1: \documentclass[12pt,preprint]{aastex}
2: \begin{document}
3: \title{Far Ultraviolet Spectral Analysis of the Prototype Nova-Like Variable VY Sculptoris
4: from the High State to the Low State}
5:
6: \author{Ryan T. Hamilton}
7: \affil{Dept. of Astronomy,
8: New Mexico State University,
9: Las Cruces, NM 88003,
10: email: rthamilt@nmsu.edu}
11:
12: \author{Edward M. Sion}
13: \affil{Dept. of Astronomy \& Astrophysics,
14: Villanova University,
15: Villanova, PA 19085,
16: e-mail: edward.sion@villanova.edu}
17:
18: \begin{abstract}
19:
20: The prototype nova-like variable VY Sculptoris was observed by the IUE during four different optical
21: brightness states of the system. The FUV flux level from the highest state to the lowest state declines
22: by a factor of 28. We have carried out model accretion disk and white dwarf atmosphere fitting to the
23: spectra. The corresponding accretion rates range from $\dot{M}= 8 \times 10^{-9}$M$_{\sun}$ yr$^{-1}$
24: at the highest FUV flux level down to $\dot{M}= 1.9\times 10^{-10}$M$_{\sun}$ yr$^{-1}$ at the lowest flux level.
25: We report tentative evidence for the detection of the underlying white dwarf with $T_{\rm eff} = 45,000$K
26: in the spectrum with the lowest flux level.
27: \end{abstract}
28:
29: Subject Headings: keyword{Stars} Stars: cataclysmic variables, white dwarfs, Physical
30: Processes: accretion, accretion disks
31:
32: \section{Introduction}
33:
34: Cataclysmic variables (CVs) are short-period binaries in which a
35: late-type, Roche-lobe-filling main-sequence dwarf transfers gas through an
36: accretion disk onto a rotating, accretion-heated white dwarf (WD). The
37: nova-like variables are a subclass of CVs in which the mass-transfer rate
38: is high and the light of the system is dominated by a very bright
39: accretion disk. The spectra of nova-like variables resemble those of
40: classical novae (CNe) that have settled back to quiescence. However,
41: nova-like variables have never had a recorded CN outburst. Hence their
42: evolutionary status is a mystery. They could be close to having their next
43: CN explosion, or they may have had an unrecorded explosion, possibly
44: hundreds or thousands of years ago. Adding to the mystery of nova-like
45: variables is that some of them (known as the VY Sculptoris stars after
46: their prototype) show the curious behavior of being in a high optical
47: brightness state most of the time, but then, for no apparent reason,
48: plummeting into a deep low-brightness state with little or no ongoing
49: accretion. Then, just as unpredictably, they return to the high-brightness
50: state. These drops are possibly related to cessation of mass transfer from
51: the K-M dwarf secondary star either by starspots positioned under L$_{1}$
52: (Livio and Pringle 1994) or irradiation feedback in which an inflated
53: outer disk can modulate the mass transfer from the secondary by blocking
54: irradiation of the hot inner accretion disk region(Wu et al. 1995).
55:
56: VY Scl's brightest state has V = 12.9 and its low state has V = 18.5
57: (Warner 1987). Its orbital period is 0.1662 days but little else is known
58: (Hollander, A., Kraakman, H., van Paradijs, J.1993 and references
59: therein). Martinez-Pais et al (2000) found the white dwarf mass to be
60: $M_{(wd)} = 1.22\pm{0.22}M_{\sun}$, the mass of the secondary $M_{(2)} =
61: 0.43\pm{0.13} M_{\sun}$ and the binary inclination $i = 30^{o}\pm{10} $ In
62: its low state, VY Scl shows emission lines similar to dwarf novae in
63: quiescence (Szkody 1987). A reddening correction of E(B-V) = 0.06 is given
64: by Bruch and Engle (1995). Unfortunately, the distance is unknown.
65:
66: During the low states of VY Scl stars, it is possible to obtain nearly
67: uncontaminated spectra of the WD component because the accretion disk is
68: very faint and cool if present at all. Thus, through model-atmosphere
69: fitting during the low states, it is possible to determine temperatures,
70: gravities, and abundances for the WD. This paper reports an analysis of
71: FUV archival spectra which sample several different brightness states of
72: the prototype, VY Scl itself. Among the key questions we seek to answer
73: are the following: What are the accretion rates during outburst and
74: quiescence? Is this rate of accretion consistent with the suppression of
75: the disk instability mechanism? How much flux does the white dwarf
76: contribute to the FUV? How much flux does the accretion disk contribute
77: to the FUV? How hot is the white dwarf?
78:
79: \section{Far Ultraviolet Spectroscopic Observations}
80:
81: Far ultraviolet spectra of VY Scl were obtained over a period of four
82: years with the International Ultraviolet Explorer Spacecraft short
83: wavelength prime camera (SWP) through the large aperture at low dispersion
84: with a resolution of 5\AA. The SWP spectra covered the wavelength range
85: 1170 \AA\ - 2000 \AA. Fortuitously, the series of observations included
86: four spectra, SWP32594, SWP29755, SWP06123 and SWP21523, which covered a
87: variety of brightness states from the optical high brightness state when
88: the system was most luminous to intermediate brightness states down the
89: low optical brightness state when, in VY Scl nova-like variables, the
90: underlying white dwarf is thought to be exposed and the accretion disk had
91: greatly declined in brightness or gone away altogether. Due to large gaps
92: in the AAVSO observations archive for VY Scl, we cannot display the
93: placement of the observations on a light curve. However, the AAVSO data
94: do indicate that at the time that SWP21523 was taken, the system was in a
95: deep low state with V < 15.8. Surprisingly, only one of the spectra,
96: SWP32594, has been discussed in the literature (LaDous 1991; Hoare and
97: Drew 1993). In Table 1, an observing log of the IUE archival spectra is
98: presented in which by column: (1) lists the SWP spectrum number, (2) the
99: aperture diameter, (3) the exposure time in seconds, (4) the date and time
100: of the observation, (5) the continuum to background counts, and (6) the
101: brightness state of the system.
102:
103: \begin{deluxetable}{lccccccrcl}
104: \tablecaption{IUE Observing Log
105: }
106: \tablenum{1}
107: \tablehead{
108: \colhead{SWP}
109: &\colhead{t$_{exp}$}
110: &\colhead{Disp.}
111: &\colhead{Ap.}
112: &\multicolumn{2}{c}{Date of Observation}
113: &\colhead{F($\lambda$1350)}
114: &\colhead{C}
115: &\colhead{B}
116: &\colhead{State}
117: }
118: \startdata
119: 32594 & 2100& LOW& Lg& 1987-12-23 &12:51:00& $ 5.5\times10^{-13}$& 144& 20 & High State\\
120: 29755 & 420 & LOW& Lg& 1986-11-26 &14:44:00& $3.5\times10^{-13}$& 41& 17 & High State\\
121: 06123 & 2700 & LOW& Lg& 1979-08-09& 01:02:00&$ 1.7\times10^{-13}$& 196 & 24 & Intermediate\\
122: 21523 & 7200 & LOW& Lg& 1983-11-12& 15:01:00&$ 2.0\times10^{-14}$& 65& 30 & Deep Low State\\
123: \hline
124: \enddata
125: \end{deluxetable}
126:
127:
128: The archival IUE NEWSIPS spectra were flux calibration-corrected using the
129: algorithm of Massa and Fitzpatrick (2000). Massa and Fitzpatrick (2000)
130: have shown that the absolute flux calibration of the NEWSIPS low
131: dispersion data was inconsistent with its reference model and subject to
132: time-dependent systematic effects, which together, amount to as much as
133: 10-15\%. Therefore, in order to correct the data and optimize the
134: signal-to-noise, we used the IDL programs which apply Massa-Fitzpatrick
135: corrections to VY Scl's low dispersion IUE data.
136:
137: The reddening of VY Scl was determined based upon all estimates listed in
138: the literature. The three principal sources of reddening values are the
139: compilations of la Dous (1991), Verbunt (1987) and Bruch \& Engel (1999).
140: The compilation of Bruch \& Engel listed E(B-V) = 0.06 for VY Scl.
141: Therefore the spectra were de-reddened with this color excess using the
142: IUERDAF routine UNRED. The brightness states of VY Scl at the time of the
143: IUE observations were assessed by comparison with the AAVSO light curve
144: data (visual magnitude versus Julian Date).
145:
146: In figure 1, we display the four spectra together on the same flux scale.
147: The spectrum with highest flux level (the top spectrum), SWP32594 is
148: followed successively downward by two spectra of intermediate brightness
149: and finally the bottom-most spectrum taken during a low state of VY Scl.
150:
151: The outburst spectrum is dominated by absorption lines including P Cygni
152: profile structure at C IV 1550, blue-shifted absorption at Si IV and
153: absorption features due to C II 1335, Si II 1260, and N IV 1718. In
154: SWP29755, the flux level has fallen to $3.5 \times 10^{-13}$ and the
155: absorption lines have weakened although N V 1240 and Si IV 1400 remain
156: prominent. In spectrum SWP06123, C IV emission is seen but, in general,
157: there is little convincing evidence for any other absorption features and
158: the flux at 1350\AA\ has declined to $1.7 \times 10^{-13}$. In the low
159: state spectrum, SWP21523, all regions where absorption appeared in
160: outburst and intermediate spectra have been replaced by emission lines and
161: there is an underlying continuum. The strongest such emission lines are
162: Si IV (1400) and C IV (1550).
163:
164: \section{Synthetic Spectral Fitting}
165:
166: We adopted model accretion disks from the optically thick disk model grid
167: of Wade \& Hubeny (1998). In these accretion disk models, the innermost
168: disk radius, R$_{in}$, is fixed at a fractional white dwarf radius of $x =
169: R_{in}/R_{wd} = 1.05$. The outermost disk radius, R$_{out}$, was chosen so
170: that T$_{eff}(R_{out})$ is near 10,000K since disk annuli beyond this
171: point, which are cooler zones with larger radii, would provide only a very
172: small contribution to the mid and far UV disk flux, particularly the SWP
173: FUV bandpass. The mass transfer rate is assumed to be the same for all
174: radii. Thus, the run of disk temperature with radius is taken to be:
175:
176: \begin{equation}
177: T_{eff}(r)= T_{s}x^{-3/4} (1 - x^{-1/2})^{1/4}
178: \end{equation}
179:
180: where $x = r/R_{wd}$
181: and $\sigma T_{s}^{4} = 3 G M_{wd}\dot{M}/8\pi R_{wd}^{3}$
182:
183: Limb darkening of the disk is fully taken into account in the manner
184: described by Diaz et al. (1996) involving the Eddington-Barbier relation,
185: the increase of kinetic temperature with depth in the disk, and the
186: wavelength and temperature dependence of the Planck function. The
187: boundary layer contribution to the model flux is not included. However,
188: the boundary layer is expected to contribute primarily in the extreme
189: ultraviolet below the Lyman limit.
190:
191: Model spectra with solar abundances were created for high gravity stellar
192: atmospheres using TLUSTY (Hubeny 1988) and SYNSPEC (Hubeny \& Lanz 1995).
193: Using IUEFIT, a $\chi^{2}$ minimization routine, both $\chi^{2}$ values
194: and a scale factor were computed for each model. The scale factor,
195: normalized to a kiloparsec, can be related to the white dwarf radius
196: through: $F_{\lambda(obs)} = 4 pi (R^{2}/d^{2})H_{\lambda(model)}$.
197:
198: After masking emission lines in the spectra, we determined separately for
199: each spectrum, the best-fitting white dwarf-only models and the
200: best-fitting disk-only models using IUEFIT, a $\chi^{2}$ minimization
201: routine. A dense grid of model spectra with solar abundances was created
202: for high gravity stellar atmospheres using TLUSTY (Hubeny 1988) and
203: SYNSPEC (Hubeny \& Lanz 1995). We took a range of gravities in the fitting
204: from Log $g = 7.0 - 9.0$ in steps of 0.5. For the white dwarf radii, we
205: use the mass-radius relation from the evolutionary model grid of Matt Wood
206: (private communication) for C-O cores.
207:
208: Taking the best-fitting white dwarf model and combining it with the
209: best-fitting disk model, we varied the accretion rate of the best-fitting
210: disk model by a small multiplicative factor in the range 0.1 to 10 using a
211: $\chi^{2}$ minimization routine called DISKFIT. Using this method the
212: best-fitting composite white dwarf plus disk model is determined based
213: upon the minimum $\chi^{2}$ value achieved and consistency of the scale
214: factor-derived distance with the adopted distance for each system. The
215: scale factor, $S$, normalized to a kiloparsec and solar radius, can be
216: related to the white dwarf radius R through: $F_{\lambda(obs)} = S
217: H_{\lambda(model)}$, where $S=4\pi R^2 d^{-2}$, and $d$ is the distance to
218: the source.
219:
220: The results of our fitting are summarized as follows. The spectra with the
221: three highest flux levels are all dominated by an accretion disk. The best
222: fitting disk models to SWP32594, SWP29755 and SWP06123 are displayed in
223: figures 2, 3, and 4. In each case the accretion disk provides essentially
224: all of the FUV flux even as the flux level at its peak in SWP32594 is
225: lower by a factor of 3 in SWP06123.
226:
227: In figure 2, the best-fitting disk model to SWP32594 corresponded to the
228: following parameters: accretion rate $\dot{M}= 8 \times
229: 10^{-9}$M$_{\sun}$ yr$^{-1}$, $M_{(wd)} = $1.0M$_{\sun}$, and an
230: inclination of $41^{o}$ with a $\chi^{2}$ = 5.96 The scale factor of this
231: best fit yields a distance to VY Scl of 620 parsecs.
232:
233: At the lower flux level of SWP29755, the best-fitting disk model is
234: displayed in figure 3. This fit corresponded to the following parameters:
235: accretion rate $\dot{M}= 5\times 10^{-9}$M$_{\sun}$ yr$^{-1}$, $M_{(wd)} =
236: $1.0M$_{\sun}$, inclination of $41^{o}$ with a $\chi^{2}$ = 7.86. The
237: scale factor of this fit yielded a distance of 616 pc.
238:
239: The best-fitting disk model to SWP06123 corresponded to the following
240: parameters: accretion rate $\dot{M}= 1.6\times 10^{-9}$M$_{\sun}$
241: yr$^{-1}$, $M_{(wd)} = $1.0M$_{\sun}$, and an inclination of $41^{o}$ with
242: $\chi^{2}$ = 5.74. The scale factor S yielded a distance to VY Scl of 536
243: parsecs. This fit is seen in figure 4
244:
245: The spectrum with the lowest flux level, SWP21523 was obtained during a
246: low state of VY Scl. The flux level at 1350A is lower by a factor of 28
247: than the spectrum in Figure 1 with the highest flux level. This spectrum
248: contains a very significant contribution (34\% of the flux) from a a hot
249: WD but with the flux of an accretion disk still accounting for the
250: majority (66\%) of the FUV flux. If this spectrum recorded the deepest low
251: state of VY Scl, then the presence of a strong disk component is a
252: departure from the totally white dwarf-dominated low state spectra of the
253: low state nova-like systems MV Lyra, TT Ari and DW UMa. Our best-fitting
254: white dwarf plus accretion disk combination fit to SWP21523 consisted of a
255: hot solar composition WD with $T_{\rm eff} = 45,000$K$\pm3000K$, log$g =
256: 8.5$ with a $\chi^{2}$ = 3.38 and accretion disk model with accretion rate
257: $\dot{M}= 1.9\times 10^{-10}$M$_{\sun}$ yr$^{-1}$, $M_{(wd)} =
258: $1.0M$_{\sun}$, and an inclination of $41^{o}$ with $\chi^{2}$ = 3.38. The
259: scale factor S yielded a distance to VY Scl of 536 parsecs. This fit is
260: seen in figure 5.
261:
262: It is important to explore the uncertainties in our estimated accretion
263: rates, $\dot{M}$. In order to assess the errors in $\dot{M}$ through a
264: formal
265: error analysis, it is necessary to know the errors in the individual input
266: parameters used to determine $\dot{M}$. These inexact parameters are the
267: white dwarf mass, the orbital inclination, the distance to the system and
268: the reddening. Each would produce errors in $\dot{M}$ and each parameter
269: has
270: its own uncertainties. Since each of these parameters affects either the
271: continuum slope or the flux level or both, they affect the accretion rate.
272: For example, Puebla et al. (2007) using similar disk models as ours but
273: with a more sophisticated four dimensional multiparameter optimization
274: fitting method, cite an error in E(B-V) of 0.02 to 0.05 producing errors
275: in $\dot{M}$ of 30 to 50\%. For a sense of the uncertainty of the
276: accretion
277: disk fitting method used in this paper, the reader is referred to Winter
278: and Sion (2003) who carried out a formal error analysis with error
279: contours for $\dot{M}$.
280:
281: \section{Conclusions}
282:
283: Our multi-component white dwarf photosphere plus accretion disk model
284: analysis of the archival IUE spectra of VY Scl covered four different
285: optical brightness states of the system. We find strong evidence that the
286: accretion rate has declined from $\dot{M}= 8 \times 10^{-9}$M$_{\sun}$
287: yr$^{-1}$ at the highest FUV flux level down to $\dot{M}= 1.9\times
288: 10^{-10}$M$_{\sun}$ yr$^{-1}$ during the lowest brightness state
289: spectroscopically recorded by the IUE. During this latter state of lowest
290: flux, the white dwarf contributes substantially to the FUV flux. This is
291: not unexpected for a nova-like low state since the underlying white dwarfs
292: in the nova-like systems TT Ari (Gaensicke et al. 2001), MV Lyr (Linnell
293: et al.2005), DW UMa (Knigge et al.2005) and V794 Aql (Godon et al.2007)
294: are also exposed. The temperature estimate of 45,000K we obtain for the
295: white dwarf in VY Scl must be regarded as preliminary. It is however
296: interesting that the temperature for the WD in VY Scl is within the range
297: of white dwarf temperatures measured for the other nova-like systems as
298: seen in Table 2.
299:
300: \begin{deluxetable}{lccl}
301: \tablecaption{Temperatures of WDs in VY Sculptoris-Type Nova-like Variables
302: }
303: \tablenum{2}
304: \tablehead{
305: \colhead{Star}&\colhead{P$_{orb}$}&\colhead{d(pc)}&\colhead{Te}
306: }
307: \startdata
308: V794 Aql& 0.2323 &250& > 47,000 K\\
309: TT Ari& 0.1375 &185 &40,000 K\\
310: MV Lyr &0.1329& 335& 47,000 K\\
311: DW UMa & 0.1375 & 830 & 46,000K\\
312: VY Scl &0.1662 &300 &45,000 K\\
313: \enddata\\
314: References: V794 Aql: Godon et al. (2007), ApJ, in press; TT Ari: Gaensicke et al. (2001); MV Lyr: Linnell et al. (2005); VY Scl: This paper.
315: \end{deluxetable}
316:
317: Since dwarf nova behavior is not seen in VY Scl-type and UX UMa-type
318: nova-like variables, it is widely held that the supression of the disk
319: instability mechanism requires that the accretion rate be higher than the
320: critical mass transfer rate for dwarf nova outbursts. This threshhold
321: value as a function of white dwarf mass and orbital period is discussed in
322: detail by Shafter, Wheeler and Cannizzo (1986). Referring to Figure 2 in
323: Shafter et al.(1986), the accretion rates we estimated for VY Scl in its
324: high states lie above the dwarf nova instability line for white dwarf
325: masses from 0.4 to 1.0 Msun.
326:
327: It is clear that higher quality FUV spectra are required to confirm the
328: tentative results herein for VY Scl. Moreover, given the small size of
329: the sample of exposed white dwarfs in nova-like systems, it is
330: particularly important to catch more nova-like systems in their low states
331: for both ground-based optical and space observations.
332:
333: \acknowledgments
334:
335: We thank an anonymous referee for helpful comments. This work was
336: supported by NSF grants AST05-07514, NNG04GE78G and by summer
337: undergraduate research support from the Delaware Space Grant Consortium.
338: Some or all of the data presented in this paper were obtained from the
339: Multi mission Archive at the Space Telescope Science Institute (MAST).
340: STScI is operated by the Association of Universities for Research in
341: Astronomy, Inc., under NASA contract NAS5-26555. Support for MAST for
342: non-HST data is provided by the NASA Office of Space Science via grant
343: NAG5-7584 and by other grants and contracts.
344:
345:
346: \begin{thebibliography}{}
347: \bibitem[]{}
348: Bruch, A., \& Engel, A. 1994, A\&A Suppl.Ser., 104, 79
349:
350: \bibitem[]{}
351: Diaz et al. 1996, ApJ, 459, 236
352:
353: \bibitem[]{}
354: Gaensicke et al. 2001, A\&A, 347, 178
355:
356: \bibitem[]{}
357: Godon et al. 2007, ApJ, in press
358:
359: \bibitem[]{}
360: Hoare, M., and Drew, J., 1993, MNRAS, 260, 647
361:
362: \bibitem[]{}
363: Hollander, A., Kraakman, H., van Paradijs, J.1993, A\&AS, 101, 87.
364:
365: \bibitem[]{}
366: Hubeny, I. 1988, Comput. Phys. Comun., 52, 103
367:
368: \bibitem[]{}
369: Hubeny, I., and Lanz, T. 1995, ApJ, 439, 875
370:
371: \bibitem[]{}
372: Hutchings and Cowley 1991 (1994), PASP, 96, 559
373:
374: \bibitem[]{}
375: Knigge, C. et al. 2000, ApJ, 539, L39
376:
377: \bibitem[]{}
378: LaDous, C. 1991, A\&A, 252, 100
379:
380: \bibitem[]{}
381: Linnell et al. 2005, ApJ, 624, 923
382:
383: \bibitem[]{}
384: Livio, M., \& Pringle, J.E. 1998, ApJ, 505, 339
385:
386: \bibitem[]{}
387: Massa, F., \& Fitzpatrick, E. 2000, ApJS, 126, 517
388:
389: bibitem[]{}
390: Martínez-Pais, I. G., Martín-Hern\'{a}ndez, N. L., Casares, J.,
391: Rodríguez-Gil, P.2000, ApJ, 538, 315
392:
393:
394: \bibitem[]{}
395: Patterson, J.1984, ApJS, 54, 443
396:
397: \bibitem[]{}
398: Puebla, R., Diaz, M., \& Hubeny, I.2007, AJ, 134, 192
399:
400:
401: \bibitem[]{}
402: Ritter, H., \& Kolb, U. 1998, A\&A, 129, 83
403:
404: \bibitem[]{}
405: Shafter, A., Wheeler, J.C., \& Cannizzo, J. 1986, ApJ, 305, 261
406:
407: \bibitem[]{}
408: Szkody 1987, ApJS, 63, 685
409:
410: \bibitem[]{}
411: Verbunt, F.1987, A\&A, 71, 339
412:
413: \bibitem[]{}
414: Wade, R.A. \& Hubeny, I. 1998, ApJ, 509, 350.
415:
416: \bibitem[]{}
417: Warner, B.1987, MNRAS, 227, 23
418:
419: \bibitem[]{}
420: Warner, B. 1995, Cataclysmic Variable Stars (Cambridge: Cambridge Univ. Press)
421:
422: \bibitem[]{}
423: Wu, K., Warner, B., Wickramasinghe, D.T. 1995, Proc.Ast.Soc.Austral.12:1, p.60
424:
425: \end{thebibliography}
426:
427: Figure Captions
428:
429:
430: \figcaption{The four IUE spectra corresponding to four different brightness states
431: from the peak of outburst (top spectrum, highest flux level) down to
432: a deep low state (bottom spectrum, lowest flux level).
433: Note the transition of the line spectrum as a function of brightness state
434: from the outburst spectrum where absorption lines dominate
435: down to the low state FUV spectrum when emission lines dominate the spectrum.}
436:
437: \figcaption{The best-fit accretion disk
438: synthetic fluxes to the spectrum SWP32594 of VY Scl during its high state. The accretion
439: disk corresponds to $\dot{M} = 8\times 10^{-9} M_{\sun}$/yr, $i = 41
440: ^{o}$, and M$_{wd} = 1.0 M_{\sun}$. The top solid curve is the
441: best-fitting combination, the dotted curve is the negligible contribution of the white dwarf
442: alone and the dashed curve is the accretion disk synthetic spectrum alone.
443: In this fit, the white dwarf the accretion disk contributes 98\% of the
444: far UV flux and the white dwarf 2\% of the flux.}
445:
446: \figcaption{The best-fit accretion disk
447: synthetic fluxes to the spectrum SWP29755 of VY Scl during its high state. The accretion
448: disk corresponds to $\dot{M} = 5\times 10^{-9} M_{\sun}$/yr, $i = 41
449: ^{o}$, and M$_{wd} = 1.0 M_{\sun}$. The top solid curve is the
450: best-fitting combination, the dotted curve is the contribution of the white dwarf
451: alone and the dashed curve is the accretion disk synthetic spectrum alone.
452: In this fit, the white dwarf the accretion disk contributes 98\% of the
453: far UV flux and the white dwarf 2\% of the flux.}
454:
455: \figcaption{The best-fit accretion disk
456: synthetic fluxes to the spectrum SWP of VY Scl during its high state. The accretion
457: disk corresponds to $\dot{M} = 1.6\times 10^{-9} M_{\sun}$/yr, $i = 41
458: ^{o}$, and M$_{wd} = 1.0 M_{\sun}$. The top solid curve is the
459: best-fitting combination, the dotted curve is the small contribution of the white dwarf
460: alone and the dashed curve is the accretion disk synthetic spectrum alone.
461: In this fit, the white dwarf the accretion disk contributes 92\% of the
462: far UV flux and the white dwarf 8\% of the flux.}
463:
464: \figcaption{The best-fit combination of white dwarf plus accretion disk
465: synthetic fluxes to the spectrum of VY Scl during its low state. The white
466: dwarf model has $T_{\rm eff} = 45,000$K, log $g =8.5$ and the accretion
467: disk corresponds to $\dot{M} = 1.9\times 10^{-10} M_{\sun}$/yr, $i = 41
468: ^{o}$, and M$_{wd} = 1.0 M_{\sun}$. The top solid curve is the
469: best-fitting combination, the dotted curve is the white dwarf spectrum
470: alone and the dashed curve is the accretion disk synthetic spectrum alone.
471: In this fit, the accretion disk contributes 66\% of the
472: far UV flux and the white dwarf 34\% of the flux.}
473:
474: %\clearpage
475:
476: \plotone{f1.ps}
477: %\clearpage
478:
479: \plotone{f2.ps}
480: %\clearpage
481:
482: \plotone{f3.ps}
483: %\clearpage
484:
485: \plotone{f4.ps}
486: %\clearpage
487:
488: \plotone{f5.ps}
489: \end{document}
490:
491:
492: