1: 

2: \documentclass[12pt,preprint]{aastex}

3: %\documentclass[manuscript]{aastex}

4: %\usepackage{latexsym}

5: %\usepackage{graphicx}

6: 

7: 

8: \begin{document}

9: 

10: \title{Identification, Classifications, and Absolute Properties of\\ 773 Eclipsing Binaries Found in the TrES Survey}

11: 

12: \author{Jonathan Devor\altaffilmark{1,2}, David Charbonneau\altaffilmark{1,3},\\ Francis T. O'Donovan\altaffilmark{4}, Georgi Mandushev\altaffilmark{5}, and Guillermo Torres\altaffilmark{1}}

13: 

14: \altaffiltext{1}{Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138}

15: \altaffiltext{2}{Email: jdevor@cfa.harvard.edu}

16: \altaffiltext{3}{Alfred P. Sloan Research Fellow}

17: \altaffiltext{4}{California Institute of Technology, 1200 East California Boulevard, Pasadena, CA 91125}

18: \altaffiltext{5}{Lowell Observatory, 1400 West Mars Hill Road, Flagstaff, AZ 86001}

19: 

20: \begin{abstract}

21: 

22: In recent years we have witnessed an explosion of photometric

23: time-series data, collected for the purpose of finding a small

24: number of rare sources, such as transiting extrasolar planets and

25: gravitational microlenses. Once combed, these data are often set

26: aside, and are not further searched for the many other variable

27: sources that they undoubtedly contain. To this end, we describe a

28: pipeline that is designed to systematically analyze such data,

29: while requiring minimal user interaction. We ran our pipeline on a

30: subset of the Trans-Atlantic Exoplanet Survey dataset, and used it

31: to identify and model 773 eclipsing binary systems. For each

32: system we conducted a joint analysis of its light curve, colors,

33: and theoretical isochrones. This analysis provided us with

34: estimates of the binary's absolute physical properties, including

35: the masses and ages of their stellar components, as well as their

36: physical separations and distances. We identified three types of

37: eclipsing binaries that are of particular interest and merit

38: further observations. The first category includes 11 low-mass

39: candidates, which may assist current efforts to explain the

40: discrepancies between the observation and the models of stars at

41: the bottom of the main-sequence. The other two categories include

42: 34 binaries with eccentric orbits, and 20 binaries with abnormal

43: light curves. Finally, this uniform catalog enabled us to identify

44: a number of relations that provide further constraints on binary

45: population models and tidal circularization theory.

46: 

47: \end{abstract}

48: 

49: \keywords{binaries: eclipsing --- catalogs --- methods: data analysis --- stars: statistics --- techniques: photometric}

50: 

51: 

52: \section{Introduction}

53: 

54: Since the mid 1990s there has been an explosion of large scale

55: photometric variability surveys. The search for gravitational

56: microlensing events, which were predicted by \citet{Paczynski86},

57: motivated the first wave of surveys [e.g. OGLE: \citet{Udalski94};

58: EROS: \citet{Beaulieu95}; DUO: \citet{Alard97}; MACHO:

59: \citet{Alcock98}]. Encouraged by their success, additional

60: surveys, searching for Gamma-Ray Bursts [e.g. ROTSE:

61: \citet{Akerlof00}] and general photometric variabilities [e.g.

62: ASAS: \citet{Pojmanski97}] soon followed.

63: 

64: Shortly thereafter, with the discovery of the first transiting

65: extrasolar planet \citep{Charbonneau00, Henry00, Mazeh00}, a

66: second wave of photometric surveys ensued [e.g. OGLE-III:

67: \citet{Udalski03}; TrES: \citet{Alonso04}; HAT: \citet{Bakos04};

68: SuperWASP: \citet{Christian06}; XO: \citet{McCullough06}; for

69:  a review, see \citet{Charbonneau07}]. Each

70: of these projects involved intensive efforts to locate a few

71: proverbial needles hidden in a very large data haystack. With few

72: exceptions, once the ``needles'' were found, thus fulfilling the

73: survey's original purpose, the many gigabytes of photometric light

74: curves (LCs) collected were not made use of in any other way. In

75: this paper we demonstrate how one can extract a great deal more

76: information from these survey datasets, with comparably little

77: additional effort, using automated pipelines. To this end, we have

78: made all the software tools described in this paper freely

79: available (see web links to the source code and working examples),

80: and they are designed to be used with any LC dataset.

81: 

82: In the upcoming decade, a third wave of ultra-large ground-based

83: synoptic surveys [e.g. Pan-STARRS: \citet{Kaiser02}; LSST:

84: \citet{Tyson02}], and ultra-sensitive space-based surveys [e.g.

85: KEPLER: \citet{Borucki97}; COROT: \citet{Baglin98}; GAIA:

86: \citet{Gilmore98}] are expected to come online. These surveys are

87: designed to produce photometric datasets that will dwarf all

88: preceding efforts. To make any efficient use of such large

89: quantities of data, it will become imperative to have in place a

90: large infrastructure of automated pipelines for performing even

91: the most casual data mining query.

92: 

93: In this paper, we focus exclusively on the identification and

94: analysis of eclipsing binary (EB) systems. EBs provide favorable

95: targets, as they are abundant and can be well modeled using

96: existing modeling programs [e.g. WD: \citet{Wilson71}; EBOP:

97: \citet{Popper81}]. Once modeled, EBs can provide a wealth of

98: useful astrophysical information, including constraints on binary

99: component mass distributions, mass-radius-luminosity relations,

100: and theories describing tidal circularization and synchronization.

101: These findings, in turn, will likely have a direct impact on our

102: understanding of star formation, stellar structure, and stellar

103: dynamics. These physical distributions of close binaries may even

104: help solve open questions relating to the progenitors of Type Ia

105: supernovae \citep{Iben84}. In additional to these, EBs can be used

106: as tools; both as distance indicators \citep{Stebbing10,

107: Paczynski97} and as sensitive detectors for tertiary companions

108: via eclipse timing \citep{Deeg00, Holman05, Agol05}.

109: 

110: In order to transform such large quantities of data into useful

111: information, one must construct a robust and computationally

112: efficient automated pipeline. Each step along the pipeline will

113: either measure some property of the LC, or filter out LCs that do

114: not belong, so as to reduce the congestion in the following, more

115: computationally intensive steps. One can achieve substantial gains

116: in speed by dividing the data into subsets, and processing them in

117: parallel on multiple CPUs. The bottlenecks of the analysis are the

118: steps that require user interaction. In our pipeline, we reduce

119: user interaction to essentially yes/no decisions regarding the

120: success of the EB models, and eliminate any need for interaction

121: in all but two stages. We feel that this level of interaction

122: provides good quality control, while minimizing its detrimental

123: subjective effects.

124: 

125: The data that we analyzed originate from 10 fields of the

126: Trans-atlantic Exoplanet Survey [TrES ; \citet{Alonso04}]. TrES

127: employs a network of three automated telescopes to survey $6^\circ

128: \times 6^\circ$ fields-of-view. To avoid potential systematic

129: noise we use the data from only one telescope, Sleuth, located at

130: the Palomar Observatory in Southern California \citep{ODonovan04}.

131: This telescope has a 10 cm physical aperture and a photometric

132: aperture of radius of 30". The number of LCs in each field ranges

133: from 10405 to 26495 (see Table \ref{tableFieldsObs}), for a total

134: of 185445 LCs. The LCs consist of $\sim$2000 $r$-band photometric

135: measurements at a 9-minute cadence. These measurements were

136: created by binning the image-subtraction results of 5 consecutive

137: 90-second observations, thus improving their non-systematic

138: photometric noise. As a result $\sim$16\% of the LCs have an RMS

139: $<$1\%, and $\sim$38\% of the LCs have an RMS $<$2\% (see Table

140: \ref{tableFieldsYield}). The calibration of TrES images,

141: identification of stars therein, extraction, and decorrelation of

142: the LCs is described elsewhere \citep{Dunham04, Mandushev05,

143: ODonovan06, ODonovan07}. TrES is currently an active survey that

144: is continuously observing new fields, though for this paper we

145: have limited ourselves to these 10 fields.

146: 

147: 

148: \section{Method}

149: \label{secMethod}

150: 

151: The pipeline we have developed is an extended version of the

152: pipeline described by \citet{Devor05}. At the heart of this

153: analysis lie two computational routines that we have described in

154: earlier papers: the Detached Eclipsing Binary Light curve

155: fitter\footnote{The DEBiL source code, utilities, and running

156: example files are available online at:\newline

157: http://www.cfa.harvard.edu/$\sim$jdevor/DEBiL.html} [DEBiL ;

158: \citet{Devor05}], and the Method for Eclipsing Component

159: Identification\footnote{The MECI source code and running examples

160: are available online at:\newline

161: http://www.cfa.harvard.edu/$\sim$jdevor/MECI.html} [MECI ;

162: \citet{Devor06a, Devor06b}]. DEBiL fits each LC to a

163: $\it{geometric}$ model of a detached EB (steps 3 and 5 below).

164: This model consists of two luminous, limb-darkened spheres that

165: orbit in a Newtonian 2-body orbit. MECI restricts the DEBiL fit

166: along theoretical isochrones, and is thus able to create a

167: $\it{physical}$ model of each EB (step 9). This second model

168: describes the masses and absolute magnitudes of the EB's stellar

169: components, which are then used to determine the EB's distance and

170: absolute separation.

171: 

172: The pipeline consists of 10 steps. We elaborate on each of these

173: steps below:

174: 

175: \begin{enumerate}

176: \item Determine the period.

177: \item If a distinct secondary eclipse is not observed, an entry with twice the period is added.

178: \item Fit the orbital parameters with DEBiL.

179: \item Fine-tune the period using eclipse timing.

180: \item Refine the orbital parameters with DEBiL using the revised period.

181: \item Remove contaminated LCs.

182: \item Visually assess the quality of the EB models.

183: \item Match the LC sources with external databases.

184: \item Estimate the absolute physical properties of the binary components using MECI.

185: \item Classify the resulting systems using both automatic and manual criteria.

186: \end{enumerate}

187: 

188: We use the same filtering criteria as described in

189: \citet{Devor05}, both for removing LCs that are not periodic (step

190: 1) and then for removing non-EB LCs (step 3). Together, these

191: automated filters remove approximately 97\% of the input LCs. In

192: addition to these filters, we perform stringent manual inspections

193: (steps 7 and 10) whereby we removed all the LCs we were not

194: confident were EBs. These inspections ultimately removed

195: approximately 86\% of the remaining LCs. Thus only 1 out of every

196: 240 input LCs, were included in the final catalog.

197: 

198: In step (1) we use both the Box-fitting Least Squares (BLS) period

199: finder \citep{Kovacs02}, and a version of the analysis of

200: variances (AoV) period finder \citep{SchwarzenbergCzerny89,

201: SchwarzenbergCzerny96} to identify the periodic LCs within the

202: dataset and to measure their periods. In our AoV implementation,

203: we scan periods from 0.1 days up to the duration of each LC. We

204: then select the period that minimizes the variance of a linear fit

205: within 8 phase bins. We removed all systems with weak

206: periodicities [see \citet{Devor05} for details], and with one

207: exception (T-Lyr1-14413), all the systems whose optimal period was

208: found to be longer than half their LC duration. In this way we

209: were able to filter out many of the non-periodic variables.

210: 

211: The AoV algorithm is most effective in identifying the periods of

212: LCs with long duration features, such as semi-detached EBs and

213: pulsating stars. The BLS algorithm, in contrast, is effective at

214: identifying periodic systems whose features span only a brief

215: portion of the period, such as detached EBs and transiting planets

216: (see Figure \ref{periodDiffFrac}). However, the BLS algorithm is

217: easily fooled by outlier data points, identifying them as short

218: duration features. For this reason the BLS algorithm has a

219: significantly higher rate of false positives than AoV, especially

220: for long periods, which have only a few cycles over the duration

221: of the observations. Therefore we limit the search range of the

222: BLS algorithm to periods shorter than 12 days, although as Figure

223: \ref{periodDiffFrac} illustrates, its efficiency at locating EBs

224: rapidly declines at periods greater than 10 days.

225: 

226: \placefigure{periodDiffFrac}

227: 

228: In step (2), we address the ambiguity between EBs with identical

229: components in a circular orbit, and EBs with extremely disparate

230: components. The phased LC of EBs with identical components

231: contains two identical eclipses, whereas the phased LC of EBs with

232: disparate components will have a secondary eclipse below the

233: photometric noise level. These two cases are degenerate, since

234: doubling the period of a disparate system will result in a LC that

235: looks like an equal-component system. In the pipeline, we handle

236: this problem by doubling such entries; one with the period found

237: in step (1), and another with twice that period. Both of these

238: entries proceed through the pipeline independently. In many cases,

239: after additional processing by the following steps, one of these

240: entries will emerge as being far less likely than the other (see

241: appendix \ref{appendixSingleEclipse}), at which point it is

242: removed. But in cases where photometry alone cannot determine

243: which is correct, one needs to perform spectroscopic follow up to

244: break the ambiguity. In particular, a double-lined spectrum would

245: support the equal-component hypothesis.

246: 

247: Step (3) is performed using DEBiL, which fits the fractional radii

248: ($r_{1,2}$) and observed magnitudes ($mag_{1,2}$) of the EB's

249: stellar components, their orbital inclination ($i$) and

250: eccentricity ($e$), and their epoch ($t_0$) and argument of

251: periastron ($\omega$). DEBiL first produces an initial guess for

252: these parameters, and then iteratively improves the fit using the

253: downhill simplex method \citep{Nelder65} with simulated annealing

254: \citep{Kirkpatrick83, Press92}.

255: 

256: In step (4), we fine-tune the period ($P$) using a method based on

257: eclipse timing\footnote{The source code and running examples are

258: available online at\newline

259: http://www.cfa.harvard.edu/$\sim$jdevor/Timing.html}, which we

260: describe below. In order to produce an accurate EB model in step

261: (9) it is necessary to know the system's period with greater

262: accuracy than that produced in step (1). If we neglect to

263: fine-tune the period, the eclipses may be out of phase with

264: respect to one another, and so the phased eclipses will appear

265: broadened. Our timing method employs the DEBiL model produced in

266: step (3), and uses it to find the difference between the observed

267: and calculated ($O-C$) eclipse epochs. This is done by minimizing

268: the chi-squared fit of the model to the data points in each

269: eclipse, while varying only the model's epoch of periastron. When

270: the period estimate is off by a small quantity ($\Delta P$), the

271: $O-C$ difference increases by $\Delta P$ each period. This change

272: in the $O-C$ over time can be measured from the slope of the

273: linear regression, which is expected to equal $\Delta P / P$. Thus

274: measuring such an $O-C$ slope will yield the desired period

275: correction (see Figure \ref{figTimingVariations}).

276: 

277: If the EB has an eccentric orbit, the primary and secondary

278: eclipse will separate on the $O-C$ plot, and form two parallel

279: lines with a vertical offset of $\Delta t$ (see Figure

280: \ref{figTimingVariationsEcc}). We measure this offset and use it

281: as a sensitive method to detect orbital eccentricities. In

282: particular, the value of $\Delta t$ constrains $e \cos \omega$,

283: which in turn provides a lower limit for the system's eccentricity

284: \citep{Tsesevich73}:

285: 

286: \begin{equation}

287: \label{eqOmC}

288: e \cos \omega \simeq \frac{\pi}{2} \frac{\Delta t}{P}

289: \end{equation}

290: 

291: This formula assumes an orbital inclination of $i=90^\circ$,

292: making it a good approximation for eclipsing binaries. We use this

293: method, in combination with DEBiL, to identify the eccentric EBs

294: in the catalog (see Table \ref{tableEccentric}). However, in cases

295: where the eclipse timing measures $|e \cos \omega| < 0.005$, or

296: when the eccentricity is consistent with zero, we assume that the

297: EB is non-eccentric, and model it using a circular orbit. We

298: further discuss the physics of these systems in

299: \S\ref{subsecEccentricEBs}.

300: 

301: \placefigure{figTimingVariations}

302: \placefigure{figTimingVariationsEcc}

303: 

304: Step (5) is identical to step (3), except that it uses the revised

305: period from step (4). This step provides an improved fit to the

306: LCs, as evidenced by an improved chi-squared value in over 70\% of

307: the cases.

308: 

309: In step (6) we locate and remove non-EB sources that seem to be

310: periodic due to photometric contamination by true EBs. Such

311: contaminations result from overlapping point spread functions

312: (PSF) that cause each source to partially blend into the other.

313: These cases can be easily identified with a program that scans

314: through pairs of targets\footnote{We ran a brute force scan, which

315: required O($N^2$) iterations. But by employing a data structure

316: that can restrict the scan to nearby pairs, it is possible to

317: perform this scan in only O($N$) iterations, assuming that such

318: pairs are rare.}, and selects the ones that both have similar

319: periods (see description below) and are separated by an angle that

320: is smaller than twice the PSF. We found 14 such pairs, all of

321: which were separated by less than $41"$, which is well within

322: twice the TrES PSF ($60"$), while the remaining pairs with similar

323: periods were separated by over $450"$. Upon inspection, all 14 of

324: the pairs we found had similar eclipse shapes, indicating that we

325: had no false positives. Among each pair, we identify the LC with

326: shallower eclipses (in magnitudes) as being contaminated and

327: remove it from the catalog.

328: 

329: We define periods as being similar if the difference between them

330: is smaller than their combined uncertainty. We estimate the period

331: uncertainty using the relation: $\varepsilon_P \propto P^2 / T$,

332: where $T$ is the time interval between the initial and the final

333: observations. One arrives at this relation by noticing that when

334: phasing the LC, the effect of any perturbation from the true

335: period will grow linearly with the number of periods in the LC

336: (see step 4). This amplified effect will become evident once it

337: reaches some fraction of the period itself, in other words, when

338: $\varepsilon_P (T / P) \propto P$. A typical TrES LC with a

339: revised period will have a proportionality constant of

340: approximately $1/1000$. In order to avoid missing contaminated

341: pairs (false negatives) we adopt in this step, the extremely

342: liberal proportionality constant of unity.

343: 

344: In step (7) we conduct a visual inspection of all the LC fits.

345: Most EBs were successfully modeled and were included into the

346: catalog as is. About $1\%$ of the LCs analyzed had misidentified

347: periods, as a result of failures of the period finding method of

348: step (1). In most of these cases the period finder indicated

349: either a harmonic of the true period or a rational multiple of a

350: solar or sidereal day. In such cases we use an interactive

351: periodogram\footnote{LC, created by Grzegorz Pojmanski.} to find

352: the correct period and then reprocess the LCs through the

353: pipeline. Some entries were misidentified at step (2) as being

354: ambiguous, even though they have a detectable secondary eclipse or

355: have slightly unequal eclipses. In these cases the erroneous

356: doubled entry was removed. Lastly, some of the EBs were not fit

357: sufficiently well with DEBiL in step (5). These cases were

358: typically due to clustered outlier data points, systematic noise,

359: or severe activity of a stellar component (e.g. flares or spots),

360: which caused DEBiL to produce erroneous initial model parameters.

361: These cases were typically handled by having DEBiL produce the

362: initial model parameters from a more smoothed version of the LC.

363: 

364: In step (8) we match each system, through its coordinates, with

365: the corresponding source in the Two Micron All Sky Survey catalog

366: [2MASS ; \citep{Skrutskie06}]. This was done to obtain both

367: accurate target positions and observational magnitudes. These

368: magnitude measurements are then used to derive the colors of each

369: EB, which are incorporated into the MECI analysis, as well as to

370: estimate the EB's distance modulus (step 9). To this end, 2MASS

371: provides a unique combination of high astrometric accuracy

372: ($\sim$0.1") together with high photometric accuracy ($\sim$0.015

373: mag) at multiple near-infrared bands, all while maintaining a

374: decent photometric resolving power ($\sim$3"). By employing these

375: near-infrared bands we both inherently reduce the detrimental

376: effects of stellar reddening, and are able to correct for much of

377: the remaining extinction by fitting for the Galactic interstellar

378: absorbtion.

379: 

380: In order to use the measurements from the 2MASS custom $J$, $H$,

381: and $K_s$ filters, we converted them to the equivalent ESO-filter

382: values so that they could be compared to the isochrone table

383: values used in the MECI analysis. This conversion was done using

384: approximate linear transformations \citep{Carpenter01}. However,

385: the colors of three EBs (T-And0-10336, T-Cyg1-02304, and

386: T-Per1-05205) were so anomalous that they did not permit a

387: reasonable model solution, thus we chose not to include any color

388: information in their MECI analyses.

389: 

390: In addition to its brightness, we also look up each EB's proper

391: motion. Although proper motion is not required for any of the

392: pipeline analyses, it provides a useful verification for low-mass

393: candidates (see \S\ref{subsecLowMassEBs}). These systems are

394: expected to have large proper motions, since they must be nearby

395: to be observable in this magnitude-limited survey. The most

396: extreme such case in the catalog is CM Draconis (T-Dra0-01363),

397: which has a proper motion of over 1300 mas/yr \citep{Salim03}, and

398: is probably the lowest mass system in our catalog. To this end, we

399: match each system to the Second U.S. Naval Observatory CCD

400: Astrograph Catalog [UCAC release 2.4 ; \citet{Zacharias04}]. When

401: there was no match with UCAC, we use the more comprehensive but

402: less accurate U.S. Naval Observatory photographic sky survey

403: [USNO-B release 1.0 ; \citet{Monet03}]. These matches were made

404: using the more accurate aforementioned adopted 2MASS coordinates.

405: However, because of their increased observational depth, and the

406: fact that some high-proper motion targets are expected to have

407: moved multiple arcseconds in the intervening decades, we chose to

408: match each target to the brightest ($R$-band) source within 7.5".

409: It should be noted that the position of CM Draconis shifted by

410: more than 22" and had to be matched manually, though 90\% of the

411: matches were separated by less than 0.6", and 98\% were separated

412: by less than 2" (see Figure \ref{figPosErr}).

413: 

414: The proper motions garnered from these databases can be combined

415: with distance estimates ($D$), to calculate the absolute

416: transverse velocity ($v_{tr}$) of a given EB:

417: 

418: \begin{equation}

419: v_{tr} \simeq 4.741\: km/sec \left(\frac{PM}{1\: mas/yr}\right)\left(\frac{D}{1\: kpc}\right),

420: \end{equation}

421: 

422: where $PM$ is the system's angular proper motion. In the catalog

423: we list the right ascension and declination components

424: ($PM_{\alpha}$ and $PM_{\delta}$, respectively), so as to allow

425: one to compute the system's direction of motion in the sky. The

426: value of $PM$ can be computed from its components, using: $PM^2 =

427: PM_{\delta}^2 + PM_{\alpha}^2 \cos^2 \delta$, where $\delta$ is

428: the system's declination. When applying this formula, one should

429: be aware that USNO-B folds the $\cos \delta$ coefficient into its

430: listed $PM_{\alpha}$, while UCAC does not.

431: 

432: Finally, we incorporate the USNO-B photometric $B$- and

433: $R$-magnitude measurements into our catalog to provide a rough

434: estimate of the optical brightness of each target. USNO-B lists

435: two independent measurements in each of these filter, however in

436: some cases one or both of these measurements failed. When both

437: measurements are available, we average them for improved accuracy.

438: However, each measurement has a large photometric uncertainty of

439: $\sim$0.3 mag, thus even these averaged values will have errors

440: that are over an order of magnitude larger that the photometric

441: measurements of 2MASS. For this reason, and because of the

442: increased effect of stellar reddening, we chose not to incorporate

443: these data into the MECI analysis. However, USNO-B's high

444: photometric resolution ($\sim$1") enabled us to detect many

445: sources that blended with our targets in the TrES exposures. By

446: summing the $R$-band fluxes of all the USNO-B sources within 30"

447: of each target, we estimated the fraction of third-light included

448: in each LC (see Figure \ref{figBlending}). Note that this measure

449: provides only a lower bound to the true third-light fraction, as

450: some EBs are expected to have additional close hierarchical

451: components that would not be resolved by USNO-B. For most of the

452: catalog targets, the third-light flux fraction was found to be

453: small ($<$10\%). We therefore conclude that stellar blending will

454: usually have only a minor effect on the MECI analysis results,

455: however users should be aware of the potential biases in the

456: calculated properties of highly blended targets. Though it was not

457: applied to this catalog, in principal, given a third-light flux

458: fraction at a well-determined LC phase, one could correct for the

459: effects of blending.

460: 

461: \placefigure{figPosErr}

462: \placefigure{figBlending}

463: 

464: In step (9) we analyze the LCs with MECI. We refer the reader to

465: the full description of this method in \citet{Devor06a, Devor06b},

466: and provide here only a brief outline. Given an observed EB LC and

467: out-of-eclipse colors, MECI will iterate through a range of values

468: for the EB age and the masses of its two components. By looking up

469: their radii and luminosities in theoretical isochrone tables, MECI

470: simulates the expected LC and combined colors, and selects the

471: model that best matches the observations, as measured by the

472: chi-squared statistic. Or more concisely, MECI searches the ($M_1,

473: M_2, age$)-parameter space for the chi-squared global minimum of

474: each EB. Figures \ref{figMECI1} and \ref{figMECI2} show

475: constant-age slices through such a parameter space. Once found,

476: the curvature of the global minimum along the parameter space axes

477: is used to determine the uncertainties of the corresponding

478: parameters.

479: 

480: The MECI analysis makes two important assumptions. The first is

481: that EB stellar components are coeval, which has been shown to

482: generally hold for close binaries \citep{Claret02}. When this

483: assumption is violated, MECI will often not be able to find an EB

484: model that successfully reproduces the LC eclipses. Such systems,

485: which may be of interest in their own right, make up $\sim$3\% of

486: the catalog and are further discussed later in this section. The

487: second assumption is that there is no significant reddening, or

488: third-light blended into the observations (i.e. from a photometric

489: binary or hierarchical triple). Such blending in the LC will make

490: the eclipses shallower, which produces an effect very similar to

491: that of the EB having a grazing orbit. Thus, it will cause the

492: measured orbital inclination to be erroneous, although it should

493: rarely otherwise affect the results of the MECI analysis

494: significantly. However, the MECI analysis is sensitive to color

495: biases caused by stellar reddening and blending.

496: 

497: We reduce both these biases by incorporating 2MASS colors (see

498: step 8), which are both less suspectable to reddening than optical

499: colors, and suffers from significantly less blending than TrES, as

500: the radius of the 2MASS photometric aperture is $\sim$20 times

501: smaller that that of TrES. We then attempt to further mitigate

502: this problem by analyzing each EB twice, using different relative

503: LC/color information weighting values [see \citet{Devor06b} for

504: further details]. We first run MECI with the default weighting

505: value ($w = 10$), and then run MECI again with an increased LC

506: weighting ($w = 100$) thereby decreasing the relative color

507: weighting. Finally, we adopt the solution that has a smaller

508: reduced chi-square. Typically, the results of the two MECI

509: analyses are very similar, indicating that the observed colors are

510: consistent with the ones predicted by the theoretical isochrones.

511: In such cases, the color information provides an important

512: constraint, which significantly reduces the parameter

513: uncertainties. However, when there is a significant color bias,

514: the default model will not fit the observed data as well as the

515: model that uses a reduced weighting of the color information. In

516: such a case, the reduced color information model, which has a

517: smaller chi-squared, is adopted. Following this procedure, we find

518: that in $\sim$9\% of our EBs, the reduced color information model

519: provided a better fit, indicating that while significant

520: color-bias is uncommon, it is a source of error that should not be

521: ignored.

522: 

523: By default, we had MECI use the Yonsei-Yale \citep{Yi01, Kim02}

524: isochrone tables of solar metallicity stars. Although they

525: successfully describe stars in a wide range of masses, these

526: tables become increasingly inaccurate for low-mass stars, as the

527: stars become increasingly convective. For this reason we

528: re-analyze EBs for which both components were found to have masses

529: below $0.75M_{\sun}$, using instead the \citet{Baraffe98}

530: isochrone tables, assuming a convective mixing length equal to the

531: pressure scale height. Our EB models also take into account the

532: effects of the limb darkening of each of the stellar components.

533: To this end we employ the ATLAS \citep{Kurucz92} and PHOENIX

534: \citep{Claret98, Claret00} tables of quadratic limb-darkening

535: coefficients.

536: 

537: \placefigure{figMECI1}

538: \placefigure{figMECI2}

539: 

540: As previously mentioned, once we know the absolute properties of

541: an EB system, we are able to estimate its distance

542: \citep{Stebbing10, Paczynski97}, and thus such systems can be

543: considered standard candles. We use \citet{Cox00} for the

544: extinction coefficients, assuming the standard Galactic ISM

545: optical parameter, $R_V = 3.1$, to create the following system:

546: 

547: \begin{eqnarray}

548: mag_J - Mag_J & = & \Delta Mag + 0.282 \cdot A(V)\\

549: mag_H - Mag_H & = & \Delta Mag + 0.176 \cdot A(V)\\

550: mag_K - Mag_K & = & \Delta Mag + 0.108 \cdot A(V)

551: \end{eqnarray}

552: 

553: Where $\Delta Mag$ is the extinction-corrected distance modulus,

554: and $A(V)$ is the V-mag absorption due to Galactic interstellar

555: extinction. The estimated distance can then be solved using: $D =

556: 10pc \cdot 10^{\Delta Mag / 5}$. Because we have three equations

557: for only two unknowns, we adopt the solution that minimizes the

558: sum of the squares of the residuals. In some cases we remove one

559: of the bands as being an outlier (i.e. if it would have resulted

560: in a negative absorption), after which we are still able to solve

561: the systems. But in cases where we need to remove two bands, we

562: set $A(V) = 0$ in order to solve for the distance modulus.

563: Although this method has a typical uncertainty of 10\% to 20\%, it

564: can be applied to EBs that are far more distant and dim than are

565: accessible in other methods, such as parallax measurement. It can

566: be used to map broad features of the Galaxy, and identify binaries

567: that are in the Galactic halo. This method can also be used on

568: large group of systems, so that if the EBs are clustered, one can

569: average their distances, and thus reduce the cluster distance

570: uncertainty as the inverse square root of the number of systems.

571: 

572: In step (10) we perform a final quality check for the EB model

573: fits, and classify them into 7 groups:

574: 

575: \renewcommand{\theenumi}{\Roman{enumi}}

576: \begin{enumerate}

577: \item Eccentric: EBs with unequally-spaced eclipses

578: \item Circular: EBs with equally-spaced but distinct eclipses

579: \item Ambiguous-unequal: EBs with undetected secondary eclipses

580: \item Ambiguous-equal: EB with equally-spaced and indistinguishable eclipses

581: \item Inverted: detached EBs that are not successfully modeled by MECI

582: \item Roche-lobe-filling: non-detached EBs that are filling at least one Roche-lobe

583: \item Abnormal: EBs with atypical out-of-eclipse distortions

584: \end{enumerate}

585: 

586: We list the model parameters for the EBs of groups I-IV in the

587: electronic version of this catalog (see full description in

588: appendix \ref{appendixCatalogDescription}). The EBs of groups

589: V-VII could not be well modeled by MECI, therefore we list only

590: their coordinates and periods, so that they can be followed-up.

591: 

592: Figure \ref{figPeriodDistrib} illustrates the period distribution

593: of these seven groups. Note however that both the orbital geometry

594: of EBs (eclipse probability $\propto P^{-2/3}$), and the limited

595: duration of the TrES survey data ($\leq 90\:days$ ; varies from

596: field to field ; see Table \ref{tableFieldsObs}), act to suppress

597: the detection of binaries with longer periods. An added

598: complication for single-telescope surveys is that about half of

599: the EBs with periods close to an integer number of days will not

600: be detectable, as they eclipse only during the daytime. This EB

601: distribution is consistent with the far deeper OGLE II field

602: catalog \citep{Devor05}, where the long tail of Roche-lobe-filling

603: systems has recently been explained by \citet{Derekas07} as being

604: the result of a strong selection towards detecting eclipsing giant

605: stars.

606: 

607: \placefigure{figPeriodDistrib}

608: 

609: Group [I] contains the eccentric EBs identified in step (4) as

610: having centers of eclipse that are separated by a duration

611: significantly different from half an orbital period (see Figures

612: \ref{figEcc1}, \ref{figEcc2}, and \ref{figEcc3}). This criterion

613: is sufficient for demonstrating eccentricity, but not necessary,

614: since we miss systems for which $\cos \omega \simeq 0$ (see

615: equation \ref{eqOmC}). Fortunately, we are able to detect

616: eccentricities in well-detached EBs with $|e \cos \omega| \geq

617: 0.005$, using eclipse timing. Therefore, assuming that $\omega$ is

618: uniformly distributed, we are approximately 67\% complete for $e

619: = 0.01$, and over 92\% complete for $e = 0.04$. In

620: principle, it would be possible to be 100\% complete for these

621: systems by measuring the differences in their eclipse durations,

622: however this measurement is known to be unreliable \citep{Etzel91}

623: and so would likely contaminate this group with false positives.

624: Group [II] consists of all such circular-orbit EBs that were

625: successfully fit by a single MECI model (see Figure

626: \ref{figE0Cat}).

627: 

628: \placefigure{figEcc1}

629: \placefigure{figEcc2}

630: \placefigure{figEcc3}

631: \placefigure{figE0Cat}

632: 

633: EBs with only one detectable eclipse can potentially be modeled in

634: two alternative ways. One way is to assume very unequal stellar

635: components, which have a very shallow undetected secondary eclipse

636: (group [III]). Since we cannot estimate the eccentricity of such

637: systems, we assume that they have circular orbits. The other way

638: is to assume that the period at hand is twice the correct value,

639: and that the components are nearly equal (group [IV]). The entries

640: of such ambiguous LCs were doubled in step (2), so that these two

641: solutions would be independently processed through the pipeline

642: (see Figure \ref{figAmbigCat}). Therefore, these two groups have a

643: one-to-one correspondence between them, although only one entry of

644: each pair can be correct. Resolving this ambiguity may not always

645: be possible without spectroscopic data. In some cases we were able

646: to resolve this ambiguity using either a morphological or a

647: physical approach. The morphological approach consists of manually

648: examining the LCs of group [IV] for any asymmetries in the two

649: eclipses (e.g. width, depth, or shape), or in the two plateaus

650: between the eclipses (e.g. perturbations due to tidal effects,

651: refections, or the ``O'Connell effect''). The physical approach

652: consists of applying our understanding of stellar evolution in

653: order to exclude entries that cannot be explained through any

654: coeval star pairing (see appendix \ref{appendixSingleEclipse}).

655: Either way, once one of the two models has been eliminated, the

656: other model is moved into group [II] and is adopted as a

657: non-ambiguous solution. It is interesting to note that when

658: analyzing the two models with MECI, the equal-component solution

659: (group [IV]) has masses approximately equal to the primary

660: component of the unequal-component solution (group [III]). The

661: mass of the unequal-component solution's secondary component will

662: typically be the smallest value listed in the isochrone table, as

663: this configuration will produce the least detectable secondary

664: eclipse.

665: 

666: 

667: \placefigure{figAmbigCat}

668: 

669: Group [V] consists of detached EBs that cannot be modeled by two

670: coeval stellar components. As mentioned earlier, we can reject the

671: single-eclipse solution for EBs with sufficiently deep eclipses

672: (see appendix \ref{appendixSingleEclipse}). This argument can be

673: further extended to cases where we can detect both eclipses in the

674: LC, but where one is far shallower than the other. In some cases,

675: no two coeval main-sequence components will reproduce such a LC,

676: but unlike the previous case, since both eclipses are seen, we

677: cannot conclude that the period needs to be doubled. Such systems

678: are likely to have had mass transfer from a sub-giant component

679: onto a main-sequence component through Roche-lobe overflow, to the

680: point where currently the main-sequence component has become

681: significantly more massive and brighter than it was originally

682: \citep{Crawford55}. This process will cause the components to

683: effectively behave as non-coeval stars, even though they have in

684: fact the same chronological age. In extreme cases, the originally

685: lower-mass main-sequence component can become more massive than

686: the sub-giant, and thus swap their original primary/secondary

687: designations, so that the main-sequence component is now the

688: primary component. We call such systems ``inverted'' EBs, and

689: place them into group [V] (see Figure \ref{figInvertedCat}). This

690: phenomenon is often referred to in the literature as the ``Algol

691: paradox'', though we chose not to adopt this term so as to avoid

692: confusing it with the term ``Algol-type EB'' (EA), which is

693: defined by the General Catalogue of Variable Stars [GCVS ;

694: \citep{Kukarkin48, Samus06}] as being the class of all

695: well-detached EBs.

696: 

697: \placefigure{figInvertedCat}

698: 

699: Group [VI] contains the EBs that have at least one component

700: filling its Roche-lobe (see Figure \ref{figRocheFilledCat}). Such

701: system cannot be well-fit by either DEBiL or MECI since they

702: assume that the binary components are detached, and so neglect

703: tidal and rotational distortions, gravity darkening, and

704: reflection effects. These systems must be separated from the rest

705: of the catalog since their resulting best-fit models will be poor

706: and therefore their evaluated physical attributes will likely be

707: erroneous. In a similar fashion to \citet{Tamuz06}, we detect

708: these systems automatically by applying the \citet{Eggleton83}

709: approximation for the Roche-lobe radius, and place in group [VI]

710: all the systems for which at least one of the EB components has

711: filled its Roche-lobe (see Figure \ref{figRoche}), that is, if

712: either one of the following two inequalities occurs:

713: 

714: \begin{eqnarray}

715: \label{eqRoche1}

716: r_1 &>& \frac{0.49\: q^{-2/3}}{0.6\: q^{-2/3} + \ln \left(1 + q^{-1/3}\right)} {\rm \ \ \ or}\\

717: \label{eqRoche2}

718: r_2 &>& \frac{0.49\: q^{2/3}}{0.6\: q^{2/3} + \ln \left(1 + q^{1/3}\right)}\ ,

719: \end{eqnarray}

720: 

721: where $q = M_2 / M_1$ is the EB components' mass ratio. Since we

722: expect non-detached EBs to be biased towards evolving, higher mass

723: stellar components, we estimated $q$ using the early-type

724: mass-radius power law relation found in binaries \citep{Gorda98}:

725: $q \simeq (r_2/r_1)^{1.534}$. Although in principle, we could have

726: estimated $q$ directly from the EB component masses resulting from

727: the MECI analysis, we chose not to, since as stated above, the

728: analysis of such systems is inaccurate. The analytic approximation

729: we used, though crude, proved to be remarkably robust, as we found

730: only 5 false negatives and no false positives when visually

731: inspecting the LCs. We found many more false positives/negatives

732: when using the alarm criteria suggested by \citet{Devor05} or

733: \citet{Mazeh06}, both of which attempt to identify bad model fits

734: by evaluating spatial correlations of the model's residuals.

735: 

736: \placefigure{figRocheFilledCat}

737: \placefigure{figRoche}

738: 

739: Finally, group [VII] contains systems visually identified as EBs

740: (i.e. having LCs with periodic flux dips), yet having atypical LC

741: perturbations that indicate the existence of additional physical

742: phenomena (see Figures \ref{figAbnormalEBs1} and

743: \ref{figAbnormalEBs2}). For lack of a better descriptor, we call

744: such systems ``abnormal'' (see further information in

745: \S\ref{subsecAbnomalEBs}). This group is different from the

746: previous six in that we cannot automate their classification, and

747: their selection is thus inherently subjective. In 15 of the 20

748: systems, we were able to approximately model the LCs, and included

749: them in one of the aforementioned groups. In these cases users

750: should be aware that these model may be biased by the phenomenon

751: that brought about their LC distortion.

752: 

753: \placefigure{figAbnormalEBs1}

754: \placefigure{figAbnormalEBs2}

755: 

756: 

757: \section{Results}

758: \label{secResults}

759: 

760: We identified and classified a total of 773 EBs\footnote{The

761: observed LCs, fitted models, and model residuals of each of these

762: EBs are shown at:\newline

763: http://www.cfa.harvard.edu/$\sim$jdevor/Catalog.html}. These

764: systems consisted of 734 EBs with circular orbits, 34 detached EBs

765: with eccentric orbits (group [I] ; Table \ref{tableEccentric}),

766: and 5 unclassified abnormal EBs (group [VII] ; Table

767: \ref{tableAbnormal}). We marked 15 of the detached EBs with

768: circular orbits as also being abnormal. Of the 734 EBs with

769: circular orbits, we classify 290 as unambiguous detached EBs

770: (group [II] ; Table \ref{tableCircular}), 103 as ambiguous

771: detached EBs, for which we could not determine photometrically if

772: they consisted of equal or disparate components (groups [III] and

773: [IV] ; Table \ref{tableAmbig}), 23 as inverted EBs (group [V] ;

774: Table \ref{tableInverted}), and 318 as non-detached (group [VI] ;

775: Table \ref{tableFillRoche}). With the exception of the abnormal

776: EBs, which were selected by eye, we use an automated method to

777: classify each of these groups (see \S\ref{secMethod} for details).

778: Our mass estimates for the primary and secondary components are

779: plotted in Figure \ref{figMassMass}.

780: 

781: \placefigure{figMassMass}

782: 

783: The EB discovery yield (the fraction of LCs found to be EBs),

784: varies greatly from field to field, ranging from 0.72\% for

785: Cygnus, to 0.15\% for Corona Borealis (see Table

786: \ref{tableFieldsYield}). This variation is strongly correlated

787: with Galactic latitude, where fields near the Galactic plane have

788: larger discovery yields than those that are farther from it (see

789: Figure \ref{figDiscoveryYield}). This effect is likely due to the

790: fact that fields closer to the Galactic plane contain a higher

791: fraction of early-type stars. These early-type stars are both

792: physically larger, making them more likely to be eclipsed, and are

793: more luminous, which has them produce brighter and less noisy LCs,

794: thereby enabling the detection of EBs with shallower eclipses.

795: Furthermore, much of the residual scatter can be attributed to the

796: variation in the observed duration of each field (see Table

797: \ref{tableFieldsObs}). That is, we find additional EBs, with

798: longer periods, in fields that were observed for a longer

799: duration.

800: 

801: \placefigure{figDiscoveryYield}

802: 

803: Currently, 88 of the cataloged EBs (11\%) appear in either the

804: International Variable Star Index\footnote{Maintained by the

805: American Association of Variable Star Observers (AAVSO).} (VSX),

806: or in the SIMBAD\footnote{Maintained by the Centre de Données

807: astronomiques de Strasbourg (CDS).} astronomical database (Table

808: \ref{tableSIMBAD}). However, only 49 systems (6\%) have been

809: identified as being variable. Not surprisingly, with few

810: exceptions, these targets were among the brightest sources of the

811: catalog. Using only photometry, it is often notoriously difficult

812: to distinguish non-detached EBs from pulsating variables that vary

813: sinusoidally in time, such as type-C RR Lyrae. Furthermore,

814: unevenly spotted stars may also cause false positive

815: identifications, especially in surveys with shorter durations.

816: Ultimately, spectroscopic follow-up will always be necessary to

817: confirm the identification of such variables.

818: 

819: We highlight three groups of EBs as potentially having special

820: importance as test beds for current theory. For more accurate

821: properties, these EBs will likely need to be followed up both

822: photometrically and spectroscopically. The brightness of these EBs

823: will considerably facilitate their follow-up.

824: 

825: 

826: \subsection{Low-mass EBs}

827: \label{subsecLowMassEBs}

828: 

829: The first group consists of 11 low-mass EB candidates, including

830: 10 newly discovered EBs with either K or M-dwarf stellar

831: components. Our criteria for selecting these binaries were that

832: they be well-detached, and that both components have estimated

833: masses below $0.75M_\sun$ (see Table \ref{tableLowMass} and Figure

834: \ref{figLowMassCat}). Currently, only 7 such detached low-mass EBs

835: have been confirmed [YY~Gem \citep{Kron52, Torres02}, CM~Dra

836: \citep{Lacy77a, Metcalfe96}, CU~Cnc \citep{Delfosse99, Ribas03},

837: T-Her0-07621 \citep{Creevey05}, GU~Boo \citep{LopezMorales05},

838: NSVS01031772 \citep{LopezMorales06}, and UNSW-TR-2

839: \citep{Young06}].

840: 

841: Despite a great deal of work that has been done to understand the

842: structure of low-mass stars [e.g. \citet{Chabrier00}], models

843: continue to underestimate their radii by as much as $15$\%

844: \citep{Lacy77b, Torres02, Creevey05, Ribas06}, a significant

845: discrepancy considering that for solar-type stars the agreement

846: with the observations is typically within $1-2$\%

847: \citep{Andersen91, Andersen98}. In recent years an intriguing

848: hypothesis has been put forward that strong magnetic fields may

849: have bloated these stars through chromospheric activity

850: \citep{Ribas06, Torres06, LopezMorales07, Chabrier07}.

851: Furthermore, \citet{Torres06} find that such bloating occurs even

852: for stars with nearly a solar mass, and suggest that this effect

853: may also be due to magnetically induced convective disruption. In

854: either case, these radius discrepancies should diminish for widely

855: separated binaries with long periods, as they become

856: non-synchronous and thus rotate slower, which according to dynamo

857: theory would reduce the strength of their magnetic fields.

858: 

859: Unfortunately, the small number of well characterized low-mass EBs

860: makes it difficult to provide strong observational constraints to

861: theory. Despite the fact that such stars make up the majority of

862: the Galactic stellar population, their intrinsic faintness renders

863: them extremely rare objects in magnitude-limited surveys. In

864: addition, once found, their low flux severely limits the ability

865: of observing their spectra with both sufficiently high resolution

866: and a high signal-to-noise ratio. To this end, the fact that the

867: TrES survey was made with small-aperture telescopes is a great

868: advantage, as any low-mass EB candidate found is guaranteed to be

869: bright, and thus require only moderate-aperture telescopes for

870: their follow-up. Thus we propose multi-epoch spectroscopic study

871: of the systems listed here, in order to confirm their low mass and

872: to estimate their physical properties with an accuracy sufficient

873: to test models of stellar structure. Moreover, two of our

874: candidates (T-Cyg1-12664 and T-Cas0-10450), if they are in fact

875: ambiguous-equal (group [IV]), have periods greater than 8 days,

876: making them prime targets for testing the aforementioned

877: magnetic-bloating hypothesis.

878: 

879: \placefigure{figLowMassCat}

880: 

881: 

882: 

883: \subsection{Eccentric EBs}

884: \label{subsecEccentricEBs}

885: 

886: The second group of EBs consists of 34 binaries with eccentric

887: orbits (see Table \ref{tableEccentric}, and Figures \ref{figEcc1},

888: \ref{figEcc2}, and \ref{figEcc3}). We were able to reliably

889: measure values of $|e \cos \omega|$ as low as $\sim0.005$ by using

890: the eclipse timing technique (see \S\ref{secMethod} and Figure

891: \ref{figTimingVariationsEcc}). Since this measure provides a lower

892: limit to the eccentricity, it is well-suited to identify eccentric

893: EBs, even though the actual value of the eccentricity may be

894: uncertain. As mentioned earlier, in an effort to avoid

895: false-positives, we do not include in this group EBs whose eclipse

896: timing measures $|e \cos \omega| < 0.005$, or EBs with an

897: eccentricity consistent with zero.

898: 

899: Our interest in these eccentric binaries stems from their

900: potential to constrain tidal circularization theory

901: \citep{Darwin1879}. This theory describes how the eccentricity of

902: a binary orbit decays over time due to tidal dissipation, with a

903: characteristic timescale ($t_{circ}$) that is a function of the

904: components' stellar structure and orbital separation. As long as

905: the components' stellar structure remains unchanged, the orbital

906: eccentricity is expected to decay approximately exponentially over

907: time [$e \propto \exp (-t/t_{circ})$]. However, once the

908: components evolve off the main sequence, this time scale may vary

909: considerably \citep{Zahn89}. Thus, to understand the

910: circularization history of binaries with circularization

911: timescales similar to or larger than their evolutionary

912: timescales, one must integrate over the evolutionary tracks of

913: both stellar components.

914: 

915: Three alternate tidal dissipation mechanisms have been proposed:

916: dynamical tides \citep{Zahn75, Zahn77}, equilibrium tides

917: \citep{Zahn77, Hut81}, and hydrodynamics \citep{Tassoul88}.

918: Despite its long period of development, the inherent difficulty of

919: observing tidal dissipation has prevented definitive conclusions.

920: \citet{Zahn89} add a further complication by maintaining that most

921: of the orbital circularization process takes place at the

922: beginning of the Hayashi phase, and that the eccentricity of a

923: binary should then remain nearly constant throughout its lifetime

924: on the main sequence.

925: 

926: Observational tests of these tidal circularization theories,

927: whereby $t_{circ}$ is measured statistically in coeval stellar

928: populations, have so far proved inconclusive. \citet{North03}

929: found that short-period binaries in both the Large and Small

930: Magellanic Clouds seem to have been circularized in agreement with

931: the theory of dynamical tides. However, \citet{Meibom05} show that

932: with the exception of the Hyades, the stars in the clusters that

933: they observed were considerably more circularized than any of the

934: known dissipation mechanisms would predict. Furthermore, they find

935: with a high degree of certainty, that older clusters are more

936: circularized than younger ones, thereby contradicting the Hayashi

937: phase circularization model.

938: 

939: Encouraged by the statistical effect of circularization that can

940: be seen in our catalog (Figure \ref{figPeriodEcc}), we further

941: estimated $t_{circ}$ for each of the eccentric systems as follows.

942: \citet{Zahn77, Zahn78} provides an estimate for the orbital

943: circularization timescale due to turbulent dissipation in stars

944: possessing a convective envelope, assuming that corotation has

945: been achieved:

946: 

947: \begin{equation}

948: t_{circ} = \frac{1}{21q(1+q)k_2} \left(\frac{MR^2}{L}\right)^{1/3}\left(\frac{a}{R}\right)^8

949: \end{equation}

950: 

951: where $M, R, L$ are the star's mass, radius and luminosity, and

952: $k_2$ is the apsidal motion constant of the star, which is

953: determined by its internal structure and dynamics.

954: 

955: More massive stars, which do not have a convective envelope but

956: rather develop a radiative envelope, are thought to circularize

957: their orbit using radiative damping \citep{Zahn75, Claret97}. This

958: is a far slower mechanism, whose circularization timescale can be

959: estimated by:

960: 

961: \begin{equation}

962: t_{circ} = \frac{2}{21 q (1+q)^{11/6} E_2} \left(\frac{R^3}{GM}\right)^{1/2} \left(\frac{a}{R}\right)^{21/2}

963: \end{equation}

964: 

965: where $E_2$ is the tidal torque constant of the star, and $G$ is

966: the universal gravitational constant. We can greatly simplify

967: these expressions by applying Kepler's law [$a^3 =

968: GM(1+q)(P/{2\pi})^2$], and adopt the \citet{Cox00} power law

969: approximations for the main-sequence mass-radius and

970: mass-luminosity relations. For the convective envelope case, we

971: adopt the late-type mass-radius relation ($M < 1.3 M_\sun$), and

972: for the radiative envelope case we adopt the early-type

973: mass-radius relation ($M \geq 1.3 M_\sun$), thus arriving at:

974: 

975: \begin{equation}

976: \label{eqCircTime}

977: t_{circ} \simeq \cases{

978: 0.53 Myr \; (k_2 / 0.005)^{-1}q^{-1}(1+q)^{5/3} \left(P / day\right)^{16/3} \left(M / M_\sun \right)^{-4.99}, \ M < 1.3 M_\sun\cr

979: 1370 Myr \; (E_2 / 10^{-8})^{-1}q^{-1}(1+q)^{5/3} \left(P / day\right)^7 \left(M / M_\sun \right)^{-2.76}, \ M \geq 1.3 M_\sun \cr}

980: \end{equation}

981: 

982: Determining the value of $k_2$ and $E_2$ is the most difficult

983: part of this exercise, since their values are a function of the

984: detailed structure and dynamics of the given star, which in turn

985: changes significantly as the star evolves \citep{Claret97,

986: Claret02}. In our calculation, we estimate these values by

987: interpolating published theoretical tables [$k_2$: \citet{Zahn94},

988: $E_2$: \citet{Zahn75, Claret97}]. Since both stellar components

989: contribute to the circularization process, the combined

990: circularization timescale becomes $t_{circ} = 1/(t_{circ,1}^{-1} +

991: t_{circ,2}^{-1})$, where the subscripts $1$ and $2$ refer to the

992: primary and secondary binary components \citep{Claret97}. In Table

993: \ref{tableEccentric}, we list the combined circularization

994: timescale for each of the eccentric EBs we identify.

995: 

996: The value of $t_{circ}$ for most of the eccentric systems (21 of

997: 34) is larger than the Hubble time, indicating that no significant

998: circularization is expected to have taken place since they settled

999: on the main sequence. About a quarter of the eccentric systems (8

1000: of 34) have a $t_{circ}$ smaller than the Hubble time but larger

1001: than $1\:Gyr$. While circularization is underway, the fact that

1002: they are still eccentric is consistent with theoretical

1003: expectations. The remaining systems (5 of 34) all have $t_{circ} <

1004: 1\:Gyr$, have periods less than 3.3 days, and unless they are

1005: extremely young, require an explanation for their eccentric

1006: orbits. Two of these EBs (T-Tau0-02487 and T-Tau0-03916) are

1007: located near the star forming regions of Taurus, supporting the

1008: hypothesis that they are indeed young. However, this hypothesis

1009: does not seem to be adequate for T-Cas0-02603, which has a period

1010: of only 2.2 days and $t_{circ} \simeq 0.26\:Gyr$, while possessing

1011: a large eccentricity of $e \simeq 0.25$. An alternative

1012: explanation is that some of these binaries were once further

1013: apart, having larger orbital periods, and thus larger

1014: circularization timescales. These systems may have been involved

1015: in a comparably recent interaction with a third star (a collision

1016: or near miss), or have been influenced by repeated resonant

1017: perturbations of a tertiary companion.

1018: 

1019: Finally, we would like to draw the reader's attention to our

1020: shortest-period eccentric EB, T-Cas0-00394, whose period is a mere

1021: 1.7 days. Notably, this system is entirely consistent with theory,

1022: since its mass falls in a precarious gap, where the stellar

1023: envelopes of its components are no longer convective, yet their

1024: radiative envelopes are not sufficiently extended to produce

1025: significant tidal drag (see Figure \ref{figPeriodM1}).

1026: 

1027: \placefigure{figPeriodEcc}

1028: \placefigure{figPeriodM1}

1029: 

1030: 

1031: \subsection{Abnormal EBs}

1032: \label{subsecAbnomalEBs}

1033: 

1034: The third group of EBs consists of 20 abnormal systems (see Table

1035: \ref{tableAbnormal}, and Figures \ref{figAbnormalEBs1} and

1036: \ref{figAbnormalEBs2}). While possessing the distinctive

1037: characteristics of EBs, these LCs stood out during manual

1038: inspection for a variety of reasons. These systems underline the

1039: difficulty of fully automating any LC pipeline, as any such system

1040: will inevitably need to recognize atypical EBs that were not

1041: encountered before.

1042: 

1043: The LCs we listed can be loosely classified into groups according

1044: to the way they deviate from a simple EB model. A few cases

1045: exhibited pulsation-like fluctuations that were not synchronized

1046: with the EB period (shorter-period: T-Dra0-00398, longer-period:

1047: T-Lyr1-00359, T-Per1-00750). These fluctuations may be due either

1048: to the activity of an EB component, or to a third star whose light

1049: is blended with the binary. In principle, one can identify the

1050: active star by examining the amplitude of the fluctuations during

1051: the eclipses. If the fluctuations originate from one of the

1052: components, their observed amplitude will be reduced when the

1053: component is being eclipsed. In such a case, if the fluctuations

1054: are due to pulsations, they can further provide independent

1055: constraints to the stellar properties through astro-seismological

1056: models \citep{Mkrtichian04}. To identify such fluctuating EBs one

1057: must subtract the fitted EB model from the LC, and evaluate the

1058: residuals [e.g. \citet{Pilecki07}]. When the fluctuation period is

1059: fixed, one can simply search the residual LC using a periodogram,

1060: as was done in step (1) of our pipeline (see \S\ref{secMethod}).

1061: However, when the fluctuation period varies (i.e. non-coherent),

1062: as in the aforementioned LCs, one must employ alternate methods,

1063: since simply phasing their LC will not produce any discernable

1064: structure. For LCs with long-period fluctuations one can directly

1065: search the residuals for time dependencies, while for LCs with

1066: short-period fluctuations one can search the residuals for

1067: non-Gaussian distributions. However, in practice these

1068: measurements will likely not be robust, as there are many

1069: instrumental effects that can produce false positives. Thus, we

1070: employ a search for auto-correlations in the residual time series,

1071: which overcomes most instrumental effects, while providing a

1072: reliable indicator for many types of pseudo-periodic fluctuations.

1073: 

1074: The remaining systems had LC distortions that appear to be

1075: synchronized with the orbital period. The source of these

1076: fluctuations is likely due to long-lasting surface inhomogeneities

1077: on one or both of the rotationally synchronized components. When

1078: the LC has brief periodic episodes of darkening (T-And0-11476,

1079: T-Cas0-13944, T-Cyg1-07584, T-Dra0-04520), they can usually be

1080: explained as stable star spots, but brief periodic episodes of

1081: brightening (T-And0-04594, T-Her0-08091), which may indicate the

1082: presence of stable hot-spots, are more difficult to interpret.

1083: This phenomenon is especially puzzling in the aforementioned two

1084: cases, in which the brightening episodes are briefer than one

1085: would expect from a persistent surface feature and repeat at the

1086: middle of both plateaus.

1087: 

1088: When the two plateaus of a LC are not flat, they are usually

1089: symmetric about the center of the eclipses. This is due to the

1090: physical mirror symmetry about the line intersecting the binary

1091: components' centers. When the axis of symmetry does not coincide

1092: with the center of eclipse (T-And0-00920, T-Cyg1-08866,

1093: T-Dra0-03105, T-Lyr1-07584, T-Lyr1-15595), a phenomenon we term

1094: ``eclipse offset'', we conclude that this symmetry must somehow be

1095: broken. This may occur if the EB components are not rotationally

1096: synchronized, or have a substantial tidal lag. Another form of

1097: this asymmetry can appear as an amplitude difference between the

1098: two LC plateaus (T-Her0-03497, T-Lyr1-13166, T-Per1-08789,

1099: T-UMa0-03090). This phenomenon, which was originally called the

1100: ``periastron effect'' and has since been renamed the ``O'Connell

1101: effect'', has been known for over a century, and has been

1102: extensively studied [e.g. \citet{OConnell51, Milone86}]. Classic

1103: hypotheses suggest an uneven distribution of circumstellar

1104: material orbiting with the binary \citep{Struve48} or surrounding

1105: the stars \citep{Mergentaler50}, either of which could induce a

1106: preferential $H^-$ absorption on one side. \citet{Binnendijk60}

1107: was the first of many to suggested that the this asymmetry is due

1108: to subluminous regions of the stellar surface (i.e. star spots).

1109: However, this explanation also requires the stars to be

1110: rotationally synchronized, and for the spots to be stable over the

1111: duration of the observations. Alternate models abound, including a

1112: hot spot on one side of a component brought about through mass

1113: transfer from the other component, persistent star spots created

1114: by an off-axis magnetic field, and circumstellar material being

1115: captured by the components and heating one side of both stars

1116: \citep{Liu03}. As with many phenomena that have multiple possible

1117: models, the true answer may involve a combination of a number of

1118: these mechanisms, and will likely vary from system to system

1119: \citep{Davidge84}.

1120: 

1121: Finally, a few particularly unusual LCs (T-Dra0-03105,

1122: T-Lyr1-05984) display a very large difference between their

1123: eclipse durations. Although a moderate difference could be

1124: explained by an eccentric orbit, such extreme eccentricities in

1125: systems with such short orbital periods (0.5 and 1.5 days) are

1126: highly unlikely.

1127: 

1128: \section{Conclusions}

1129: 

1130: We presented a catalog of 773 eclipsing binaries found in 10

1131: fields of the TrES survey, identified and analyzed using an

1132: automated pipeline. We described the pipeline we used to identify

1133: and model them. The pipeline was designed to be mostly automated,

1134: with manual inspections taking place only once the vast majority

1135: of non-EB LCs had been automatically filtered out. At the final

1136: stage of the pipeline we classified the EBs into 7 groups:

1137: eccentric, circular, ambiguous-equal, ambiguous-unequal, inverted,

1138: Roche-lobe-filling, and abnormal. The former four groups were all

1139: successfully modeled with our model fitting program. However, the

1140: latter three groups possessed significant additional physical

1141: phenomena (tidal distortions, mass-transfer, and surface

1142: activity), which did not conform to the simple detached-EB model

1143: we employed.

1144: 

1145: We highlighted three groups of binaries, which may be of

1146: particular interest and warrant follow-up observations. These

1147: groups are: low-mass EBs, EBs with eccentric orbits, and abnormal

1148: EBs. The low-mass EBs (both components $< 0.75M_{\sun}$) allow one

1149: to probe the mass-radius relation at the bottom of the

1150: main-sequence. Only 7 such EBs have previously been confirmed, and

1151: the physical properties of many of them are inconsistent with

1152: current theoretical models. Our group of 10 new candidates will

1153: likely provide considerable additional constraints to the models,

1154: and the discovery of 2 long period systems could help confirm a

1155: recent hypothesis that this inconsistency is due to stellar

1156: magnetic activity. The eccentric-orbit EBs may help confirm and

1157: constrain tidal circularization theory, as many of them have

1158: comparably short circularization timescales. We demonstrated that,

1159: as one would predict from the theory, the shortest period systems

1160: fall within a narrow range of masses, in which their stellar

1161: envelopes cease to be convective yet their envelopes are not

1162: extended enough to produce significant tidal drag. The abnormal

1163: EBs seem to show a plethora of effects that are indicative of

1164: asymmetries, stellar activity, persistent hot and cold spots, and

1165: a host of other physical phenomena. Some of these systems may

1166: require dedicated study to be properly understood.

1167: 

1168: In the future, as LC datasets continue to grow, it will become

1169: increasingly necessary to use such automated pipelines to identify

1170: rare and interesting targets. Such systematic searches promise a

1171: wealth of data that can be used to test and constrain theories in

1172: regions of their parameter-space that were previously

1173: inaccessible. Furthermore, even once the physics of ``vanilla''

1174: systems has been solved, more complex cases will emerge to

1175: challenge us to achieve a better understanding of how stars form,

1176: evolve and interact.

1177: 

1178: \acknowledgments

1179: 

1180: We would like to thank Tsevi Mazeh for many useful discussions, as

1181: well as S{\o}ren Meibom for his repeated help. We would also like

1182: to thank Sarah Dykstra for her continuous support throughout the

1183: preparation of this paper. We are thankful to the staff of the

1184: Palomar Observatory for their assistance in operating the Sleuth

1185: instrument, and we acknowledge support from NASA through grant

1186: NNG05GJ29G issued through the Origins of Solar Systems Program.

1187: This research has made use of the NASA's Astrophysics Data System

1188: Bibliographic Services, the SIMBAD database, operated at CDS,

1189: Strasbourg, and the VSX database, which was created by Christopher

1190: Watson for the AAVSO. This publication also utilizes data products

1191: from the Two Micron All Sky Survey, which is a joint project of

1192: the University of Massachusetts and the Infrared Processing and

1193: Analysis Center/California Institute of Technology, funded by NASA

1194: and NSF. Finally, we would like to thanks the anonymous referee

1195: for very insightful comments and suggestions, which significantly

1196: improved this manuscript.

1197: 

1198: 

1199: \appendix

1200: 

1201: \section{Rejecting single-eclipse EB Models}

1202: \label{appendixSingleEclipse}

1203: 

1204: An EB LC comprised of a deep eclipse and a very shallow eclipse,

1205: can occur in one of two ways. Either the secondary component is

1206: luminous but extremely small (e.g. a white dwarf observed in UV),

1207: thus producing a shallow primary eclipse, or the secondary

1208: component is comparably large but extremely dim, thus producing a

1209: shallow secondary eclipse. The first case, though possible [e.g.

1210: \citet{Maxted04}], is extremely rare, and will have a signature

1211: ``flat bottom'' to the eclipse. We have not encountered such a LC

1212: in our dataset. The second case will have a rounded eclipse

1213: bottom, due to the primary component's limb darkening. Assuming

1214: this latter contingency, in which the secondary component is dark

1215: in comparison to the primary component, we can place a lower bound

1216: to its radius ($R_2$):

1217: 

1218: \begin{equation}

1219: R_2 \geq R_1 \sqrt{1 - 10^{-0.4 \Delta mag_1}}\ ,

1220: \end{equation}

1221: 

1222: where $R_1$ is the radius of the primary component, and $\Delta

1223: mag_1$ is the magnitude depth of the primary eclipse. Thus, if the

1224: eclipse is very deep, the size of the secondary component must

1225: approach the size of the primary component. However, coeval

1226: short-period detached EBs with components of similar sizes yet

1227: desperate luminosities are expected to be very rare, assuming they

1228: follow normal stelar evolution. Therefore, if only one eclipse is

1229: detected, and it is both rounded and sufficiently deep, we may

1230: conclude that this configuration entry is likely to be incorrect,

1231: and that the correct configuration has double the orbital period

1232: and produces two equal eclipses. Only when we cannot apply such a

1233: period doubling solution (i.e. when the secondary eclipse is

1234: detectable), do we resort to questioning our assumption of normal

1235: stelar evolution (see classification group V, described in

1236: \S\ref{secMethod}).

1237: 

1238: 

1239: \section{Description of the Catalog Fields}

1240: \label{appendixCatalogDescription}

1241: 

1242: Due to the large size of the catalog we were only able to list

1243: small excerpts of it in the body of this paper. Readers interested

1244: in viewing the catalog in its entirety can download it

1245: electronically. Note that although the catalog lists 773 unique

1246: systems, each of the 103 ambiguous EBs appear in both possible

1247: configurations (see \S\ref{secMethod}), raising the total number

1248: of catalog entries to 876. Below, we briefly describe the

1249: catalog's 38 columns. The column units, if any, are listed in

1250: square brackets.

1251: 

1252: \renewcommand{\theenumi}{\arabic{enumi}}

1253: \begin{enumerate}

1254: \item $Category$ - the EB's classification (see \S\ref{secMethod}).

1255: \item $Binary\ name$ - the EB's designation, which is composed of its TrES field (see Table \ref{tableFieldsObs}) and index.

1256: \item $\alpha$ - the EB's right ascension (J2000).

1257: \item $\delta$ - the EB's declination (J2000).

1258: \item $Period$ [days] - the EB's orbital period.

1259: \item $Period \ uncertainty$ [days] - the uncertainty in the EB's orbital period.

1260: \item $Mass_1$ [$M_\sun$] - the mass of the EB's primary (more massive) component.

1261: \item $Mass_1\ uncertainty$ [$M_\sun$] - the uncertainty in the primary component's mass.

1262: \item $Mass_2$ [$M_\sun$] - the mass of the EB's secondary (less massive) component.

1263: \item $Mass_2\ uncertainty$ [$M_\sun$] - the uncertainty in the secondary component's mass.

1264: \item $Age$ [Gyr] - the age of the EB (assumed to be coeval).

1265: \item $Age\ uncertainty$ [Gyr] - the uncertainty in the EB's age.

1266: \item $Score$ - a weighted reduced $\chi^2$ of the MECI model fit [see \citet{Devor06b} for further details].

1267: \item $Isochrone\ source$ - isochrone tables used [Y2: \citet{Kim02}, or Baraffe: \citet{Baraffe98}].

1268: \item $Color\ weighting$ - the relative weight ($w$) of the LC fit, compared to the color fit [see \citet{Devor06b} for further details].

1269: \item $PM\ source$ - the database that provided the proper motion measurement [UCAC: \citet{Zacharias04}, USNO-B: \citet{Monet03}, or Salim03: \citet{Salim03}].

1270: \item $PM_{\alpha}$ [mas/yr] - the right ascension component of the EB's proper motion.

1271: \item $PM_{\delta}$ [mas/yr] - the declination component of the EB's proper motion.

1272: \item $Location\ error$ [arcsec] - the distance between our listed location (columns 3 and 4) and the location listed by the proper motion database.

1273: \item $mag_B$ - the USNO-B $B$-band observational magnitude of the EB (average of both magnitude measurements, if available).

1274: \item $mag_R$ - the USNO-B $R$-band observational magnitude of the EB (average of both magnitude measurements, if available).

1275: \item $Third-light\ fraction$ - the fraction of third-light flux ($R$-band) blended into the LC (i.e. the flux within 30", excluding the target, divided by the total flux within 30").

1276: \item $mag_J$ - the 2MASS observational $J$-band magnitude of the EB, converted to ESO $J$-band.

1277: \item $mag_H$ - the 2MASS observational $H$-band magnitude of the EB, converted to ESO $H$-band.

1278: \item $mag_K$ - the 2MASS observational $K_s$-band magnitude of the EB, converted to ESO $K$-band.

1279: \item $Mag_J$ - the absolute ESO $J$-band magnitude of the EB listed in the isochrone tables.

1280: \item $Mag_H$ - the absolute ESO $H$-band magnitude of the EB listed in the isochrone tables.

1281: \item $Mag_K$ - the absolute ESO $K$-band magnitude of the EB listed in the isochrone tables.

1282: \item $Distance$ [pc] - the distance to the EB, as calculated from the extinction-corrected distance modulus.

1283: \item $A(V)$ - the EB's V-mag absorption due to Galactic interstellar extinction (assuming $R_V = 3.1$).

1284: \item $\sin(i)$ - the sine of the EB's orbital inclination.

1285: \item $|e \cos(\omega)|$ - a robust lower limit for the EB's eccentricity (see equation \ref{eqOmC}).

1286: \item $Eccentricity$ - the orbital eccentricity of the EB.

1287: \item $Eccentricity\ uncertainty$ - the uncertainty in the orbital eccentricity of the EB.

1288: \item $\Delta mag_1$ - the $r$-band primary (deeper) eclipse depth in magnitudes.

1289: \item $Epoch_1$ - the Heliocentric Julian date (HJD) at the center of a primary eclipse, minus 2400000.

1290: \item $\Delta mag_2$ - the $r$-band secondary (shallower) eclipse depth in magnitudes.

1291: \item $Epoch_2$ - the Heliocentric Julian date (HJD) at the center of a secondary eclipse, minus 2400000.

1292: \end{enumerate}

1293: 

1294: Note that the value of the uncertainties (columns 6, 8 10, 12, and

1295: 34), were calculated by measuring the curvature of the

1296: parameter-space $\chi^2$ contour, near its minimum. This method

1297: implicitly assumes a Gaussian distribution of the parameter

1298: likelihood. If the likelihood distribution not Gaussian, but

1299: rather has a flattened (boxy) distribution, then the computed

1300: uncertainty becomes large. In extreme cases the estimated formal

1301: uncertainty can be larger than the measurement itself.

1302: 

1303: 

1304: \begin{thebibliography}{}

1305: 

1306: \bibitem[Agol et al.(2005)]{Agol05} Agol, E., Steffen, J., Sari, R., \& Clarkson, W.\ 2005, \mnras, 359, 567

1307: 

1308: \bibitem[Akerlof et al.(2000)]{Akerlof00} Akerlof, C., et al.\ 2000, \aj, 119, 1901

1309: 

1310: \bibitem[Alard \& Guibert(1997)]{Alard97} Alard, C.~\& Guibert, J.\ 1997, \aap, 326, 1

1311: 

1312: \bibitem[Alcock et al.(1998)]{Alcock98} Alcock, C., et al.\ 1998, \apj, 492, 190

1313: 

1314: \bibitem[Alonso et al.(2004)]{Alonso04} Alonso, R., et al.\ 2004, \apjl, 613, L153

1315: 

1316: \bibitem[Andersen(1991)]{Andersen91} Andersen, J.\ 1991, \aapr, 3, 91

1317: 

1318: \bibitem[Andersen(1998)]{Andersen98} Andersen, J.\ 1998, IAU Symp.~189: Fundamental Stellar Properties, 189, 99

1319: 

1320: \bibitem[Baglin \& The COROT Team(1998)]{Baglin98} Baglin, A., \& The COROT Team 1998, New Eyes to See Inside the Sun and Stars, 185, 301

1321: 

1322: \bibitem[Bakos et al.(2004)]{Bakos04} Bakos, G., Noyes, R.~W., Kov{\' a}cs, G., Stanek, K.~Z., Sasselov, D.~D.,~\& Domsa, I.\ 2004, \pasp, 116, 266

1323: 

1324: \bibitem[Baraffe et al.(1998)]{Baraffe98} Baraffe, I., Chabrier, G., Allard, F., \& Hauschildt, P.~H.\ 1998, \aap, 337, 403

1325: 

1326: \bibitem[Beaulieu et al.(1995)]{Beaulieu95} Beaulieu, J.~P., et al.\ 1995, \aap, 299, 168

1327: 

1328: \bibitem[Binnendijk(1960)]{Binnendijk60} Binnendijk, L.\ 1960, \aj, 65, 358

1329: 

1330: \bibitem[Borucki et al.(1997)]{Borucki97} Borucki, W.~J., Koch, D.~G., Dunham, E.~W., \& Jenkins, J.~M.\ 1997, Planets Beyond the Solar System and the Next Generation of Space Missions, 119, 153

1331: 

1332: \bibitem[Carpenter(2001)]{Carpenter01} Carpenter, J.~M.\ 2001, \aj, 121, 2851

1333: 

1334: \bibitem[Chabrier \& Baraffe(2000)]{Chabrier00} Chabrier, G., \& Baraffe, I.\ 2000, \araa, 38, 337

1335: 

1336: \bibitem[Chabrier et al.(2007)]{Chabrier07} Chabrier, G., Gallardo, J., \& Baraffe, I.\ 2007, ArXiv e-prints, 707, arXiv:0707.1792

1337: 

1338: \bibitem[Charbonneau et al.(2000)]{Charbonneau00} Charbonneau, D., Brown, T.~M., Latham, D.~W., \& Mayor, M.\ 2000, \apjl, 529, L45

1339: 

1340: \bibitem[Charbonneau et al.(2007)]{Charbonneau07} Charbonneau, D., Brown, T.~M., Burrows, A., \& Laughlin, G.\ 2007, Protostars and Planets V, 701

1341: 

1342: \bibitem[Christian et al.(2006)]{Christian06} Christian, D.~J., et al.\ 2006, \mnras, 372, 1117

1343: 

1344: \bibitem[Claret \& Cunha(1997)]{Claret97} Claret, A., \& Cunha, N.~C.~S.\ 1997, \aap, 318, 187

1345: 

1346: \bibitem[Claret(1998)]{Claret98} Claret, A.\ 1998, \aap, 335, 647

1347: 

1348: \bibitem[Claret(2000)]{Claret00} Claret, A.\ 2000, \aap, 363, 1081

1349: 

1350: \bibitem[Claret \& Willems(2002)]{Claret02} Claret, A., \& Willems, B.\ 2002, \aap, 388, 518

1351: 

1352: \bibitem[Crawford(1955)]{Crawford55} Crawford, J.~A.\ 1955, \apj, 121, 71

1353: 

1354: \bibitem[Creevey et al.(2005)]{Creevey05} Creevey, O.~L., et al.\ 2005, \apjl, 625, L127

1355: 

1356: \bibitem[Cox(2000)]{Cox00} Cox, A.~N.\ 2000, Allen's astrophysical quantities, (4th ed.; New York: AIP Press; Springer)

1357: 

1358: \bibitem[Darwin(1879)]{Darwin1879} Darwin, G.~H.\ 1879, The Observatory, 3, 79

1359: 

1360: \bibitem[Davidge \& Milone(1984)]{Davidge84} Davidge, T.~J., \& Milone, E.~F.\ 1984, \apjs, 55, 571

1361: 

1362: \bibitem[Deeg et al.(2000)]{Deeg00} Deeg, H.~J., Doyle, L.~R., Kozhevnikov, V.~P., Blue, J.~E., Mart{\'{\i}}n, E.~L., \& Schneider, J.\ 2000, \aap, 358, L5

1363: 

1364: \bibitem[Delfosse et al.(1999)]{Delfosse99} Delfosse, X., Forveille, T., Mayor, M., Burnet, M., \& Perrier, C.\ 1999, \aap, 341, L63

1365: 

1366: \bibitem[Derekas et al.(2007)]{Derekas07} Derekas, A., Kiss, L.~L., \& Bedding, T.~R.\ 2007, ArXiv Astrophysics e-prints, arXiv:astro-ph/0703137

1367: 

1368: \bibitem[Devor(2005)]{Devor05} Devor, J.\ 2005, \apj, 628, 411

1369: 

1370: \bibitem[Devor \& Charbonneau(2006a)]{Devor06a} Devor, J., \& Charbonneau, D.\ 2006a, \apss, 304, 351

1371: 

1372: \bibitem[Devor \& Charbonneau(2006b)]{Devor06b} Devor, J., \& Charbonneau, D.\ 2006b, \apj, 653, 648

1373: 

1374: \bibitem[Dunham et al.(2004)]{Dunham04} Dunham, E.~W., Mandushev, G.~I., Taylor, B.~W., \& Oetiker, B.\ 2004, \pasp, 116, 1072

1375: 

1376: \bibitem[Eggleton(1983)]{Eggleton83} Eggleton, P.~P.\ 1983, \apj, 268, 368

1377: 

1378: \bibitem[Etzel(1991)]{Etzel91} Etzel, P.~B.\ 1991, International Amateur-Professional Photoelectric Photometry Communications, 45, 25

1379: 

1380: \bibitem[Gilmore et al.(1998)]{Gilmore98} Gilmore, G., Perryman, M. A. C., Lindegren, L., et al.\ 1998, Proc. SPIE Conf., 3350, 541

1381: 

1382: \bibitem[Gorda \& Svechnikov(1998)]{Gorda98} Gorda, S.~Y.~\& Svechnikov, M.~A.\ 1998, Astronomy Reports, 42, 793

1383: 

1384: \bibitem[Henry et al.(2000)]{Henry00} Henry, G.~W., Marcy, G.~W., Butler, R.~P., \& Vogt, S.~S.\ 2000, \apjl, 529, L41

1385: 

1386: \bibitem[Holman \& Murray(2005)]{Holman05} Holman, M.~J., \& Murray, N.~W.\ 2005, Science, 307, 1288

1387: 

1388: \bibitem[Hut(1981)]{Hut81} Hut, P.\ 1981, \aap, 99, 126

1389: 

1390: \bibitem[Iben \& Tutukov(1984)]{Iben84} Iben, I., Jr., \& Tutukov, A.~V.\ 1984, \apjs, 54, 335

1391: 

1392: \bibitem[Kaiser et al.(2002)]{Kaiser02} Kaiser, N., et al.\ 2002, \procspie, 4836, 154

1393: 

1394: \bibitem[Kim et al.(2002)]{Kim02} Kim, Y., Demarque, P., Yi, S.~K.,~\& Alexander, D.~R.\ 2002, \apjs, 143, 499

1395: 

1396: \bibitem[Kirkpatrick et al.(1983)]{Kirkpatrick83} Kirkpatrick, S., Gelatt, C.~D.,~\& Vecchi, M.~P.\ 1983, Science, 220, 671

1397: 

1398: \bibitem[Kov{\'a}cs et al.(2002)]{Kovacs02} Kov{\'a}cs, G., Zucker, S., \& Mazeh, T.\ 2002, \aap, 391, 369

1399: 

1400: \bibitem[Kron(1952)]{Kron52} Kron, G.~E.\ 1952, \apj, 115, 301

1401: 

1402: \bibitem[Kukarkin(1948)]{Kukarkin48} Kukarkin B.~V., Parengo P.~P.\ 1948, General Catalogue of Variable Stars, Nauka Publ. House, Moscow

1403: 

1404: \bibitem[Kurucz(1992)]{Kurucz92} Kurucz, R.~L.\ 1992, IAU Symp.~149: The Stellar Populations of Galaxies, 149, 225

1405: 

1406: \bibitem[Lacy(1977a)]{Lacy77b} Lacy, C.~H.\ 1977a, \apj, 218, 444

1407: 

1408: \bibitem[Lacy(1977b)]{Lacy77a} Lacy, C.~H.\ 1977b, \apjs, 34, 479

1409: 

1410: \bibitem[Liu \& Yang(2003)]{Liu03} Liu, Q.-Y., \& Yang, Y.-L.\ 2003, Chinese Journal of Astronomy and Astrophysics, 3, 142

1411: 

1412: \bibitem[L{\'o}pez-Morales \& Ribas(2005)]{LopezMorales05} L{\'o}pez-Morales, M., \& Ribas, I.\ 2005, \apj, 631, 1120

1413: 

1414: \bibitem[L{\'o}pez-Morales et al.(2006)]{LopezMorales06} L{\'o}pez-Morales, M., Orosz, J.~A., Shaw, J.~S., Havelka, L., Arevalo, M.~J., McIntyre, T., \& Lazaro, C.\ 2006, ArXiv Astrophysics e-prints, arXiv:astro-ph/0610225

1415: 

1416: \bibitem[L{\'o}pez-Morales(2007)]{LopezMorales07} L{\'o}pez-Morales, M.\ 2007, \apj, 660, 732

1417: 

1418: \bibitem[Mandushev et al.(2005)]{Mandushev05} Mandushev, G., et al.\ 2005, \apj, 621, 1061

1419: 

1420: \bibitem[Maxted et al.(2004)]{Maxted04} Maxted, P.~F.~L., Marsh, T.~R., Morales-Rueda, L., Barstow, M.~A., Dobbie, P.~D., Schreiber, M.~R., Dhillon, V.~S., \& Brinkworth, C.~S.\ 2004, \mnras, 355, 1143

1421: 

1422: \bibitem[Mazeh et al.(2000)]{Mazeh00} Mazeh, T., et al.\ 2000, \apjl, 532, L55

1423: 

1424: \bibitem[Mazeh et al.(2006)]{Mazeh06} Mazeh, T., Tamuz, O., \& North, P.\ 2006, \mnras, 367, 1531

1425: 

1426: \bibitem[McCullough et al.(2006)]{McCullough06} McCullough, P.~R., et al.\ 2006, \apj, 648, 1228

1427: 

1428: \bibitem[Meibom \& Mathieu(2005)]{Meibom05} Meibom, S., \& Mathieu, R.~D.\ 2005, \apj, 620, 970

1429: 

1430: \bibitem[Mergentaler(1950)]{Mergentaler50} Mergentaler, J.\ 1950, Wroclaw Cont., 4, 1

1431: 

1432: \bibitem[Metcalfe et al.(1996)]{Metcalfe96} Metcalfe, T.~S., Mathieu, R.~D., Latham, D.~W., \& Torres, G.\ 1996, \apj, 456, 356

1433: 

1434: \bibitem[Milone(1986)]{Milone86} Milone, E.~F.\ 1986, \apjs, 61, 455

1435: 

1436: \bibitem[Mkrtichian et al.(2004)]{Mkrtichian04} Mkrtichian, D.~E., et al.\ 2004, \aap, 419, 1015

1437: 

1438: \bibitem[Monet et al.(2003)]{Monet03} Monet, D.~G., et al.\ 2003, \aj, 125, 984

1439: 

1440: \bibitem[Nelder(1965)]{Nelder65} Nelder, J.~A.,~\& Mead 1965, R., Computer Journal, 7, 308

1441: 

1442: \bibitem[North \& Zahn(2003)]{North03} North, P., \& Zahn, J.-P.\ 2003, \aap, 405, 677

1443: 

1444: \bibitem[O'Connell(1951)]{OConnell51} O'Connell, D.~J.~K.\ 1951, Publications of the Riverview College Observatory, 2, 85

1445: 

1446: \bibitem[O'Donovan et al.(2004)]{ODonovan04} O'Donovan, F.~T., Charbonneau, D., \& Kotredes, L.\ 2004, The Search for Other Worlds, 713, 169

1447: 

1448: \bibitem[O'Donovan et al.(2006)]{ODonovan06} O'Donovan, F.~T., et al.\ 2006, \apj, 644, 1237

1449: 

1450: \bibitem[O'Donovan et al.(2007)]{ODonovan07} O'Donovan, F.~T., et al.\ 2007, \apj, 662, 658

1451: 

1452: \bibitem[Paczynski(1986)]{Paczynski86} Paczynski, B.\ 1986, \apj, 304, 1

1453: 

1454: \bibitem[Paczynski(1997)]{Paczynski97} Paczynski, B.\ 1997, The Extragalactic Distance Scale, 273

1455: 

1456: \bibitem[Pedoussaut et al.(1988)]{Pedoussaut88} Pedoussaut, A., Carquillat, J.~M., Ginestet, N., \& Vigneau, J.\ 1988, \aaps, 75, 441

1457: 

1458: \bibitem[Pilecki \& Szczygiel(2007)]{Pilecki07} Pilecki, B., \& Szczygiel, D.~M.\ 2007, Informational Bulletin on Variable Stars, 5768, 1

1459: 

1460: \bibitem[Pojmanski(1997)]{Pojmanski97} Pojmanski, G.\ 1997, Acta Astronomica, 47, 467

1461: 

1462: \bibitem[Pont et al.(2006)]{Pont06} Pont, F., Zucker, S., \& Queloz, D.\ 2006, \mnras, 373, 231

1463: 

1464: \bibitem[Popper \& Etzel(1981)]{Popper81} Popper, D.~M.~\& Etzel, P.~B.\ 1981, \aj, 86, 102

1465: 

1466: \bibitem[Press et al.(1992)]{Press92} Press, W.~H., Teukolsky, S.~A., Vetterling, W.~T.,~\& Flannery, B.~P.\ 1992, Numerical Recipes in C. The Art of Scientific Computing (2nd ed.; Cambridge: University Press)

1467: 

1468: \bibitem[Ribas(2003)]{Ribas03} Ribas, I.\ 2003, \aap, 398, 239

1469: 

1470: \bibitem[Ribas(2006)]{Ribas06} Ribas, I.\ 2006, \apss, 304, 87

1471: 

1472: \bibitem[Salim \& Gould(2003)]{Salim03} Salim, S., \& Gould, A.\ 2003, \apj, 582, 1011

1473: 

1474: \bibitem[Samus(2006)]{Samus06} Samus, N.~N.\ 2006, Astronomical and Astrophysical Transactions, 25, 223

1475: 

1476: \bibitem[Schwarzenberg-Czerny(1989)]{SchwarzenbergCzerny89} Schwarzenberg-Czerny, A.\ 1989, \mnras, 241, 153

1477: 

1478: \bibitem[Schwarzenberg-Czerny(1996)]{SchwarzenbergCzerny96} Schwarzenberg-Czerny, A.\ 1996, \apjl, 460, L107

1479: 

1480: \bibitem[Skrutskie et al.(2006)]{Skrutskie06} Skrutskie, M.~F., et al.\ 2006, \aj, 131, 1163

1481: 

1482: \bibitem[Stebbing(1910)]{Stebbing10} Stebbing, J.\ 1910, \apj, 32, 185

1483: 

1484: \bibitem[Struve(1948)]{Struve48} Struve, O.\ 1948, \pasp, 60, 160

1485: 

1486: \bibitem[Tamuz et al.(2006)]{Tamuz06} Tamuz, O., Mazeh, T., \& North, P.\ 2006, \mnras, 367, 1521

1487: 

1488: \bibitem[Tassoul(1988)]{Tassoul88} Tassoul, J.-L.\ 1988, \apjl, 324, L71

1489: 

1490: \bibitem[Torres \& Ribas(2002)]{Torres02} Torres, G., \& Ribas, I.\ 2002, \apj, 567, 1140

1491: 

1492: \bibitem[Torres et al.(2006)]{Torres06} Torres, G., Lacy, C.~H., Marschall, L.~A., Sheets, H.~A., \& Mader, J.~A.\ 2006, \apj, 640, 1018

1493: 

1494: \bibitem[Tsesevich(1973)]{Tsesevich73} Tsesevich, V.~P.\ 1973, Eclipsing Variable Stars (New York: J.~Wiley)

1495: 

1496: \bibitem[Tyson(2002)]{Tyson02} Tyson, J.~A.\ 2002, \procspie, 4836, 10

1497: 

1498: \bibitem[Udalski et al.(1994)]{Udalski94} Udalski, A., et al.\ 1994, Acta Astronomica, 44, 165

1499: 

1500: \bibitem[Udalski(2003)]{Udalski03} Udalski, A.\ 2003, Acta Astronomica, 53, 291

1501: 

1502: \bibitem[Wilson \& Devinney(1971)]{Wilson71} Wilson, R.~E.~\& Devinney, E.~J.\ 1971, \apj, 166, 605

1503: 

1504: \bibitem[Yi et al.(2001)]{Yi01} Yi, S., Demarque, P., Kim, Y.-C., Lee, Y.-W., Ree, C.~H., Lejeune, T., \& Barnes, S.\ 2001, \apjs, 136, 417

1505: 

1506: \bibitem[Young et al.(2006)]{Young06} Young, T.~B., Hidas, M.~G., Webb, J.~K., Ashley, M.~C.~B., Christiansen, J.~L., Derekas, A., \& Nutto, C.\ 2006, \mnras, 370, 1529

1507: 

1508: \bibitem[Zacharias et al.(2004)]{Zacharias04} Zacharias, N., Urban, S.~E., Zacharias, M.~I., Wycoff, G.~L., Hall, D.~M., Monet, D.~G., \& Rafferty, T.~J.\ 2004, \aj, 127, 3043

1509: 

1510: \bibitem[Zahn(1975)]{Zahn75} Zahn, J.-P.\ 1975, \aap, 41, 329

1511: 

1512: \bibitem[Zahn(1977)]{Zahn77} Zahn, J.-P.\ 1977, \aap, 57, 383

1513: 

1514: \bibitem[Zahn(1978)]{Zahn78} Zahn, J.-P.\ 1978, \aap, 67, 162

1515: 

1516: \bibitem[Zahn \& Bouchet(1989)]{Zahn89} Zahn, J.-P., \& Bouchet, L.\ 1989, \aap, 223, 112

1517: 

1518: \bibitem[Zahn(1994)]{Zahn94} Zahn, J.-P.\ 1994, \aap, 288, 829

1519: 

1520: \end{thebibliography}{}

1521: 

1522: 

1523: \clearpage

1524: 

1525: \begin{deluxetable}{cccccccc}

1526: \tabletypesize{\tiny}

1527: \rotate

1528: \tablecaption{Observational parameters of the TrES fields}

1529: \tablewidth{0pt}

1530: \tablehead{\colhead{Field} &

1531:            \colhead{Constellation} &

1532:            \colhead{\begin{tabular}{c} $\alpha$\\ (J2000)\tablenotemark{a}\end{tabular}} &

1533:            \colhead{\begin{tabular}{c} $\delta$\\ (J2000)\end{tabular}} &

1534:            \colhead{\begin{tabular}{c} Galactic\\ coordinates (l,b)\end{tabular}} &

1535:            \colhead{\begin{tabular}{c} Starting\\ epoch (HJD)\end{tabular}} &

1536:            \colhead{\begin{tabular}{c} Ending\\ epoch (HJD)\end{tabular}} &

1537:            \colhead{\begin{tabular}{c} Duration\\ (days)\end{tabular}}}

1538: \startdata

1539: And0& Andromeda      & 01 09 30.1255& +47 14 30.453& (126.11, -015.52)& 2452878.9& 2452934.9& 56.0\\

1540: Cas0& Cassiopeia     & 00 39 09.8941& +49 21 16.519& (120.88, -013.47)& 2453250.8& 2453304.6& 53.8\\

1541: CrB0& Corona Borealis& 16 01 02.6616& +33 18 12.634& (053.49, +048.92)& 2453493.8& 2453536.8& 43.0\\

1542: Cyg1& Cygnus         & 20 01 21.5633& +50 06 16.902& (084.49, +010.28)& 2453170.7& 2453250.0& 79.3\\

1543: Dra0& Draco          & 16 45 17.8177& +56 46 54.686& (085.68, +039.53)& 2453093.8& 2453163.0& 69.2\\

1544: Her0& Hercules       & 16 49 14.2185& +45 58 59.963& (071.61, +039.96)& 2452769.9& 2452822.0& 52.1\\

1545: Lyr1& Lyra           & 19 01 26.3713& +46 56 05.325& (077.15, +017.86)& 2453541.8& 2453616.7& 74.9\\

1546: Per1& Perseus        & 03 41 07.8581& +37 34 48.712& (156.37, -014.04)& 2453312.8& 2453402.8& 90.0\\

1547: Tau0& Taurus         & 04 20 21.2157& +27 21 02.713& (169.83, -015.94)& 2453702.7& 2453770.9& 68.2\\

1548: UMa0& Ursa Major     & 09 52 06.3560& +54 03 51.596& (160.87, +047.70)& 2453402.9& 2453487.8& 84.9\\

1549: \enddata

1550: \tablenotetext{a}{ICRS 2000.0 coordinates of the guide star, which is located at the center of the field of view.}

1551: \label{tableFieldsObs}

1552: \end{deluxetable}

1553: 

1554: 

1555: 

1556: \begin{deluxetable}{ccccccc}

1557: \tabletypesize{\tiny}

1558: \rotate

1559: \tablecaption{The number of sources and yield of the TrES fields}

1560: \tablewidth{0pt}

1561: \tablehead{\colhead{Field} &

1562:            \colhead{\begin{tabular}{c} Number\\ of LCs\end{tabular}} &

1563:            \colhead{\begin{tabular}{c} Number of observations\\ in each LC\end{tabular}} &

1564:            \colhead{\begin{tabular}{c} Fraction\\ RMS $<$ 1\%\end{tabular}} &

1565:            \colhead{\begin{tabular}{c} Fraction\\ RMS $<$ 2\%\end{tabular}} &

1566:            \colhead{\begin{tabular}{c} Found\\ EBs\end{tabular}} &

1567:            \colhead{\begin{tabular}{c} EB discovery\\ yield\end{tabular}}}

1568: \startdata

1569: And0& 26495& 2357& 16.5\%& 40.4\%& 111& 0.42\%\\

1570: Cas0& 22615& 2069& 11.0\%& 38.2\%& 119& 0.53\%\\

1571: CrB0& 18954& 1287& 11.0\%& 22.4\%&  28& 0.15\%\\

1572: Cyg1& 17439& 3256& 30.3\%& 65.7\%& 125& 0.72\%\\

1573: Dra0& 15227& 2000& 11.8\%& 26.4\%&  42& 0.28\%\\

1574: Her0& 15916&  974& 16.8\%& 35.0\%&  28& 0.18\%\\

1575: Lyr1& 22964& 2815& 19.4\%& 49.0\%& 135& 0.59\%\\

1576: Per1& 20988& 1647& 15.9\%& 38.4\%&  93& 0.44\%\\

1577: Tau0& 14442& 1171& 13.1\%& 32.5\%&  68& 0.47\%\\

1578: UMa0& 10405& 1343& 13.6\%& 29.5\%&  24& 0.23\%\\

1579: \enddata

1580: \label{tableFieldsYield}

1581: \end{deluxetable}

1582: 

1583: 

1584: \begin{deluxetable}{ccccccccccc}

1585: \tabletypesize{\tiny}

1586: \rotate

1587: \tablecaption{Eccentric EBs}

1588: \tablewidth{0pt}

1589: \tablehead{\colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$\tablenotemark{a}} & \colhead{$|e \cos\omega|_{timing}$\tablenotemark{b}} & \colhead{$|e \cos\omega|_{adopted}$\tablenotemark{c}} & \colhead{$e$\tablenotemark{d}} & \colhead{$M_1/M_\sun$} & \colhead{$M_2/M_\sun$}& \colhead{age $[Gyr]$} & \colhead{$t_{circ}$ $[Gyr]$}}

1590: \startdata

1591: T-And0-04144& 01 17 35.247& 49 46 16.97&  7.869& 0.0072& 0.0068& $0.14^{+0.08}_{-0.08}$  & 0.84 (-1)\tablenotemark{e}      & 0.54 (-1)      & 10.0 (-3)     & 140\\

1592: T-And0-17158& 01 10 09.143& 48 18 19.68& 11.415& 0.0182& 0.0180& $0.038^{+0.12}_{-0.02}$ & 1.03 (-1)      & 0.92 (-1)      & 10.0 (-3)     & 370\\

1593: T-And0-24609& 00 58 29.826& 49 25 08.88& 17.997& 0.0794& 0.0799& $0.10^{+0.10}_{-0.02}$  & 1.22 $\pm$ 0.10& 1.10 $\pm$ 0.30& 5.4 $\pm$ 12.0& 6400\\

1594: T-Cas0-00394& 00 32 51.608& 49 19 39.36&  1.746& 0.0235& 0.0242& $0.024^{+0.03}_{-0.001}$& 1.46 $\pm$ 0.01& 1.44 $\pm$ 0.01& 3.4 $\pm$ 0.3 & 260\\

1595: T-Cas0-02603& 00 47 08.610& 50 37 19.32&  2.217& 0.2098& 0.2143& $0.25^{+0.14}_{-0.04}$  & 1.25 $\pm$ 0.01& 0.75 $\pm$ 0.04& 5.4 $\pm$ 5.0 & 0.26\\

1596: T-Cas0-04534& 00 31 04.585& 51 52 10.88&  6.909& 0.0057& 0.0048& $0.014^{+0.03}_{-0.01}$ & 1.17 $\pm$ 0.04& 0.96 $\pm$ 0.15& 6.4 $\pm$ 6.9 & 29\\

1597: T-Cas0-04947& 00 47 10.336& 50 45 12.36&  3.285& 0.0845& 0.0845& $0.10^{+0.04}_{-0.02}$  & 1.04 (-1)      & 0.86 (-1)      & 10.0 (-3)     & 0.53\\

1598: T-Cas0-05165& 00 43 59.256& 51 14 00.07&  2.359& 0.0311& 0.0327& $0.15^{+0.08}_{-0.08}$  & 1.50 $\pm$ 0.21& 0.76 $\pm$ 0.17& 2.7 $\pm$ 2.9 & 0.34\\

1599: T-Cas0-07630& 00 37 23.347& 47 19 20.68&  5.869& 0.0200& 0.0298& $0.038^{+0.15}_{-0.008}$& 1.15 $\pm$ 0.12& 0.87 $\pm$ 0.34& 5.9 $\pm$ 9.7 & 13\\

1600: T-Cyg1-01364& 20 09 38.211& 49 05 08.02& 12.233&   N/A & 0.3254& $0.53^{+0.04}_{-0.04}$  & 1.03 $\pm$ 0.18& 0.50 $\pm$ 0.09& 0.4 $\pm$ 1.2 & 1100\\

1601: T-Cyg1-01373& 19 55 44.105& 52 13 34.61&  4.436& 0.0059& 0.0054& $0.010^{+0.02}_{-0.005}$& 0.97 (-1)      & 0.82 (-1)      & 10.0 (-3)     & 3.0\\

1602: T-Cyg1-01994& 20 03 03.111& 52 42 04.17& 14.482&   N/A & 0.0107& $0.15^{+0.15}_{-0.14}$  & 1.80 (-1)      & 1.06 (-1)      & 0.20 (-2)     & 2300\\

1603: T-Cyg1-02304& 20 02 04.388& 47 34 14.75&  5.596& 0.1549& 0.1529& $0.23^{+0.10}_{-0.08}$  & 2.20 $\pm$ 1.28& 0.72 $\pm$ 0.41& 0.7 $\pm$ 4.8 & 46\\

1604: T-Cyg1-02624& 19 59 25.926& 52 23 59.91& 11.608& 0.0172& 0.0172& $0.068^{+0.03}_{-0.03}$ & 2.11 $\pm$ 0.05& 1.52 $\pm$ 0.03& 0.3 $\pm$ 0.1 & $10^7$\\

1605: T-Cyg1-06677& 20 07 25.526& 52 22 00.54&  6.512& 0.0077& 0.0069& $0.062^{+0.03}_{-0.03}$ & 1.54 $\pm$ 0.20& 1.31 $\pm$ 0.22& 1.6 $\pm$ 1.9 & $10^6$\\

1606: T-Cyg1-07248& 19 54 45.937& 50 24 05.32&  6.058& 0.1674& 0.1681& $0.17^{+0.07}_{-0.001}$ & 1.68 $\pm$ 0.01& 0.87 $\pm$ 0.20& 2.0 $\pm$ 2.1 & 33\\

1607: T-Cyg1-07297& 20 10 46.910& 49 09 29.42& 11.613& 0.3019& 0.3010& $0.38^{+0.08}_{-0.08}$  & 0.97 (-1)      & 0.55 (-1)      & 10.0 (-3)     & 830\\

1608: T-Cyg1-07584& 19 58 58.012& 47 38 19.26&  4.925& 0.0074& 0.0074& $0.022^{+0.08}_{-0.01}$ & 0.94 (-1)      & 0.90 (-1)      & 10.0 (-3)     & 4.7\\

1609: T-Cyg1-09934& 20 10 44.209& 51 07 51.77&  4.549& 0.0505& 0.0501& $0.11^{+0.06}_{-0.06}$  & 1.35 $\pm$ 0.64& 0.94 $\pm$ 0.41& 3.5 $\pm$ 5.6 & 5.6\\

1610: T-Cyg1-15752& 20 13 52.454& 50 52 23.12&  9.372& 0.2402& 0.2402& $0.35^{+0.05}_{-0.05}$  & 1.31 $\pm$ 0.04& 1.05 $\pm$ 0.11& 3.6 $\pm$ 4.9 & 230\\

1611: T-Lyr1-09931& 18 59 08.441& 48 36 00.04& 11.632& 0.2207& 0.2209& $0.25^{+0.04}_{-0.03}$  & 0.91 $\pm$ 0.09& 0.67 $\pm$ 0.08& 2.7 $\pm$ 3.3 & 730\\

1612: T-Lyr1-13841& 19 06 26.558& 48 28 47.04&  6.640& 0.0362& 0.0362& $0.075^{+0.11}_{-0.04}$ & 1.01 $\pm$ 0.27& 1.01 $\pm$ 0.24& 8.7 $\pm$ 13.1 & 19\\

1613: T-Lyr1-14413& 19 03 41.143& 47 36 55.78& 39.861& 0.5922& 0.6240& $0.64^{+0.006}_{-0.006}$& 1.08 $\pm$ 0.34& 0.96 $\pm$ 0.26& 6.4 $\pm$ 18.9 & $10^5$\\

1614: T-Lyr1-14508& 18 57 40.271& 48 40 51.28&  8.050& 0.1861& 0.1862& $0.31^{+0.16}_{-0.12}$  & 1.34 $\pm$ 0.28& 1.20 $\pm$ 0.78& 2.9 $\pm$ 8.4 & 220\\

1615: T-Lyr1-22359& 19 10 54.290& 49 26 06.95& 12.319& 0.1990& 0.1984& $0.33^{+0.05}_{-0.05}$  & 0.97 $\pm$ 0.48& 0.97 $\pm$ 0.46& 6.9 $\pm$ 29.3 & 550\\

1616: T-Per1-00769& 03 31 43.915& 36 31 52.36&  3.648& 0.0248& 0.0263& $0.055^{+0.05}_{-0.03}$ & 1.06 $\pm$ 0.01& 0.65 $\pm$ 0.03& 7.6 $\pm$ 2.1 & 1.4\\

1617: T-Per1-04218& 03 35 33.667& 40 00 49.18&  4.070& 0.0072& 0.0079& $0.10^{+0.19}_{-0.09}$  & 0.94 (-1)      & 0.72 (-1)      & 10.0 (-3)     & 2.4\\

1618: T-Per1-05205& 03 34 19.432& 39 32 44.41&  8.472& 0.0558& 0.0592& $0.095^{+0.11}_{-0.04}$ & 2.22 $\pm$ 0.01& 1.17 $\pm$ 0.28& 0.9 $\pm$ 1.7 & 210\\

1619: T-Per1-08252& 03 52 00.670& 40 03 47.73&  4.457& 0.0656& 0.0645& $0.065^{+0.06}_{-0.001}$& 1.56 $\pm$ 0.01& 1.40 $\pm$ 0.34& 2.4 $\pm$ 2.5 & $10^5$\\

1620: T-Per1-11424& 03 47 56.473& 37 31 31.83&  4.247& 0.2403& 0.2404& $0.24^{+0.02}_{-0.006}$ & 1.01 (-1)      & 0.82 (-1)      & 10.0 (-3)     & 2.3\\

1621: T-Per1-17327& 03 40 45.644& 34 47 57.26&  3.946& 0.0332& 0.0305& $0.069^{+0.25}_{-0.04}$ & 1.10 $\pm$ 0.02& 1.09 $\pm$ 0.09& 8.4 $\pm$ 16.4& 1.2\\

1622: T-Tau0-02487& 04 21 55.933& 25 35 49.28&  2.826& 0.0125& 0.0054& $0.014^{+0.005}_{-0.005}$&1.74 $\pm$ 0.07& 1.01 $\pm$ 0.08& 1.6 $\pm$ 0.7 & 0.39\\

1623: T-Tau0-03916& 04 23 37.351& 25 46 36.00&  3.217& 0.0713& 0.0706& $0.071^{+0.02}_{-0.004}$& 1.18 $\pm$ 0.01& 1.15 $\pm$ 0.03& 6.0 $\pm$ 4.4 & 0.56\\

1624: T-UMa0-01822& 09 53 37.710& 52 45 44.72&  9.551& 0.1502& 0.1503& $0.31^{+0.02}_{-0.02}$  & 1.01 $\pm$ 0.04& 1.00 $\pm$ 0.05& 8.3 $\pm$ 4.8 & 130\\

1625: \enddata

1626: \tablenotetext{a}{The full precision of the measured period is

1627: listed in the electronic version of the catalog, together with its

1628: uncertainty and the epoch of the center of eclipse (see appendix

1629: \ref{appendixCatalogDescription}).}

1630: \tablenotetext{b}{Measurements

1631: made using the eclipse timing of step (4). Although these values

1632: are approximations, they do not suffer from nearly as much

1633: numerical error as the DEBiL measurement, and are therefore

1634: usually accurate. ``N/A'' marks LCs for which there were too few

1635: eclipses to be able to apply the timing method.}

1636: \tablenotetext{c}{The adopted value is a combination of the values

1637: measured with the timing method and with DEBiL.}

1638: \tablenotetext{d}{The uncertainties of the eccentricities are

1639: non-Gaussian, since they have a strict lower bound ($e \geq |e

1640: \cos\omega|$). We truncated the quoted lower uncertainties at this

1641: value, though even at this truncated value the real uncertainty is

1642: beyond $1\sigma$.}

1643: \tablenotetext{e}{When the most likely model is

1644: at the edge of the parameter space, MECI is not able to bound the

1645: solution, and therefore cannot estimate the uncertainties. We mark

1646: (-3) when the upper limit was reached, (-2) when the lower limit

1647: was reached, and (-1) if one of the other parameters is at its

1648: limit.} \label{tableEccentric}

1649: \end{deluxetable}

1650: 

1651: 

1652: 

1653: \begin{deluxetable}{cccc}

1654: \tabletypesize{\tiny}

1655: \rotate

1656: \tablecaption{EBs that fill at least one of their Roche-lobes (first 20)}

1657: \tablewidth{0pt}

1658: \tablehead{\colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$}}

1659: \startdata

1660: T-And0-03774& 00 59 01.029& 46 47 17.08&  1.362\\

1661: T-And0-04813& 01 16 37.880& 47 33 23.43&  0.552\\

1662: T-And0-05140& 01 03 22.258& 44 56 24.31&  0.981\\

1663: T-And0-05153& 01 18 48.278& 49 39 36.86&  0.492\\

1664: T-And0-05343& 00 52 55.122& 48 01 37.68&  0.824\\

1665: T-And0-07638& 01 09 27.871& 49 20 33.81&  0.403\\

1666: T-And0-07892& 00 56 15.567& 48 39 10.73&  0.380\\

1667: T-And0-08330& 01 19 15.949& 48 00 17.45&  0.630\\

1668: T-And0-08652& 00 56 58.855& 49 05 05.00&  0.335\\

1669: T-And0-09528& 01 22 09.328& 47 14 29.86&  0.918\\

1670: T-And0-10071& 01 14 50.412& 49 17 46.28&  0.387\\

1671: T-And0-10206& 00 55 55.724& 49 49 46.56&  0.859\\

1672: T-And0-10511& 01 19 16.430& 47 07 46.27&  0.563\\

1673: T-And0-10722& 01 04 03.859& 48 37 13.04&  1.062\\

1674: T-And0-11354& 01 18 05.168& 46 10 14.66&  0.331\\

1675: T-And0-11476& 01 07 32.106& 45 55 44.93&  6.380\\

1676: T-And0-11599& 01 09 28.113& 46 18 24.85&  0.280\\

1677: T-And0-11617& 01 07 28.020& 45 22 40.35&  0.503\\

1678: T-And0-12453& 01 17 12.316& 46 42 35.43&  0.448\\

1679: T-And0-12769& 00 52 58.164& 44 44 11.26&  0.325\\

1680: \enddata

1681: \label{tableFillRoche}

1682: \end{deluxetable}

1683: 

1684: 

1685: \begin{deluxetable}{ccccccl}

1686: \tabletypesize{\tiny}

1687: \rotate

1688: \tablecaption{Abnormal EBs}

1689: \tablewidth{0pt}

1690: \tablehead{

1691: \colhead{Object} &

1692: \colhead{$\alpha$ (J2000)} &

1693: \colhead{$\delta$ (J2000)} &

1694: \colhead{\begin{tabular}{c} Period\\ (days) \end{tabular}} &

1695: \colhead{\begin{tabular}{c} Classified\\ in catalog?\end{tabular}} &

1696: \colhead{\begin{tabular}{c} In SIMBAD/VSX?\\ (see table \ref{tableSIMBAD}) \end{tabular}} &

1697: \colhead{Notes}}

1698: \startdata

1699: T-And0-00920& 01 17 30.677& 47 03 31.61&24.073& no & no & Large asymmetric reflection ($0.1$ mag) \ offset eclipse\\

1700: T-And0-04594& 01 16 10.713& 48 52 18.97& 3.910& yes& no & Spots / active\\

1701: T-And0-11476& 01 07 32.106& 45 55 44.93& 6.380& yes& no & Tilted plateaus (spots?)\\

1702: T-Cas0-13944& 00 29 48.990& 50 49 54.06& 1.739& yes& no & Irregular eclipse depths\\

1703: T-Cyg1-07584& 19 58 58.012& 47 38 19.26& 4.925& yes& no & Large persistent spot\\

1704: T-Cyg1-08866& 20 08 36.448& 49 29 35.79& 2.876& yes& no & Offset eclipse\tablenotemark{a}\\

1705: T-Dra0-00398& 16 57 33.875& 59 31 51.98& 1.046& yes& yes& Active (has $0.2$ mag fluctuations with periods of a few hours)\\

1706: T-Dra0-03105& 16 23 02.558& 59 27 23.44& 0.485& no & yes& Unequal eclipses\tablenotemark{b} / semi-detached\\

1707: T-Dra0-04520& 16 49 57.960& 56 26 45.56& 3.113& yes& no & Tilted plateaus (spots?)\\

1708: T-Her0-03497& 16 52 28.391& 44 51 29.63& 7.853& yes& no & Unequal plateaus\tablenotemark{c}\\

1709: T-Her0-08091& 16 51 52.608& 47 01 47.98& 2.694& yes& no & Offset eclipse\\

1710: T-Lyr1-00359& 19 15 33.695& 44 37 01.30& 1.062& yes& yes& Large recurring spots ($\sim0.05$ mag)\\

1711: T-Lyr1-02800& 19 08 18.809& 47 12 48.16& 4.876& no & no & Semi-detached / unequal plateaus (spots?)\\

1712: T-Lyr1-05984& 18 53 50.481& 45 33 20.90& 1.470& no & no & Unequal eclipses\tablenotemark{b} / semi-detached\\

1713: T-Lyr1-08305& 18 56 43.798& 48 07 02.86&14.081& yes& no & Large asymmetric reflection ($0.05$ mag) ; offset eclipse\\

1714: T-Lyr1-13166& 19 02 28.120& 46 58 57.75& 0.310& no & no & Unequal plateaus ; misshapen eclipse (persistent spot?)\\

1715: T-Lyr1-15595& 19 06 05.267& 49 04 08.95& 9.477& yes& no & Offset eclipse\\

1716: T-Per1-00750& 03 47 45.543& 35 00 37.08& 1.929& yes& yes& Spots / active\\

1717: T-Per1-08789& 03 54 33.282& 39 07 41.53& 2.645& yes& no & Tilted plateaus\\

1718: T-UMa0-03090& 10 08 52.180& 52 45 52.49& 0.538& yes& yes& Unequal plateaus\\

1719: \enddata

1720: \tablenotetext{a}{Even when the LC plateaus are

1721: not flat, due to tidal distortion or reflections, the system's

1722: mirror symmetry normally guarantees that the eclipses will occur

1723: during a plateau minimum or maximum. When, as in these cases, the

1724: eclipses are significantly offset from the plateau minima/maxima

1725: we can conclude that some mechanism, perhaps severe tidal lag, is

1726: breaking the system's symmetry.}

1727: \tablenotetext{b}{Might not be an EB. This LC could be due to

1728: non-sinusoidal pulsations.}

1729: \tablenotetext{c}{The two LC plateaus

1730: between the eclipses, have a significantly different mean

1731: magnitude. This may be due to one or both components being tidally

1732: locked, and having a persistent spot or surface temperature

1733: variation at specific longitudes.}

1734: \label{tableAbnormal}

1735: \end{deluxetable}

1736: 

1737: 

1738: 

1739: \begin{deluxetable}{cccccccc}

1740: \tabletypesize{\tiny}

1741: \rotate

1742: \tablecaption{Ambiguous EBs (first 10)}

1743: \tablewidth{0pt}

1744: \tablehead{\colhead{Ver.} & \colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$} & \colhead{$M_1/M_\sun$} & \colhead{$M_2/M_\sun$}& \colhead{age $[Gyr]$}}

1745: \startdata

1746: A& T-And0-00657& 01 06 06.159& 47 31 59.37&  6.725&  2.50       (-1)& 0.74       (-1)& 0.20       (-2)\\

1747: B& T-And0-00657& 01 06 06.159& 47 31 59.37& 13.456&  1.92       (-1)\tablenotemark{c}& 1.92       (-1)& 0.20       (-2)\\

1748: A& T-And0-01203& 01 03 34.745& 48 32 39.27&  3.505&  1.86 $\pm$ 0.09& 0.56 $\pm$ 0.10& 0.89 $\pm$ 0.83\\

1749: B& T-And0-01203& 01 03 34.745& 48 32 39.27&  7.011&  1.90 $\pm$ 0.12& 0.66 $\pm$ 0.19& 0.80 $\pm$ 1.13\\

1750: A& T-And0-06017& 01 12 48.217& 49 58 07.16&  2.543&  1.40 $\pm$ 0.35& 0.52 $\pm$ 0.77& 3.49 $\pm$ 4.28\\

1751: B& T-And0-06017& 01 12 48.217& 49 58 07.16&  5.085&  1.18 $\pm$ 0.71& 1.12 $\pm$ 0.85& 3.12 $\pm$ 11.15\\

1752: A& T-And0-06500& 01 25 56.083& 49 23 31.74&  5.337&  0.97 $\pm$ 0.20& 0.49 $\pm$ 0.53& 7.71 $\pm$ 16.33\\

1753: B& T-And0-06500& 01 25 56.083& 49 23 31.74& 10.674&  1.01 $\pm$ 0.30& 0.93 $\pm$ 0.45& 0.74 $\pm$ 1.60\\

1754: A& T-And0-06680& 00 55 48.153& 45 02 48.57&  4.551&  1.16 $\pm$ 0.04& 0.51 $\pm$ 0.20& 6.09 $\pm$ 8.91\\

1755: B& T-And0-06680& 00 55 48.153& 45 02 48.57&  9.104&  1.16 $\pm$ 0.09& 0.96 $\pm$ 0.29& 6.24 $\pm$ 10.78\\

1756: A& T-And0-08053& 01 13 59.402& 45 51 43.43&  4.116&  1.14       (-1)& 0.40       (-2)& 6.00       (-1)\\

1757: B& T-And0-08053& 01 13 59.402& 45 51 43.43&  8.231&  1.09 $\pm$ 0.55& 1.05 $\pm$ 0.64& 3.22 $\pm$ 16.37\\

1758: A& T-And0-08417& 01 01 39.041& 45 03 32.98&  2.053&  1.01       (-1)& 0.47       (-1)& 10.00       (-3)\\

1759: B& T-And0-08417& 01 01 39.041& 45 03 32.98&  4.106&  1.01       (-1)& 0.90       (-1)& 10.00       (-3)\\

1760: A& T-And0-09365& 01 01 00.459& 45 14 24.77&  1.887&  1.05 $\pm$ 0.03& 0.43 $\pm$ 0.39& 8.74 $\pm$ 16.06\\

1761: B& T-And0-09365& 01 01 00.459& 45 14 24.77&  3.774&  1.05 $\pm$ 0.05& 0.93 $\pm$ 0.52& 9.47 $\pm$ 23.55\\

1762: A& T-And0-10518& 01 07 44.417& 48 44 58.11&  0.194&  0.90       (-1)& 0.40       (-2)& 0.40       (-1)\\

1763: B& T-And0-10518& 01 07 44.417& 48 44 58.11&  0.387&  0.45 $\pm$ 0.27& 0.45 $\pm$ 0.28& 0.27 $\pm$ 0.54\\

1764: A& T-And0-11453& 01 05 42.744& 44 54 02.26&  0.784&  1.12       (-1)& 0.40       (-2)& 7.00       (-1)\\

1765: B& T-And0-11453& 01 05 42.744& 44 54 02.26&  1.568&  1.02 $\pm$ 0.43& 1.01 $\pm$ 0.32& 8.81 $\pm$ 14.54\\

1766: \enddata

1767: \tablenotetext{A}{Unequal eclipse model, assuming an unseen secondary eclipse.}

1768: \tablenotetext{B}{Equal eclipse model, with double the period of the unequal model.}

1769: \tablenotetext{c}{When the most likely model is at the edge of the parameter space, MECI is not

1770: able to bound the solution, and therefore cannot estimate the uncertainties. We mark (-3) when

1771: the upper limit was reached, (-2) when the lower limit was reached, and (-1) if one of the other

1772: parameter is at its limit.}

1773: \label{tableAmbig}

1774: \end{deluxetable}

1775: 

1776: 

1777: 

1778: \begin{deluxetable}{cccccccccc}

1779: \tabletypesize{\tiny}

1780: \rotate

1781: \tablecaption{Circular EBs (first 20)}

1782: \tablewidth{0pt}

1783: \tablehead{\colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$} & \colhead{$M_1/M_\sun$} & \colhead{$M_2/M_\sun$}& \colhead{age $[Gyr]$} &

1784: \colhead{\begin{tabular}{c} Proper motion \\ source catalog \end{tabular}} & \colhead{\begin{tabular}{c} $PM_{\alpha}$ \\ $[MAS/year]$ \end{tabular}} & \colhead{\begin{tabular}{c} $PM_{\delta}$\\ $[MAS/year]$\end{tabular}}}

1785: \startdata

1786: T-And0-00194& 01 20 12.816& 48 36 41.36&  2.145&  2.07 $\pm$ 0.02& 0.97 $\pm$ 0.02& 0.59 $\pm$ 0.12& UCAC& 28.4& -12.2\\

1787: T-And0-00459& 01 11 24.845& 46 57 49.44&  3.655&  1.20 $\pm$ 0.01& 1.19 $\pm$ 0.01& 5.35 $\pm$ 1.13& UCAC& -1.6& -20.6\\

1788: T-And0-00745& 01 03 45.076& 44 50 41.14&  2.851&  1.86 $\pm$ 0.23& 1.02 $\pm$ 0.22& 1.04 $\pm$ 0.72& UCAC& -6.6& -4.8\\

1789: T-And0-01461& 01 06 15.353& 45 08 25.66&  5.613&  1.47 $\pm$ 0.01& 1.45 $\pm$ 0.08& 2.76 $\pm$ 2.72& UCAC& -11.4& 2.8\\

1790: T-And0-01554& 01 17 04.999& 45 54 06.20&  1.316&  0.90    (-1)\tablenotemark{a}   & 0.84     (-1)  & 10.00    (-3)  & UCAC& -44.6& -40.8\\

1791: T-And0-01597& 01 10 32.071& 46 49 53.18&  3.503&  1.55 $\pm$ 0.03& 1.54 $\pm$ 0.01& 2.37 $\pm$ 0.76& UCAC& 2.9& -5.5\\

1792: T-And0-02462& 01 18 00.594& 49 27 12.47&  3.069&  1.97 $\pm$ 0.69& 1.10 $\pm$ 1.31& 1.02 $\pm$ 1.58& UCAC& 5.8& -1.1\\

1793: T-And0-02699& 01 06 44.813& 47 31 08.61&  1.759&  1.18 $\pm$ 0.02& 0.53 $\pm$ 0.07& 5.21 $\pm$ 3.37& UCAC& 0.2& -6.8\\

1794: T-And0-02798& 01 21 18.345& 48 48 05.63&  2.860&  1.04 $\pm$ 0.10& 0.65 $\pm$ 0.13& 6.14 $\pm$ 9.51& UCAC& 6.3& -8.1\\

1795: T-And0-03526& 01 20 17.451& 47 39 23.32&  1.536&  1.04 $\pm$ 0.02& 0.84 $\pm$ 0.02& 6.29 $\pm$ 2.37& UCAC& 17.9& -11.1\\

1796: T-And0-04046& 00 55 20.157& 47 44 53.20&  3.916&  1.30 $\pm$ 0.09& 1.25 $\pm$ 0.12& 3.10 $\pm$ 4.31& UCAC& -3.8& -7.3\\

1797: T-And0-04594& 01 16 10.713& 48 52 18.97&  3.910&  1.05    (-1)   & 0.82    (-1)   & 10.00    (-3)  & UCAC& 1.5& -1.9\\

1798: T-And0-04829& 01 15 15.228& 47 45 58.97&  0.678&  0.99    (-1)   & 0.92    (-1)   & 10.00    (-3)  & UCAC& -23.8& 44.4\\

1799: T-And0-05241& 00 56 34.679& 46 37 02.91&  1.454&  1.56 $\pm$ 0.01& 1.47 $\pm$ 0.31& 2.69 $\pm$ 7.01& UCAC& -4.5& -0.5\\

1800: T-And0-05375& 01 10 58.225& 49 52 48.69&  1.640&  2.13    (-1)   & 1.85    (-2)   & 1.00    (-1)   & UCAC& -6.3& 0.1\\

1801: T-And0-05794& 01 12 11.763& 47 32 30.94&  1.053&  2.06 $\pm$ 0.19& 1.08 $\pm$ 0.54& 1.08 $\pm$ 1.92& UCAC& -0.4& -1.4\\

1802: T-And0-06039& 01 23 37.548& 48 25 37.73&  4.923&  1.22 $\pm$ 0.05& 1.08 $\pm$ 0.31& 5.33 $\pm$ 7.17& UCAC& -2.5& -5.0\\

1803: T-And0-06340& 01 01 55.269& 49 18 38.23&  5.437&  1.33    (-1)   & 0.40    (-2)   & 4.00    (-1)   & UCAC& 0.3& -2.9\\

1804: T-And0-06538& 01 20 58.907& 49 29 08.89& 18.669&  1.33 $\pm$ 0.15& 0.97 $\pm$ 0.17& 3.38 $\pm$ 3.45& UCAC& 1.1& -6.8\\

1805: T-And0-06632& 01 22 36.840& 47 52 53.29&  1.669&  1.69 $\pm$ 0.01& 1.45 $\pm$ 0.24& 2.21 $\pm$ 1.02& UCAC& -7.2& -7.6\\

1806: \enddata

1807: \label{tableCircular}

1808: \tablenotetext{a}{When the most likely model is at the edge of the parameter space, MECI is not

1809: able to bound the solution, and therefore cannot estimate the uncertainties. We mark (-3) when

1810: the upper limit was reached, (-2) when the lower limit was reached, and (-1) if one of the other

1811: parameter is at its limit.}

1812: \end{deluxetable}

1813: 

1814: 

1815: 

1816: \begin{deluxetable}{cccc}

1817: \tabletypesize{\tiny}

1818: \rotate

1819: \tablecaption{Inverted EBs}

1820: \tablewidth{0pt}

1821: \tablehead{\colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$}}

1822: \startdata

1823: T-And0-13653& 00 59 57.881& 45 03 41.53&  3.342\\

1824: T-Cas0-02069& 00 49 17.959& 50 39 02.92&  2.830\\

1825: T-Cas0-03012& 00 45 41.832& 51 01 35.40&  1.108\\

1826: T-Cas0-04618& 00 46 22.661& 50 39 17.57&  2.798\\

1827: T-Cas0-07780& 00 34 18.779& 52 00 35.72&  1.852\\

1828: T-Cas0-19045& 00 21 44.707& 50 32 29.55&  0.785\\

1829: T-Cas0-19668& 00 48 01.342& 47 06 11.58&  1.848\\

1830: T-Cas0-21651& 00 26 34.895& 46 38 42.69&  1.155\\

1831: T-Cyg1-01956& 19 53 29.106& 47 48 49.86&  2.045\\

1832: T-Cyg1-02929& 20 11 57.009& 48 07 03.59&  4.263\\

1833: T-Cyg1-17342& 19 49 54.197& 50 53 28.08&  2.220\\

1834: T-Her0-05469& 16 54 51.245& 43 20 35.89&  0.899\\

1835: T-Lyr1-04431& 19 12 16.047& 49 42 23.58&  0.903\\

1836: T-Lyr1-05887& 18 52 10.489& 47 48 16.67&  1.802\\

1837: T-Lyr1-07179& 18 49 14.039& 45 24 38.61&  1.323\\

1838: T-Lyr1-10989& 19 06 22.791& 45 41 53.82&  2.015\\

1839: T-Lyr1-11067& 18 52 53.489& 47 51 26.58&  2.241\\

1840: T-Per1-04353& 03 45 04.887& 37 47 15.91&  2.953\\

1841: T-Per1-06993& 03 40 59.668& 39 12 35.90&  2.125\\

1842: T-Per1-09366& 03 49 20.305& 39 55 41.97&  2.374\\

1843: T-Per1-12217& 03 28 59.454& 37 37 42.14&  1.690\\

1844: T-Tau0-00686& 04 07 13.870& 29 18 32.44&  5.361\\

1845: T-UMa0-00127& 09 38 06.716& 56 01 07.32&  0.687\\

1846: \enddata

1847: \label{tableInverted}

1848: \end{deluxetable}

1849: 

1850: 

1851: 

1852: \begin{deluxetable}{cccccll}

1853: \tabletypesize{\tiny}

1854: \rotate

1855: \tablecaption{EBs that appear in either the VSX or the SIMBAD astronomical databases}

1856: \tablewidth{0pt}

1857: \tablehead{\colhead{Category} & \colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Spectral type} & \colhead{Classification} & \colhead{Identifiers}}

1858: \startdata

1859: Circular  & T-And0-00194& 01 20 12.816& 48 36 41.36& A5  & Star                  & BD+47 378 ; GSC 03269-00662 ; SAO 37126 ; AG+48 143\\

1860:           &             &             &            &     &                       & PPM 43886 ; TYC 3269-662-1\\

1861: Circular  & T-And0-00459& 01 11 24.845& 46 57 49.44& F8  & EB of Algol type      & CO And ; GSC 03268-00398 ; TYC 3268-398-1 ; BD+46 281 ; BV 74\\

1862: Ambiguous & T-And0-00657& 01 06 06.159& 47 31 59.37& K0  & Star                  & BD+46 254 ; GSC 03267-01349 ; TYC 3267-1349-1\\

1863:           &             &             &            &     &                       & AG+47 120 ; PPM 43637\\

1864: Circular  & T-And0-00745& 01 03 45.076& 44 50 41.14&     & Star                  & TYC 2811-470-1 ; GSC 02811-00470\\

1865: Ambiguous & T-And0-01203& 01 03 34.745& 48 32 39.27&     & Star                  & TYC 3267-1176-1 ; GSC 03267-01176\\

1866: Circular  & T-And0-04046& 00 55 20.157& 47 44 53.20&     & Star                  & GPM 13.833991+47.748193\\

1867: Roche-fill& T-And0-05153& 01 18 48.278& 49 39 36.86&     & EB of W UMa type      & QW And\\

1868: Roche-fill& T-And0-05343& 00 52 55.122& 48 01 37.68&     & Star                  & GPM 13.232700+48.019757\\

1869: Roche-fill& T-And0-07892& 00 56 15.567& 48 39 10.73&     & EB                    & NSVS 3757820\\

1870: Circular  & T-And0-23792& 00 54 09.254& 47 45 19.91&     & Star                  & GPM 13.538629+47.755510\\

1871: Roche-fill& T-Cas0-00170& 00 53 37.847& 48 43 33.83&     & Star                  & TYC 3266-195-1 ; GSC 03266-00195\\

1872: Eccentric & T-Cas0-00394& 00 32 51.608& 49 19 39.36& B3  & EB of $\beta$ Lyr type& V381 Cas ; BD+48 162 ; BV 179\\

1873: Roche-fill& T-Cas0-00430& 00 40 06.247& 50 14 15.64& K4  & EB of W UMa type      & V523 Cas ; GSC 03257-00167 ; WR 16 ; CSV 5867\\

1874:           &             &             &            &     &                       & 1RXS J004005.0+501414 ; TYC 3257-167-1\\

1875: Circular  & T-Cas0-00640& 00 47 06.277& 48 31 13.14&     & Star                  & TYC 3266-765-1 ; GSC 03266-00765\\

1876: Circular  & T-Cas0-00792& 00 48 26.554& 51 35 02.52&     & Star                  & TYC 3274-664-1 ; GSC 03274-00664\\

1877: Roche-fill& T-Cas0-02013& 00 40 46.427& 46 56 57.41&     & Star                  & TYC 3253-1767-1 ; GSC 03253-01767\\

1878: Inverted  & T-Cas0-02069& 00 49 17.959& 50 39 02.92&     & EB                    & V385 Cas\\

1879: Roche-fill& T-Cas0-08802& 00 51 32.351& 47 16 42.57&     & Star                  & GPM 12.884787+47.278540\\

1880: Roche-fill& T-CrB0-00654& 16 00 14.507& 35 12 31.56&     & EB of W UMa type      & AS CrB ; GSC 02579-01125 ; NSVS 7847829\\

1881:           &             &             &            &     &                       & ROTSE1 J160014.54+351228.4\\

1882: Roche-fill& T-CrB0-00705& 15 55 51.838& 33 11 00.39&     & EB of W UMa type      & ROTSE1 J155551.87+331100.5\\

1883: Roche-fill& T-CrB0-01589& 16 10 09.313& 35 57 30.57&     & Variable of $\delta$ Sct type& ROTSE1 J161009.33+355730.8\\

1884: Roche-fill& T-CrB0-01605& 16 00 58.472& 34 18 54.34&     &EB of W UMa or RR Lyr-C& NSVS 7848126 ; ROTSE1 J160058.45+341854.5\\

1885: Roche-fill& T-CrB0-04254& 16 09 19.589& 35 32 11.48&     & EB of W UMa type      & ROTSE1 J160919.62+353210.8\\

1886: Circular  & T-Cyg1-00246& 19 44 01.777& 50 13 57.42&     & Star                  & TYC 3565-643-1 ; GSC 03565-00643\\

1887: Roche-fill& T-Cyg1-00402& 19 54 39.939& 50 36 41.91&     & Star                  & TYC 3566-606-1 ; GSC 03566-00606\\

1888: Ambiguous & T-Cyg1-01385& 20 15 21.936& 48 17 14.14&     & Star                  & TYC 3576-2035-1 ; GSC 03576-02035\\

1889: Circular  & T-Cyg1-01627& 19 45 20.426& 51 35 07.22&     & Star                  & TYC 3569-1752-1 ; GSC 03569-01752\\

1890: Roche-fill& T-Cyg1-04652& 20 07 07.305& 50 34 01.34&     &EB of W UMa type       & GSC 03567-01035\\

1891: Roche-fill& T-Cyg1-04852& 19 51 59.208& 50 05 29.61&     &EB of W UMa type       & NSVS 5645908\\

1892: Circular  & T-Cyg1-09274& 20 16 06.814& 51 56 26.07&     & EB of W UMa type      & V1189 Cyg ; CSV 8488 ; GSC 03584-01600 ; SON 7885\\

1893: Roche-fill& T-Cyg1-11279& 19 59 53.377& 49 23 27.86&     & X-ray source          & 1RXS J195954.0+492318\\

1894: Roche-fill& T-Cyg1-12518& 19 58 15.339& 48 32 15.79&     & Variable star         & Mis V1132\\

1895: Roche-fill& T-Cyg1-14514& 19 48 05.077& 52 51 16.25&     &EB of W UMa or RR Lyr-C& V997 Cyg ; GSC 03935-02233 ; ROTSE1 J194804.79+525117.6 ; SON 7839\\

1896: Ambiguous & T-Dra0-00240& 17 03 52.919& 57 21 55.54&     & Star                  & TYC 3894-898-1  ; GSC 03894-00898\\

1897: Ambiguous & T-Dra0-00358& 16 45 38.339& 54 31 32.02&     & Star                  & TYC 3879-2689-1 ; GSC 03879-02689\\

1898: Circular  & T-Dra0-00398& 16 57 33.875& 59 31 51.98&     & EB of Algol type/X-ray source& RX J1657.5+5931 ; 1RXS J165733.5+593156\\

1899:           &             &             &            &     &                       & VSX J165733.8+593151 ; GSC 03898-00272\\

1900: Roche-fill& T-Dra0-00405& 16 27 49.103& 58 50 23.30&     & Star                  & TYC 3884-1488-1 ; GSC 03884-01488\\

1901: Roche-fill& T-Dra0-00959& 16 27 44.159& 56 45 59.30&     & EB of W UMa type/X-ray source& NSVS 2827877 ; 1RXS J162743.9+564557\\

1902: Circular  & T-Dra0-01363& 16 34 20.417& 57 09 48.95&M4.5V& EB of BY Dra type     & CM Dra ; CSI+57-16335 1 ; LSPM J1634+5709 ; G 225-67 ; G 226-16\\

1903:           &             &             &            &     &High proper-motion Star& IDS 16326+5721 A ; [RHG95] 2616 ; SBC7 580 ; CCDM J16343+5710A\\

1904:           &             &             &            &     &                       & GJ 630.1 A ; LP 101-15 ; IDS 16325+5721 A ; [GKL99] 324 ; LHS 421\\

1905:           &             &             &            &     &                       & 2MASS J16342040+5709439 ; CCABS 108 ; CABS 134 ; GEN\# +9.80225067\\

1906:           &             &             &            &     &                       & RX J1634.3+5709 ; 1RXH J163421.2+570941 ; 1RXS J163421.2+570933\\

1907:           &             &             &            &     &                       & PM 16335+5715 ; USNO 168 ; USNO-B1.0 1471-00307615 ; NLTT 43148\\

1908: Roche-fill& T-Dra0-01346& 16 52 12.345& 57 43 31.70&     & EB of Algol type      & BPS BS 16080-0095 ; VSX J165212.3+574331 ; GSC 03885-00583\\

1909: Roche-fill& T-Dra0-02224& 16 30 01.408& 54 45 55.80&     & Star                  & BPS BS 16084-0159\\

1910: Circular  & T-Dra0-03021& 17 01 03.618& 55 14 54.70&     & EB of Algol type      & VSX J170103.5+551455 ; GSC 03890-01216 \\

1911: Abnormal  & T-Dra0-03105& 16 23 02.558& 59 27 23.44&     & X-ray source          & 1RXS J162303.6+592717\\

1912: Roche-fill& T-Dra0-05259& 16 41 48.751& 56 22 34.40&     & EB of W UMa type      & VSX J164148.7+562234 ; GSC 03882-02264 ; USNO-B1.0 1463-0278621\\

1913: Ambiguous & T-Her0-00274& 17 00 51.150& 45 25 35.94&     & Star                  & TYC 3501-2245-1 ; GSC 03501-02245\\

1914: Roche-fill& T-Her0-01086& 16 48 15.539& 44 44 28.73&     & EB of W UMa type      & GSC 03082-00896 ; NSVS 5252572 ; 1RXS J164817.3+444430\\

1915: Roche-fill& T-Her0-03579& 16 35 47.390& 45 24 58.19&     & EB of W UMa type      & GSC 03499-01631\\

1916: Inverted  & T-Her0-05469& 16 54 51.245& 43 20 35.89&     & EB                    & V747 Her ; SVS 2066\\

1917: Circular  & T-Lyr1-00359& 19 15 33.695& 44 37 01.30& G0V & EB                    & V2277 Cyg ; GSC 03133-01149 ; ROTSE1 J191533.92+443704.9\\

1918:           &             &             &            &     & X-ray source          & BD+44 3087 ; ILF1+44 155 ; 1RXS J191533.7+443704\\

1919: Circular  & T-Lyr1-00687& 18 55 27.911& 47 13 41.76&     & Star                  & TYC 3544-1392-1 ; GSC 03544-01392\\

1920: Circular  & T-Lyr1-01013& 18 55 03.963& 47 49 08.39&     & Star                  & TYC 3544-2565-1 ; GSC 03544-02565\\

1921: Circular  & T-Lyr1-01439& 19 06 13.439& 46 57 26.42&     & Star                  & TYC 3545-2716-1 ; GSC 03545-02716\\

1922: Circular  & T-Lyr1-02109& 18 57 35.415& 45 07 44.10&     & Cepheid variable star & ROTSE1 J185735.99+450752.5\\

1923: Roche-fill& T-Lyr1-02166& 19 05 07.448& 46 15 07.51&     & X-ray source          & 1RXS J190504.8+461512\\

1924: Roche-fill& T-Lyr1-03173& 18 59 45.531& 47 20 07.34&     & EB of W UMa type      & ROTSE1 J185945.43+472007.0\\

1925: Roche-fill& T-Lyr1-03211& 18 45 56.939& 47 19 09.54&     & EB of W UMa type/X-ray source& ROTSE1 J184556.86+471914.4 ; 1RXS J184557.9+471906\\

1926: Roche-fill& T-Lyr1-03270& 18 57 33.098& 48 05 22.49&     & EB of W UMa type      & ROTSE1 J185733.12+480522.5\\

1927: Roche-fill& T-Lyr1-03783& 18 50 12.684& 45 35 44.05&     & Star                  & GPM 282.552858+45.595521\\

1928: Inverted  & T-Lyr1-04431& 19 12 16.047& 49 42 23.58&     & EB of Algol type      & NSV 11822 ; GSC 03550-01770 ; NSVS 5578839 ; SON 9371\\

1929: Roche-fill& T-Lyr1-05706& 18 47 57.211& 44 38 11.30&     & EB of W UMa type      & ROTSE1 J184757.18+443810.8\\

1930: Inverted  & T-Lyr1-05887& 18 52 10.489& 47 48 16.67&     & EB of Algol type      & WX Dra ; AN 24.1925\\

1931: Roche-fill& T-Lyr1-06583& 18 52 26.837& 44 55 20.86&     & EB                    & ROTSE1 J185226.53+445527.8\\

1932: Inverted  & T-Lyr1-07179& 18 49 14.039& 45 24 38.61&     & Star                  & GPM 282.308454+45.410868\\

1933: Roche-fill& T-Lyr1-08406& 18 50 06.942& 45 41 05.95&     & Star                  & GPM 282.528833+45.685035\\

1934: Roche-fill& T-Lyr1-10276& 18 46 55.088& 45 00 52.27&     & EB of W UMa type      & V596 Lyr ; GPM 281.729421+45.014635 ; GSC 03540-00085\\

1935:           &             &             &            &     &                       & ROTSE1 J184654.98+450054.7\\

1936: Inverted  & T-Lyr1-10989& 19 06 22.791& 45 41 53.82&     & EB of Algol type      & V512 Lyr ; SON 10931\\

1937: Roche-fill& T-Lyr1-11226& 18 45 21.748& 45 53 28.79&     & EB of W UMa type or $\delta$ Sct & V594 Lyr ; GPM 281.340617+45.891326 ; GSC 03540-01842\\

1938:           &             &             &            &     &                       & ROTSE1 J184522.47+455321.0\\

1939: Roche-fill& T-Lyr1-12772& 18 52 25.096& 44 55 40.23&     & EB of W UMa type      & ROTSE1 J185226.53+445527.8\\

1940: Abnormal  & T-Lyr1-13166& 19 02 28.120& 46 58 57.75& F9V & EB                    & V361 Lyr ; SON 9349\\

1941: Roche-fill& T-Per1-00328& 03 41 57.108& 39 07 29.60& G5  & EB of Algol type      & HD 275743 ; BD+38 787 ; GSC 02863-00755 ; TYC 2863-755-1\\

1942: Circular  & T-Per1-00459& 03 34 57.745& 39 33 18.70& G5  & Star                  & HD 275547 ; GSC 02866-01995 ; TYC 2866-1995-1\\

1943: Circular  & T-Per1-00750& 03 47 45.543& 35 00 37.08&     &Double or multiple star& TYC 2364-2327-1 ; GSC 02364-02327 ; CCDM J03478+3501BC\\

1944:           &             &             &            &     &                       & ADS 2771 BC ; BD+34 732B ; CSI+34 732 2 ; NSV 1302\\

1945: Roche-fill& T-Per1-00974& 03 34 43.738& 38 40 22.22&  A  & Star                  & HD 275481\\

1946: Circular  & T-Per1-01218& 03 42 33.165& 39 06 03.63&  A  & EB                    & HU Per ; HD 275742 ; SVS 922\\

1947: Roche-fill& T-Per1-01482& 03 48 45.999& 35 14 10.05& F0  & Star                  & HD 279025\\

1948: Circular  & T-Per1-02597& 03 44 32.202& 39 59 34.94& K4V & T Tau type Star       & [LH98] 94 ; 1RXS J034432.1+395937 ; 1SWASP J034433.95+395948.0\\

1949: Inverted  & T-Per1-04353& 03 45 04.887& 37 47 15.91&     & EB of Algol type      & HV Per ; SVS 368 ; P 107\\

1950: Roche-fill& T-Tau0-00397& 04 30 09.466& 25 32 27.05& A3  & EB of $\beta$ Lyr type& GW Tau ; SVS 1421 ; HD 283709 ; ASAS 043009+2532.4\\

1951: Inverted  & T-Tau0-00686& 04 07 13.870& 29 18 32.44&     & EB of Algol type      & IL Tau ; SON 9543\\

1952: Roche-fill& T-Tau0-00781& 04 12 51.218& 24 41 44.26& G9  &Eruptive/T Tau-type Star& V1198 Tau ; NPM2+24.0013 ; 1RXS J041250.9+244201\\

1953:           &             &             &            &     &                       & GSC 01819-00498 ; RX J0412.8+2442 ; [WKS96] 14\\

1954: Roche-fill& T-Tau0-01262& 04 16 28.109& 28 07 35.81& K7V &Variable Star of Orion Type& V1068 Tau ; EM StHA 25 ; JH 165 ; EM LkCa 4\\

1955:           &             &             &            &     &                       & HBC 370 ; ASAS 041628+2807.6\\

1956: Roche-fill& T-Tau0-01715& 04 19 26.260& 28 26 14.30& K7V &T Tau-type Star/X-ray source& V819 Tau ; HBC 378 ; NAME WK X-Ray 1 ; 1E 0416.3+2830\\

1957:           &             &             &            &     &                       & IRAS C04162+2819 ; TAP 27 ; [MWF83] P1 ; WK81 1\\

1958:           &             &             &            &     &                       & 1RXS J041926.1+282612 ; X 04163+283\\

1959: Roche-fill& T-Tau0-06463& 04 07 27.415& 27 51 06.36&     & EB of W UMa type      & V1022 Tau ; HV 6199 ; NSV 1464\\

1960: Inverted  & T-UMa0-00127& 09 38 06.716& 56 01 07.32& A2V & EB of Algol type      & VV UMa ; GEN\# +0.05601395 ; HIP 47279 ; TYC 3810-1290-1\\

1961:           &             &             &            &     &                       & GSC 03810-01290 ; SBC7 384 ; GCRV 6211 ; BD+56 1395\\

1962:           &             &             &            &     &                       & HIC 47279 ; SVS 770 ; AAVSO 0931+56\\

1963: Circular  & T-UMa0-00222& 10 07 18.023& 56 12 37.12& A0  & Star                  & HD 237866 ; GSC 03818-00504 ; SAO 27524 ; AG+56 778 ; HIC 49581\\

1964:           &             &             &            &     &                       & BD+56 1432 ; HIP 49581 ; YZ 56 6209 ; TYC 3818-504-1\\

1965: Roche-fill& T-UMa0-01701& 10 03 02.856& 55 47 53.34&     & X-ray source          & RX J100303.4+554752 ; [PTV98] H22 ; [PTV98] P29\\

1966: Circular  & T-UMa0-03090& 10 08 52.180& 52 45 52.49& K2e & Star                  & GSC 03815-01151 ; RIXOS 229-302 ; RX J100851.6+524553\\

1967: Roche-fill& T-UMa0-03108& 10 04 16.780& 54 12 02.83&     & EB of W UMa type      & NSVS 2532137\\

1968: \enddata

1969: \label{tableSIMBAD}

1970: \end{deluxetable}

1971: 

1972: 

1973: 

1974: \begin{deluxetable}{lcccccccccc}

1975: \tabletypesize{\tiny}

1976: \rotate

1977: \tablecaption{Low-mass EB candidates  ($M_{1,2} < 0.75M_\sun$ ; sorted by mass)}

1978: \tablewidth{0pt}

1979: \tablehead{\colhead{Category} & \colhead{Object} & \colhead{$\alpha$ (J2000)} & \colhead{$\delta$ (J2000)} & \colhead{Period $[days]$} & \colhead{$M_1/M_\sun$} & \colhead{$M_2/M_\sun$} & \colhead{age $[Gyr]$} &

1980:  \colhead{\begin{tabular}{c} Proper motion \\ source catalog\tablenotemark{a} \end{tabular}} & \colhead{\begin{tabular}{c} $PM_{\alpha}$ \\ mas/year \end{tabular}} & \colhead{\begin{tabular}{c} $PM_{\delta}$ \\ mas/year \end{tabular}}}

1981: \startdata

1982: Circular& T-Dra0-01363\tablenotemark{b}& 16 34 20.417& 57 09 48.95&  1.268&   0.27 $\pm$ 0.02&  0.24 $\pm$ 0.03&  1.6 $\pm$ 1.6  & \citet{Salim03} & -1121& 1186\\

1983: AmbigEq\tablenotemark{c}&  T-And0-10518& 01 07 44.417& 48 44 58.11&  0.387&   0.45 $\pm$ 0.27&  0.45 $\pm$ 0.28&  0.3 $\pm$ 0.5  & UCAC  &  2.7& -2.0\\

1984: AmbigEq&  T-Cyg1-12664& 19 51 39.824& 48 19 55.38&  8.257&   0.50 $\pm$ 0.20&  0.48 $\pm$ 0.19&  0.3 $\pm$ 0.4  & USNO-B&  -18&  -6\\

1985: AmbigEq&  T-CrB0-14232& 16 10 22.495& 33 57 52.33&  0.971&   0.60 $\pm$ 0.24&  0.55 $\pm$ 0.29&  4.4 $\pm$ 8.8  & UCAC  &-15.2& -24.2\\

1986: AmbigEq&  T-CrB0-14543& 15 57 45.926& 33 56 07.28&  1.506&   0.60 (-1)\tablenotemark{d}      &  0.60 (-1)      &  0.2 (-2)       & UCAC&  -13.9&  13.3\\

1987: Circular& T-Per1-13685& 03 53 51.217& 37 03 16.73&  0.384&   0.60 (-1)      &  0.50 (-1)      &  10.0 (-3)      & UCAC  &-24.1&-15.9\\

1988: AmbigEq&  T-CrB0-10759& 15 52 18.455& 30 35 32.13&  1.901&   0.63 $\pm$ 0.24&  0.62 $\pm$ 0.21&  7.3 $\pm$ 49.6 & UCAC  &  3.6&-19.4\\

1989: AmbigEq&  T-UMa0-08238& 10 09 25.384& 53 57 01.31&  1.250&   0.69 $\pm$ 0.54&  0.61 $\pm$ 0.51&  4.1 $\pm$ 15.0 & USNO-B&    6&   -4\\

1990: AmbigEq&  T-Cas0-10450& 00 29 16.288& 50 27 38.58&  8.656&   0.71 $\pm$ 0.21&  0.67 $\pm$ 0.20&  0.3 $\pm$ 0.4  & UCAC  & -3.1& -4.2\\

1991: AmbigEq&  T-Dra0-07116& 17 02 53.025& 55 07 47.44&  1.369&   0.71 $\pm$ 0.22&  0.69 $\pm$ 0.22&  2.1 $\pm$ 3.6  & USNO-B&   -2&  -16\\

1992: Circular& T-Tau0-04859& 04 08 11.608& 24 51 10.18&  3.068&   0.74 $\pm$ 0.10&  0.66 $\pm$ 0.10&  8.8 $\pm$ 14.8 & UCAC  &  3.4& -8.0\\

1993: \enddata

1994: \tablenotetext{a}{Where possible, we used the more accurate UCAC

1995: catalog, otherwise we reverted to the USNO-B catalog. Since they

1996: are dim and nearby, we expect most of the low-mass binaries to

1997: have comparably large proper motions.}

1998: \tablenotetext{b}{This binary is CM Draconis, which has been extensively

1999: studied and found to have a masses of $M_1 = 0.2307 \pm 0.0010 M_\sun$ and

2000: $M_2 = 0.2136 \pm 0.0010 M_\sun$ \citep{Lacy77a, Metcalfe96}. For consistency,

2001: we listed the MECI results, which are off by less than $0.04 M_\sun$

2002: ($\sim1.5\sigma$). We also adopted an alternate proper motion

2003: estimate, as its USNO-B values seems to be erroneous, probably due

2004: to its very high angular velocity.}

2005: \tablenotetext{c}{For clarity

2006: we list for the ambiguous systems, only the solution with

2007: approximately equal components. But it is likely that at least a

2008: few of the ambiguous systems may be unequal, with half the period.

2009: Such cases can be identified as single-line spectroscopic

2010: binaries, with the secondary component being no larger than a few

2011: $0.1M_\sun$.}

2012: \tablenotetext{d}{When the most likely model is at

2013: the edge of the parameter space, MECI is not able to bound the

2014: solution, and therefore cannot estimate the uncertainties. We mark

2015: (-3) when the upper limit was reached, (-2) when the lower limit

2016: was reached, and (-1) if one of the other parameter is at its

2017: limit.}

2018: \label{tableLowMass}

2019: \end{deluxetable}

2020: 

2021: 

2022: \clearpage

2023: 

2024: \begin{figure}

2025: \plotone{f1.eps}

2026: \caption{The fraction of the EBs in the final catalog found using

2027: the BLS algorithm and the AoV algorithm. The number

2028: of EBs in each bin is shown in Figure \ref{figPeriodDistrib}. The

2029: BLS method excelled at identifying EBs with short-duration eclipses

2030: (compared to the orbital period), which predominately occur at periods

2031:  $> 0.75$ days. The AoV method fared better with EBs that have

2032: long-duration eclipses, which predominately occur in sub-day periods. The AoV

2033: method also does well with EBs with period longer than $10$ days,

2034: which may be dominated by giant-giant binaries \citep{Derekas07}, and

2035: so also have broad eclipses. This plot demonstrates the importance

2036: of using multiple independent methods of identifying EB, otherwise

2037: the results will have a significant selection effect that may

2038: bias any statistical results.}

2039: \label{periodDiffFrac}

2040: \end{figure}

2041: 

2042: 

2043: \begin{figure}

2044: \plotone{f2.eps}

2045: \caption{An eclipse timing plot produced

2046: in step (4), showing the $O-C$ residuals of the primary eclipses

2047: (circles) and the secondary eclipses (triangles). Here, T-Lyr1-14962

2048: is shown with an assumed period of $5.710660\: days$, as measured

2049: with an AoV periodogram. The slope of the residuals indicates that

2050: the assumed period is inaccurate. The grey solid line is predicted

2051: by the best circular-orbit model, whereas the dashed lines are

2052: predicted by the best eccentric-orbit model (compare to Figure

2053: \ref{figTimingVariationsEcc}). After correction, we get a

2054: fine-tuned period of $5.712516\: days$. This 0.03\% correction is

2055: small but significant in that without having had this correction,

2056: the eclipses would have smeared out and widened.}

2057: \label{figTimingVariations}

2058: \end{figure}

2059: 

2060: 

2061: \begin{figure}

2062: \plotone{f3.eps}

2063: \caption{An eclipse timing plot for

2064: T-Cyg1-01373, with an assumed period of $4.436013\: days$. In contrast

2065: to Figure \ref{figTimingVariations}, the slope here is consistent

2066: with zero, thus indicating that the period does not need to be

2067: fine-tuned. However, the $O-C$ offset between the primary

2068: (circles) and secondary (triangle) eclipses is significant

2069: ($1449\: seconds$), indicating that this EB has an eccentric

2070: orbit. The reduced chi-squared of the best circular-orbit model

2071: (grey solid line) is $\chi^2_\nu = 12.9$, while the reduced

2072: chi-squared of the best eccentric-orbit model (dashed lines) is

2073: $\chi^2_\nu = 0.95$. Applying the $O-C$ timing offset to equation

2074: \ref{eqOmC} provides a lower limit to the binary's orbital

2075: eccentricity: $e \geq | e \cos \omega | \simeq 0.00594$.}

2076: \label{figTimingVariationsEcc}

2077: \end{figure}

2078: 

2079: 

2080: \begin{figure}

2081: \plotone{f4.eps}

2082: \caption{The distribution of the catalog position errors when

2083: matching targets to the proper motion databases. In some cases,

2084: the position errors are dominated by the

2085: motion of the EB during the intervening years.}

2086: \label{figPosErr}

2087: \end{figure}

2088: 

2089: 

2090: \begin{figure}

2091: \plotone{f5.eps}

2092: \caption{The distribution of the $R$-band third-light flux fraction in

2093: the catalog LCs. This fraction was calculated by summing

2094: the fluxes of all the USNO-B sources within 30" of the target, excluding the

2095: target, and dividing this value by the total flux within 30", including the

2096: target. The resulting fraction ranges from 0 to 1.}

2097: \label{figBlending}

2098: \end{figure}

2099: 

2100: 

2101: \begin{figure}

2102: \plottwo{f6a.eps}{f6b.eps}

2103: \caption{MECI likelihood contour plots of a typical circular-orbit

2104: EB (T-And0-00745 ; left) and eccentric-orbit EB (T-UMa0-01822 ;

2105: right). There is no significant difference in the way MECI handles

2106: these cases, and both usually have a single contour

2107: minimum. The plots shown here have the ages set to the values

2108: that produced the lowest MECI minima.}

2109: \label{figMECI1}

2110: \end{figure}

2111: 

2112: 

2113: \begin{figure}

2114: \plottwo{f7a.eps}{f7b.eps}

2115: \caption{MECI likelihood contour plots of a typical ambiguous EB

2116: (T-Tau0-03579). These plots show the effect of assuming that the

2117: binary components are equal (left) or unequal (right). Note that

2118: the equal-component solution will have a nearly symmetric contour

2119: around the diagonal, while the unequal-component solution can

2120: provide only an upper limit to the secondary component's mass,

2121: in this case $M_2 \lesssim 1\:M_\sun$. The plots shown here have

2122: the ages set to the values that produced the lowest MECI minima.}

2123: \label{figMECI2}

2124: \end{figure}

2125: 

2126: 

2127: \begin{figure}

2128: \plotone{f8.eps}

2129: \caption{The EB orbital period distribution within the catalog.

2130: Each bin is subdivided to show the number of binaries belonging to

2131: each of the classification groups described in \S\ref{secMethod}.

2132: Note that the ambiguous-equal and ambiguous-unequal entries

2133: represent the same stars, with entries in the former group

2134: having double the period of the latter. Notice also how the Roche-lobe-filling

2135: EBs dominate the sub-day bins, and have a long tail stretching well above

2136: 10-day periods. Furthermore, the circular orbit EBs have a period

2137: distribution peak of at $\sim$2 days, while the eccentric orbit EBs

2138: peak at $\sim$5 days. This is likely due to the orbital

2139: circularization that occurs preferentially in short-period systems

2140: (see also Figures \ref{figPeriodEcc} and \ref{figPeriodM1}).}

2141: \label{figPeriodDistrib}

2142: \end{figure}

2143: 

2144: \clearpage

2145: 

2146: \begin{figure}

2147: \plotone{f9.eps}

2148: \caption{Eccentric EBs (panel 1).

2149: Note how the secondary eclipse is not at phase 0.5, as would be in circular orbit EBs.}

2150: \label{figEcc1}

2151: \end{figure}

2152: 

2153: \begin{figure}

2154: \plotone{f10.eps}

2155: \caption{Eccentric EBs (panel 2).}

2156: \label{figEcc2}

2157: \end{figure}

2158: 

2159: \begin{figure}

2160: \plotone{f11.eps}

2161: \caption{Eccentric EBs (panel 3).}

2162: \label{figEcc3}

2163: \end{figure}

2164: 

2165: 

2166: \begin{figure}

2167: \plotone{f12.eps}

2168: \caption{Examples of unambiguous

2169: EBs with circular orbits, with their best-fit MECI models (solid line).}

2170: \label{figE0Cat}

2171: \end{figure}

2172: 

2173: 

2174: \begin{figure}

2175: \plotone{f13.eps}

2176: \caption{Examples of ambiguous EBs. Left

2177: column: assuming very unequal components. Right column: assuming

2178: approximately equal components with double the period.}

2179:  \label{figAmbigCat}

2180: \end{figure}

2181: 

2182: \clearpage

2183: 

2184: \begin{figure}

2185: \plotone{f14.eps}

2186: \caption{Examples of EBs classified as inverted EBs. We included the

2187: unsuccessful best-fit MECI model (solid curve) as an approximate reference to

2188: illustrate the LC of a corresponding binary that has had no mass transfer.

2189: Notice how the model LC is unable to achieve a sufficiently deep primary

2190: eclipse, while producing a secondary eclipse that is too deep.}

2191: \label{figInvertedCat}

2192: \end{figure}

2193: 

2194: 

2195: \begin{figure}

2196: \plotone{f15.eps}

2197: \caption{Examples of EBs that are assumed to have filled at least one of their Roche lobes.

2198: We included, for illustration purposes only, their best-fit MECI models (solid line).

2199: These models were not adopted since they neglect tidal distortions, reflections, and gravity

2200: darkening effects, and so produce a poor fit to the data.}

2201: \label{figRocheFilledCat}

2202: \end{figure}

2203: 

2204: 

2205: \begin{figure}

2206: \plotone{f16.eps}

2207: \caption{The criterion applied in

2208: equations \ref{eqRoche1} and \ref{eqRoche2} to determine whether one or both the EB

2209: components have filled their Roche lobe, and thus need to be

2210: placed into group (6).}

2211: \label{figRoche}

2212: \end{figure}

2213: 

2214: \clearpage

2215: 

2216: \begin{figure}

2217: \plotone{f17.eps}

2218: \caption{LCs of abnormal EBs (panel 1).}

2219: \label{figAbnormalEBs1}

2220: \end{figure}

2221: 

2222: 

2223: \begin{figure}

2224: \plotone{f18.eps}

2225: \caption{LCs of abnormal EBs (panel 2).}

2226: \label{figAbnormalEBs2}

2227: \end{figure}

2228: 

2229: \begin{figure}

2230: \plotone{f19.eps}

2231: \caption{The mass-mass relation for the

2232: detached EBs of the TrES dataset. Each category is represented by

2233: a different symbol. Note that the ambiguous EBs are plotted twice,

2234: where only one of the solution can be correct. Notice also that in

2235: the equal-component solutions are clustered along the diagonal,

2236: while the unequal-component solutions are clustered along the

2237: minimum available mass of the Yonsei-Yale isochrones ($0.4

2238: \:M_\sun$). Some of the ambiguous solutions deviate from these

2239: clusters due to poor constraints on the secondary eclipse, which

2240: brings about a large uncertainty. Finally, notice the sparsity of

2241: EBs populating the low-mass corner of this plot ($M_{1,2} <

2242: 0.75\:M_\sun$). These systems, whose importance is outlined in

2243: \S\ref{subsecLowMassEBs}, were modeled using the Baraffe

2244: isochrones. CM Draconis (T-Dra0-01363) clearly sets itself apart,

2245: being the lowest-mass binary in the catalog (circle at

2246: bottom-left).}

2247: \label{figMassMass}

2248: \end{figure}

2249: 

2250: 

2251: \begin{figure}

2252: \plotone{f20.eps}

2253: \caption{The relation between the EB

2254: discovery yield (the fraction of LCs found to be EBs) and the

2255: absolute value of the Galactic latitude, or $|b|$, for the 10 TrES

2256: fields used in this catalog (see Tables \ref{tableFieldsObs} and

2257: \ref{tableFieldsYield}). The solid line is the linear regression

2258: of the log of the EB discovery yield ($r^2 = 0.867$). Some of the

2259: residual scatter can be explained as being due to differences in

2260: the duration of observations in each field. By including the

2261: duration in a bi-linear regression, we get a substantially improved

2262: fit ($r^2 = 0.911$).}

2263: \label{figDiscoveryYield}

2264: \end{figure}

2265: 

2266: 

2267: \begin{figure}

2268: \plotone{f21.eps}

2269: \caption{Low mass candidates ($M_1 < 0.75 M_\sun$), with their best-fit MECI models (solid line).}

2270: \label{figLowMassCat}

2271: \end{figure}

2272: 

2273: 

2274: \begin{figure}

2275: \plotone{f22.eps}

2276: \caption{The period-eccentricity

2277: relation. The lower end of the error bars were truncated, where

2278: needed, by the measured lower limit, $|e \cos \omega|$. Note the

2279: lack of eccentric short period systems. The diagonal line is

2280: provided to guide the eye.}

2281: \label{figPeriodEcc}

2282: \end{figure}

2283: 

2284: 

2285: \begin{figure}

2286: \plotone{f23.eps}

2287: \caption{The period-primary mass relation

2288: for eccentric EBs. We included all systems with well-determined

2289: masses. The area of the grey circles is proportional to the EB's

2290: eccentricity. All the curves are theoretical boundaries, assuming

2291: that the binary components are both on the main sequence and have

2292: equal masses ($q=1$). The left-most

2293: dot-dash line demarcates the binary contact limit, and the

2294: remaining curves mark systems with increasing circularization time

2295: (see equation \ref{eqCircTime}). Note the abrupt increase in the

2296: circularization time for systems more massive than

2297: $\sim1.25M_\sun$, at which point the stellar convective envelope

2298: becomes radiative, and thus far less efficient at tidal

2299: dissipation.}

2300: \label{figPeriodM1}

2301: \end{figure}

2302: 

2303: \end{document}

2304: