1: %% version proofed AA/2007/7702 12/07/07
2: %\documentclass{../../../../Dir_ARTICLES/DirAA/aa}
3: %\documentclass{../DirAA/aa}
4: %\documentclass[referee]{../DirAA/aa}
5: \documentclass{aa}
6: %\usepackage{graphicx}
7: \usepackage{epsfig,amsmath,amssymb}
8:
9: \def\fconv{F_{\rm conv}}
10: \def\msol{M_\odot}
11: \def\lsol{L_\odot}
12: \def\rjup{R_{\rm Jup}}
13: \def\lbol{L_{\rm bol}}
14: \def\te{T_{\rm eff}}
15: \def\vrot{{\rm v}_{rot}}
16: \def\vconv{{\rm v}_{conv}}
17: \def\tconv{{\rm t}_{conv}}
18: \def\hp{{H}_{P}}
19: \def\kms{\,{\rm km}\,{\rm s}^{-1}}
20: \def\cms{\,{\rm cm}\,{\rm s}^{-1}}
21: \def\cm2s{\,{\rm cm}^2\,{\rm s}^{-1}}
22: \def\dyn{\,{\rm dyn}\,{\rm cm}^{-2}}
23:
24: \def\ga{\,\hbox{\hbox{$ > $}\kern -0.8em \lower 1.0ex\hbox{$\sim$}}\,}
25: \def\la{\,\hbox{\hbox{$ < $}\kern -0.8em \lower 1.0ex\hbox{$\sim$}}\,}
26:
27:
28: \begin{document}
29:
30: \title{Evolution of low-mass star and brown dwarf eclipsing binaries}
31:
32: \author{Gilles Chabrier, Jos\'e Gallardo, Isabelle Baraffe }
33:
34: \institute{Ecole Normale Sup\'erieure de Lyon, CRAL (UMR CNRS 5574), Universit\'e de Lyon, France}
35:
36: \date{Received/Accepted}
37:
38: \titlerunning{Evolution of low-mass star and brown dwarf eclipsing binaries}
39: \authorrunning{Chabrier et al. }
40: \abstract{
41: We examine the evolution of low-mass star and brown dwarf eclipsing binaries. These objects are rapid
42: rotators and are believed to shelter large magnetic fields.}
43: { We suggest that reduced convective efficiency, due to fast rotation and large field
44: strengths, and/or to magnetic spot coverage of the radiating surface
45: significantly affect their evolution,
46: leading to a reduced heat flux and thus larger radii and cooler effective temperatures than for regular objects.}{We have considered such processes in our evolutionary calculations, using a phenomenological approach.}
47: {This yields mass-radius and effective temperature-radius
48: relationships in agreement with the observations. We also reproduce the effective temperature ratio
49: and the radii of the two components of
50: the recently discovered puzzling eclipsing brown dwarf system.}
51: {These calculations show that fast rotation and/or magnetic activity may significantly affect the evolution of eclipsing binaries and
52: that the mechanical and thermal properties of these objects depart from the ones of
53: non-active low-mass objects. We find that, for internal field strengths compatible with the
54: observed surface value of a few kiloGauss, convection can be severely inhibited.
55: The onset of a central radiative zone for rapidly rotating active low-mass stars
56: might thus occur below the usual $\sim 0.35\,\msol$ limit. }
57:
58: \keywords{}
59:
60: \maketitle
61:
62: \section{Introduction}
63:
64: Low-mass stars (LMS), i.e., M-type stars in the present context, represent an overwhelming fraction of the Galactic stellar population (\cite{Chabrier03}).
65: Observational determination of their
66: mass-radius relationship provides a
67: stringent testing of the theoretical description of their structure and evolution. Although observations of this
68: relationship agree well with the
69: theoretical predictions in a large number of cases (\cite{Segransan03}; \cite{CBAH05}), the radius determinations of eclipsing binaries (EBs) depart significantly ($\sim$10-15\%)
70: from these predictions
71: (\cite{Torres}; \cite{CBAH05}).
72: These discrepancies are unlikely to be due to inaccurate equation of state or opacities, whose treatment
73: is relatively well mastered for $m\ga 0.4\msol$ and $\te\ga 4000$ K (Chabrier \& Baraffe 2000, CB00).
74: Furthermore, active LMS are observed to be redder, i.e. cooler, than the
75: other LMS (\cite{Hawley96}). These particular behaviours pose a challenge to theorists. On the other hand, the majority of LMS, notably
76: EBs, are fast rotators and exhibit strong persistent
77: coronal (X-ray) and chromospheric (H$_\alpha$) activity or large
78: flares (\cite{Gizis02}), indicating the presence of a strong magnetic field (\cite{Donati06}). In this Letter, we suggest that the discrepancy between the observed mass-radius relationship of EBs
79: and of other LMS does not stem from inaccurate equation-of-state or opacity problems, but is due
80: to a rotation or magnetic field induced reduction of the efficiency of large-scale thermal convection in their interior, leading to less efficient heat transport. We show
81: that the reduction of the star's radiating surface due to magnetic spot coverage also yields a
82: smaller $\te$ and a larger radius.
83:
84:
85: \section{Effect of rotation and magnetic field}
86: The EBs and, more generally, active LMS are fast rotators
87: with rotation periods of
88: $P\la 3$ days and rotation velocities of $\vrot \ga 10\kms$ (\cite{Delfosse98}; \cite{Reid02,MB03}). Typical convective time scales in their interior are $\tconv$$\sim$${H_P \over \vconv}$$\sim$${10^9\over (10^2-10^3)}$$\sim$$10^6$-$10^7$ s, where $H_P$$\sim$$ R_\star$
89: is the pressure scale height (\cite{CB00}). This yields Rossby numbers
90: $Ro$=${P\over \tconv}\la 10^{-2} $. At a very small Rossby number, the Proudman-Taylor theorem enforces fluid motions to columnar or sheet-like structures
91: with a characteristic length scale perpendicular to the rotation axis that is much smaller than the one
92: along the rotation axis (of the order of $R_\star$).
93: Although the conditions of this theorem do not exactly apply for spherical objects
94: with finite viscosity,
95: rotation-dominated convective motions are severely affected by rotation,
96: leading to highly anisotropic patterns with
97: large-scale fluid motions being confined to columns
98: along the rotation axis and motions in the other directions being strongly inhibited
99: (\cite{ZhangJones97}).
100: As a whole, this reduces the characteristic length scale along these directions and the mean convective velocity amplitudes,
101: thus the efficiency of thermal convection to transport the internal heat flux.
102:
103: On the other hand, fully ionised interiors of LMS are excellent electrical conductors.
104: The interaction between the magnetic field and the fluid motions induces an electrical current and a
105: Lorentz force, which in turn affects the motions.
106: In the presence of a magnetic field,
107: the anisotropy of rotation-dominated convection is reduced and the large-scale flows reach the so-called
108: MAC (Magnetic, Archimedean, Coriolis) balance between the buoyancy, Coriolis, and Lorentz forces (\cite{StarchenkoJones}).
109: Equipartition
110: between the buoyancy force, $\sim \rho g\delta (\Delta T/T)$ (where $\delta$=$-({\partial \ln \rho \over \partial \ln T})_P$ and $\Delta T$
111: is the temperature excess over a pressure scale height), and the Coriolis force, $\sim 2\rho\Omega \vconv$
112: (where $\Omega={\vrot\over R_\star}$
113: is the star angular velocity), yields
114:
115: \begin{eqnarray}
116: v_{conv} \approx {g\delta \Delta T\over T \Omega}
117: \label{vconv}
118: \end{eqnarray}
119:
120: \noindent for the typical {\it average} large-scale velocity. For $\Omega \sim 10^{-4}$ s$^{-1}$, typical values of the
121: various quantities over the LMS domain give $\vconv \sim$10-100$\cms$, about
122: a factor of 10 lower than the predictions of the usual mixing length theory.
123:
124: There is presently no clear understanding of the
125: interaction between convection and magnetic field under stellar interior conditions.
126: We can, however, estimate the conditions for generation of a dynamo
127: in LMS or BD interiors. The microscopic magnetic diffusivity of metallic hydrogen is $\eta \approx 10^2$-$10^3$ $\cm2s$. According to Ohm's law and Maxwell's equations (hydromagnetic induction equation), a magnetic field will decay unless a velocity field can counteract or balance the diffusive effects.
128: For typical values of the convective velocity (see above), the characteristic magnetic Reynolds
129: number over a star-size conducting region is
130: $R_m=v_{conv} \,R_\star/\eta \gg 100$. According to dynamo theory, $R_m$ in LMS and BD interiors is
131: thus large enough for dynamo to occur, providing both rotation and convection are present.
132: Once the criterion for dynamo onset is satisfied, the field will grow and is supposed to equilibrate when
133: the Lorentz and the Coriolis forces become comparable (Elsasser number of order unity), reaching the aforementioned MAC balance. This yields an amplitude for the internal field,
134: $B_{eq} \approx (8\pi \bar{\rho} \eta \Omega)^{1/2}\ga 10\,{\rm G}$, for fast rotators ($\vrot \ga 10\kms$).
135:
136: At large $R_m$, i.e. in the dissipationless regime, however,
137: magnetic diffusion is mainly due to turbulent rather than
138: molecular diffusion, with $\eta \equiv \eta_t\sim l\,v_{conv}$\footnote{Observations
139: of sunspot decay indeed suggest that the solar surface diffusivity is many
140: orders of magnitude higher than the atomic value.}.
141: In that case,
142: non-linear saturation occurs when turbulence, enhancing the diffusive processes, is strong enough to reduce $R_m$ down to the critical value for dynamo action, $R_{m}\ga 50$. This yields typical magnetic length scales $l\la R_\star/50$, and thus $B_{eq}\approx {\rm a\, few}\,10^4$ G for LMS average conditions,
143: in good agreement with {\it surface} field determinations of a few kG (\cite{Donati06}; \cite{ReinersBasri}).
144: This corresponds to Alfven velocities $v_A={B\over \sqrt{4\pi \rho}} \ga \vconv$,
145: so that for such strong fields the Lorentz force will impede the convection by reducing the flow speed.
146: The main effect of a strong magnetic field is to inhibit motions across it in
147: comparison with motions along it.
148: An ideally conducting fluid is tied to the fluid lines.
149: In a fluid of
150: finite conductivity, motion across the field is possible at a rate governed by the conductivity.
151: In a highly conductive medium with a strong magnetic field, the motion will be slow. Stevenson's (1979)
152: stability analysis shows that the combination of fast rotation ($Ro\la 0.1$) and a magnetic field
153: with finite diffusivity {\it enhances} convection, because of the reduction by the Lorentz force of the flow anisotropy
154: due to the Proudman-Taylor constraint. Stevenson's approach, however, applies to planar
155: geometry, i.e. thin convection zones, and to uniform density and magnetic fields and is likely to break down for large (star-size) convective zones and strong fields. Magneto-convection 3D simulations indeed show that the magnetic field inhibits the magnitude of the velocity fluctuations and reduces the heat flux (\cite{Stein92}).
156: On the basis of the aforementioned
157: field strength values, it seems unavoidable to suppose that, even though
158: the magnetic field will not necessarily stabilise
159: the fluid against convection in a fluid of finite electrical resistivity, it will
160: cause a serious reduction of convective efficiency.
161:
162: These estimates, on the other hand, show that the
163: magnetic pressure in LMS or BD interiors, ${B^2\over 8\pi}\la 10^7\dyn$, is orders of magnitude lower than the gas pressure, $P_{g}\ga 10^{12}$-$10^{16}\dyn$ (\cite{CB00}), and it can be safely ignored in the internal structure equations.
164: Indeed, if the fluid is convectively unstable, the ratio of the magnetic pressure inside the flux tube over the surrounding mean gas pressure is expected to be of the order of the superadiabaticity,
165: ${(B^2/8\pi)\over P_{g}}\la (\nabla - \nabla_{ad})\la 10^{-7}$ (\cite{GoughTayler66}; \cite{Meyer}).
166: An estimate for complete inhibition of convection is obtained when the Lorentz force is strong enough to balance the buoyancy force:
167:
168: \begin{eqnarray}
169: {B^2\over 4\pi l}\ga \rho g \delta (\nabla - \nabla_{ad})\, ,
170: \label{crit}
171: \end{eqnarray}
172:
173: \noindent where $l\ll R_\star$ is the aforementioned characteristic magnetic length scale in a dynamo
174: turbulent medium. This yields field amplitudes of the order of $10^4$ G, comparable to the value of $B_{eq}$. Magnetic fields can thus in principle severely inhibit convection in the interior of
175: active LMS and BDs.
176: The criterion (\ref{crit}) for
177: stability against convection in the presence of a magnetic field is similar to the one derived by
178: Stevenson in the dissipationless regime and by
179: Gough \& Tayler (their Eq.(1.2)),
180: except for the reducing
181: factor $\sim l/R_\star$ for the gas pressure term (assuming $\nabla P_g\sim P_g/R_\star$).
182: The Gough and Tayler approach, however, is primarily devoted to the study of surface spots, where
183: the magnetic and gas pressures are comparable.
184: Applying this criterion to LMS interiors yields
185: field strengths $\ga 10^7$ G for the inhibition of convection in the core of a 0.3 $\msol$ star (\cite{Mullan}).
186: It seems rather difficult to generate such strong fields. Together, near-equipartition (within a factor $\sim$10) between turbulent and
187: magnetic energy, $B^2/8\pi \sim \rho v^2_{conv}$, and
188: the fact that
189: $L(r)\propto 4\pi r^2(\rho \vconv^3)$ is a slowly varying quantity, yield
190: an amplification factor $\sim$10-100 from the surface to the central regions.
191: A full field of several megaGauss would thus be in super-equipartition
192: and, if confined to the interior, would be unstable (Elsasser numbers $\gg 1$) (\cite{Tayler73}; \cite{MarkeyTayler73}).
193: Therefore, it does not seem realistic
194: to apply the Gough-Tayler criterion to the entire stellar structure, in particular for uniformly dense objects like LMS and BDs.
195:
196:
197: A real picture of a magnetised, convectively unstable medium is cooling flows along the magnetic flux tubes passing through the convective medium.
198: Part of the thermal flux, confined to the tubes, is thus carried by diffusion. This contraint on the flow patterns leads to a substantial reduction in the transport of energy and reduces the maximum possible heat flux.
199: Given the absence of a proper treatment of heat transport in a magnetised medium for LMS conditions, which requires multidimensional simulations over a characteristic convective length scale,
200: and given the exploratory nature of this Letter, we decided to stick to a minimalist approach, focusing on the reduced efficiency of global thermal
201: convection due to the presence of strong rotation and/or magnetic fields in EBs.
202: Such an approach is based on the phenomenological representation of this network of diffusive tubes surrounded by impeded field-free convective regions as a global convective system - since the large Reynolds numbers illustrate the overwhelming importance
203: of macroscopic motions over microscopic diffusive processes - with reduced efficiency. In the framework of the standard MLT
204: formalism, this translates into a mixing length parameter $\alpha =l/\hp < 1$.
205:
206: \section{Effect of spot coverage}
207:
208: The strong chromospheric and coronal emission in LMS of spectral types $\ga $M2-M3,
209: with $L_{{\rm H}_\alpha}$ and $L_X$ about 100 times
210: the solar value (\cite{Gizis02}; \cite{MB03}), can be associated with an average large fraction of the radiating surface being covered with magnetic
211: spots. Cool spots are the illustration of the inhibition of energy transport by convective motions
212: (buoyancy) in a (rotating) highly conducting medium, but they also illustrate the fact that the field is not strong enough to suppress convection completely over the entire structure. This brings some justification to our admitedly simplistic approach where the aim is to explore the effect of reduced convective efficiency in EBs.
213: We denote $\beta=S_s/S_\star$ as the (time-averaged) fraction of the stellar surface
214: covered by cool spots,
215: ${\mathcal F_s}$ the {\it total} flux emerging from the spots, and ${\mathcal F_\star}=\sigma {\te}^4_\star$ the one associated with a spot-free star of effective temperature $\te$.
216: Cool spots imply ${\mathcal F_s} < \beta {\mathcal F_\star}$. The total flux is
217:
218: \begin{eqnarray}
219: {\mathcal F}&=&(1- \beta) {\mathcal F_\star} + {\mathcal F_s}\, < \,{\mathcal F_\star}.
220: \end{eqnarray}
221:
222: \noindent Cool spot coverage
223: thus yields an effective temperature $\te <{\te}_\star $,
224: a consequence of the reduced heat flow reaching the surface (\cite{Cowling76}; \cite{Stein92}).
225: Given the presently undetermined spot temperature for LMS and the exploratory nature of the present calculations, we simply assumed black (zero-temperature) spots. The effect of spot coverage for a given value of $\beta$ is thus an upper limit.
226:
227: \section{Results}
228:
229: \begin{figure}
230: \centering
231: \includegraphics[width=8cm]{f1.eps}
232: \caption{Mass-radius relationships for 1 Gyr old LMS for various values of the mixing length parameter
233: $\alpha$=$l/\hp$ ((a) upper panel) and of the fractional surface area $\beta$ covered with cool spots
234: ((b) lower panel). Inset upper panel: dash-dot line : $\alpha=2$, $10^8$ yr. Solid thick line: standard case, no spot coverage.
235: Low-mass eclipsing binary observational determinations are displayed with their $1\sigma$ error bars.
236: }
237: \label{fig-mR}
238: \end{figure}
239:
240: \subsection{Effect of reduced convective efficiency}
241:
242: Figure \ref{fig-mR}(a) portrays the $m$-$R$ relationship for LMS at 1 Gyr\footnote{Objects with
243: $m\ge 0.09\msol$ are on the ZAMS at this age (\cite{CB00}).} for various values of
244: the mixing length parameter
245: $\alpha$. Observed radii of EBs are displayed as well. Objects below $\sim 0.35 \msol$ are fully convective (\cite{CB97}); convection is nearly adiabatic, so changing $\alpha$ for these objects has a modest impact.
246: Since superadiabaticity increases with mass, in particular in the outermost regions
247: of the star, reducing the convective efficiency, i.e. decreasing $\alpha$, for higher masses leads to an
248: increasingly greater effect. The reduced convective flux requires a larger fraction of the heat to
249: be transported by radiation in the outermost regions, yielding a steeper outer thermal gradient
250: and thus a cooler $\te$. The immediate consequence is to decrease the luminosity and thus the
251: central temperatures, i.e. the nuclear energy
252: production needed to maintain thermal equilibrium. This in turn yields
253: an expansion of the star, i.e. a larger radius.
254: For BDs, the reduced heat flux implies a slower contraction rate.
255:
256: We have also explored the fate of a 0.3 $\msol$ star when drastically reducing the convective flux
257: in the interior ($\alpha = 0.05$) as a result of condition (\ref{crit}). The star indeed develops a stable inner radiative zone ($\nabla_{rad}<\nabla_{ad}$) over $\sim 20\%$ of its mass, yielding a
258: $\sim 7\%$ larger radius and a cooler $\te$.
259:
260: \subsection{Effect of spot coverage}
261:
262: Figure \ref{fig-mR}(b) illustrates the effect of surface spot coverage. As mentioned above,
263: spot coverage yields a smaller heat flux output, i.e. cooler $\te$ and thus larger radii (since $L$
264: is unaffected) at a given age compared with spot-free objects. A spot
265: coverage fraction $\beta \sim 30$-50\%, for a value of $\alpha =1$, can by itself reproduce
266: most of the observed EB radii within 1$\sigma$.
267:
268: \begin{figure}
269: \centering
270: \includegraphics[width=6cm]{f2.eps}
271: \caption{Effective temperature vs radius at 1 Gyr for various values of $\alpha$ and $\beta$. Solid thick line: standard case, no spot coverage.
272: }
273: \label{fig-TR}
274: \end{figure}
275:
276: \begin{figure}
277: \centering
278: \includegraphics[width=6cm]{f3.eps}
279: \caption{Same as Fig. 1 for the eclipsing BD system (Stassun et al. 2006) for different values of $\alpha$ and $\beta$.
280: }
281: \label{fig-BD}
282: \end{figure}
283:
284:
285: Figure \ref{fig-TR} displays the effective temperature as a function of the radius for low-mass EBs for different values of
286: $\alpha$ and $\beta$. Both reduced convective efficiency and spot coverage yield larger radii and cooler $\te$ compared with regular objects and provide a simple explanation for
287: the particular mechanical and thermal properties of EBs.
288: Figure \ref{fig-BD} portrays $m$-$R$ relations for the recently discovered
289: eclipsing BD system (\cite{Stassun06}). A solution
290: with $\alpha=0.5$, $\beta=0.5$ for the most massive object ($m_1$=$0.054\msol$), and
291: $\alpha$=$1$, $\beta$=$0.2$ for the least massive one ($m_2$=$0.034\msol$) yields radius values within the error bars and a temperature {\it ratio} ${{\te}_2/ {\te}_1}$=${2440\,{\rm K}/ 2320\,{\rm K}}$=$1.05$, in agreement with the observational determination, for the age of the system $\sim 10^6$ yr\footnote{The larger $\te$ determinations obtained by Stassun et al. (2006) with the usual Sp-$\te$-colour and Sp-gravity relations suggest that these relations cannot be applied to active objects and to stellar surfaces exhibiting magnetic spots.}. This possible solution suggests that the evolution of the most massive BD has been
292: significantly affected by the presence of a strong magnetic field, whereas the least massive BD has
293: been less affected.
294: Indeed, BDs with spectral type $\ga$M8-M9, i.e.
295: $\te \la 2400$ K, are rapid rotators and are believed to shelter large-scale magnetic fields in spite of showing no sign of persistent activity (\cite{MB03}; \cite{CK06}).
296: The reason invoked for this is
297: the low electrical conductivity and thus the lack of substantial current generation in their atmosphere (\cite{Mohanty02}; \cite{ReinersBasri}).
298: Spot coverage for cool ($\te \la 2400$ K) BDs is thus expected to be small, even in the presence of a
299: strong magnetic field. A possible, although speculative, explanation for the
300: BD system of interest is that the most massive object was originally hot enough (its $\te$ for $\alpha$=$2$, $\beta$=$0$ is $\te=2800$ K) for significant spot coverage to occur,
301: whereas the least massive BD remained too cool for atmospheric coupling between the magnetic field
302: and the gas to be significant, yielding less spot coverage.
303:
304: \section{Conclusion}
305:
306: In this paper, we have examined the consequences of (i) inhibiting convection due to rotation and/or internal
307: magnetic field and (ii) the presence of surface magnetic spot coverage, on the evolution of LMS and BDs. We have focused on the
308: particular case of EBs. Our approach is phenomenological and the present paper makes no
309: claim to present a consistent description of the effect of a large magnetic field on the evolution of
310: dominantly convective objects. A proper approach to such a complex process for LMS
311: interior conditions requires presently unavailable
312: numerical tools. Our calculations show that rotation or magnetic field induced inhibition
313: of convection leads to a reduced heat flow and thus significantly larger radii
314: and cooler $\te$ than for regular objects.
315: Spot coverage of the stellar radiating surface by itself has a major impact on the
316: evolution. Either one or a combination of these effects explains the observed mechanical and thermal
317: properties of EBs. Work is under way to explore the spot modulation in EBs and to derive
318: observational values of the spot coverage fraction $\beta$ for M-stars (Morales et al., in preparation).
319: A spot coverage fraction $\beta \sim 30$-50\%, however, is consistent with observations of rapidly rotating, more massive active stars (\cite{Jeffers05}).
320: These calculations strongly suggest that the evolution of EBs, or of very magnetically active
321: LMS and BDs, differs noticeably from the one of objects for which
322: rotational or magnetic effects are negligible, a fact supported by observational analysis (\cite{Morales}). This
323: should be taken into consideration when comparing observations with theoretical models.
324: Global averaging of the mass-radius relationship for LMS
325: is thus incorrect and highly misleading.
326: We show that the
327: puzzling eclipsing brown dwarf system that was recently discovered, with the more massive companion being
328: the cooler one, can be explained if the evolution of the most massive
329: object has been strongly affected by magnetic effects.
330:
331: We suggest that
332: internal field strengths consistent with the observed surface value, about a few kG,
333: might be sufficient for severely inhibiting convection in parts of LMS interiors, pushing the limit
334: for the onset of an inner radiative zone
335: below $\sim 0.35\msol$, with important consequences for the field geometry (\cite{CK06}; \cite{Donati06}).
336:
337: \begin{thebibliography}{}
338: \bibitem[Chabrier 2003]{Chabrier03} Chabrier, G., 2003, PASP, 2003, 115, 763
339: \bibitem[Chabrier \& Baraffe 1997]{CB97} Chabrier, G. \& Baraffe, I., 1997, \aap, 327, 1039
340: \bibitem[CB00]{CB00} Chabrier, G. \& Baraffe, I., 2000, ARA$\&$A, 38, 337, CB00
341: \bibitem[Chabrier et al. 2005]{CBAH05} Chabrier, G., Baraffe, I., Allard, F. \& Hauschildt, P., 2005, in {\it Resolved Stellar Populations}, Ed. Valls-Gabaud, D., astro-ph/0509798
342: \bibitem[Chabrier \& K\"uker 2006]{CK06} Chabrier, G. \& K\"uker, M., 2006, \aap, 446, 1027
343: \bibitem[Cowling 1976]{Cowling76} Cowling, T.G., 1976, \mnras, 177, 409
344: \bibitem[Delfosse et al. 1998]{Delfosse98} Delfosse, X., Forveille, T., Perrier, C. \& Mayor, M., 1998, \aap, 331, 581
345: \bibitem[Donati et al. 2006]{Donati06} Donati, J-F., Forveille, T., Cameron, A., Barnes, J. R., Delfosse, X., Moira, J. M. \& Valenti, J. A., 2006, Science, 311, 633
346: \bibitem[Gizis et al. 2002]{Gizis02} Gizis, J., Reid, I. \& Hawley, S., 2002, \aj, 123, 3356
347: \bibitem[Gough \& Tayler 1966]{GoughTayler66} Gough, D. \& Tayler, R., 1966, \mnras, 133, 85
348: \bibitem[Hawley et al. 1996]{Hawley96} Hawley, S. L., Gizis, J. E. \& Reid, N. I., 1996, \aj, 112, 2799
349: \bibitem[Jeffers et al. 2005]{Jeffers05} Jeffers, S.V., Cameron, A., Barnes, J.R, Aufdenberg, J. \& Hussain, G., 2005, \apj, 621, 425 \mnras
350: \bibitem[Markey \& Tayler 1973]{MarkeyTayler73} Markey, P. \& Tayler, R., 1973, \mnras, 163, 77
351: \bibitem[Meyer 1994]{Meyer} Meyer, F., 1994, in {\it Cosmical Magnetism}, Ed. D. Lynden-Bell, Kluwer
352: \bibitem[Mohanty et al. 2002]{Mohanty02} Mohanty, S., Basri, G., Shu, F., Allard, F. \& Chabrier, G., 2002 \apj, 571, 469
353: \bibitem[Mohanty \& Basri 2003]{MB03} Mohanty, S. \& Basri, G., 2003, \apj, 583, 451
354: \bibitem[Morales et al. 2007] {Morales} Morales, J.C., Ribas, I. \& Jordi, C., 2007, \aap, submitted
355: \bibitem[Mullan \& MacDonald 2001]{Mullan} Mullan, D. \& MacDonald, J., 2001, \apj, 559, 353
356: \bibitem[Reid et al. 2002]{Reid02} Reid, N. I., Gizis \& Hawley, S. L., 2002, \aj, 124, 2721
357: \bibitem[Reiners \& Basri 2007]{ReinersBasri} Reiners, \& Basri, G., 2007, \apj, 656, 1121
358: \bibitem[S\'egransan et al. 2003]{Segransan03} S\'egransan, D., Kervella, P., Forveille, T., Queloz, D., 2003, \aap, 397, L5
359: \bibitem[Starchenko \& Jones 2002]{StarchenkoJones} Starchenko, S.V. \& Jones, C.A.,, 2002, \icarus, 157, 426
360: \bibitem[Stassun et al. 2006]{Stassun06} Stassun, K.G., Mathieu, R.D. \& Valenti, J.A., 2006, Nature, 440, 311
361: \bibitem[Stein et al. 1992]{Stein92} Stein, R., Brandenburg, A. \& Nordlund, A., 1992, ASP. Conf. Ser., 26, 148
362: \bibitem[Stevenson 1979]{Stevenson79} Stevenson, D.J., 1979, Geophys. Astrophys. Fluid Dynamics, 12, 139
363: \bibitem[Tayler 1973]{Tayler73} Tayler, R., 1973, \mnras, 161, 365
364: \bibitem[Torres \& Ribas 2002]{Torres} Torres, G. \& Ribas, I., 2002, \apj, 567, 1140
365: \bibitem[Zhang \& Jones 1997]{ZhangJones97} Zhang, K., Jones, C.A., 1997, Geophys. Res. Lett., 24, 2869
366:
367: \end{thebibliography}
368:
369: \clearpage
370:
371: \begin{figure}
372: %\epsscale{2.0}
373: %\plotone{f1.eps}
374: \epsfxsize=180mm
375: \epsfysize=200mm
376: \epsffile{f1.eps}
377: %\caption{Mass-radius relationships for 1 Gyr old LMS for various values of the mixing length parameter $\alpha$=$l/\hp$ ((a) upper panel) and of the fractional surface area $\beta$ covered with cool spots ((b) lower panel). Inset upper panel: dash-dot line : $\alpha=2$, $10^8$ yr. Solid thick line: standard case, no spot coverage. Low-mass eclipsing binary observational determinations are displayed with their $1\sigma$ error bars.}
378: \end{figure}
379:
380: \clearpage
381:
382: \begin{figure}
383: %\epsscale{2.0}
384: %\plotone{f1.eps}
385: \epsfxsize=180mm
386: \epsfysize=200mm
387: \epsffile{f2.eps}
388: %\caption{Effective temperature vs radius at 1 Gyr for various values of $\alpha$ and $\beta$. Solid thick line: standard case, no spot coverage.}
389: \end{figure}
390:
391: \clearpage
392:
393: \begin{figure}
394: %\epsscale{2.0}
395: %\plotone{f1.eps}
396: \epsfxsize=180mm
397: \epsfysize=200mm
398: \epsffile{f3.eps}
399: %\caption{Same as Fig. 1 for the eclipsing BD system (Stassun et al. 2006) for different values of $\alpha$ and $\beta$.}
400: \end{figure}
401:
402: \end{document}
403:
404:
405: