astro-ph0001442/u5.tex
1: %latex    u5.tex
2: %\title{RR Lyrae -- Theory vs Observation}
3: % revised version
4: 
5: \documentstyle[emulateapj,flushrt,epsf,rotate]{article}
6: 
7: \long\def\jumpover#1{{}}
8: 
9: \def\approxgt{\,\raise2pt \hbox{$>$}\kern-8pt\lower2.pt\hbox{$\sim$}\,}
10: \def\approxlt{\,\raise2pt \hbox{$<$}\kern-8pt\lower2.pt\hbox{$\sim$}\,}
11: \def \th{\thinspace}
12: \def \ngth{\negthinspace}
13: \def \ni{\noindent}
14: %\def \ub#1{{\underbar {#1}}}
15: %\def \ul#1{{\underbar {#1}}}
16: \def \Teff{{$T_{\rm {ef\!f}} $}}
17: \def \Mo{{$M_\odot $}}
18: \def \Lo{{$L_\odot $}}
19: \def \at{{\rm\char'100}}
20: %\def \ahalf   {{\frac 12}}
21: %\def \athird {{\frac 13}}
22: %\def \quarter{{\frac 14}}
23: %\def \aquarter{{\frac 14}}
24: %\def \nth{{$^{th}$}}
25: %\def \first{{1$^{st}$}}
26: \def \apriori{{\it a priori\ }}
27: \def \eg{{{\it e.g.},\ }}
28: \def \etal{{\it et al.\ }}
29: \def \etc{{\it etc.\ }}
30: \def \cf{{\it cf.\ }}
31: \def \ia{{{\it inter alia},\ }}
32: \def \ie{{{\it i.e.},\ }}
33: \def \iff{{{\it iff},\ }}
34: \def \via{{\it via\ }}
35: \def \viz{{\it viz.\ }}
36: \def \vs{{\it vs.\ }}
37: %  . with d above for day
38: \def \dotd{\hbox{$.\!\!^{\rm d}$}}
39: \def \dotm{\hbox{$.\!\!^{\rm m}$}}
40: \def \dd{{$^d$}}
41: %\def \ML{{{\it M \ngth -- \ngth L}}}
42: \def\Log{{\mathrm Log}}
43: \def\LogL{{Log$\th L$}}
44: \def \LTeff{{Log$\th T_{{\rm ef\!f}} $}}
45: \def\LogP{{Log$\th P$}}
46: 
47: 
48: \def\rn{\noindent\parshape 2 0truecm 8.8truecm 0.3truecm 8.5truecm}
49: 
50: \begin{document}
51: 
52: \submitted{ASTROPHYSICAL JOURNAL, revised version, in press}
53: 
54: \title{RR Lyrae -- Theory \lowercase{vs} Observation}
55: \author{Zolt\'an Koll\'ath$^{1}$,
56: J. Robert Buchler$^{2}$
57:  \& Michael Feuchtinger$^{2}$
58: }
59:  \begin{abstract}
60: 
61: The luminosities, effective temperatures and metallicities that are derived
62: empirically by Kov\'acs and Jurcsik from the light curves of a large number of
63: globular cluster and field RRab and RRc stars are compared to theoretical RR Lyrae
64: models.  The strong luminosity dependence of the empirical blue and red edges
65: (\LogL\ \vs \LTeff\ diagram) is in disagreement with that of both radiative and
66: convective models.  A reexamination of the theoretical uncertainties in the
67: modelling leads us to conclude that the disagreement appears irreconcilable.
68: 
69: \end{abstract}
70: 
71:  \keywords{
72: stars: variable, 
73: stars: oscillations, 
74: stars: RR Lyrae,
75: stars: horizontal-branch, 
76: stars; abundances, 
77: stars: distances,
78: stars: atmospheres, 
79: stars: fundamental parameters,
80: globular cluster}
81: 
82: 
83: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
84: {\bigskip
85:         {\footnotesize
86: \noindent $^1$Konkoly Observatory, Budapest, HUNGARY; kollath\at
87: konkoly.hu  \\
88: \noindent $^2$Physics Department, University of Florida, Gainesville, FL, USA;
89: buchler\at phys.ufl.edu, fm\at phys.ufl.edu
90: }}
91: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
92: 
93: 
94:  \section{Introduction}
95: 
96: The recent work of Kov\'acs \& Jurcsik (Kov\'acs \& Jurcsik 1996 [KJ96], 1997
97: [KJ97], Jurcsik 1998 [J98]) proposes an almost purely empirical method of
98: extracting the absolute magnitudes, colors and effective temperatures (\Teff)
99: and metallicities directly from the observed periods and light curves of RR
100: Lyrae stars.  The only theoretical input appears in the transformation from
101: M$_V$ and V -- K to L and \Teff\ via Kurucz's static model atmospheres.  The
102: very empirical nature and potential usefulness of the approach has attracted a
103: great deal of attention from observers.
104: 
105: The most recent work (J98) compiles and analyzes a sizeable set of
106: observational data of RR Lyrae cluster variables and RR Lyrae field stars.
107: This study thus includes RR Lyrae stars with metallicities ranging from
108: Z=0.00001 to almost solar Z=0.020.  The end products of her analysis that we
109: are most concerned about here are \LogL-\LTeff\ and \LogL-\LogP\ plots, and the
110: byproducts which are relations between luminosity L, mass M and metallicity Z.
111: 
112: The J98 \LogL-\LTeff\ and \LogL-\LogP\ data are shown in Figure~1 as small dots,
113: circles for RRab and triangles for RRc stars.  The figures indicate well
114: defined slopes for the fundamental and overtone blue edges and red edges, all
115: four of which have approximately the same values, although there may be a
116: slight broadening of the instability strip with luminosity.  Hereafter we call
117: this slope $\Xi$ ($=\Delta$~\LogL / $\Delta$~\LTeff).
118: 
119: J98 pointed out that no model calculations can explain the strong dependence of
120: the temperature on the luminosity.  In \S2 we first confirm that indeed the
121: slope $\Xi$ of this empirical \LogL\ \vs \LTeff\ diagram is in irreconcilable
122: disagreement with that of radiative models.  Next, in \S3 we show that the
123: inclusion of turbulent convection in the models only shifts the \LogL\ \vs
124: \LTeff\ line, but does not change the slope.  The discrepancy is therefore very
125: basic because it is already inherent in models that are as fundamental as
126: purely radiative ones.  In \S4 we reexamine the uncertainties of both radiative
127: and turbulent model calculations and conclude that the theoretical $\Xi$ is
128: very robust and cannot be changed very much without introducing new physics.
129: In \S5 we discuss the theory that goes into the empirical relations of Jurcsik,
130: in particular the influence of the use of the static Kurucz color --
131: temperature transformation.  We conclude in \S6.
132: 
133: 
134:  \section{Radiative Models of RR Lyrae Stars}
135: 
136: In the following we examine how well RR Lyrae models agree with the J98 data.
137: We first examine radiative models, and thus limit ourselves to the vicinity of
138: the blue edges.
139: 
140: 
141: \begin{figure*}
142: \vspace{0cm}
143: \centerline{{\vbox{\epsfxsize=9cm\epsfbox{fig1a.ps}}}
144:             {\vbox{\epsfxsize=9cm\epsfbox{fig1b.ps}}}}
145: \vspace{0.2cm}
146: \noindent{\small
147:  Fig.~1 :
148: Left: \LogL-\LTeff\ plot:
149:  RRab (filled circles) and RRc (triangles) stars from Jurcsik 
150: ([J98] and priv. comm.);
151:  Theoretical fundamental and first overtone blue edges for radiative models:
152: {\it solid lines}: fixed mass and Z;  
153: {\it dotted lines} have M and Z  from Jurcsik's relation; 
154: {\it dashed lines} have M from Jurcsik with fixed Z=0.001; 
155: Tuggle \& Iben's overtone and fundamental blue edges are reported as open
156: circles and squares, respectively.
157: Right:  Same data in a \LogL-\LogP\ plot.
158:  }
159:  \label{fig1}
160: \end{figure*}
161: 
162: 
163: 
164: Figure~1 shows a \LogL-\LTeff\ plot on the left.  On these we superpose the
165: linear fundamental and first overtone blue edges of radiative RR Lyrae models.
166: We recall here that among the linear edges of the instability strip only the
167: overtone linear blue edge and the fundamental linear red edge coincide with the
168: observable edges of the instability strip.  Nonlinear dynamical effects shift
169: the observable fundamental blue edge and the overtone red edge to the red and
170: to the blue, respectively, compared to their corresponding linear edges, by up
171: to several 100\th K (\eg \cf Buchler 2000).  Strictly speaking, for radiative
172: models, it is only the linear overtone blue edge that is relevant, but the
173: linear fundamental blue edge can give a rough indication of the slope $\Xi$.
174: Furthermore, the red edges are determined by convection, and radiative red
175: edges are not relevant and are therefore not shown in Fig.~1.
176: 
177: The solid lines have been computed for models with M=0.65, X=0.75 and Z=0.001.
178: The dotted lines represent models with M and Z from Jurcsik's \{L, M, Z\}
179: relations, the dashed  lines those with M from Jurcsik, but with Z=0.001.
180: Finally, the large open circles are models from Tuggle and Iben (1972) for
181: M=0.6 and Z=0.001 (with the old Los Alamos opacities).  They have essentially
182: the same $\Xi$ as ours which are computed with the OPAL opacities (Iglesias \&
183: Rogers 1996) merged with the low temperature Alexander-Ferguson (1994) opacities.
184: 
185: Again, because we are computing radiative models we do not expect the location
186: of the blue edge to be in perfect agreement with the edge of the instability
187: strip, but we note a very large discrepancy in the slope $\Xi$.
188: 
189: Figure~1 also displays a \LogL\ -- \LogP\ plot on the right.  Only the slope of
190: the constant mass models is in almost acceptable agreement with the J98 data,
191: but those calculated with the J98 $\{M, L, Z\}$ are in strong disagreement.
192: 
193: In Figures~2 we display again a \LogL\ -- \LTeff\ plot which shows the effect of
194: composition on the location of the linear blue edges.  The (radiative) models
195: were computed with M=0.65. Here the solid lines have X=0.75, Z=0.001, the dotted
196: lines have X=0.75, Z=0.004 and the dashed lines have X=0.70, Z=0.001.
197: 
198: The location of the blue edge displays very little sensitivity to either
199: metallicity Z or helium content Y, within a reasonable range of values.
200: Adjusting the helium content or metallicity does not provide a resolution of
201: the slope discrepancy.  We  describe further tests in \S4.
202: 
203: We conclude this section by noting that purely radiative RR Lyrae models are in
204: severe disagreement with Jurcsik's \th \LogL-\LTeff\ data.
205: 
206: 
207:  \section{Convective  Models of RR Lyrae Stars}
208: 
209: The addition of time-dependent turbulent convection in the models has led to
210: some spectacular successes compared to radiative models, in particular, in
211: finally predicting double mode behavior both in RR Lyrae (Feuchtinger 1998)
212: and in Cepheids (Koll\'ath, Beaulieu, Buchler \& Yecko 1998).  One would hope
213: therefore that the inclusion of turbulent convection might also remove the
214: discrepancy between observations and theory described in the previous section.
215: 
216: Yet, in that respect all of our modelling efforts with turbulent convection
217: have proved in vain, despite the flexibility afforded by the 8 free order unity
218: ($\alpha$) parameters that the turbulent convective equations contain (\eg
219: Yecko, Koll\'ath \& Buchler 1997, Buchler\& Koll\'ath 2000).  We find that
220: turbulent convection can shift the blue edges, but cannot produce the
221: differential effect with respect to luminosity that is required to give the
222: right slope $\Xi$ in the \LogL\ \vs \LTeff\ plots.  In Figure~3 we show the
223: results for three different combinations of these parameters.
224: 
225: We note that different codes and slightly different recipes for convection
226: give essentially the same theoretical slopes.  For example, even though no
227: linear models are computed, the nonlinear hydrodynamical models of Bono \etal
228: (1997) indicate a similarly steeper slope than that of J98.
229: 
230: 
231:  \section{Discussion of the Modeling Assumptions}
232: 
233:  In this section we discuss the effects of several of the approximations that
234:  are inherent in the numerical modeling to see how robust the theoretical
235:  \LogL-\LTeff\ slope is.
236: 
237: 
238:  \subsection{Equation of State}
239: 
240: Our code uses as equation of state a simple iteration of the Saha equations
241: (e.g. Stellingwerf 1982).  This equation of state is very similar to the OPAL
242: (Rogers \etal 1996) and the MHD (D\"appen \etal 1988) equations of state and we
243: do not believe that the tiny differences can be responsible for the discrepancy
244: between the models and the J98 data.
245: 
246: \vfill\eject
247: 
248:  \subsection{Radiative Transport}
249: 
250: Our radiative code uses a standard equilibrium diffusion approximation for
251: the radiative transport, \ie
252: 
253: \begin{equation}
254:  L = (4\pi R^2)^2 {c\over 3\kappa} {\partial\over \partial m} aT^4
255: \label{L_eq}
256: \end{equation}
257: 
258: \ni in which one uses for the opacity $\kappa$ the Rosseland mean (\eg Mihalas
259: \& Mihalas 1984).  Eq.~\ref{L_eq} has several shortcomings in the region above
260: the photosphere.  First, it implies an Eddington factor of $f_E=1/3$, which is
261: not correct in the optically thin outer region where $f_E$ approaches 0.4--0.5
262: (Feuchtinger \& Dorfi 1994, Fig.~3).  Second, in this regime, the opacities
263: should be higher than those given by the Rosseland mean (Alexander \& Ferguson
264: 1994).  Finally, for the computation of the periods and growth rates it would
265: be more appropriate to linearize the radiation hydrodynamics equations rather
266: than the equilibrium diffusion equation.  The magnitude of the errors
267: introduced by the first effect can easily be estimated.
268: 
269: 
270: 
271: 
272: 
273: One can approximate the effect of the Eddington factor in Eq.~\ref{L_eq}. by
274: replacing $3\kappa$ by $\kappa/f_E$.  The Eddington factor also appears in the
275: radiation pressure which becomes $p_{\rm rad}\rightarrow 3f_E \th p_{\rm rad}$.
276: We can disregard sphericity effects in the momentum equation contained in the
277: term $(p_{\rm rad}-f_E\th e_{\rm rad})/r$.
278: 
279: To simulate the effect of an increasing $f_E$ we assume that for $T <$ \Teff,
280: $f_E$ increases smoothly from 1/3 to 1/2 with decreasing temperature.  The
281: results confirm our hunch that an increase of $f_E$ in the very outer region
282: has very little influence on the growth rates and almost none on the
283: periods. More importantly it has no differential sensitivity to luminosity.
284: 
285: 
286: 
287: 
288: %\begin{figure}
289: \centerline{\vbox{\epsfxsize=9cm\epsfbox{fig2.ps}}}
290: \noindent{\small
291:  Fig.~2 :
292:  \LogL-\LTeff\ plot:
293:  RRab (filled circles) and RRc (triangles) stars of Jurcsik;
294:  Radiative blue edges for M=0.65;
295:  solid line:   X=0.75 Z=0.001,
296:       dotted line:  X=0.75 Z=0.004,
297:       dashed line:  X=0.70 Z=0.001.
298:  }
299:  \label{fig2}
300: %\end{figure}
301: 
302: 
303: 
304: % \begin{figure}
305:  \centerline{\vbox{\epsfxsize=9cm\epsfbox{fig3.ps}}}
306:  \noindent{\small
307:  Fig.~3 :
308:  \LogL-\LTeff\ plot:
309:  RRab (filled circles) and RRc (triangles) stars:
310:   Convective models:  Solid, dotted and dashed lines are for three different
311:  combinations of the $\alpha$ parameters in the convective model equations,
312:  chosen to give a reasonable width for the instability strip.
313:  }
314:  \label{fig3}
315:  \vspace{5mm}
316: % \end{figure}
317: 
318: 
319: 
320: \begin{figure*}
321: \centerline{{\vbox{\epsfxsize=9cm\epsfbox{fig4a.ps}}}
322:             {\vbox{\epsfxsize=9cm\epsfbox{fig4b.ps}}}}
323: \vspace{0.2cm} \noindent{\small
324:  Fig.~4 :
325:  Left: \Teff-(V-I) plot and right: (V-I)-M$_V$ plot:
326:  Theoretical color variations for two nonlinear full amplitude RRab models with
327:  different luminosities of L = 40 \Lo (dashed line) and L = 52 \Lo (full
328:  line). Crosses indicate the time steps of the nonlinear
329:  calculations. See text for details.}
330:  \label{fig4}
331: \vskip 0.5cm
332: \end{figure*}
333: 
334: 
335: We have artificially, and somewhat arbitrarily, increased the opacity from the
336: vicinity of \Teff\ outward.  We have not found an appreciable differential
337: effect with \Teff, and have therefore not pursued this avenue with more
338: detailed modelling.
339: 
340: Finally, the insensitivity of the slope to $f_E$ makes us believe that the
341: linearization of the radiation hydrodynamics equations instead of the common
342: equilibrium diffusion equation (requiring a serious coding effort) would also
343: make little difference.
344: 
345: We conclude that the discrepancy between the models and the observations 
346: is not due to an inadequate treatment of the radiation transport.
347: 
348: 
349:  \subsection{Composition}
350: 
351: Could compositional make-up be more important than expected?  We have computed
352: a number of RR Lyrae models with various combinations of Y and Z, but find that
353: the effect on $\Xi$ is negligible.  Next we have artificially increased the
354: abundances of the easily ionizable light elements such as Mg and Na in the OPAL
355: opacities.  Again this has such a negligible effect on $\Xi$ that the inclusion
356: of this data would unnecessarily clutter Figure~2.
357: 
358: An inhomogeneous composition is unlikely to exist in RR Lyrae envelopes because
359: of convection, but even if existed our tests with various changes in
360: composition make us doubt that it would resolve the discrepancy.
361: 
362:  \subsection{Rotation}
363: 
364: Could rotation be responsible for the discrepancy?  In order to estimate the
365: magnitude of the effect of rotation we have included a spherical
366: pseudo-centrifugal acceleration $\omega^2\th r$ in the equilibrium model and in
367: the computation of the periods and growth rates.  We find that a rather short
368: rotation period, of order of a few days, would be necessary to have an impact.
369: Furthermore the disagreement is worsened by rotation because the slope
370: steepens.
371: 
372: We conclude that moderate rotation rates cannot be the cause of the
373: discrepancy.
374: 
375: \subsection{Evolutionary Effects}
376: 
377:  The blueward moving evolutionary paths turn around at some \Teff\ and then
378: move upward (\eg Dorman 1992, Lee, Demarque \& Zinn 1990).  If this happened
379: inside the linear instability strip, then the leftmost blue edge, that of the
380: overtone mode, would be defined by the topology of these paths, rather than by
381: the pulsational stability of the models.  However, the fact that the Jurcsik
382: blue and red edges of the fundamental and the overtone are all essentially
383: parallel eliminates this possibility as an explanation for the shallow observed
384: slope.  Moreover, for some of the clusters, full photometry of the horizontal
385: branch exists, and there is no indication of any gap next to the RR Lyrae
386: region.  Clearly there are stars next to the empirical red edge for which the
387: amplitude (if not zero) is less than the observational limit.
388: 
389: \subsection{Conclusion}
390: 
391: The theoretical $\Xi$ slope is very robust with respect to the uncertainties or
392: approximations inherent in our code, but is in disagreement with the
393: Jurcsik empirical relations.  We have therefore been led to reexamine
394: the latter for their robustness, especially since they also make use of some
395: theory, \viz color to temperature transformation.
396: 
397: 
398: \vskip 20pt
399: 
400:  \section{Discussion of the Empirical Relations}
401: 
402: Because of the strong disagreement between Jurcsik's processed observational
403: data and the predictions from something as basic as radiative models, it is
404: worth while to reexamine some of the most uncertain points in the reduction of
405: the observational data.
406: 
407: The observational material consists of RRab periods and lightcurves.  KJ96,
408: JK97 and J98 deduce the color, the visual magnitude M$_v$ and the metallicity
409: Z ([Fe/H] in the form of linear fully empirical relations
410: 
411: 
412: %\begin{eqnarray}
413: % \pmatrix{M_v -1.221     \cr
414: %          B-V -0.308     \cr
415: %          [Fe/H] +5.038  \cr}
416: %       &=&  \pmatrix{-1.39  & -0.477 & 0.103 \cr
417: %                      0.163 & -0.187 & 0     \cr
418: %                     -5.394 &  0     & 1.345 \cr}
419: %                             \cdot  \pmatrix{P        \cr
420: %                                             A_1      \cr
421: %                                             \phi_{31}\cr} \cr
422: %      &\equiv&  A           \cdot   \pmatrix{P        \cr
423: %                                             A_1      \cr
424: %                                             \phi_{31}\cr}
425: %\label{eq_empir}
426: %\end{eqnarray}
427: 
428: \begin{eqnarray}
429:  \pmatrix{M_v    \cr
430:            V-K    \cr
431:           [Fe/H] \cr}
432:       &\equiv&  c_0 + A     \cdot   \pmatrix{P        \cr
433:                                              A_1      \cr
434:                                              \phi_{31}\cr
435:                                              \phi_{41}\cr}
436: \label{eq_empir}
437: \end{eqnarray}
438: 
439: 
440: The connection with \LTeff\ and with \LogL\ was then made through static
441: envelope models of Kurucz (1993, \cf J98 for details).  This is the only
442: theoretical input into the otherwise empirical relations.
443: 
444: \begin{equation}
445:  \pmatrix{\Log L         \cr
446:           \L T_{ef\!f} \cr}
447: %          [Fe/H]     \cr}
448: %       &=&  \pmatrix{ a & b & c \cr
449: %                      d & e & f \cr
450: %                      g & h & i \cr}
451: %                             &\cdot&  \pmatrix{M_v\cr
452: %                                             B-V \cr
453:        = c_1 + C
454:                              \cdot  \pmatrix{M_v\cr
455:                                              V-K \cr
456:                                             [Fe/H] \cr}
457:        = c_2 + C\cdot A
458:                              \cdot  \pmatrix{P       \cr
459:                                              A_1      \cr
460:                                              \phi_{31}\cr
461:                                              \phi_{41}\cr}
462: \label{eq_kur}
463: \end{equation}
464: 
465: 
466: 
467: 
468: \begin{figure*}
469: \centerline{{\vbox{\epsfxsize=9cm\epsfbox{fig5a.ps}}}
470:             {\vbox{\epsfxsize=9cm\epsfbox{fig5b.ps}}}}
471: \vspace{0.2cm}
472: \noindent{\small
473:  Fig.~5 : Left: RRab Period -- Radius plot;
474:  {\it open circles}: Baade-Wesselink radii, and
475:  {\it crosses}: same stars from J98;
476:  The solid and dashed lines are linear regressions through the Baade-Wesselink
477: and the Jurcsik data, respectively.
478:  Right: 
479:  {\it thick lines}: radiative models with Jurcsik L, M, Z relation;
480:  {\it thin lines}: radiative models with constant mass and luminosity;
481:  {\it filled squares}: same stars from J98; 
482: }
483:  \label{fig5}
484: \vskip 1cm
485: \end{figure*}
486: 
487: \vspace{0.5cm}
488: 
489: \subsection{Color -- \Teff\ Transformation}
490: 
491: J98 uses the {\it static} envelope calculations of Kurucz's ATLAS code (for
492: details cf. J98), to make the transformation from average color to average
493: \Teff.  The \Teff\ however varies considerably during one pulsation cycle,
494: compared to the width of the instability strip, and the question arises whether
495: the correct mean color and mean temperature are still related through the
496: static Kurucz relations.  To check the validity of this assumption we have
497: computed the spectra of two full amplitude RR Lyrae models.
498: 
499: We investigate the theoretical color variation for two RR Lyrae models with
500: different luminosities (L = 40 \Lo\ and L = 52 \Lo, both for M = 0.65 \Mo,
501: \Teff = 6500 K, X = 0.76 and Z=0.001).  Nonlinear full amplitude models are
502: taken from Feuchtinger (1999) and the color variations in V and I are computed
503: as outlined in Dorfi \& Feuchtinger (1999).  The results can be inferred from
504: Fig.~\ref{fig4} which displays the color difference V -- I as a function of the
505: effective temperature (left panel) and the corresponding color-magnitude
506: diagram (right panel), both over one oscillation cycle.  Full and dotted lines
507: refer to the L = 40 \Lo~and L = 52 \Lo\ models, respectively.  The differential
508: effect with L is seen to be small.  Dynamical atmospheric effects thus cannot
509: account for the discrepancy.
510: 
511: There remains the question of the consistency of the static Kurucz atmospheres
512: with our frequency dependent calculation.  For that purpose we have checked the
513: color -- temperature transformation against the Kurucz tables at selected
514: points on the pulsation cycle.  The agreement is quite satisfactory, and any
515: discrepancies are irrelevant for our purpose here.
516: 
517: 
518: \vspace{0.9cm}
519: 
520: \subsection{RR Lyrae Period - Radius Relation}
521: 
522: The radii of RR Lyrae stars can be obtained independently either from a
523: Baade-Wesselink approach or from the J98 empirical L and \Teff.  In
524: Fig.~\ref{fig5}, on the left, we have plotted as open circles the
525: Baade-Wesselink radii of the 15 RRab stars of Jones \etal (1992) (we have
526: omitted RR~Leo).  Superposed as crosses are the radii for the same stars
527: derived from the J98 \th L and \Teff.  The agreement is remarkably good, but
528: worsens for the longer periods.  The solid and dashed lines represent linear
529: regression lines for the BW and for the J98 period -- radius relations.
530: 
531: We have also derived the radii of the additional J98 stars that were not in the
532: Jones sample and show them as small dots in the righthand figure.  The thin
533: lines represent constant mass and constant luminosity radiative models, and the
534: thick lines radiative models that obey the J98 \th L, M, Z relations
535: Since the radiative models do not provide a red edge, the upward extent of the
536: thin or thick lines is not significant.  The latter models are seen to give
537: better agreement with the swarm of the J98 empirical radii.
538: 
539: We conclude that the agreement between the empirical J98 and Jones'
540: Baade-Wesselink radii is remarkably good, and so is the agreement that can be
541: achieved with theoretical models.  A consideration of the period -- luminosity
542: relations therefore does not give us any clues as to the origin of the slope
543: discrepancy.
544: 
545: 
546: \vfill\eject
547: 
548: 
549: \section{Conclusions}
550: 
551: We have shown that the Jurcsik relations lead to RR Lyrae radii that are in
552: good agreement with both Baade-Wesselink radii and with theoretical radii.  On
553: the other hand, there exists a strong discrepancy between the slope of the
554: theoretical \LogL -- \LTeff\ relation and the slope of the empirical Jurcsik
555: relation.  
556: 
557: We have reviewed the physical and numerical uncertainties that enter the
558: theoretical calculations.  The discrepancy exists already at the level of
559: purely radiative modelling.  We have further shown that our model equations for
560: turbulent convection do not alter the slope of the \LogL--\LTeff\ relation
561: despite the large number (8) of adjustable parameters ($\alpha$'s) that we have
562: at our disposal.  Consequently, apart from the uncertainties inherent in a 1D
563: recipe for turbulent convection, it appears that the discrepancy is not caused
564: by a deficiency of the theoretical models.  We note that the same slope
565: discrepancy is implicit in older and more recent independent calculations
566: (Tuggle \& Iben 1972, Bono \etal 1997).
567:   
568: The derivation of the empirical relations makes use of a {\it static} Kurucz
569: color - \Teff\ transformation.  By computing the behavior of color versus
570: temperature over the pulsation cycle we have shown that the use of static
571: envelopes is in fact a very good approximation, and therefore cannot be the
572: culprit.
573: 
574: Finally, the shape of the evolutionary tracks, through their potential
575: avoidance of certain regions of the instability strip cannot be responsible
576: either.  
577: 
578: At this time the origin of this disturbing discrepancy constitutes an unsolved
579: puzzle.  It remains to be seen whether an improved treatment of turbulent
580: convection or more complete observations will resolve the difficulty.
581: 
582: 
583:  \section{Acknowledgements} The authors would like to thank Johanna Jurcsik for
584: kindly providing us with copies of her data sets.  They also acknowledge
585: fruitful discussions with Johanna Jurcsik and G\'eza Kov\'acs.  This work has
586: been supported by NSF (AST 95-28338 and AST 98-19608) and the Hungarian OTKA
587: (T-026031).
588: 
589: 
590:  \vskip 10pt
591: 
592:  \begin{references}
593: 
594:  \rn
595:  Alexander, D. R.\& Ferguson, J. W. 1994, ApJ 437, 879
596: 
597:  \rn
598:  Bono, G., Caputo, F., Castellani, V. \& Marconi, M. 1997, 
599:  A\&ASuppl 121, 327
600:  % -341
601:  %Nonlinear investigation of the pulsational properties of rr lyrae variables
602: 
603:  \rn
604:  Buchler, J. R. 2000, in {\sl The Impact of Large-Scale Surveys on Stellar
605:  Pulsations}, Eds. L. Szabados \& D. Kurtz, ASP Conf. Ser. (in press) 
606: 
607:  \rn
608:  Buchler, J. R. \& Koll\'ath, Z., 2000, {\sl Turbulent Convection in the
609:      Classical Variable Stars}, in {\sl Astrophysical Turbulence and
610:      Convection}, Eds. J.R. Buchler \& H. Kandrup, Annals of the New York
611:      Academy of Sciences, 898, 39.
612: 
613:  \rn
614:  Cohen, J. G. 1992, ApJ 400, 528
615: 
616:  \rn
617:  D\"appen, W., Mihalas, D., Hummer, D. G., Mihalas, B. W., 1988, ApJ 332, 261
618: % MHD EOS
619: 
620:  \rn
621:  Dorman, B. 1992, AJ Suppl Ser. 81, 221
622: 
623:  \rn
624:  Dorfi, E. A. \&  Feuchtinger, M. U. 1999, AA 348, 815
625:  % spectra
626: 
627:  \rn
628:  Feuchtinger, M. U. 1998, AA 337, L29
629:  % double mode RR Lyrae
630: 
631:  \rn
632:  Feuchtinger, M. U. 1999, AA 351, 103
633:  % -- 118
634:  % conv rr lyrae 
635: 
636:  \rn
637:  Feuchtinger, M. U. \& Dorfi, E. A. 1994, AA 291, 209
638:  % variable eddington factor
639: 
640:  \rn
641:  Iglesias, C.A. \& Rogers, F. J., 1996, {\sl ApJ}, 464, 943
642: 
643:  \rn
644:  Jones, R. V., Carney, B. W., Storm, J. \& Latham, D.W. 1992, ApJ 386, 646
645: 
646:  \rn
647:  Jurcsik, J., 1998, AA 333, 571
648:  %fundamental physical parameters of rrab stars
649: 
650: % \rn
651: % Jurcsik, J. \& Kov\'acs, G. 1996, AA 312, 111
652: % %determination of fe/h from lightcurves of rr lyr stars
653: 
654:  \rn
655:  Kov\'acs, G. \& Jurcsik, J. 1996, ApJLett 466, L17
656:  % on the lightcurve - luminos relation of rr lyr stars
657: 
658:  \rn
659:  Kov\'acs, G. \& Jurcsik, J. 1997, AA 322, 218
660:  % comput of the dist modul of rr lyr stars from their light and color  curves
661: 
662:  \rn
663:  Koll\'ath, Z., Beaulieu, J.P., Buchler, J. R. \& Yecko, P., 1998, 
664:    ApJ Lett 502, L55
665:  %DM
666: 
667: % \rn
668: % Kurucz, R. 1993, ATLAS9 code (\cf Jurcsik 1999)
669: 
670:   \rn
671:  Lee, Y. W., Demarque, P. \& Zinn, R. 1990, ApJ 350, 155
672: 
673:   \rn
674:  Liu, T. \& Janes, U. A. 1990, ApJ 354, 273
675: 
676:  \rn
677:  Mihalas, D. \& Mihalas, B. 1984, {\it Foundations of Radiation Hydrodynamics},
678:  (Oxford: University Press)
679: 
680:  Rogers, F.J., Swenson, F.J. and Iglesias, C. A., 1996, {\sl ApJ} 456,902
681: % OPAL EOS
682: 
683:  \rn
684:  Stellingwerf, R.F., 1982, ApJ 262, 330
685: % - 338, EOS
686: % Convection in pulsating stars I: Nonlinear Hydrodynamics
687: 
688:  \rn
689:  Tuggle, R. S.  \& Iben, I. 1972, ApJ 178, 455 
690:  % -466
691:  %On the Location of Pulsational Blue Edges and Estimates of the Luminosity and
692:  %Helium Content of RR Lyrae Stars
693: 
694:  \rn
695:  Yecko, P., Koll\'ath, Z. \& Buchler, J. R. 1998, {\sl Turbulent Convective
696:      Cepheid Models: Linear Properties}, Astronomy \& Astrophysics 336,
697:      553--564
698: 
699: \end{references}
700: 
701: 
702: 
703: \end{document}
704: \end\bye
705: 
706: