of
283: weaker ($\Delta B \sim 100-150$ G) and more inclined ($\Delta \gamma_{\rm LRF}
284: \sim 4^\circ$) fields than the lateral brightenings or surrounding structures.
285: In general, we find a very smooth magnetic topology in this part of the
286: penumbra, which is only altered by the presence of the dark cores. Note that
287: the inclination map shows extended patches in the penumbra where the polarity
288: of the magnetic field is opposite to that of the spot, i.e., field lines
289: returning back to the solar surface ($\gamma_{\rm LRF} <90^{\rm o}$). The
290: existence of such field lines has been inferred from inversions of Stokes
291: profiles at lower resolution, but this is the first time they are detected so
292: close to the umbra (see also Ichimoto et al.\ 2007). As pointed out by Bellot
293: Rubio (2007), the existence of field lines returning to the solar surface well
294: within the penumbra may pose a serious problem for the gappy model of Spruit
295: \& Scharmer (2006).
296:
297:
298: %
299: %________________________________________________________________
300:
301: \section{Conclusions}
302: \label{sec:con}
303:
304: The nearly diffraction-limited (0\farcs3) spectropolarimetric measurements
305: performed by Hinode reveal that dark-cored penumbral filaments are conspicuous
306: structures in polarized light, showing higher contrasts than in continuum
307: intensity. The total polarization signal emerging from the dark cores is
308: smaller than that observed in the lateral brightenings, which is highly
309: suggestive of lower field strengths. We have confirmed this hyphotesis by
310: carrying out an inversion of the spectra observed in a sunspot very close
311: to disk center.
312:
313: The dark cores possess $100-150$~G weaker fields than the lateral brightenings
314: and surrounding structures, i.e., they are not field-free. In addition, the
315: field is deduced to be more horizontal in the dark cores than in the lateral
316: brightenings, by some $4^\circ$. The field strength and inclination
317: differences are surprisingly small, which might indicate still insufficient
318: angular resolution and/or simplistic modeling (the inversions performed here
319: just provide {\em average} values of the atmospheric parameters along the
320: LOS). High-resolution magnetograph observations (Langhans et al.\ 2007)
321: suggest that the actual differences could be larger.
322:
323: Our results are consistent with the idea that dark-cored filaments represent
324: magnetic flux tubes embedded in an ambient field. Their properties coincide
325: with those inferred from uncombed inversions of visible and infrared lines
326: (weaker and more inclined fields in the tubes as compared with the background
327: atmosphere; large area asymmetries and multi-lobed Stokes $V$ profiles caused
328: by strong Evershed flows along the tubes; see, e.g., Beck 2006 and Borrero et
329: al.\ 2006). They also agree with the properties of moving penumbral tubes as
330: deduced from numerical simulations (Schlichenmaier et al.\ 1998;
331: Schlichenmaier 2002). According to the simulations, the field is weaker in the
332: tubes to maintain horizontal mechanical equilibrium. It is also more inclined
333: because the tubes rapidly bend when they reach the subadiabatic layers of the
334: photosphere during their ascent. In the moving tube simulations, Evershed
335: flows exceding 10 km~s$^{-1}$ are driven by gas pressure gradients that build
336: up as a consequence of the radiative cooling of the plasma. By contrast, no
337: mechanism capable of explaining Evershed flows of several km~s$^{-1}$ has been
338: identified in the gappy penumbral model as yet. The Eveshed flow is an
339: ingredient without which the observed Stokes $V$ shapes and area asymmetries
340: cannot be understood.
341:
342: Hopefully, further analyses of Hinode spectropolarimetric observations will
343: help distinguish between competing models of the penumbra (Bellot Rubio
344: 2007). Initial efforts in this direction are presented by Jur\v{c}ak et
345: al.\ (2007) and Ichimoto et al.\ (2007).
346:
347:
348: \acknowledgments
349:
350: Hinode is a Japanese mission developed and launched by ISAS/JAXA, with NAOJ as
351: domestic partner and NASA and STFC (UK) as international partners. It is
352: operated by these agencies in co-operation with ESA and NSC (Norway). This
353: work has been partially funded by the Spanish MEC through project
354: ESP2006-13030-C06-02.
355:
356:
357: \begin{thebibliography}{}
358:
359: \bibitem[]{} Beck, C.\ 2006, PhD thesis, Alberts-Ludwig Universit\"at,
360: Freiburg
361:
362: \bibitem[]{298} Bellot Rubio, L.R.\ 2007, in: Highlights of Spanish Astrophysics
363: IV, eds.\ F.\ Figueras et al.\ (Springer), in press [astro-ph/0611471]
364:
365: \bibitem[Bellot Rubio et al.(2005)]{2005A&A...443L...7B} Bellot Rubio,
366: L.R., Langhans, K., \& Schlichenmaier, R.\ 2005, \aap, 443, L7
367:
368: \bibitem[Bellot Rubio et al.(2006)]{2006A&A...453.1117B} Bellot Rubio,
369: L.R., Schlichenmaier, R., \& Tritschler, A.\ 2006, \aap, 453, 1117
370:
371: \bibitem[Borrero(2007)]{2007A&A...471..967B} Borrero, J.~M.\ 2007, \aap,
372: 471, 967
373:
374: \bibitem[Borrero et al.(2006)]{2006A&A...450..383B} Borrero, J.M.,
375: Solanki, S.K., Lagg, A., Socas-Navarro, H., \& Lites, B.\ 2006, \aap,
376: 450, 383
377:
378: \bibitem[Cabrera Solana et al.(2005)]{2005A&A...439..687C} Cabrera Solana,
379: D., Bellot Rubio, L.R., \& del Toro Iniesta, J.C.\ 2005, \aap, 439, 687
380:
381: \bibitem[2007]{ichimoto} Ichimoto, K.\ et al.\ 2007, PASJ, submitted
382:
383: \bibitem[2007]{jurcak} Jur\v{c}ak, J., Bellot Rubio, L.R., Ichimoto, K.,
384: et al.\ 2007, PASJ, submitted [arXiv:0707.1560]
385:
386: \bibitem[2007]{kosugi} Kosugi, T., Matsuzaki, K., Sakao, T., et al.\
387: 2007, \solphys, submitted
388:
389: \bibitem[2005]{langhans} Langhans, K., Scharmer, G.B., Kiselman, D.,
390: L\"ofdahl, M.G., \& Berger, T.E.\ 2005, \aap, 436, 1087
391:
392: \bibitem[2007]{langhans} Langhans, K., Scharmer, G.B., Kiselman, D.,
393: L\"ofdahl, M.G.\ 2007, \aap, 464, 763
394:
395: \bibitem[Lites et al.(2001)]{2001ASPC..236...33L} Lites, B.W., Elmore,
396: D.F., \& Streander, K.V.\ 2001, ASP Conf.\ Ser.\ 236: Advanced Solar
397: Polarimetry, 33
398:
399: \bibitem[Rimmele \& Marino(2006)]{2006ApJ...646..593R} Rimmele, T., \&
400: Marino, J.\ 2006, \apj, 646, 593
401:
402: \bibitem[Ruiz Cobo \& del Toro Iniesta(1992)]{1992ApJ...398..375R} Ruiz
403: Cobo, B., \& del Toro Iniesta, J.C.\ 1992, \apj, 398, 375
404:
405: \bibitem[Scharmer \& Spruit(2006)]{2006A&A...460..605S} Scharmer, G.~B., \&
406: Spruit, H.~C.\ 2006, \aap, 460, 605
407:
408: \bibitem[2002]{scharmer} Scharmer, G.B.,
409: Gudiksen, B.V., Kiselman, D., L\"ofdahl, M.G., \& Rouppe van der
410: Voort, L.H.M.\ 2002, \nat, 420, 151
411:
412: \bibitem[Schlichenmaier(2002)]{2002AN....323..303S} Schlichenmaier, R.\
413: 2002, Astronomische Nachrichten, 323, 303
414:
415: \bibitem[Schlichenmaier et al.(1998)]{1998A&A...337..897S} Schlichenmaier,
416: R., Jahn, K., \& Schmidt, H.~U.\ 1998, \aap, 337, 897
417:
418: \bibitem[]{351} Shimizu, T., Nagata, S., Tsuneta, S., et al.\ 2007, \solphys,
419: submitted
420:
421: \bibitem[Spruit \& Scharmer(2006)]{2006A&A...447..343S} Spruit, H.~C., \&
422: Scharmer, G.~B.\ 2006, \aap, 447, 343
423:
424: \bibitem[2004]{sutterlin} S{\" u}tterlin,
425: P., Bellot Rubio, L.R., \& Schlichenmaier, R.\ 2004, \aap, 424, 1049
426:
427: %\bibitem[2007]{360} Tsuneta, S., Suematsu, Y., Ichimoto, K., et al.\ 2007,
428: %\solphys, submitted
429:
430: \end{thebibliography}
431:
432: \end{document}
433:
434: \begin{figure}
435: \centering
436: \resizebox{!}{.8\vsize}{\includegraphics{f1.eps} }
437: \caption{Dark-cored penumbral filaments observed on the limb side (left) and
438: center side (right) of AR 10923 on November 10, 2006, between 16:01 and 17:26
439: UT. The heliocentric angle of the spot was $50^\circ$. North is up and West to
440: the right. Minor tick marks are separated by 0\farcs5. The black arrows point
441: to disk center. From top to bottom: maps of continuum intensity, total
442: circular polarization, total linear polarization, total polarization, and area
443: asymmetry for the two regions. White arrows indicate the position of
444: dark-cored filaments.}
445: \label{fig1}
446: \end{figure}
447:
448:
449:
450: \begin{figure}[t]
451: \centering \resizebox{.19\hsize}{!}{\includegraphics[bb=107 352 326
452: 540]{f2a.ps}}
453: \resizebox{.8\hsize}{!}{\includegraphics[bb=60 365 702 515]
454: {f2b.ps} }
455: \caption{\ion{Fe}{1}~630.15 and 630.25~nm Stokes $V$ profiles emerging from a
456: dark-cored filament in the limb-side penumbra (black) and the surrounding umbra
457: (red) as a function of distance to the filament head. The distance increases
458: in steps of 0\farcs48 from right to left (total polarization map) and from
459: left to right (Stokes spectra panels). The umbra is sampled 0\farcs64 away
460: from the dark core; the corresponding Stokes profiles are shifted upward by
461: 2\% for better visibility. The dark-cored filament makes an angle of
462: $7^\circ$ to the line of symmetry. The polarity of the spot is negative.}
463: \label{fig2}
464: \end{figure}
465:
466: \begin{figure}
467: \centering
468: \resizebox{.7\hsize}{!}{\includegraphics[bb=20 60 450 690]
469: {f3.eps}}
470: \caption{Observed (dots) and best-fit (solid lines) profiles from a
471: one-component inversion with SIR, for a pixel inside a dark core (left) and
472: its lateral brightening (right). Both structures were observed in the
473: limb-side penumbra of AR 10923 at $8^\circ$ from disk center (Fig.~\ref{fig4}).}
474: \label{fig3}
475: \end{figure}
476:
477: \begin{figure*}[t]
478: \centering
479: \resizebox{1\hsize}{!}{
480: \includegraphics[bb=95 380 527 620]{f4a.ps}
481: \includegraphics[bb=95 380 527 620]{f4b.ps}}
482: \resizebox{1\hsize}{!}{
483: \includegraphics[bb=95 380 527 620]{f4c.ps}
484: \includegraphics[bb=95 380 527 620]{f4d.ps}}
485: \vspace*{1.5em}
486: \caption{{\em Top:} Continuum intensity (left) and total polarization
487: (right) maps of the limb-side penumbra of AR 10923 as observed by
488: Hinode/SP on November 14, 2006 between 16:30 and 17:15 UT. Four
489: dark-cored filaments are indicated with small horizontal arrows. The
490: long arrows point to disk center. {\em Bottom:} Maps of
491: magnetic field strength and inclination in the LRF, as
492: deduced from an inversion of the spectra.
493: \vspace*{1em}}
494: \label{fig4}
495: \end{figure*}
496:
497:
498:
499:
500: \end{document}
501:
502:
503: