1: %\documentstyle[emulateapj]{article}
2: \documentclass[12pt,preprint]{aastex}
3:
4: %% manuscript produces a one-column, double-spaced document:
5:
6: %\documentclass[manuscript]{aastex}
7:
8: %% preprint2 produces a double-column, single-spaced document:
9:
10: %\documentclass[preprint2]{aastex}
11:
12:
13: %% \documentclass[preprint2,longabstract]{aastex}
14:
15:
16: \newcommand{\vdag}{(v)^\dagger}
17: \newcommand{\jcd}{Christensen-Dalsgaard}
18:
19: %% You can insert a short comment on the title page using the command below.
20:
21:
22: \shorttitle{Helioseismic constraints on solar abundances}
23: \shortauthors{Basu and Antia}
24:
25:
26: \begin{document}
27:
28:
29: \title{Constraining solar abundances using helioseismology}
30:
31:
32: \author{Sarbani Basu}
33: \affil{Astronomy Department, Yale University, P. O. Box 208101,
34: New Haven CT 06520-8101, U.S.A.}
35: \email{basu@astro.yale.edu}
36:
37: \and
38:
39: \author{H. M. Antia}
40: \affil{Tata Institute of Fundamental Research, Homi Bhabha Road,
41: Mumbai 400005, India}
42: \email{antia@tifr.res.in}
43:
44:
45: \begin{abstract}
46: Recent analyses of solar photospheric abundances suggest that the oxygen
47: abundance in the solar atmosphere needs to be revised downwards. In this
48: study we investigate the consequence of this revision on helioseismic
49: analyses of the depth of the solar convection zone and the helium abundance
50: in the solar envelope and find no significant effect.
51: We also find that the revised abundances along with
52: the current OPAL opacity tables are
53: not consistent with seismic data.
54: A significant upward revision of the opacity tables is required
55: to make solar models with lower oxygen abundance consistent with seismic
56: observations.
57:
58:
59: \end{abstract}
60:
61:
62: \keywords{Sun: abundances --- Sun: oscillations --- Sun: interior}
63:
64:
65: \section{Introduction}
66:
67: Recent analyses of spectroscopic data using modern atmospheric models
68: have suggested that the solar abundance of oxygen and other abundant elements
69: needs to be revised downwards (Allende Prieto, Lambert \& Asplund 2001, 2002;
70: Asplund et al.~2004). Asplund et al.~(2004) claim that the oxygen abundance should
71: be reduced by a factor of about 1.48 from the earlier estimates of
72: Grevesse \& Sauval (1998). The abundances of C, N, Ne
73: and Ar are also reduced.
74: The measured ratio of oxygen to hydrogen abundance, $[O/H]=8.66\pm0.05$,
75: is different from earlier estimates at approximately $3\sigma$ level.
76: As a result, the ratio (by mass) of heavy element to
77: hydrogen abundance, $Z/X$, reduces from 0.023 to 0.0171, which causes
78: the heavy element abundance
79: in the solar envelope to reduce from $Z=0.017$ to 0.0126.
80: This will cause the opacity
81: of solar material to decrease, in turn reducing the depth of the
82: convection zone (henceforth CZ) in solar models. Bahcall \& Pinsonneault (2004) have
83: constructed a solar model using these revised abundances to find that
84: the depth of the CZ is indeed reduced significantly as the CZ base
85: is found to be at $r_b=0.726R_\odot$, which is
86: inconsistent with the seismic estimate of $(0.713\pm0.001)R_\odot$
87: (Basu \& Antia 1997, henceforth BA97; Basu 1998). Asplund et al.~(2004) have argued
88: that their estimate refers to the abundances near the surface, while
89: because of diffusion the abundance could be much higher in the radiative
90: interior.
91: %
92: While this may be true, it should be
93: noted that since the CZ is mixed on a rather small time-scale,
94: the abundances at the surface and above the base of the CZ should
95: be the same. It is these abundances, coupled with the corresponding
96: opacity tables, that determine the depth of the CZ in a
97: solar model.
98: It may be possible to increase the hydrogen abundance
99: in the convection zone by increasing diffusion, thereby increasing the
100: convection zone depth, but in that case the helium content of
101: convection zone may reduce below the seismically estimated value.
102:
103: Since the position of the CZ base, $r_b$,
104: has been determined very accurately through seismic analysis
105: (\jcd, Gough \& Thompson 1991; BA97) it provides a
106: constraint on the heavy element abundance or the opacities. In fact,
107: BA97 have shown that the then accepted solar abundances
108: along with the OPAL opacities (Iglesias \& Rogers 1996) are consistent with
109: the helioseismic data. Thus it is of interest to test how
110: the reduction in abundance affects the conclusions.
111: Before we do this, we examine whether a decrease in the abundances affects
112: the seismic estimate of the CZ depth or the estimate of
113: the helium abundance, $Y$.
114:
115: \section{The technique}
116:
117: To estimate the depth of the CZ, we use the technique
118: described by BA97. This technique requires the construction
119: of solar envelope models with prescribed values of $r_b$
120: and abundances along with
121: the known input physics. Solar envelope models are more useful than
122: full solar models since envelope
123: models can be constructed with specific values of the CZ depth and $X$ (or $Y$)
124: and the CZ of the models do not depend on other uncertainties like opacities,
125: treatment of diffusion, etc., in the radiative interior.
126: Since both the hydrogen abundance, $X$ and
127: the depth of the CZ can be determined seismically, we
128: can construct an envelope model with the seismic estimates of
129: $r_b$ and $X$,
130: and then compare the sound speed and density profiles of the
131: model with the seismically obtained solar sound-speed and density
132: profiles to check for consistency.
133:
134: The relative abundances of heavy elements are modified by the
135: reduction in the abundances of oxygen and other elements
136: and we need to reconstruct the OPAL opacity tables for this new mixture of heavy elements
137: (with $[O/H]=8.66$) which we refer to as MIX1.
138: %
139: In this mixture the
140: logarithmic abundances of C, N, O, Ne and Ar are reduced by 0.17
141: as compared to those given by Grevesse \& Sauval (1998), while abundances
142: of other elements are unchanged.
143: It is possible that the abundances of
144: some of these elements change by a slightly different factor,
145: but that does not affect our conclusions.
146: We use the OPAL equation
147: of state (EOS) (Rogers \& Nayfonov 2002).
148: In principle, the EOS tables also need to be modified
149: in view of the change in mixture of heavy elements. We have not done
150: that since the EOS is
151: not particularly sensitive to the detailed breakup of heavy element abundance.
152: Opacity, on the other hand, needs recalculation since in regions
153: of fully ionized H and He, the heavy elements are the predominant
154: contributors to opacity, though their contribution to the EOS is small.
155: Below the CZ we use the $X$
156: profile determined from seismic data using the method of Antia \&
157: Chitre (1998) with the new heavy element abundances. We construct
158: solar envelope models with CZ base position,
159: $r_b$, at $0.709R_\odot$, $0.711R_\odot$, $0.713R_\odot$,
160: $0.715R_\odot$ and $0.717R_\odot$ to serve as calibration models
161: for determining $r_b$. All these models have $Z/X=0.0171$ and
162: $X=0.74$ which is close to
163: the estimated value of hydrogen abundance. To determine
164: the helium abundance we use calibration models with $r_b=0.713R_\odot$
165: and $X=0.70,0.72,0.74,0.76,0.78$ and use the technique described
166: by Basu \& Antia (1995).
167:
168: We also examine the consistency of similar models that have smaller reduction
169: (by 0.03) in the logarithmic abundances
170: ($[O/H]=8.80$ for these models).
171: We refer to this mixture of heavy element abundances as MIX2.
172: The changes in abundances result in $Z/X=0.0218$ for these models.
173: We also construct a few full solar models
174: to check how they compare with seismic data.
175:
176: For this work we use the observed frequencies obtained by the
177: Global Oscillations Network Group (GONG) (Hill et al.~1996) between
178: months 4 to 14, as well as frequencies obtained by the Michelson Doppler Imager (MDI)
179: from the first 360 days of its observation (Schou et al.~1998).
180:
181: \section{Results}
182:
183: Using the technique described by Basu \& Antia (1995) we first estimate
184: the helium abundance in the solar envelope using models with MIX1 abundance.
185: We find $X=0.7392\pm 0.0034$ using GONG data and $X=0.7385\pm 0.0034$ using
186: MDI data or $Y=0.2482$ and 0.2489 respectively. The error-bars
187: include systematic errors including those caused by uncertainties in the
188: EOS (see Basu \& Antia 1995; Basu 1988 for details of uncertainties).
189: These results are consistent with earlier estimates that used models with completely
190: different heavy element abundances (D\"appen et al.~1991; Basu \& Antia
191: 1995; Richard et al.~1996; Basu 1998).
192: We repeat the same exercise
193: using MIX2 abundances to find $X=0.7394$ (GONG) and 0.7386 (MDI).
194: Thus it appears that the inferred hydrogen
195: abundance is insensitive to heavy element abundance of the calibration models, while the
196: helium abundance changes slightly due to change in $Z$.
197:
198: To estimate the depth of the CZ
199: we use the technique described by BA97 that requires
200: the decomposition of the frequency differences between the calibration
201: models and the Sun in terms of two functions $H_1(w)$ and $H_2(\nu)$ that
202: depend on the sound-speed differences and surface structure respectively
203: (\jcd, Gough \& Thompson 1989). Here, $\nu$ is the frequency
204: and $w=\nu/(\ell+1/2)$, where $\ell$ is the degree of the mode.
205: The function $H_1(w)$ can be used to
206: determine the position of the CZ base.
207: Figure~1(a) shows the function $H_1(w)$ for
208: calibration models using MIX1 abundances. From this we can estimate
209: the position of the base of the CZ to be
210: $r_b=(0.7133\pm 0.0005)R_\odot$ (GONG) and $(0.7132\pm0.0005)R_\odot$ (MDI).
211: The error bars here include systematic errors as determined by Basu (1998).
212: These values are in good agreement with earlier
213: estimates and are evidently not affected by the revision of abundances.
214: The function $H_1(w)$ is a measure of the sound speed
215: differences between the models and the Sun, and Fig.~1(a) shows
216: that although the sound speed differences are small in the CZ,
217: there are significant systematic
218: differences below the CZ.
219: There is a small difference in
220: the sound speed even within the CZ.
221: This figure can be compared with Fig.~10 of BA97
222: to compare $H_1(w)$ for higher $Z$ models.
223: The difference is most likely due to reduction in opacities because
224: of reduced $Z$.
225:
226: Although the seismic estimate of $X$ or CZ depth
227: is not affected by the revision in abundances,
228: solar models constructed using the current abundances may not have
229: the correct depth of the CZ or the correct density profile
230: in the \hbox{CZ}. To check for this we
231: compare the density profiles in these envelope models with that
232: obtained through seismic inversions, the results are shown in Fig.~1(b).
233: It is clear that the differences are very large.
234: The estimated error in density inversion
235: in most of the CZ is about 0.005.
236: In addition to this there could be systematic errors due to uncertainties
237: in $X$, $r_b$ and the EOS. An error of 0.0034 in $X$ causes
238: $\delta\rho/\rho\approx 0.012$, an error of $0.0005R_\odot$ in
239: $r_b$ yields $\delta\rho/\rho\approx 0.005$, while uncertainties in
240: the EOS also give $\delta\rho/\rho\approx 0.005$.
241: Assuming that these errors are uncorrelated, we get a total error of
242: 0.015 in $\delta\rho/\rho$.
243: Other uncertainties like the treatment of convection, turbulence and
244: the atmosphere etc., only affect the outermost layers
245: and hence are not included.
246: We have not included the effect of uncertainties in $Z$ and opacities as
247: we use seismic data to constrain these quantities by matching the density
248: profile (BA97) in a model with correct $X$ and $r_b$.
249: To get the correct density profile we need to increase the opacity
250: near the CZ base. We find that if the opacities
251: are increased by 19\%, then a MIX1 model with $r_b=0.7133$ and $X=0.739$
252: has the correct density
253: profile (see Fig. 1b). Alternately, as in BA97 we can
254: estimate the value of $Z/X$ for the same relative abundances as MIX1
255: that will give the correct density
256: assuming that the OPAL opacities are correct
257: (i.e., we increase opacity by
258: increasing $Z$). We find that we need $Z/X=0.0214$ to get the
259: correct density profile (see Fig.~1b).
260: We can also determine the range of
261: opacities that give the density profile within acceptable limits
262: for each value of $Z/X$ and Fig.~3 shows the result.
263: Only
264: the $Z/X$ and opacity values in the shaded region are consistent
265: with seismic constraints. The point with error bars shows the
266: measured values assuming an uncertainty of 5\% in the opacity at
267: the CZ base.
268:
269: We have repeated the analysis using models with MIX2 composition.
270: The results are shown in Fig.~2.
271: The position of the base of the CZ estimated using
272: GONG and MDI data is $0.7135R_\odot$
273: and $0.7134R_\odot$ respectively. Thus
274: once again the $r_b$ estimate is not significantly affected
275: by $Z$. From Fig.~2(b), it is clear that now we have better
276: agreement with solar density in the \hbox{CZ}.
277: The function $H_1(w)$ is also essentially flat in the \hbox{CZ}.
278: However, the model with correct CZ depth
279: does not have the correct density.
280: In this case increasing opacity by 3.5\% or increasing $Z/X$ to 0.0228
281: makes it possible to get the correct density profile (see
282: Fig. 2b).
283: This value of $Z/X$ is close to
284: that found by Grevesse \& Sauval (1998).
285: The allowed region in the $Z/X$--opacity plane for MIX2 mixture is
286: also shown in Fig.~3 by the vertically shaded region.
287: The allowed region is just above that for MIX1 mixture.
288:
289: The above results were obtained with envelope models and it could be argued that
290: full solar models obtained from evolutionary calculations may give better results.
291: To check this we construct two
292: full solar models, FULL1 and FULL2, with $Z/X$ and relative heavy element
293: abundance as in MIX1 and MIX2 respectively.
294: In these models, as with other standard solar models,
295: the surface $X$ and
296: the mixing-length parameter are adjusted to match the present day
297: solar radius and luminosity at an age of 4.6 Gyr.
298: These models were constructed with
299: the Yale Rotating
300: Evolution Code in its non-rotating configuration (Guenther et al.~1992) and
301: includes diffusion of helium and heavy elements as per
302: Thoul, Bahcall \& Loeb (1994).
303: The sound-speed and density differences between these models and the Sun
304: are shown in Fig.~4.
305: The sound-speed differences are dominated
306: by differences in the \hbox{CZ} depth. The
307: density differences are also affected by the differences in composition.
308: We can see that these models are worse than the envelope models.
309: Models FULL1 and FULL2 have
310: $r_b=0.7320R_\odot$ and $0.7217R_\odot$, and surface $X$ of 0.7505 and 0.7422 respectively.
311: We can see that
312: the model with the higher $Z/X$ is closer to the Sun, and we can argue that the
313: agreement improves with even higher oxygen abundance and higher overall $Z/X$,
314: which is in fact the case
315: (see, for example, model STD of
316: Basu et al.~2000, or model 20 of Winnick et al.~2002). Unlike models with higher
317: $Z/X$, these models do
318: not reproduce the seismically determined value of $X$ in the solar envelope either.
319:
320: Since Asplund et al.~(2004) have suggested that increased diffusion
321: may yield seismically consistent solar models with the revised abundances,
322: we have constructed two models FULL1M and FULL2M with
323: diffusion coefficients increased by a factor of 1.65 and these models
324: are also shown in Fig.~4.
325: The positions of
326: the CZ bases are $0.7233R_\odot$ and $0.7138R_\odot$ for FULL1M and
327: FULL2M respectively,
328: and the envelope $X$ is 0.7626 and 0.7519 respectively. Thus FULL2M has the
329: observed value of $r_b$, but the value of $X$ in the envelope is more
330: than $3\sigma$ beyond the seismically measured value.
331: It should be noted that we have increased the diffusion coefficients without any
332: physical justification.
333: We have also constructed static, full models of the Sun using
334: the seismically determined abundance profile (Antia \& Chitre 1998)
335: and these
336: are labeled INV1 and INV2 for MIX1 and MIX2 abundances.
337: These models have surface $X$ of 0.7680 and 0.7447 respectively and are also shown in Fig.~4.
338: Despite having the correct CZ depth, the $X$ values in these models
339: are respectively, about $9\sigma$ and $2\sigma$ away from the seismically
340: inferred value.
341: Thus with the new abundances, we could not get the correct values of both $X$ and
342: CZ depth in a full solar model.
343: Models FULL1M, FULL2M, INV1, and INV2 are not standard solar
344: models since the $X$ profile has been calculated using non-standard
345: procedures.
346:
347:
348:
349: \section{Conclusions}
350:
351: We have investigated the effects that the revision of the abundance of oxygen
352: and related elements in the solar atmosphere may have on seismic estimates
353: of the solar helium abundance and the depth of the \hbox{CZ}. We find
354: neither of these estimates to be sensitive to variations in $Z$.
355: We find that solar envelope models that have reduced abundances of oxygen
356: and related elements do not have the correct density profile in the CZ despite
357: having the seismically determined CZ depth and surface $X$.
358: The density difference is about 10\%, which is more than 6 times
359: the estimated uncertainties in density.
360: In order to get a seismically consistent solar model it is necessary
361: to increase either the abundances of heavy elements or the
362: computed opacities for a given abundance ratio.
363: Even for a much smaller reduction in the oxygen abundance ($[O/H]=8.80$, the MIX2 abundances)
364: it turns out that $Z/X$ needs to be increased
365: from 0.0218 to 0.0228 to match the sound speed and density
366: in the solar \hbox{CZ}.
367: The region in the $Z/X$--opacity plane for both mixtures that is consistent with
368: seismic constraints
369: is shown in Fig.~3.
370: The allowed region is not too sensitive
371: to variations in the oxygen abundance, but to match the
372: seismic constraints either the opacity or the heavy element abundances,
373: or both need to be increased. From seismic constraints it
374: is not possible to decide between these possibilities.
375:
376: We find that the recent estimate of the value of
377: the oxygen abundance by Asplund et al.~(2004) along with the OPAL
378: opacities are not consistent with
379: seismic data. In fact, if current opacity tables are accepted,
380: no significant reduction in the oxygen abundance
381: from the values of Grevesse \& Sauval (1998) is favored
382: by helioseismology.
383: Models constructed with the new abundances either have incorrect
384: CZ depth, or incorrect $X$, or incorrect density profile in the \hbox{CZ}.
385: Increasing the diffusion of heavy elements below the CZ base
386: to compensate for a reduction in the atmospheric abundance does not work,
387: the models fail to satisfy the $X$ constraint in the
388: solar envelope, even though they may have the correct CZ depth.
389: If the new values of abundances are confirmed, then
390: opacity sources in the Sun will need to be re-investigated. The intrinsic opacities
391: will need to be increased to counteract the decrease in opacity due to
392: the reduction in heavy element abundances.
393: It could be argued that the abundances
394: in the atmosphere where the spectral lines are formed is different
395: from those in the CZ, due to some fractionation. This could resolve the problem
396: we face now. However we will be faced with the more fundamental problem
397: of having no way of determining the CZ abundances of the Sun.
398:
399:
400: \acknowledgments
401:
402: This work utilizes data obtained by the Global Oscillation
403: Network Group project, managed by the National Solar Observatory
404: which is
405: operated by AURA, Inc. under a cooperative agreement with the
406: NSF.
407: This work also utilizes data from the Solar Oscillations
408: Investigation/ Michelson Doppler Imager (SOI/MDI) on the Solar
409: and Heliospheric Observatory (SOHO). SOHO is a project of
410: international cooperation between ESA and NASA.
411: MDI is supported by NASA grants NAG5-8878 and NAG5-10483
412: to Stanford University.
413: We thank the OPAL group for the opacity tables for the
414: different heavy element mixtures.
415: This work was supported by
416: NSF grant ATM 0206130 to SB.
417:
418:
419: \begin{thebibliography}{}
420:
421: \bibitem[Allende Prieto et al.(2001)]{ap01}
422: Allende Prieto, C., Lambert, D. L., \& Asplund, M. 2001, ApJ, 556, L63
423:
424: \bibitem[Allende Prieto et al.(2002)]{ap02}
425: Allende Prieto, C., Lambert, D. L., \& Asplund, M. 2002, ApJ, 573, L137
426:
427: \bibitem[Antia \& Chitre(1998)]{ac98} Antia, H. M., \& Chitre, S. M.
428: 1998, A\&A, 339, 239
429:
430: \bibitem[Asplund et al.(2004)]{asp04} Asplund, M., Grevesse, N.,
431: Sauval, A. J., Allende Prieto, C., \& Kiselman, D. 2004, A\&A, (in press)
432: astro-ph/0312290
433:
434: \bibitem[Bahcall \& Pinsonneault(2004)]{bp04} Bahcall, J. N., \& Pinsonneault, M. H.
435: 2004, Phys.\ Rev.\ Lett., (in press) astro-ph/0402114
436:
437: \bibitem[Basu(1998)]{b98} Basu, S. 1998, \mnras, 298, 719
438:
439: \bibitem[Basu \& Antia(1995)]{ba95} Basu, S., \& Antia, H. M. 1995,
440: \mnras, 276, 1402
441:
442: \bibitem[Basu \& Antia(1997)]{ba97} Basu, S., \& Antia, H. M. 1997,
443: \mnras, 287, 189 (BA97)
444:
445: \bibitem[Basu, Pinsonneault \& Bahcall(2000)]{bbp00} Basu, S.,
446: Pinsonneault, M. H., \& Bahcall, J. N. 2000, ApJ, 529, 1084
447:
448:
449: \bibitem[Christensen-Dalsgaard et al.(1989)]{jcd89}
450: Christensen-Dalsgaard, J., Gough, D. O., \& Thompson, M. J.
451: 1989, \mnras, 238, 481
452:
453: \bibitem[Christensen-Dalsgaard et al.(1991)]{jcd91}
454: Christensen-Dalsgaard, J., Gough, D. O., \& Thompson, M. J.
455: 1991, \apj, {378}, 413
456:
457: \bibitem[D\"appen et al.(1991)]{dap91}
458: D\"appen, W., Gough, D. O., Kosovichev, A. G., \& Thompson, M. J. 1991,
459: in {Challenges to theories of the structure of moderate-mass stars},
460: {Lecture Notes in Physics}, vol.~{388}, p. 111,
461: eds D. O. Gough, \& J. Toomre, Springer, Heidelberg
462:
463: \bibitem[Guenther et al.(1992)]{gue92}
464: Guenther, D. B., Demarque, P., Kim, Y.-C., \& Pinsonneault, M. H. 1992,
465: ApJ, 387, 372
466:
467: \bibitem[Grevesse \& Sauval(1998)]{gre98} Grevesse, N., \& Sauval, A. J.
468: 1998, in Solar composition and its evolution --- from core to corona,
469: eds., C. Fr\"ohlich, M. C. E. Huber, S. K. Solanki, \&
470: R. von Steiger, Kluwer, Dordrecht, p. 161
471:
472: \bibitem[Hill et al(1996)]{hill96} Hill, F. et al.~1996, Science, 272, 1292
473:
474: \bibitem[Iglesias \& Rogers(1996)]{opal96}
475: Iglesias, C. A., \& Rogers, F. J. 1996, \apj, {464}, 943
476:
477: \bibitem[Richard et al.(1996)]{ric96}
478: Richard, O., Vauclair, S., Charbonnel, C., \& Dziembowski, W. A. 1996,
479: A\&A, {312}, 1000
480:
481: \bibitem[Rogers \& Nayfonov(2002)]{opal02} Rogers, F. J., \& Nayfonov, A.
482: 2002, \apj, 576, 1064
483:
484: \bibitem[1998]{sch98} Schou, J., Christensen-Dalsgaard, J., Howe, R.,
485: Larsen, R. M., Thompson, M. J., \& Toomre, J. 1998,
486: in Proc. SOHO6/GONG98 workshop,
487: Structure and Dynamics of the Interior of the
488: Sun and Sun-like Stars, ed. S. G. Korzennik, \& A. Wilson
489: (Noordwijk: ESA), ESA SP-418, vol 2, 845
490:
491: \bibitem[Thoul et al.(1994)Thoul, Bahcall \& Loeb]{tbl94}
492: Thoul, A. A., Bahcall, J. N., \& Loeb, A. 1994, ApJ, 421, 828
493:
494: \bibitem[Winnick et al.(2002)]{wi02}
495: Winnick, R. A., Demarque, P., Basu, S., \& Guenther, D. B. 2002, ApJ,
496: 576, 1075
497:
498:
499: \end{thebibliography}
500:
501: \clearpage
502:
503:
504: \begin{figure}
505: %\epsscale{.80}
506: \plotone{f1.eps}
507: \caption{Panel (a) --- The function $H_1(w)$ obtained from frequency differences
508: between envelope models with MIX1 composition and the MDI
509: frequencies plotted as a function of lower turning point of the acoustic
510: modes. Panel (b) --- The relative difference in density between the solar
511: envelope models with MIX1 composition and the Sun. The different line styles
512: are the same as in Panel(a).
513: The panel also shows the density differences for two other models,
514: one with $Z/X=0.0214$ and the other with opacities increased by 19\%.
515: \label{fig1}}
516: \end{figure}
517:
518: \clearpage
519: \begin{figure}
520: %\epsscale{.80}
521: \plotone{f2.eps}
522: \caption{
523: The same as Fig.~1, but for models with MIX2 composition.
524: The lower panel also shows the density differences for two other models,
525: one with $Z/X=0.0228$ and another with opacities increased by 3.5\%.
526: \label{fig2}}
527: \end{figure}
528:
529: \clearpage
530: \begin{figure}
531: %\epsscale{.80}
532: \plotone{f3.eps}
533: \caption{The shaded area shows the allowed region in $Z/X$--opacity
534: plane that is consistent with seismic constraints. The horizontal
535: and vertical shadings, shows the region for MIX1 and MIX2 mixtures
536: respectively. The point with error bars shows current values of
537: opacity and abundances.
538: \label{fig3}}
539: \end{figure}
540:
541:
542: \clearpage
543: \begin{figure}
544: %\epsscale{.80}
545: \plotone{f4.eps}
546: \caption{ The relative sound-speed and density differences between different
547: full solar models and the Sun. The points represent standard solar models, while the
548: lines show non-standard ones. Models FULL1, FULL1M and INV1 have
549: MIX1 composition and FULL2, FULL2M and INV2 have MIX2 composition.
550: \label{fig4}}
551: \end{figure}
552:
553:
554:
555:
556: \end{document}
557:
558: