1: %%
2: %% Beginning of file 'sample.tex'
3: %%
4: %% Modified 2004 January 9
5: %%
6: %% This is a sample manuscript marked up using the
7: %% AASTeX v5.x LaTeX 2e macros.
8:
9: \documentclass[12pt,preprint]{aastex}
10: %%\documentclass{emulateapj}
11: \newcommand{\vdag}{(v)^\dagger}
12: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
13:
14: %%\slugcomment{}
15:
16: \shorttitle{NSVS01031772}
17: \shortauthors{L\'opez-Morales et al.}
18:
19: \begin{document}
20:
21: \title{NSVS01031772: A New 0.50+0.54 $M_{\sun}$ Detached Eclipsing Binary}
22:
23: \author{Mercedes L\'opez-Morales\altaffilmark{1,7},
24: Jerome A. Orosz\altaffilmark{2},
25: J. Scott Shaw\altaffilmark{3,4},
26: Lauren Havelka\altaffilmark{2},
27: Maria Jes\'us Ar\'evalo\altaffilmark{5},
28: Travis McIntyre\altaffilmark{6}, Carlos L\'azaro\altaffilmark{5}}
29:
30: \email{mercedes@dtm.ciw.edu, orosz@sciences.sdsu.edu, jss@hal.physast.uga.edu, havelka@sciences.sdsu.edu, mam@iac.es, travis@clemson.edu, clh@iac.es}
31:
32:
33: \altaffiltext{1}{Carnegie Fellow.
34: Carnegie Institution of Washington, Department of Terrestrial
35: Magnetism, 5241 Broad Branch Rd. NW, Washington D.C., 20015, USA}
36: \altaffiltext{2}{Department of Astronomy,
37: San Diego State University, San Diego, CA, 92182, USA}
38: \altaffiltext{3}{Department of Physics and Astronomy,
39: University of Georgia, Athens, GA, 30602, USA}
40: \altaffiltext{4}{Southeastern Association for Research in Astronomy}
41: \altaffiltext{5}{Dept. de Astrof\'isica, Universidad de La Laguna;
42: Instituto de Astrof\'isica de Canarias, E38205, La Laguna, Tenerife, Spain}
43: \altaffiltext{6}{Department of Physics and
44: Astronomy, Clemson University, Clemson,
45: SC, 29634, USA}
46: \altaffiltext{7}{Visiting Astronomer,
47: Kitt Peak National Observatory, NOAO, which is operated by
48: the Association of Universities for Research in Astronomy, Inc.
49: (AURA) under cooperative agreement with the National Science Foundation.}
50:
51: \begin{abstract}
52: We report the discovery of a new detached eclipsing binary
53: known as NSVS01031772 with component masses
54: $M_1$= 0.5428 $\pm$ 0.0027$M_{\sun}$, $M_2$= 0.4982 $\pm$
55: 0.0025$M_{\sun}$, and radii $R_1$= 0.5260 $\pm$ 0.0028$R_{\sun}$,
56: $R_2$= 0.5088 $\pm$ 0.0030$R_{\sun}$. The system has an orbital period
57: of 0.3681414(3) days and an apparent magnitude V $\simeq$
58: 12.6. The estimated effective temperatures of the stars are $T_{{\rm
59: eff}_1} = 3615 \pm 72K$ and $T_{{\rm eff}_2} = 3513 \pm 31K$. The space velocities of
60: the system suggest that it is a main sequence binary and with a
61: metallicity that is approximately solar. The two stars in this binary
62: are located in a region of the Mass-Radius relation where no accurate
63: observational data was previously available. Similarly to other
64: low-mass binaries recently studied, the radius of each star in
65: NSVS01031772 exceeds the best evolutionary model predictions by about 8.5\% on average.
66: \end{abstract}
67:
68: \keywords{binaries: eclipsing --- binaries: spectroscopic --- stars:
69: fundamental parameters --- stars: late-type --- stars:
70: individual (NSVS01031772)}
71:
72: \section{Introduction} \label{sec:intro}
73:
74: Low-mass stellar structure models have improved fast in the past few
75: years as reviewed by Chabrier \& Baraffe (2000) and Chabrier et
76: al.~(2005). However, that improvement has not yet been matched by
77: observations. Accurate parameters of low-mass stars are difficult to
78: obtain, with the best source of precise data being double-lined
79: eclipsing binaries (DDEBs), but those systems are scarce. Only two
80: low-mass DDEBs had been found until 1998, CM Dra (Lacy 1977; Metcalfe
81: et al.~1996) and YY Gem (Leung \& Schneider 1978; Torres \& Ribas
82: 2002). A few more have been discovered since then, CU Cnc (Delfosse et
83: al.~1999; Ribas 2003), BW3 V38 (Maceroni \& Montalb\'an 2004),
84: TrES-HerO-07621 (Creevey et al.~2005), GU Boo (L\'opez-Morales \&
85: Ribas 2005),
86: 2MASS J05162281+2607387 (Bayless \& Orosz 2006),
87: UNSW-TR-2 (Young et al. 2006),
88: and a new binary found in the open cluster NGC1647 by
89: Hebb et al.~(2006). However, the number of low-mass DDEBs is still small,
90: and the error bars on the parameters of some of those systems are too large
91: to place rigorous constraints on the evolutionary models. There are also
92: portions of the Mass-Radius relation where no good observations exist
93: (e.g.~0.25--0.35 and 0.5--0.6$M_{\sun}$).
94:
95: Clearly, the sample size of well-studied low-mass binaries
96: needs to be increased.
97: Over the past two years we have searched for candidate low-mass binaries
98: in the Northern Sky Variability Survey (NSVS)
99: database (Wozniak et al. 2004). Our search algorithm uses two period-searching
100: methods: the ``string/rope'' method based on the Lafler-Kinman
101: statistic (Clarke 2002), and
102: the analysis of variance method (Schwarzenberg-Czerny 1989).
103: The first candidate system, known as NSVS01031772 (hereafter NSVS0103),
104: with J2000 coordinates of $\alpha=$~13:45:35, $\delta=+79$:23:48 (Figure 1),
105: was identified as a possible low-mass binary
106: based on its short orbital period
107: ($P\approx0.368$ days), the duration of its eclipses
108: ($\approx 16$\% of phase), and its red
109: near-infrared colors, obtained from the 2 Micron All Sky Survey (2MASS)
110: database
111: (Skrutskie et al. 2006).
112: In this work we present follow-up spectroscopic and
113: photometric observations which confirm the low-mass nature of NSVS0103.
114: We derive accurate fundamental parameters for the component stars
115: and compare our results to evolutionary models.
116:
117:
118: \section{Radial Velocity Curves} \label{sec:rv}
119:
120: We collected a total of 108 spectra
121: in two nights during May 2005 with the
122: echelle spectrograph at the 4-m Mayall telescope at the Kitt Peak
123: National Observatory (KPNO). The wavelength coverage of each spectrum
124: is 5700--8160~\AA, with a resolving power of 18,750 at 6000~\AA\ and
125: an average signal-to-noise ratio (S/N) of 7--10
126: per pixel. We also obtained a high S/N spectrum of the
127: M dwarf GJ 740 (M1.5V) for use as a template in the derivation of the
128: radial velocities. By using a real star as template we avoid the
129: problems that the low-mass stellar atmosphere models have reproducing
130: some spectral features of the stars.
131:
132: The radial velocities were extracted using the implementation of the
133: cross-correlation algorithm TODCOR (Zucker \& Mazeh 1994) kindly
134: supplied by Guillermo Torres. The analysis of NSVS01031 with TODCOR
135: was analogous to those described by Torres \& Ribas (2002) and
136: L\'opez-Morales \& Ribas (2005) in their analyses of YY Gem and GU Boo.
137: The template was rotationally broadened by 70 km~s$^{-1}$ to
138: optimize the results of TODCOR. The standard errors of the
139: resulting radial velocities are 10 km~s$^{-1}$ or less. We tested
140: for systematic effects in the derived velocities
141: and found no obvious ones.
142: The radial velocities for both stars phased
143: on the eclipse ephemeris derived below are shown in Figure~\ref{fig:rv}.
144: The data on that figure are available online. A sample of the contents of the
145: online table is shown in Table 1.
146:
147: \section{Light Curves}\label{sec:lc}
148:
149: We obtained complete $V$, $R$, and $I$-band light curves of NSVS0103
150: over 12 nights between March and May 2005, using the Apogee U55
151: 512x1152 CCD on the Southeastern Association for Research in Astronomy
152: (SARA) 0.9-m telescope at KPNO. The data were
153: reduced and analyzed using standard
154: aperture photometry packages in IRAF\footnote{IRAF is distributed by
155: the NOAO, which are operated by the Association of Universities for
156: Research in Astronomy, Inc., under cooperative agreement with the
157: NSF.}, with no differential extinction effects taken into account
158: given the relative small separation between the target and the
159: comparison and check stars in the field
160: (see Figure 1). The final light curves, which
161: contain 649 points in $V$, 843 points in $R$, and 1210 points
162: in $I$, are illustrated in Figure 3. The average photometric
163: precision per data point is 0.017 mags in $V$, 0.013 mags in $R$, and 0.007 mags in $I$.
164:
165: NSVS0103 was also observed on the nights of June 29 and 30, 2006
166: using the
167: 0.6m telescope at the Mount Laguna Observatory (MLO), equipped with an SBIG
168: CCD and $R$ and $I$ filters. IRAF was again used to apply the flat-fielding and
169: dark current corrections, and to derive differential light curves.
170: The MLO light curves, which contain
171: 366 points in $R$ and 364 points in $I$, are shown in Figure 4. The average photometric
172: precision per data point is in this case 0.009 mags in $R$ and 0.0011 mags in $I$. All the data shown in figures 3 and 4 are available online. Table 2 shows a sample of the contents of the online table.
173:
174: Times of minimum light derived from our photometry and from the NSVS
175: database are given in Table 3.
176:
177: From those times of minima we derive the following ephemeris equation
178: \begin{equation}
179: T_{Min I} = \mbox{HJD}\, 2,453,456.6796(2) + 0.3681414(3)E
180: \end{equation}
181: With the exception of one point, the typical ($O-C$) residuals are
182: less than one minute. There are no obvious trends.
183: Using the orbital
184: period above, we find an average phase difference between primary and
185: secondary minima of $\Delta\phi$ = 0.4989$\pm$0.0027,
186: which is consistent with a circular orbit.
187:
188:
189: \section{Analysis} \label{sec:anal}
190:
191: Each set of light curves was simultaneously modeled with the velocity curves
192: using the ELC code (Orosz \& Hauschildt 2000) and its various optimizers
193: using updated model atmospheres for
194: low-mass stars and brown dwarfs (Hauschildt, priv.~comm.).
195: The model has two main types of free parameters: those
196: related to the geometry of the binary such as the stellar masses,
197: radii, and separation, and those related to the radiative
198: properties of the stars such as their effective temperatures,
199: gravity and limb darkening, and locations of spots (if any).
200: For the geometrical parameters, we assumed a fixed orbital period
201: of 0.3681414 days, zero eccentricity, and synchronous rotation for
202: both stars. The free geometrical parameters
203: were the inclination $i$, the mass and radius of the primary,
204: $M_1$ and $R_1$, the ratio of the radii $R_2/R_1$,
205: and the $K$-velocity of the primary $K_1$. For a given inclination
206: $i$ and orbital period $P$, the orbital separation $a$ and
207: the mass ratio $Q=M_2/M_1$ can be found specifying $M_1$ and $K_1$.
208: The value of $a$ sets the absolute scale of the binary, while $Q$
209: gives a unique Roche geometry. We find that $M_1$ and $K_1$ are more
210: efficient to optimize
211: the solutions for well-detached binaries than $a$ and $Q$,
212: since $K_1$ can be inferred from the radial velocity
213: curve and $M_1$, roughly, from the spectral type. Once the scale
214: and the mass ratio are known, specifying $R_1$ and the ratio of the
215: radii
216: $R_2/R_1$ gives the specific values of the equipotentials.
217:
218: For the radiative properties of the stars, we eliminate the need for a
219: parameterized limb darkening law by using tabulated model atmosphere with
220: specific intensities over a wide range of temperatures and gravities.
221: The gravity darkening
222: exponents were set according to the temperatures of the stars following
223: Claret (2001). The stars in NSVS0103 are nearly spherical
224: so there is essentially no gravity darkening. We used
225: ``simple reflection'' (see Wilson 1990). The free radiative-properties
226: parameters were the temperature of
227: the primary $T_1$, the temperature ratio $T_2/T_1$, and
228: the parameters to describe two spots on the primary (this is the model that provided the best results in both the SARA and MLO datasets, see below).
229: The spots in ELC are specified by a temperature factor, the
230: longitude and latitude of the spot center,
231: and the angular radius of the spot, similarly to the Wilson-Devinney
232: code (1971). Finally, ELC has a phase-shift parameter to account for
233: small errors in the ephemeris. In total, we have 16 free parameters.
234:
235: We began by modelling the radial velocity curves
236: simultaneously with the SARA light curves since these light
237: curves have better sampling and more filters than the MLO light curves.
238: It became immediately clear that the model radial velocity curves did not
239: match the observed radial velocities near the conjunction
240: phases (e.g.\ during the primary and secondary eclipses).
241: During partial eclipse, the spectral line profiles may no longer
242: symmetric, which may lead to a difference between the measured
243: radial velocity and the actual radial velocity of the star.
244: ELC computes the change in radial velocity
245: (generally known as the Rossiter effect) using the technique outlined
246: in Wilson \& Sofia (1976), which accounts for the shift in the
247: ``center of light''. Since this simple technique may not fully mimic
248: the way TODCOR measures velocities during partial eclipse, we
249: did not include in the fit the primary and secondary radial velocities
250: between phases 0.425--0.575 and 0.925--1.075. In all, a total of 38 points
251: were excluded from each radial velocity curve, leaving 71 points per curve.
252: We note that since the mass constraints come mainly from observations
253: near the quadrature phases, excluding the points near the conjunction phases
254: will have very little if any effect on our results. The points not included
255: in the fit are represented as open symbols in figure 2.
256:
257: The SARA light curves are not symmetric about phase 0.5.
258: Usually asymmetries in the light curves are attributed to the presence
259: of spots on one or both stars.
260: We tried models with a single spot on
261: the primary and models with a single
262: spot on the secondary, but could not find an acceptable solution.
263: We then tried models with two spots on the primary, models
264: with two spots on the secondary,
265: and models with a spot on each star.
266: The best
267: solution was found with a model with two bright spots on the primary. That model is illustrated in the top diagram of Figure 5.
268: We run the ELC genetic optimizer
269: for 700 iterations, using the final solution to scale the error bars so
270: that $\chi^2=N-1$ for each data set.
271: The resulting mean errors per point were 0.017 mag in $V$, 0.013 mag in $R$,
272: and 0.006 mag in $I$.
273: After scaling the error bars, we ran the genetic
274: code and also a simple $grid$ $search$ optimizer to establish the $1\sigma$
275: errors on the fitted and derived parameters
276: using brute-force (see Orosz et al. 2002).
277: The results are summarized in the second column of Table 4.
278: Figures \ref{fig:rv} and \ref{fig:lcsSARA} show the best model
279: radial velocity and light curves plotted with the observations.
280:
281:
282: The MLO light curves were modeled in the same way as the SARA light curves
283: described above. The results are summarized in the third column of Table 4.
284: Figure 4 shows the best model fits for those observations.
285: The two bottom diagrams in figure 5 show the best spot configuration in this
286: case.
287: The light curves from MLO turned out to be not as useful as the SARA
288: light curves for deriving precise parameters of the system.
289: The statistical errors on the fitted and derived
290: parameters from the MLO data are typically a factor of three larger
291: than the statistical errors from the SARA data.
292: The SARA light curves have a total of 2702
293: points in three filters obtained with a 0.9m telescope, whereas
294: the MLO light curves have a total of 730 points in two filters
295: obtained with a 0.6m telescope, so it is not surprising that
296: much more precise results were obtained from the SARA data.
297:
298:
299: \section{Parameters of NSVS01031772}
300:
301: We use the ELC solutions discussed
302: in \S 4 to derive the physical parameters
303: of NSVS0103. The masses of the stars derived from the orbital
304: solution of the SARA data are $M_1$= 0.5416 $\pm$ 0.0068$M_{\sun}$ and
305: $M_2$= 0.4988 $\pm$ 0.0048$M_{\sun}$. For the radii we obtain
306: $R_1$= 0.5273 $\pm$ 0.0029$R_{\sun}$ and $R_2$= 0.5058 $\pm$ 0.0032$R_{\sun}$.
307: The values of those parameters from the MLO data are
308: $M_1$= 0.543 $\pm$ 0.003$M_{\sun}$ and $M_2$= 0.498 $\pm$ 0.003$M_{\sun}$
309: and
310: $R_1$= 0.510 $\pm$ 0.010$R_{\sun}$ and $R_2$= 0.538 $\pm$ 0.010$R_{\sun}$.
311: Given the discrepancies between the two datasets, mainly in the values of the stellar radii, we decided to adopt a weighted average of those parameters as our final values. The results are shown in Table 5.
312:
313: The projected rotational
314: velocities of the stars, derived from their radii and the orbital
315: period of the system (assuming synchronous rotation), are ${v_{\rm
316: sync}}_1 \sin i= 72.13\pm0.38$ and
317: ${v_{\rm sync}}_2 \sin i=69.77\pm0.41$ km s$^{-1}$,
318: consistent with the results from
319: TODCOR. The values of $T_{eff1}$ and $T_{eff2}$ in Table 5 correspond to
320: the average of the SARA and MLO results. Note that the effective temperatures
321: reported in Table 4 correspond to the temperature of the photosphere of the stars without spots. We recomputed the values of $T_{eff1}$ to account for the presence of spots. The resultant mean effective temperature of the system is therefore $T_{eff}= 3564 \pm 73K$. This value agrees within the errors with the mean effective temperature of $T_{eff}= 3750 \pm 185$K derived using the color-$T_{eff}$ relations in Table 2 of L\'opez-Morales \& Ribas (2005).
322:
323:
324: The luminosity and absolute magnitude of the stars were
325: computed from their effective temperatures, radii, and surface
326: gravities using filter-integrated {\sc NextGen}
327: model atmospheres provided by France
328: Allard. A simple computer code was written that uses
329: the temperature, gravity, and radius plus
330: the $1 \sigma$ errors of each component as
331: inputs to compute the absolute magnitudes of the binary
332: in the standard UBVRIJHK filters. We find,
333: for example, $M_V = 9.08\pm 0.22$,
334: $M_J = 5.77\pm 0.09$, and
335: $M_K = 4.89\pm 0.08$.
336: The distance can be computed by using the observed near infrared
337: magnitudes given in the 2MASS database, where NSVS0103 has the designation
338: 2MASS J13453489+7923482.
339: The observed magnitudes are listed as $J = 9.692
340: \pm 0.021$,
341: $H = 9.021 \pm 0.018$,
342: and
343: $K = 8.778 \pm 0.016$. The $K$-band
344: extinction to infinity in this direction is 0.012 mag, according to the
345: extinction maps (see http://irsa.ipac.caltech.edu/applications/DUST/).
346: Assuming the temperatures above, and the best-fitting radii and gravities,
347: we find a distance of $d=60.3 \pm 1.0$ pc using the observed $K$ magnitude and
348: extinction and a distance of
349: $d=60.9 \pm 0.9$ using the $J$-band magnitude and
350: extinction.
351:
352: \subsection{Age, Space Velocities and
353: Stellar Activity Indicators}\label{sec:age}
354:
355: NSVS0103 does not seem to be related to any known cluster, stellar
356: association, or star formation region. Therefore, we can only evaluate
357: its age from its space velocities $(U,V,W)$\footnote{Positive values
358: of $U$, $V$, and $W$ correspond to velocities toward the Galactic
359: center, Galactic rotation and North Galactic pole.}. The heliocentric
360: space velocities of NSVS0103 were computed from its position, radial
361: velocity ($\gamma=19.0\pm1.0$ km~s$^{-1}$), distance ($d=60.6\pm 0.7$
362: pc), and proper motions; the latter retrieved from the USNO-B1.0
363: catalog (Monet et al.~2003):
364: $\mu_{\alpha} =100 \pm 4$ and
365: $\mu_{\delta} =62 \pm 1$ mas yr$^{-1}$. The obtained components
366: of the space motion are
367: $U =-21.07 \pm 0.16$,
368: $V =16.54 \pm 0.04$ and
369: $W = -1.19 \pm 0.15$
370: km~s$^{-1}$, which correspond to a total space velocity of
371: $S =26.81 \pm 0.13$ km~s$^{-1}$. The value of $W$ indicates that NSVS0103
372: is confined to the galactic plane. The location of the binary in the
373: $U - V$ plane does not fall within the young disk stars area defined
374: by Eggen (1989), nor the area occupied by the young population tracers
375: (Skuljan et al.~1999). Also the space velocities of NSVS0103 do not
376: match any known moving group (Montes et al.~2001). We conclude therefore that
377: NSVS0103 is not a young object and has most likely already reached the
378: main sequence. In addition we expect the metal abundance of NSVS0103
379: to be close to solar, since its space motions agree with a disk
380: population.
381:
382: NSVS0103 appears in the ROSAT All-Sky Bright Source Catalog (Voges et
383: al.~1999) as X-ray source 1RXS J134540.6+792332. Using the calibration
384: equation by Schmitt et al.~(1995) we estimate an X-ray
385: flux of
386: $(5.97 \pm 1.25) \times 10^{-13}$ ergs~cm$^{-2}$~$s^{-1}$,
387: which results in an X-ray
388: luminosity of
389: $\log L_{X} ({\rm ergs~s}^{-1}) = 29.39 \pm 0.09$,
390: which is similar to the
391: X-ray luminosities of YY Gem,
392: $\log L_{X} ({\rm ergs~s}^{-1}) = 29.27 \pm 0.02$,
393: and GU Boo,
394: $\log L_{X} ({\rm ergs~s}^{-1}) = 29.3 \pm 0.2$,
395: computed using the same calibration equation. Finally,
396: we observe strong $H_{\alpha}$ emission lines in our spectra, with an
397: average equivalent width of $5.4\pm 1.1$~\AA\
398: over all phases. That
399: $H_{\alpha}$ emission level is higher than those of YY Gem
400: (2.0~\AA), CU Cnc (3.85--4.05~\AA) and GU Boo (1.7~\AA).
401:
402: \section{Comparison with Models} \label{sec:models}
403: We compare the masses and radii of the stars in
404: NSVS0103 to the predictions by the models of
405: Baraffe et al.~(1998) (hereafter B98) and
406: Siess et al.~(2000) (hereafter S00), which are
407: the only models that explicitly attempt to
408: reproduce the properties of main sequence low-mass stars. We also include
409: in this comparison the other known binaries compiled in Table 10
410: of L\'opez-Morales \& Ribas (2005), and the new systems found by Hebb et al.~
411: (2006), Bayless \& Orosz (2006) and Young et al. (2006). After exploring a range of metallicities and ages,
412: we conclude that the Z=0.02 models provide the closest
413: values to the measured masses
414: and radii of NSVS0103. Different isochrones give practically identical
415: results, assuming the system has already reached the main
416: sequence\footnote{Chabrier \& Baraffe (1995) estimate a time of
417: arrival to the ZAMS of about 0.3 Gyr for stars below 0.6 $M_{\odot}$},
418: so we chose a 0.35 Gyr isochrone which also allows us to compare
419: NSVS0103 to the other binaries. This comparison is illustrated in
420: Figure~\ref{fig:mr}. We find that the models consistently predict
421: smaller radii than the ones observed. The discrepancies are of the
422: order of 5\% for the primary and 11.9\% for the secondary for B98. In the case of the S00 models, those discrepancies are 15.5\% and 22\%. This result fully agrees with
423: the trend observed in previous studies of YY Gem, CU Cnc, and GU Boo. We also show in figure 6 the empirical Mass-Radius relation derived by Bayless \& Orosz (2006), using the parameters of previously studied low-mass binaries. The parameters of the secondary in NSVS0103 fit well in this case, but the radius of the primary is overestimated by about 5\%
424:
425: We compare next the $T_{\rm eff}$ and $M_{V}$ of the stars in NSVS0103 to
426: the two models above. The error bars of these parameters are much
427: larger than in $M$ and $R$, since we have to use external calibrations
428: to compute their values. However, they are still useful to check for
429: significant discrepancies between models and observations. The
430: temperatures predicted by B98 for 0.50 and 0.54 $M_{\sun}$ stars are
431: 3657K and 3750K, while the predictions by S00 for those same stars are
432: 3810K and 3880K. In both cases the models predict hotter temperatures than the ones obtained for NSVS0103. Finally, B98 predict
433: absolute magnitudes of 9.89 and 9.48 for 0.50 and 0.54 $M_{\sun}$
434: stars, while the values predicted by S00 are 9.97 and 9.56. In both cases the absolute magnitudes of NSVS0103 agree with the models, given the uncertainties.
435:
436:
437: \section{Summary and Discussion} \label{sec:sum}
438:
439: We have found a new detached eclipsing binary composed of two M-type
440: stars with masses between 0.49 and 0.53 $M_{\sun}$. Both stars fall in
441: a stretch of the mass-radius relation where no accurate data were
442: available before. The radii of these stars are the most accurate
443: derive to date for low-mass stars, with uncertainties of only 0.55 and
444: 0.63\%. The estimated temperatures of the stars are
445: $T_{{\rm eff}_1} =
446: 3615 \pm 72K$ and $T_{{\rm eff}_2} = 3513 \pm 31K$, after accounting for the effect of
447: spots. We also conclude from the space velocities that the binary is
448: most likely on the main sequence stage. Therefore the parameters of
449: the stars are suitable to test stellar structure models.
450:
451: The models that most closely reproduce the radius of the stars in
452: NSVS0103 are B98 ($Z=0.02$). However, they still underestimate the
453: observed values by 8.5\% on average; a result similar to previous
454: studies of other
455: binaries. This discrepancy between models and observations could be
456: attributed to innacuracies in the equation of state. However, the same
457: models describe well stars in long period ($> 10$ days) spectroscopic
458: binaries and single field stars (Delfosse et al.~2000; Ribas
459: 2005). One difference between short period binaries and the long
460: period binaries and single stars is the higher rotational velocities
461: of the former as result of orbital synchronization. The higher
462: rotational speeds enhance the magnetic activity, and hence the
463: appearance of large spots, X-ray activity, and emission lines in the
464: stars. The observed larger radii and lower temperatures could be a
465: consequence of that enhanced activity. Mullan \& MacDonald (2001), for
466: example, suggest that the larger radii may be caused by the inhibition
467: of convection in stars with strong magnetic fields. The stars would appear cooler as a consequence of the
468: larger radii. If this hypothesis is correct, it affects not only the
469: components of short period binaries, but also any active low-mass
470: star. If the radius of low-mass stars is correlated to their magnetic
471: activity (equivalently to their rotational velocity), the current
472: models may only apply to slow, inactive rotators. For younger, faster
473: rotating stars, their magnetic activity level may have to be included
474: in the models as an additional parameter to compute their fundamental
475: properties.
476:
477: \acknowledgments
478:
479: We thank the Southeastern Association for Research in Astronomy for
480: making their facilities available to us. We also thank G. Torres for
481: providing us with his implementation of TODCOR, and our referee for
482: useful comments and suggestions. M.~L-M. receives research and travel
483: support from the Carnegie Institution of Washington through a Carnegie
484: Fellowship. L.~H. was supported by NSF-REU grant AST-0453609 to
485: San Diego State University.
486: T.~M. was supported by NSF-REU grant AST-0097616 to the
487: Southeastern association for Research in Astronomy. This project was
488: partially supported by the National Aeronautics and Space
489: Administration grant NAG5-12182. We made use of data from the Northern
490: Sky Variability Survey created jointly by the Los Alamos National
491: Laboratory and University of Michigan. The NSVS was funded by the US
492: Department of Energy, the National Aeronautics and Space
493: Administration and the National Science Foundation.
494:
495: \begin{thebibliography}{}
496:
497:
498: \bibitem[Baraffe et al.(1998)]{bar98}
499: Baraffe, I., Chabrier, G., Allard, F., \& Hauschildt, P.H. 1998,
500: A\&A, 337, 403
501:
502: \bibitem[Chabrier \& Baraffe(1995)]{cha95}
503: Chabrier, G. \& Baraffe, I.
504: 1995, ApJ, 451, L29
505:
506: \bibitem[Chabrier \& Baraffe(2000)]{cha00}
507: Chabrier, G. \& Baraffe, I.
508: 2000, ARA\&A, 38, 337
509:
510: \bibitem[Chabrier et al.(2005)]{cha05}
511: Chabrier, G., Baraffe, I., Allard, F., \&
512: Hauschildt, P. H. 2005, arXiv: astro-ph/0509798
513:
514: \bibitem[Claret(2001)]{cla01}
515: Claret, A. 2001, MNRAS, 327, 989
516:
517: \bibitem[Clarke(2002)]{clar02}
518: Clarke, D. 2002, A\&A, 386, 763
519:
520: \bibitem[Creevey et al.(2005)]{cre05}
521: Creevey, O. L., Benedict, G. F,
522: Brown, T. R., et al.\ 2005, ApJL, 625, 127
523:
524: \bibitem[Delfosse et al.(1999)]{del99}
525: Delfosse, X., Forveille, T., Mayor,
526: M., et al.\ 1999, A\&A, 341, L63
527:
528: \bibitem[Delfosse et al.(2000)]{del00}
529: Delfosse, X., Forveille, T.,
530: S\'egransan, D., et al.\ 2000, A\&A, 364, 217
531:
532: \bibitem[Eggen(1989)]{egg89}
533: Eggen, O. 1989, PASP, 101, 366
534:
535: \bibitem[Hebb et al.(2006)]{heb06}
536: Hebb, L., Wyse, R. F. G., Gilmore, G. \&
537: Holtzman, J. 2006, AJ, 131, 555
538:
539: \bibitem[Lacy(1977)]{lac77}
540: Lacy, C. H. 1977, ApJ, 218, 444
541:
542: \bibitem[Leung \& Schneider(1978)]{leu78}
543: Leung, K. C. \& Schneider, D. P.
544: 1978, AJ, 83, 618
545:
546: \bibitem[L\'opez-Morales \& Ribas(2005)]{lop05}
547: L\'opez-Morales, M. \&
548: Ribas, J. C. 2005, ApJ, 631, 1120
549:
550: \bibitem[Maceroni \& Montalb\'an(2004)]{mac04}
551: Maceroni, C. \&
552: Montalb\'an, J. 2004, A\&A, 426, 577
553:
554: \bibitem[Metcalfe et al.(1996)]{met96}
555: Metcalfe, T. S., Mathieu, R. D.,
556: Latham, D. W., \& Torres, G. 1996, ApJ, 456, 356
557:
558: \bibitem[Monet et al.(2003)]{mon03}
559: Monet, D. G., et al.\ 2003, AJ, 125,
560: 984 %The USNO-B1.0 Catalog
561:
562: \bibitem[Montes et al.(2001)]{mon01}
563: Montes, D., L\'opez-Santiago, J.,
564: G\'alvez, M. C., et al.\ 2001, MNRAS, 328, 45
565:
566: \bibitem[Mullan \& MacDonald(2001)]{mul01}
567: Mullan, D. J. \& MacDonald, J. 2001, ApJ, 559, 353
568:
569: \bibitem[Orosz \& Hauschildt(2000)]{oro00}
570: Orosz, J. A. \& Hauschildt, P. H. 2000, A\&A, 364, 265
571:
572: \bibitem[Orosz et al.(2002)]{oro02}
573: Orosz, J. A., Groot, P. J.,
574: van der Klis, M., et al.\ 2002, \apj, 568, 845
575:
576: \bibitem[Ribas(2003)]{rib03}
577: Ribas, I. 2003, A\&A, 398, 239
578:
579: \bibitem[Ribas(2005)]{rib05}
580: Ribas, I. 2005, arXiv:astro-ph/0511431
581:
582: \bibitem[Schmitt et al.(1995)]{sch95}
583: Schmitt, J. H. M. N.,
584: Fleming, T. A., \& Giampapa, M. S. 1995, ApJ, 450, 392
585:
586: \bibitem[Skuljan et al.(1999)]{sku99}
587: Skuljan, J., Hearnshaw, J. B.,\&
588: Cottrell, P. L. 1999, MNRAS, 308, 731
589:
590: \bibitem[Siess et al.(2000)]{sie00}
591: Siess, L., Dufour, E., \& Forestini, M. 2000, A\&A, 358, 593
592:
593: \bibitem[Schwarzenberg-Czerny(1989)]{sch89}
594: Schwarzenberg-Czerny, A. 1989, MNRAS, 241, 153
595:
596: \bibitem[Torres \& Ribas(2002)]{tor02}
597: Torres, G. \& Ribas, I. 2002, ApJ,
598: 567, 1140
599:
600: \bibitem[Voges et al.(1999)]{vog99}
601: Voges, W., Aschenbach, B., Boller, T.,
602: et al.\ 1999, A\&A, 349, 389
603:
604: \bibitem[Wilson \& Devinney (1971)]{wil71}
605: Wilson, R. E. \& Devinney, E. J. 1971, ApJ, 166, 605
606:
607: \bibitem[Wilson \& Sofia(1976)]{wil76}
608: Wilson, R. E. \& Sofia, S. 1976, \apj, 203, 182
609:
610: \bibitem[Wilson (1990)]{wil90}
611: Wilson, R. E. 1990, ApJ, 356, 613
612:
613: \bibitem[Wo\'zniak et al.(2004)]{woz04}
614: Wo\'zniak, P. R.,
615: Vestrand, W. T.,
616: Akerlof, C. W., et al.\ 2004, AJ, 127, 2436
617:
618: \bibitem[Young et al. (2006)]{you06} Young, T. B., Hidas, M. G.,
619: Webb, J. K., Ashley, M. C. B., Christiansen, J. L., Derekas, A. \&
620: Nutto, C. 2006, MNRAS, 370, 1529
621:
622: \bibitem[Zucker \& Mazeh(1994)]{zuc94}
623: Zucker, S. \& Mazeh, T. 1994, ApJ,
624: 420,806
625:
626: \end{thebibliography}
627:
628: \clearpage
629: %%%%%%%%% TABLES
630:
631: \begin{table}[t]
632: \centering
633: \footnotesize
634: \caption{Sample of radial velocities table. A full version of this table is available online}
635: \label{tab:times}
636: \begin{tabular}{lrrrrr}
637: \hline\hline
638: HJD (days) & phase &$K_{1}$ (km/s)&$K_{1}$ err (km/s)&$K_{2}$ (km/s) & $K_{2}$ err (km/s)\\
639: \hline
640: 2453493.7683& 0.746 & 160.51 & 9.29 &-134.44 &14.74\\
641: 2453493.7710& 0.753 & 158.48 &11.32 &-136.25 & 9.47\\
642: 2453493.7737& 0.760 & 164.10 &14.24 &-135.40 &12.21\\
643: ... & ... & ... & ... & ... & ... \\
644: \hline\hline
645: \end{tabular}
646: \end{table}
647:
648: \begin{table}[t]
649: \centering
650: \footnotesize
651: \caption{Sample of SARA and MLO light curve data. A full version of this table is available online}
652: \label{tab:times}
653: \begin{tabular}{cccc}
654: \hline\hline
655: HJD (days) & phase & mag &mag err\\
656: \hline
657: 2453456.92852& 0.676& 0.466& 0.018\\
658: 2453456.92971& 0.679& 0.458& 0.016\\
659: 2453456.93091& 0.683& 0.470& 0.016\\
660: 2453456.93212& 0.686& 0.481& 0.016\\
661: 2453456.93331& 0.689& 0.464& 0.016\\
662: ... & ... & ... & ... \\
663: \hline\hline
664: \end{tabular}
665: \end{table}
666:
667: \begin{table}[t]
668: \centering
669: \footnotesize
670: \caption{Observed Times of Minimum Light}
671: \label{tab:times}
672: \begin{tabular}{ccc}
673: \hline\hline
674: Time of minima & cycle \# & O-C \\
675: (HJD) & (since $T_{0}$) & (min) \\
676: \hline
677: 2453473.98266 & 47.0 & $ 0.596448$ \\
678: 2453478.95221 & 60.5 & $ 0.079632$ \\
679: 2453480.97698 & 66.0 & $ 0.068544$ \\
680: 2453480.79303 & 65.5 & $ 0.242352$ \\
681: 2453490.91543 & 93.0 & $ -1.901088$ \\
682: 2453492.94170 & 98.5 & $ 0.247824$ \\
683: 2453493.86213 & 101.0 & $ 0.357984$ \\
684: 2453520.73664 & 174.0 & $ 0.628416$ \\
685: 2453520.92006 & 174.5 & $ -0.308592$ \\
686: 2453916.67214 & 1249.5 & $ -0.200592$ \\
687: 2453916.85669 & 1250.0 & $ 0.489600$ \\
688: 2453917.77661 & 1252.5 & $ -0.134640$ \\
689: \hline\hline
690: \end{tabular}
691: \end{table}
692:
693:
694: \begin{table}[t]
695: \centering
696: \footnotesize
697: \caption{Model Parameters for Light Curve Solution}
698: \label{tab:parm}
699: \begin{tabular}{lr@{\,$\pm$\,}lr@{\,$\pm$\,}l}
700: \hline\hline
701: Parameter & \multicolumn{2}{c}{Value from SARA} &
702: \multicolumn{2}{c}{Value from MLO}\\
703: \hline
704: orbital period $P$ (days) \dotfill & \multicolumn{4}{c}{fixed at 0.368141} \\
705: eccentricity $e$ \dotfill & \multicolumn{4}{c}{fixed at 0.0} \\
706: rotational velocities \dotfill & \multicolumn{4}{c}{fixed at synchronous} \\
707: gravity darkening, primary \dotfill & \multicolumn{4}{c}{0.0450244} \\
708: gravity darkening, secondary \dotfill & \multicolumn{4}{c}{0.0450994} \\
709: limb darkening coefficients \dotfill & \multicolumn{4}{c}{from model
710: atmospheres} \\
711: reflection \dotfill & \multicolumn{4}{c}{``simple'' reflection} \\
712: inclination $i$ (deg) \dotfill & 85.91&0.03 & 85.86&0.05 \\
713: mass ratio $Q\equiv M_2/M_1$ \dotfill & 0.9217&0.0048 & 0.9166&0.0050 \\
714: radius ratio $R_2/R_1$ \dotfill & 0.9638&0.0066 & 1.054&0.051 \\
715: temperature ratio $T_2/T_1$ \dotfill & 1.01129&0.00036 & 0.9918&0.0012 \\
716: primary temperature $T_1$ (K) \dotfill & 3505&6 & 3512&14 \\
717: secondary temperature $T_2$ (K) \dotfill & 3545&6 & 3482&18 \\
718: separation $a$ ($R_{\odot}$) \dotfill & 2.1870&0.0089 & 2.1910&0.0050 \\
719: $K_1$ (km s$^{-1}$) \dotfill & 143.85&0.37 & 143.5&0.6 \\
720: $K_2$ (km s$^{-1}$) \dotfill & 156.06&0.88 & 156.6&0.7 \\
721: Omega-potential 1 \dotfill & 5.121&0.015 & 5.24&0.11 \\
722: Omega-potential 2 \dotfill & 5.053&0.022 & 4.81&0.08 \\
723: temperature factor, spot 1 \dotfill & 1.048&0.004 & 1.19&0.02 \\
724: latitude spot 1 (deg) \dotfill & 26.1&0.3 & 104&17\\
725: longitude spot 1 (deg) \dotfill & 310.9&0.5 & 291.5&2.0 \\
726: angular radius spot 1 (deg) \dotfill & 88.4&0.5 & 8.5&1.0 \\
727: temperature factor, spot 2 \dotfill & 1.196&0.002 & 1.075&0.010 \\
728: latitude spot 2 (deg) \dotfill & 42.4&1.0 & 46&6 \\
729: longitude spot 2 (deg) \dotfill & 90.6&0.5 & 63.5&2.0\\
730: angular radius spot 2 (deg) \dotfill & 14.9&0.2 & 24.5&0.5 \\
731: \hline\hline
732: \end{tabular}
733: \end{table}
734:
735: %
736: %
737: %
738:
739: \begin{table}[t]
740: \centering
741: \footnotesize
742: \caption{Absolute dimensions and main physical parameters of the
743: components of NSVS0103}
744: \label{tab:AbsDim}
745: \begin{tabular}{lcc}
746: \hline\hline
747: Parameter & Primary & Secondary\\
748: \hline
749: Mass ($M_{\odot}$) \dotfill & $0.5428 \pm 0.0027$ & $0.4982 \pm 0.0025$\\
750: Radius ($R_{\odot}$) \dotfill & $0.5260 \pm 0.0028$ & $0.5088 \pm 0.0030$ \\
751: $\log g$ (cgs) \dotfill &$ 4.730 \pm 0.005 $ & $4.722 \pm 0.006$ \\
752: $v_{\rm sync} \sin i$ (km s$^{-1}$) \dotfill & $72.13\pm 0.38$ & $69.77 \pm 0.41$ \\
753: $T_{\rm eff}$ (K) \dotfill & 3615 $\pm$ 72 & 3513 $\pm$ 31 \\
754: $L/L_{\odot}$ \dotfill & 0.0426 $\pm$ 0.0034& 0.0356 $\pm$ 0.0013\\
755: $M_{\rm bol}$ (mag) \dotfill & 8.08 $\pm$ 0.25& 8.27 $\pm$ 0.12\\
756: $M_{V}$ (mag) \dotfill & 9.74 $\pm$ 0.22& 10.04 $\pm$ 0.10\\
757: \hline\hline
758: \end{tabular}
759: \end{table}
760:
761: %%%%%%%%%%%% FIGURES
762:
763: \begin{figure}[t]
764: \epsscale{1.0}
765: \plotone{fig1.eps}
766: \caption{Finding chart for NSVS0103, made from the Digitized Sky
767: Survey. NSVS0103 is marked with the hash marks.}
768: \label{fig:fc}
769: \end{figure}
770:
771: \begin{figure}[t]
772: \epsscale{1.0}
773: \plotone{fig2.eps}
774: \caption{Radial velocity curves of NSVS0103. The filled circles and squares correspond, respectively, to the velocities of the primary and the secondary used in the orbital fit in \S 4. The open symbols show points excluded from that fit. The solid and dotted lines represent the orbital solution obtained with ELC.
775: The dashed line shows the velocity of the center of mass of the system ($\sim$ 19 km/s). These data are available in the electronic edition of this journal (Table 1).}
776: \label{fig:rv}
777: \end{figure}
778:
779: \begin{figure}[t]
780: \epsscale{1.0}
781: \plotone{fig3.eps}
782: \caption{V, R, and I light curves from SARA. The dots represent individual observations in each passband. The solid lines show the best fit to the data using ELC. The O-C diagrams on the top show the residuals of those fits. These light curves are available in the electronic edition of this journal (Table 2).}
783: \label{fig:lcsSARA}
784: \end{figure}
785:
786: \begin{figure}[t]
787: \epsscale{1.0}
788: \plotone{fig4.eps}
789: \caption{R and I light curves from MLO. The dots represent individual observations in each passband. The solid lines show the best fit to the data using ELC. The O-C diagrams on the top show the residuals of those fits. These light curves are available in the electronic edition of this journal (Table 2).}
790: \label{fig:lcsMLO}
791: \end{figure}
792:
793: \begin{figure}[t]
794: \epsscale{0.9}
795: \plotone{fig5.eps}
796: \caption{Spot models that best reproduce the SARA light curves (top diagram) and the MLO light curves (two bottom diagrams). In both cases the best solution correspond to two bright spots on the primary. The parameters of the spots are listed in Table 4}\label{fig:spot}
797: \end{figure}
798:
799: \begin{figure}[t]
800: \epsscale{0.9}
801: \plotone{fig6.eps}
802: \caption{Mass-radius relations of stars between 0.1 and 0.8 $M_{\odot}$
803: predicted by the models of Baraffe et al. (1998) ($solid$ $line$) and Siess
804: et al.(2000) ($dashed$ $line$), for an age of 0.35 Gyrs and $Z$ = 0.02, and the empirical relation of Bayless \& Orosz (2006) (($dotted$ $line$). The open triangles show the location of all the low-mass main sequence binaries previously found. The open circles show the two stars in NSVS0103. Each point includes error bars.}\label{fig:mr}
805: \end{figure}
806:
807: \end{document}
808: