1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass[12pt]{emulateapj}
3:
4: \shorttitle{Predicted and empirical radii of RR Lyrae stars}
5: \shortauthors{Marconi et al.}
6:
7: \begin{document}
8:
9: \title{Predicted and empirical radii of RR Lyrae stars}
10:
11: \author{Marconi, M.\altaffilmark{1},
12: Nordgren, T.\altaffilmark{2},
13: Bono, G.\altaffilmark{3},
14: Schnider, G.\altaffilmark{2}, and
15: Caputo, F.\altaffilmark{3}}
16:
17: \altaffiltext{1}{INAF - Osservatorio Astronomico di Capodimonte, via
18: Moiarello 16, 80131 Napoli, Italy; marcella@na.astro.it}
19: \altaffiltext{2}{Department of Physics, University of Redlands, 1200 East
20: Colton Avenue, Redlands, CA 92373, USA; Tyler\_Nordgren@redlands.edu,
21: her\_own\_wings@hotmail.com}
22: \altaffiltext{3}{INAF - Osservatorio Astronomico di Roma, Via Frascati 33,
23: 00040 Monte Porzio Catone Italy; bono,caputo@mporzio.astro.it}
24:
25: \date{\centering drafted \today\ / Received / Accepted }
26:
27: \begin{abstract}
28: We present new theoretical Period-Radius-Metallicity (PRZ) relations for
29: RR Lyrae stars. Current predictions are based on a large set of nonlinear,
30: convective models that cover a broad range of chemical abundances and
31: input parameters. We also provide new and homogeneous estimates of
32: angular diameters for a sample of field RR Lyrae stars using a recent
33: calibration of the Barnes-Evans surface brightness relation. Predicted
34: and empirical radii are, within the errors, in reasonable agreement,
35: but in the short-period range the latter present a larger scatter.
36: As a working hypothesis we suggest that this discrepancy might be
37: due to the occurrence either of nonlinear features such as bumps or
38: a steep rising branch. New distance determination for RR Lyr itself is
39: in very good agreement with HST trigonometric parallax and with
40: pulsation parallax.
41: \end{abstract}
42:
43: \keywords{stars: evolution -- stars: horizontal branch
44: -- stars: oscillations -- stars: variables: others}
45:
46: \maketitle
47:
48: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
49: \section{Introduction}
50:
51: RR Lyrae stars are widely adopted not only as tracers of old, low-mass
52: stellar populations but also as standard candles to estimate Galactic
53: and extragalactic distances. They are ubiquitous across the Galaxy and
54: they have been detected in all stellar systems that host a well-defined
55: old population. This means that they can be adopted to constrain the
56: intrinsic accuracy of current primary distance indicators, such as
57: classical Cepheids, Tip of the Red Giant Branch, and Main Sequence
58: fitting. These are the reasons why a countless number of theoretical
59: and empirical investigations have been devoted to the RR Lyrae distance
60: scale (Caputo et al. 1999; Bono et al. 2002,2003; Cacciari \&
61: Clementini 2003; Walker 2003; Catelan, Pritzl, \& Smith 2004;
62: Gratton et al. 2004).
63:
64: Even though a paramount observational effort has been devoted to obtain
65: Baade-Wesselink (BW) distances to a sizable sample of cluster and field
66: RR Lyrae (Cacciari et al. 1989; Clementini et al. 1990;
67: Carney et al. 1992, hereafter C92; Jones et al. 1992; Storm et al. 1994a,b)
68: we still lack detailed empirical and theoretical Period-Radius
69: (PR) relations for RR Lyrae stars. Note that radii are a by product
70: of the BW method. There is only one exception, Burki \& Meylan (1986)
71: derived, using BW measurements, an empirical PR relation for
72: Type II Cepheids that according to the authors could also be
73: applied to RR Lyrae stars.
74:
75: The same outcome applies to theoretical models. Pulsation properties
76: of RR Lyrae stars have been investigated using both linear
77: and nonlinear models, we still lack detailed theoretical predictions.
78: To fill this gap we present new theoretical
79: PR relations for RR Lyrae stars based on a detailed and homogeneous set
80: of nonlinear pulsation models that cover a wide range of stellar masses
81: and chemical compositions.
82: We also investigate the dependence of the PR relation on metallicity and
83: compare theoretical predictions with empirical radius estimates.
84:
85: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
86: \section{Predicted and Empirical radii of RR Lyrae}
87:
88: During the last few years we have been developing an homogeneous theoretical
89: scenario for RR Lyrae stars by constructing an extensive grid of nonlinear,
90: convective models (Bono et al. 1997, 2001, 2003). These models cover a wide
91: range of stellar masses (0.53 $\le$ M/M$_\odot$ $\le$ 0.80),
92: and chemical compositions (0.24 $\le$ Y(He abundance) $\le$ 0.28;
93: 0.0001 $\le$ Z (metal abundance) $\le$ 0.02).
94: The physical and numerical assumptions adopted in the model computations
95: have already been discussed in previous papers
96: (Bono \& Stellingwerf 1994; Bono et al. 1997; Bono, Castellani, \& Marconi 2000).
97: Note that current nonlinear, convective models predict pulsation
98: observables such as the variation along the pulsation cycle of luminosity,
99: radius, velocity, and temperature to be compared with actual empirical data.
100: We estimated for each model the mean radius as a time
101: average over the predicted surface radius curve.
102: Figure 1 shows predicted radii for fundamental (F) pulsators, the solid
103: line displays the linear regression over the entire set of models:
104:
105: $$\log R = 0.90 (\pm0.03) + 0.65 (\pm0.03) \log P \;\;\;\; \;\;\;\;\;\sigma=0.03$$
106:
107: where $R$ is the mean RR Lyrae radius (solar units), $P$ the pulsation period (days),
108: and $\sigma$ the intrinsic dispersion.
109: For comparison Fig. 1 also shows the empirical PR relation (dashed line) for Type II
110: Cepheids derived by Burki \& Meylan (1986). The vertical error bar is the
111: standard deviation of the predicted PR relation. The agreement between
112: the empirical relation and current predictions is, within the intrinsic dispersion,
113: quite good. This evidence supports the suggestion by Burki \& Meylan concerning
114: the similarity between the PR relation of Type II Cepheids and RR Lyrae.
115: However theoretical radii present, at fixed period, a substantial
116: spread along the best-fit line, thus strongly suggesting the dependence of the
117: PR relation on a {\it second parameter}.
118:
119:
120: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
121: %\section{The Period-Radius-Metallicity relation}
122:
123: To further improve the intrinsic accuracy of the predicted PR relation we accounted
124: for metal abundance. The dependence on this parameter is expected, since both theory
125: and observations support the evidence that the mean magnitude of RR Lyrae stars
126: depends on the metal content. We performed a linear regression over the entire set
127: of F models and we found the following Period-Radius-Metallicity (PRZ) relation:
128:
129: $$\log R = 0.774(\pm0.009)\, +\, 0.580(\pm0.007) \log P\, -\, 0.035(\pm0.001) \log Z \;\;\;\;\; \;\;\;\;\;\sigma=0.008$$
130:
131: where the symbols have their usual meaning.
132: Data plotted in the bottom panel of Fig. 2 show that the inclusion of the metallicity
133: term causes a decrease in the intrinsic dispersion from 0.03 to 0.008 dex. To supply
134: a homogeneous theoretical scenario for RR Lyrae radii we also estimated the PRZ
135: relation for first overtone (FO) pulsators (see top panel of Fig. 2), and we
136: found:
137:
138: $$\log R = 0.883(\pm0.004)\, +\, 0.621(\pm0.004) \log P\, -\, 0.0302(\pm0.001) \log Z \;\;\;\;\; \;\;\;\;\; \sigma=0.004$$
139:
140: The reason why we estimated an independent PR relation for FOs is twofold:
141: {\em i))} the width in temperature of the FO instability strip is narrower when
142: compared with the F one. This means that the FO PR relation presents a smaller
143: intrinsic dispersion when compared with the F one.
144: {\em ii))} FO pulsators are systematically hotter than F ones. This means that
145: the predicted FO PR relation is marginally affected by uncertainties in the
146: treatment of convection.
147:
148:
149: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
150: %\section{Empirical radii of RR Lyrae Stars}
151:
152: To validate current predictions we collected a sample of RR Lyrae stars
153: for which are available accurate photometric and spectroscopic data,
154: namely the BW sample (Bono et al. 2003, hereafter B03). From these, we calculated
155: angular diameters using the latest calibration (Nordgren et al. 2002) of the
156: Barnes-Evans surface brightness relation (Barnes \& Evans 1976). Together with
157: radial velocities from spectroscopic data, we determined linear radii
158: (and distances) for selected RR Lyrae (see Table 1).
159: While all of these stars already had radii determined by various authors
160: (Jones et al. 1992, C92, and see references in Table 1) this
161: new sample has the benefit of uniformity and the most recent stellar
162: interferometric angular diameter measurements.
163:
164: We compiled published V and K photometry for the stars in Table 1 (see
165: references in column 9).
166: The mean values of these V and K magnitudes agree with those in B03 to within an
167: average of 0.04 magnitudes (about 0.4\% given an average V and K magnitude of 10).
168: We performed dereddening of the photometric data
169: using values for E(B-V) provided in Table 1, and extinction correction
170: constants from Cardelli et al. (1989).
171: We used both linear interpolation and polynomial-fitting in order to
172: calculate K, and thus (V-K) values at those
173: pulsation phases with V photometry. For RR Lyr itself, we used V and (V-R) photometry.
174: We calculated angular diameters as a function of pulsation phase for stars in Table 1
175: using equations (1), (2), (5) and (6) of Nordgren et al. (2002) we obtain:
176:
177: %$$\log\theta = 0.5734 - 0.2V + 0.246(V-K)$$
178: %$$\log\theta = 0.5914 - 0.2V + 0.730(V-R)$$
179: \begin{center}
180: $\log\theta = 0.5734 - 0.2V + 0.246(V-K)$
181:
182: $\log\theta = 0.5914 - 0.2V + 0.730(V-R)$
183: \end{center}
184:
185: where $\theta$ is the angular diameter in milliarcseconds (mas).
186: This version of the Barnes-Evans relation was calibrated using
187: interferometric angular diameter observations of 57 non-variable giant stars.
188: Where available, angular diameters were calculated using V and (V-K) pairs,
189: as opposed to V and (V-R), as the former relation has been shown to yield more
190: precise results (Fouqu\'{e} \& Gieren 1997).
191: Using polynomial fitting (with polynomial orders ranging from 9 to 11) we calculated
192: radial velocities from spectroscopic measurements at the same pulsation phases as the
193: calculated angular diameters (see references in Column 9 of Table 1).
194: We calculated linear displacements of the stellar surface from:
195:
196: $$\Delta R = - p \int{(V_r - V^\ast)} d\phi $$
197:
198: where $V^\ast$ is the radial velocity of the center of mass of the star and
199: $p$ is the pulsation projection factor. The value of $V^\ast$ was found by
200: integrating the $V_r$ curve over the entire phase cycle (phase = 0 - 1), and
201: demanding that $\Delta R_{0-1}$ = 0. For all but RS Boo, our value of $V^\ast$
202: agrees within the uncertainties with those published by Beers et al. (2000).
203: For the pulsation factor, Fernley (1994) argues for $p = 1.38$ for field
204: RR Lyrae stars and so we have used this value.
205:
206: The linear radius ($R_o$) and distance ($d$) for the star is found from
207: $R_o + \Delta R = 1000 d (\theta/2)$, where the angular diameter $\theta$
208: is in mas and the linear radius and radius displacement are
209: in AU (yielding the distance, $d$, in parsecs). From this equation the
210: radii (and distances) in Table 1 were calculated using a least-squares fit
211: to our calculated angular diameters and linear displacements.
212:
213: The above equation for $\Delta R$ shows that uncertainties in $V^\ast$ will be
214: propagated into the uncertainty in radius and distance. For each individual star
215: this is a systematic error, but it is an error that is random from star to star
216: within the sample. The uncertainties in radius and distance in Table 1 are the
217: random errors, with the additional systematic error due to $V^\ast$ given in
218: parentheses. The total error for each star is the quadrature sum of the two
219: errors. The comparison of random and systematic errors in Table 1 supports
220: the importance of accurate radial velocity measurements in BW type analyses.
221:
222: It should also be noted that there is an uncertainty in the value for the
223: pulsation projection factor $p$. As the linear radius displacement scales
224: linearly with $p$, so too will the final radius and distance. For instance, if one were
225: to use $p = 1.30$ (Jones, Carney \& Latham, 1988a,b) instead of our value of $1.38$,
226: all radii and distances in Table 1 will be smaller by a factor of 1.38/1.30 = 1.08.
227: For each star in Table 1, our new empirical radius estimate is in agreement, within
228: the errors, to that found by the authors of the original published photometry and
229: radial velocities (see references in Table 1). This includes the star with the
230: largest estimated radius, SS Leo. Fernley et al. (1990) estimated a radius of
231: 6.63 $R_\odot$, while Jones et al. (1992) estimate a radius of 7.32 $R_\odot$,
232: both of which, without better knowledge of their uncertainties, are in general
233: agreement with our estimate of $7.2 \pm 0.4 R_\odot$.
234:
235: Figure 2 also shows the comparison between predicted and new empirical
236: (open circles) radius estimates. Triangles display the radius estimates
237: provided by C92. Homogeneous radius determinations for a larger sample of
238: BW RR Lyrae with accurate radial velocity measurements will be provided in
239: a forthcoming investigation.
240: Observed radii have been plotted using the homogeneous compilation of metal
241: abundances provided by Fernley et al. (1998) and listed in column 3) of
242: Table 1. The reader interested in a detailed discussion concerning the
243: metallicity measurements and the metallicity scale of RR Lyrae stars is
244: referred to Dall'Ora et al. (2004) and to Gratton et al. (2004).
245: Data plotted in Fig. 2 indicate that empirical radius estimates are affected by
246: large scatter. However, theory and observations are, within current uncertainties,
247: in reasonable agreement for periods longer than 0.42 days. The radius measurements
248: by C92 do not include individual error estimates, and therefore,
249: it is not clear whether the three objects with $P > 0.63$ days present a real
250: discrepancy. However, observed radii show a larger scatter when moving toward
251: shorter periods. The reason for this drift is not clear, however four shorter
252: period RR Lyrae present a well-defined bump along the decreasing branch and
253: a steep rising branch in both light and velocity curves (RS Boo, TW Her,
254: Jones et al. 1988b; V445 Oph, Fernley et al. 1990; W Crt, Skillen et al. 1993).
255: Moreover, DH Peg is a FO RR Lyrae.
256:
257:
258: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
259: \section{Discussion and final remarks}
260:
261: Recent improvements in optical and infrared interferometry have allowed for
262: direct evaluation of the accuracy of radii and distances determined
263: by BW analyses of pulsating stars, in particular Cepheid variables
264: (Nordgren et al. 2002; Lane et al. 2002; Kervella et al. 2004). Since
265: no RR Lyrae star currently has an angular diameter directly measured
266: by interferometry, such a comparison of the accuracy of our RR Lyrae
267: radii is not possible. However, one may compare the distances calculated
268: using our method as a test of the accuracy of our surface brightness
269: relation, provided there is a known distance to any of the stars in our sample.
270: Alone of the stars in Table 1, RR Lyr has a distance known from high precision
271: trigonometric parallax observations. Benedict et al. (2002, hereinafter B02)
272: used the Hubble Space Telescope Fine Guidance Sensor to obtain a parallax of
273: $\pi = 3.82 \pm 0.2$ mas yielding a distance of $262 \pm 14$ pc. As a check
274: of our surface brightness analysis we compare this relatively model independent
275: distance to the distance calculated in Table 1: $270 \pm 35$ pc (this uncertainty
276: is the quadrature sum of the listed random and
277: systematic uncertainties). Our distance is in excellent agreement with the HST
278: distance and with the distance obtained using the K-band Period-Luminosity-Metallicity
279: ($PLZ_{\rm K}$) relation ($260\pm 5$ pc) obtained by B03. This agreement gives us confidence in
280: the accuracy of surface brightness relations and their results. It should be noted that
281: no horizontal branch stars were included in the calibration of the surface brightness
282: relation used here (Nordgren et al. 2002). That RR Lyrae radii determined from this calibration
283: agree so well with the theoretically computed radii argues that the surface brightnesses of
284: horizontal branch stars may be well computed from calibrations based on giant stars.
285:
286: According to this evidence, we compared current distance determinations with distances
287: estimated using the RR Lyrae visual magnitude metallicity relation
288: ($M_V\; vs\; [Fe/H]$) provided by B02. Data plotted
289: in the top panel of Fig. 3, show that the relative difference is within an average of
290: 10\%. As expected, the discrepancy is significantly larger in the short-period range.
291: To constrain the intrinsic accuracy of current distances, the bottom panel
292: of Fig. 3 shows the relative difference with the distances based on the $PLZ_{\rm K}$
293: relation. A glance at the data plotted in this panel confirms the quoted result.
294:
295: Current findings suggest that predicted and observed radii of RR Lyrae stars are in
296: reasonable agreement. The accuracy of empirical estimates do not allow us to constrain
297: the plausibility of nonlinear, convective RR Lyrae models. Needless to say, that this
298: analysis shall be extended to the entire sample of cluster and field RR Lyrae stars
299: for which are available accurate spectroscopic and photometric (optical, NIR)
300: measurements. In the future, more precise trigonometric and pulsation parallaxes
301: together with new angular diameter measurements, will certainly improve the
302: observational scenario not only for radii and distances but also for the
303: pulsation factor $p$.
304:
305: It is a pleasure to thank an anonymous referee for his/her positive comments
306: and suggestions. This work was partially supported by PRIN~2003 and INAF~2003.
307:
308: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
309: \begin{references}
310: \reference{} Barnes, T. G. \& Evans, D. S. 1976, MNRAS, 174, 489
311: \reference{} Beers, T. C., Chiba, M., Yoshi, Y., Platais, I., Hanson, R. B.,
312: Fuchis, B., \& Rossi, R. 2000, AJ, 119, 2866
313: \reference{} Benedict, G. F. et al., 2002, AJ, 123, 473 (B02)
314: \reference{} Bono, G., Caputo, F., Castellani, V., \& Marconi, M. 1997, A\&AS, 121, 327
315: \reference{} Bono, G., Caputo, F., Castellani, V., Marconi, M., \& Storm, J. 2001,
316: MNRAS, 326, 1183
317: \reference{} Bono, G., Caputo, F., Castellani, V., Marconi, M., \& Storm, J. 2002,
318: MNRAS, 332, 78
319: \reference{} Bono, G., Caputo, F., Castellani, V., Marconi, M., Storm, J., \&
320: Degl'Innocenti, S. 2003, MNRAS, 344, 1097 (B03)
321: \reference{} Bono, G., Castellani, V., \& Marconi, M. 2000, ApJ, 532, 129
322: \reference{} Bono, G., \& Stellingwerf, R. F. 1994, ApJS, 193, 233
323: \reference{} Burki, G. \& Meylan, G. 1986, A\&A, 159, 261
324: \reference{} Cacciari, C. \& Clementini, G. 2003, in Stellar Candles for the
325: Extragalactic Distance Scale, ed. D. Alloin \& W. Gieren (Berlin: Springer \& Verlag), 105
326: \reference{} Cacciari, C., Clementini, G., Prevot, L. \& Buser, R. 1989, A\&A, 209, 141
327: \reference{} Caputo, F., Castellani, V., Marconi, M., \& Ripepi, V. 2000, MNRAS,
328: 316, 819
329: \reference{} Cardelli, J. A., Clayton, G. C., \& Mathis, J. S., 1989, ApJ, 345, 245
330:
331: \reference{} Carney, B. W., Storm, J., \& Jones, R. V. 1992, ApJ, 386, 663 (C92)
332: \reference{} Catelan, M., Pritzl, B. J., \& Smith, H. A. 2004, ApJS, 154, 633
333: \reference{} Clementini, G., Cacciari, C. \& Lindgren, H. 1990, A\&AS, 85, 865
334: \reference{} Dall'Ora, M., et al. 2004, ApJ, 610, 269
335: \reference{} Fernley, J. A. 1994, A\&A, 284, L16
336: \reference{} Fernley, J. A., Skillen, I., Jameson, R. F., Barnes, T. G.,
337: Kilkenny, D., \& Hill, G. 1990, MNRAS, 247, 287
338: \reference{} Fernley, J. et al. 1998, A\&A, 330, 515
339: \reference{} Fouqu\'{e}, P. \& Gieren, W. P. 1997, A\&A, 320, 799
340: \reference{} Gratton, R. G., Bragaglia, A., Clementini, G., Carretta, E.,
341: Di Fabrizio, L., Maio, M., \& Taribello, E. 2004, A\&A, 421, 937
342: \reference{} Jones, R. V., Carney, B. W., Storm, J., \& Latham, D. W. 1992, ApJ, 386, 646
343: \reference{} Jones, R. V., Carney, B. W., \& Latham, D. W., 1988a, ApJ, 326, 312
344: \reference{} Jones, R. V., Carney, B. W., \& Latham, D. W., 1988b, ApJ, 332, 206
345: \reference{} Kervella, P., Bersier, D., Mourard, D., Nardetto, N., Fouqu{\' e},
346: \& P., \& Coud{\' e} du Foresto, V., 2004, A\&A, 428, 587
347: \reference{} Lane, B. F., Creech-Eakmon, M. J., \& Nordgren, T. E., 2002, ApJ, 573, 330
348: \reference{} Manduca, A. \& Bell, R. A., 1981, ApJ, 250, 306
349: \reference{} Nordgren, T. E., Lane, B. F., Hindsley, R.~B. \& Kervella, P. 2002, AJ, 123, 3380
350: \reference{} Skillen, I., Fernley, J. A., Stobie, R. S., \& Jameson, R. F., 1993a, MNRAS, 265, 301
351: \reference{} Skillen, I., Fernley, J. A., Stobie, R. S., Marang, F., \& Jameson, R. F., 1993b, SAAOC, 15, 90
352: \reference{} Storm, J., Carney, B. W., \& Latham, D. W. 1994a, A\&A, 290, 443
353: \reference{} Storm, J., Nordstr\"om, B., Carney, B. W., \& Andersen, J. 1994b, A\&A, 291, 121
354: \reference{} Walker, A. R. 2003, in Stellar Candles for the Extragalactic Distance
355: Scale, ed. D. Alloin \& W. Gieren (Berlin: Springer \& Verlag), 265
356: \reference{} Wilson, R. E. 1953, ``General Catalogue of Stellar Radial Velocities'',
357: Carnegie Inst. Wash. D.C. Pub., 601
358:
359: \end{references}
360:
361:
362: \clearpage
363: %table 1
364: \begin{deluxetable}{lcccrrrrl}
365: \scriptsize
366: \tablecolumns{9}
367: \tablewidth{0pt}
368: \tablecaption{Selected Baade-Wesselink RR Lyrae}
369: \tablehead{
370: \colhead{Star} & \colhead{Period} & \colhead{[Fe/H]} & \colhead{E(B-V)} & \colhead{$V^\ast$\tablenotemark{a}} &
371: \colhead{Phase} & \colhead{Distance} & \colhead{Radius} & \colhead{Ref.}\\
372: \colhead{} & \colhead{(days)} & \colhead{} & \colhead{} & \colhead{} &
373: \colhead{} & \colhead{(pc)} & \colhead{(R$_\odot$)} & \colhead{}
374: }
375: \startdata
376: DH Peg\tablenotemark{b}&0.2555&-1.24 & 0.08 & -71 $\pm$ 1& 0.0-0.8 & 470 $\pm$ 60(50) & 3.8 $\pm$ 0.5(0.5) & e \\
377: RS Boo & 0.3773 & -0.36 & 0.02 & -4 $\pm$ 2 & 0.35-0.8 & 770 $\pm$ 20(120) & 4.1 $\pm$ 0.1(0.7) & g \\
378: TW Her & 0.3996 & -0.69 & 0.05 & -5 $\pm$ 2 & 0.1-0.9 &1150 $\pm$ 35(100) & 4.2 $\pm$ 0.1(0.5) & g \\
379: V445 Oph& 0.4000 & -0.19 & 0.27 & -18 $\pm$ 1 & 0.0-1.0 & 700 $\pm$ 15(30) & 4.4 $\pm$ 0.1(0.3) & h \\
380: W Crt & 0.4120 & -0.54 & 0.05 & 59 $\pm$ 1 & 0.35-0.8 &1170 $\pm$ 25(90) & 3.7 $\pm$ 0.1(0.4) & c,d \\
381: UU Vir & 0.4756 & -0.87 & 0.03 & -8 $\pm$ 1 & 0.1-0.9 & 880 $\pm$ 20(40) & 4.6 $\pm$ 0.2(0.4) & g \\
382: BB Pup & 0.4805 & -0.64 & 0.10 & 130 $\pm$ 1 & 0.0-1.0 &1520 $\pm$ 70(30) & 4.3 $\pm$ 0.2(0.1) & c,d \\
383: RR Lyr & 0.5668 & -1.39 & 0.07 & -72 $\pm$ 1 & 0.0-1.0 & 270 $\pm$ 25(25) & 5.2 $\pm$ 0.5(0.5) & i,f \\
384: RV Oct & 0.5711 & -1.71 & 0.13 & 136 $\pm$ 1 & 0.0-0.85 & 960 $\pm$ 20(40) & 5.3 $\pm$ 0.2(0.2) & c,d \\
385: WY Ant & 0.5743 & -1.48 & 0.05 & 204 $\pm$ 1 & 0.2-0.9 &1120 $\pm$ 35(60) & 5.7 $\pm$ 0.2(0.4) & c,d \\
386: SS Leo & 0.6263 & -1.50 & 0.01 & 161 $\pm$ 1 & 0.0-1.0 &1620 $\pm$ 40(70) & 7.2 $\pm$ 0.2(0.4) & h \\
387: \enddata
388: \tablenotetext{a}{Mean radial velocity (kms$^{-1}$).
389: \hspace*{2.5mm} $^b$ First overtone pulsator.
390: References:
391: (c) Skillen et al. (1993a);
392: (d) Skillen et al. (1993b);
393: (e) Jones, Carney \& Latham (1988a);
394: (f) Manduca \& Bell (1981);
395: (g) Jones, Carney \& Latham (1988b);
396: (h) Fernley et al. (1990);
397: (i) Wilson (1953).
398: }
399: \end{deluxetable}
400:
401: \clearpage
402: %fig1
403: \begin{figure}
404: %\epsscale{0.90}
405: \plotone{f1.ps}
406: \caption{Predicted radii for fundamental RR Lyrae models as a function of the
407: logarithmic period. Solid line shows the predicted linear regression over the
408: entire set of models, while the dashed one the empirical PR relation for
409: Type II Cepheids derived by Burki \& Meylan (1986).
410: The vertical error bar plotted in the left top corner shows the large
411: intrinsic dispersion of predicted radii.\label{fig1}}
412: \end{figure}
413:
414: %fig2
415: \begin{figure}
416: \plotone{f2.ps}
417: \caption{Period-Radius-Metallicity relation for first overtone (top) and
418: fundamental (bottom) RR Lyrae models projected onto a two-dimensional plane.
419: Filled and open circles display predicted and new radius estimates, while
420: triangles the radius determinations by Carney et al. (1992). Empirical radii
421: have been plotted assuming metal abundances by Fernley et al. (1998) and
422: $Z_\odot=0.02$. Individual error bars account for both random and systematic
423: errors. \label{fig2}}
424: \end{figure}
425:
426: %fig3
427: \begin{figure}
428: \plotone{f3.ps}
429: \caption{Top panel - Relative difference between current distances and distances
430: estimated using the calibration of the $M_V vs [Fe/H]$ relation provided by
431: B02. Individual error bars account for both random and
432: systematic errors. The period of DH Peg (open circle) was fundamentalized.
433: Bottom panel - Same as the top, but with distances estimated using the
434: $PLZ_{\rm K}$ relation provided by B03.\label{fig3}}
435: \end{figure}
436:
437: \end{document}
438:
439: