1: \documentclass[12pt,preprint]{aastex}
2:
3: \def\vmi{\hbox{\it V--I\/}}
4: \def\vmr{\hbox{\it V--R\/}}
5: \def\vmk{\hbox{\it V--K\/}}
6: \def\bmv{\hbox{\it B--V\/}}
7: \def\mmm{\hbox{\it m--M\/}}
8:
9: \newcommand\msol{{\cal M_{\odot}}}
10: \newcommand\mbol{M_{\rm bol}}
11: \newcommand\mstar{{\cal M}_*}
12: \newcommand\mtr{{{\cal M}_{\rm tr}}}
13: \newcommand\teff{{T_{\rm eff}}}
14: \newcommand\amlt{{\alpha_{\rm MLT}}}
15: \newcommand\delv{\Delta V^{\rm HB}_{\rm TO}}
16: \newcommand\delc{\Delta (\bmv)_{\rm TO,RGB}}
17: \newcommand\lta{\mathrel{\hbox{\raise 0.6 ex \hbox{$<$}\kern
18: -1.8 ex\lower .5 ex\hbox{$\sim$}}}}
19: \newcommand\gta{\mathrel{\hbox{\raise 0.6 ex \hbox{$>$}\kern
20: -1.7 ex\lower .5 ex\hbox{$\sim$}}}}
21:
22: %\received{}
23: %\accepted{}
24: %\journalid{}{}
25: %\articleid{}{}
26:
27: %\slugcomment{The Astrophysical Journal, submitted}
28:
29: \shortauthors{VandenBerg et al.}
30: \shorttitle{Constraint on $Z_\odot$ from M$\,$67}
31:
32: \begin{document}
33:
34: \title{A Constraint on $Z_\odot$ From Fits of Isochrones to the Color-Magnitude
35: Diagram of M$\,$67}
36:
37: \author{Don A.~VandenBerg\altaffilmark{1}, Bengt Gustafsson\altaffilmark{2},
38: Bengt Edvardsson\altaffilmark{2}, Kjell Eriksson\altaffilmark{2}, and \\
39: Jason Ferguson\altaffilmark{3}}
40:
41: \altaffiltext{1}{Department of Physics \& Astronomy, University of Victoria,
42: P.O.~Box 3055, Victoria, B.C., V8W~3P6, Canada (email: vandenbe@uvic.ca)}
43: \altaffiltext{2}{Uppsala Astronomical Observatory, Box 515, SE-751 20 Uppsala,
44: Sweden \\ (email: firstname.lastname@astro.uu.se)}
45: \altaffiltext{3}{Department of Physics, Wichita State University, Wichita,
46: Kansas, 67260-0032, U.S.A. \\ (email: jason.ferguson@wichita.edu)}
47:
48: \begin{abstract}
49: The mass at which a transition is made between stars that have radiative {\it
50: or} convective cores throughout the core H-burning phase is a fairly sensitive
51: function of $Z$ (particularly, the CNO abundances). As a consequence, the
52: $\sim 4$ Gyr, open cluster M$\,$67 provides a constraint on $Z_\odot$ (and
53: the solar heavy-element mixture) because (i) high-resolution spectroscopy
54: indicates that this system has virtually the same metal abundances as the Sun,
55: and (ii) its turnoff stars have masses just above the lower limit for sustained
56: core convection on the main sequence. In this study, evolutionary tracks and
57: isochrones using the latest MARCS model atmospheres as boundary conditions have
58: been computed for 0.6--$1.4 \msol$ on the assumption of a metals mix (implying
59: $Z_\odot \approx 0.0125$) based on the solar abundances derived by M.~Asplund
60: and collaborators using 3-D model atmospheres. These calculations do not
61: predict a turnoff gap where one is observed in M$\,$67. No such difficulty is
62: found if the analysis uses isochrones for $Z_\odot = 0.0165$, assuming the
63: Grevesse \& Sauval (1998) mix of heavy elements. Our findings, like the
64: inferences from helioseismology, indicate a problem with the Asplund et
65: al.~abundances. However, it is possible that low-$Z$ models with diffusive
66: processes taken into account will be less problematic.
67: \end{abstract}
68:
69: \keywords{Hertzsprung-Russell diagram --- open clusters: individual (M$\,$67)
70: --- stars: atmospheres --- stars: evolution --- Sun: abundances}
71:
72: \section{Introduction}
73: \label{sec:intro}
74: The metallicity of the Sun (both the mix of heavy elements, and the total
75: mass-fraction abundance, $Z$) is presently a subject of considerable
76: controversy. Until a few years ago, the generally accepted value of $Z_\odot$
77: was $\approx 0.017$--0.02, and there was good consistency between the
78: predictions of Standard Solar Models (SSMs) for metallicities within this range
79: (see, e.g., \citealt{trm98}, \citealt{bpb01}) and the measured neutrino flux as
80: well as helioseismological data. However, significantly reduced abundances for
81: several of the most abundant heavy elements (including C, N, O, and Ne),
82: resulting in $Z_\odot \approx 0.0125$, have been derived by M.~Asplund and
83: collaborators from their analyses of the photospheric spectrum using 3-D,
84: non-LTE model atmospheres (see \citealt{ags06}; and references therein).
85: Whereas high-$Z$ solar models are able to reproduce the inferred radial
86: dependence of the square of the sound speed (down to $\sim 0.1 R_\odot$) from
87: solar oscillations to $\lta 0.3$\% (see \citealt{cd02}), those computed for the
88: Asplund et al.~metallicity are unable to do so to better than $\sim 3$\% (e.g.,
89: (\citealt{tcp04}).
90:
91: It seems unlikely that this difficulty is due to problems with current opacities
92: (\citealt{ab05}) or the presently accepted rates of diffusive processes
93: (\citealt{gu05}). As noted by \citet{mmt06}, any substantial increases to the
94: diffusion coefficients that are made in order to recover good agreement with the
95: inferred sound speed profile result in predictions for the surface helium
96: content of the Sun that are below the values obtained by the inversion of solar
97: oscillations. In fact, turbulent mixing processes appear to operate in stellar
98: envelopes to {\it reduce} the effects of diffusion on the surface abundances of
99: $\sim 1 \msol$ main-sequence (MS) stars (see \citealt{kgr06}; \citealt{rmr02}).
100:
101: While the low solar $Z$ derived by Asplund et al.~may be erroneous, the 3-D
102: model atmospheres used in their studies have had unprecedented success in
103: modelling the properties of the solar atmosphere (see \citealt{asp05}), and the
104: predicted spectral line profiles based on these atmospheres provide superb
105: matches to the observed profiles (\citealt{ant00}). In addition, the reduced
106: abundances of the CNO elements are in excellent agreement with the values
107: measured in the interstellar medium (\citealt{tcp04}) and in solar neighborhood
108: B stars (see \citealt{ags05}; and references therein). Thus, their findings
109: are not easily dismissed.
110:
111: There is another, albeit indirect, way of constraining the metallicity of the
112: Sun that has not yet been investigated. The large reduction in the CNO and Ne
113: abundances found by Asplund et al.~(relative to previous estimates) affects the
114: mass at which a transition is made between stars that have radiative cores
115: during the MS phase and those possessing convective cores when their central
116: hydrogen fuel is exhausted. Consequently, a cluster like M$\,$67, which has
117: [m/H] $\approx 0.0$ and turnoff stars with masses close to the transition mass,
118: $\mtr$, provides a probe of both stellar physics {\it and} the chemical
119: composition of the material out of which the stars formed. As shown by
120: \citet{vbd06}, $\mtr$ is predicted to increase with decreasing metal abundance,
121: and to be a fairly sensitive function of $Z$ (see their Fig.~2). Whether or
122: not an old open cluster is expected to have a gap in its distribution of
123: near-turnoff stars on the color-magnitude diagram (CMD) will therefore depend
124: on its metallicity. (A gap is the observational manifestation of the rapid
125: contraction phase that accompanies the exhaustion of hydrogen in stars that have
126: convective cores at the end of their MS lifetimes.)
127:
128: The M$\,$67 CMD (see \citealt{mmj93}; \citealt{sa04}) contains such a gap,
129: and stellar models for $Z \gta 0.017$ have been quite successful in reproducing
130: these data. \citet{vs04} have shown that a 4.0 Gyr isochrone for $Z = 0.0173$,
131: assuming the \citet{gn93} metal abundances, provides a satisfactory match to the
132: cluster CMD, if a small amount of convective core overshooting is assumed (also
133: see VBD06). A nearly identical fit was obtained by \citet{mrr04} using
134: diffusive isochrones for 3.7 Gyr and $Z_{\rm initial} = 0.02$, without having to
135: postulate any core overshooting. These calculations predict that the cluster
136: MS stars have $\sim 6$--12\% surface underabundances of the metals, depending
137: on their temperatures --- compared with a $\lta 8$\% reduction in the case of
138: the Sun (\citealt{trm98}), which is much too small to simulate the difference
139: between the Asplund et al.~solar composition and that reported by, e.g.,
140: Grevesse \& Noels.
141:
142: Given the apparent success of relatively metal-rich models in explaining the
143: CMD of M$\,$67, which has very close to the solar metallicity, is it possible
144: for $Z=0.0125$ isochrones to provide a comparable fit? This is the question
145: that has motivated the present investigation.
146: %The models that have been
147: %computed to investigate this consistency test are described in the next section,
148: %while the application of these models to the M$\,$67 CMD is presented in
149: %\S~\ref{sec:m67}. A brief summary of the main results of this study is given
150: %in \S~\ref{sec:conclude}.
151:
152: \section{Stellar Models for Solar Abundances}
153: \label{sec:models}
154: The evolutionary tracks needed for this investigation have been computed using
155: the Victoria stellar evolution code (see VBD06, and references therein). Model
156: atmospheres produced by the latest version of the MARCS code (e.g.,
157: \citealt{gee03}) have been used as boundary conditions. [Differences in the
158: treatment of the surface layers do not alter the tracks for the MS and early
159: subgiant branch (SGB) phases of solar metallicity stars in any significant way
160: if the value of the mixing-length parameter, $\amlt$, is chosen so that the
161: solar constraint is satisfied (see, e.g., Fig.~3 in the study by
162: \citealt{vee07}). The assumed value of $\amlt$ does affect the predicted
163: location of the red-giant branch (RGB), but that is not a concern for the
164: present work.]
165:
166: The standard abundance distribution reported by \citet{gs98}, as determined from
167: analyses of solar photospheric spectra and meteoritic data using classical 1-D
168: hydrostatic models, was the last such compilation prior to the use of 3-D
169: hydrodynamical model atmospheres by Asplund et al.~(2006) to determine the solar
170: composition. The GS98 abundances for the 19 metals normally considered in
171: stellar evolutionary computations (because they are the only heavy elements that
172: are taken into account in the calculation of OPAL opacities for stellar interior
173: conditions; see \citealt{ir96}) are listed in the second column of
174: Table~\ref{tab:tab1}. The third column contains the the abundance distribution
175: that is obtained when the revised values derived by Asplund et al.~for several
176: of the heavy elements are adopted: note that the C and N abundances in
177: Table~\ref{tab:tab1} are 0.02 dex higher than their published 2006 estimates.
178:
179: OPAL opacities were computed for both heavy-element mixtures
180: using the Livermore Laboratory website
181: facility (see http://www-phys.llnl.gov/Research/OPAL). Because these
182: calculations do not include the contributions from molecular sources or grains,
183: complementary opacity data for $T \le 10^4$ K were computed for the same
184: abundances using the code described by \citet{faa05}. The tabulated values of
185: $\log N_{\rm He}$ were obtained from SSMs that were constructed for the two
186: sets of abundances: the resultant values of $Y$ and $Z$ are listed in the last
187: two rows of Table~\ref{tab:tab1}. To satisfy the solar constraint, the
188: $1.0 \msol$ models for the GS98 and Asplund et al.~abundances required
189: $\amlt = 1.84$ and 1.80, respectively (when MARCS model atmospheres are used as
190: boundary conditions; see \citealt{vee07}).
191:
192: The value of $\mtr$ is a function of $Z$ (especially the CNO abundances; see
193: below). The transition occurs when the CNO-cycle becomes an important source of
194: nuclear energy production, and since both the decrease in the {\it abundances}
195: of the CNO elements and the concomitant reduction in opacity will serve to
196: reduce the rate of the CNO-cycle, the minimum mass for sustained core convection
197: on the MS is higher in stars of lower $Z$. By trial and error, $\mtr$ was found
198: to be $1.195 \msol$ if $Z = 0.0125$ (Asplund et al.~metals mix), whereas $1.155
199: \msol$ is obtained if $Z = 0.0165$ (GS98 metal abundances). [The transition
200: mass also depends on the adopted helium content; e.g., $\mtr = 1.181 \msol$ if
201: $Z=0.0125$ and the value of $Y$ required by an SSM for $Z=0.0165$ is assumed
202: (i.e., $Y=0.2676$; see Table~\ref{tab:tab1}).]
203:
204: Indeed, the character of the tracks changes abruptly when core convection
205: persists until H exhaustion. This is shown in Figure~\ref{fig:vfig1}, which
206: plots the evolutionary tracks for 1.0--$1.4 \msol$ that were computed for the
207: two values of $Z_\odot$; including, in particular, sequences for $\mtr$ and
208: $(\mtr + 0.001) \msol$. Each of the higher mass tracks possesses a blueward
209: hook, which arises when a star contracts following the depletion of hydrogen in
210: a convective core in order to ignite H-burning in a shell around the He core.
211: This morphology is not seen in the lower mass tracks because the central regions
212: are radiative when hydrogen is exhausted and there is a smooth transition to
213: H shell burning.
214:
215: A difference of $0.04 \msol$ in $\mtr$ is surprisingly large given that VBD06
216: found an increase of only $0.009 \msol$ when $Z$ was decreased from 0.0173 to
217: 0.0125. However, the latter calculations assumed the same relative abundances
218: of the metals (those given by \citealt{gn93}), whereas the tracks plotted in
219: Fig.~\ref{fig:vfig1} assume two very different heavy-element mixtures. Indeed,
220: a large reduction in the CNO abundances is mainly responsible for the downward
221: revision of the solar value of $Z$ from 0.0165 (GS98) to 0.0125 (Asplund et
222: al.) --- which demonstrates that $\mtr$ is a strong function of the abundances
223: of the CNO elements.
224:
225: An increase in $\mtr$ from 1.155 to $1.195 \msol$ will be accompanied by a
226: decrease in the maximum age at which a gap near the turnoff is expected in an
227: observed CMD. Fig.~\ref{fig:vfig1} shows that the biggest differences between
228: the two sets of models occur at masses of $\approx 1.2 \msol$. Because $\mtr$
229: is lower in the case of the GS98 computations, and because the amount of
230: overshooting in stars of mass ${\cal M_*} > \mtr$ is assumed to vary directly
231: with ${\cal M_*} - \mtr$ (see VBD06), the tracks represented by the dashed
232: curves for masses of 1.2--$1.4 \msol$ have larger convective cores during MS
233: evolution, and cooler blueward hooks at H exhaustion, than those plotted as
234: solid curves. Such differences will impact the isochrones (computed in this
235: study using the interpolation software described by VBD06) and the degree to
236: which they are able to reproduce the M$\,$67 CMD.
237:
238: \section{Application to the M$\,$67 CMD}
239: \label{sec:m67}
240:
241: According to high-resolution spectroscopy, M$\,$67 has [Fe/H] $= 0.0 \pm 0.03$
242: (see \citealt{tet00}; \citealt{rsp06}). For the cluster reddening, we have
243: adopted $E(B-V) = 0.038$ mag, as found from the \citet{sfd98} dust maps: this
244: estimate is probably accurate to within $\pm 0.01$ mag, given that it agrees so
245: well with the values derived using alternative methods (see \citealt{ntc87};
246: \citealt{svk99}). As far as the distance is concerned, the modulus that is
247: obtained from a main-sequence fit of the dereddened photometry to the lower MS
248: segments of our isochrones should also be quite accurate (probably to within
249: $\pm 0.1$ mag) because the model temperatures satisfy the solar constraint and
250: the adopted color transformations (by \citealt{vc03}) are in very good agreement
251: with the empirical relations derived by \citet{sf00} from their study of solar
252: neighborhood stars having well-determined properties (see the VandenBerg \& Clem
253: paper). The application of the MS-fitting technique yields $(m-M)_V = 9.70$.
254:
255: With these choices for the basic parameters of M$\,$67, the isochrone for each
256: value of $Z_\odot$ that provides the best fit to the subgiants is readily
257: identified. As shown in Figure~\ref{fig:vfig2}, the 3.9 Gyr isochrone for
258: $Z=0.0165$ (GS98 metal abundances) provides a good match to the morphology of
259: the M$\,$67 CMD all the way from the ZAMS to the base of the giant branch,
260: including, in particular, the location of the gap near the turnoff. (Many of
261: the stars just brighter than the gap, at $M_V \approx 3.1$, are presumably
262: binaries, given that M$\,$67 is known to have an unusually large binarity
263: fraction; at least 63\%, according to Montgomery et al.~1993. The $\sim 0.06$
264: mag offset in color at a given $M_V$ along the lower RGB is discussed by
265: VandenBerg et al.~2007.) In stark contrast with this, the 4.2 Gyr isochrone
266: for $Z=0.0125$ (the Asplund et al.~metallicity) fails to reproduce the location
267: of the near-turnoff stars of M$\,$67, despite providing a comparably good fit
268: to the ZAMS and the SGB. This isochrone shows no indication at all of a
269: blueward hook feature, and thus it does not predict a gap in the distribution
270: of stars near the turnoff where one is observed.\footnote{Although one might
271: question the reality of the gap in the Montgomery et al.~(1993) CMD, there is
272: no doubt about its existence (in the same magnitude range as that indicated in
273: Fig.~\ref{fig:vfig2}) in the CMD published by \citet{sa04} for the
274: high-probability single-star members. Moreover, the observed distribution of
275: single stars along the cluster fiducial sequence (see Sandquist's Fig.~14) is
276: very reminiscent of that predicted by best-fitting isochrones (see Fig.~14 in
277: the study by \citealt{mrr04}). [The CMD obtained by Sandquist is not used here
278: because he did not provide $BV$ photometry (only $VI$), and his $V-I$ colors,
279: but not his $V$ magnitudes, are at odds with other measurements (see the
280: discussion by \citealt{vs04}). Perhaps the main advantage of using $BV$ data
281: in our analysis is that, as mentioned above, the $(B-V)$--$\teff$ relations
282: that have been used to transpose the models to the observed plane appear to be
283: especially reliable.]}
284:
285: The oldest isochrone in the $Z=0.0125$ set that has the required turnoff
286: morphology is one for an age of 3.6 Gyr, and indeed, it provides a satisfactory
287: fit to the CMD of M$\,$67 if $E(B-V) \approx 0.065$ and $(m-M)_V \approx 9.90$.
288: However, such a high reddening seems to be ruled out and it is doubtful that
289: the model colors are in error by as much as 0.03 mag. Isochrones for the
290: Asplund metallicity thus appear to pose similar difficulties for our
291: understanding of M$\,$67 as for helioseismology, in the sense that {\it stellar
292: models with higher values of $Z_\odot$ are better able to explain the
293: observations}.
294:
295: As the CNO elements play such a key role in the interpretation of the cluster
296: data, one must first consider whether their abundances, relative to iron, are
297: higher in M$\,$67 than in the Sun. An increase in the abundances of C, N,
298: {\it and} O by $\sim 0.1$ dex (or more, if the increase is limited to one or
299: two of these elements) may be enough to reconcile the predicted and observed
300: CMDs. However, while a true cosmic spread among solar metallicity disk stars
301: with an amplitude of about $0.1$ dex in [C/Fe], [N/Fe], and [O/Fe] cannot be
302: excluded (see the recent surveys by \citealt{rtl03} and \citealt{eis06}), the
303: abundance analysis of M$\,$67 main-sequence stars carried out by \citet{rsp06}
304: does not support the possibility that M$\,$67 has [CNO/Fe] $\gta 0.1$. These
305: authors derived a mean oxygen abundance corresponding to [O/Fe] $= -0.01\pm
306: 0.03$, which agrees well with the findings of \citet{tet00} for cluster giants.
307: The latter also derived C and N abundances, though, in the case of giant stars,
308: they are expected to have been altered by CN burning and the first dredge-up.
309: They find that [C/H] and [N/H] are typically $-0.2$ and $+0.2$, respectively,
310: which is roughly consistent with initial C and N abundances slightly {\it lower}
311: than solar. In fact, \citet{fb92} found a mean [C/H] of $-0.09 \pm 0.03$ for
312: three F dwarfs of M$\,$67. [Based on the work of \citet{npa02}, we find that
313: the expected differential effects due to the use of 3-D model atmospheres in
314: the determination of [O/Fe] in the cluster turnoff stars, like those analyzed
315: by Randich et al., would amount to $+0.03$ dex relative to the solar value.]
316: Thus, it seems improbable that the CNO abundances relative to iron are enhanced
317: in the cluster with respect to the Sun by as much as $0.1$ dex. We also note
318: that an increase of the Ne abundance, which has been suggested (by
319: \citealt{bbs05}) to solve the discrepancy between solar oscillation data and
320: the Asplund et al.~abundances would not resolve the M$\,$67 conundrum.
321:
322: %Because the turnoff stars in M$\,$67 are more massive than the Sun by $\sim
323: %0.2$--$0.3 \msol$, it is not necessarily the case that the same refinements
324: %will explain both discrepancies. One can speculate, for instance, that an
325: %improved treatment of {\it convective core overshooting} could lead to a
326: %reconciliation of low-$Z$ solar abundance models with the observed CMD of
327: %M$\,$67. However, this seems unlikely because the empirical calibration of the
328: %dependence of overshooting on stellar mass that has been assumed here is based
329: %on, and provides a consistent interpretation of, observations of a number of
330: %binary stars as well as the CMDs of several open clusters encompassing wide
331: %ranges in age and chemical composition (see, e.g., VBD06). Overshooting is not
332: %a factor in the current dilemma concerning solar oscillations because the mass
333: %of the Sun is too low for it to have a convective core.
334:
335: Although our present understanding of diffusive processes precludes the
336: possibility that there is a large difference between the interior and
337: photospheric metallicities of the Sun, \citet{mrr04} have found from their
338: calculations for $Z_\odot \ge 0.0175$ that the lowest mass diffusive model with
339: a convective core on the MS is $\approx 1.10 \msol$, as opposed to $1.14 \msol$
340: in the case of non-diffusive models. The difference in mass is comparable to
341: the difference in $\mtr$ predicted by our calculations when the assumed value
342: of $Z$ is increased from 0.0125 (Asplund et al.) to 0.0165 (GS98). Consequently,
343: it seems quite possible that low-$Z$ {\it diffusive} models (with or without
344: some core overshooting) may be able to satisfy the M$\,$67 constraint. Further
345: investigation is needed. However, even if this leads to a consistent
346: explanation for both the Asplund et al.~solar abundances and the M$\,$67 CMD,
347: the large discrepancies with helioseismology would remain.
348:
349: %(Of course, if diffusion has had a large impact
350: %on the metallicity of the solar photosphere, then the apparent consistency
351: %between the abundances of CNO in the Sun and in the interstellar medium cannot
352: %be used as an argument in support of Asplund's abundance determinations.
353: %Anyway, it seems doubtful that the local interstellar medium is the same as
354: %that where the Sun was born at $\sim 1/2$ of the present age of the Galactic
355: %disk.)
356:
357: \section{Conclusions}
358: \label{sec:conclude}
359: The new Asplund et al.~metallicity for the Sun presents some difficulties for
360: fits of solar abundance models to the M$\,$67 CMD, in that they do not predict
361: a gap near the turnoff where one is observed. Whether or not stellar models
362: predict a gap at the observed $M_V$ in this open cluster depends quite
363: sensitively on the assumed CNO abundances, and it is mainly the revision of
364: these elemental abundances (along with that of Ne) that is responsible for the
365: decrease in $Z_\odot$ from 0.0165 (GS98) to 0.0125. Isochrones for the higher
366: value of $Z$ are able to reproduce the detailed CMD morphology of M$\,$67 in
367: the vicinity of the turnoff without apparent problems. Interestingly, it is
368: primarily the reduction in the abundances of CNO that has resulted in
369: substantial difficulties for helioseismology; e.g., an increase from
370: $\sim 0.3$\% to $\sim 3$\% in the differences between the predicted and inferred
371: sound speed squared profiles. Are solar oscillation studies and our
372: investigation of M$\,$67 telling us that the low solar $Z$ determined by
373: Asplund et al.~is wrong?
374:
375: Not necessarily. It is possible, judging from the work of \citet{mrr04},
376: that models which take diffusive processes into account may not have the same
377: difficulties as the models used in this study, which neglect such processes.
378: That is, if the Asplund et al.~solar abundances are correct, {\it only} those
379: low-$Z$ models that treat diffusion may be able to reproduce the M$\,$67 CMD.
380: This possibility, which would {\it not} resolve the quandary presented by
381: solar oscillations, needs to be carefully studied. Diffusion clearly adds
382: another level of complexity to the problem since, e.g., the initial abundances
383: of the gas out of which the Sun and M$\,$67 formed would have been somewhat
384: different, if they have the same abundances today, because the Sun is up to 1
385: Gyr older than M$\,$67. Importantly, MS stars in M$\,$67 should show systematic
386: variations in their surface metallicities as a function of $\teff$ due to the
387: operation of diffusive processes (see Michaud et al.).
388:
389:
390:
391: \acknowledgements
392: We thank Martin Asplund and Santi Cassisi for valuable suggestions and comments
393: on the manuscript. This work has been supported by the Natural Sciences and
394: Engineering Research Council of Canada through a Discovery Grant to DAV, and by
395: grants from the Swedish Research Council to BE, KE, and BG. JF acknowledges
396: support from NSF grant AST-0239590.
397:
398: \newpage
399:
400: \begin{thebibliography}{}
401:
402: \bibitem[Antia \& Basu(2005)]{ab05}
403: Antia, H.~M., \& Basu, S.~2005, ApJ, 620, L129.
404:
405: \bibitem[Asplund(2005)]{asp05}
406: Asplund, M.~2005, ARA\&A, 43, 481.
407:
408: \bibitem[Asplund, Grevesse, \& Sauval(2006)]{ags06}
409: Asplund, M., Grevesse, N., \& Sauval, A.~J.~2006, Comm.~in Asteroseismology,
410: 147, 76.
411:
412: \bibitem[Asplund et al.(2005)]{ags05}
413: Asplund, M., Grevesse, N., \& Sauval, A.~J., Allende Prieto, C., \& Blomme,
414: R.~2005, A\&A, 431, 693.
415:
416: \bibitem[Asplund et al.(2000)]{ant00}
417: Asplund, M., Nordlund, A., Trampedach, R., Allende Prieto, C., \& Stein,
418: R.~F.~2000, A\&A, 359, 729.
419:
420: \bibitem[Bahcall, Basu \& Serenelli(2005)]{bbs05}
421: Bahcall, J.~N., Basu, S., \& Serenelli, A.~M.~2005, ApJ, 631, 1281.
422:
423: \bibitem[Bahcall, Pinsonneault, \& Basu(2001)]{bpb01}
424: Bahcall, J.~N., Pinsonneault, M.~H., \& Basu, S.~2001, ApJ, 555, 990.
425:
426: %\bibitem[Boothroyd \& Sackmann(2003)]{bs03}
427: %Boothroyd, A.~I., \& Sackmann, J.~J.~2003, ApJ, 583, 1004.
428:
429: \bibitem[Christensen-Dalsgaard(2002)]{cd02}
430: Christensen-Dalsgaard, J.~2002, Rev.~Mod.~Phys., 74, 1073.
431:
432: \bibitem[Ecuvillon et al.(2006)]{eis06}
433: Ecuvillon, A., Israelian, G., Santos, N.~C., Shchukina, N.~G., Mayor, M., \&
434: Rebolo, R.~2006, A\&A, 445, 633.
435:
436: \bibitem[Ferguson et al.(2005)]{faa05}
437: Ferguson, J.~W., Alexander, D.~R., Allard, F., Barman, T., Bodnarik, J.~G.,
438: Hauschildt, P.~H., Heffner-Wong, A., \& Tamani, A.~2005, ApJ, 623, 585.
439:
440: \bibitem[Friel \& Boesgaard(1992)]{fb92}
441: Friel, E.~D., \& Boesgaard, A.~M.~1992, ApJ, 387, 170.
442:
443: \bibitem[Grevesse \& Noels(1993)]{gn93}
444: Grevesse, N., \& Noels, A.~1993, Phys.~Scr., T47, 133.
445:
446: \bibitem[Grevesse \& Sauval(1998; hereafter GS98)]{gs98}
447: Grevesse, N., \& Sauval, A.~J.~1998, Space Sci.~Rev., 85, 161.\ \ \ \ (GS98)
448:
449: \bibitem[Gustafsson et al.(2003)]{gee03}
450: Gustafsson, B., Edvardsson, B., Eriksson, K., Mizuno-Wiedner, M.,
451: Jorgensen, U.~G., \& Plez, B.~2003, in Stellar Atmosphere Modelling,
452: eds.~I.~Hubeny, D.~Mihalas, \& K.~Werner, ASP Conf.~Ser., 288, 331.
453:
454: \bibitem[Guzik(2005)]{gu05}
455: Guzik, J.~2005, ApJ, 627, 1049.
456:
457: %\bibitem[Hobbs \& Thorburn(1991)]{ht91}
458: %Hobbs, L.~M., \& Thorburn, J.~A.~1991, AJ, 102, 1070.
459: %
460: \bibitem[Iglesias \& Rogers(1996)]{ir96}
461: Iglesias, C.~A., \& Rogers, F.~J.~1996, ApJ, 464, 943.
462:
463: \bibitem[Korn et al.(2006)]{kgr06}
464: Korn, A.~J., Grundahl, F., Richard, O., Barklem, P.~S., Mashonkina, L.,
465: Collet, R., Piskunov, N., \& Gustafsson, B.~2006, Nature, 442, 657.
466:
467: \bibitem[Michaud et al.(2004)]{mrr04}
468: Michaud, G., Richard, O., Richer, J., \& VandenBerg, D.~A.~2004, ApJ, 606, 452.
469:
470: \bibitem[Montalb\'an et al.(2006)]{mmt06}
471: Montalb\'an, J., Miglio, A., Th\'eado, Noels, A., \& Grevesse, N.~2006,
472: Comm.~in Asteroseismology, 147, 80.
473:
474: \bibitem[Montgomery, Marschall, \& Janes(1993)]{mmj93}
475: Montgomery, K.~A., Marschall, L.~A., \& Janes, K.~A.~1993, AJ, 106, 181.
476:
477: \bibitem[Nissen, Twarog, \& Crawford(1987)]{ntc87}
478: Nissen, P.~E., Twarog, B.~A., \& Crawford, D.~L.~1987, AJ, 93, 634.
479:
480: \bibitem[Nissen et al.(2002)]{npa02}
481: Nissen, P.~E., Primas, F., Asplund, M., \& Lambert, D.~L.~2002, A\&A, 390, 235.
482:
483: \bibitem[Randich et al.(2006)]{rsp06}
484: Randich, S., Sestito, P., Primas, F., Pallavicini, R., \& Pasquini, L.~2006,
485: A\&A, 450, 557.
486:
487: \bibitem[Reddy et al.(2003)]{rtl03}
488: Reddy, B.~E., Tomkin, J., Lambert, D.~L., \& Allende Prieto, C.~2003, MNRAS,
489: 340, 304.
490:
491: \bibitem[Richard et al.(2002)]{rmr02}
492: Richard, O., Michaud, G., Richer, J., Turcotte, S., Turck-Chi\'eze, \&
493: VandenBerg, D.~A.~2002, ApJ, 568, 979.
494:
495: \bibitem[Rood \& Crocker(1985)]{rc85}
496: Rood, R.~T., \& Crocker, D.~A.~1985, in Production and Distribution of C, N, O
497: Elements, eds.~I.~J.~Danziger, F.~Matteucci, \& K.~Kj\"ar (Garching: ESO),
498: p.~61.
499:
500: \bibitem[Sandquist(2004)]{sa04}
501: Sandquist, E.~2004, MNRAS, 347, 101.
502:
503: \bibitem[Sarajedini et al.(1999)]{svk99}
504: Sarajedini, A., von Hippel, T., Kozhurina-Platais, V., \& Demarque, P.~1999,
505: AJ, 118, 2894.
506:
507: \bibitem[Schlegel, Finkbeiner, \& Davis(1998)]{sfd98}
508: Schlegel, D.~J., Finkbeiner, D.~P., \& Davis, M.~1998, ApJ, 500, 525.
509:
510: \bibitem[Sekiguchi \& Fukugita(2000)]{sf00}
511: Sekiguchi, M., \& Fukugita, M.~2000, AJ, 120, 1072.
512:
513: \bibitem[Tautvai{\v s}ien{\.e} et al.(2000)]{tet00}
514: Tautvai{\v s}ien{\.e}, G., Edvardsson, B., Tuominen, I., \& Ilyin, I.~2000,
515: A\&A, 360, 499.
516:
517: \bibitem[Turck-Chi\`eze et al.(2004)]{tcp04}
518: Turck-Chi\`eze, S., Couvidat, S., Piau, L., Ferguson, J., Lambert, P., Ballot,
519: J., Garc\'ia, J., \& Ngheim, P.~2004, Phys.~Rev.~Lett., 93, 211102.
520:
521: \bibitem[Turcotte et al.(1998)]{trm98}
522: Turcotte, S., Richer, J., Michaud, G., Iglesias, C.~A., \& Rogers, F.~J.~1998,
523: ApJ, 504, 539.
524:
525: \bibitem[VandenBerg, Bergbusch, \& Dowler(2006; hereafter VBD06)]{vbd06}
526: VandenBerg, D.~A., Bergbusch, P.~A., \& Dowler, P.~D.~2006, ApJS, 162,
527: 375.\ \ \ \ (VBD06)
528:
529: \bibitem[VandenBerg \& Clem(2003)]{vc03}
530: VandenBerg, D.~A., \& Clem, J.~L.~2003, AJ, 126, 778.
531:
532: \bibitem[VandenBerg et al.(2007)]{vee07}
533: VandenBerg, D.~A., Edvardsson, B., Eriksson, K., \& Gustafsson, B.~2007, ApJ,
534: submitted.
535:
536: \bibitem[VandenBerg \& Stetson(2004)]{vs04}
537: VandenBerg, D.~A., \& Stetson, P.~B.~2004, PASP, 116, 997.
538:
539: \end{thebibliography}
540:
541: \clearpage
542: \begin{figure}
543: \plotone{f1.eps}
544: \caption{Grids of evolutionary tracks for the indicated values of $Z$ and masses
545: (in solar units). The sequences for 1.155 and $1.156 \msol$ (dashed curves) and
546: for 1.195 and $1.196 \msol$ (solid curves) show how rapidly the morphology of
547: the tracks changes near $\mtr$ (see the text). Although not plotted, both grids
548: were extended down to $0.6 \msol$.}
549: \label{fig:vfig1}
550: \end{figure}
551:
552: \clearpage
553: \begin{figure}
554: \plotone{f2.eps}
555: \caption{Comparison of isochrones for the indicated and ages and metallicities
556: with the CMD of M$\,$67 (Montgomery et al.~1993), on the assumption of the
557: reddening and apparent distance modulus specified in the lower left-hand
558: corner. The vertical line bounded by short horizontal lines, located adjacent
559: to the turnoff, indicates our estimate of the luminosity spanned by the gap
560: in the observed CMD. Only the dashed isochrone has a blueward hook in the
561: same magnitude range.}
562: \label{fig:vfig2}
563: \end{figure}
564:
565: \clearpage
566: \begin{deluxetable}{ccc}
567: \tabletypesize{\footnotesize}
568: \tablecaption{Solar Elemental Abundances \label{tab:tab1}}
569: \tablewidth{0pt}
570: \tablehead{ & \multispan2\hfil $\log N$\hfil \\ & \multispan2\hrulefill \\
571: Element & Grevesse \& Sauval (1998) & Asplund et al.\tablenotemark{a} }
572: \startdata
573: H\phantom{e} & 12.00\phantom{38} & 12.00\phantom{87} \\
574: He\tablenotemark{b} & 10.9738 & 10.9487 \\
575: C\phantom{e} & \phantom{1}8.52\phantom{38} & \phantom{1}8.41\phantom{87} \\
576: N\phantom{e} & \phantom{1}7.92\phantom{38} & \phantom{1}7.80\phantom{87} \\
577: O\phantom{e} & \phantom{1}8.83\phantom{38} & \phantom{1}8.66\phantom{87} \\
578: Ne & \phantom{1}8.08\phantom{38} & \phantom{1}7.84\phantom{87} \\
579: Na & \phantom{1}6.33\phantom{38} & \phantom{1}6.33\phantom{87} \\
580: Mg & \phantom{1}7.58\phantom{38} & \phantom{1}7.58\phantom{87} \\
581: Al & \phantom{1}6.47\phantom{38} & \phantom{1}6.47\phantom{87} \\
582: Si & \phantom{1}7.55\phantom{38} & \phantom{1}7.51\phantom{87} \\
583: P\phantom{e} & \phantom{1}5.45\phantom{38} & \phantom{1}5.45\phantom{87} \\
584: S\phantom{e} & \phantom{1}7.33\phantom{38} & \phantom{1}7.33\phantom{87} \\
585: Cl & \phantom{1}5.50\phantom{38} & \phantom{1}5.50\phantom{87} \\
586: Ar & \phantom{1}6.40\phantom{38} & \phantom{1}6.18\phantom{87} \\
587: K\phantom{e} & \phantom{1}5.12\phantom{38} & \phantom{1}5.12\phantom{87} \\
588: Ca & \phantom{1}6.36\phantom{38} & \phantom{1}6.36\phantom{87} \\
589: Ti & \phantom{1}5.02\phantom{38} & \phantom{1}5.02\phantom{87} \\
590: Cr & \phantom{1}5.67\phantom{38} & \phantom{1}5.67\phantom{87} \\
591: Mn & \phantom{1}5.39\phantom{38} & \phantom{1}5.39\phantom{87} \\
592: Fe & \phantom{1}7.50\phantom{38} & \phantom{1}7.45\phantom{87} \\
593: Ni & \phantom{1}6.25\phantom{38} & \phantom{1}6.25\phantom{87} \\
594: \multispan3\hrulefill \\
595: $Y$ & 0.2676 & 0.2559 \\
596: $Z$ & 0.0165 & 0.0125 \\
597: \enddata
598: \tablenotetext{a}{Abundances for C, N, O, Ne, Si, Ar, and Fe were provided
599: by M.~Asplund (2004, priv.~comm.): Grevesse \& Sauval (1998) abundances are
600: assumed for all other elements heavier than helium.}
601: \tablenotetext{b}{Determined from Standard Solar Models.}
602: \end{deluxetable}
603:
604: \end{document}
605: