1: \documentclass[12pt,preprint]{aastex}
2: \shorttitle{Magnetic field morphology of the white dwarf WD\,1953-011}
3: \shortauthors{Valyavin et al.}
4: \begin{document}
5:
6: \title{The pecular magnetic field morphology of the white dwarf WD\,1953-011:
7: evidence for a large-scale magnetic flux tube?}
8:
9: \author{G. Valyavin\altaffilmark{1}, G.A.~Wade\altaffilmark{2},
10: S.~Bagnulo\altaffilmark{3}, T.~Szeifert\altaffilmark{4},
11: J.D.~Landstreet\altaffilmark{5}, Inwoo Han\altaffilmark{1},
12: A.~Burenkov\altaffilmark{6}}
13:
14: \altaffiltext{1}{Korea Astronomy and Space Science Institute, 61-1,
15: Whaam-Dong, Youseong-Gu, Taejeon, Republic of Korea 305-348 }
16: \altaffiltext{2}{Physics Department, Royal
17: Military College of Canada, Kingston, Ontario, Canada}
18: \altaffiltext{3}{Armagh Observatory, Northern Irland}
19: \altaffiltext{4}{European Southern Observatory, Alonso de
20: C\'ordova 3107, Santiago, Chile}
21: \altaffiltext{5}{Physics \& Astronomy Department, University of Western
22: Ontario, London, Canada}
23: \altaffiltext{6}{Special Astrophysical Observatory, Russian Academy
24: of Sciences, Nizhnii Arkhyz, Karachai Cherkess Republic, 357147,
25: Russia}
26:
27:
28: \begin{abstract}
29: We present and interpret new spectropolarimetric
30: observations of the magnetic white dwarf WD\,1953-011.
31: Circular polarization and intensity spectra of the H$\alpha$ spectral
32: line demonstrate the presence of two-component magnetic field
33: in the photosphere of this star. The geometry consists of a weak, large
34: scale component, and a strong, localized component. Analyzing
35: the rotationally modulated low-field component, we establish a rotation
36: period $P_{rot} = 1.4480 \pm 0.0001$~days.
37: Modeling the measured
38: magnetic observables, we find that the low-field component
39: can be described by the superposition of a dipole and quadrupole.
40: According to the best-fit model, the inclination of the stellar rotation axis with
41: respect to the line of sight is $i \approx 20^\circ$, and the angle between
42: the rotation axis and the dipolar axis is $\beta \approx 10^\circ$.
43: The dipole strength at the pole is about 180\,kG, and the
44: quadrupolar strength is about 230\,kG. These data suggest a
45: fossil origin of the low-field component. In contrast,
46: the strong-field component exhibits a peculiar, localized structure
47: (``magnetic spot'') that confirms the conclusions of Maxted and co-workers.
48: The mean field modulus of the spot ($|B_{spot}| = 520 \pm 7$~kG)
49: together with its variable longitudinal magnetic field
50: having a maximum of about $+400$~kG make it difficult to
51: describe it naturally as a high-order component of the star's global poloidal
52: field. Instead, we suggest that the
53: observed strong-field region has a geometry similar to a magnetic flux
54: tube.
55: \end{abstract}
56: \keywords{stars: individual (\objectname{WD1953-011}
57: --- stars: magnetic fields --- stars: white dwarfs}
58:
59: \section{Introduction}
60: \label{Intro}
61:
62: At present, there are more than one hundred known isolated
63: magnetic white dwarfs (MWDs) with
64: magnetic field strengths from a few tens of kilogauss to several hundreds of
65: megagauss \citep{ABL81,SS95,LBH03,VBFM03,AZC04,VBF06,KAW07,JAN07}. It is generally
66: assumed that the
67: magnetic fields of the strong-magnetic MWDs (those with MG-strength fields)
68: are organized as
69: low-order multipolar fields with dominating dipolar components
70: \citep{P99}.
71: The rotation periods and surface magnetic fields of the strong-magnetic
72: MWDs are believed to be stable on long time scales
73: \citep{SN91}, suggesting that their
74: fields are fossil remnants of the fields of their progenitor stars.
75: A comparison of the field strengths and incidence statistics of the strong-magnetic
76: MWDs with magnetic fields of Ap/Bp main sequence stars support this
77: assumption \citep{ABL81}.
78:
79: Despite the progress with the strong-magnetic MWDs, the
80: magnetic properties of the weak-field degenerates are only poorly known.
81: Presently, only a few white dwarfs with kilogauss magnetic fields has been
82: identified \citep{SS95,FVB03,VBFM03,AZC04,VBF06,KAW07,JAN07}. Their rotation and field
83: geometries are poorly studied, although some progress has been achieved
84: by \citet{MFMW00} and \citet{Wad03}
85: with a study of WD\,1953-011 and by \citet{VBMF05} with WD\,0009+501.
86: \citet{MFMW00} established that the magnetic morphology of
87: WD\,1953-011 can be described by both low-field ($B \sim 90$\,kG) and
88: strong-field ($B \sim 500$\,kG) components. Some evidence for the presence
89: of a non-dipolar (quadrupolar) component
90: was also found in WD\,0009+501 by \citet{VBMF05}.
91:
92: Motivated by the results of \citet{MFMW00}, we have undertaken
93: collaborative spectropolarimetric monitoring of WD1953-011. In this paper
94: we report results of these observations and analyse them in the manner
95: presented by \citet{Wad03} and \citet{VBMF05}.
96: Our goal is to determine precisely the magnetic morphology of this degenerate.
97:
98: \section{A few preliminary remarks}
99: \label{Rem}
100:
101: Our preliminary analysis of the spectropolarimetric data obtained with FORS1
102: at the VLT \citep{Wad03} revealed significant variability of the
103: Stokes~$I$ and $V$ spectra of WD\,1953-011 due to rotation, with a period of about
104: 1.45 days. In Fig.~\ref{fig1} we show those results which illustrate the
105: variation
106: of the Stokes~$I,V,Q$ and $U$ profiles of the H$\alpha$ line with rotational
107: phase (phase increases from top to bottom in the figure).
108: As can be seen, the Stokes~$I$ profile is strongly variable. The central S-wave of
109: the Stokes~$V$ profile is almost constant. However, near rotational
110: phase 0.6 additional broadened Stokes~$V$ signatures appear in the
111: H$\alpha$ wings. These signatures correspond to the weak ``satellite
112: features'' observed by \citet{MFMW00} in the wings of
113: the H$\alpha$ profile at these phases, and were attributed to
114: the presence of a high-field magnetic structure. Linear polarization
115: Stokes~$Q$ and $U$ signatures are only marginally detected at several
116: rotational phases.
117:
118: In this paper we extend this analysis using addition observational
119: material obtained with the AAT \citep{MFMW00} and with the
120: 6-m Russian telescope BTA, and using more sophisticated modeling techniques.
121:
122: \section{Observations}
123: \label{Obs}
124:
125: Spectropolarimetric observations of WD~1953$-$011 were obtained in
126: service mode between May and June 2001 with FORS1 on the ESO
127: VLT. FORS1 is a multi-mode instrument for imaging and multi-object
128: spectroscopy equipped with polarimetric optics, and is described by
129: \citet{AFF98}. For this work, FORS1 was used to measure
130: Stokes~\textit{IQUV} profiles of WD~1953$-$011 at 12 different
131: rotation phases, using grism 600\,R (plus order separation filter
132: GG\,435), which covers the interval 5250\,\AA -- 7450\,\AA. With
133: a slit width of 0.7", the spectral resolving power was about 1650.
134: To perform circular polarization measurements, a $\lambda/4$ retarder
135: waveplate and a Wollaston prism are inserted in the FORS1 optical path
136: (see \citet{A67}). The $\lambda/4$ retarder waveplate can be
137: rotated in 45\degr\ steps.
138:
139: To perform linear polarization
140: measurements, a $\lambda/2$ retarder waveplate is used, which can be
141: rotated in $22.5^\circ$ steps. At each epoch, Stokes~$V$ was measured
142: taking two 420\,s exposures: one with the $\lambda/4$ retarder
143: waveplate at $-45^\circ$, and one with the $\lambda/4$ retarder
144: waveplate at $+45^\circ$. Stokes~$Q$ and $U$ were measured taking four
145: 600\,s exposures with the $\lambda/2$ retarder waveplate at $0^\circ$,
146: $22.5^\circ$, $45^\circ$, and $67.5^\circ$. The Stokes $V/I$
147: circular polarization spectrum was then obtained by calculating
148: \begin{equation}
149: P_V = \frac{V}{I} = \frac{1}{2} \left(r_{-45} - r_{+45}\right)
150: \end{equation}
151: where
152: \begin{equation}
153: \label{erre}
154: r_\alpha = \frac{f^{\rm o} -f^{\rm e}}{f^{\rm o} +f^{\rm e}}.
155: \end{equation}
156: In Eq.(2) $f^{\rm o}$ is the flux measured in the ordinary beam and
157: $f^{\rm e }$
158: is the flux measured in the extra-ordinary beam, obtained with the
159: $\lambda/4$ retarder waveplate at angle $\alpha$. Similarly, the linear
160: polarization was obtained by calculating
161: \begin{equation}
162: \begin{array}{rcl}
163: P_Q = \frac{Q}{I} &=& \frac{1}{2} \left(r_{0} - r_{45}\right) \\
164: P_U = \frac{U}{I} &=& \frac{1}{2} \left(r_{22.5} - r_{67.5}\right) \\
165: \end{array}
166: \end{equation}
167: where $r_\beta$ is defined by Eq.~\ref{erre}, except that $\beta$
168: refers to the position angle of the $\lambda/2$ retarder
169: waveplate. Fluxes $f^{\rm o}$ and $f^{\rm e}$ were obtained from the
170: raw data after bias correction and wavelength calibrations performed
171: using standard IRAF routines.
172:
173: These observations were supported by a short
174: observing run at the 6-m Russian telescope BTA where we obtained
175: additional $I, V$ series of spectra of WD\,1953-011 using the UAGS
176: spectropolarimeter, with nearly the same resolving power as
177: in the observations with FORS1. The instrument is described in detail
178: by \citet{ABVD95} and by \citet{NVF02}. The
179: observational technique and data reduction are similar to
180: those described by \citet{Bag00,Bag02} and by
181: \citet{VBMF05}. A comparative analysis of the spectropolarimetric data
182: obtained from the different telescopes showed identical results. A comparison
183: of the Stokes~$V$ spectra obtained with the VLT and BTA is illustrated in
184: Fig.~\ref{fig2}. The spectra are obtained at different times but
185: similar rotational phases.
186:
187: In addition to the spectropolarimetric data from the VLT and BTA, in
188: this paper we also use high-resolution spectroscopic data (Stokes~$I$)
189: obtained at the AAT and described by \citet{MFMW00}. Together
190: with the spectropolarimetry, these data extend the analysis presented by
191: \citet{Wad03} to a much longer time base.
192: Table~\ref{tbl1} gives an overview of all the observations. In
193: the table: {\bf JD} is the Julian Date; {\bf Exp} is an equivalent exposure time of an
194: observation; {\bf Stokes} is the observed Stokes parameter ($I,V,Q$ or $U$);
195: and {\bf Telescope} is telescope used (VLT, AAT or BTA).
196:
197: \section{Mean field modulus of the magnetic field of
198: WD\,1953-011}
199: \label{Meanf}
200:
201: We begin with an analysis of the low and high-resolution Stokes~$I$ spectra,
202: extending over 5 years. These spectra are used for
203: establishing the rotation period of the star,
204: as well as the
205: mean field modulus of the low- and strong-field components.
206:
207:
208: \subsection{The low-field component}
209:
210: The low-field component of WD\,1953-011 was first discovered
211: spectroscopically by \citet{KDWA98} and described
212: in detail by \citet{MFMW00}. The mean field modulus,
213: $|B_G|$\footnote{For distinctness we label all magnetic observables related
214: to the low-field component with the subscript ``$G$'', assuming its large scale (Global
215: ) geometry. Observables related to the strong-field component will be labeled
216: with the subscript ``$Spot$'' or ``$S$''.} exhibits a low-amplitude variation due to the
217: star's rotation, with a period estimated between hours and days
218: \citep{MFMW00}. These conclusions were made on the basis of the
219: high-resolution spectroscopy of the Zeeman pattern in the H$\alpha$ core.
220:
221: In our low-resolution FORS1 and BTA observations, Zeeman splitting attributed to
222: the low-field component cannot be resolved spectroscopically.
223: In the spectra the splitting is revealed as
224: an additional variable broadening and desaturation of the H$\alpha$ core.
225: In this case, measurements of the low-field component
226: can be carried out by an analysis of the equivalent widths ($EW_{core}$)
227: of the H$\alpha$ core. Using field modulus $|B_G|$ values determined
228: by \citet{MFMW00} from an analysis of individual high-resolution H$\alpha$
229: line profiles and measuring equivalent widths of the H$\alpha$ cores, we may try
230: to calibrate the relationship $EW_{core}$\,--\,$|B_G|$ to allow us to
231: determine $|B_G|$ in the low-resolution spectra. To obtain the required calibration, we
232: estimated equivalent widths of the H$\alpha$ core
233: from the high resolution spectra obtained by \citet{MFMW00}.
234: In order to work with measurements having a uniform resolution, all high-resolution spectra were convolved
235: with a gaussian instrumental profile to reproduce the spectral
236: resolution of FORS1 and UAGS. The resultant spectra are presented in
237: Fig.~\ref{fig3}.
238:
239: As one can see in Fig.~\ref{fig3}, the profiles are
240: strongly variable. The central intensity of the core also correlates with the
241: intensity of the strong-field Zeeman features which are found in the wings of
242: the H$\alpha$ profile (Fig.~\ref{fig3}: the two satellite features at
243: $\pm$10~\AA \, around the H$\alpha$ core). This correlation (the higher the
244: intensity of the features, the weaker the central intensity) is due to the
245: fact that the spot, which appears periodically on the visible disc due to rotation,
246: redistributes the flux according its projected area.
247: It is also seen that the width of the H$\alpha$ core
248: is variable itself due to the variable Zeeman pattern of the line core.
249: Therefore, to minimize the influence of the variable high-field spectral
250: features in measurements of the central Zeeman pattern attributed to the
251: low-field component, we artificially re-normalized all the profiles to equal
252: residual intensities ($r_c = 0.47$ at the line center) and measured equivalent widths of
253: the central narrow portion ($r_c \le 0.6$) of the resultant H$\alpha$
254: profiles. In these conditions, the variation of the $H\alpha$ core is
255: attributed only to the rotationally modulated low-field component. As hoped, we find a close correlation between the $EW_{core}$ measured in this way and the value of $|B_G|$ measured by Maxted. This relation is shown in Fig.\,4.
256:
257: Finally, using the $|B_G|-EW$ relationship derived from the
258: high-resolution spectra as illustrated in Fig.\,4, we inferred the field
259: modulus $|B_G|$ associated with each of the low-resolution spectra
260: (see Table~\ref{tbl2}).
261:
262: \subsection{The strong-field component}
263:
264: In order to measure the magnetic field modulus $|B_{Spot}|$ of the
265: strong-field component we
266: deblended the H$\alpha$ profile by means of a simultaneous fit of five
267: Gaussian profiles (three central profiles used to fit the H$\alpha$ core,
268: and two satellite gaussians to reproduce the strong-field Zeeman pattern).
269: This method enabled us to reproduce the Zeeman splitting of the strong-field
270: component and the corresponding magnetic field strength in those
271: spectra where the strong-field spectral features are seen.
272: The method also allows us to estimate the projected fractional area $S$ of the strong-field
273: area on the disc. Reconstructing by gaussians and extracting the Zeeman pattern
274: of the strong field component from the observed H$\alpha$ line profiles,
275: we determined $S$, the fraction of the flux absorbed by the strong-field pattern
276: relative to the total H$\alpha$ absorption. The method is rather rough and can be
277: considered as a first-guess approximation that is necessary for
278: the analysis described below. A more realistic calculation of the size of the
279: strong-field area is performed in Sect.\,8 where we model the spectra.
280: The results ($|B_{spot}|$ and $S$) are presented
281: in Table~\ref{tbl3}. $S$ is given in per cent of the disc area.
282:
283: \section{
284: Mean longitudinal magnetic field of WD1953-011}
285:
286:
287: From the Stokes~$I$ and $V$ spectra obtained with the VLT and BTA we
288: determined longitudinal fields through the weak-field approximation
289: \citep{AMS73} modified to the analysis of the two-component
290: Stokes~$V$ spectra:
291:
292: \begin{equation}
293: V(\lambda) \sim
294: (1 - S) B_G^l \, \bigg{(} \frac{\lambda}{\lambda_0} \bigg{)}^2
295: \frac{1}{I(\lambda)} \frac{{\rm d} I(\lambda) }{{\rm d} \lambda} \,\,
296: + S V(\lambda)_{Spot}
297: \,
298: \end{equation}
299: where $B_G^l$ is the longitudinal field of the low-field component,
300: $\frac{d I(\lambda) }{d \lambda}$ describes the gradient of the flux
301: profile, $S$ is the relative area of the spot projected on the disc,
302: $\lambda_0$ is the H$\alpha$ rest wavelength and $V(\lambda)_{Spot}$ is
303: the Stokes~$V$ profile from the strong-field component observed in the
304: H$\alpha$ wings.
305:
306: In Eq.(4) we effectively separate the disc
307: into two equivalent areas with different averaged magnetic field strengths.
308: The first term in the equation describes the weak-field area and the second
309: term is attributed to the strong-field component. The first
310: term is used in the usual manner according to which
311: the flux and its gradient are taken directly from the observed spectra.
312: (Inaccuracies due to the presence of the strong-field features in the wings
313: are comparatively weak: these features are
314: located quite far from the line core and do not affect the narrow
315: central low-field polarization profile). However, circular polarization
316: $V(\lambda)_{Spot}$ from the strong-field component (the second term in the equation)
317: cannot be fitted in the same way.
318:
319: In order to fit $V(\lambda)_{Spot}$ and estimate the
320: longitudinal field of the strong-field area, we compute:
321:
322: \begin{equation}
323: V(\lambda)_{Spot} = \frac{I(\lambda)^{L}_{Spot} - I(\lambda)^{R}_{Spot}}{I(\lambda)}
324: \end{equation}
325: \\
326:
327: \noindent
328: where the flux $I(\lambda)$ is the observed H$\alpha$ flux profile, and
329: $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$ are the left- and right-hand
330: polarized parts of the H$\alpha$ profile from the strong-field
331: equivalent area of the disc.
332: In the observed polarization spectra
333: $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$ are mixed with fluxes
334: from the weak-field equivalent area and therefore can not be extracted
335: directly. However, we may estimate them with some simplifications.
336:
337: Individually, $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$ are
338: Zeeman-split profiles of the
339: circularly polarized satellite $\sigma$ components.
340: Due to the fact that the $\sigma_-$ component
341: is absent in $I(\lambda)^{R}_{Spot}$, and the $\sigma_+$ component is absent in
342: $I(\lambda)^{L}_{Spot}$, their centers of gravity are displaced, indicating the
343: presence of the longitudinal field from the strong-field area. Their
344: difference provides the non-zero circular polarization (Eqn.\,5).
345:
346:
347: Because $I(\lambda)^{L}_{Spot}$ and
348: $I(\lambda)^{R}_{Spot}$ are not resolved in the total left- and
349: right-circularly polarized observed fluxes, the
350: determination of their true shapes requires
351: detailed modeling the field geometry. However,
352: as a first-guess approximation we may describe them by simulating an
353: equivalent ``mean'' Zeeman-broadened H$\alpha$
354: profile magnetically displaced to the left- and right-sides from the rest
355: wavelength
356: \footnote{The low resolving power of the FORS1
357: and UAGS, as well as unresolved circular polarization features
358: attributed to the strong-field area, enable us to consider the problem in
359: terms of Zeeman broadening instead of detailed analysis of the strong-field
360: Zeeman pattern.}. With this simplification only two
361: parameters~-- Zeeman broadening
362: and their magnetic displacement due to the
363: averaged longitudinal field from the spot should be varied to reproduce the
364: observed circular polarization.
365:
366: In the simulation procedure we may use a template H$\alpha$
367: profile, artificially broadened to unresolved Zeeman patterns
368: typical for $I(\lambda)^{R}_{Spot}$ and $I(\lambda)^{L}_{Spot}$ and
369: magnetically displaced. This template can be
370: taken from a zero magnetic field solution for the atmosphere of
371: WD\,1953-011 or from the observed spectra.
372: For example, assuming the pressure-temperature
373: conditions in the spot area to be similar to conditions in the other parts
374: of the white dwarf's surface we may choose as the template one of
375: the observed weak-field H$\alpha$ profiles (obtained at those moments
376: when the spot is not seen). In our analysis we proceed this way.
377:
378: Thus, to simulate $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$
379: in order to fit the circular polarization (5) from the strong-field area and
380: estimate its longitudinal magnetic field we used the following iterative
381: method: \\
382:
383: \begin{itemize}
384:
385: \item
386: {\bf Step-1:} We construct the reference weak-field ``template'' H$\alpha$ profile
387: from the observed I-profiles obtained at those rotational phases where
388: the strong-field Zeeman pattern is not seen.\\
389:
390: \item
391: {\bf Step-2:} We artificially broaden the template profile by a gaussian
392: filter with an arbitrary half-width
393: to an unresolved strong-field Zeeman pattern and displace the result
394: by the magnetic displacement factor $\Delta \lambda$ to the shorter / longer wavelengths
395: to estimate the
396: $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$ profiles.
397:
398: \item
399: {\bf Step-3:} Varying the magnetic broadening of the estimated profiles
400: $I(\lambda)^{L}_{Spot}$ and $I(\lambda)^{R}_{Spot}$ and their Zeeman
401: displacement we finally fit the strong-field circular polarization (5) in
402: the working equation (4). The displacement found
403: $\Delta \lambda = 4.67 \cdot 10^{-13} B_S^l \lambda_0^2$ \citep{L80},
404: gives an estimate of the longitudinal field $B_S^l$.
405: (In other words, taking $S$ measured from the Stokes~$I$ spectra and presented
406: in Table~\ref{tbl3} we simultaneously fit the observed
407: combined Stokes~$V$ (4) varying $B_G^l$ and circular polarization (5) of the
408: strong-field component, where $B_S^l$ is one of the parameters.)
409: \end{itemize}
410:
411:
412: In the fit procedure, the associated error bars are obtained
413: using the Monte Carlo modeling method presented by
414: \citet{SS94}. An example of the fit is presented in Fig.~\ref{fig5}.
415: The results are collected in Table~\ref{tbl4}.
416:
417: This method gives quite robust estimates of the longitudinal magnetic
418: field of the low-field component. In the case of the strong-field component,
419: the real intensities of the fields could be slightly over or underestimated due
420: to the simplifications described above.
421: For these reasons, estimates of the strong-field
422: component given here could be considered to be approximate.
423: As an alternative, the two-component circular polarization spectra
424: could be analyzed by using Zeeman tomography
425: \citep[for instance]{EJB02}. To provide
426: more precise modeling,
427: below (Sec.\,8) we analyze our data again in the framework
428: of simplified Zeeman tomography.
429:
430: \section{Period determination}
431: \label{Temporal}
432:
433: To search for the star's rotation period we used the
434: equivalent widths $EW_{core}$ of the H$\alpha$
435: core determined in Sect.\,4 . This observable is the most sensitive indicator
436: for the determination of the rotation period.
437: To determine the rotation period we applied the Lafler-Kinman
438: method \citep{LK65}, as modified by
439: \citet{G04}. Analysis of the power spectrum of the data revealed a
440: signal indicating a probable period between 1.4 and 1.5~days
441: (Fig.~\ref{fig6}). This is consistent with the period estimate
442: ($P \approx 1.45$\,days) given by \citet{Wad03} and
443: \citet{BM05}.
444:
445: Detailed study
446: of the periodogram showed that the most significant sinusoidal signal
447: corresponds to a period P~=~1.4480~$\pm$~0.0001~days. Other peaks are
448: located around 1.447 days and 1.442 days.
449: An examination of these periods reveals distorted,
450: non-sinusoidal
451: signals and we do not consider these periods further.
452:
453:
454: The phase variation of the H$\alpha$ core equivalent widths $EW_{core}$ derived
455: with this period is presented in Fig.~\ref{fig7}.
456: The derived period shows a very good agreement among all the observations
457: taken from different telescopes (the VLT, BTA and AAT). For the minimum of
458: $EW_{core}$ we obtain the following ephemeris:
459: \[
460: \mathrm{JD} = 2452048.801 \pm 0.03 + 1^d _\cdot 4480 \pm 0.0001\,\mathrm{E}
461: \]
462:
463: The corresponding phase curves of the mean field modulus $|B_G|$
464: and longitudinal field $|B_G^l|$ of the weak-field components
465: are presented in Fig.~\ref{fig8}. The phase curves
466: are almost
467: sinusoidal, and symmetric about the values of about $|B_G|$~=~$+87$\,kG
468: and $|B_G^l|$~=~$-43$\,kG. The
469: modulus of the weak-field component varies from $+77 \pm 1.5$\,kG to $+97 \pm
470: 1.5$\,kG; the longitudinal magnetic field shows variation in the
471: range $-39 \pm 2$\,kG to $-47 \pm 2$\,kG.
472:
473: As one can see in Fig.~\ref{fig8} the behavior of the longitudinal
474: magnetic field and modulus of the weak-field component suggest the field
475: geometry to
476: be a simple, low-order poloidal field, which supports our view
477: that the period is correct. Below we use this period in analyzing the
478: magnetic field morphology of WD\,1953-011.
479: At the same time it is important to note that another estimate of
480: the rotational period
481: (P~=~1.4418~days) obtained by \citet{BM05} from
482: differential
483: photometry of this MWD is similar to, but formally different from, our
484: result. Following the next section, where we
485: establish the magnetic field morphology of the low-field component,
486: we will discuss this difference in more detail.
487:
488: \section{Modeling the weak-field component of the magnetic field of
489: WD\,1953-011}
490: \label{Model}
491:
492: To verify that the behavior of the weak-field component of
493: WD\,1953-011 is consistent
494: with a nearly dipolar geometry, we have followed the schematic
495: method proposed by \citet{Lan97}. This method has already been
496: described and applied to establish the
497: magnetic field morphology of the weak field
498: white dwarf WD0009+501 (see \citet{VBMF05} for details).
499: For this reason here we do not explain all the modeling details, but restrict
500: ourselves to the presentation of the results.
501:
502: In this paper we model the phase-resolved
503: measurements of the mean longitudinal field and mean field modulus of the
504: weak-field component within the framework of a pure dipole and
505: dipole+quadrupole field.
506: The phase-resolved observables for the weak-field component
507: which we use as input data are obtained by binning measurements in phase
508: and averaging. The binned data are presented in Table~\ref{tbl5}.
509:
510: The dipole or dipole plus quadrupole models depend on
511: the following 10 parameters:
512: \begin{itemize}
513: \item[--] $B_\mathrm{d}$ and $B_\mathrm{q}$, the dipole and quadrupole
514: strength, respectively;
515: \item[--] $v_\mathrm{e}$, the stellar equatorial velocity;
516: \item[--] $i$, the inclination of the stellar rotation axis to the line
517: of sight;
518: \item[--] $\beta$, the angle between the dipolar axis and
519: the rotation axis;
520: \item[--] $\beta_1$ and $\beta_2$, the analogues of $\beta$ for the
521: directions identified by the quadrupole;
522: \item[--] $\gamma_1$ and $\gamma_2$, the azimuthal angles of the unit
523: vectors of the quadrupole;
524: \item[--] $f_0$, the ``reference'' rotational phase of the model;
525: \item[--] $v_\mathrm{e} sin i$, the projected stellar rotation velocity.
526: \end{itemize}
527: The angles $i$, $\beta$, $\beta_1$, $\beta_2$ range from $0\degr$ to
528: $180\degr$, while $\gamma_1$, $\gamma_2$, $f_0$ range from $0\degr$ to
529: $360\degr$. The rotational period $P = 1.448$ and the limb-darkening
530: constant
531: $u$, which also affects the expressions of the magnetic observables,
532: are taken as fixed. (Note that the pure dipole model would retain as
533: free parameters only $B_\mathrm{d}$, $v_\mathrm{e}$, $i$, $\beta$,
534: and $f_0$.)
535:
536: For the stellar mass, \citet{BRG95}
537: gave the value of $0.844\,M_\odot$, which together with known
538: surface gravity of WD\,1953-011 ($log\, g = 8.412$, \citet{BRG95})
539: correspond to a stellar radius of about $0.0095\,R_\odot$.
540: This parameter and the period were then used
541: to estimate the equatorial and projected velocities of the star.
542:
543: For the limb-darkening coefficient, we adopted the value of $u=0.5$.
544: Note that, as discussed by \citet{Bag00}, the results
545: of the modeling are only slightly influenced by the $u$ value.
546: The best-fit parameters are:\\
547:
548: {\it A) Dipole }\\
549: %%%%%%%%%%%%%%%%%%%%%%%%%%%%ARRAY1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
550: \begin{displaymath}
551: \begin{array}{lcrcl}
552: %\hline
553: i &=& 14\degr &\pm& 10\degr \\
554: \beta &=& 14\degr &\pm& 10\degr \\
555: f_0 &\approx& 352\degr & & \\
556: B_{\rm d} &=& 108 &\pm& 5\,kG \\
557: v_{\rm e} &=& 0.33 &\pm& 0.05\,km\,s^{-1} \\
558: v_{\rm e} sin i &=& 0.08&\pm& 0.03\,km\,s^{-1} \\
559: %\hline
560: \end{array}
561: \end{displaymath}
562: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%\\
563: \vspace*{0.5cm}
564:
565: {\it B) Dipole + quadrupole}\\
566:
567: %%%%%%%%%%%%%%%%%%%%%%%%%%%%ARRAY2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
568: \begin{displaymath}
569: \begin{array}{lcrcl}
570: %\hline
571: i &=& 18\degr &\pm& 10\degr \\
572: \beta &=& 8\degr &\pm& 10\degr \\
573: f_0 &\approx& 357 && \\
574: \beta_1 &=& 22\degr &\pm& 10\degr \\
575: \beta_2 &=& 24\degr &\pm& 10\degr \\
576: \gamma_1 &\approx& 77\degr && \\
577: \gamma_2 &\approx& 243\degr && \\
578: B_{\rm d} &=& 178 &\pm& 30\,kG \\
579: B_{\rm q} &=& 233 &\pm& 30\,kG \\
580: v_{\rm e} &=& 0.33 &\pm& 0.05\,km\,s^{-1} \\
581: v_{\rm e} sin i &=& 0.1 &\pm& 0.05\,km\,s^{-1} \\
582: %\hline
583: \end{array}
584: \end{displaymath}
585: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
586:
587: Note that, as explained by \citet{Bag00}, the available
588: observations do not allow one to distinguish between two magnetic
589: configurations symmetrical about the plane containing the rotation
590: axis and the dipole axis. Such configurations are characterized by the
591: same values of $B_\mathrm{d}$, $B_\mathrm{q}$, $v_\mathrm{e}$,
592: $\gamma_1$, $\gamma_2$, $f_0$, while the remaining angles are related by
593: %%%%%%%%%%%%%%%%%%%%%%%%%%%%ARRAY%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
594: \[
595: \begin{array}{lccccr}
596:
597: (&i, &\beta, &\beta_1, &\beta_2 &)
598: \phantom{\;.} \\[0.1cm]
599: (&180\degr -i,&180\degr -\beta,&180\degr -\beta_1,&180\degr -\beta_2&)\;.
600: \end{array}
601: \]
602: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
603:
604: Due to the fact that the spin axis angle is close
605: to a pole-on orientation and due to the small number of available
606: observables, the error bars on the derived quantities are fairly large.
607: For the same reasons, there are some uncertainties in the
608: determinations of all the parameters considered together. However,
609: despite these weakness, we are able to obtain some conclusions about the
610: most probable geometry of the white dwarf's global field.
611:
612: The best fit of the dipole+quadrupole model applied to the observations is
613: shown by solid lines in Fig.~\ref{fig9}. For comparison, the dashed line
614: shows the fit obtained using the pure dipolar morphology. As is evident,
615: the dipolar model does not reproduce the observations well.
616: Examination of the reduced $\chi_r^2$ statistics shows that the quality of
617: the dipole+quadrupole fit ($\chi_r^2=0.22, 0.43$ for longitudinal field
618: and field modulus, respectively)
619: is significantly better than the pure dipole fit ($\chi_r^2=1.9, 2.2$).
620: We therefore conclude that the large-scale weak-field component of
621: WD\,1953-011 is better modeled by the superposition of a
622: dipole and quadrupole components.
623:
624: \section{High-field component}
625: \label{Active}
626:
627: \subsection{Migrating magnetic flux tube?}
628:
629: In contrast to the well-organized, nearly sinusoidal variation of the
630: weak-field component, the phase behavior of the high-field structure
631: exhibits a number of peculiar features that made it impossible to model
632: these data as a simple low-order multipole:
633:
634:
635: \begin{itemize}
636:
637: \item
638: According to the
639: measurements of the Zeeman-split satellite spectral features in the
640: H$\alpha$ wings, the mean field modulus of the strong-field component
641: does not show any noticeable variation during the star's rotation.
642: (Due to rotation, we see variation of the flux intensities from the
643: strong-field area, but the corresponding Zeeman displacement is nearly
644: constant.) The most likely explanation \citep{MFMW00}
645: is that there is an area with a nearly uniformly distributed
646: strong magnetic field. This explanation, if true, suggests that
647: the strong-field component has a localized geometry and cannot
648: be understood as a high-field term in the multipolar expansion
649: of the star's general field. Averaging all the data we determine
650: $<|B_{spot}|> = 515 \pm 7$~kG.
651:
652: \item
653: The Zeeman pattern attributed to the strong-field component becomes visible
654: at rotational phases $\phi = 0.25 - 0.7$ and demonstrates variation in the
655: flux intensities that suggests rotational variability of the projected effective
656: size $S$ of the magnetic spot. The projected area of the strong-field
657: structure varies from zero to about 12\% of the
658: disk, consistent with the study of \citet{MFMW00}.
659: This observable can be used as an additional
660: parameter to test the rotational period of the star.
661: However, using this quantity to search for the period we did
662: not find a regular signal at any period within the
663: tested 5-year time base. Moreover, phasing the data with
664: the magnetic ephemeris characterized by the rotational period of
665: 1.448 days, the resultant phase curve of the spot size variation (Fig.\,10)
666: appears distorted in comparison to the well-organized
667: behavior of the weak-field component phased with the same ephemeris.
668: We observe a small relative phase shift between the data obtained
669: with different telescopes that
670: may indicate a possible secular longitudinal drift
671: of the strong-field component.
672:
673:
674: \item
675: The averaged longitudinal field
676: of the strong field area varies from zero (when the area is invisible)
677: to about 450~kG (Table~\ref{tbl4}) which is comparable to the averaged
678: mean field modulus ($\approx 515$\,kG) of the field. This suggests a
679: deviation of the strong field component from any of low-order multipolar
680: geometries for which the difference between the full vector and
681: its longitudinal projection should much larger
682: (for example, for a centered dipole field the difference should be at least
683: 2.5 times, \citet{S50}).
684: \end{itemize}
685:
686: The last point suggests the presence of an essentially
687: vertical orientation of the magnetic field lines relative to
688: the star's surface, typical for local magnetic flux tubes in cool,
689: convective stars (the Sun for example). If the
690: geometry is a tube seen as a local magnetic
691: spot in the photosphere, we may also expect the above-mentioned secular
692: drift. To our knowledge and by an analogy to the
693: Sun, such fields are expected to show dynamical activity like migration
694: over the star's surface and be associated with dark spots that might
695: produce photometric variability of the star. Significant photometric
696: variability of WD\,1953-011 has been established
697: \citep{Wad03,BM05}.
698: However, in this paper we are unable to establish the association of the
699: darkness and magnetic spots for reasons which we discuss below.
700:
701: \subsection{Geometry and location of the magnetic spot}
702:
703: Measuring the longitudinal magnetic field of the strong-field component
704: as described in Sect.\,5 we noted that the estimates of the
705: longitudinal field of the strong-field area might be
706: affected by our simplifying assumptions. In order to control
707: these measurements we have also directly modeled the observed polarization and
708: intensity H$\alpha$ spectra of the star without parametrizing the surface field
709: components. In addition to the analysis of the Stokes $I$ and $V$
710: spectra, the observed $Q$ and $U$ spectra were also taken into consideration.
711: As can be seen in Fig.\,1, linear polarization Stokes $Q,U$ signatures are
712: detected only marginally at a few rotational phases. However,
713: this information can also be used to constrain the magnetic geometry of the
714: degenerate. Note that different magnetic geometries may produce similar
715: Stokes $V$ spectral features due to axial symmetry of circular polarization
716: provided by the longitudinal projection of the field.
717: Linear polarization restricts the strength and orientation of the
718: transverse field that, together with circular polarization, makes it
719: possible to resolve the geometry spatially. The observed $Q$ and $U$
720: spectra are mainly noise, but we may try to use them in terms of the upper limits.
721:
722: In our model we examined several
723: low order multipolar magnetic field geometries, integrating over the surface elementary
724: (taken at a single surface element) Stokes $I,V,Q$, and $U$ spectra calculated
725: for various field strengths and
726: orientations of the magnetic field lines. The technique we used
727: to calculate the elementary spectra deserves some special explanation.
728: \\
729:
730: {\it a) Simulation of the Stokes $I,V,Q$ and $U$ synthetic H$\alpha$ spectra.}
731: \\
732:
733:
734: Generally, accurate simulation of the split Balmer profiles in spectra of
735: strong magnetic
736: white dwarfs requires detailed computations of the main opacity sources
737: under the influence of strong magnetic fields. To our knowledge, these
738: computations have not yet been tabulated for practical use. For this reason,
739: a self-consistent solution of the transfer equation for the line profiles in
740: spectra of strong-magnetic white dwarfs cannot be performed without
741: special consideration of
742: additional parameters (related, for example, to the Stark
743: broadening in the presence of a strong magnetic field, \citet{J92}).
744: However, in case of the weak-field degenerates we may restrict
745: ourself to a zero-field solution similar to that presented by
746: \citet{WM79} or by \citet{SBJ92}. The method assumes that if
747: the Stark broadening dominates the line opacity, the total opacity
748: can be calculated as the sum of individual Stark-broadened Zeeman components.
749: The Stark broadening is suggested to be taken as ``non-magnetic'' in this
750: case. Under these simplifying assumptions we may simulate the local,
751: elementary Zeeman spectra using one of the following two ways:
752:
753:
754: \begin{itemize}
755: \item[i] to compute the transfer equation for all Stokes parameters
756: at given strength and orientation of a local magnetic field line
757: calculating the H$\alpha$ opacities as described, or (alternatively)
758: \item[ii]
759: to select a ``template'' H$\alpha$ profile typical for
760: a zero-field white dwarf with the same pressure-temperature
761: conditions as in WD\,1953-011 for construction of the
762: elementary Zeeman spectra. (In other words, we may try to construct from this template
763: profile individual Zeeman $\pi-$ and $\sigma-$ components, parametrizing
764: their magnetic displacement and relative intensities, and additively combine
765: them to obtain the elementary $I,V,Q,U$ H$\alpha$ profiles.)
766: \end{itemize}
767:
768: The first, direct method of atmospheric calculations for WD\,1953-011
769: requires special theoretical tools which are outside the scope of
770: this observational paper.
771: The second, simplified method, which we will use, seems to be rather rough
772: due to the fact that the
773: fluxes from the individual $\pi-$ and $\sigma-$ components obtained
774: by using the zero-field template profile are generally not additive
775: (whereas their corresponding opacities can be added in the transfer equation).
776: Nevertheless, in the linear guess approximation they can be taken
777: as additive and the method can also be applied.
778: Besides, testing this method on some standard, well-studied
779: magnetic Ap/Bp stars we have obtained satisfactory results modeling their
780: observed polarizations. This allowed us to conclude that
781: the method is reasonably accurate.
782:
783: Thus, to calculate the elementary $I,V,Q,U$ H$\alpha$ profiles
784: we adopted the use of the zero-field H$\alpha$ template profile
785: which was constructed from the observed I-spectra obtained at those moments,
786: when the strong-field Zeeman pattern is not seen. The Stark parts of the profile
787: were obtained by averaging the $I$ profiles at the rotation phases 0 and
788: 0.91 (see Fig.\,1) in which the strong-field features are not seen. The
789: central ``zero-field'' Doppler profile
790: was adopted to reproduce in the model procedure the observed
791: low-field magnetic broadening of the H$\alpha$ cores at phases 0 and 0.91.
792: The necessary individual profiles of the non-displaced $\pi-$ and displaced
793: $\sigma-$ components were obtained by entering the normal Zeeman displacements
794: according to the orientation of the local magnetic field: the circularly polarized
795: $\sigma-$ components are displaced according to the longitudinal
796: projection of the local magnetic field, and the linearly polarized
797: $\sigma-$ components are displaced by the transverse field.
798:
799: In order to model the polarization H$\alpha$ profiles, the
800: relative intensities of the central $\pi-$ and displaced $\sigma \pm$
801: components were computed from the prjection of the local field vector onto
802: the plane on the sky and on the line of sight
803: as described by \citet{U56} (for a qualitative explanation see also
804: \citet{L80}). The final intensities of the
805: $\pi-$ and $\sigma-$ components were
806: obtained by renormalisation such that
807: the total sum of the fluxes from
808: all the components be equal to the flux from the zero-field template
809: H$\alpha$ profile.
810:
811: Finally, the elementary $\pi-$ and $\sigma-$ components were combined
812: to construct from them
813: the elementary $I,V,Q,U$ spectra by simulation of the ordinary
814: and extraordinary beams given by a polarimetric analyzer.
815: For example, simulating the H$\alpha$ Stokes $V$ profile,
816: all the components except the circularly polarized $\sigma-$ components
817: are equally distributed between the beams. The circularly polarized
818: $\sigma_+$ components are absent in one of the beams, and the oppositely
819: polarized $\sigma_-$ components are absent in the other beam. The final
820: Stokes $V$ H$\alpha$ profile was obtained by subtraction of the ordinary
821: from extraordinary beams and devision of the result by the total flux.
822: The $Q$ and $U$ spectra were obtained in a similar way: the linearly polarized
823: $\sigma$ components and central $\pi$ component are distributed
824: between the beams according to the projection of the local magnetic field
825: onto the plane of the sky.
826:
827: After the determination of the elementary $I,V,Q,U$ spectra given by
828: a magnetic geometry in all surface elements, we finally integrated and
829: averaged them over the disc. For the limb-darkening coefficient
830: we adopted the value of $u=0.5$.\\
831:
832: {\it b) Modeling the field geometry in WD\,1953-011 by simulation of
833: the observed polarization spectra.}
834:
835: In Sect.\,5) we have concluded, that estimates of the longitudinal magnetic
836: field based on the weak-field approximation are accurate enough to make it
837: possible to model the geometry of the low-field component separately from the
838: strong-field component in the manner as demostrated in Sect.\,7. For this
839: reason, and in order to reduce a number of variables, we use those results
840: (obtained in Sect.\,7) as input and non-changeable parameters in the tomography.
841: We just note, that modeling
842: the observed spectra obtained at those time moments, where the spot is not
843: seen, we have confirmed, that
844: the observations (Stokes-V spectra) can be better fit by the
845: dipole+quadrupole geometry of the low-field component with parameters performed in Sect.\,7 (
846: case {\it B}).
847: Examination of the pure dipole model (case {\it A}) gives no satisfactory results
848: and we do not use this case here.
849:
850: Modeling the strong-field area as an additional
851: harmonic in the low-order (lower than octupole) multipolar expansion
852: we were unable to reproduce the observations.
853: The fit does not provide the necessary contrast in the observed
854: Zeeman patterns at those phases where the polarization and intensity spectra
855: demonstrate the weak- and strong-field Zeeman features together.
856: However, assuming the strong-field component to be concentrated
857: into a localized area having maximum projected size of about 12 per
858: cent of the disk, the strong-field Zeeman spectral features can be
859: well-reproduced with an average magnetic field of $550\pm 50$\,kG.
860: Practically the same result has been obtained by
861: \citet{MFMW00}.
862:
863:
864: To model the strong-field area we tested two simplified localized geometries:
865: a ``contrast spot'' with a homogeneously distributed, essentially
866: vertical magnetic field, and a ``sagittal'' geometry with a
867: strong vertically-oriented central magnetic field, that smoothly decreased
868: to zero at the spot edges. Generally, both geometries are able to describe
869: the Stokes $I$ and $V$ spectra with more or less acceptable accuracy.
870: The first model, however, does not provide us with a good fit of the
871: $Q$ and $U$ spectra due to the presence of the sharp (and non-physical) jump
872: of the field intensity at the edges of the strong-field area. For this
873: reason we do not discuss this case in detail.
874:
875: The ``sagittal'' geometry of the strong-field area was constructed
876: by using a modified model of a centered dipole: about 45\% of the spot's area (central parts)
877: have the dipolar distribution with polar field $Bp = +810$\,kG at the center.
878: The remaining 55\% of the external dipolar field is artificially modulated
879: to have a gradual decrease to zero at the edges of the area. This model provides
880: a good fit of the Stokes $I,V$ spectra and reasonable reproduction of
881: the linear polarization Q-,U-spectra, as shown in Fig.\,11 where we also
882: illustrate the tomographic portrait of the white dwarf's magnetosphere.
883:
884: Despite the fact that we obtain such a good agreement of the ``sagittal'' geometry
885: with the observables from the strong-field area, we do not claim that
886: this geometry is fully correct in all details (for example, the model does not
887: control conservation of magnetic flux). Similar to the case of the
888: weak-field component, the most natural way to study
889: the strong-field area is to describe it as a strong-field
890: feature resulting from the superposition of several high-order
891: harmonics in the multipolar expansion. At this time we are unable
892: to study this term using any combination of the first hamonics higher than
893: octupole, but we do not exclude that the use of the highest terms
894: of different polarities and intensities will resolve the problem.
895:
896: However, this result clearly demonstrates a qualitative difference in the
897: morphologies of the strong-field area and the global field of the white dwarf.
898: From the model we establish with a very high probability that the
899: strong-field area has a localized structure with essentially vertical
900: orientation of the magnetic field lines. The physical size of the
901: area is about 20\% of the star's surface giving maximum 12\% projection
902: on the disc. The spot is located at an angle of about $67^\circ$ with respect
903: to the spin axis, providing a maximum longitudinal field strength
904: of about 400\,kG. These results are in good agreement with the
905: measurements of the longitudinal field of the spot.
906:
907: \section{Discussion}
908: \label{Discuss}
909:
910: We have presented new low-resolution spectropolarimetric observations
911: of the magnetic white dwarf WD\,1953-011. From these
912: observations and observations of previous authors we have determined
913: the star's rotation
914: period, mean longitudinal field, mean
915: field modulus, and surface field morphology.
916: Let us finally summarize these results.
917:
918: \begin{itemize}
919: \item[1)] Our present picture of WD\,1953-011 consists of a MWD with
920: relatively smooth,
921: low-field global magnetic field component, and a high-field
922: magnetic area.
923: \item[2)]
924: The low-field component demonstrates regular periodicity with
925: period $P~=~1.4480 \pm 0.0001$~days.
926: We interpret this as the rotational period
927: of the white dwarf. The long-term stability
928: of the surface and longitudinal magnetic fields of the low-field component
929: enable us to interpret this component as a fossil poloidal
930: magnetic field consisting of dipolar and quadrupolar harmonics with
931: the following basic parameters:\\
932:
933: 1) the inclination of stellar rotation axis
934: $i = 18\degr \pm 10\degr$; \\
935: 2) the angle between the dipolar axis and the rotation axis
936: $\beta = 8\degr \pm 10\degr$; \\
937: 3) the dipole strength $B_{\rm d} = 178 \pm 30$\,kG; \\
938: 4) the quadrupole strength $B_{\rm q} = 233 \pm 30$\,kG. \\
939:
940: \item[3)]
941: The strong-field component exhibits a peculiar localized structure.
942: The mean field modulus of the spot
943: $|B_{spot}|$ is estimated to be $515 \pm 7$~kG, which
944: is consistent with the results presented by \citet{MFMW00}.
945: The longitudinal magnetic field of the spot varies with
946: rotational phase from $< 300$~kG to about $400$~kG.
947: Comparing the mean field modulus with the maximum longitudinal field
948: we suggest that the geometry of the high-field spot may be
949: similar to a magnetic flux tube with vertically-oriented magnetic field
950: lines. The spot is located at an angle of
951: $\approx 67\degr$ with respect to the spin axis.
952: \end{itemize}
953:
954: Our results suggest that the magnetic field
955: of WD\,1953-011 consists of two physically different morphologies -
956: the fossil poloidal field and an apparently induced magnetic spot. To our knowledge,
957: fossil, slowly decaying global magnetic fields are organized in a nearly
958: force-free poloidal configuration \footnote{According to the basic
959: properties of the Maxwell stress tensor (see, for instance,
960: \citet{PA79}) the magnetic field ${\bf B}$ creates in the atmospheric plasma
961: an isotropic pressure $\frac{B^2}{8 \pi}$ and tension $\frac{B_i B_j}{4 \pi}$
962: directed along the magnetic lines of force. While neighboring lines of force
963: of a magnetic field try to expand due to the magnetic pressure, tension tends
964: to compensate for this effect. The force-free configuration is possible only
965: when the gradient of the magnetic pressure is fully compensated by the
966: tension forces.}.
967: In contrast, if the suggested vertical orientation of the magnetic
968: field lines in the spot is correct, the uncompensated ``magnetic
969: pressure''${^3}$ of such a localized field may dominate against the tension
970: causing a strong impact on the pressure-temperature balance in the photosphere
971: of the degenerate. This may produce a temperature difference between the
972: strong-field area and other parts of the star's surface. As a result we
973: may expect rotationally-modulated photometric variability of WD\,1953-011
974: (as observed by \citet{Wad03} and \citet{BM05}).
975: For these reasons (and by analogy to sunspots) such
976: fields might be unstable if not supported by other
977: dynamical processes such as differential rotation, and
978: may therefore exhibit secular drift with respect to the stellar rotation axis.
979:
980: In the above context we note that significant photometric variability of
981: WD\,1953-011 has been established \citep{Wad03,BM05}. Also, remarkably,
982: the authors had encountered
983: problems analyzing the periodicity of the variable differential flux.
984: When individual epochs of their photometric data are phased
985: according to the rotation period of about 1.45 days, the resultant folded
986: lightcurves are smooth and approximately sinusoidal. However,
987: they had difficulty obtaining an acceptable fit to all epochs
988: of photometric data considered simultaneously \citep{Wad03}.
989: Besides, periodograms \citep{BM05} obtained separately
990: for their 7 individual observing
991: runs indicate {\it a significant spread in the period
992: distribution}. The individual peaks are stochastically distributed
993: around a rotation period $P \approx 1^d.45$ from $P \approx 1^d.415$
994: to $P \approx 1^d.48$, also suggesting a probable phase shift from epoch
995: to epoch with characteristic times from tens to hundred of days.
996: Combining all the data, they establish their version ot the rotation period
997: $P = 1.^d441769(8)$\,days. This period is significantly different from
998: the period derived by us from the behavior of the global field of the
999: star (P~=~1.448~$\pm$~0.0001~days).
1000:
1001: Indirectly, these facts suggest a
1002: physical relationship between the darkness and magnetic spots, and their
1003: possible secular migration. Unfortunately, our spectroscopy and
1004: the available photometry were obtained at different epochs, making it impossible
1005: to study this relationship in this paper. Examination of these
1006: problems will be among the goals of our further study of this magnetic
1007: degenerate upon carrying out the necessary simultaneous photometric
1008: and spectral observations of this star.
1009:
1010: \begin{acknowledgements}
1011: Our thanks to L.~Ferrario, P.~Maxted, and C.~Brinkworth for providing details of
1012: individual spectroscopic and photometric measurements of WD\,1953-011.
1013: We are also especially grateful to Stefan Jordan
1014: (our referee) for valuable comments,
1015: suggestions and his high estimation of our work.
1016: GV is grateful to the Korean
1017: MOST (Ministry of Science and Technology, grant M1-022-00-0005) and KOFST
1018: (Korean Federation of Science and Technology Societies) for providing
1019: him an opportunity to work at KAO through the Brain Pool program.
1020: GAW and JDL acknowledge Discovery Grant support from the Natural Sciences and
1021: Engineering Research Council of Canada. This study was also partially
1022: supported by KFICST (grant 07-179). Based on observations collected at the
1023: European Southern Observatory, Chile (ESO program 67.D-0306(A))
1024: \end{acknowledgements}
1025:
1026: \begin{thebibliography}{}
1027: \bibitem[Angel et al.(1973)]{AMS73}
1028: Angel, J.R.P., McGraw, J.T. \& Stockman, H.S., 1973,
1029: ApJ, 184, L79
1030: \bibitem[Angel et al.(1981)]{ABL81}
1031: Angel, J.R.P., Borra, E.F., \& Landstreet, J.D., 1981, ApJS, 45, 457
1032: \bibitem[Appenzeller(1967)]{A67}
1033: Appenzeller, I., 1967, PASP, 79, 136
1034: \bibitem[Appenzeller et al.(1998)]{AFF98}
1035: Appenzeller, I., Fricke, K., Furtig, W., et al., 1998, The Messenger,
1036: 94, 1.
1037: \bibitem[Afanasief et al.(1995)]{ABVD95}
1038: Afanasiev, V.L., Burenkov, A.N., Vlasyuk, V.V. \& Drabek S.V.,
1039: 1995, SAO Ras internal rep., 234
1040: \bibitem[Aznar Cuadrado et al.(2004)]{AZC04}
1041: Aznar Cuadrado, R., Jordan, S., Napiwotzki, R., Schmid, H.M., Solanki, S.K.,
1042: \& Mathys, G., 2004, A\&A, 423, 1081
1043: \bibitem[Bagnulo et al.(2000)]{Bag00}
1044: Bagnulo, S., Landolfi, M., Mathys, G., \& Landi Degl'Innocenti, M.,
1045: 2000, A\&A, 358, 929
1046: \bibitem[Bagnulo et al.(2002)]{Bag02}
1047: Bagnulo, S., Szeifert, T., Wade, G., Landstreet, J., Mathys, G., 2002,
1048: A\&A, 389, 191
1049: \bibitem[Barstow et al.(2005)]{BJD95}
1050: Barstow, M.A., Jordan, S., O'Donohue, D., Burleigh, M.R., Napitowski, R.,
1051: \& Harrop-Alin, M.K., 1995, MNRAS, 277, 971
1052: \bibitem[Bragaglia et al.(1995)]{BRG95}
1053: Bragaglia, A., Renzini, A., \& Bergeron, P., 1995, ApJ, 443, 735
1054: \bibitem[Brinkworth et al.(2005)]{BM05}
1055: Brinkworth, C.S., Marsh, T.R., Morales-Rueda, L., Maxted, P.F.L.,
1056: Burleigh, M.R., \& Good, S.A., 2005, MNRAS, 356, 89
1057: \bibitem[Euchner et al.(2002)]{EJB02}
1058: Euchner, F., Jordan, S., Beuermann, K., G\"ansicke, B.T., and
1059: Hessman, F.V., 2002, A\&A 390, 633
1060: \bibitem[Fabrika et al.(2003)]{FVB03}
1061: Fabrika, S.N., Valyavin, G.G., \& Burlakova, T.E., 2003,
1062: Astronomy Letters, 29, 737
1063: \bibitem[Goransky(2004)]{G04}
1064: Goransky, V., 2004, private communication.
1065: %\bibitem[Hamada \& Salpeter(1961)]{HS61}
1066: %Hamada, T., Salpeter, E.E., 1961, ApJ, 134, 683
1067: \bibitem[Jordan(1992)]{J92}
1068: Jordan, S., 1992, A\&A, 265, 570
1069: \bibitem[Jordan et al.(2007)]{JAN07}
1070: Jordan, S., Aznar Cuadrado, R., Napiwotzki, R.,
1071: Schmidt, H.M., \& Solanki, S.K., 2007, A\&A, 462, 1097
1072: \bibitem[Kawka et al.(2007)]{KAW07}
1073: Kawka, A., Vennes, S., Schmidt, G.D., Wickramasinghe, D.T., \&
1074: Koch, R., 2007, ApJ, 654, 499
1075: \bibitem[Koester et al.(1998)]{KDWA98}
1076: Koester, D., Dreizler, S., Weidemann, V., \& Allard, N.F. 1998,
1077: A\&A, 338, 612
1078: \bibitem[Landolfi et al.(1997)]{Lan97}
1079: Landolfi, M., Bagnulo, S., Landi Degl'Innocenti, E.,
1080: Landi Degl'Innocenti, M., \& Leroy, J.L.,
1081: 1997, A\&A 322, 197
1082: \bibitem[Landstreet(1980)]{L80}
1083: Landstreet, J.D. 1980, Astronomical Journal, 85, 611
1084: \bibitem[Liebert et al.(2003)]{LBH03}
1085: Liebert, J., Bergeron, P., \& Holberg, J.B. 2003, AJ, 125, 348
1086: \bibitem[Lafler \& Kinman(1965)]{LK65}
1087: Lafler, J., \& Kinman, T.D., 1965, ApJS, 11, 216
1088: \bibitem[Maxted et al.(2000)]{MFMW00}
1089: Maxted, P.F.L., Ferrario, L., Marsh, T.L., Wickramasinghe, D.T.,
1090: 2000, MNRAS, 315, L41
1091: %\bibitem[1995]{M95}
1092: %Mathys, G., 1995, A\& A, 293, 733
1093: \bibitem[Naydenov et al.(2002)]{NVF02}
1094: Naydenov, I.D., Valyavin, G.G., Fabrika, S.N., et al
1095: 2002, Bull. Spec. Astrophys. Obs., 53, 124
1096: \bibitem[Parker(1979)]{PA79}
1097: Parker, E. N. 1979, {\it Cosmical magnetic fields},
1098: Oxford: Clarendon press
1099: %\bibitem[1966]{PI66}
1100: %Pikelner, S. B. 1966, {\it Principles of cosmic electrodynamics},
1101: %(in Russian), Moscow: Nauka
1102: \bibitem[Putney(1999)]{P99}
1103: Putney, A. 1999, 11th. European Workhop on White Dwarfs, ASP Conf.
1104: Series, 196, 195. J.E., Solheim, E.G., Meistas ed.
1105: \bibitem[Schmidt \& Northworthy(1991)]{SN91}
1106: Schmidt, G.D. \& Northworthy, J.N., 1991, ApJ, 366, 270
1107: \bibitem[Schmidt et al.(1992)]{SBJ92}
1108: Schmidt, G.D., Bergeron, P., Liebert, J., \& Saffer, Rex A., 1992,
1109: ApJ, 394, 603
1110: \bibitem[Schmidt \& Smith(1994)]{SS94}
1111: Schmidt, G.D. \& Smith, P.S., 1994, ApJ, 423, L63
1112: \bibitem[Schmidt \& Smith(1995)]{SS95}
1113: Schmidt, G.D. \& Smith, P.S., 1995, ApJ, 448, 305
1114: \bibitem[Stibbs(1950)]{S50}
1115: Stibbs, D.W.N., 1950, MNRAS, 110, 395
1116: \bibitem[Unno(1956)]{U56}
1117: Unno, W., 1956, PASJ, 8, 108
1118: \bibitem[Valyavin et al.(2003)]{VBFM03}
1119: Valyavin, G.G., Burlakova, T.E., Fabrika, S.N, \& Monin, D.N.,
1120: 2003, Astronomy Reports, 47, 589
1121: \bibitem[Valyavin et al.(2005)]{VBMF05}
1122: Valyavin, G.G., Bagnulo, S., Monin, D.N., Fabrika, S.N,
1123: Lee, B.-C., Galazutdinov, G., Wade, G.A., \& Burlakova, T.
1124: 2005, A\& A, 439, 1099
1125: \bibitem[Valyavin et al.(2006)]{VBF06}
1126: Valyavin, G.G., Bagnulo, S., Fabrika, S.N,
1127: Reisenegger, A., Wade, G.A., Han, Inwoo, \& Monin, D.N.,
1128: 2006, ApJ, 648, 559
1129: \bibitem[Wade et al.(2003)]{Wad03}
1130: Wade, G.A., Bagnulo, S., Szeifert, T., Brinkworth, C., Marsh, T., Landstreet,
1131: J.D., Maxted, P. 2003, in: Solar Polarization, J.~Trujillo
1132: Bueno \& J.~S\'{a}nchez Almeida (eds.), ASP Conference Series
1133: No.~307, p.~565
1134: \bibitem[Wickramasinghe \& Martin(1979)]{WM79}
1135: Wickramasinghe, D.T., \& Martin, B. 1979, MNRAS, 188, 165
1136: \end{thebibliography}
1137:
1138: \newpage
1139: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%TABLE1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1140: \begin{table}
1141: \caption{\label{tbl1} Spectral and spectropolarimetric observations
1142: of WD\,1953-011: column~1 is the Julian Date of the midpoint of the
1143: observation, column~2 is the exposure time, and column~3 reports
1144: the telescope used in the observations ({\bf AAT} indicates the high
1145: resolution spectroscopy presented by \citet{MFMW00}).
1146: For data obtained with the VLT
1147: the exposure times are presented for the three consecutive
1148: $I,V$ / $Q$ / $U$ modes of observations (in this case the midpoint
1149: corresponds to observations of the Stokes $I,V$ parameters).
1150: }
1151: \begin{tabular}{|l|c|c|l|}
1152: \hline
1153: \hline
1154: JD & $Exp$ (sec)& Stokes&Telescope\\
1155: \hline
1156: 2450676.955& 600 &$I$& AAT \\
1157: 2451391.948& 600 &$I$& AAT \\
1158: 2451391.955& 600 &$I$& AAT \\
1159: 2451391.962& 600 &$I$& AAT \\
1160: 2451392.059& 1800&$I$& AAT \\
1161: 2451392.957& 1800&$I$& AAT \\
1162: 2451393.066& 1800&$I$& AAT \\
1163: 2451393.106& 1800&$I$& AAT \\
1164: 2451393.943& 1200&$I$& AAT \\
1165: 2451393.958& 1200&$I$& AAT \\
1166: 2451393.973& 1200&$I$& AAT \\
1167: 2451393.988& 1200&$I$& AAT \\
1168: 2451394.003& 1200&$I$& AAT \\
1169: 2452048.801 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1170: 2452048.893 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1171: 2452076.671 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1172: 2452076.883 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1173: 2452078.722 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1174: 2452078.879 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1175: 2452079.672 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1176: 2452079.892 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1177: 2452087.621 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1178: 2452087.670 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1179: 2452087.722 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1180: 2452087.768 & 840/1200/1200 &$I,V/Q/U$& VLT\\
1181: 2452505.290 & 3600 &$I,V$& BTA\\
1182: 2452505.327 & 3600 &$I,V$& BTA\\
1183: 2452505.360 & 3600 &$I,V$& BTA\\
1184: 2452505.397 & 3600 &$I,V$& BTA\\
1185: \hline
1186: \hline
1187: \end{tabular}
1188: \end{table}
1189: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%TABLE1%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1190: \newpage
1191:
1192: %%%%%%%%%%%%%%%%%%%%%%%%%%Table2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1193: \begin{table}
1194: \caption{\label{tbl2} Determinations of the mean modulus $|B_G|$
1195: of the weak-field component. Column\,1 is the Julian Date, Col.\,2
1196: and Col.\,3 are the equivalent widths of the H$\alpha$ core ($EW_{core}$)
1197: and associated error bar, Col.\,4 and Col.\,5 are the inferred
1198: field strength $|B_G|$ and its error bar $\sigma$\,(kG), Col.\,6
1199: is the telescope used. Uncertainties at the measured equivalent
1200: widths are calculated as a noise fraction of the flux (due to Poisson noise)
1201: in the total flux under the line profile.
1202: The mean field modulus and its uncertainty obtained from the high-resolution
1203: spectroscopy with the {\bf AAT} are taken from \citet{MFMW00}.
1204: Uncertainties at the calibrated field strengths (observations with the
1205: {\bf VLT} and {\bf BTA}) result from regression errors in the
1206: $EW_{core}$\,--\,$|B_G|$ relationship shown in Fig.\,4.
1207: }
1208: \begin{tabular}{|l|c|c|c|c|l|}
1209: \hline
1210: \hline
1211: JD & $EW_{core}$& $\sigma$ & $|B_G|$ $ (kG)$& $\sigma (kG)$ &Telescope\\
1212: \hline
1213: 2450676.955 & 1.040 & 0.008 & 91 & 5 & AAT\\
1214: 2451391.948 & 1.073 & 0.008 & 93 & 4 & AAT\\
1215: 2451391.955 & 1.105 & 0.016 & 100 & 4 & AAT\\
1216: 2451391.962 & 1.045 & 0.008 & 93 & 4 & AAT\\
1217: 2451392.059 & 1.068 & 0.012 & 93 & 2 & AAT\\
1218: 2451392.957 & 0.905 & 0.012 & 83 & 1 & AAT\\
1219: 2451393.066 & 0.943 & 0.012 & 80 & 2 & AAT\\
1220: 2451393.106 & 0.933 & 0.012 & 83 & 1 & AAT\\
1221: 2451393.947 & 1.008 & 0.020 & 92 & 3 & AAT\\
1222: 2451393.958 & 1.013 & 0.016 & 87 & 2 & AAT\\
1223: 2451393.973 & 0.945 & 0.016 & 84 & 2 & AAT\\
1224: 2451393.988 & 0.935 & 0.020 & 84 & 2 & AAT\\
1225: 2451394.003 & 0.981 & 0.020 & 83 & 2 & AAT\\
1226: 2452048.801 & 0.901 & 0.012 & 80 & 3 & VLT\\
1227: 2452048.893 & 0.917 & 0.012 & 81 & 3 & VLT\\
1228: 2452076.671 & 1.012 & 0.012 & 89 & 3 & VLT\\
1229: 2452076.883 & 1.057 & 0.016 & 93 & 3 & VLT\\
1230: 2452078.722 & 1.013 & 0.016 & 89 & 3 & VLT\\
1231: 2452078.879 & 0.939 & 0.012 & 83 & 3 & VLT\\
1232: 2452079.672 & 1.052 & 0.016 & 93 & 3 & VLT\\
1233: 2452079.892 & 1.102 & 0.012 & 97 & 3 & VLT\\
1234: 2452087.621 & 0.952 & 0.016 & 82 & 3 & VLT\\
1235: 2452087.670 & 0.939 & 0.016 & 83 & 3 & VLT\\
1236: 2452087.722 & 0.911 & 0.016 & 78 & 3 & VLT\\
1237: 2452087.768 & 0.921 & 0.016 & 79 & 3 & VLT\\
1238: 2452505.290 & 0.978 & 0.020 & 84 & 4 & BTA\\
1239: 2452505.327 & 1.000 & 0.020 & 88 & 4 & BTA\\
1240: 2452505.360 & 1.028 & 0.020 & 91 & 4 & BTA\\
1241: 2452505.397 & 1.034 & 0.020 & 91 & 4 & BTA\\
1242: \hline
1243: \hline
1244: \end{tabular}
1245: \end{table}
1246: %%%%%%%%%%%%%%%%%%%%%%%%%%Table2%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1247: \newpage
1248:
1249: %%%%%%%%%%%%%%%%%%%%%%%%%%%Table3%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1250: \begin{table}
1251: \caption{\label{tbl3} Determinations of the mean modulus $|B_{spot}|$
1252: of the strong-field component. Column\,1 is the Julian Date, Col.\,2
1253: and Col.\,3 are relative area of the spot on the disc $S$ (in percent
1254: of the full disc area)
1255: and associated error bar obtained as a noise fraction of the flux in the
1256: total flux under the strong-field satellite features. Col.\,4 and Col.\,5 are the
1257: magnetic field strength $|B_{spot}|$ and its error bar $\sigma$\,(kG) obtained
1258: as uncertainty in the determination of the satellite positions deblended by
1259: Gaussians. Col.\,6 is the telescope used.)}
1260: \begin{tabular}{|l|c|c|c|c|l|}
1261: \hline
1262: \hline
1263: JD & $S $(\%)& $\sigma$ (\%) & $|B_{spot}|$ $(kG)$& $\sigma (kG)$ &Telescope\\
1264: \hline
1265: 2450676.955 & 13.1 & 0.6 & 521 & 40 & AAT\\
1266: 2451391.948 & 9.5 & 0.6 & 513 & 30 & AAT\\
1267: 2451391.955 & 12.0 & 1.2 & 495 & 30 & AAT\\
1268: 2451391.962 & 11.7 & 0.6 & 494 & 30 & AAT\\
1269: 2451392.059 & 12.0 & 0.6 & 527 & 15 & AAT\\
1270: 2451392.957 & 0.9 & 0.9 &invisible & & AAT\\
1271: 2451393.066 & 0.6 & 0.9 &invisible & & AAT\\
1272: 2451393.106 & 0.7 & 0.9 &invisible & & AAT\\
1273: 2451393.947 & 5.4 & 1.5 &invisible & & AAT\\
1274: 2451393.958 & 2.4 & 1.2 &invisible & & AAT\\
1275: 2451393.973 & 0.6 & 1.2 &invisible & & AAT\\
1276: 2451393.988 & 0.7 & 1.5 &invisible & & AAT\\
1277: 2451394.003 & 0.7 & 1.5 &invisible & & AAT\\
1278: 2452048.801 & 0.6 & 0.9 &invisible & & VLT\\
1279: 2452048.893 & 0.6 & 0.9 &invisible & & VLT\\
1280: 2452076.883 & 7.8 & 1.2 & 520 & 15 & VLT\\
1281: 2452078.622 & 3.2 & 1.2 &invisible & & VLT\\
1282: 2452078.879 & 0.5 & 1.2 &invisible & & VLT\\
1283: 2452079.672 & 8.4 & 1.2 & 529 & 30 & VLT\\
1284: 2452079.892 & 12.3 & 0.9 & 511 & 15 & VLT\\
1285: 2452087.621 & 0.6 & 1.2 &invisible & & VLT\\
1286: 2452087.670 & 0.6 & 1.2 &invisible & & VLT\\
1287: 2452087.722 & 0.7 & 1.2 &invisible & & VLT\\
1288: 2452087.768 & 0.6 & 1.2 &invisible & & VLT\\
1289: 2452505.290 & 10.4 & 1.2 & 500 & 35 & BTA\\
1290: 2452505.327 & 12.3 & 1.2 & 492 & 40 & BTA\\
1291: 2452505.360 & 12.3 & 1.2 & 524 & 35 & BTA\\
1292: 2452505.397 & 10.8 & 1.5 & 502 & 45 & BTA\\
1293: \hline
1294: \hline
1295: \end{tabular}
1296: \end{table}
1297: %%%%%%%%%%%%%%%%%%%%%%%%%%%Table3%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1298: \newpage
1299:
1300:
1301: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%Table4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1302: \begin{table}
1303: \caption{\label{tbl4} Determinations of the longitudinal magnetic field
1304: of the low- and high-field components of WD\,1953-011.
1305: Column\,1 is the Julian Date, Col.\,2
1306: and Col.\,3 are the longitudinal field of the low-field component
1307: and associated error bar, Col.\,4 and Col.\,5 are the deduced
1308: longitudinal magnetic field of the strong-field component and its error
1309: bar (``no'' means ``below the detection level''),
1310: Col.\,6 is the telescope used.)}
1311: \begin{tabular}{|l|c|c|c|c|l|}
1312: \hline
1313: \hline
1314: JD & $B_G^l$ $(kG)$& $\sigma$ $(kG)$ & $B_l^s$ $(kG)$& $\sigma$ $(kG)$ &OBS\\
1315: \hline
1316: 2452048.801 & -41.5 & 1.5 & no & & VLT\\
1317: 2452048.893 & -39.6 & 1.6 & no & & VLT\\
1318: 2452076.883 & -41.0 & 1.6 & 430 & 70 & VLT\\
1319: 2452078.722 & -42.9 & 1.8 & & & VLT\\
1320: 2452078.879 & -42.2 & 1.7 & no & & VLT\\
1321: 2452079.672 & -41.9 & 1.6 & 360 & 60 & VLT\\
1322: 2452079.892 & -46.8 & 1.7 & 460 & 60 & VLT\\
1323: 2452087.621 & -41.5 & 1.6 & no & & VLT\\
1324: 2452087.670 & -39.8 & 1.7 & no & & VLT\\
1325: 2452087.722 & -40.1 & 1.7 & no & & VLT\\
1326: 2452087.768 & -40.1 & 1.5 & no & & VLT\\
1327: 2452505.290 & -46.2 & 2.3 & 440 & 80 & BTA\\
1328: 2452505.327 & -44.8 & 2.3 & 450 & 80 & BTA\\
1329: 2452505.360 & -45.0 & 2.5 & no & & BTA\\
1330: 2452505.397 & -42.0 & 2.7 & no & & BTA\\
1331: \hline
1332: \hline
1333: \end{tabular}
1334: \end{table}
1335: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%Table4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1336: \newpage
1337:
1338: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%Table5%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1339: \begin{table}
1340: \caption{\label{tbl5} Phase-resolved observable magnetic
1341: quantities of the weak-field component of WD\,1953-011. The first
1342: column is rotation phase $\phi$ obtained according to the magnetic ephemeris
1343: characterized by the rotational period of 1.448 days found here;
1344: the second and third columns are the mean field
1345: modulus or ``surface magnetic field'' $|B_G|$ and associated error bar;
1346: the fourth and fifth columns are
1347: the longitudinal field $B_l^G$ and its error bar.
1348: }
1349: \hspace*{1.5cm}
1350: \begin{tabular}{|l|c|c|c|l|}
1351: \hline
1352: \hline
1353: $\phi $ & $|B_G|$ & $\sigma $ & $B_G^l$ & $\sigma$ \\
1354: & $(kG)$ & $(kG)$ & $(kG)$ & $(kG)$ \\
1355: \hline
1356: 0.05 & 81.5 & 4 & -40.5 &1.3 \\
1357: 0.15 & 82 & 2 & & \\
1358: 0.25 & 89 & 3 & & \\
1359: 0.35 & 94 & 3 & -44 &1.5 \\
1360: 0.45 & 93 & 2 & & \\
1361: 0.55 & 97 & 3 & -47 &2 \\
1362: 0.65 & 89 & 3 & & \\
1363: 0.75 & 86 & 4 & -42.2 &2 \\
1364: 0.85 & 83 & 1 & -41 &1 \\
1365: 0.95 & 78 & 2 & -40 &1 \\
1366: \hline
1367: \hline
1368: \end{tabular}
1369: \end{table}
1370:
1371:
1372:
1373: \newpage
1374:
1375: \begin{figure}
1376: \centering
1377: \includegraphics[width=18.5cm, height=14cm, angle=0]{f1.eps}
1378: \caption{Stokes $IQUV$ H$\alpha$ profile timeseries of WD1953-011,
1379: obtained using the FORS1 spectropolarimeter at the ESO VLT. Phases
1380: correspond to the magnetic ephemeris obtained in this paper
1381: and are expressed in part per mil at right. The thin lines
1382: represent the observations obtained at phase 0, and are reproduced
1383: to emphasise the variability of the Stokes profiles.
1384: }
1385: \label{fig1}
1386: \end{figure}
1387:
1388:
1389: \begin{figure}
1390: \centering
1391: \hspace*{-0.5cm}
1392: \includegraphics[width=16cm, height=19cm, angle=270]{f2.eps}
1393: \caption{Stokes $V$ at the H$\alpha$ line obtained with the VLT
1394: (the solid line) and the BTA (the dashed line). The spectra correspond
1395: to phases of the maximum visible strong-field Zeeman satellite features
1396: at the H$\alpha$ wings ($\phi \approx 0.3$ in observations with the BTA and
1397: $\phi \approx 0.5$ in observations with the VLT: the observed phase shift is
1398: discussed in Sect.\,8 and Sect.\,9 of this study.)
1399: }
1400: \label{fig2}
1401: \end{figure}
1402:
1403: \begin{figure}
1404: \centering
1405: %\hspace*{-1.cm}
1406: \includegraphics[width=6.8cm, height=10.0cm, angle=270]{f3.eps}
1407: \caption{The H$\alpha$ profiles obtained at the VLT, AAT and BTA.
1408: High resolution spectra are convolved to the spectral resolution of
1409: the FORSE1 and UAGS. The solid lines illustrate profiles at two extreme
1410: rotation phases at which the spot component is most clearly seen
1411: (the shallowest profile) and where the spot component is
1412: absent (the deepest profile).
1413: }
1414: \label{fig3}
1415: \end{figure}
1416:
1417: \begin{figure}
1418: \centering
1419: \hspace*{-1.cm}
1420: \includegraphics[width=6.8cm, height=9.0cm, angle=0]{f4.eps}
1421: \caption{
1422: The relationship $EW_{core}$ -- $|B_G|$ obtained from the convolved
1423: high-resolution spectra for which $|B_G|$ values are estimated by
1424: \citet{MFMW00}. The dotted line is a linear fit of
1425: the relationship.
1426: }
1427: \label{fig4}
1428: \end{figure}
1429:
1430: \begin{figure}
1431: \centering
1432: %\hspace*{-0.9cm}
1433: \includegraphics[width=7.8cm, height=10.0cm, angle=270]{f5.eps}
1434: \caption{
1435: An example of the model technique as described in sect.~5. The thick
1436: solid line is the Stokes $V$ observed spectrum containing strong circular
1437: polarization from the strong-field component; the dotted and dashed lines are
1438: modeled Stokes $V$ spectra of the weak- and strong-field components
1439: respectively; the thin solid line is their sum.
1440: }
1441: \label{fig5}
1442: \end{figure}
1443:
1444: \begin{figure}
1445: \centering
1446: \includegraphics[width=8.5cm, height=5.5cm, angle=0]{f6.eps}
1447: \caption{Power spectrum of the magnetic field variations in WD\,1953-011.
1448: }
1449: \label{fig6}
1450: \end{figure}
1451:
1452: \begin{figure}
1453: \vspace*{0.5cm}
1454: \centering
1455: \includegraphics[width=8.5cm, height=5.5cm, angle=0]{f7.eps}
1456: \vspace*{0.5cm}
1457: \caption{The phase curve of the equivalent widths at the H$\alpha$ core phased
1458: with the 1.4480-day period.
1459: }
1460: \label{fig7}
1461: \end{figure}
1462:
1463: \begin{figure}
1464: \vspace*{0.5cm}
1465: \centering
1466: \includegraphics[width=8.5cm, height=6.5cm, angle=0]{f8.eps}
1467: \vspace*{0.1cm}
1468: \caption{
1469: The magnetic phase curves of WD\,1953-011 and their fits with the
1470: 1.448-day period. The upper plot illustrates variation of
1471: the field modulus of the weak-field
1472: component; the lower plot is variation of the its longitudinal magnetic field.
1473: }
1474: \label{fig8}
1475: \end{figure}
1476:
1477:
1478: \begin{figure}
1479: \centering
1480: \includegraphics[width=9cm, height=14cm, angle=0]{f9.eps}
1481: %\vspace*{-3.1cm}
1482: \caption{
1483: Observations and modeling of mean longitudinal field (bottom panel) and
1484: mean field modulus, or surface field (top panel). The dashed lines
1485: show a fit obtained with dipole model. The solid lines show the best-fit
1486: obtained by means of a dipole + quadrupole model.
1487: }
1488: \label{fig9}
1489: \end{figure}
1490:
1491: \begin{figure}
1492: \centering
1493: \includegraphics[width=12.5cm, height=8.5cm, angle=0]{f10.eps}
1494: %\vspace*{-3.1cm}
1495: \caption{
1496: Phase-resolved projection S of the strong-field area:
1497: open circles illustrate the data from the AAT, filled circles
1498: and triangles are the observations with the VLT and BTA respectively.
1499: All the data have been phased according to the magnetic ephemeris obtained
1500: in Sect.~6.
1501: }
1502: \label{fig10}
1503: \end{figure}
1504:
1505: \begin{figure*}
1506: \centering
1507: \includegraphics[width=13.5cm, height=16.0cm, angle=0]{f11a.eps}
1508: %\vspace*{-3.1cm}
1509: \caption{Model fits (red lines) of the observed (black lines)
1510: Stokes $IVQU$ spectra obtained in observations with the VLT.
1511: The corresponding tomographic portraits of the star's
1512: magnetosphere and rotational phase (marked as {\bf Phase} ) are presented
1513: at left. The strong-field area is shown by white. The magnetic field line of
1514: force are red lines.
1515: }
1516:
1517: \label{fig11}
1518: \end{figure*}
1519: \newcounter{10}
1520:
1521: \begin{figure*}
1522: \centering
1523: \includegraphics[width=13.5cm, height=16.0cm, angle=0]{f11b.eps}
1524: %\vspace*{-3.1cm}
1525: \caption{
1526: continued.
1527: }
1528: \label{fig12}
1529: \end{figure*}
1530:
1531:
1532: \end{document}
1533:
1534: