astro-ph0211454/ms.tex
1: % Letter with the main results of the speckle reconstructed
2: %	quiet Sun magnetograms
3: %
4: %
5: % Modification History:
6: %
7: %	ver0	written by Franz. October 2002
8: %			(aa latex)
9: %
10: %	ver1	translated to AAS latex ... and else by jos
11: %
12: %	ver2	includes Franz modifications
13: %		includes Hector suggestions
14: %
15: %	ver3	1st version submitted to ApJ on Oct 24, 2002
16: %
17: %	ver 4 	2nd version submitted to ApJL
18: %
19: 
20: 
21: %%%%%%%%%%%
22: 
23: %\documentclass[12pt,preprint]{aastex}
24: 
25: %% manuscript produces a one-column, double-spaced document:
26: 
27: % \documentclass[manuscript]{aastex}
28: 
29: %% preprint2 produces a double-column, single-spaced document:
30: 
31: % \documentclass[preprint2]{aastex}
32: 
33: 
34: %% emulateapj5: closest possible to apj.
35: \documentclass{aastex}
36: \usepackage{emulateapj5}
37: \usepackage{apjfonts}
38: 
39: 
40: %%%%%%%%%%%%%
41: 
42: 
43: 	% required for astrobib
44: 	\newcommand{\citeN}[1]{\citeauthor{#1} (\citeyear{#1})}
45: 	\newcommand{\citeNP}[1]{\citeauthor{#1} \citeyear{#1}}
46: 	\newcommand{\citeyearNP}[1]{\citeyear{#1}}
47: 	\newcommand{\citeANP}[1]{\citeauthor{#1}}
48: 	%
49: 	\newcommand\lineone{\rm Fe~{\sc i}~$\lambda$6301.5~\AA}
50: 	\newcommand\linetwo{\rm Fe~{\sc i}~$\lambda$6302.5~\AA}
51: 	\newcommand\josa{JOSA}
52: 
53: 
54: 
55: %%%%%%%%%%%%%
56: 
57: %\slugcomment{ver4}
58: \shorttitle{Quiet Sun magnetic fields at high spatial resolution}
59: \shortauthors{Dom\'\i nguez Cerde\~na et al.}
60: 
61: 
62: 
63: %%%%%%%%%%%%
64: \begin{document}
65: %   \headnote{Letter}
66:    \title{Quiet Sun magnetic fields at high spatial resolution}
67: 
68:    \author{I. Dom\'\i nguez Cerde\~na}
69:    \affil{Universit\"ats-Sternwarte,
70:               Geismarlandstra\ss e 11, D-37083 G\"ottingen, Germany}
71:    \email{ita@uni-sw.gwdg.de}
72: 
73:    \author{F. Kneer}
74:    \affil{Universit\"ats-Sternwarte,
75:               Geismarlandstra\ss e 11, D-37083 G\"ottingen, Germany}
76:    \email{kneer@uni-sw.gwdg.de}
77:           \and
78:     \author{J. S\'anchez Almeida}
79:     \affil{Instituto de Astrof\'\i sica de Canarias, 
80:               E-38200 La Laguna, Spain}
81:    \email{jos@ll.iac.es}
82:  
83: 
84: \begin{abstract}
85: We present spectro-polarimetric observations of Inter-Network magnetic fields 
86: at the solar disk center. A Fabry-Perot spectrometer 
87: was used to scan  the two Fe\,{\sc i} lines at $\lambda$6301.5~\AA\ and $\lambda$6302.5~\AA .
88: High spatial resolution  (0\farcs5) magnetograms were obtained after
89: speckle reconstruction. The patches with magnetic fields above noise 
90: cover approximately  
91: 45\,\% of the observed area.  Such large coverage renders a mean
92: unsigned magnetic flux density of  some 20 G (or 20 Mx cm$^{-2}$), which exceeds 
93: all previous measurements.
94: Magnetic signals occur predominantly in intergranular spaces. 
95: The systematic difference between the flux densities measured in the two 
96: iron lines leads to the conclusion that, typically, we detect structures with
97: intrinsic field strengths larger than 1\,kG occupying only
98: 2\% of the surface.
99: %
100: \end{abstract}
101: \keywords{
102:           Sun: magnetic fields --
103:           Sun: photosphere}
104: 
105: 
106: %
107: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
108: %
109: 
110: 
111: \section{Introduction\label{intro}}
112: 
113: 
114: We study magnetic fields of the quiet Sun. Our analysis will exclude the 
115: magnetism in photospheric network regions known to contain strong kG 
116: magnetic fields (e.g., \citeNP{ste94}). We will concentrate on the low flux features 
117: in the interior of 
118: the network which we will call, for brevity, IN fields (Inter-Network fields). 
119: Magnetic fields within the network cells were
120: already discovered in the seventies (\citeNP{liv75};
121: \citeNP{smi75}). 
122: During the last decade, with improved observational and diagnostic
123: capabilities, these IN fields have obtained increasing attention. 
124: They will allow to test our ideas on the magnetic field generation
125: by convective processes, a natural step towards  understanding the principles
126: of astrophysical dynamos 
127: (e.g. \citeNP{cat99a}; \citeNP{emo01}).
128: In addition, it has been found that the very quiet IN areas
129: harbor a sizeable fraction of the solar magnetic flux 
130: (see, e.g., \citeNP{ste82}; \citeNP{yi93};
131: \citeNP{soc02}).
132: Such estimates are based on lower limits to the still unknown amount 
133: of flux,
134: which suggests the potential importance of the IN component of
135: the solar magnetism. 
136: 
137: So far only a fraction of the IN field has been detected. This conclusion
138: follows from the presence of unresolved mixed polarities in  
139: 1\arcsec\ angular resolution  observations
140: (e.g., \citeNP{san96}; \citeNP{san00}; \citeNP{lit02}). The mixing
141: of polarities reduces the polarization,
142: making the magnetic structures difficult to identify.
143: Consequently,
144: it is to be expected finding more sites with magnetic
145: fields when increasing the spatial resolution. This idea triggered 
146: our investigation: we aimed at observing the IN fields with the highest possible
147: spatial resolution yet keeping a good polarimetric sensitivity. 
148: 
149: Many of the recent observational studies find that the IN fields have 
150: magnetic field strengths substantially lower than 1~kG
151: 	(\citeNP{kel94};
152: 	\citeNP{lin95};
153: 	\citeNP{lin99};
154: 	\citeNP{bia98};
155: 	\citeNP{bia99};
156: 	\citeNP{kho02}).
157: Yet, on the 
158: basis of different observing and interpretation techniques, 
159: one can find in IN areas strong kG magnetic 
160: fields (\citeNP{gro96};
161: 	\citeNP{sig99};
162: 	\citeNP{san00};
163: 	\citeNP{soc02}).
164: This seeming discrepancy probably
165: points out that the IN regions present a continuous distribution of field strengths. 
166: Depending on the specificities of the diagnostic technique,
167: one selects only a particular part of such distribution (see \citeNP{cat99a};
168: 	\citeNP{san00}).
169: 
170: 
171: In this study we confirm the presence of kG IN magnetic fields. In addition,
172: we detect an amount of unsigned magnetic flux which exceeded any value previously 
173: reported in the literature. 
174: Our conclusions are based on spectro-polarimetric observations achieving both high spatial 
175: resolution (0\farcs5) and fair sensitivity (20 G). 
176: In this 
177: letter we give a short description of the observations and the data 
178: analysis. Our aim is to communicate the results bearing on strong IN 
179: fields and large flux densities.
180: %
181: A detailed discussion, including additional properties on the time
182: evolution,
183: will be given in a forthcoming more extended contribution. 
184: 
185: %                                      
186: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
187: %
188: 
189: \section{Observations and data reduction}
190: 
191: The observations were obtained in April 2002 with the {\em G\"ottingen}
192: Fabry-Perot Interferometer
193: (FPI; \citeNP{ben92}; \citeNP{ben93})
194: mounted at the Vacuum
195: Tower Telescope of the Observatorio del Teide (Tenerife). The seeing was excellent 
196: (Fried parameter $r_0$\,=\,13--14\,cm). A very quiet region 
197: near disk center was selected using G band video images to avoid network 
198: regions. The observational setup is described in \citeN{kos01}, which includes a 
199: Stokes $V$ polarimeter. While scanning across a spectral 
200: region, the spectrometer takes two-dimensional narrow-band images with short exposure times 
201: and, strictly simultaneously, broadband images from the same Field Of View (FOV).
202: 
203: The data for the present study stem from a 17\,mininutes time series of spectral scans 
204: in the 6302\,\AA\ region, containing the \lineone\ line (Land\'e 
205: factor $g=1.67$), the \linetwo\ line ($g=2.5$), and 
206: the telluric O$_2$ 6302.76\,\AA\ line.
207: The FWHM of the FPI and the sampling interval were 44\,m\AA\ and 
208: 32\,m\AA, respectively. Each FPI scan consists of 28 
209: wavelength positions for the two iron lines, and 
210: 5 wavelength positions for the telluric line. 
211: The duration of a 
212: spectral scan is 35\,s,  plus 15s needed for storage 
213: onto hard disk. The FOV corresponds to 15\arcsec$\times$25\arcsec
214: on the Sun, sampled with a pixel size of 
215: 0\farcs1$\times$0\farcs1.
216: The data reduction includes subtraction of dark current, flat fielding, and
217: speckle reconstruction of the broadband images 
218: (\citeNP{deb92};  \citeNP{deb96})
219: using the spectral ratio method (\citeNP{von84}) and the speckle
220: masking method (\citeNP{wei77}). The reconstruction of the
221: narrow-band images proceeds the way proposed by 
222: \citeauthor{kel92b} (\citeyear{kel92b}; see also \citeNP{kri99};
223: \citeNP{hir01}).
224: %
225: We use the 
226: code by \citeN{jan03} with
227: minor modifications.
228: %
229: To have the noise filtering independent of the seeing, which may vary during a 
230: spectral scan, we apply a filter which limits the spatial resolution of the 
231: reconstructed spectro-polarimetric images to 0\farcs5. 
232: To further suppress noise, the 
233: images  were processed with a 5$\times$5 pixel boxcar 
234: smoothing after the filtering. 
235: The FPI simultaneously collects the left and right circularly
236: polarized beams onto different areas of the CCD. They are added and 
237: subtracted to obtain the Stokes $I$ and $V$ profiles. This
238: task demands a careful superposition of the two beams,
239: which we carry out after a  
240: sub-pixel interpolation. 
241: The spatial filters were not applied to the broad-band images.
242: 
243: 
244: \subsection{Calibration of the magnetograms}
245: 
246: 	Magnetic flux densities are computed starting from the 
247: magnetograph equation, which
248: relates the circular polarization $V$ and
249: the longitudinal component
250: of the magnetic field $B$ (e.g., \citeNP{unn56}; \citeNP{lan92}),
251: \begin{equation} 
252: V(\lambda) = C(\lambda)\ B, 
253: \label{eeqmag} 
254: \end{equation} 
255: %
256: with the calibration constant $C(\lambda)$ proportional to $dI(\lambda)/d\lambda$.
257: First, the
258: derivative of Stokes $I$ required to evaluate 
259: $C(\lambda)$ is computed numerically for each iron line.
260: Then, the wavelengths of the two extrema of the 
261: derivative, plus the two wavelengths immediately adjacent to 
262: each one of them, are selected. 
263: The flux density of each solar point, $B_\mathrm{eff}$, is obtained solving 
264: equation (\ref{eeqmag}) for these six wavelengths $\lambda_i$.
265: The least-squares fit renders
266: \begin{equation}
267: B_\mathrm{eff}=\sum_i V(\lambda_i)\ C(\lambda_i)\Big/\sum_i C^{2}(\lambda_i).
268: \label{llest}
269: \end{equation} 
270: Obviously, if 
271: the observed Stokes $I$ and $V$ profiles satisfy equation (\ref{eeqmag}) 
272: then $B_\mathrm{eff}=B$.
273: In general this is not the case and $B_\mathrm{eff}$ represents a mere
274: biased estimate of the true flux density. 
275: 
276: 	Two main sources of noise limit the precision of $B_\mathrm{eff}$.
277: The measure is affected by the random noise of the 
278: Stokes $V$ spectra.
279: We estimate its influence applying to equation (\ref{llest}) the law of propagation 
280: of errors (e.g., \citeNP{mar71}). Assuming
281: the Stokes $V$ noise to be constant and equal to $0.5\%I_c$ ($I_c$ being 
282: the continuum intensity),
283: the typical value for the random error of $B_\mathrm{eff}$ turns out to be
284: 20 G (Table \ref{table}; 1~G~$\equiv$~1~Mx~cm$^{-2}$).
285: %
286: The adopted $V$-noise comes from the
287: r.m.s. fluctuations of the Stokes $V$ profiles when $B_\mathrm{eff}\longrightarrow 0$.
288: A second source of error affecting $B_\mathrm{eff}$ is due to the
289: contamination of the polarization signals with intensity (the so-called
290: $I$-to-$V$ crosstalk). It can be created in many ways during the reduction
291: procedure (e.g., errors in setting the continuum intensity level). 
292: Since $V \ll I$, a small crosstalk
293: compromises the polarimetric accuracy of the measurements. Fortunately,
294: the procedure that we employ to determine $B_\mathrm{eff}$ is 
295: insensitive to this contamination. The calibration constant $C(\lambda)$
296: has different signs in the two wings of the spectral lines, consequently,
297: any symmetric contribution to $V$ (e.g., the contamination with 
298: Stokes $I$) leaves only a small residual in equation (\ref{llest}). 
299: Using quiet Sun Stokes $I$ profiles,
300: we deduce a residual of at most a few G, which is
301: regarded as negligible.
302: %                                     
303: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
304: %
305: 
306: \section{Results and discussion}
307: %
308: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
309: 
310: \subsection{Magnetograms}
311: %
312: 
313: 
314: 
315: 
316: Figure \ref{fig1} presents two magnetograms taken simultaneously in the two iron 
317: lines. They exhibit a {\em salt and pepper} pattern. The close 
318: similarity of the two magnetograms demonstrates that we have found magnetic 
319: field signals above 
320: noise. We sought and found further evidences for the consistency
321: of the magnetograms. Among them (a) 
322: most of the patches have sizes of 
323: the resolution limit or larger, and (b)
324: the patches maintain their identity between successive snapshots of the
325: time sequence that we obtained.
326: Approximately 45\,\% of the FOV contains 
327: polarimetric signal.
328: The mean unsigned magnetic field density averaged over the FOV is
329: 21~G, for the 
330: magnetogram based on \lineone , and 17 G for 
331: \linetwo\ (Table \ref{table}).
332: In this average we set to zero all those pixels with $B_\mathrm{eff}$
333: below the noise level.
334: The unsigned flux densities that we measure are larger than the values found 
335: hitherto (e.g., see the values compiled by \citeNP{san03}, \S 4.1). We detect more signals 
336: probably due to the unique combination of high
337: spatial resolution and good sensitivity of our observation.
338: The magnetic flux detected in the IN 
339: critically depends on the polarimetric
340: sensitivity and the angular resolution. For example,
341: if the magnetograms in Figure \ref{fig1} had
342: a sensitivity of 100~G, then the mean unsigned flux density
343: drops to 1-2 G (Table \ref{table};
344: the new lower sensitivity is modeled  setting to zero all the observed signals 
345: below 100 G).
346: The  drastic decrease of signals with decreasing sensitivity 
347: can easily explain why the IN signals that we detect
348: have been missed in previous
349: high resolution magnetograms 
350: (\citeNP{kel95}; \citeNP{sto00}; \citeNP{ber01}).
351: A decrease of angular resolution also reduces the signals 
352: due to cancellation between close 
353: opposite polarities.
354: If the angular resolution of the observation is artificially degraded 
355: by convolving the magnetograms in Figure \ref{fig1} with a 1\arcsec~FWHM Gaussian, 
356: then the mean unsigned flux densities  become of the order of 7-9 G.
357: These values agree with previous determinations 
358: based on scanning spectro-polarimeters that reach 1\arcsec\ resolution 
359: (\citeNP{san00} obtain 10\,G, whereas \citeNP{lit02} gives 
360: 9\,G).  The mean {\em signed}
361: magnetic flux density in Figure \ref{fig1} is found to be 
362: +3~G for \linetwo ,  
363: and +2~G for \lineone. 
364: 
365: 
366: 
367: 
368: 
369: %
370: 
371: 
372: Figure \ref{fig2} shows the speckle reconstructed broad-band image 
373: and, overlaid to it, the 
374: \linetwo\ magnetogram.  
375: It is important to note that most of the magnetic fields, 
376: $\approx65$\,\%, are located in intergranular spaces. This was observed before 
377: (\citeNP{lin99}; \citeNP{lit02}; \citeNP{soc03b}) and is expected from 
378: numerical simulations of magneto-convection (e.g., \citeNP{cat99a}).
379: Our high spatial resolution observations 
380: leave no doubt on these results. However, magnetic fields can also be found in 
381: granules (\citeNP{sto00}; \citeNP{kos01}).
382: Patches of opposite polarity are often located close to each other, again in 
383: agreement with expectations from numerical simulations. 
384: %
385: 
386:  
387: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
388: %
389: \subsection{Magnetic field strengths \label{mfs}}
390: 
391: The polarization signals obtained from the two spectral
392: lines  are correlated (see Fig. \ref{fig1}). However,  
393: the magnitude of the effective flux density of \lineone, 
394: $B_\mathrm{eff}(6301)$, is systematically larger than the
395: effective field derived using \linetwo, $B_\mathrm{eff}(6302)$.
396: We have estimated
397: this excess by several means:
398: least-squares fit of $B_\mathrm{eff}(6301)$ versus $B_\mathrm{eff}(6302)$,
399: least-squares fit of $B_\mathrm{eff}(6302)$ versus $B_\mathrm{eff}(6301)$, 
400: mean ratio among all points in the field of view, etc.
401: Our best estimate for this ratio is,
402: \begin{equation}
403: B_\mathrm{eff}(6301) /B_\mathrm{eff}(6302)\simeq 1.28\pm 0.10,
404: 	\label{ratio0}
405: \end{equation}
406: where the error bar accounts for all the individual estimates. 
407: We interpret this systematic difference 
408: as an indication that the magnetic field strengths in the
409: magnetograms are, typically, in the kG regime.
410: For weak fields, say 500\,G, 
411: the magnetograph equation (\ref{eeqmag}) holds, since
412: the basic conditions for the approximation to be valid are satisfied (\citeNP{soc02}).
413: Thus, for weak fields one expects 
414: $B_\mathrm{eff}(6301)/B_\mathrm{eff}(6302)\simeq1$. Since this is not the observed ratio,
415: one is forced to conclude 
416: that the field strengths have to be in the kG range.
417: We have modeled the ratio (\ref{ratio0})
418: using synthetic polarized spectra formed in atmospheres whose magnetic fields are known.
419: Such numerical calibration confirms the qualitative argument posed above, namely,
420: that the 
421: ratio (\ref{ratio0}) corresponds  to kG fields.
422: 
423: Note that structures with intrinsic kG magnetic field strength showing
424: 20 G flux density have to
425: occupy only a small fraction of the solar
426: surface. The simplest estimate yields 2\% of the area, which
427: comes from
428: fill factor $\sim B_\mathrm{eff}/B\sim 20~{\rm G} / 1000~$G.
429: 
430: 
431: Magnetic structures with kG fields should be bright at the granulation level 
432: because they have lower density and thus lower opacity than the ambient 
433: atmosphere (\citeNP{spr76}). Except for a few cases (e.g.,
434: Fig. \ref{fig2}, upper left corner at [1\farcs5,~22\arcsec]),
435: we do not see such brightening in the broadband 
436: images. This implies that the structures must thus be smaller 
437: than the resolution of these images, of approximately 
438: 0\farcs25 or 180\,km.
439: 
440: 
441: %                                     
442: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
443: %
444: 
445: \section{Conclusion}
446: 
447: 
448: We have demonstrated that polarimetric observations with high spatial resolution 
449: reveal a wealth of magnetic structures even in the very quiet Sun, away from 
450: activity or network features. Our study reveals a mean unsigned flux density
451: of 20 G, which is at least a factor two larger than the flux found in previous studies
452: having lower angular resolution. Even more, it is almost twice the flux density
453: detected during the solar maximum in the form of
454: active regions and network using conventional techniques (see \citeNP{soc02};
455: 	\citeNP{san02b}).
456: Our analysis also leads to the conclusion that the 
457: IN fields contain strong kG fields. 
458: Since only some 2\% of the 
459: solar surface produces these kG signals,
460: we do not exclude the
461: existence of weaker fields in the region
462: (of the order of tens of G, as found from Hanle diagnostics, or hundreds of G, as deduced
463: from Fe {\sc i} 15648~\AA\ measurements; see \S \ref{intro}).
464: 
465: We find no evidence that the angular resolution and sensitivity
466: of our magnetograms suffice to single out all the magnetic features existing in
467: the IN region. Consequently, the flux that we detect
468: should be regarded only as a lower
469: limit.
470: %
471: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
472: %
473: 
474: \acknowledgements
475: %
476: Thanks are due to T. Berger, H. Socas-Navarro and A. Title for discussions
477: on the work. 
478: %
479: Thanks are also due to K. Jan\ss en for letting us use her image reconstruction
480: routines.
481: %
482: IDC acknowledges support by the Deutsche Forschungsgemeinschaft (DFG) through
483: grant 418\,SPA-112/14/01. 
484: %
485: The Vacuum Tower Telescope is operated by the Kiepenheuer-Institut f\"ur
486: Sonnenphysik, Freiburg, at the Spanish Observatorio del Teide of the
487: Instituto de Astrof\'\i sica de Canarias.
488: %
489: The work has been partly funded by the Spanish Ministry of Science
490: and Technology, project AYA2001-1649.
491: 
492: %
493: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
494: %
495: %\bibliography{apjmnemonic,/home/jos/texto/papers/sun}
496: %\bibliographystyle{apj}
497: %\input{letter.bbl}
498: %
499: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
500: 
501: 
502: %\clearpage
503: \begin{thebibliography}{}
504: 
505: \bibitem[\protect\citeauthoryear{{Bendlin}}{{Bendlin}}{1993}]{ben93}
506: {Bendlin}, C. 1993, Ph.D. thesis, University of G\"ottingen, G\"ottingen
507: 
508: \bibitem[\protect\citeauthoryear{{Bendlin}, {Volkmer}, \& {Kneer}}{{Bendlin}
509:   et~al.}{1992}]{ben92}
510: {Bendlin}, C., {Volkmer}, R.,  \& {Kneer}, F. 1992, \aap, 257, 817
511: 
512: \bibitem[\protect\citeauthoryear{{Berger} \& {Title}}{{Berger} \&
513:   {Title}}{2001}]{ber01}
514: {Berger}, T.~E.,  \& {Title}, A.~M. 2001, \apj, 553, 449
515: 
516: \bibitem[\protect\citeauthoryear{{Bianda}, {Stenflo}, \& {Solanki}}{{Bianda}
517:   et~al.}{1998}]{bia98}
518: {Bianda}, M., {Stenflo}, J.~O.,  \& {Solanki}, S.~K. 1998, \aap, 337, 565
519: 
520: \bibitem[\protect\citeauthoryear{{Bianda}, {Stenflo}, \& {Solanki}}{{Bianda}
521:   et~al.}{1999}]{bia99}
522: {Bianda}, M., {Stenflo}, J.~O.,  \& {Solanki}, S.~K. 1999, \aap, 350, 1060
523: 
524: \bibitem[\protect\citeauthoryear{{Cattaneo}}{{Cattaneo}}{1999}]{cat99a}
525: {Cattaneo}, F. 1999, \apjl, 515, L39
526: 
527: \bibitem[\protect\citeauthoryear{{de Boer}}{{de Boer}}{1996}]{deb96}
528: {de Boer}, C.~R. 1996, \aaps, 120, 195
529: 
530: \bibitem[\protect\citeauthoryear{{de Boer} \& {Kneer}}{{de Boer} \&
531:   {Kneer}}{1992}]{deb92}
532: {de Boer}, C.~R.,  \& {Kneer}, F. 1992, \aap, 264, L24
533: 
534: \bibitem[\protect\citeauthoryear{{Emonet} \& {Cattaneo}}{{Emonet} \&
535:   {Cattaneo}}{2001}]{emo01}
536: {Emonet}, T.,  \& {Cattaneo}, F. 2001, \apjl, 560, L197
537: 
538: \bibitem[\protect\citeauthoryear{{Grossmann-Doerth}, {Keller}, \&
539:   {Sch\"ussler}}{{Grossmann-Doerth} et~al.}{1996}]{gro96}
540: {Grossmann-Doerth}, U., {Keller}, C.~U.,  \& {Sch\"ussler}, M. 1996, \aap, 315,
541:   610
542: 
543: \bibitem[\protect\citeauthoryear{{Hirzberger} et~al.}{{Hirzberger}
544:   et~al.}{2001}]{hir01}
545: {Hirzberger}, J., {Koschinsky}, M., {Kneer}, F.,  \& {Ritter}, C. 2001, \aap,
546:   367, 1011
547: 
548: \bibitem[\protect\citeauthoryear{{Jan\ss en}}{{Jan\ss en}}{2003}]{jan03}
549: {Jan\ss en}, K. 2003, Ph.D. thesis, University of G\"ottingen, G\"ottingen, in
550:   press
551: 
552: \bibitem[\protect\citeauthoryear{{Keller}}{{Keller}}{1995}]{kel95}
553: {Keller}, C.~U. 1995, Rev. Mod. Astron., 8, 27
554: 
555: \bibitem[\protect\citeauthoryear{{Keller} et~al.}{{Keller}
556:   et~al.}{1994}]{kel94}
557: {Keller}, C.~U., {Deubner}, F.-L., {Egger}, U., {Fleck}, B.,  \& {Povel}, H.~P.
558:   1994, \aap, 286, 626
559: 
560: \bibitem[\protect\citeauthoryear{{Keller} \& {von der L\"uhe}}{{Keller} \& {von
561:   der L\"uhe}}{1992}]{kel92b}
562: {Keller}, C.~U.,  \& {von der L\"uhe}, O. 1992, \aap, 261, 321
563: 
564: \bibitem[\protect\citeauthoryear{{Khomenko} et~al.}{{Khomenko}
565:   et~al.}{2002}]{kho02}
566: {Khomenko}, E.~V., {Collados}, M., {Solanki}, S.~K., {Lagg}, A.,  \&
567:   {Trujillo-Bueno}, J. 2002, \aap, in preparation
568: 
569: \bibitem[\protect\citeauthoryear{{Koschinsky}, {Kneer}, \&
570:   {Hirzberger}}{{Koschinsky} et~al.}{2001}]{kos01}
571: {Koschinsky}, M., {Kneer}, F.,  \& {Hirzberger}, J. 2001, \aap, 365, 588
572: 
573: \bibitem[\protect\citeauthoryear{{Krieg} et~al.}{{Krieg} et~al.}{1999}]{kri99}
574: {Krieg}, J., {Wunnenberg}, M., {Kneer}, F., {Koschinsky}, M.,  \& {Ritter}, C.
575:   1999, \aap, 343, 983
576: 
577: \bibitem[\protect\citeauthoryear{{Landi Degl'Innocenti}}{{Landi
578:   Degl'Innocenti}}{1992}]{lan92}
579: {Landi Degl'Innocenti}, E. 1992, in Solar Observations: Techniques and
580:   Interpretation, ed. F.~{S\'anchez}, M.~{Collados}, \& M.~{V\'azquez}
581:   (Cambridge: Cambridge University Press), 71
582: 
583: \bibitem[\protect\citeauthoryear{{Lin}}{{Lin}}{1995}]{lin95}
584: {Lin}, H. 1995, \apj, 446, 421
585: 
586: \bibitem[\protect\citeauthoryear{{Lin} \& {Rimmele}}{{Lin} \&
587:   {Rimmele}}{1999}]{lin99}
588: {Lin}, H.,  \& {Rimmele}, T. 1999, \apj, 514, 448
589: 
590: \bibitem[\protect\citeauthoryear{{Lites}}{{Lites}}{2002}]{lit02}
591: {Lites}, B.~W. 2002, \apj, 573, 431
592: 
593: \bibitem[\protect\citeauthoryear{{Livingston} \& {Harvey}}{{Livingston} \&
594:   {Harvey}}{1975}]{liv75}
595: {Livingston}, W.~C.,  \& {Harvey}, J.~W. 1975, \baas, 7, 346
596: 
597: \bibitem[\protect\citeauthoryear{{Martin}}{{Martin}}{1971}]{mar71}
598: {Martin}, B.~R. 1971, Statistics for Physicists (London: Academic Press)
599: 
600: \bibitem[\protect\citeauthoryear{{S\'anchez Almeida}}{{S\'anchez
601:   Almeida}}{2002}]{san02b}
602: {S\'anchez Almeida}, J. 2002, in Solar Wind 10, ed. M.~{Velli}, AIP Conf. Proc.
603:   (New York: American Institute of Physics), in press
604: 
605: \bibitem[\protect\citeauthoryear{{S\'anchez Almeida}, {Emonet}, \&
606:   {Cattaneo}}{{S\'anchez Almeida} et~al.}{2003}]{san03}
607: {S\'anchez Almeida}, J., {Emonet}, T.,  \& {Cattaneo}, F. 2003, \apj, 585, in
608:   press
609: 
610: \bibitem[\protect\citeauthoryear{{S\'anchez Almeida} et~al.}{{S\'anchez
611:   Almeida} et~al.}{1996}]{san96}
612: {S\'anchez Almeida}, J., {Landi Degl'Innocenti}, E., {Mart\'\i nez Pillet}, V.,
613:    \& {Lites}, B.~W. 1996, \apj, 466, 537
614: 
615: \bibitem[\protect\citeauthoryear{{S\'anchez Almeida} \& {Lites}}{{S\'anchez
616:   Almeida} \& {Lites}}{2000}]{san00}
617: {S\'anchez Almeida}, J.,  \& {Lites}, B.~W. 2000, \apj, 532, 1215
618: 
619: \bibitem[\protect\citeauthoryear{{Sigwarth} et~al.}{{Sigwarth}
620:   et~al.}{1999}]{sig99}
621: {Sigwarth}, M., {Balasubramaniam}, K.~S., {Kn\"olker}, M.,  \& {Schmidt}, W.
622:   1999, \aap, 349, 941
623: 
624: \bibitem[\protect\citeauthoryear{{Smithson}}{{Smithson}}{1975}]{smi75}
625: {Smithson}, R.~C. 1975, \baas, 7, 346
626: 
627: \bibitem[\protect\citeauthoryear{{Socas-Navarro}}{{Socas-Navarro}}{2003}]{soc0%
628: 3b}
629: {Socas-Navarro}, H. 2003, in Solar Polarization Workshop 3, ed.
630:   J.~{Trujillo-Bueno} \& J.~{S\'anchez Almeida}, ASP Conf. Ser. (San Francisco:
631:   ASP), in preparation, revise
632: 
633: \bibitem[\protect\citeauthoryear{{Socas-Navarro} \& {S\'anchez
634:   Almeida}}{{Socas-Navarro} \& {S\'anchez Almeida}}{2002}]{soc02}
635: {Socas-Navarro}, H.,  \& {S\'anchez Almeida}, J. 2002, \apj, 565, 1323
636: 
637: \bibitem[\protect\citeauthoryear{{Spruit}}{{Spruit}}{1976}]{spr76}
638: {Spruit}, H.~C. 1976, \solphys, 50, 269
639: 
640: \bibitem[\protect\citeauthoryear{{Stenflo}}{{Stenflo}}{1982}]{ste82}
641: {Stenflo}, J.~O. 1982, \solphys, 80, 209
642: 
643: \bibitem[\protect\citeauthoryear{{Stenflo}}{{Stenflo}}{1994}]{ste94}
644: {Stenflo}, J.~O. 1994, Solar Magnetic Fields, ASSL 189 (Dordrecht: Kluwer)
645: 
646: \bibitem[\protect\citeauthoryear{{Stolpe} \& {Kneer}}{{Stolpe} \&
647:   {Kneer}}{2000}]{sto00}
648: {Stolpe}, F.,  \& {Kneer}, F. 2000, \aap, 353, 1094
649: 
650: \bibitem[\protect\citeauthoryear{{Unno}}{{Unno}}{1956}]{unn56}
651: {Unno}, W. 1956, \pasj, 8, 108
652: 
653: \bibitem[\protect\citeauthoryear{{von der L\"uhe}}{{von der
654:   L\"uhe}}{1984}]{von84}
655: {von der L\"uhe}, O. 1984, \josa, A1, 510
656: 
657: \bibitem[\protect\citeauthoryear{{Weigelt}}{{Weigelt}}{1977}]{wei77}
658: {Weigelt}, G.~P. 1977, Optics Comm., 21, 55
659: 
660: \bibitem[\protect\citeauthoryear{{Yi}, {Jensen}, \& {Engvold}}{{Yi}
661:   et~al.}{1993}]{yi93}
662: {Yi}, Z., {Jensen}, E.,  \& {Engvold}, O. 1993, in ASP Conf. Ser., Vol.~46, The
663:   Magnetic and Velocity Fields of Solar Active Regions, ed. H.~{Zirin},
664:   G.~{Ai}, \& H.~{Wang}, San Francisco, 232
665: 
666: \end{thebibliography}
667: 
668: %
669: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
670: %
671: %\clearpage
672: 
673: \begin{figure}
674: \epsscale{0.6}
675: \plotone{f1.eps}
676:    \caption{Magnetograms in the two Fe\,{\sc i} lines at $\lambda$6302.5\,\AA\ (left)
677:    and 
678: $\lambda$6301.5\,\AA\ (right).   
679: Dark and light correspond to different polarities
680: (see the vertical bar). The distance between
681: tickmarks is 1\arcsec.}
682: 	\label{fig1}
683: \end{figure}
684: %
685: %\clearpage
686: 
687: \begin{figure}
688: \epsscale{.60}
689: \plotone{f2.eps}
690: \caption{Speckle reconstructed broadband image overlaid with the magnetogram  of
691: \linetwo\ with contours at $\vert B_\mathrm{eff}\vert=$~30, 50, 
692: 70, and 90\,G. The solid and dotted contours indicate opposite polarities. The 
693: distance between tickmarks is 1\arcsec.}
694: \label{fig2}
695: \end{figure}
696: %________________________________________________________________
697: %
698: %
699: %\clearpage
700: 
701: \begin{deluxetable}{lccccc}
702: \tablewidth{0pt}
703: \tablecaption{Mean unsigned flux density in the magnetograms\label{table}}
704: \tablehead{line&full data& $>100$~G\tablenotemark{a}& 1\arcsec\ seeing& 
705: 	signed flux\tablenotemark{b}& random noise\tablenotemark{c}} 
706: \startdata
707: \lineone&21 G&2 G&9 G&2 G&23 G\\
708: \linetwo&17 G&1 G&7 G&3 G&17 G\\
709: \enddata
710: \tablenotetext{a}{Only signals above 100 G are considered.}
711: \tablenotetext{b}{Contrarily to the other estimates in the table,
712: 	the signs of the signals are considered
713: 	in the average.}
714: \tablenotetext{c}{Corresponding to a single pixel.}
715: \end{deluxetable}
716: %________________________________________________________________
717: 
718: \end{document}
719: 
720: 
721: 
722: 
723: 
724: 
725: 
726: 
727: