1: % latex ms ; bibtex ms ; latex ms ; latex ms ; dvips ms -o
2: % latex ms ; latex ms ; dvips ms -o
3: \documentclass[12pt,preprint]{aastex}
4:
5: %% manuscript produces a one-column, double-spaced document:
6: % \documentclass[manuscript]{aastex}
7: %% preprint2 produces a double-column, single-spaced document:
8: %\documentclass[preprint2]{aastex}
9:
10: %\usepackage{natbib}
11: %\usepackage{graphics}
12: %\usepackage{epsfig}
13: %\usepackage{amssymb}
14: %\usepackage{rotating}
15: %\usepackage{lscape}
16: %\usepackage{longtable}
17:
18: \shorttitle{STELLAR ROTATION IN M35}
19: \shortauthors{Meibom et al.}
20:
21: \begin{document}
22:
23: \title{STELLAR ROTATION IN M35: MASS-PERIOD RELATIONS, SPIN-DOWN RATES,
24: AND GYROCHRONOLOGY\altaffilmark{1}}
25:
26: \author{S{\o}ren Meibom\altaffilmark{2,3,4} and Robert D. Mathieu
27: \altaffilmark{4}}
28: \affil{Department of Astronomy, University of Wisconsin - Madison, Madison,
29: WI, 53706, USA}
30:
31: \author{Keivan G. Stassun\altaffilmark{4}}
32: \affil{Physics and Astronomy Department, Vanderbilt University, Nashville, TN - 32735, USA}
33:
34: \altaffiltext{1}{WIYN Open Cluster Study. XXXIII.}
35: \altaffiltext{2}{{\it smeibom@cfa.harvard.edu}}
36: \altaffiltext{3}{{Now at Harvard-Smithsonian Center for Astrophysics, Cambridge, MA, 02138, USA}}
37: \altaffiltext{4}{Visiting Astronomer, Kitt Peak National Observatory,
38: National Optical Astronomy Observatory, which is operated by the
39: Association of Universities for Research in Astronomy, Inc. (AURA)
40: under cooperative agreement with the National Science Foundation.}
41:
42:
43: % *****************************************************************************
44:
45: \begin{abstract} \label{abs}
46:
47: We present the results of a 5 month photometric time-series survey
48: for stellar rotation over a $40\arcmin \times 40\arcmin$ field centered
49: on the 150\,Myr open cluster M35. We report rotation periods for 441 stars
50: within this field and determine their cluster membership and binarity
51: based on a decade-long radial-velocity survey, proper-motion measurements,
52: and multi-band photometric observations. We find that 310 of the stars
53: with measured rotation periods are late-type members of M35. The distribution
54: of rotation periods for cluster members span more than two orders of
55: magnitude from $\sim 0.1 - 15$ days, not constrained by the sampling
56: frequency and the time-span of the survey. With an age between the zero-age
57: main-sequence and the Hyades, and with $\sim$6 times more rotation
58: periods than measured in the Pleiades, M35 permit detailed
59: studies of early rotational evolution of late-type stars. Nearly 80\%
60: of the 310 rotators lie on two distinct sequences in the color-period
61: plane, and define clear relations between stellar rotation period and color
62: (mass). The M35 color-period diagram enables us to determine timescales
63: for the transition between the two rotational states, of $\sim$60\,Myr
64: and $\sim$140\,Myr for G and K dwarfs, respectively. These timescales
65: are inversely related to the mass of the convective envelope, and offer
66: constraints on the rates of internal and external angular momentum
67: transport and of the evolution of stellar dynamos. A comparison to the
68: Hyades, confirm the Skumanich (1972) spindown-dependence for G dwarfs
69: on one rotational state, but suggest that K dwarfs spin down more slowly.
70: The locations of the rotational sequences in the M35 color-period diagram
71: support the use of rotational isochrones to determine ages for coeval
72: stellar populations. We use such gyrochronology to determine ``gyro-ages''
73: for M35 from 134\,Myr to 161\,Myr. We use the M35 data to evaluate new color
74: dependencies for the rotational isochrones.
75:
76: \end{abstract}
77:
78: \keywords{open clusters and associations:general - stars:late-type
79: - stars:rotation - stars:ages - stars:spots - binary stars:rotation}
80:
81:
82: % *****************************************************************************
83:
84: \section{INTRODUCTION}
85: \label{intro}
86:
87: Observations of coeval populations of late-type stars younger
88: than the Hyades have revealed that they rotate with periods
89: ranging over two orders of magnitude - from near breakup to
90: periods similar to the Sun. Understanding why some stars
91: deplete their angular momentum faster than others, which
92: physical processes are at work, when, and to what extent,
93: is a primary mandate for stellar evolution research.
94:
95: The discovery from photometric and spectroscopic measurements
96: in the Pleiades of sub 1-day rotation periods for K dwarfs
97: \citep{av81,va82,mav82,sjw83} challenged prior
98: understanding of the early angular momentum evolution for late-type stars,
99: and sparked a renewed interest in the topic. Observations of mainly projected
100: rotation velocities ($v \sin(i)$) of late-type stars in $\alpha$ Persei
101: (50 Myr; \citet{shb+85,shj89}), the Pleiades (100 Myr;
102: \citet{sjw83,shs+84,bmm84,sh87,ssh+93,tsp+00}), and the Hyades (625 Myr;
103: \citet{soderblom82,bmm84,rtl+87}), and photometric studies of the Hyades
104: \citep{ltr+84,rtl+87}, confirmed the coexistence of slowly and rapidly
105: rotating stars in $\alpha$ Persei and the Pleiades, but found an absence
106: of rapid rotators in the Hyades. Furthermore, the rapid rotators in the
107: youngest clusters were found primarily among K and M dwarfs and not among
108: G dwarfs.
109:
110: The emerging evidence for an age- and mass-dependence of rapid rotation
111: prompted new ideas about the rotational evolution of late-type stars.
112: For example the suggestion of epochs of decoupling and recoupling of the
113: stellar core and envelope \citep[e.g.][]{shs+84,ssm+93,jc93}. The idea of
114: decoupling - allowing the more shallow convective envelope of G dwarfs
115: to spin down faster than the envelopes in K and M dwarfs - had developed
116: in parallel in models of stellar rotation \citep[e.g.][]{es81,pkd90,mb91,bs96}.
117: However, the concept of decoupling, if permanent, is in conflict with
118: helioseismic observations of the sun as a solid-body rotator
119: \citep{gough82,ddg+84,gdk+91,ekj02}. Furthermore, recoupling - giving
120: access to the angular momentum reservoir of the faster spinning core -
121: was suggested by \citet{ssm+93} as being necessary to explain the evolution
122: beyond the age of the Pleiades of slowly rotating stars \citep{ssm+93}.
123:
124: Understanding the formation of the rapid rotators is a separate problem.
125: The fastest spinning stars in the youngest clusters cannot be explained
126: from Skumanich-style spin-down \citep{skumanich72} of the fastest spinning
127: T Tauri stars. The rapid rotators can be explained only by introducing
128: ``magnetic saturation'' of the angular momentum loss via stellar winds
129: \citep{sh87,mb91,bs96,bfa97,kpb+97,spt00}, and by allowing the saturation
130: threshold to depend on the stellar mass. The physical meaning of
131: ``saturation'' is still unclear.
132:
133: During the pre main-sequence (PMS) phase, large dispersions and substructure
134: (bimodalities) in the rotation-period distributions has also been observed
135: for coeval stellar populations. Here, a popular explanation for coeval rapid
136: and slow rotators originates from the work of \citet{konigl91} and
137: \citet{esh+93} on interactions between T Tauri stars and their circumstellar
138: disks. ``Magnetic disk-locking'' was introduced to provide a means to brake
139: the spin-up of the central star by transferring angular momentum from the
140: star to the disk \citep[e.g.][and references therein]{sno+94,najita95}.
141: Accordingly, rotation-period dispersions (bimodalities) should result if
142: some stars lose their disks faster than others (e.g. weak vs. classic T
143: Tauri stars). Whether magnetic disk-braking is a dominant process regulating
144: stellar rotation during the early PMS remains under debate on both
145: observational and theoretical grounds.
146:
147: Recently, taking advantage of results from an increasing number of
148: photometric monitoring programs of late-type stars in young main-sequence
149: clusters, \citet{barnes03a} presented an interpretation of his own
150: and other published rotation period data. Free of the ambiguities of
151: $v \sin(i)$, \citeauthor{barnes03a} identified, in each coeval stellar
152: population, separate groups of fast and intermediate/slowly rotating
153: stars with different dependencies on color (mass). He specifically
154: proposed that coeval stars fall along two ``rotational sequences'' in
155: the color vs. rotation period plane. From an analysis of these sequences
156: and their dependencies on stellar age, \citet{barnes03a} proposed a
157: framework for connecting internal and external magneto-hydrodynamic
158: processes to explain the evolution in the observed period distributions,
159: including bimodalities. This approach combines the ideas of
160: \citep[e.g.][]{shs+84,ssm+93} of an initial decoupling of the stellar
161: core and envelope with re-connection of the two zones through a global
162: dynamo-field at a later and mass-dependent time. It does not (yet)
163: interface with PMS angular momentum evolution.
164: Importantly, \citet{barnes03a} also proposed that the age-dependence of
165: the location of the rotational sequences in the color-period plane,
166: could be used to measure the age of a stellar population, much like
167: the sequences in the color-magnitude diagram. \citet{barnes07} further
168: developed this idea of ``gyrochronology''.
169: Determining the age of a late-type star from its rotation
170: and color, had previously been suggested by \citet{kawaler89}.
171: % based on theoretical models of angular momentum loss.
172:
173: Interpretation aside, it has become desirable and increasingly
174: possible to eliminate the ambiguity of projected rotation velocities
175: ($v\sin i$) by measuring photometric rotation periods from light
176: modulation by spots on the surfaces of young late-type stars.
177: While more labor and time intensive, photometric measurements of
178: rotation periods in coeval populations,
179: promise to reveal important details about dependencies of rotation
180: on other stellar properties - most obviously mass and age, but
181: likely also stellar activity, internal/external magnetic configurations,
182: and binarity.
183:
184: We present in this paper the results of an extensive time-series
185: photometric survey for rotation periods and a decade-long spectroscopic
186: surveys for membership and binarity for late-type stars in the field
187: of the open cluster M35 (NGC\,2168). The combination of time-series
188: and multi-band photometry with time-series radial-velocity data enable
189: us to explore the distribution of rotation periods vs. stellar color
190: (mass) for {\it bona fide} single and binary members of M35.
191:
192: M35 is a rich ($\ga$ 2500 stars; \citet{nds01}) northern hemisphere
193: cluster located $\sim$800-900 pc \citep{nds01,kfr+03} toward the
194: galactic anti-center ($\alpha_{2000} = 6^{h}~9^{m}$, $\delta_{2000}
195: = 24\degr~20\arcmin$; $l = 186\fdg59$, $b = 2\fdg19$). The age of
196: M35 has been estimated to 150\,Myr \citep{vss+02}, 175\,Myr \citep{nds01},
197: and 180\,Myr \citep{kfr+03}. We adopt an age of 150\,Myr, in agreement
198: with the most recent age determination based on the isochrone method
199: \citep{deliyannis08}.
200: At the distance of M35 the majority of cluster members are confined
201: to within a $\sim 0.5\degr$ diameter field, facilitating photometric
202: and spectroscopic observations of a large number of stars through
203: wide-field CCD imaging and multi-object spectroscopy. Older and much
204: more populous than the Pleiades, and younger than the Hyades, M35
205: nicely bridges a gap in the age sequence of star clusters with
206: comprehensive information about rotation and membership, permitting
207: a more detailed study of the rotational evolution of late-type stars
208: beyond the zero-age main-sequence (ZAMS).
209:
210: We begin in Section~\ref{obs} by describing our time-series photometric
211: observations, our methods for data-reduction and for photometric period
212: detection, and the information about cluster membership available from
213: our spectroscopic survey and the M35 color-magnitude diagram. In
214: Section~\ref{distr} we present the distribution of rotation periods
215: in M35, discuss the short- and long-period tails in the context of our
216: period detection limits, and assess the stability/lifetime of spots or
217: groups of spots by comparison of our short- and long-term photometric data.
218: Section~\ref{cpd} introduces the M35 color-period diagram and the
219: observed dependencies of stellar rotation on mass. In Section~\ref{discuss}
220: we discuss and interpret the M35 color-period diagram in the context
221: of present ideas for stellar angular momentum evolution. In particular
222: we use the diagram to estimate spin-down rates for G and K dwarfs, to
223: test the time-dependence on rotational evolution from a comparison with
224: measured rotation periods in the Hyades, determine M35's gyrochronology
225: age, and to evaluate the best functional representation of the color-period
226: dependence in the M35 data. Section~\ref{conclusions} summarizes and
227: presents our conclusions.
228:
229:
230: % *****************************************************************************
231:
232:
233: \section{OBSERVATIONS, DATA REDUCTION, AND METHODS} \label{obs}
234:
235: \subsection{Time-Series Photometric Observations} \label{programs}
236:
237: We photometrically surveyed stars in a region approximately
238: $40\arcmin \times 40\arcmin$ centered on the open cluster M35
239: over a timespan of 143 days. The photometric data were
240: obtained in the Johnson V-band with the WIYN \footnote{The WIYN
241: Observatories are joint facilities of the University of Wisconsin-Madison,
242: Indiana University, Yale University, and the National Optical Astronomy
243: Observatories.} 0.9m telescope \footnote{The 0.9m telescope is operated
244: by WIYN Inc. on behalf of a Consortium of ten partner Universities
245: and Organizations (see http://www.noao.edu/0.9m/general.html)} on
246: Kitt Peak equipped with a $2k \times 2k$ CCD
247: camera. The field of view of this instrument is $20.5\arcmin
248: \times 20.5\arcmin$ and observations where obtained over a $2
249: \times 2$ mosaic.
250:
251: The complete dataset presented is composed of images from
252: high-frequency (approximately once per hour for 5-6 hours per night)
253: time-series photometric observations over 16 full nights from 2-17 December
254: 2002, complemented with a queue-scheduled observing program over
255: 143 nights from 22 October 2002 to 11 March 2003, obtaining one
256: image per night interrupted only by bad weather and scheduled
257: instrument changes. The result is a database of differential V-band
258: light curves for more than 14,000 stars with $12 \la V \la 19.5$.
259: The sampling frequency of the December 2002 observations allow
260: us to detect photometric variability with periods ranging from
261: less than a day to about 10 days. The long time-span of the
262: queue-scheduled observations provide data suitable for detecting
263: periodic variability of up to $\sim$ 75 days, and for testing
264: the long-term stability of short-period photometric variations.
265: From this database we derive rotation periods for 441 stars.
266:
267: Figure~\ref{spatial} shows the surveyed region (solid square)
268: and the spatial distribution of the 441 rotators. The region
269: is roughly coincident with that of \citet{deliyannis08} in which
270: they obtained UBVRI CCD photometry for $\sim$19,000 stars (dashed
271: square). Also shown is the circular target region of our spectroscopic
272: survey described in Section~\ref{spec} and in \citet{mm05}. The
273: photometric survey was carried out within the region of the
274: spectroscopic survey to optimize information about spectroscopic
275: membership and binarity\footnote{All of these mutually supportive
276: studies are parts of the WIYN Open Cluster Study (WOCS; \citet{mathieu00}).}.
277:
278: %\clearpage
279: \begin{figure}[ht!]
280: \epsscale{1.0}
281: \plotone{f1.eps}
282: \caption{The locations and spatial extents of the photometric
283: and spectroscopic surveys used in this study. The innermost
284: solid square denotes the $40\arcmin \times 40\arcmin$ region
285: of our time-series photometric survey. Within it we show the
286: distribution of the 441 stars with measured rotation periods
287: (black dots). The dashed rectangle
288: displays the region of the multi-band photometric study by
289: \citet{deliyannis08} and the circle represents the 1-degree
290: diameter field of our spectroscopic survey of M35 \citep{mm05}.
291: \label{spatial}}
292: \end{figure}
293: %\clearpage
294:
295: Figure~\ref{sampling} displays the time-series data from both
296: programs for a photometrically non-variable star. Filled symbols
297: represent the high-frequency observations and open symbols
298: represent the queue-scheduled observations.
299:
300: %\clearpage
301: \begin{figure}[ht!]
302: \epsscale{1.0}
303: \plotone{f2.eps}
304: \caption{Sample time-series data from from our photometric database
305: for a non-variable $V \simeq$14th magnitude star. Filled symbols
306: represent measurements from the high-frequency December 2002 observing
307: run and open symbols represent the low-frequency queue-scheduled
308: observations. The data span a total of 143 days. The star was observed
309: in all 157 images of the north-east quadrant of the $2 \times 2$ mosaic.
310: The standard deviation ($\sigma_{\delta V}$) of the 157 measurements
311: is $0\fm004$, representative of our best photometric precision. The
312: horizontal dotted lines denote $\delta V = \pm 0\fm01$.
313: \label{sampling}}
314: \end{figure}
315: %\clearpage
316:
317: % -----------------------------------------------------------------------------
318:
319: \subsection{Basic Reductions, PSF Photometry, and Light Curves} \label{psf}
320:
321: Basic reductions of our CCD frames, identification of stellar sources,
322: and computations of equatorial coordinates\footnote{We used data from
323: the STScI Digitized Sky Survey; The Digitized Sky Surveys were produced
324: at the Space Telescope Science Institute under U.S. Government grant
325: NAG W-2166.} were done using standard IRAF packages.
326: %
327: Instrumental magnitudes were determined from Point Spread Function
328: (PSF) photometry using the IRAF DAOPHOT package. The analytical PSF
329: and a residual lookup table were derived for each frame based on
330: $\sim$30 evenly distributed isolated stars. The number of measurements
331: in the light curve of a given
332: star vary because stars near the edges of individual frames may be
333: missed due to telescope pointing errors, while bright stars near the
334: CCD saturation limit and faint stars near the detection threshold may
335: be excluded on some frames because of variations in seeing, sky brightness,
336: and sky transparency. To ensure our ability to perform reliable
337: time-series analysis on stars in our database, we have eliminated
338: stars that appear on fewer than half of the total number of frames.
339: The resulting database contains 14022 stars with a minimum of 75
340: measurements.
341:
342: We applied the \citet{honeycutt92} algorithm for differential
343: CCD photometry to our raw light curves to remove non-stellar
344: frame-to-frame photometric variations. We favor this technique
345: for differential photometry because it does not require a
346: particular set of comparison stars to be chosen {\it a priori},
347: nor does it require a star to appear in every frame.
348: %
349: Figure~\ref{sampling} shows the light curve for a $V \sim$14th
350: magnitude star. The standard deviation from 157 photometric
351: measurements is 0\fm004, representative of our photometric
352: precision at that brightness. Figure~\ref{m0_sigst} shows the
353: standard deviation of the photometric measurements as a function
354: of the V magnitude for each star in the field of M35. The relative
355: photometric precision is $\sim$0.5\% for stars with $12 \la V \la 14.5$.
356:
357: %\clearpage
358: \begin{figure}[ht!]
359: \epsscale{1.0}
360: \plotone{f3.eps}
361: \caption{The logarithm of the standard deviation of all instrumental
362: magnitudes as a function of V magnitude for 14022 stars in the field
363: of M35. The number of measurements for each star range from 75 to 216.
364: The solid and dashed
365: horizontal lines represent $\sigma_V$ of 0.01 (1\%) and 0.005 (0.5\%),
366: respectively. A relative photometric precision of $\sim$ 0.5\% is
367: obtained for stars with $12\fm0 \la V \la 14.5\fm0$.
368: \label{m0_sigst}}
369: \end{figure}
370: %\clearpage
371:
372: % -----------------------------------------------------------------------------
373:
374: \subsection{Photometric Period Detection} \label{periods}
375:
376: We employed the \citet{scargle82} periodogram analysis to detect
377: periodic variability in the light curves because of its ability to
378: handle unevenly sampled data. We searched a grid of 5000 frequencies
379: corresponding to periods between 0.1 day and 90 days. The lower
380: search limit was set at a period ensuring critical sampling based
381: on the Nyquist critical frequency for our high-frequency data
382: ($f_{c} = 1/(2 \delta t)$, where $\delta t$ is the sampling interval
383: of $\sim$1 hour. The upper limit was set at 90 days because a star
384: with a 90-day period would complete about 1.5 cycle over the 143
385: nights of the survey.
386:
387: A false alarm probability (FAP), the probability that a signal
388: detected at a certain power level can be produced by statistical
389: fluctuation, was calculated as the measure of confidence in a
390: detected period. An analytical expression for estimating a FAP
391: is given by \citet{scargle82} and \citet{hb86}. However, these
392: methods are not entirely suitable when applied to time-series
393: photometric studies of young stars because they only test against
394: random fluctuations of a purely statistical nature (i.e., measurement
395: errors) and do not account for correlated fluctuations intrinsic
396: to the source such as variability on timescales long compared to
397: the sampling frequency. For young stars our repeated measurements
398: during a single night are not necessarily independent and uncorrelated.
399: Consequently, the analytical expressions estimating a FAP will
400: likely overestimate the significance of any measured periodic
401: variability. Hence, we performed a two-dispersion Monte Carlo
402: calculation to estimate the FAP of our detected periods, as per
403: \citet{hw96} and \citet{smm+99}.
404: For each star, we generated a set of 100 synthetic
405: light curves, each consisting of normally distributed noise with
406: two dispersions: one representing the variability of the star during
407: a night and one representing the night-to-night variability of the
408: star. The former was estimated by taking the mean of each night's
409: standard deviation, and the latter by taking the standard deviation
410: of nightly means. With this approach, the test light curves can
411: vary on timescales that are long compared to our sampling interval,
412: allowing them to mimic the random slow variability of stellar origin
413: that could produce spurious periodic behavior over our limited observing
414: window. The maximum power of the 100 periodograms of the test light
415: curves was adopted as the level of 1\% FAP, and used as the initial
416: threshold for detecting significant photometric variability. For all
417: stars that met the FAP criterion we examined (by eye) the periodogram
418: and raw and phased light curves. We report stellar rotation periods
419: for 441 stars in our database (see Table 1 in Appendix B).
420:
421: We do not have multiple seasons of observations or observations
422: in multiple pass-bands at our disposal by which to confirm rotation
423: periods of individual stars. However, the reliability of the
424: derived periods is supported by an observed correlation between
425: photometric period and rotational line broadening within a subset
426: of 16 single cluster members. Figure~\ref{vsini} shows the projected rotation
427: velocities measured by \citet{nds01} for 16 stars for which we
428: have determined rotation periods. The shortest period stars
429: ($P_{r} \la 2.5$ days) show increasing $v \sin(i)$ with decreasing
430: rotation period. For rotation periods of $\sim$ 4 days or longer
431: the upper limits on the projected rotation velocities are consistent
432: with slower rotation. For comparison, the solid, dashed, and dotted
433: curves in Figure~\ref{vsini} indicate the relation between rotation
434: period and the projected rotational velocity for a solar-like star
435: with a $90\degr$, $70\degr$, and $50\degr$ inclination of the rotational
436: axis, respectively. Thus, for all 16 stars, the projected rotation
437: velocities are consistent with the measured rotation periods.
438:
439: %\clearpage
440: \begin{figure}[ht!]
441: \epsscale{1.0}
442: \plotone{f4.eps}
443: \caption{Projected rotation velocities \citep{nds01} plotted
444: against the measured rotation period for 16 stars in M35. All
445: stars have $P_{RV} \ge 60\%$ and none of the 16 stars are spectroscopic
446: binaries. For comparison, the solid, dashed, and dotted curves
447: indicate the relation between rotation period and the projected
448: rotational velocity for a solar-like star with a $90\degr$,
449: $70\degr$, and $50\degr$ inclination of the rotational axis,
450: respectively. The rotation periods and the projected rotation
451: velocities are consistent for all 16 stars.
452: \label{vsini}}
453: \end{figure}
454: %\clearpage
455:
456: % -----------------------------------------------------------------------------
457:
458: \subsection{The Spectroscopic Survey} \label{spec}
459:
460: M35 has been included in the WIYN Open Cluster Study
461: (WOCS; \citet{mathieu00}) since 1997. As part of WOCS
462: more than 6000 spectra has been obtained of approximately
463: 1500 solar-type stars within a 1-degree field centered on
464: M35. The selection of survey target stars was based on
465: photometric (\citet{deliyannis08}, see Section~\ref{phot})
466: and proper-motion \citep{ms86,cudworth71} membership data.
467: Stars on or less than $\sim 1\fm0$ above the cluster main sequence
468: were selected, with brightness and color ranges corresponding to
469: a range in mass from $\sim0.7~M_{\odot}$ to $\sim1.4~M_{\odot}$.
470: All spectroscopic data were obtained using the WIYN 3.5m
471: telescope equipped with a multi-object fiber optic positioner
472: (Hydra) feeding a bench mounted spectrograph. Observations were
473: done at central wavelengths of 5130\AA\ or 6385\AA\ with a wavelength
474: range of $\sim$200\AA\, providing many narrow absorption
475: lines. Radial velocities with a precision of $\la 0.5~km~s^{-1}$
476: \citep{gmh+08,mbd+01} were derived from the spectra via
477: cross-correlation with a high $S/N$ sky spectrum. From this
478: extensive radial-velocity survey we have 1) calculated the cluster
479: membership probability; 2) detected the cluster binary stars; and 3)
480: determined the orbital parameters for the closest binaries.
481:
482: Of the 441 stars with rotation periods presented in this
483: study, 259 have one or more radial-velocity measurements
484: (the remainder being below the faint limit of the spectroscopic
485: survey or photometric non-members). The radial-velocity
486: cluster membership probability of each star is calculated
487: using the formalism by \citet{vkp58}. The mean or center-of-mass
488: radial velocity of a single or binary star was used when calculating
489: the membership probability. We have adopted 50\% as the threshold
490: for assigning radial-velocity and proper-motion cluster membership.
491: Of the 259 rotators with one or more radial-velocity measurement,
492: 203 are radial-velocity members of M35 and 20 of those 203 stars
493: are also proper-motion members. More detailed descriptions of the
494: radial-velocity survey and membership determination can be found
495: in \citet{mm05}, \citet{mms06}, and \citet{bmm08}.
496:
497: % -----------------------------------------------------------------------------
498:
499: \subsection{The M35 Color-Magnitude Diagram and Photometric Membership}
500: \label{phot}
501:
502: Figure~\ref{cmd} shows the (V-I) vs. V color-magnitude diagram (CMD)
503: for M35. The photometry was kindly provided by \citet{deliyannis08}
504: who obtained UBVRI data in a $23\arcmin \times 23\arcmin$ central
505: field and BVRI data in a $2 \times 2$ mosaic for a total of $\sim40\arcmin
506: \times 40\arcmin$ using the WIYN 0.9m telescope. In the CMD the 441
507: stars for which we have measured rotation periods are highlighted in
508: black. The solid lines enclosing stars within or above the cluster
509: sequence (allowing for inclusion of equal-mass binaries) show our criteria
510: for photometric membership. The insert in Figure~\ref{cmd} shows the
511: location in the M35 CMD of only radial-velocity members (open symbols),
512: and radial-velocity and proper-motion members (filled symbols),
513: the location of which was used to define the criteria for photometric
514: membership. There are 23 photometric members with 3 or more radial-velocity
515: measurements and a radial-velocity membership probability of less
516: than 50\%. Those 23 stars were removed from the list of cluster
517: members. The final number of stars with measured rotation periods
518: selected as radial-velocity and/or photometric members of M35 is 310.
519:
520: \placefigure{cmd}
521:
522: %\clearpage
523: \begin{figure}[ht!]
524: \epsscale{1.0}
525: \plotone{f5.eps}
526: \caption{The M35 $(V-I)_0$ vs. $V$ color-magnitude diagram.
527: Photometry was provided by \citet{deliyannis08}. The 441 stars
528: with rotation periods are highlighted in black. Stars located between
529: the solid lines are considered photometric members of M35. Note that
530: the faint limits for proper-motion and radial-velocity surveys are
531: $V \simeq 14.5$ and $V \simeq 17.5$, respectively. The insert shows
532: the location of stars that are radial-velocity members (open symbols),
533: and radial-velocity and proper-motion members (filled symbols). These
534: kinematic members of M35 were used to define the boundaries for photometric
535: membership. The isochrone shown represents a 150 Myr Yale model
536: \citep{ykd03}.
537: \label{cmd}}
538: \end{figure}
539: %\clearpage
540:
541: % *****************************************************************************
542:
543:
544: \section{THE ROTATION-PERIOD DISTRIBUTION} \label{distr}
545:
546: The 310 members of M35 with determined rotation periods correspond to
547: $\sim$12\% of the photometric cluster population within the brightness
548: and areal limits of our photometric survey. Figure~\ref{phist}a shows
549: the distribution of rotation periods, which spans more than 2 orders of
550: magnitude from $\sim$0.1 days to $\sim$15 days. The distribution peaks
551: shortward of 1 day and has a broader and shallower peak centered at about
552: 6 days.
553:
554:
555: \placefigure{phist}
556:
557: %\clearpage
558: \begin{figure}[ht!]
559: \epsscale{1.0}
560: \plottwo{f6a.eps}{f6b.eps}
561: \caption{{\bf a)} The distribution of rotation periods for the sample
562: of 310 cluster members with masses from $\sim0.6-1.4~M_{\odot}$ (spectral
563: type late K to mid F). The distribution show a large dispersion from
564: $\sim$0.1 days to $\sim$15 days, and a distinct peak at $\le 1$ day
565: and a shallower and broader peak centered at $\sim$6 days.
566: {\bf b)} The distribution of rotation periods shortward of 1 day binned
567: in 0.1 day bins. The dashed line histogram represents member as well as
568: non-member stars with measured rotation periods in our sample. The
569: grey histogram represents only members of M35.
570: \label{phist}}
571: \end{figure}
572: %\clearpage
573:
574: Figure~\ref{phist}b displays with an increased resolution of 0.1 day
575: the distribution of rotation periods shortward of 1 day. The dashed
576: and grey histograms, respectively, represent all stars and all cluster
577: members with detected rotation periods. The distribution
578: shows that we are capable of detecting rotation periods down to the
579: pseudo-Nyquist period-limit of about 2 hours ($\sim 0.08$ day) resulting
580: from our typical sampling cadence of about $1~hr^{-1}$ in December 2002.
581: The distribution of rotation periods for cluster members falls off shortward
582: of 0.3-0.4 days. Two member stars have rotation periods between 0.1 days and
583: 0.2 days, corresponding to surface rotational velocities of 50\% or more
584: of their breakup velocities ($v_{br} = \sqrt{(GM_{\star}/R_{\star})}$).
585: We argue based on this inspection of the short-period tail of the
586: distribution that the lower limit of 0.1 days for our period search was
587: set appropriately for the stars in M35.
588:
589: The long-period ends of the period distributions for members and
590: non-members (Figure~\ref{nmd}, Appendix~\ref{nm}) show that the
591: long time-span of the queue-scheduled data enable us to detect rotation
592: periods beyond the $\sim$10 days typically found to be the upper limit
593: in photometric surveys with durations similar to our short-term December
594: 2002 observing run. We report the detection of 18 stars with rotation
595: periods longer than 10 days, 7 of which are members of M35. The longest
596: rotation period among members is 15.3 days, and among the field stars
597: rotation periods of up to $\sim$ 17 and 23 days have been measured.
598: In the M35 period distribution
599: we see a drop-off at $\sim$10 days. If the $\sim$12\% of the cluster's
600: late-type population with measured rotation periods is a representative
601: sample of the late-type stellar population in M35, then the $\sim$10 day
602: cutoff may represent a physical upper limit on the rotation-period
603: distribution at 150 Myr. However, it is also possible that we are not
604: capable of detecting the slowest rotators despite our long-baseline
605: photometric survey. Indeed, the modest number of rotation periods longer
606: than 10 days found in the much larger sample of field stars may reflect
607: that the frequency and size of spots on stars rotating slower than
608: $\sim$10 days is insufficient for detection with the photometric precision
609: of our data. Indeed, X-ray observations in Orion \citep{sab+04} indicate
610: that rotation-period studies of young stars may in general be biased
611: against very slow rotators because such stars likely do not generate
612: strong activity, and thus are not sufficiently spotted to allow detection of
613: photometric periods."
614: Measuring the rotation for such slowly rotating stars will likely
615: require either higher photometric precision or high resolution ($R \ga
616: 50,000$) spectroscopic observations to measure projected rotation velocities.
617: The small sample of M35 member stars with periods above 10 days will
618: be discussed in Section~\ref{prediction}.
619:
620: % -----------------------------------------------------------------------------
621:
622: \subsection{Long-term stability of the number, sizes, and configurations
623: of stellar spots and spot-groups}
624:
625: We find that for almost all cluster and field stars with measured rotation
626: periods, the long-term queue-scheduled data, spanning $\sim$5 months in time,
627: phase up with and coincide with the short-term data (16 nights in December
628: 2002) in the light curves. We tested further the agreement between the
629: short-term and long-term photometric variability, by measuring the rotation
630: period separately from the short- and long-term data for 20 randomly chosen
631: stars. For all but one star we found a difference between the two rotation
632: periods that was less than 1\% of the period measured from the short-term
633: data alone (in most cases the difference was less than 0.1\%). We examined
634: the 20 light curves with all data phased to the period derived from only
635: the short-term data. For all but one star, the long-term data produced light
636: curves of the same shape and phase as the light curves based only on the
637: short-term data. Even for two light curves with clear signs of multiple
638: spots (spot-groups), the short- and long-term data coincided very well.
639: Because it is unlikely that the disappearance and recurrences of
640: spots will result in a light curve with the same shape, amplitude, and
641: phase, the agreement suggests stability of individual spots and/or
642: spot-groups over the $\sim$5 month time-span of our photometric observations.
643:
644: We find that the stability of the sizes and configurations of spots on
645: young stars have recently been studied for e.g. the solar analog PMS star
646: V410 Tau \citep{sfc+03} and for stars in the PMS cluster IC348
647: \citep{nhr+06}. For V410 Tau data has been collected for over two decades,
648: showing changes in the shape of the light curve over the last decade.
649: The authors suggest that the observed changes reflect variations of the
650: structure of the active regions over timescales of years. However, stability
651: in the rotation period and the recurrence of the light curve minimum,
652: suggest stability of the largest spots over years, and either a lack
653: of latitudinal differential rotation in V410 Tau, or confinement of its
654: spots to a narrow range of latitudes. Similarly, \citet{nhr+06} finds
655: a remarkable stability over 7 years in the rotations periods for stars
656: in IC348, suggesting again that these PMS stars do not have significant
657: differential rotation, or that their spots are constrained to a narrow
658: range of stellar latitudes. However, contrary to what is observed over
659: 5 months in M35, all periodic stars in IC348, as well as V410 Tau, do
660: show changes in the light curve shape and amplitude from year to year.
661:
662:
663: % *****************************************************************************
664:
665:
666: \section{THE M35 COLOR-PERIOD DIAGRAM - THE DEPENDENCE OF STELLAR
667: ROTATION ON MASS} \label{cpd}
668:
669: In Figure~\ref{pbv} we display the rotation periods for the 310 members
670: plotted against their $B-V$ color indices, or equivalently their
671: masses. The color indices derive from the deep multi-band photometry
672: by \citet[][Section~\ref{phot}]{deliyannis08} and the corresponding
673: stellar mass estimates (upper x-axis) from a fit of a 150\,Myr Yale
674: stellar evolutionary model \citep{ykd03} to the M35 color-magnitude diagram.
675: Dark blue symbols represent stars that are both photometric and
676: radial-velocity members of M35. Light blue symbols are used for stars
677: that are photometric members only. Proper-motion members are marked
678: with additional squares.
679:
680: \placefigure{pbv}
681:
682: %\clearpage
683: \begin{figure}[ht!]
684: \epsscale{1.0}
685: \plotone{f7.eps}
686: \caption{The distribution of stellar rotation periods with (B-V) color
687: index for 310 members of M35. Dark blue symbols represent stars that are
688: both photometric and radial-velocity members of M35. Light blue symbols
689: are used for stars that are photometric members only. Proper-motion members
690: are marked with additional squares. The upper x-axis gives a stellar mass
691: estimate corresponding to the color on the lower axis. Masses are derived
692: using a 150 Myr Yale isochrone.
693: \label{pbv}}
694: \end{figure}
695: %\clearpage
696:
697: The M35 color-period diagram shows striking structure. The coeval stars
698: fall along two well-defined sequences apparently representing two different
699: rotational states. One sequence displays clear dependence between
700: period and color, starting at the blue end at $(B-V)_0 \simeq 0.5$
701: ($M_{\star} \simeq 1.2~M_{\odot}$) and $P_{rot} \simeq 2$ days and forming
702: a rich diagonal band of stars whose periods are increasing with increasing
703: color index (decreasing mass). This sequence terminates at about $(B-V)_0
704: \simeq 1.2$ ($M_{\star} \simeq 0.65~M_{\odot}$) and $P_{rot} \simeq 10$ days.
705: The second sequence consists of rapidly rotating ($P_{rot} \la 1$ day) stars
706: and extends from $(B-V)_0 \simeq 0.7-0.8$ ($M_{\star} \simeq
707: 0.9-1.0~M_{\odot}$) to $(B-V)_0 \simeq 1.5$ ($M_{\star} \la 0.5~M_{\odot}$).
708: This well defined sequence of rapidly rotating stars shows a small but
709: steady decrease in rotation period with increasing color (decreasing mass).
710: Finally, a subset of stars are distributed in between the two sequences,
711: and 10 stars have rotation periods that are unusually long, placing them
712: above the diagonal sequence in Figure~\ref{pbv}.
713:
714: The M35 color-period diagram gives a clear picture of preferred
715: stellar rotation periods as a function of color for 150\,Myr late-type
716: dwarfs. With the added dimension of color, the diagram
717: take us beyond the one-dimensional period distribution and shows
718: which stars are responsible for the structure observed in
719: Figure~\ref{phist}. The short-period peak at $\la$1 day is due primarily
720: to the rapidly rotating late G- and K-dwarfs ($M \la 0.9\,M_{\odot}$),
721: and the sharpness of this peak is the result of little dependence
722: of rotation on color within this group. The more slowly rotating
723: mid to late G- and K-dwarfs give rise to the broader peak at $\sim$6
724: days, while the early to mid G-dwarfs and some cooler stars fill in
725: the distribution between the peaks.
726:
727: The two sequences of stars in the color-period diagram
728: represent the most likely/stable rotation period(s) for a given stellar
729: mass at the age of M35. Under the assumption that rotation periods
730: increase with time for all stars, the more sparsely populated area
731: between the two sequences must then represent a phase of rotational
732: evolution of shorter duration. We will discuss the different loci
733: in the color-period diagram in more detail in Section~\ref{discuss}.
734:
735:
736: % *****************************************************************************
737:
738:
739: \section{ANGULAR MOMENTUM EVOLUTION AND THE COLOR-PERIOD DIAGRAM}
740: \label{discuss}
741:
742: Sequences similar to those observed in the M35 color-period diagram
743: were noted by \citet[][hereinafter B03]{barnes03a} from careful examination
744: of compilations of rotation-period data from photometric monitoring
745: campaigns on open clusters and field stars. B03 named the diagonal
746: sequence of stars on which rotation periods increase with color the
747: {\it interface} sequence (or I sequence), and the sequence of rapidly
748: rotating stars the {\it convective} sequence (or C sequence). In what
749: follows we will adopt these names for the two sequences in the M35
750: color-period diagram.
751:
752: B03 argues that the rapidly rotating stars on the C sequence have
753: radiative cores and convective envelopes that are decoupled. For
754: these stars he suggests that the evolution of their surface rotation
755: rates is governed primarily by the moments of inertia of the convective
756: envelope and by inefficient wind-driven loss of angular momentum linked
757: to small-scale convective magnetic fields. For stars on the I sequence,
758: large-scale (sun-like) magnetic fields provided by an interface dynamo
759: couple the core and envelope, and the rotational evolution of the I
760: sequence stars is thus primarily governed by the moments of inertia
761: of the entire star and more efficient angular momentum loss (i.e., a
762: \citet{skumanich72} style spin-down).
763: %
764: Accordingly, B03 suggests that a late-type star,
765: in which the core and envelope are decoupled as it settles on the ZAMS,
766: will begin its main-sequence life on the C sequence and evolve onto
767: the I sequence when rotational shear between the stellar core and
768: envelope establish a large-scale dynamo field that couples the two zones
769: and provide efficient magnetic wind loss. Higher mass stars have thinner
770: convective envelopes with smaller moments of inertia than low mass stars
771: and thus leave the C sequence sooner. Stars that are either fully radiative
772: or fully convective will remain as rapid rotators.
773:
774: Color-period diagrams for coeval populations of different ages
775: allow us to examine the rotational properties of late-type coeval stars
776: as a function of their mass and age, and may bring us closer to understanding
777: the physical mechanisms (internal and/or external) regulating their
778: rotational evolution. The M35 color-period diagram, rich in stars and
779: cleaned for spectroscopic and photometric non-members, reveals the
780: morphology described by B03 more clearly than any published stellar
781: populations. We therefore begin with a discussion of the M35
782: result in the context of the framework developed by B03.
783:
784: % -----------------------------------------------------------------------------
785:
786: \subsection{Timescales for migration from the C to the I sequence}
787: \label{timescales}
788:
789: In Figure~\ref{fractions} we add M35 to Fig. 3 in B03 which shows the
790: relative fractions of stars with $0.5 \le (B-V)_0 \le 1.5$ on the I and
791: C sequences for stellar populations of distinct ages. With rotation periods
792: measured over more than two orders of magnitude for confirmed
793: spectroscopic and photometric cluster members, M35 adds the statistically
794: most secure datapoints to this figure. M35 fit well with the evolutionary
795: trends of increasing relative fractions of C sequence (and gap) stars
796: and decreasing fractions of I sequence stars for younger cluster populations.
797: The almost linear trends in Figure~\ref{fractions} suggest that the
798: decrease and increase in the number of stars on the C and I sequences,
799: respectively, are approximately exponential with time. Under the presumption
800: of an exponential time dependence and that all stars start on the C sequence
801: at the ZAMS,
802: we can estimate the characteristic timescale for the rotational evolution
803: of stars off the C sequence and onto the I sequence, by counting stars on
804: both sequences and in the gap in the M35 color-period diagram. Such
805: timescales may offer valuable constraints on the rates of internal and
806: external angular momentum transport and on the evolution rates of stellar
807: dynamos in late-type stars of different masses.
808:
809: \placefigure{fractions}
810:
811: %\clearpage
812: \begin{figure}[ht!]
813: \epsscale{1.0}
814: \plotone{f8.eps}
815: \caption{Figure 3 from B03 with M35 added. The figure shows
816: the fractions of stars with $0.5 \le (B-V)_0 \le 1.5$ on the I sequence
817: (circles) and the C sequence (stars) for clusters of different ages.
818: The squares represent the relative fraction of the sum of C sequence
819: and gap stars. The filled symbols show the relative fractions for M35.
820: We follow B03 in estimating the uncertainties in the
821: fractions by the square root of the number of stars.
822: \label{fractions}}
823: \end{figure}
824: %\clearpage
825:
826: When counting the number of stars on the I and C sequences and in the gap,
827: we use the following criteria. Stars located in the color-period diagram
828: between the lines represented by $P_{rot} = 10 (B-V)_0 - 2.5 \pm 2.0$ and
829: with periods above 1.5 days were counted as I sequence stars (see dotted
830: lines in Figure~\ref{pbv_iso2}). Stars redder than $(B-V)_0 = 0.6$ and with
831: periods between 0 and 1.5 days were counted as C sequence stars. Stars
832: located below $P_{rot} = 10 (B-V)_0 - 4.5$ and with periods above 1.5 days
833: were counted as gap-stars. These selection criteria are subjective and
834: although the sequences are well defined, the I sequence becomes broader
835: redward of $(B-V)_0 \simeq 1.0$ making the distinction between I sequence
836: and gap stars more difficult. However, due to the large number of rotation
837: periods in M35, the small number of stars that might be moved from the gap
838: to the I sequence or vice versa by using slightly different criteria will
839: not influence the relative fractions and thus the timescales in any
840: significant way.
841:
842: The number of C sequence stars ($N_c$) at a time $t$ can then be expressed
843: by:
844:
845: \begin{equation}
846: N_c = N_{c_{0}} e^{-t/\tau_c}
847: \end{equation}
848:
849: \noindent where $N_{c_{0}}$ and $\tau_c$ are, respectively, the total number
850: of stars on the I and C sequences and in the gap, and the characteristic
851: exponential timescale. We use the B-V color index to divide the stars into
852: G-dwarfs ($0.6 < B-V < 0.8$) and K-dwarfs ($0.8 < B-V < 1.3$) We count all
853: stars within each color-interval in the color-period diagram as $N_{c_{0}}$,
854: all C sequence stars within each color-interval as $N_c$, and adopt
855: 150\,Myr as the age of M35. We derive from equation [1] $\tau_c^G = 60$ Myr
856: and $\tau_c^K = 140$ Myr as the characteristic exponential timescales for
857: transition between the C and the I sequence for G and K dwarfs, respectively.
858:
859: We can qualitatively verify these time scales by a comparison between the
860: M35 color-period diagram and those of the younger Pleiades cluster and
861: the older cluster NGC3532 presented in B03. In M35 only 7 G dwarfs ($0.6
862: \la (B-V)_0 \la 0.8$) are found on the C sequence and in the gap, while the
863: G dwarf I sequence is well defined and rich. In contrast, In contrast, the
864: M35 C sequence and gap are rich in K dwarfs ($0.8 \la (B-V)_0 \la 1.3$),
865: whereas the K dwarf I sequence is less densely populated and less well defined.
866: The lack of G dwarfs on the C sequence seen in M35 is already apparent at
867: 100 Myr in the Pleiades color-period diagram, indicating that the
868: characteristic timescale for G dwarfs to evolve off the C sequence
869: and onto the I sequence is less than $\sim$100-150 Myr. The rich population
870: of early and mid K dwarfs on the M35 C sequence have evolved off the C
871: sequence and onto a well defined I sequence by the age of NGC\,3532 (300 Myr).
872: The NGC\,3532 C sequence and gap, however, are populated by late K dwarfs,
873: suggesting that early to mid K dwarfs evolve onto the I sequence on a
874: timescale between 150 and 300 Myr, or approximately twice the time
875: required for G dwarfs. Finally, by the age of the Hyades only 3 late K or early
876: M dwarfs have been found off the I sequence, or in the gap (see Fig. 1
877: in B03 or Figure~\ref{m35_hyades_pbv} below), suggesting that such stars
878: evolve off the C sequence and possibly onto the I sequence on a timescale
879: of $\sim$600 Myr, or approximately twice the time required for the early
880: to mid K dwarfs.
881:
882: There is thus good agreement between the exponential timescales derived
883: from the M35 color-period diagram alone and the estimated timescales based
884: on a comparison of color-period diagrams of different ages. We note that
885: although the relative fractions displayed in Figure~\ref{fractions} represent
886: all rotators with $0.5 \le (B-V)_0 \le 1.5$, a least-squares fit of an
887: exponential function to the C sequence fractions in Figure~\ref{fractions},
888: gives a timescale of 106\,Myr for a decrease in the number of C sequence
889: stars by a factor of $e$.
890:
891: % -----------------------------------------------------------------------------
892:
893: \subsection{Testing the Skumanich $\sqrt{t}$ spin-down rate between
894: M35 and Hyades} \label{skumanich}
895:
896: The color-period diagram for the Hyades contains 25 stars
897: \citep{rtl+87,psd+95}, 22 of which form an I sequence of G and K dwarfs.
898: Despite the smaller number of stars, the blue part of the Hyades I sequence,
899: populated by G dwarfs, is well defined. By comparing the rotation periods
900: for the I sequence G dwarfs in M35 to those of the I sequence G dwarfs in
901: the $\sim$4 times older Hyades, we can directly test the \citet{skumanich72}
902: $\sqrt{t}$ time-dependence on the rotation-period evolution for stars in
903: this mass-range. We follow B03
904: in assuming separate mass and time dependencies for stellar rotation,
905: and that the same mass dependence can be applied to different stellar
906: populations. Adopting an age of 625\,Myr for the Hyades \citep{pbl+98}
907: and of 150\,Myr for M35, we decrease the Hyades rotation periods
908: by $\sqrt{625/150} \simeq 2$. We show in Fig.~\ref{m35_hyades_pbv}
909: the color-period diagram with the 310 M35 members (all grey symbols) and
910: with the locations of the 25 Hyades stars overplotted (black symbols).
911: The spun-up Hyades I sequence G dwarfs coincide nicely with the M35
912: I sequence G dwarfs, in support of the Skumanich $\sqrt{t}$ time-dependence
913: for such stars.
914: Curiously, the Hyades K-dwarfs, also spun-up according to the Skumanich
915: law, have rotation periods systematically shorter than the M35 K dwarfs.
916: At face value, this suggests that the time-dependence for spin-down of
917: K dwarfs is different and slower than for G dwarfs between 150\,Myr and
918: 625\,Myr.
919:
920: \placefigure{m35_hyades_pbv}
921:
922: %\clearpage
923: \begin{figure}[ht!]
924: \epsscale{1.0}
925: \plotone{f9.eps}
926: \caption{The M35 color-period diagram (grey symbols) with 25 Hyades stars
927: overplotted \citep[black;][]{rtl+87,psd+95}. All but the 3 reddest Hyades
928: stars fall on a sequence similar to the M35 I sequence. All Hyades rotation
929: periods were spun-up by a factor $\sqrt{625/150} \simeq 2$ in accordance
930: with the Skumanich $\sqrt{t}$ time-dependence on stellar rotation evolution,
931: assuming ages of 625\,Myr and 150\,Myr for the Hyades and M35, respectively.
932: \label{m35_hyades_pbv}}
933: \end{figure}
934: %\clearpage
935:
936: % -----------------------------------------------------------------------------
937:
938: %\subsection{The gyro-age of M35 - fitting rotational isochrones to
939: %the M35 color-period diagram} \label{isochrones}
940:
941: \subsection{The gyro-age of M35} \label{isochrones}
942:
943: Arguing that the rotation of stars on the C and I sequences follow
944: separate mass (M) and age (t) dependencies ($P(t,M) = g(t) \times f(M)$),
945: B03 introduced heuristic functional forms to represent these separate
946: dependencies of the I and C sequences. \citet[][hereinafter B07]{barnes07}
947: presents a modified functional form for the I sequence. These
948: functions define one-parameter families, with that parameter being the
949: age of the stellar population, and the resulting curves in the color-period
950: plane represent a set of rotational isochrones. We note that B03 use
951: $g(t) = \sqrt{t}$ (Skumanich), while B07 derive $g(t) = t^{0.52}$ by
952: requiring that a solar-like star spin down to solar rotation at solar age.
953:
954: \citet{kawaler89} used his own calibrated angular momentum loss law
955: \citep{kawaler88} and the assumption of solid body rotation after
956: $\sim$100\,Myr \citep{pks+89}, to derive a relationship between stellar
957: age, rotation period, and color. Kawaler's age-period-color relation
958: is thus based on models calibrated to the sun, and the assumption that the
959: Skumanich relationship is valid for all masses. He express the stellar
960: mass, radius, and moment of inertia, in terms of observables such as
961: stellar color, via stellar models.
962:
963: We note that our result in Section~\ref{skumanich}, suggest that the time
964: dependence of stellar rotation is not independent of stellar mass, as
965: assumed by both B03, B07, and \citet{kawaler89}.
966:
967: The well-defined sequences in the M35 color-period diagram make possible
968: a test of the period-color relations proposed by B03, B07, and
969: \citet{kawaler89}. We show in Figure~\ref{pbv_iso} the M35 color-period
970: diagram with the B03 and B07 rotational isochrones for the independently
971: determined stellar-evolution age for M35 of 150 Myr
972: \citep{vss+02,deliyannis08}. The rotational isochrones match the M35 I
973: and C sequences well, suggesting that they can indeed provide a consistent
974: age estimate (gyro-age) for a cluster based on a well populated color-period
975: diagram cleaned for non-members. To illustrate the sensitivity to age,
976: rotational isochrones for 130 Myr and 170 Myr are also displayed in
977: Figure~\ref{pbv_iso}.
978:
979: \placefigure{pbv_iso}
980:
981: %\clearpage
982: \begin{figure}[ht!]
983: \epsscale{1.0}
984: \plotone{f10.eps}
985: \caption{The M35 color-period diagram with the 150\,Myr rotational isochrones
986: from B03 (left) and B07 (right) overplotted as a thick black solid curves.
987: To illustrate the sensitivity to age we show the 130\,Myr and the 170\,Myr
988: isochrones as thinner curves flanking the 150\,Myr isochrones.
989: \label{pbv_iso}}
990: \end{figure}
991: %\clearpage
992:
993: Assuming no prior knowledge about the age of M35 (thus letting age be a
994: free parameter), we perform a non-linear least squares fit to the I
995: sequence stars (enclosed by the dotted lines in Figure~\ref{pbv_iso2})
996: of the functional form of the rotational I sequence isochrones from
997: B03:
998:
999: \begin{equation}
1000: P(t,(B-V)) = \sqrt{t} \times \bigl(\sqrt{((B-V)-a)} - b((B-V)-a\bigr)
1001: \end{equation}
1002:
1003: \noindent with $a = 0.50$, and $b = 0.15$, and from B07:
1004:
1005: \begin{equation}
1006: P(t,(B-V)) = t^{0.52} \times \bigl(c((B-V)-d)^{f}\bigr)
1007: \end{equation}
1008:
1009: \noindent with $c = 0.77$, $d = 0.40$, and $f = 0.60$.
1010:
1011: When fitting, the I sequence stars were weighted them according to their
1012: cluster membership, with most weight given to confirmed radial-velocity
1013: and proper motion members and least weight given to stars with only
1014: photometric membership. The weights given to individual stars are listed
1015: in Table~\ref{tab2} in Appendix B. Figure~\ref{pbv_iso2} shows the best
1016: fits of both eq. [2] and eq. [3] to the M35 I sequence. The derived age
1017: is 134\,Myr for both functional forms, each with a formal $1-\sigma$
1018: uncertainty of $\sim$3\,Myr. The close agreement of the two ages likely
1019: reflects the similar shape of the two isochrones over the color interval
1020: from $\sim0.5-1.0$ where the M35 I sequence is most densely populated.
1021: The small formal uncertainties reflect a well-defined I sequence rich
1022: in stars. However, the gyro-ages may still be affected by systematic
1023: errors in the stellar evolutionary ages for the young open clusters.
1024:
1025: \placefigure{pbv_iso2}
1026:
1027: %\clearpage
1028: \begin{figure}[ht!]
1029: \epsscale{1.0}
1030: \plotone{f11.eps}
1031: \caption{The least squares fit of the B03 I sequence isochrone
1032: (left; eq. [2]) and the B07 I sequence isochrone (right; eq. [3])
1033: to the M35 I sequence with age ($t$) as a free parameter. The gyro-ages
1034: corresponding to the fits are $133.9 \pm 3$Myr and $133.5 \pm 3$Myr,
1035: respectively. The I sequence stars to which the isochrones were fitted
1036: are enclosed by the dotted lines in both figures.
1037: \label{pbv_iso2}}
1038: \end{figure}
1039: %\clearpage
1040:
1041: Alternatively, we show in Figure~\ref{age-distr} the distribution
1042: of gyro-ages of the M35 I sequence stars, calculated using the
1043: age-period-color relations of B03 (left panel) and B07 (right panel).
1044: The mean gyro-ages of 137\,Myr and 161\,Myr, are both close to the
1045: 150\,Myr derived for the cluster using the isochrone method (see
1046: Section~\ref{intro}). Assuming that all I sequence stars are truely
1047: coeval, the standard errors of 3.8\,Myr and 4.1\,Myr give uncertainties
1048: on the calculated mean gyro-ages for M35 of 2.8\% and 2.5\%, close to
1049: the formal uncertainty on the least squares fits. The close agreement
1050: between the mean gyro-age (137\,Myr) and the gyro-age determined from
1051: the least square fit (134\,Myr), as well as a smaller standard deviation
1052: in the gyro-age distribution, suggest that the B03 I sequence isochrone
1053: (eq. [2]) provide a better match to the color dependence of stellar
1054: rotation on the I sequence than the B07 isochrone (eq. [3]).
1055:
1056: \placefigure{age-distr}
1057:
1058: %\clearpage
1059: \begin{figure}[ht!]
1060: \epsscale{1.0}
1061: \plottwo{f12a.eps}{f12b.eps}
1062: \caption{The distribution of gyro-ages for M35 I sequence stars.
1063: The panels show the distributions of M35 gyro-ages calculated using
1064: the B03 (left panel) and B07 (right panel) age-rotation-color
1065: relations. The distribution mean, standard deviation, and standard
1066: error on the mean, are given in the upper right corner of each panel.
1067: \label{age-distr}}
1068: \end{figure}
1069: %\clearpage
1070:
1071:
1072: In Figure~\ref{age-dist_kawalerr} we show the corresponding distribution
1073: of gyro-ages calculated using the \citet{kawaler89} age-period-color
1074: relation. The mean age is equal to that derived for the B07 relation,
1075: while the larger $\sigma$ and standard error (3.3\%) reflects primarily
1076: a poorer fit to the M35 I sequence for the late F and early G type
1077: stars, resulting in gyro-ages that are too low for those stars.
1078:
1079: \placefigure{age-distr_kawaler}
1080:
1081: %\clearpage
1082: \begin{figure}[ht!]
1083: \epsscale{1.0}
1084: \plotone{f13.eps}
1085: \caption{The distribution of gyro-ages for M35 I sequence stars
1086: calculated using the \citet{kawaler89} age-rotation-color relation.
1087: The distribution mean, standard deviation, and standard error on the
1088: mean, are given in the upper right corner of each panel.
1089: \label{age-dist_kawalerr}}
1090: \end{figure}
1091: %\clearpage
1092:
1093:
1094: % -----------------------------------------------------------------------------
1095:
1096: \subsection{Improving the I sequence mass-rotation relation using M35}
1097: \label{mass-rotation}
1098:
1099: The method of gyro-chronology relies on fitting the I sequence
1100: rotational isochrone, with age as a free parameter, to populations
1101: of cluster stars or to individual field stars in the color-period plane
1102: B07. The functional dependence between stellar color
1103: and rotation period of the isochrone will thus directly affect the
1104: derived gyro-age, and will, if not accurately determined, introduce a
1105: systematic error. It is therefore important to constrain and test
1106: the mass-rotation relation for stars on the I sequence as new data
1107: of sufficiently high quality becomes available.
1108:
1109: Our data for M35 are well suited for such a test because
1110: of the rich and well-defined I sequence, the extensive knowledge about
1111: cluster membership, and the independent stellar evolution age for the
1112: cluster. To constrain the color-period relation, we fit equations [2]
1113: and [3] to the M35 I sequence, using the same selection of stars and
1114: the same fitting weights as described in Section~\ref{isochrones}. We
1115: determine all coefficients in equations [2] and [3] for the fixed cluster
1116: age ($t$) of 150\,Myr. The coefficients with $1\sigma$ uncertainties
1117: are listed in Table~\ref{coeffs} and the corresponding rotational
1118: isochrones are shown in Figure~\ref{newiso}. To illustrate how closely
1119: the isochrones trace the selected I sequence stars, Figure~\ref{newiso}
1120: also displays a dashed curve representing the moving average of the rotation
1121: periods along the I sequence.
1122:
1123: \placefigure{newiso}
1124:
1125: %\clearpage
1126: \begin{figure}[ht!]
1127: \epsscale{1.0}
1128: \plotone{f14.eps}
1129: \caption{The least squares fits (solid curves) of 150\,Myr B03 (left)
1130: and B07 (right) I sequence isochrones (eq. [2] and eq. [3]) to the M35
1131: I sequence. The corresponding new values for the isochrone coefficients
1132: $a$, $b$, $c$, $d$, and $f$ were determined from the fits and are listed
1133: in Table~\ref{coeffs}. The moving average of the rotation periods for the
1134: I sequence stars is also shown as a dashed curve for comparison.
1135: \label{newiso}}
1136: \end{figure}
1137: %\clearpage
1138:
1139: We can compare the coefficients derived using the M35 data to those
1140: chosen and/or derived by B03 and B07. Our best fit of the B03 isochrone
1141: confirms the value of 0.5 for the $a$ coefficient chosen (not fitted)
1142: by B03 to best represent the color-period data included in his study.
1143: $a$ is a translational term that determines the color for which the
1144: isochrone gives a period of zero days. For the $b$ coefficient our best
1145: fit give a value of 0.20 compared to the choice of 0.15 by B03. Our
1146: larger value of the $b$ coefficient results in an isochrone with slightly
1147: more curvature.
1148:
1149: From our best fit of the B07 rotational isochrone to the M35 I sequence,
1150: we determined a $c$ coefficient of 0.77, equal to the value used by
1151: B07, while our value 0.55 for the $f$ coefficient is smaller than the
1152: value of 0.60 used by B07. In the case of the $c$ and $f$ coefficients,
1153: B07 also determined their values from least squares fitting to the I
1154: sequence stars of several young open clusters. However, for the
1155: translational term $d$, he chose a fixed value of 0.4 to allow for
1156: more blue stars to be fitted. We left the $d$ coefficient as a free
1157: parameter when fitting to the M35 I sequence, and got a value of 0.47.
1158:
1159: The new value of 0.47 for $d$ is particularly interesting as it
1160: corresponds to the approximate $B-V$ color for F-type stars at the
1161: transition from a radiative to a convective envelope. This transition
1162: was noted from observations of stellar rotation (known as the break
1163: in the Kraft curve \citep{kraft67}), and is associated with the onset
1164: of effective magnetic wind breaking \citep[e.g.][]{schatzman62}.
1165: The value of 0.47 for the $d$ coefficient therefore suggest that,
1166: for M35, the blue (high-mass) end of the I sequence begins at the
1167: break in the Kraft curve.
1168:
1169: %\clearpage
1170: \begin{deluxetable}{cccc}
1171: \tabletypesize{\normalsize}
1172: \tablecaption{New coefficients for the I sequence rotational isochrones
1173: \label{coeffs}}
1174: \tablewidth{0pt}
1175: \tablehead{
1176: \colhead{Isochrone} & \colhead{Coefficient} & \colhead{Value} & \colhead{1$\sigma$ error}\\
1177: %\colhead{} & \colhead{} & \colhead{} & \colhead{}
1178: }
1179: \startdata
1180: B03 & a & 0.507 & 0.005 \\
1181: B03 & b & 0.204 & 0.013 \\
1182: B07 & c & 0.770 & 0.014 \\
1183: B07 & d & 0.472 & 0.027 \\
1184: B07 & f & 0.553 & 0.052
1185: \enddata
1186: \end{deluxetable}
1187: %\clearpage
1188:
1189: % -----------------------------------------------------------------------------
1190:
1191: \subsection{Prediction: Tidal Evolution is Responsible for the Unusually
1192: Slow Rotators} \label{prediction}
1193:
1194: Ten stars fall above the M35 I sequence, and thus rotate unusually
1195: slowly in comparison to other members of M35 with similar masses.
1196: All 10 stars are photometric members of M35 and 2 are also spectroscopic
1197: members. We have no reason to believe that the rotation periods for
1198: these stars are due to aliases in the power spectra, and we note that
1199: a similar pattern is seen in NGC\,3532 with 7 stars located above
1200: the I sequence B03 and in M34 with 6 stars above the I sequence
1201: \citep{mms08b}.
1202:
1203: What causes the rotational evolution of these stars to deviate
1204: significantly from that of most similar-color stars in M35? We propose
1205: here that tidal interactions with a close stellar companion has
1206: acted to partially or fully synchronize the stellar spins of these
1207: stars to the orbital motions, and that such tidal synchronization
1208: is responsible for their slower-than-expected rotation. We thus
1209: predict that these 10 stars are the primary stars in binaries with
1210: periods of $\sim$10-15 days.
1211:
1212: This proposition finds support from the star of M35 located in the
1213: color-period diagram at $(B-V)_0 = 0.68$ and $P_{rot} = 10.13$ days.
1214: This star is the primary in a circular binary with an orbital period
1215: of 10.33 days. The rotation of this star has been synchronized to the
1216: orbital motion of the companion \citep{mms06}, forcing it to rotate
1217: more slowly than stars of similar mass.
1218: In addition to the spectroscopic binary, 3 of the remaining
1219: 9 slow rotators are photometric binaries. Spectroscopic observations
1220: has begun of those stars and of the remaining 5 stars as of fall
1221: 2007 to determine their status as binary or single stars, and in
1222: the case of binarity, the degree of tidal evolution.
1223:
1224: % -----------------------------------------------------------------------------
1225:
1226: \subsection{Stellar angular momentum evolution near the ZAMS}
1227: \label{hypothesis}
1228:
1229: The trend in Figure~\ref{fractions} of an increasing fraction of C
1230: sequence (and gap) stars for younger cluster populations leads
1231: naturally to the suggestion that most, if not all, late-type stars
1232: pass through a phase of rapid rotation (the C sequence) at the ZAMS.
1233: We note that even should this be the case, an {\it observed} C sequence
1234: fraction of 1 is not expected for even the youngest coeval populations,
1235: as stars of different masses will reach the ZAMS, and thus hypothetically
1236: the C sequence, at different times. For example, late-F stars will be
1237: the first to arrive at/on the ZAMS and C sequence, and leave it before
1238: the arrival of G and K type stars.
1239:
1240: The color-period diagrams for the youngest stellar populations
1241: presented in Figure~\ref{fractions} (see also Figure 1 in B03)
1242: show that most stars lay either on the C sequence or in the gap.
1243: B03 finds only 25\% of the stars at 30 Myr to be on the I sequence.
1244: In fact the I sequence is not clearly identifiable at this age,
1245: and the stars identified as being on the I sequence are early-type
1246: rapid rotators near the intersection of the two sequences.
1247:
1248: By 30 Myr very few of the cluster members have rotation periods longer
1249: than 5 days. This is in marked contrast to fractions of $\sim$60\% and
1250: $\sim$40\%, respectively, for such slow rotators in the PMS populations
1251: of the Orion Nebula cluster (ONC) and NGC\,2264 \citep[see][]{hem+07}.
1252: The difference in the numbers of slowly rotating stars pre- and post-ZAMS,
1253: suggests that most, if not all, of the stars rotating slowly at $\sim$1-3\,Myr,
1254: spin up as they evolve onto the ZAMS. Such spin-up may have been observed.
1255: Comparison of the rotation period distributions for stars in the $\sim$1\,Myr
1256: ONC and the $\sim$2-3\,Myr NGC\,2264 \citep[][and references therein]{hem+07}
1257: shows a spin-up with time by a factor $\sim$2, presumably due to conservation
1258: of angular momentum as the stars contract on the PMS. On the other hand,
1259: the distribution of a smaller sample of rotation periods in the $\sim$2-3\,Myr
1260: IC\,348 \citep{nhr+06} does not show similar evidence for spin-up when
1261: compared to the ONC.
1262:
1263: From the point of view of modeling stellar angular momentum evolution,
1264: we emphasize the narrowness of the C sequence, with all rotation periods
1265: between 0.5 days and 1.5 days. We suggest that the broad distribution of
1266: rotation period among solar-like stars in the PMS populations must collapse
1267: into a narrow C sequence of similar rotation periods independent of mass.
1268: Indeed, we suggest that in the two 30\,Myr clusters of B03 (Figure 1),
1269: the gap stars with $(B-V)_{0} \ga 0.9$ may in fact be evolving {\it toward}
1270: the C sequence, and point out that in the 50\,Myr clusters in B03, mostly C
1271: sequence stars are observed redward of $(B-V)_{0} \simeq 0.9$.
1272:
1273:
1274: % *****************************************************************************
1275:
1276: \section{SUMMARY AND CONCLUSIONS} \label{conclusions}
1277:
1278: We present the results of an extensive time-series photometric survey
1279: over $\sim$5 months of late-type members in the 150\,Myr open cluster
1280: M35 (NGC\,2168). We have obtained photometric light curves for 14022
1281: stars with $12 \la V \la 19.5$ over a $40\arcmin \times 40\arcmin$ field
1282: centered on M35. We have determined the rotation periods for 441 stars.
1283: Cluster membership and binarity for stars with rotation periods are
1284: determined from the results of a decade long spectroscopic survey in M35.
1285: Of the 441 rotators 310 stars are radial-velocity and/or photometric
1286: members of M35.
1287:
1288: With an age slightly older than the Pleiades but with a much larger
1289: population of late-type stars, M35 is particularly interesting for
1290: studying stellar rotational evolution during this active phase of
1291: angular momentum evolution between the ZAMS and the age of the Hyades.
1292: The rotation periods of the 310 late-type members span over two
1293: orders of magnitude from 0.1 day ($\ga50\%$ of their breakup velocities),
1294: up to $\sim$15 days. A drop-off in the period distribution is found
1295: at $\sim$10 days, well below the upper limit of our period search.
1296: The $\sim$10-day cutoff may represent a physical upper limit on the
1297: rotation-period distribution at 150 Myr. However, it is also possible
1298: that detecting more slowly rotating stars in M35 will require higher
1299: photometric precision or higher resolution spectroscopic observations.
1300:
1301: We find in the phased light curves for almost all stars with measured
1302: rotation periods that the long-baseline ($\sim$5 months), low-frequency
1303: (1/night) photometric measurements match the short-baseline
1304: (16 nights), high-frequency ($\sim$ 1/hours) measurements in both
1305: phase, shape, and amplitude. Further tests on a subset of stars
1306: show that the same rotation periods are derived from the short-
1307: and long-term data to within 1\%. This stability in the modulation
1308: of the stellar brightness suggest a similar stability in the
1309: configuration, size, and number of starspots.
1310:
1311: In the color-period plane, the 310 M35 rotators reveal striking
1312: dependencies between surface rotation period and stellar color
1313: (mass). More than 75\% of the stars lay along two distinct sequences
1314: in the color-period diagram, apparently representing two different
1315: states in their rotational evolution. Similar sequences
1316: were identified by \citet{barnes03a} for stars
1317: in other clusters. Comparison between M35 and these clusters of the
1318: locations of the sequences in color-period diagram, as well as the
1319: relative numbers of stars on each, support for the idea (proposed by
1320: \citet{barnes03a}) that stars evolve from one sequence (C) to the
1321: other sequence (I) at a rate that is inversely proportional to the
1322: stellar mass.
1323:
1324: We determine from the M35 color-period diagram that the characteristic
1325: exponential timescale for rotational evolution off the C sequence and
1326: onto the I sequence is $\sim$60 Myr and $\sim$140 Myr for G and K dwarfs,
1327: respectively. These timescales may offer valuable constraints on the
1328: rates of internal and external angular momentum transport and on the
1329: evolution rates of stellar dynamos in late-type stars of different masses.
1330:
1331: From the emerging trend (supported by M35) of an increasing relative
1332: fraction of rapidly rotating C sequence stars with decreasing population
1333: age, we propose the hypothesis that most, if not all, late-type stars
1334: pass through a phase of rapid rotation (C sequence) on the ZAMS.
1335: By conjecture, there may not be a need for a direct connection between
1336: slowly rotating stars observed in the early PMS and slowly rotating
1337: stars at $\sim$100\,Myr post the ZAMS. Such a connection has often
1338: been assumed and set as a constraint on models of stellar angular
1339: momentum evolution, motivating the introduction of mechanisms to
1340: prevent slowly rotating PMS stars from spinning up as they evolve
1341: onto the main-sequence.
1342:
1343: By comparison with measured rotation periods in the Hyades, we
1344: put to the test the empirical Skumanich $\sqrt{t}$ time-dependence
1345: on the stellar rotation period for G dwarfs. By reducing the Hyades
1346: rotation periods by a factor $\sqrt{Age_{Hyades}/Age_{M35}}$ we find
1347: that the $\sqrt{t}$ law accounts very well for the rotational evolution
1348: of G dwarfs between M35 and the Hyades, whereas among the K dwarfs
1349: the $\sqrt{t}$ time-dependence predicts a spin-down rate that is
1350: faster than observed between M35 and the Hyades.
1351:
1352: We find that the heuristic rotational isochrones proposed by
1353: \citet{barnes03a} and \citet{barnes07} match the location of M35
1354: I and C sequences using the independently determined stellar evolution
1355: isochrone age for M35. A non-linear least-squares fit of the rotational
1356: isochrones to the M35 I sequence sets the cluster's gyro-age to 134\,Myr
1357: with a formal $1\sigma$ uncertainty of 3 Myr. We use the age-period-color
1358: relations by \citet{barnes03a}, \citet{barnes07}, and \citet{kawaler89},
1359: to calculate the distributions of gyro-ages for the M35 I sequence
1360: stars. The mean gyro-ages have standard errors of order 3\% and agree
1361: well with the $\sim$150\,Myr age derived for the cluster using the
1362: isochrone method.
1363: These results suggest that a well-populated color-period diagram,
1364: cleaned for non-members, in combination with rotational isochrones,
1365: can provide a precise age estimate that is consistent with the age derived
1366: from isochrone fitting in the CMD. We also use the M35 I sequence to improve
1367: the coefficients for the color-dependence of the rotational isochrones.
1368:
1369: Finally, to explain the ten M35 stars rotating with rates that are
1370: unusually slow compared to similar stars in the cluster, we propose
1371: that tidal synchronization in binary stars with orbital periods of
1372: order 10-15 days is responsible. Two of the 10 stars have already been
1373: found to be primary stars in tidally evolved spectroscopic binaries,
1374: while 3 other stars are photometric binaries. Accordingly, we predict
1375: that the remaining stars are also primary stars in spectroscopic binaries
1376: with orbital periods of $\sim$10-15 days.
1377:
1378:
1379: % *****************************************************************************
1380:
1381:
1382: \acknowledgments
1383:
1384: We wish to thank the University of Wisconsin - Madison Astronomy
1385: Department and NOAO for the time granted on the WIYN 0.9m and
1386: 3.5m telescopes. We express our appreciation to the site managers
1387: and support staff at both telescopes for their exceptional and
1388: friendly support. We are thankful to all observers in the WIYN
1389: 0.9m consortium who provided us with high-quality data through
1390: the queue-scheduled observing program. This work has been supported
1391: by NSF grant AST-0406615 to the UW-Madison, a Ph.D fellowship from
1392: the Danish Research Academy to S.M., partial support to S.M. from
1393: the Kepler mission via NASA Cooperative Agreement NCC2-1390, and
1394: by the Cottrell Scholarship from the Research Corporation to K.G.S.
1395:
1396: \clearpage
1397:
1398:
1399: % *****************************************************************************
1400: % A P P E N D I X
1401: % *****************************************************************************
1402: %% Appendix material should be preceded with a single \appendix command.
1403: %% There should be a \section command for each appendix. Mark appendix
1404: %% subsections with the same markup you use in the main body of the paper.
1405:
1406: %% Each Appendix (indicated with \section) will be lettered A, B, C, etc.
1407: %% The equation counter will reset when it encounters the \appendix
1408: %% command and will number appendix equations (A1), (A2), etc.
1409:
1410: \appendix
1411:
1412: \section{PHASED LIGHT CURVES}
1413:
1414: This appendix presents the light curves for the stars in the field of M35
1415: for which we measured rotation periods. In the printed journal,
1416: Figure~\ref{f15} below show examples of our light curves and light
1417: curve plots. Phased light curves for all 441 stars can be found in the
1418: electronic edition of the Journal. The light curves have
1419: been divided into 3 groups according to the amplitude of the photometric
1420: variation. For each group the light curves are sorted by the rotation period
1421: and are presented with the same $\delta V$ range on the ordinate. The group
1422: of stars with the largest photometric variability are shown first.
1423:
1424: For each star we plot the data from the high-frequency survey (December 2002)
1425: as black symbols and data from the low-frequency survey (October 2002 through
1426: March 2003) as grey symbols. A running ID number corresponding to the ID
1427: number in Table 1 Appendix B is given in the upper left hand corner in each
1428: plot. The period to which the data are phased (the rotation period listed in
1429: Table 1 as $P_{rot}$) is given in the upper right corner. The 2-5 letter code
1430: in the lower right corner informs about the stars membership status. The
1431: codes have the following meaning: Photometric Member (PM; described in
1432: Section~\ref{phot}), Photometric Non-Member (PNM), Photometric and
1433: Spectroscopic Member (PSM), Photometric and Proper-motion Member (PPM),
1434: and Photometric Member but Spectroscopic Non-Member (PMSNM). For each star
1435: a horizontal grey line in each plot mark $\delta V = 0.0$ and a vertical
1436: grey line marks a phase of 1.0.
1437:
1438: \notetoeditor{The first panel (f15a.eps) of Figure 14 could be shrunk
1439: slightly and rotated 90 degrees to fit below the text above}
1440:
1441: \placefigure{f15}
1442:
1443: \clearpage
1444: \pagestyle{empty}
1445: \begin{figure}
1446: \epsscale{.92}
1447: \plotone{f15a.eps}
1448: \caption{Phased light curves for stars with measured rotation periods.
1449: %This figure is intended to show examples of our light curves and light
1450: %curve plots. Phased light curves for all 441 stars can be found in the
1451: %electronic edition of the Journal.
1452: \label{f15}}
1453: \end{figure}
1454: \clearpage
1455: {\plotone{f15b.eps}}\\
1456: \centerline{Fig. 14. --- Continued.}
1457: \clearpage
1458: {\plotone{f15c.eps}}\\
1459: \centerline{Fig. 14. --- Continued.}
1460: \clearpage
1461: {\plotone{f15d.eps}}\\
1462: \centerline{Fig. 14. --- Continued.}
1463: \clearpage
1464: {\plotone{f15e.eps}}\\
1465: \centerline{Fig. 14. --- Continued.}
1466: \clearpage
1467: {\plotone{f15f.eps}}\\
1468: \centerline{Fig. 14. --- Continued.}
1469: \clearpage
1470: {\plotone{f15g.eps}}\\
1471: \centerline{Fig. 14. --- Continued.}
1472: \clearpage
1473: {\plotone{f15h.eps}}\\
1474: \centerline{Fig. 14. --- Continued.}
1475: \clearpage
1476: {\plotone{f15i.eps}}\\
1477: \centerline{Fig. 14. --- Continued.}
1478: \clearpage
1479: {\plotone{f15j.eps}}\\
1480: \centerline{Fig. 14. --- Continued.}
1481: \clearpage
1482: {\plotone{f15k.eps}}\\
1483: \centerline{Fig. 14. --- Continued.}
1484: \clearpage
1485: {\plotone{f15l.eps}}\\
1486: \centerline{Fig. 14. --- Continued.}
1487: \clearpage
1488: {\plotone{f15m.eps}}\\
1489: \centerline{Fig. 14. --- Continued.}
1490: \clearpage
1491: {\plotone{f15n.eps}}\\
1492: \centerline{Fig. 14. --- Continued.}
1493: \clearpage
1494: {\plotone{f15o.eps}}\\
1495: \centerline{Fig. 14. --- Continued.}
1496: \clearpage
1497: {\plotone{f15p.eps}}\\
1498: \centerline{Fig. 14. --- Continued.}
1499: \clearpage
1500: {\plotone{f15q.eps}}\\
1501: \centerline{Fig. 14. --- Continued.}
1502: \clearpage
1503: {\plotone{f15r.eps}}\\
1504: \centerline{Fig. 14. --- Continued.}
1505: \clearpage
1506: {\plotone{f15s.eps}}\\
1507: \centerline{Fig. 14. --- Continued.}
1508: \clearpage
1509: \pagestyle{plaintop}
1510:
1511: \clearpage
1512:
1513: % ***************************************************************************
1514:
1515:
1516: \section{DATA FOR THE 441 STARS WITH MEASURED ROTATION PERIODS IN THE
1517: FIELD OF M35}
1518:
1519: Table~\ref{tab2} presents the results from this study together with
1520: information relevant to this paper for 441 stars in the field of M35.
1521: In the printed journal a stub version of Table 2 show the form of the
1522: full table and a sample the first 5 lines of its contents. The full
1523: version of the table can be found online. The stars appear in order
1524: of increasing rotation period, and
1525: the running number in the first column corresponds to the number in
1526: the upper left hand corner of the stars light curve in Appendix A.
1527: Columns 2 and 3 give the stellar equatorial coordinates (equinox 2000).
1528: Column 4 lists the measured stellar rotation period in decimal days.
1529: Columns 5, 6, and 7 gives the stellar V magnitude and B-V and V-I color
1530: indices, respectively, corrected for extinction and reddening. Column
1531: 8 presents the number of radial-velocity measurements for the star and
1532: columns 9 and 10 give the mean radial-velocity and the velocity standard
1533: deviation, respectively. Column 11 list the radial-velocity cluster
1534: membership probability calculated using the formalism by \citet{vkp58}.
1535: Column 12 contains a proper-motion cluster membership probability from
1536: either \citet{cudworth71} or \citet{ms86}. In column 13 we give the
1537: abbreviated membership codes (initialisms) also found in the light
1538: curves in Appendix A. The codes denote the type of membership information
1539: available for the star and have the following meaning: Photometric
1540: Member (PM; described in Section~\ref{phot}), Photometric Non-Member
1541: (PNM), Photometric and Spectroscopic Member (PSM), Photometric and
1542: Proper-motion Member (PPM), and Photometric Member but Spectroscopic
1543: Non-Member (PMSNM). In column 14 we give the weights used for each star
1544: when fitting the rotational isochrones in Section~\ref{isochrones} and
1545: Section~\ref{mass-rotation}. Finally, in column 15 the rotational state
1546: of the star is indicated by a 1-letter code representing, respectively,
1547: the I sequence (``i''), the C sequence (``c''), and the gap (``g'').
1548: Stars with a ``-'' in column 15 have locations in the color-period diagram
1549: that do not correspond to either of the sequences or the gap.
1550:
1551: \notetoeditor{An example of Table 2 with the first 5 lines can go here.
1552: tab2.tex contains just the first 5 lines of data. tab2_online.tex is
1553: the full table. NOTE: in the "b", and "c" table notes, the first letter
1554: is missing.}
1555:
1556: \clearpage
1557: \input{tab2.tex}
1558: \clearpage
1559:
1560: % ***************************************************************************
1561:
1562:
1563: \section{THE ROTATION PERIOD DISTRIBUTION OF NON-MEMBERS} \label{nm}
1564:
1565: In this section, we display and briefly comment upon the rotation
1566: period distribution of the non-members among the 441 rotators, which
1567: presumably are
1568: mostly field stars belonging to the Galactic disk. Unlike
1569: the cluster members we do not know the ages, distances, or masses
1570: of these stars. We will therefore only comment on a comparison
1571: between the two distributions and on distinct features in the
1572: period distribution of the 131 non-members shown in Figure~\ref{nmd}.
1573: First, rotation periods are detected over approximately the same
1574: range ($\sim$0.1 - 15 days) as for the 150 Myr cluster members,
1575: with only a few stars rotating slower. Second, there appears to
1576: be no indication of a bimodal distribution, but rather a distribution
1577: with a peak of ultra fast rotators and a long tail of periods
1578: up to and beyond 15 days. Third, and most strikingly, as shown
1579: by the 0.1 day resolution of the insert in Figure~\ref{nmd}, the
1580: non-member distribution exhibits a very distinct peak between 0.1
1581: and 0.2 days. The phased light curves for these stars (Appendix A)
1582: shows that the majority of these stars have large and well defined
1583: photometric variability with with peak-to-peak amplitudes of $0\fm1$
1584: or higher. It is possible that most of these stars are contact binaries
1585: of, e.g., the W UMa-type. Such systems typically have orbital periods
1586: of order 0.2-0.5 days and occur with a frequency of $\sim$0.2\%
1587: \citep[OGLE Variable Star Catalog;][]{rucinski97}. The frequency
1588: of rapidly varying non-members found in the field of M35 is $\sim$20
1589: out of 13700 or $\sim$0.15\% and thus in good agreement with that
1590: found from the OGLE Catalog. Indeed, if the light curves for these
1591: stars represent eclipses rather than spot-modulation, then the
1592: measured periods must be doubled bringing them into the expected
1593: range for W Uma systems.
1594:
1595: \clearpage
1596: \begin{figure}[ht!]
1597: \epsscale{1.0}
1598: \plotone{f16.eps}
1599: \caption{The rotation period distribution of the 131 radial-velocity
1600: and/or photometric non-members. The insert shows the distribution of
1601: periods less than 1 day with increased resolution of 0.1 day.
1602: \label{nmd}}
1603: \end{figure}
1604: \clearpage
1605:
1606: % ***************************************************************************
1607: %\bibliographystyle{apj}
1608: %\bibliography{refs}
1609:
1610: \begin{thebibliography}{66}
1611: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
1612:
1613: \bibitem[{{Alphenaar} \& {van Leeuwen}(1981)}]{av81}
1614: {Alphenaar}, P., \& {van Leeuwen}, F. 1981, Informational Bulletin on Variable
1615: Stars, 1957, 1
1616:
1617: \bibitem[{{Barnes} \& {Sofia}(1996)}]{bs96}
1618: {Barnes}, S., \& {Sofia}, S. 1996, \apj, 462, 746
1619:
1620: \bibitem[{{Barnes}(2003)}]{barnes03a}
1621: {Barnes}, S.~A. 2003, \apj, 586, 464
1622:
1623: \bibitem[{{Barnes}(2007)}]{barnes07}
1624: ---. 2007, \apj, 669, 1167
1625:
1626: \bibitem[{{Barrado y Navascu{\' e}s} {et~al.}(2001){Barrado y Navascu{\' e}s},
1627: {Deliyannis}, \& {Stauffer}}]{nds01}
1628: {Barrado y Navascu{\' e}s}, D., {Deliyannis}, C.~P., \& {Stauffer}, J.~R. 2001,
1629: \apj, 549, 452
1630:
1631: \bibitem[{{Benz} {et~al.}(1984){Benz}, {Mayor}, \& {Mermilliod}}]{bmm84}
1632: {Benz}, W., {Mayor}, M., \& {Mermilliod}, J.~C. 1984, \aap, 138, 93
1633:
1634: \bibitem[{{Bouvier} {et~al.}(1997){Bouvier}, {Forestini}, \& {Allain}}]{bfa97}
1635: {Bouvier}, J., {Forestini}, M., \& {Allain}, S. 1997, \aap, 326, 1023
1636:
1637: \bibitem[{{Braden} {et~al.}(2008){Braden}, {Mathieu}, \& {Meibom}}]{bmm08}
1638: {Braden}, E., {Mathieu}, R.~D., \& {Meibom}, S. 2008, in preparation.
1639:
1640: \bibitem[{{Cudworth}(1971)}]{cudworth71}
1641: {Cudworth}, K.~M. 1971, \aj, 76, 475
1642:
1643: \bibitem[{{Deliyannis}(2008)}]{deliyannis08}
1644: {Deliyannis}, C.~P. 2008, private comm.
1645:
1646: \bibitem[{{Duvall} {et~al.}(1984){Duvall}, {Dziembowski}, {Goode}, {Gough},
1647: {Harvey}, \& {Leibacher}}]{ddg+84}
1648: {Duvall}, T.~L., {Dziembowski}, W.~A., {Goode}, P.~R., {Gough}, D.~O.,
1649: {Harvey}, J.~W., \& {Leibacher}, J.~W. 1984, \nat, 310, 22
1650:
1651: \bibitem[{{Edwards} {et~al.}(1993){Edwards}, {Strom}, {Hartigan}, {Strom},
1652: {Hillenbrand}, {Herbst}, {Attridge}, {Merrill}, {Probst}, \&
1653: {Gatley}}]{esh+93}
1654: {Edwards}, S., {Strom}, S.~E., {Hartigan}, P., {Strom}, K.~M., {Hillenbrand},
1655: L.~A., {Herbst}, W., {Attridge}, J., {Merrill}, K.~M., {Probst}, R., \&
1656: {Gatley}, I. 1993, \aj, 106, 372
1657:
1658: \bibitem[{{Eff-Darwich} {et~al.}(2002){Eff-Darwich}, {Korzennik}, \& {Jim{\'
1659: e}nez-Reyes}}]{ekj02}
1660: {Eff-Darwich}, A., {Korzennik}, S.~G., \& {Jim{\' e}nez-Reyes}, S.~J. 2002,
1661: \apj, 573, 857
1662:
1663: \bibitem[{{Endal} \& {Sofia}(1981)}]{es81}
1664: {Endal}, A.~S., \& {Sofia}, S. 1981, \apj, 243, 625
1665:
1666: \bibitem[{{Geller} {et~al.}(2008){Geller}, {Mathieu}, Harris, \&
1667: McClure}]{gmh+08}
1668: {Geller}, A., {Mathieu}, R.~D., Harris, H.~C., \& McClure, R. 2008, submitted
1669: to AJ
1670:
1671: \bibitem[{{Goode} {et~al.}(1991){Goode}, {Dziembowski}, {Korzennik}, \&
1672: {Rhodes}}]{gdk+91}
1673: {Goode}, P.~R., {Dziembowski}, W.~A., {Korzennik}, S.~G., \& {Rhodes}, E.~J.
1674: 1991, \apj, 367, 649
1675:
1676: \bibitem[{{Gough}(1982)}]{gough82}
1677: {Gough}, D.~O. 1982, \nat, 298, 334
1678:
1679: \bibitem[{{Herbst} {et~al.}(2007){Herbst}, {Eisl{\"o}ffel}, {Mundt}, \&
1680: {Scholz}}]{hem+07}
1681: {Herbst}, W., {Eisl{\"o}ffel}, J., {Mundt}, R., \& {Scholz}, A. 2007, in
1682: Protostars and Planets V, ed. B.~{Reipurth}, D.~{Jewitt}, \& K.~{Keil},
1683: 297--311
1684:
1685: \bibitem[{{Herbst} \& {Wittenmyer}(1996)}]{hw96}
1686: {Herbst}, W., \& {Wittenmyer}, R. 1996, Bulletin of the American Astronomical
1687: Society, 28, 1338
1688:
1689: \bibitem[{{Honeycutt}(1992)}]{honeycutt92}
1690: {Honeycutt}, R.~K. 1992, \pasp, 104, 435
1691:
1692: \bibitem[{{Horne} \& {Baliunas}(1986)}]{hb86}
1693: {Horne}, J.~H., \& {Baliunas}, S.~L. 1986, \apj, 302, 757
1694:
1695: \bibitem[{{Jianke} \& {Collier Cameron}(1993)}]{jc93}
1696: {Jianke}, L., \& {Collier Cameron}, A. 1993, \mnras, 261, 766
1697:
1698: \bibitem[{{Kalirai} {et~al.}(2003){Kalirai}, {Fahlman}, {Richer}, \&
1699: {Ventura}}]{kfr+03}
1700: {Kalirai}, J.~S., {Fahlman}, G.~G., {Richer}, H.~B., \& {Ventura}, P. 2003,
1701: \aj, 126, 1402
1702:
1703: \bibitem[{{Kawaler}(1988)}]{kawaler88}
1704: {Kawaler}, S.~D. 1988, \apj, 333, 236
1705:
1706: \bibitem[{{Kawaler}(1989)}]{kawaler89}
1707: ---. 1989, \apjl, 343, L65
1708:
1709: \bibitem[{{Koenigl}(1991)}]{konigl91}
1710: {Koenigl}, A. 1991, \apjl, 370, L39
1711:
1712: \bibitem[{{Kraft}(1967)}]{kraft67}
1713: {Kraft}, R.~P. 1967, \apj, 150, 551
1714:
1715: \bibitem[{{Krishnamurthi} {et~al.}(1997){Krishnamurthi}, {Pinsonneault},
1716: {Barnes}, \& {Sofia}}]{kpb+97}
1717: {Krishnamurthi}, A., {Pinsonneault}, M.~H., {Barnes}, S., \& {Sofia}, S. 1997,
1718: \apj, 480, 303
1719:
1720: \bibitem[{{Lockwood} {et~al.}(1984){Lockwood}, {Thompson}, {Radick}, {Osborn},
1721: {Baggett}, {Duncan}, \& {Hartmann}}]{ltr+84}
1722: {Lockwood}, G.~W., {Thompson}, D.~T., {Radick}, R.~R., {Osborn}, W.~H.,
1723: {Baggett}, W.~E., {Duncan}, D.~K., \& {Hartmann}, L.~W. 1984, \pasp, 96, 714
1724:
1725: \bibitem[{{MacGregor} \& {Brenner}(1991)}]{mb91}
1726: {MacGregor}, K.~B., \& {Brenner}, M. 1991, \apj, 376, 204
1727:
1728: \bibitem[{{Mathieu}(2000)}]{mathieu00}
1729: {Mathieu}, R.~D. 2000, in ASP Conf. Ser. 198: Stellar Clusters and
1730: Associations: Convection, Rotation, and Dynamos, 517
1731:
1732: \bibitem[{{McNamara} \& {Sekiguchi}(1986)}]{ms86}
1733: {McNamara}, B., \& {Sekiguchi}, K. 1986, \aj, 91, 557
1734:
1735: \bibitem[{{Meibom} {et~al.}(2001){Meibom}, {Barnes}, {Dolan}, \&
1736: {Mathieu}}]{mbd+01}
1737: {Meibom}, S., {Barnes}, S.~A., {Dolan}, C., \& {Mathieu}, R.~D. 2001, in ASP
1738: Conf. Ser. 243: From Darkness to Light: Origin and Evolution of Young Stellar
1739: Clusters, 711
1740:
1741: \bibitem[{{Meibom} \& {Mathieu}(2005)}]{mm05}
1742: {Meibom}, S., \& {Mathieu}, R.~D. 2005, \apj, 620, 970
1743:
1744: \bibitem[{{Meibom} {et~al.}(2006){Meibom}, {Mathieu}, \& {Stassun}}]{mms06}
1745: {Meibom}, S., {Mathieu}, R.~D., \& {Stassun}, K.~G. 2006, \apj, 653, 621
1746:
1747: \bibitem[{{Meibom} {et~al.}(2008){Meibom}, {Mathieu}, \& {Stassun}}]{mms08b}
1748: ---. 2008, in preparation.
1749:
1750: \bibitem[{{Meys} {et~al.}(1982){Meys}, {Alphenaar}, \& {van Leeuwen}}]{mav82}
1751: {Meys}, J.~J.~M., {Alphenaar}, P., \& {van Leeuwen}, F. 1982, Informational
1752: Bulletin on Variable Stars, 2115, 1
1753:
1754: \bibitem[{{Najita}(1995)}]{najita95}
1755: {Najita}, J. 1995, in Revista Mexicana de Astronomia y Astrofisica Conference
1756: Series, 293
1757:
1758: \bibitem[{{Nordhagen} {et~al.}(2006){Nordhagen}, {Herbst}, {Rhode}, \&
1759: {Williams}}]{nhr+06}
1760: {Nordhagen}, S., {Herbst}, W., {Rhode}, K.~L., \& {Williams}, E.~C. 2006, \aj,
1761: 132, 1555
1762:
1763: \bibitem[{{Perryman} {et~al.}(1998){Perryman}, {Brown}, {Lebreton}, {Gomez},
1764: {Turon}, {de Strobel}, {Mermilliod}, {Robichon}, {Kovalevsky}, \&
1765: {Crifo}}]{pbl+98}
1766: {Perryman}, M.~A.~C., {Brown}, A.~G.~A., {Lebreton}, Y., {Gomez}, A., {Turon},
1767: C., {de Strobel}, G.~C., {Mermilliod}, J.~C., {Robichon}, N., {Kovalevsky},
1768: J., \& {Crifo}, F. 1998, \aap, 331, 81
1769:
1770: \bibitem[{{Pinsonneault} {et~al.}(1990){Pinsonneault}, {Kawaler}, \&
1771: {Demarque}}]{pkd90}
1772: {Pinsonneault}, M.~H., {Kawaler}, S.~D., \& {Demarque}, P. 1990, \apjs, 74, 501
1773:
1774: \bibitem[{{Pinsonneault} {et~al.}(1989){Pinsonneault}, {Kawaler}, {Sofia}, \&
1775: {Demarque}}]{pks+89}
1776: {Pinsonneault}, M.~H., {Kawaler}, S.~D., {Sofia}, S., \& {Demarque}, P. 1989,
1777: \apj, 338, 424
1778:
1779: \bibitem[{{Prosser} {et~al.}(1995){Prosser}, {Shetrone}, {Dasgupta}, {Backman},
1780: {Laaksonen}, {Baker}, {Marschall}, {Whitney}, {Kuijken}, \&
1781: {Stauffer}}]{psd+95}
1782: {Prosser}, C.~F., {Shetrone}, M.~D., {Dasgupta}, A., {Backman}, D.~E.,
1783: {Laaksonen}, B.~D., {Baker}, S.~W., {Marschall}, L.~A., {Whitney}, B.~A.,
1784: {Kuijken}, K., \& {Stauffer}, J.~R. 1995, \pasp, 107, 211
1785:
1786: \bibitem[{{Radick} {et~al.}(1987){Radick}, {Thompson}, {Lockwood}, {Duncan}, \&
1787: {Baggett}}]{rtl+87}
1788: {Radick}, R.~R., {Thompson}, D.~T., {Lockwood}, G.~W., {Duncan}, D.~K., \&
1789: {Baggett}, W.~E. 1987, \apj, 321, 459
1790:
1791: \bibitem[{{Rucinski}(1997)}]{rucinski97}
1792: {Rucinski}, S.~M. 1997, \aj, 113, 1112
1793:
1794: \bibitem[{{Scargle}(1982)}]{scargle82}
1795: {Scargle}, J.~D. 1982, \apj, 263, 835
1796:
1797: \bibitem[{{Schatzman}(1962)}]{schatzman62}
1798: {Schatzman}, E. 1962, Annales d'Astrophysique, 25, 18
1799:
1800: \bibitem[{{Shu} {et~al.}(1994){Shu}, {Najita}, {Ostriker}, {Wilkin}, {Ruden},
1801: \& {Lizano}}]{sno+94}
1802: {Shu}, F., {Najita}, J., {Ostriker}, E., {Wilkin}, F., {Ruden}, S., \&
1803: {Lizano}, S. 1994, \apj, 429, 781
1804:
1805: \bibitem[{{Sills} {et~al.}(2000){Sills}, {Pinsonneault}, \& {Terndrup}}]{spt00}
1806: {Sills}, A., {Pinsonneault}, M.~H., \& {Terndrup}, D.~M. 2000, \apj, 534, 335
1807:
1808: \bibitem[{{Skumanich}(1972)}]{skumanich72}
1809: {Skumanich}, A. 1972, \apj, 171, 565
1810:
1811: \bibitem[{{Soderblom}(1982)}]{soderblom82}
1812: {Soderblom}, D.~R. 1982, \apj, 263, 239
1813:
1814: \bibitem[{{Soderblom} {et~al.}(1983){Soderblom}, {Jones}, \& {Walker}}]{sjw83}
1815: {Soderblom}, D.~R., {Jones}, B.~F., \& {Walker}, M.~F. 1983, \apjl, 274, L37
1816:
1817: \bibitem[{{Soderblom} {et~al.}(1993{\natexlab{a}}){Soderblom}, {Stauffer},
1818: {Hudon}, \& {Jones}}]{ssh+93}
1819: {Soderblom}, D.~R., {Stauffer}, J.~R., {Hudon}, J.~D., \& {Jones}, B.~F.
1820: 1993{\natexlab{a}}, \apjs, 85, 315
1821:
1822: \bibitem[{{Soderblom} {et~al.}(1993{\natexlab{b}}){Soderblom}, {Stauffer},
1823: {MacGregor}, \& {Jones}}]{ssm+93}
1824: {Soderblom}, D.~R., {Stauffer}, J.~R., {MacGregor}, K.~B., \& {Jones}, B.~F.
1825: 1993{\natexlab{b}}, \apj, 409, 624
1826:
1827: \bibitem[{{Stassun} {et~al.}(2004){Stassun}, {Ardila}, {Barsony}, {Basri}, \&
1828: {Mathieu}}]{sab+04}
1829: {Stassun}, K.~G., {Ardila}, D.~R., {Barsony}, M., {Basri}, G., \& {Mathieu},
1830: R.~D. 2004, \aj, 127, 3537
1831:
1832: \bibitem[{{Stassun} {et~al.}(1999){Stassun}, {Mathieu}, {Mazeh}, \&
1833: {Vrba}}]{smm+99}
1834: {Stassun}, K.~G., {Mathieu}, R.~D., {Mazeh}, T., \& {Vrba}, F.~J. 1999, \aj,
1835: 117, 2941
1836:
1837: \bibitem[{{Stauffer} {et~al.}(1984){Stauffer}, {Hartmann}, {Soderblom}, \&
1838: {Burnham}}]{shs+84}
1839: {Stauffer}, J.~R., {Hartmann}, L., {Soderblom}, D.~R., \& {Burnham}, N. 1984,
1840: \apj, 280, 202
1841:
1842: \bibitem[{{Stauffer} \& {Hartmann}(1987)}]{sh87}
1843: {Stauffer}, J.~R., \& {Hartmann}, L.~W. 1987, \apj, 318, 337
1844:
1845: \bibitem[{{Stauffer} {et~al.}(1985){Stauffer}, {Hartmann}, {Burnham}, \&
1846: {Jones}}]{shb+85}
1847: {Stauffer}, J.~R., {Hartmann}, L.~W., {Burnham}, J.~N., \& {Jones}, B.~F. 1985,
1848: \apj, 289, 247
1849:
1850: \bibitem[{{Stauffer} {et~al.}(1989){Stauffer}, {Hartmann}, \& {Jones}}]{shj89}
1851: {Stauffer}, J.~R., {Hartmann}, L.~W., \& {Jones}, B.~F. 1989, \apj, 346, 160
1852:
1853: \bibitem[{{Stelzer} {et~al.}(2003){Stelzer}, {Fern{\'a}ndez}, {Costa},
1854: {Gameiro}, {Grankin}, {Henden}, {Guenther}, {Mohanty}, {Flaccomio},
1855: {Burwitz}, {Jayawardhana}, {Predehl}, \& {Durisen}}]{sfc+03}
1856: {Stelzer}, B., {Fern{\'a}ndez}, M., {Costa}, V.~M., {Gameiro}, J.~F.,
1857: {Grankin}, K., {Henden}, A., {Guenther}, E., {Mohanty}, S., {Flaccomio}, E.,
1858: {Burwitz}, V., {Jayawardhana}, R., {Predehl}, P., \& {Durisen}, R.~H. 2003,
1859: \aap, 411, 517
1860:
1861: \bibitem[{{Terndrup} {et~al.}(2000){Terndrup}, {Stauffer}, {Pinsonneault},
1862: {Sills}, {Yuan}, {Jones}, {Fischer}, \& {Krishnamurthi}}]{tsp+00}
1863: {Terndrup}, D.~M., {Stauffer}, J.~R., {Pinsonneault}, M.~H., {Sills}, A.,
1864: {Yuan}, Y., {Jones}, B.~F., {Fischer}, D., \& {Krishnamurthi}, A. 2000, \aj,
1865: 119, 1303
1866:
1867: \bibitem[{{van Leeuwen} \& {Alphenaar}(1982)}]{va82}
1868: {van Leeuwen}, F., \& {Alphenaar}, P. 1982, The Messenger, 28, 15
1869:
1870: \bibitem[{{Vasilevskis} {et~al.}(1958){Vasilevskis}, {Klemola}, \&
1871: {Preston}}]{vkp58}
1872: {Vasilevskis}, S., {Klemola}, A., \& {Preston}, G. 1958, \aj, 63, 387
1873:
1874: \bibitem[{{von Hippel} {et~al.}(2002){von Hippel}, {Steinhauer}, {Sarajedini},
1875: \& {Deliyannis}}]{vss+02}
1876: {von Hippel}, T., {Steinhauer}, A., {Sarajedini}, A., \& {Deliyannis}, C.~P.
1877: 2002, \aj, 124, 1555
1878:
1879: \bibitem[{{Yi} {et~al.}(2003){Yi}, {Kim}, \& {Demarque}}]{ykd03}
1880: {Yi}, S.~K., {Kim}, Y., \& {Demarque}, P. 2003, \apjs, 144, 259
1881:
1882: \end{thebibliography}
1883:
1884: \end{document}
1885: