1: \documentclass[10pt,preprint]{aastex}
2: %\documentclass[12pt,preprint]{aastex}
3:
4: %% This is a sample manuscript marked up using the
5: %% AASTeX v5.x LaTeX 2e macros.
6:
7: %% The first piece of markup in an AASTeX v5.x document
8: %% is the \documentclass command. LaTeX will ignore
9: %% any data that comes before this command.
10:
11: %% The command below calls the preprint style
12: %% which will produce a one-column, single-spaced document.
13: %% Examples of commands for other substyles follow. Use
14: %% whichever is most appropriate for your purposes.
15:
16:
17: %\Documentclass{aastex}
18: %\usepackage{emulateapj5}
19:
20: %% manuscript produces a one-column, double-spaced document:
21:
22: % \documentclass[manuscript]{aastex}
23:
24: %% preprint2 produces a double-column, single-spaced document:
25:
26: % \documentclass[preprint2]{aastex}
27:
28: %% If you want to create your own macros, you can do so
29: %% using \newcommand. Your macros should appear before
30: %% the \begin{document} command.
31: %%
32: %% If you are submitting to a journal that translates manuscripts
33: %% into SGML, you need to follow certain guidelines when preparing
34: %% your macros. See the AASTeX v5.x Author Guide
35: %% for information.
36:
37: \newcommand{\vdag}{(v)^\dagger}
38: \newcommand{\myemail}{skywalker@galaxy.far.far.away}
39:
40: %% You can insert a short comment on the title page using the command below.
41:
42: \slugcomment{Accepted for publication in The Astrophysical Journal}
43:
44: %% If you wish, you may supply running head information, although
45: %% this information may be modified by the editorial offices.
46: %% The left head contains a list of authors,
47: %% usually a maximum of three (otherwise use et al.). The right
48: %% head is a modified title of up to roughly 44 characters. Running heads
49: %% will not print in the manuscript style.
50:
51: \shorttitle{Ages for field stars using gyrochronology}
52: \shortauthors{Barnes}
53:
54: %% This is the end of the preamble. Indicate the beginning of the
55: %% paper itself with \begin{document}.
56:
57: %\received{2002 December 11}
58: \begin{document}
59:
60: %% LaTeX will automatically break titles if they run longer than
61: %% one line. However, you may use \\ to force a line break if
62: %% you desire.
63:
64: \title{Ages for illustrative field stars using gyrochronology: viability, limitations and errors}
65: %}
66:
67: %% Use \author, \affil, and the \and command to format
68: %% author and affiliation information.
69: %% Note that \email has replaced the old \authoremail command
70: %% from AASTeX v4.0. You can use \email to mark an email address
71: %% anywhere in the paper, not just in the front matter.
72: %% As in the title, you can use \\ to force line breaks.
73:
74: \author{Sydney A. Barnes}
75: \affil{Lowell Observatory, 1400 W. Mars Hill Rd., Flagstaff, AZ 86001, USA}
76: \email{barnes@lowell.edu}
77:
78: %% Notice that each of these authors has alternate affiliations, which
79: %% are identified by the \altaffilmark after each name. Specify alternate
80: %% affiliation information with \altaffiltext, with one command per each
81: %% affiliation.
82: %% Mark off your abstract in the ``abstract'' environment. In the manuscript
83: %% style, abstract will output a Received/Accepted line after the
84: %% title and affiliation information. No date will appear since the author
85: %% does not have this information. The dates will be filled in by the
86: %% editorial office after submission.
87:
88: \begin{abstract}
89: %\\{\bf Summary: }
90: We here develop an improved way of using a rotating star as a clock,
91: set it using the Sun, and demonstrate that it keeps time well.
92: %{\bf What gyrochronology is: }
93: This technique, called gyrochronology, permits the derivation of ages for
94: solar- and late-type main sequence stars using only their rotation periods and
95: colors.
96: %\\{\bf Additional BS: }
97: The technique is clarified and developed here, and used to derive ages for
98: illustrative groups of nearby, late-type field stars with measured rotation
99: periods.
100: %The technique's viability and limitations are investigated here
101: %using rotation period observations currently available for late-type stars in
102: %open clusters, field stars from the Mt.\,Wilson sample, and the Sun, which
103: %serves as the principal calibrator.
104: %\\{\bf The crux of the technique: }
105: We first demonstrate the reality of the interface sequence, the unifying
106: feature of the rotational observations of cluster and field stars that makes
107: the technique possible, and extends it beyond the proposal of Skumanich by
108: specifying the mass dependence of rotation for these stars.
109: We delineate which stars it cannot currently be used on.
110: %, and remove these to improve the mass dependence.
111: We then calibrate the age dependence using the Sun.
112: %(with a nod to Copernicus!)
113: %\\{\bf Errors are calculated: }
114: The errors are propagated to understand their dependence on color and period.
115: Representative age errors associated with the technique are estimated at
116: $\sim$15\% (plus possible systematic errors) for
117: late\,F, G, K, \& early\,M stars.
118: %\\{\bf Comparison with chromospheric ages for Mt.\,Wilson stars: }
119: Ages derived via gyrochronology for the Mt.\,Wilson stars are shown to be in
120: good agreement with chromospheric ages for all but the bluest stars, and
121: probably superior.
122: %\\{\bf Ages for a young field star sample (Strassmeier 2000) with known chromospheric activity: }
123: Gyro ages are then calculated for each of the active main sequence field stars
124: studied by Strassmeier and collaborators where other ages are not available.
125: %The median age for this sample is shown to be 380\,Myr.
126: These are shown to be mostly younger than 1\,Gyr, with a median age of 365\,Myr.
127: % consistent with their selection for activity and other astronomical
128: %information.
129: %\\{\bf Ages for a young- and intermediate-age sample (Pizzolato 2003) with measured X-rays: }
130: The sample of single, late-type main sequence field stars assembled by Pizzolato
131: and collaborators is then assessed, and shown to have gyro ages ranging from
132: under 100\,Myr to several\,Gyr, and a median age of 1.2\,Gyr.
133: %consistent with their selection and the available X-ray fluxes.
134: %\\{\bf Wide binaries: }
135: Finally, we demonstrate that, in contrast to the other techniques, the
136: individual components of the three wide binaries
137: $\xi$BooAB, 61CygAB, \& $\alpha$CenAB yield substantially the same gyro ages.
138: %{\bf Isochrones (out of place?) }
139: %Finally, we suggest that when both are available, gyro ages are superior to
140: %isochrone ages for single field stars.
141: \end{abstract}
142:
143: %% Keywords should appear after the \end{abstract} command. The uncommented
144: %% example has been keyed in ApJ style. See the instructions to authors
145: %% for the journal to which you are submitting your paper to determine
146: %% what keyword punctuation is appropriate.
147:
148: \keywords{open clusters and associations: general --- stars: activity ---
149: stars: evolution --- stars: late-type --- stars: magnetic fields ---
150: stars: rotation}
151:
152: %% From the front matter, we move on to the body of the paper.
153: %% In the first two sections, notice the use of the natbib \citep
154: %% and \citet commands to identify citations. The citations are
155: %% tied to the reference list via symbolic KEYs. The KEY corresponds
156: %% to the KEY in the \bibitem in the reference list below. We have
157: %% chosen the first three characters of the first author's name plus\citet{aur82}
158: %% the last two numeral of the year of publication as our KEY for
159: %% each reference.
160:
161: %\clearpage
162:
163: \section{Introduction} \label{intro}
164:
165: \subsection{Stellar age indicators, and motivation for a rotation clock}
166:
167: %{\bf Stars serve as clocks $\rightarrow$ isochrones: }\\
168: The age of a star is its most fundamental attribute apart from its mass, and
169: usually provides the chronometer that permits the study of the time evolution
170: of astronomical phenomena.
171: Consequently, a great deal of effort has been expended over the past several
172: decades on the possibility of using stars as clocks, to reveal their own
173: ages, those of the astronomical bodies associated with them, and to understand
174: how various astronomical phenomena unfold over time.
175:
176: %{\bf Beginning of isochrone talk: }\\
177: The most successful of these chronometric techniques is the isochrone method
178: (invented by Sandage, 1962; named and developed substantially by
179: Demarque and Larson, 1964),
180: based on the steady change in the color-magnitude morphology of a collection
181: of stars, in response to the consumption and dimunition of their nuclear fuel
182: (e.g. VandenBerg et al., 2006; Demarque et al., 2004;
183: Girardi et al., 2002; Kim et al., 2002; Yi et al., 2001).
184:
185: \subsubsection{The principal limitations of the Isochrone clock}
186:
187: %{\bf 2 limitations: }\\
188: The isochrone technique fashions a collection of coeval stars of differing
189: masses, i.e.
190: a star cluster, into a remarkable clock that provides the age of the system.
191: However, vast numbers of stars, including our own Sun and most of the nearby
192: stars amenable to detailed study, are no longer in identifiable
193: %all the host stars of extra-solar planets known to date,
194: clusters and spend their lives in relative isolation as field stars.
195: For these stars, the isochrone technique is less useful, because a star spends
196: most of its life burning hydrogen steadily on the main sequence, where its
197: luminosity and temperature, the primary indicators of isochrone age, are almost
198: {\it constant}\footnote{For example, in the 4.5\,Gyr since it was on the zero
199: age main sequence (ZAMS), the Sun's luminosity has increased by $\leq$50\% of
200: its ZAMS value.}.
201: Using classical isochrones to tell the ages of single, low-mass, main sequence
202: stars is akin to using gray hairs or baldness as an age
203: indicator for toddlers, adolescents and adults!
204:
205: %{\bf 3rd limitation: Need the distance for field stars }\\
206: Furthermore, the isochrone technique requires a measurement of the distance to
207: a field star to calculate its luminosity. This distance is hard to measure, and
208: in fact, even after the publication of the results of the Hipparcos satellite
209: (ESA, 1997), we know the distances to only $\sim$20,000 field stars
210: (all of them nearby) to better than 10\% (Perryman et al. 1997).
211: This imprecision leads to large errors in isochrone ages.
212: A 10\% error on the distance to a solar twin would result in $\sim$20\% errors
213: in its luminosity, and isochrone ages between 2 and 10\,Gyr\footnote{The Hipparcos satellite has indeed provided $\sim$1\% parallaxes for a group of stars, most of them bright enough to have been included in the catalog of Hipparchus himself if they were visible from Greece! To count them, you would need your own digits and those of some of your collaborators, but you wouldn't need more than a few of the latter.}.
214: Because the age of a star provides a direct link to many of its other
215: properties, this deficiency is keenly felt. Knowledge of the age of a field
216: star, however crude, is a very valuable astronomical commodity indeed.
217:
218: %{\bf Need other clocks }\\
219: Thus we need to consider the possibility of fashioning clocks using other
220: properties of (individual) stars.
221: In particular, it would be very valuable to construct an age indicator that is
222: independent of distance, and indeed, some of the activity-related indicators
223: suggested over the years, including the primary one used today, do have this
224: valuable characteristic.
225: In fact, the details of the pros and cons of the isochrone and other
226: chronometers are such that it might be useful here to step back even further
227: and consider how an age indicator is constructed,
228: and the general characteristics desirable for stellar age indicators.
229:
230:
231: \subsubsection{Steps in the construction of age indicators}
232:
233: Five major steps seem to describe the process:
234: \begin{enumerate}
235: \item One needs to find an observable, $v$, that changes sensitively and
236: smoothly, perhaps monotonically, with age. Preferably, this observable would be
237: a property of individual stars rather than that of a (co-eval) collection of
238: them.
239: \item One needs to determine the ages of suitable calibrating objects
240: independently. These would provide the connection to the fundamental units like
241: earth rotations, pendulum swings, etc.
242: \item One needs to identify and measure the functional form of the variable:
243: $v = v(t, w, x, ...)$ where $t$ is the age, and $w$, $x$, $...$, are possible
244: additional dependencies of the variable $v$. It is preferable to have fewer
245: variables, and to have separable dependencies of the form
246: $v = T(t) \times W(w) \times X(x) ....$.
247: \item One needs to invert the dependence determined experimentally, numerically
248: or otherwise, to find $t = t (v, w, x, ...)$. This is usually non-linear, and
249: sometimes has undesirable kinks.
250: \item Finally, one needs to calculate the error
251: $\delta t = \delta t (t, v, w, x, ...)$.
252: \end{enumerate}
253:
254: \subsubsection{Characteristics desired for age indicators}
255:
256: The foregoing considerations suggest that the following characteristics are
257: desirable for stellar age indicators.
258: \begin{enumerate}
259: \item {\bf Measurability for single stars: }
260: The indicator should be properly defined, measurable easily itself, and
261: preferably should not require many additional quantities to be measured,
262: otherwise it cannot be used routinely.
263: \item {\bf Sensitivity to Age: }
264: The indicator should have a sensitive dependence on age, i.e., should
265: change substantially (and preferably regularly) with age,
266: otherwise the errors will be inherently large.
267: \item {\bf Insensitivity to other parameters: }
268: The indicator should have insensitive (or separable) dependencies on other
269: parameters that affect the measured quantity,
270: otherwise there is the potential for ambiguity.
271: \item {\bf Calibration: }
272: The technique should be calibrable using an object (or set of objects)
273: whose age(s) we know very well, otherwise systematic errors will be
274: introduced.
275: \item {\bf Invertibility: } The functional dependence determined above should
276: be properly invertible to yield the age as a function of the measured
277: variables.
278: \item {\bf Error analysis: }
279: The errors on the age derived using the technique ought to be calculable,
280: otherwise no confidence can be attached to the ages.
281: \item {\bf Test of coeval stars: }
282: The technique should yield the same ages for stars expected to be coeval,
283: otherwise the validity of the technique itself must be questioned.
284: \end{enumerate}
285: We summarize in Table\,1 how (in)adequately these characteristics are satisfied
286: by the three age indicators relevant to this paper.
287: While the entries, especially for gyrochronology, anticipate the results derived
288: in this paper, the characteristics desired guide the progress of, and form a
289: continuous backdrop to this work.
290:
291: %{\bf Table\,1: }
292:
293: \clearpage
294:
295: \begin{deluxetable}{llll}
296: \tabletypesize{\scriptsize}
297: \tablecaption{Characteristics of the three major age indicators for field stars
298: \label{tbl-1}}
299: \tablewidth{0pt}
300: \tablehead{
301: \colhead{Property} & \colhead{Isochrone age} & \colhead{Chromospheric age} & \colhead{Gyrochronology}
302: %\colhead{$\Theta$\tablenotemark{b}}
303: }
304: \startdata
305: Measurable easily? & ?\tablenotemark{a} (Distance reqd.)& ?\tablenotemark{b} (Repetition reqd.) & ?\tablenotemark{c} (Repetition reqd.)\\
306: Sensitive to age? & No (on MS) & Yes & Yes \\
307: Insensitive to other parameters? & No & Yes\tablenotemark{d} & Yes \\
308: Technique calibrable? & Yes (Sun) & ?\tablenotemark{e} (Sun?) & Yes (Sun) \\
309: Invertible easily? & No & Yes & Yes \\
310: Errors calculable/provided? & ?\tablenotemark{f} (Difficult) & Yes?\tablenotemark{g} & Yes \\
311: Coeval stars yield the same age? & No (Field binaries) & ?\tablenotemark{h} & Yes \\
312: \enddata
313:
314:
315: %% Text for table notes should follow after the \enddata but before
316: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
317: %% in the table for each \tablenotetext.
318:
319: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
320: %with the deluxetable environment}
321: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
322: \tablenotetext{a}{A field star requires a good distance measurement in order to
323: determine its luminosity for comparison with isochrones. As explained in the
324: text, good distances are available to only a few such stars.}
325: \tablenotetext{b}{Another reason for this `?' is that it is not clear to an
326: innocent bystander how to transform
327: between the various quantities published as a chromospheric flux:
328: $S$, HK\,index, $R$, or $R'_{HK}$. The lack of a defined standard quantity for
329: published work is a significant drawback.}
330: \tablenotetext{c}{Another reason for this `?' is that for old stars the
331: modulation in broadband photometric filters is too small to yield a rotation
332: period, and for these stars one must resort to more onerous means such as
333: detecting the rotational modulation in the Ca\,II\,H\,\&\,K emission cores.}
334: \tablenotetext{d}{The benefit of doubt has been given but in fact, there is
335: usually some black magic in the transformation between chromospheric flux and
336: age.}
337: \tablenotetext{e}{The relationship between chromospheric emission and age in
338: Soderblom et al. (1991) is calibrated against isochrone ages of three
339: ``fundamental'' points, and those of the evolved components of visual binaries.
340: Since all isochrones are calibrated using the age of the Sun, this calibration
341: is also ultimately based on the age of the Sun, except for the additional step
342: involved.}
343: \tablenotetext{f}{Errors on isochrone ages for field stars were essentially
344: non-existent until Pont \& Eyer (2004) suggested a Bayesian scheme that allows
345: one to determine whether or not an isochrone age is `well-defined' (Jorgensen
346: \& Lindegren 2005) i.e. whether or not the probability density distribution for
347: the age has an identifiable maximum, and if so, to calculate an error based on
348: this property. This method has since been used by Takeda et al. (2007) on their
349: field star sample.}
350: \tablenotetext{g}{Soderblom et al. (1991) provide the error on their fit, in
351: this case $\sim$0.17\,dex ($\sim$40\%), of chromospheric emission to
352: (isochrone) age for their sample. Other researchers,
353: including Donahue (1998), usually do not provide errors.}
354: \tablenotetext{h}{For the eight pairs in Table\,2 of Donahue (1998), the mean
355: discrepancy is 0.85\,Gyr for a sample with a mean age of 1.85\,Gyr, so that the
356: fractional discrepancy in age is 0.46, or just under 50\%.}
357:
358: %\tablecomments{Occasionally, authors wish to append a short
359: %paragraph of explanatory notes that pertain to the entire table, but
360: %which are different than the caption. Such notes should be placed in
361: %a {\tt tablecomments} command like this.}
362:
363: \end{deluxetable}
364:
365: \subsubsection{Motivations for investigating a rotational clock}
366:
367: %{\bf The other age indicators: }\\
368: A wide array of age indicators have been developed over the past decades.
369: The most well-known are chromospheric emission (Wilson, 1963) and
370: rotation (Skumanich, 1972), but others like surface lithium abundance
371: (Vauclair, 1972; Rebolo, Martin \& Maguzzu 1992; Basri, Marcy \& Graham 1996;
372: Stauffer 2000) and coronal emission in X-rays (Kunte et al. 1988), usually
373: through its dependence on rotation (Pallavicini et al. 1981; Gudel 2004),
374: have also occasionally been suggested and used in various contexts.
375: All of these are related to stellar activity and are based on empirical
376: correlations between the property in question and stellar age.
377: They have been considered less reliable clocks than the canonical isochrone
378: technique because the underlying physics is not well understood, and in fact
379: there is a great deal of debate even about what the important underlying
380: phenomena are.
381: %Also, there are calibration issues....
382: Finally, one must also consider whether and how each of these age indicators is
383: calibrated.
384:
385: %{\bf Chromospheric emission has been the best other clock thus far}\\
386: Ever since the work of Skumanich (1972), and especially since the work of
387: Noyes et al. (1984), the relationship between chromospheric emission and age
388: has enjoyed the distinction of being the most consistent,
389: %(although there undoubtedly are calibration issues)
390: making chromospheric emission the leading age indicator for nearby field stars
391: (e.g. Soderblom 1985; Henry et al. 1996; Wright et al. 2004).
392:
393: %{\bf But chromospheric emission is not one of the fundamental stellar properties }\\
394: But there are more fundamental stellar observables than chromospheric
395: emission. In fact, of all the activity-related properties of stars, rotation
396: is undoubtedly the most fundamental, and many believe that together with
397: stellar mass (and another variable or two), it might be responsible, directly or
398: indirectly, for the observed morphology of all the other activity indicators.
399: %(The unconvinced will at least admit the weaker statement that the other
400: %activity indicators depend on rotation in various ways.)
401:
402: %{\bf Stellar rotation is the most fundamental of the activity-related indicators }\\
403: In fact, besides being obviously independent of the distance to the star,
404: stellar rotation is now known to change systematically, even predictably,
405: on the main sequence, where the isochrone technique is at its weakest.
406: Furthermore, the specific form of the rotational spindown of stars is such that
407: {\it initial variations in the rotation rates of young stars appear to become
408: increasingly unimportant with the passage of time, leading to an almost unique
409: relationship between rotation period and age for a star of a given mass}.
410: Finally, rotation is a property we can now measure to great precision;
411: rotation periods for late-type stars are sometimes determined today to better
412: than one part in ten thousand\footnote{The usefulness of this precision is less clear in the context of the differential rotation with latitude of the Sun and solar-type stars, but it is also clear that we are beginning to understand the systematics and origin of differential rotation, so that the attainment of such precision is useful in other ways as well.}!
413: These features of stellar rotation - its predictability, measurability, and
414: simplicity - suggest that some effort is warranted in improving its use
415: as an age indicator beyond the relationship suggested by Skumanich (1972).
416:
417: %{\bf Rewrite this paragraph: }\\
418: In fact, as we shall show below, and as is summarized in Table\,1,
419: gyrochronology satisfies more of the criteria required for an age indicator as
420: listed above, than any other
421: astronomical clock in use, and appears to be complementary to the isochrone
422: technique, in that it works very well on the main sequence, while the isochrone
423: method is better suited to evolved stars.
424:
425: %{\bf Reason for this paper: }\\
426: This paper addresses the issues of constructing and calibrating a rotational
427: clock. It appears that to first order stellar rotation depends only on the mass
428: and age of the star, so that jointly taking account of these dependencies of
429: rotation permits the determination of rotational ages ({\it and their errors})
430: for a substantial sample of main sequence stars, and
431: even {\it individual field stars}, a technique we suggest be called
432: ``gyrochronology.''
433:
434: %\newpage
435: \subsection{Stellar rotation as an astronomical clock}
436:
437: %{\bf The Skumanich (1972) rotation clock }\\
438: Major steps in the direction of using stellar rotation as a clock were made by
439: a series of studies in the 1960s, culminating in the famous relationship of
440: Skumanich (1972), relating the averaged surface rotational velocities,
441: $\overline{v \sin i}$, of stars in a number of open clusters to their ages, $t$,
442: via the expression: $\overline{v \sin i} \propto 1/\sqrt{t}$.
443: Skumanich noted that the equatorial surface rotation velocity of the Sun
444: at its independently derived age also matched this relationship\footnote{The
445: age of the Sun is not directly known, of course. We use the age of the formation
446: of the refractory inclusions in the Allende meteorite as an estimate of the
447: Sun's age (e.g. Allegre et al. 1995 but see also Patterson 1953; 1955; 1956;
448: Patterson et al. 1995 and Murthy \& Patterson, 1962 for the original work
449: establishing that the age of the Earth and that of the meteorites is identical
450: and can be called the ``age of the solar system'').}.
451: Over the years, astronomers have come to believe that this relationship
452: encapsulates something fundamental about the nature of winds and angular
453: momentum loss from late-type stars\footnote{It appears to be equivalent to a
454: cubic dependence on the rotation speed, $\Omega$, of the angular momentum loss
455: rate, $dJ/dt$, from solar- and late-type stars: $ dJ/dt \propto -\Omega^3$
456: (Kawaler, 1988). In fact, parameterizations based on this behavior are
457: routinely incorporated into stellar models that include rotation (e.g.
458: Pinsonneault et al 1989). Two of these three powers of $\Omega$ appear to be
459: related to the strength of the magnetic field of the star under the assumption
460: of a linear dynamo.}.
461:
462: %{\bf Two deficiencies of the Skumanich work that impact age determination: }\\
463: Skumanich (1972), however, did not specify the mass-dependence of rotation
464: - the so-called `correction for color' that he performed. Presumably this
465: correction was based on the Kraft (1967) curve or something similar.
466: There is also a measurement issue - for individual stars the ambiguity inherent
467: in using $v \sin i$ measurements, with the generally unknown angles of
468: inclination, $i$, can be expected to introduce large errors in the age
469: determinations\footnote{Projection effects are less relevant for entire
470: clusters, as with the averaged $v \sin i$ measurements that Skumanich used.
471: Presumably they average out because they are similar from cluster to cluster.}.
472:
473: %{\bf Added difficulty to second point: }\\
474: Furthermore, mass-dependent comparisons of rotation require precise values for
475: stellar radii to infer the true angular rotation speeds of stars.
476: Despite these shortcomings, various studies have occasionally used this
477: relationship for rotational ages, e.g. Lachaume et al (1999), and the ages
478: derived in this manner are in rough agreement with ages derived
479: using other techniques, but they are not noticeably better.
480:
481: %{\bf Kawaler's (1989) attempt at progress: }\\
482: Kawaler (1989) attempted an empirical color correction using a linear function
483: of the $(B-V)$ color of the star, but he provided no physical basis for such a
484: correction - indeed, there is none - and in any case it breaks down dramatically
485: for late-F to early-G stars (see especially Fig.\,1 in his paper). The specific
486: ways in which stars of different masses spin down, whether young clusters obey
487: such a spindown or not, and how observations in young clusters are related to
488: field star observations is a continuing matter of debate and discussion.
489:
490: %{\bf Possibility of addressing both deficiencies: }\\
491: If it were possible to eliminate the ambiguity in $v \sin i$ observations
492: by finding the true angular rotation rates of stars, as is routinely
493: accomplished nowadays by measuring rotation periods\footnote{Rotation periods are measured by timing the modulation of either filtered starlight, which works well for young stars (e.g. Van Leeuwen et al. 1987), or that of the chromospheric emission (e.g. Noyes et al. 1984), which works for older stars. Either of these is obviously more demanding than deriving $v \sin i$, but the effort is well worth the results, and furthermore, is being done routinely, as detailed below. As an aside, we point out that the ``rotation periods'' listed by Wright et al. (2004) are {\it not} directly measured; they are calculated from the measured chromospheric emission, and hence unsuitable for our purposes.},
494: and if the periods were to have a unique and ``correctable'' dependence on
495: color, with reasonably small scatter, rotation could become incredibly useful
496: as a stellar clock.
497:
498: %{\bf How to address these deficiencies: }\\
499: Using the (measured) rotation periods of the Mt.\,Wilson stars, Barnes (2001)
500: showed that
501: the age dependence of rotation for these stars is indeed Skumanich-type
502: ($ P \propto \sqrt{t}$), and furthermore, the mass dependence of rotation for
503: these stars is similar to that observed in the Hyades open cluster.
504: Barnes (2003a) noted that an age-increasing fraction of open cluster
505: stars and essentially all solar-type stars beyond a few hundred Myr in age,
506: including individual field stars, obeyed the same mass dependence. These two
507: facts provide the connection between rotation in clusters and in the field.
508:
509: %{\bf The meaning of the function: }\\
510: Furthermore, Barnes (2003a) wrote down this mass dependence, $f$, as a
511: convenient function\footnote{He used the the function: $f(B-V) = \sqrt{B-V-0.5} - 0.15 (B-V-0.5)$ but $f$ can of course be written in terms of any convenient function of stellar mass. We will modify the expression for $f$ below.} of $(B-V)$ color, $f(B-V)$.
512: This function, $f$, appears to be closely related to the moment of inertia,
513: $I_*$, of the entire star via $f \propto 1/\sqrt{I_*}$.
514: This identification, the rotational implications for the Sun and cluster stars,
515: and for stellar magnetic fields, are discussed at length in Barnes (2003a and
516: 2003b), but here we are concerned only with the
517: {\it universality and uniqueness} of this function, apparently separable from
518: the age dependence, a circumstance that leads to a remarkably simple way of
519: deriving ages (and their errors) for solar-type stars on the main sequence\footnote{I have learned from Ed Guinan (2006, personal communication) that he has been using the Hyades rotational sequence and the Skumanich relation to derive stellar ages. That would make it substantially similar to the technique developed here.}.
520:
521: \subsection{Proximate motivations for constructing a rotational clock}
522:
523: There are also proximate motivations for this work. It has become
524: increasingly obvious that greater precision in stellar ages than is available
525: using isochrones and chromospheric emission is required for many astronomical
526: purposes. The effort currently being expended on the host stars of planetary
527: systems is a case in point. Well determined ages would eventually permit the
528: study of the the evolution of planetary systems. This application is a proximate
529: one relevant to our time, but the method can undoubtedly be used to tackle
530: some of the deeper problems in astronomy.
531:
532: %{\bf Ground-based programs: }\\
533: The requirement of a stellar rotation period is not as onerous as might
534: initially appear\footnote{We note here that chromospheric emission measurements
535: also require repeated measurement to ensure that they are averages over the
536: variability from rotation or from stellar cycles.}.
537: {\it As opposed to the requirement for isochrones, it avoids the necessity of deriving the distance to a field star.}
538: The Vanderbilt/Tennessee State robotic photometric telescopes (e.g. Henry et
539: al. 1995) and of the University of Vienna (e.g. Strassmeier et al. 2000) in
540: Southern Arizona are designed to derive stellar rotation periods, and in fact,
541: the Strassmeier group, now in Potsdam, has almost finished the construction of
542: two 1.2m telescopes, Stella\,1\,\&\,2, to monitor active stars almost
543: exclusively (Strassmeier, 2006).
544: The ASAS project (e.g. Pojmanski, 2001) routinely monitors and catalogs stellar
545: (and other) variability in the Southern hemisphere, and a Northern counterpart
546: is the Northern Sky Variability Survey (Wozniak et al. 2004).
547:
548: %{\bf Foreign space missions: }\\
549: The Canadian MOST satellite (Matthews et al. 2000) was launched to provide
550: (and has since delivered) superb time-series photometry (witness its
551: identification of two closely spaced rotation periods for $\kappa^1$\,Ceti,
552: corresponding to two spot groups; (Rucinski et al. 2004), its detection of
553: 0.03\%-0.06\% brightness variations in a subdwarf B star; (Randall et al. 2005),
554: and its recent identification of g-modes in $\beta$ C\,Mi (Saio et al. 2007)).
555: The COROT satellite mission has been designed\footnote{In fact, the satellite has been built and launched.} to study stellar convection, rotation, and now, planetary transits.
556: A number of ground-based telescopes are planning to or already exploiting the
557: time domain, and of these the Large Synoptic Survey Telescope (LSST)
558: perhaps has the greatest visibility.
559:
560: %{\bf US space missions: }\\
561: The Kepler space mission, being readied for launch, is likely to yield not
562: only the planetary transit, but also the rotation period of the host star.
563: In fact, Kepler is likely to yield rotation periods for orders of magnitude
564: more stars than planetary transits\footnote{Assuming that we do not throw the baby out with the bath water.}. Regardless of whether or not the Kepler mission
565: delivers what it promises, stellar rotation periods will be determined
566: routinely as time domain astronomy comes into its own.
567: A very significant portion of time-domain work on stars will yield the stellar
568: rotation period (it is a by-product of all searches for planetary transits),
569: and if this measurement can be used to derive a precise stellar age, it would
570: permit us to address many problems involving chronometry that are not presently
571: solvable.
572:
573: \subsection{Overview of the paper and sequence of succeeding sections}
574: %{\bf Bottom line for this paper }\\
575: Our goal here is to specify the stars for which gyrochronology can and cannot
576: be used, to develop it to yield useful ages for individual field solar-type
577: stars, and to calculate the errors on these ages.
578: We will also show that where both are available, these new ages agree with
579: (and might even supercede) the ages provided by other methods.
580:
581: We begin by showing that rotating stars, whether in clusters or
582: in the field, are of two types\footnote{In principle, there is a third type, g,
583: representing stars in transition from the first/C to the second/I type.}:
584: fast/Convective/C and slow/Interface/I.
585: The Sun is shown to be on the interface sequence, which defines the rotational
586: connection between all solar-type stars (section\,2).
587: These stars are shown to spin down Skumanich-style,
588: with a mass dependence that
589: is shown to be universal, and for which we derive a simple functional form using
590: stars in open clusters (section\,3).
591: These functional dependencies are combined to yield a simple expression for the
592: gyro ages of stars.
593:
594: In section\,4, we derive the errors on these ages.
595: %, from the data themselves, and in analytical form.
596: %(We note that other chronometric techniques in common use do not.)
597: Section\,5 demonstrates that these ages compare favorably with chromospheric
598: ages for a well-studied sample.
599: Sections\,6 and 7 illustrate the use of gyrochronology on samples for which
600: other ages are not uniformly available, namely the field star samples of
601: Strassmeier et al. (2000) and Pizzolato et al. (2003).
602: Section\,8 demonstrates that gyrochronology yields the same age for the two
603: component stars of wide binaries.
604: Section\,9 contains a comparison to recently derived isochrone ages for a common
605: subset of the stars considered here,
606: and Section\,10 contains the conclusions.
607:
608:
609:
610: %\newpage
611: \section{The rotational connection between all solar-type stars}
612:
613: %(We show here that the sequences are real, particularly the I sequence)
614:
615: %{\bf Introduction: }\\
616: A fundamental fact of stellar rotation is that there are two major varieties,
617: C \& I, of rotating solar-type (FGKM) stars (see Barnes 2003a).
618: A third variety, g, merely represents stars making an apparently
619: unidirectional transition from one variety (C) to the other (I).
620: All three varieties of stars are normally found in young open clusters,
621: but the Sun and all old solar-type stars are of only the I variety.
622: Each of these varieties of rotating stars has separate mass- and age
623: dependencies
624: that can be clarified considerably merely by effecting the correct
625: separation of the stars by variety. One of these, called the Interface (I)
626: sequence stars, containing the Sun and all old field solar-type stars,
627: is related to the property Skumanich noticed in 1972.
628: This group is the one that we will use here to demonstrate the technique
629: of gyrochronology, because stars change into this variety over time.
630: We are fortunate that the rotational mass- and age dependencies of this
631: group of stars appear to be both separable and also particularly simple.
632:
633: %{\bf Basic features of the Interface sequences of the rotational distributions of stars - only the identification, no interpretation: }\\
634: If the mass- and age dependencies of this sequence are indeed separable,
635: as was claimed by Barnes (2003a) to be of the form
636: $ P (t, M) = g(t) . f(M) $,
637: then merely dividing the measured rotation periods $ P (t, M) $ by the
638: functional form $ g(t) $ of the age dependence should make the mass
639: dependence $f(M)$ manifestly clear.
640: For observational convenience, and also to avoid the error inherent in the
641: conversion from $B-V$ to stellar mass, we have used $f(B-V)$ instead of $f(M)$.
642: Removing an assumed Skumanich-type age dependence, where $ g(t) = \sqrt{t} $,
643: is particularly simple, and appears to bring the I sequence into sharp focus,
644: leading to the identification of the mass dependence as a function
645: $ f = f(B-V) $.
646: In the two subsections below, we effect this determination separately for
647: cluster and field stars.
648:
649:
650: \subsection{The connection between clusters themselves}
651:
652: %{\bf Preliminaries: }\\
653: Here we show that $f$ represents the connection between most rotating stars,
654: and that the functional dependence
655: of $f$ on color or stellar mass is common to all open clusters.
656: We use all the open cluster rotation periods currently available in the
657: literature; note that we are restricted to those stars for which $(B-V)$
658: colors are also available. The major sources are listed in Table 2.
659: We divide each of the measured rotation periods by $g(t)=\sqrt{t}$
660: where $t$ is the age of the cluster in Myr, as listed in Table\,2.
661: These quantities are plotted against de-reddened $(B-V)$ color
662: in Fig.\,1 , on a linear scale in the upper panel, and on a logarithmic
663: scale in the lower panel.
664: Clusters are color-coded violet through red in increasing age sequence.
665:
666: %Fig.\,1 $P/\sqrt{t}$ vs $(B-V)_0$ for only the open clusters
667:
668: \clearpage
669:
670: \begin{deluxetable}{lrl}
671: \tabletypesize{\scriptsize}
672: \tablecaption{Principal sources for open cluster rotation periods. \label{tbl-2}}
673: \tablewidth{0pt}
674: \tablehead{
675: \colhead{Cluster} & \colhead{Age} & \colhead{Rotation Period Source}
676: %\colhead{$\Theta$\tablenotemark{b}}
677: }
678: \startdata
679: IC 2391 & 30\,Myr & Patten \& Simon (1996)\\
680: IC 2602 & 30\,Myr & Barnes et al. (1999)\\
681: IC 4665 & 50\,Myr & Allain et al. (1996)\\
682: Alpha\,Per & 50\,Myr & Prosser \& Grankin (1997)\\
683: Pleiades & 100\,Myr & Van Leeuwen, Alphenaar \& Meys (1987),
684: Krishnamurthi et al. (1998)\\
685: NGC\,2516& 150\,Myr & Barnes \& Sofia (1998)\\
686: M\,34 & 200\,Myr & Barnes (2003a)\\
687: NGC\,3532& 300\,Myr & Barnes (1998)\\
688: Hyades & 600\,Myr & Radick et al. (1987)\\
689: Coma & 600\,Myr & Radick, Skiff \& Lockwood (1990)\\
690: \enddata
691:
692: %% Text for table notes should follow after the \enddata but before
693: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
694: %% in the table for each \tablenotetext.
695:
696: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
697: %with the deluxetable environment}
698: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
699:
700: %\tablecomments{Occasionally, authors wish to append a short
701: %paragraph of explanatory notes that pertain to the entire table, but
702: %which are different than the caption. Such notes should be placed in
703: %a {\tt tablecomments} command like this.}
704:
705: \end{deluxetable}
706:
707:
708: %{\bf Basic morphology: }\\
709: % {\bf I sequence: }
710: The most striking aspect of these data is the curvilinear feature representing
711: a concentration of stars in the vicinity of the solid line.
712: This is the interface sequence, I, proposed in Barnes (2003a), the one that
713: will consume our attention in this paper,
714: and whose position we will use as an age indicator for field stars.
715: %\\{\bf C sequence: }
716: Along the bottom of the upper panel one may also discern another linear
717: concentration of stars which represents the convective sequences, C, of the
718: youngest open clusters. This sequence could also potentially be used as
719: an age indicator for young stars, but its dependencies on stellar age
720: and mass are more complicated than those of the I sequence (see Barnes 2003a),
721: and we do not use it here.
722: %\\{\bf Gap: }
723: Stars located between these sequences are either on the convective sequences
724: of the older open clusters in this sample, or in the rotational gap, g, between
725: the interface and convective sequences.
726:
727: %{\bf Detailed Morphology (emphasis on Interface sequence): }\\
728: Every single cluster plotted in Fig.\,1 possesses an identifiable interface
729: sequence.
730: The fraction of stars on this sequence increases systematically with cluster
731: age, as shown earlier in Barnes (2003a); see especially Fig.\,3 there.
732: But for us the crucial feature of these data is that these age-corrected
733: sequences overlie one another. This feature is shared by all open
734: clusters, and can be represented by a function $f(B-V)$, common to
735: all clusters. A particular choice (used in Barnes 2003a) of
736: $f(B-V): \sqrt{(B-V-0.5)} - 0.15 (B-V-0.5)$ is displayed in both panels.
737: This is of the nature of a trial function, useful in locating the I sequence
738: roughly, and we will improve on this choice subsequently.
739:
740: %{\bf Aside: }\\
741: Barnes (2003a) has suggested that $f$ ought to be identified with
742: $1/\sqrt{I_*}$, where $I_*$ is the moment of inertia of the star, implying a
743: substantial mechanical coupling of the entire star on this sequence.
744: The suggestion in that publication was magnetic coupling by an interface
745: dynamo, hence the name interface sequence for this group of stars.
746:
747: %{\bf Comments on the slow rotators: }\\
748: The dotted lines in the figure are drawn at $2 f$, and $4 f$.
749: Present indications are that some of these stars are either non-members of
750: the cluster, sometimes stars with spurious/alias periods or otherwise
751: misidentified variables of another sort.
752: %For example, the slow rotators in NGC\,3532 are probably not members of the
753: %cluster, but merely fore/background stars.
754: % But the Pleiades has some slow rotators too...
755: %The situation in other clusters could be similar.
756: Note that the Hyades, where excellent membership information is available,
757: has no stars above the sequence.
758: A similar situation obtains in NGC\,2516 and M\,34, which are also relatively
759: clean samples. Good cluster membership information could resolve this issue
760: completely.
761:
762: %{\bf Summary of commonality between clusters: }\\
763: In summary, the behavior of the open cluster rotation observations suggests
764: the existence of a feature common to all open clusters, the Interface sequence,
765: which is observationally definable by its common mass
766: dependence, $f(M)$, across clusters, here represented by $f(B-V)$.
767: These observations also justify the use of the Skumanich (1972) relationship
768: between rotation and age to describe the age dependence of rotation, but
769: {\it only for rotating stars of this particular (interface) type}.
770:
771:
772: \subsection{The connection between clusters and field stars}
773:
774: %{\bf Preliminaries: Split sample into Y/O, big scatter, ages : }
775: Here we show that the mass dependence, $f$, among open clusters, is also
776: shared by field stars as exemplified by the Mt.\,Wilson stars.
777: We begin by removing from the Mt.\,Wilson sample those stars known or suspected
778: not to be dwarfs (based on Baliunas et al. 1995
779: %{\bf Also, check these in Jeff's new paper!}
780: ), to avoid any possible complexity
781: related to structural evolution off the main sequence.
782: An effective connection with open clusters requires splitting the remaining
783: main sequence Mt.\,Wilson field star sample by age, to control the age
784: variation among the stars and gain leverage over the time domain.
785: Fortunately, one such split, based on detailed studies of this sample,
786: especially of chromospheric emission, has already been made by Vaughan (1980),
787: who classified these stars into a Young (Y) and Old (O) group.
788: The simplest course of action is to use the existing divisions.
789: Although the age divisions are quite broad, subsequent work has confirmed
790: the basic classification. A cut by chromospheric activity is well-known
791: to be also a cut by rotation and age
792: (e.g. Barnes 2001, and references therein).
793: Part of the goal of this paper is to develop a way of ordering the stars by
794: age, so we cannot start by assuming chromospheric ages for individual stars.
795:
796: %{\bf 2 Groups of stars, and their ages: }\\
797: As a result of the above classification, we have two groups of stars, Y and O,
798: consisting of 43 and 49 stars respectively, equivalent to two additional open
799: clusters, each containing stars with a wide range of ages.
800: What are these ages? Barnes (2001) [see especially Fig.\,3] and Barnes (2003a)
801: [see especially Fig.\,2] suggest that, in terms of rotation,
802: the young (Y) stars range in age from less than 300\,Myr to about 2\,Gyr,
803: with a characteristic age of 800\,Myr, while the old (O) stars range in age
804: from 2\,Gyr to about 10\,Gyr, with a characteristic age of 4.5\,Gyr.
805: The age of the Y group is older than, but comparable to, young open cluster
806: ages,
807: while the age of the O group is reasonably represented by the Sun's age.
808: Effectively, we are assuming that the Sun is an appropriate representative,
809: in rotation and age, of the Old Mt.\,Wilson sample.
810:
811: %{\bf Sanity/consistency check: }\\
812: If we use the chromospheric ages for the same Y and O groups of Mt.\,Wilson
813: stars, calculated using the relationship of Donahue (1998),
814: the median ages of the same samples work out to be 780\,Myr and 4.24\,Gyr
815: respectively, reasonably close to our assumption above.
816: We will use these new values as the representative ages for the Y and O groups
817: in this paper.
818: The rotation clock can easily be recalibrated when the need arises.
819:
820: %{\bf Generation of Fig.\,2 and Fig.\,3: }\\
821: We can make the field star data comparable with the open cluster data by
822: similarly removing this approximate age dependence. Thus, we divide the
823: rotation periods of the Young Mt.\,Wilson stars by $\sqrt{t_Y=780\,Myr}$
824: and display them using small black asterisks in Fig.\,2, overplotted on the
825: open cluster data (colored circles). Similarly, we divide the rotation
826: periods of the Old Mt.\,Wilson stars by $\sqrt{t_O=4240\,Myr}$ and display
827: them in Fig.\,3 using large black asterisks, again overplotted on the open
828: cluster data.
829: %\\{\bf Consider combining Figs. 2 and 3 into one multi-panel figure}
830:
831: %{\bf Results from Fig.\,2 and Fig.\,3: }\\
832: Examination of Figs. 2 and 3 leads to several conclusions.
833: Firstly, we note that both the Young and Old samples overlie the Interface
834: sequences of the open clusters. The greater dispersion of the Y and O stars
835: relative to those of the open cluster I sequences can be traced to the age
836: dispersion in each of these samples. We can see, despite this dispersion,
837: that C sequence stars, and possibly g (gap) stars, are absent from both the
838: Young and Old samples. These data are consistent with all of the Mt.\,Wilson
839: stars being of the I variety.
840:
841: %{\bf Individual (chromospheric emission based) corrections for Mt.\,Wilson stars: }\\
842: Any doubts about the classification of the Young and Old Mt.\,Wilson stars can
843: be settled by making individual corrections for these stars based on their
844: chromospheric ages. If these ages are correct, then removing their dependence,
845: as in the open clusters, should make the mass dependence obvious, and that
846: mass dependence ought to be similar to $f$.
847: In fig.\,4 we display the result of dividing the Mt.\,Wilson star rotation
848: periods by the square root of the (individual) chromospheric ages, calculated
849: using the formula from Donahue (1998)\footnote{This formula yields ages in close agreement with those for old stars calculated using the formulae in Soderblom, Duncan \& Johnson (1991), but is generally considered to be an improvement for young stars because saturation effects are taken into account [cf. Barnes 2001].}.
850: We note that almost all of the Mt.\,Wilson stars in the color range considered
851: lie on/near $f(B-V)$.
852: Fig.\,4 displays $f(B-V)$ (solid line), $0.8f$ and $1.25f$ (dotted lines), to
853: show this proximity.
854: Indeed, a free-hand fit would be almost identical to $f$.
855: We will improve on this trial function below.
856: It is likely that this sample does not contain any C sequence stars or even any
857: gap stars.
858: (This observation is consistent with the C$\rightarrow$I transition timescale
859: of $\sim$200\,Myr observed in open clusters.)
860:
861: %{\bf Commonality of cluster and field star rotation: }\\
862: In closing this section, we reiterate that the rotation period distributions
863: of open clusters, when corrected for a Skumanich-type age dependence, display
864: two strong concentrations of stars. The slower of these consists of sequences
865: that are common to all open clusters and overlie one another.
866: Rotation period distributions of main sequence field stars, despite the
867: difficulty of correcting for their ages, also display this same sequence.
868: The other concentration of stars, present in open clusters, is absent here.
869: The feature common to cluster and field stars, called the Interface sequence,
870: can be fit by a function (as we do below),
871: giving the mass dependence of stellar rotation.
872:
873: %{\it Thus, this section has demonstrated that a rotational clock can be
874: %constructed, and we now proceed to construct it.}
875:
876:
877:
878: %\newpage
879:
880: \section{(Re-)determination of the mass- and age dependencies}
881:
882: %{\bf Philosophy of the act: }\\
883: Having determined that rotation has both mass- and age dependencies, how is one
884: to specify them independently using one set of data, and without greatly
885: compromising the determined dependencies?
886: One way forward is to realize that open clusters can specify the mass dependence
887: regardless of whether or not we make some error in their ages - after all, they
888: are all clustered near ZAMS ages - while the Sun provides a datum with a very
889: well-defined age far out, but obviously no information about the mass
890: dependence. These facts suggest the use of open clusters to decide the
891: mass dependence, and the Sun to decide the age dependence.
892: The effect is to follow the Copernican Principle and assume that the Sun is
893: the perfect representative of its class of star.
894: (We note that the same principle guides the solar calibration of the
895: classical isochrone method.)
896:
897: %{\bf Introductory paragraph: }\\
898: The construction of an appropriate fit for the mass dependence requires the
899: removal of stars that are not on the I\,sequence in open clusters.
900: %\\{\bf Rewritten: }\\
901: This cannot yet be done unambiguously using only color-period data because the
902: position of the I\,sequence has yet to be specified well.
903: That is part of the goal of this paper.
904: For clusters where X-ray data are also available, we get a additional handle on
905: classifying these stars using the correspondence noted in Barnes (2003b).
906: There, the classification in X-rays of unsaturated, saturated, and
907: super-saturated stars is shown to correspond, on a star-by-star basis with
908: I-, g-, and C\,stars respectively.
909: We therefore select the unsaturated stars, which are all I\,sequence stars in
910: the color-period diagram, and for each cluster, we plot $P/\sqrt{Cluster\,Age}$
911: against $(B-V)_0$. These stars define a sequence in color-$P/\sqrt{Age}$ space,
912: and we can now discard stars from the other clusters without X-ray information
913: that lie far away from this sequence. The aim is to do this conservatively, so
914: as to retain as many stars as possible for a proper definition of the
915: I\,sequence, while removing clear C-, g-, or alias period stars.
916: While it is true that this determination is done subjectively at present, it is
917: done as empirically as we possibly can at the present time\footnote{Judgements such as these are routinely made during classical isochrone fitting. A rich dataset or two, such as the one for M35 (S. Meibom, in prep) should eliminate much of the ambiguity within a year or two.}.
918: The remaining stars happen to lie near the trial function $f(B-V)$ that was
919: used in Barnes (2003a), but this function does not obviously pre-determine the
920: new one.
921:
922: %{\bf Details: }
923: This exercise suggests that slight modifications to the open cluster ages are
924: needed to tighten the overlap of the individual I sequences. We have made these
925: slight adjustments in order to ensure a valid result for the mass dependence.
926: The ages used are:
927: IC\,4665: 40Myr;
928: Alpha\,Per: 110Myr;
929: Pleiades: 120Myr;
930: NGC\,2516: 180Myr;
931: M\,34: 200Myr;
932: NGC\,3532: 250Myr;
933: Coma\,Ber: 600Myr;
934: Hyades: 600Myr.
935: We have not used IC\,2391 and IC\,2602 because although they possess
936: identifiable sequences, they are some distance off the sequence defined by
937: the other clusters, a fact we attribute to the residual effects of pre-main
938: sequence evolution\footnote{The 110\,Myr age for Alpha\,Per might also be a surprise to some. In fact, we guess that the underlying rotational behavior might also originate in residual effects from pre-main sequence evolution, similar to IC\,2391 \& IC\,2602. However, we have chosen to retain it in this analysis because we cannot yet afford to lose the many periods in this cluster (contributed by Prosser \& Grankin 1997).}.
939: These minor age adjustments are justifiable because, in any case, we are not
940: using the open clusters to decide the age dependence of rotation. We are
941: effectively merely using them to set the ``zero-point'' of the age dependence.
942: We know that their I sequence age dependence is roughly Skumanich-style.
943:
944: %{\bf Doing the fit}\\
945: Having removed the non-I sequence stars, we note a tight mass-dependence for
946: which we desire a functional form. The trial function
947: $f(B-V)= \sqrt{(B-V-0.5)} - 0.15(B-V-0.5)$ has an undesirable singularity at
948: $(B-V)=0.5$, which we would like to move blueward, to accommodate the late F
949: stars. We would also like to retain an analytic function.
950: A function of the form $f(B-V)=a.(B-V - 0.4)^{b}$ where $a$ and $b$ are
951: fitted constants, seems to be appropriate (and will permit appropriate error
952: analysis later).
953: Using the R statistics package (Ihaka \& Gentleman, 1996) to do the fit, we get
954: $a=0.7725 \pm 0.011 $ and $b=0.601 \pm 0.024 $.
955: This function is plotted with a solid line in Fig.\,5 over the I sequence
956: stars in the open clusters listed above. The standard error on the residuals
957: is $0.0795$ on 182 degrees of freedom.
958:
959: %{\bf Fit is appropriate}\\
960: To show that the fit is appropriate, we also display using a dashed line in
961: Fig.\,5 the result of fitting a non-parametric trend curve using the function
962: {\it lowess} in the R statistics package\footnote{The {\it lowess} function implements a locally weighted regression smoothing procedure using a polynomial. No significant difference is seen with other smoothing procedures.}.
963: The close correspondence between the
964: two curves shows that the function chosen above is appropriate for these data.
965:
966: %{\bf Figure with the newly-determined mass dependence: }\\
967: Having determined the mass-dependence using open clusters, we check that it is
968: appropriate for the field stars, which provided the motivation for improving
969: the representation of $f(B-V)$. We plot the new and old dependencies in
970: Fig.\,6, over the the Mt.\,Wilson stars (same data as in Fig.\,4), and again
971: {\it assuming that the chromospheric ages are correct}.
972: The figure displays the difference between the old and new functions, $f$, in
973: relation to the Mt.\,Wilson stars. (This discrepancy between $f$ and the
974: F\,star data is partially attributable to the assumption of correct
975: chromospheric ages for blue stars and is addressed in another section below.)
976:
977: %{\bf Now tackle the age dependence}\\
978: Having specified the mass dependence using open clusters, and having shown that
979: the Sun and field stars also follow this mass dependence, we can now determine
980: the age dependence. We know that the age dependence $g(t)$ will roughly be
981: $\sqrt t$, but the open clusters are too young to be effective calibrators,
982: nor are their ages known to sufficient precision.
983: In contrast, the rotation rate of the Sun is perhaps the most fundamental datum
984: in stellar rotation, and its parameters are the fundamental calibrators for
985: theoretical stellar models.
986: In keeping with this tradition, (and older ones of calibrating clocks by the
987: sun), we choose to specify the age dependence via a solar calibration.
988: Representing the age dependence using $g(t) = t^n$, and calibrating the index
989: $n$ using the Sun's measured mean rotation period of 26.09d
990: (Donahue et al. 1996), a solar $B-V$ color of 0.642 (Holmberg et al. 2005) and
991: a solar age of 4.566 Gyr (Allegre et al. 1995) yields $n=0.5189 \pm 0.0070$,
992: where the error on $n$ has been calculated by simply propagating the errors on
993: the other terms and assuming 1\,d and 50\,Myr errors in the period and age of
994: the sun respectively. This calculation is detailed in the appendix to this
995: paper.
996:
997: %{\bf Final result}\\
998: So, the final result works out to be:
999: $P(B-V,t) = f(B-V) . g(t)$, where
1000: \begin{equation}
1001: {
1002: f(B-V) = (0.7725 \pm 0.011) \times (B-V_0 - 0.4)^{0.601 \pm 0.024}
1003: }
1004: \end{equation}
1005: and
1006: \begin{equation}
1007: {
1008: g(t)= t^{0.5189 \pm 0.0070}
1009: }
1010: \end{equation}
1011: % n=0.51889 to get to the last decimal place.....
1012: a result which is simultaneously analytical, simple, separable, almost
1013: Skumanich, fits the mass dependence of the open clusters, and the age
1014: dependence specified by the Sun.
1015:
1016: %{\bf Therefore, the gyro age is simply given by the formula ....}\\
1017: We know that for the I sequence stars, whether in clusters or in the field,
1018: the rotation rate is given by $P(B-V,t) = f(B-V).g(t)$, where $f(B-V)$ and
1019: $g(t)$ were determined above. This is true for each star.
1020: Therefore $t = g^{-1}[P(B-V,t)/f(B-V)]$.
1021: Explicitly,
1022: \begin{equation}
1023: log(t_{gyro}) = \frac{1}{n}\{log(P)-log(a)-b \times log(B-V-0.4)\}
1024: \end{equation}
1025: where $t$ is in Myr, $B-V$ and $P$ are the measured color and rotation period
1026: (in days) respectively,
1027: $n=0.5189 \pm 0.007$, $a=0.7725 \pm 0.011$, and $b=0.601 \pm 0.024$.
1028:
1029: %{\it We have now set our clock by the Sun,
1030: %and below we show that it keeps time well.}
1031:
1032:
1033:
1034: %\newpage
1035: \section{Errors in the ages}
1036:
1037: %{\bf Intro to thinking about errors: }\\
1038: The ages from gyrochronology become truly useful only when we can estimate their
1039: errors and show that they are acceptable.
1040: A crude estimate of the error is simply the spread in the function $f(B-V)$.
1041: \begin{equation}
1042: {
1043: \frac{\delta t}{t} = \frac{1}{n} \frac{\delta f}{f} \approx 2\frac{\delta f}{f}
1044: }
1045: \end{equation}
1046:
1047: An estimate for $\delta f$ is the standard error of the residuals from the fit
1048: to $f$, which we have derived using R, and which is $0.0795$ on
1049: $182$ degrees of freedom. ($f$ itself as shown above, is of course known much
1050: better because of the number of points involved.) For a K star, at $B-V = 1$,
1051: roughly the middle of our distribution, $f = 0.57$, so that
1052: \begin{equation}
1053: {
1054: \frac{\delta t}{t} \approx 28\%
1055: }
1056: \end{equation}
1057: The errors in $f$ are heteroscedastic, as can be seen from Fig.\,5, and on the
1058: reasonable assumption that they scale with $f$, we can simply adopt this
1059: value of 28\% error in the ages for all G, K, \& early M stars.
1060: This gives a representative number, but a uniform adoption of this error
1061: overestimates the age error for our stars\footnote{This can be traced to the
1062: generous limits adopted in the present instance among the open clusters for
1063: inclusion as an I\,sequence star, resulting in considerable contamination from
1064: incorrect rotation periods, g- and perhaps even C\,stars. This contamination
1065: results in a large scatter in $f$, but $f$ itself is defined much better
1066: because of the large number of data points involved. Further work is needed in
1067: open clusters to clarify this matter.},
1068: and masks the underlying variations, which we elucidate below.
1069:
1070:
1071: \subsection{Derivation of errors}
1072:
1073: We begin with the representation
1074: \begin{equation}
1075: {
1076: P = f(B-V).g(t)
1077: }
1078: \end{equation}
1079: where $P, B-V$, and $t$ are the period, color and age of the star respectively,
1080: and $f$ and $g$ are the color and age dependencies, as before.
1081: Taking logs and differentiating, we get
1082: \begin{equation}
1083: {
1084: \frac{dP}{P} = \frac{df}{f} + \frac{dg}{g}
1085: }
1086: \end{equation}
1087: Now, $g(t) = t^n$, so $dg/g = n dt/t + ln\,t\,dn$, where $n \approx 0.5$. Thus,
1088: \begin{equation}
1089: {
1090: \frac{dP}{P} = \frac{df}{f} + n\frac{dt}{t} + ln\,t\,dn
1091: }
1092: \end{equation}
1093: Now, $f(B-V) = a x^b$, where $x=B-V-0.4$ and $a$ and $b$ are fitted constants
1094: (with associated errors).
1095: Differentiating, $df/f = da/a + b\,dx/x + ln\,x\,db$. Thus,
1096: \begin{equation}
1097: {
1098: \frac{dP}{P} = \frac{da}{a} + b\frac{dx}{x} + ln\,x\,db + n\frac{dt}{t} + ln\,t\,dn
1099: }
1100: \end{equation}
1101: Substituting, re-arranging, and adding the errors in quadrature under the usual
1102: assumption of independence yields
1103: \begin{equation}
1104: {
1105: (n \frac{\delta t}{t})^2 = (ln\,t\, \delta n)^2 + (\frac{\delta P}{P})^2
1106: + (\frac{\delta a}{a})^2 + (b\frac{\delta x}{x})^2
1107: + (ln\,x\,\delta b)^2
1108: }
1109: \end{equation}
1110:
1111: %{\bf The P term: }\\
1112: The one term above that requires further attention is the period (P) term.
1113: There are two contributions to the period error - the measurement error, and
1114: differential rotation, which can be added in quadrature:
1115: $(\delta P/P)^2 = (\delta P_{msrmnt}/P)^2 + (\delta P_{dffrtn}/P)^2$.
1116: The period determination itself is not usually a great contributor to the error,
1117: but the differential rotation term could potentially be a deal-breaker.
1118: Donahue et al. (1996) concluded that the dependence was a simple function of
1119: the rotation period alone, and their results (see esp. Fig.\,3 there) suggest
1120: that the period range, $\Delta P = P_{max} - P_{min}$,
1121: can be represented simply by $ log\,(\Delta P) = -1.25 + 1.3\,log <P> $.
1122: The long baseline of their dataset suggests that $\Delta P$ corresponds to
1123: $2 \sigma$, so that the ($1 \sigma$) period error is simply a quarter of this:
1124: $log\,(\delta P_{dffrtn}) = -1.85 + 1.3\,log <P>$
1125: so that,
1126: \begin{equation}
1127: {
1128: (\frac{\delta P}{P})^2 = (\frac{\delta P_{msrmnt}}{P})^2 + (10^{-1.85}\,P^{0.3})^2
1129: }
1130: \end{equation}
1131:
1132:
1133: Substituting this in equation (10) gives
1134: \begin{equation}
1135: {
1136: (n \frac{\delta t}{t})^2 =(ln\,t\, \delta n)^2 + (\frac{\delta P_{msrmnt}}{P})^2
1137: + (10^{-1.85}\,P^{0.3})^2 + (\frac{\delta a}{a})^2
1138: + (b\frac{\delta x}{x})^2 + (ln\,x\,\delta b)^2
1139: }
1140: \end{equation}
1141:
1142: %{\bf Color errors and numerics: }\\
1143: Putting in some of the numerical values will allow us to understand the
1144: dependencies of the errors.
1145: From equation\,(2) - see appendix\,1 for the details - $n = 0.0519 \pm 0.007$.
1146: The error in the period determination for the Mt.\,Wilson stars is
1147: 0.25\%-1\%, 0.5\%-2\%, 2\%-4\% for periods less than 20d, greater than 20d,
1148: or periods between 30d and 60d respectively (Donahue et al. 1996). A 1\% error
1149: seems to be a reasonable representation for the samples considered here.
1150: For $\delta x = \delta (B-V) $, we adopt the value of 0.01 suggested by the
1151: precision of the datasets considered below. This might need to be increased to
1152: 0.02 for data acquired through CCD photometry (assuming independent errors of
1153: 0.015 in each filter), but we note that this error could be considerably lower
1154: for data acquired through photoelectric photometry.
1155: From section\,3 and equation\,(1),
1156: $a = 0.7725 \pm 0.011 $ and $b = 0.601 \pm 0.024 $.
1157: We input these values to get (in the same order as above)
1158: \begin{equation}
1159: {
1160: (n\frac{\delta t}{t})^2 = (0.007\,ln\,t)^2 + (0.01)^2 + (0.014\,P^{0.3})^2
1161: + (\frac{0.011}{0.7725})^2 + (0.6\frac{0.01}{x})^2
1162: + (0.024\,ln\,x)^2
1163: }
1164: \end{equation}
1165: or
1166: \begin{equation}
1167: {
1168: (n\frac{\delta t}{t})^2 = 10^{-4}[ (0.7\,ln\,t)^2 + (1)^2
1169: + (1.4\,P^{0.3})^2 + (1.424)^2
1170: + (\frac{0.6}{x})^2 + (2.4\,ln\,x)^2
1171: ]
1172: }
1173: \end{equation}
1174: or
1175: \begin{equation}
1176: {
1177: (n\frac{\delta t}{t})^2 = 10^{-4}[ \frac{1}{2}(ln\,t)^2 + 1 + (1.4\,P^{0.3})^2
1178: + 2 + (\frac{0.6}{x})^2 + (2.4\,ln\,x)^2
1179: ]
1180: }
1181: \end{equation}
1182: Thus,
1183: \begin{equation}
1184: {
1185: \frac{\delta t}{t} = 2\% \times \sqrt{3 + \frac{1}{2}(ln\,t)^2 + 2\,P^{0.6}
1186: + (\frac{0.6}{x})^2 + (2.4\,ln\,x)^2
1187: }
1188: }
1189: \end{equation}
1190: which shows that the age error is always greater than $\sim 11\%$ for conditions
1191: similar to those assumed here.
1192: (Recall that $t$ is in Myr, $P$ is in days, and $x=B-V_0-0.4$.)
1193: For 1\,Gyr-old stars of spectral types late\,F, early\,G, mid\,K and early\,M
1194: respectively, we get
1195: \begin{equation}
1196: {
1197: \frac{\delta t}{t} = 2\% \times
1198: \cases{
1199: \sqrt{26.9 + 6.4 + 66.5} & when $B-V=0.5$ ($P= 7d$);\cr
1200: \sqrt{26.9 + 8.9 + 16.9} & when $B-V=0.65$ ($P=12d$);\cr
1201: \sqrt{26.9 + 12.1 + 2.5} & when $B-V=1.0$ ($P=20d$);\cr
1202: \sqrt{26.9 + 15.4 + 0.35} & when $B-V=1.5$ ($P=30d$).\cr
1203: }
1204: }
1205: \end{equation}
1206: which shows the relative contributions of the period and color errors (second
1207: and third terms, respectively), or,
1208: \begin{equation}
1209: {
1210: \frac{\delta t}{t} =
1211: \cases{
1212: 20\% & when $B-V=0.5$ ;\cr
1213: 15\% & when $B-V=0.65$ ;\cr
1214: 13\% & when $B-V=1.0$ ;\cr
1215: 13\% & when $B-V=1.5$ .\cr
1216: }
1217: }
1218: \end{equation}
1219: %{\bf Final summation: }\\
1220: The behaviors of the function $f$ and of differential rotation are such that
1221: the color and period errors dominate for blue and red stars respectively to
1222: give a total error of $\sim$15\%.
1223: The errors calculated using equation (16) are the ones quoted for the
1224: gyrochronology ages in the remainder of this paper.
1225: Setting the $P$ and $B-V$ errors equal leads to a trancendental equation which
1226: separates the color-period space into two regions, a blue one where color
1227: errors dominate, and a red one where the period errors (mostly differential
1228: rotation) dominate. The separator is a steep function in color-period space,
1229: and is roughly at solar color.
1230:
1231: How well these errors represent the true errors of this technique future work
1232: will show. We simply note here that the very possibility of calculating the
1233: errors distinguishes gyrochronology from other stellar chronometric methods.
1234:
1235: %{\it We have shown that the rotational clock runs well,
1236: %and now we proceed to test it.}
1237:
1238:
1239:
1240: %In summary, we have set our clock by the Sun, and shown that it keeps time well.
1241:
1242:
1243: %\newpage
1244: \section{Comparison with the chromospheric clock}
1245: %{ages using the Mt.\,Wilson stars}
1246:
1247: %{\bf Motivation: }\\
1248: Before calculating gyro ages for stars where other ages are not available, it
1249: is necessary to consider whether these ages agree at least roughly with others
1250: that might be available.
1251: We stress here that the ages derived through gyrochronology in this paper are
1252: {\it independent of other techniques}, except for the calibration using the Sun,
1253: whose age is determined using radioactivity in meteorites (see prior section).
1254: In particular, these ages are independent of chromospheric and isochrone ages,
1255: except for the common solar calibration point\footnote{The calibration issue is
1256: discussed later in this section.}.
1257:
1258: %{\bf Choice of test sample: }\\
1259: Potentially, the best way to test these ages would be to derive ages for open
1260: clusters where other ages are also available. This is not possible in this work
1261: because the open clusters have been used here to derive the mass dependence of
1262: gyrochronology, and this required a prior knowledge of their ages.
1263: Additional data will allow such a test in the future\footnote{In fact, two are underway using new data in M\,34 (James et al. 2007) and M\,35 (Meibom et al., in prep.)}.
1264: Isochrone ages for main sequence field stars are not reliable enough to serve
1265: as a test. (Section\,9 elaborates on this.)
1266: What is possible is a test against chromospheric ages for field stars.
1267:
1268: %{\bf The chromospheric sample: }\\
1269: Despite the obviously large errors associated with the method (see below),
1270: chromospheric ages have thus far been considered to be the best ones available
1271: at present for single field stars.
1272: Furthermore, there exists a substantial and uniform sample, the
1273: Mt.\,Wilson stars, for which the chromospheric emission is known very well
1274: (over decades), for which the chromospheric ages are believed to be relatively
1275: secure, and for which {\it measured} rotation periods are also
1276: available\footnote{There exists another sample of 19 Southern stars for which
1277: chromospheric emission (from Henry et al. 1996) and rotation periods are
1278: both available. These overlap with another sample of stars discussed here in
1279: section\,6 below, and the corresponding comparison is presented there.}.
1280: These facts allow us to compare the ages from the two (independent)
1281: techniques below.
1282:
1283: \subsection{How is the chromospheric age of a star calculated?}
1284:
1285: There has been considerable work on the determination of the rate of
1286: decay of chromospheric emission with age since the results of Skumanich (1972).
1287: The two sources generally quoted for a relationship between chromospheric age
1288: and $R'_{HK}$ are Donahue (1998) and Soderblom et al. (1991).
1289: Although we will end up using the former to calculate chromospheric age,
1290: it is necessary to discuss both to understand the relevant issues.
1291: The key feature of both relationships is that once the measurement of stellar
1292: chromospheric emission has been made (repeatedly or not),
1293: it can immediately be converted into an age, without additional information.
1294: (Wright et al. 2004 have shown subsequently that stars previously considered to
1295: be in Maunder Minima are in fact somewhat evolved, so caution is advisable
1296: with respect to the basic properties of the star.)
1297:
1298: \subsubsection{The Donahue (1998) relationship}
1299:
1300: %{\bf We use Donahue 1998 (despite its problems): }\\
1301: The relationship given in Donahue (1998) is:
1302: \begin{equation}
1303: {
1304: log[t_{Chromospheric}] = 10.725 -1.334 R_5 + 0.4085 R_5^2 - 0.0522 R_5^3
1305: }
1306: \end{equation}
1307: where $R_5 = 10^5 R'_{HK}$ and the age, $t$, is measured in Gyr.
1308: This relationship is essentially identical to the one in Soderblom et al (1991)
1309: (discussed below) for ages greater than 1\,Gyr. The deviation between the two
1310: relationships pertains to younger stars, including those in the Hyades, Coma,
1311: Ursa Major, Pleiades, and NGC\,2264 open clusters, for which it claims a better
1312: age calibration. This particular feature has prompted us to use it instead of
1313: the Soderblom et al. (1991) relationship.
1314:
1315: %{\bf Its limitations: }\\
1316: However, it does have two serious limitations:
1317: Firstly, it does not provide errors on the ages so derived. (However, Donahue
1318: (1998) does list the discrepancies in chromospheric age for a number of wide
1319: binaries and triple systems. The mean discrepancy for the systems listed is
1320: 0.85\,Gyr on a mean age of 1.85\,Gyr, which suggests a fractional age error of
1321: $\sim$46\%, in rough agreement with the errors quoted in Soderblom et al. 1991.)
1322: Secondly,
1323: there are no refereed publications that spell out the details of the derivation.
1324: Nevertheless, it has been used by Wright et al. (2004) to derive ages for stars
1325: in the sample being studied by the Marcy group for evidence of planets,
1326: and we follow suit.
1327:
1328: \subsubsection{The Soderblom et al. (1991) relationship}
1329:
1330: %{\bf But we follow the Soderblom et al work to gain understanding: }\\
1331: The relationship between chromospheric emission and age provided in
1332: Soderblom et al. (1991) is
1333: \begin{equation}
1334: {
1335: log[t_{Chromospheric}] = (-1.50 \pm 0.003) log R'_{HK} + (2.25 \pm 0.12)
1336: }
1337: \end{equation}
1338: where the age, $t$, is again in Gyr.
1339: ({\it Note especially that errors are provided.})
1340: This expression, equation\,(3) from Soderblom et al. (1991), is based on
1341: 42 data points and three ``fundamental'' points, which are the Sun, the Hyades,
1342: and the Ursa Major Group. This relationship passes through the data point for
1343: the Sun, using the value of Log\,$R'_{HK}=-4.96$, quoted there, and a solar age
1344: of 4.6\,Gyr. It is equivalent to $R'_{HK} \propto t^{-2/3}$, and
1345: the authors note that a case could be made for using slightly different
1346: relationships, including one where $R'_{HK} \propto t^{-3/4}$, equation (2) in
1347: their paper, depending on the choice of data points included. This relationship
1348: has a standard deviation of 0.17\,dex, which corresponds to an error of
1349: $\sim$40\%.
1350: In the absence of errors for the Donahue (1998) relationship above, we simply
1351: adopt this value of 0.17\,dex for the error in chromospheric ages calculated
1352: using that relationship also.
1353:
1354: %{\bf Why the reader needs to go to the Soderblom et al papers: }\\
1355: We note also that the Soderblom et al. (1991) relationship is in some ways the
1356: culmination of an extensive and self-consistent study by Soderblom and
1357: collaborators, and is explained in detail in a series of papers, including
1358: Duncan (1984), Duncan et al. (1984), Soderblom (1985), and
1359: Soderblom \& Clements (1987). The reader is referred to these for the
1360: technical details, and especially for the overall logic of the scheme.
1361:
1362: %{\bf Calibration of SDJ91 relationship: }\\
1363: One point about the calibration of the technique needs to be mentioned,
1364: because it also relates to the calibration of gyrochronology.
1365: The ages against which the above relationship is calculated are derived using
1366: isochrone fits to visual binary stars, and to the ``fundamental'' points, which
1367: are again based on isochrone fits.
1368: The entire isochrone technique itself is calibrated by ensuring
1369: that the appropriate solar model matches the solar parameters, usually the
1370: radius and the luminosity, at solar age. Thus, this technique is also ultimately
1371: calibrated on the Sun.
1372:
1373: \subsection{Comparison between chromospheric and gyro ages for the Mt.\,Wilson stars}
1374:
1375: %{\bf Details about chromospheric sources and stars retained: }\\
1376: We use the data compilation published in Baliunas et al. (1996) and
1377: Noyes et al. (1984) for the Mt.\,Wilson stars and calculate the chromospheric
1378: ages using the formula in Donahue (1998).
1379: The chromospheric ages for the Mt.\,Wilson stars, calculated using the above
1380: formula, are listed in Table\,3. For obvious reasons, stars with calculated
1381: periods have been excised, and only stars with measured periods
1382: (71 in number) have been retained for this comparison\footnote{We have also
1383: had to eliminate HD124570, which, although not considered evolved in the Mt.\,
1384: Wilson datasets, is now known to be so (e.g. Cowley, 1976; SAB thanks Brian
1385: Skiff for researching this star).}.
1386: %(37Y+34O)
1387:
1388: %{\bf Details about gyrochronology for these and other stars: }\\
1389: For the same stars, we can calculate ages via gyrochronology using
1390: equation\,(3) from Section\,3 above.
1391: These ages are calculated and listed in Table\,3.
1392: The errors on these ages, calculated using equation\,(16) from section\,4,
1393: are also listed in the table.
1394:
1395: %{\bf Initial comparison: }\\
1396: The gyro ages are plotted against the chromospheric ages for the same stars
1397: in Fig.\,7, with small green and large red symbols marking the young (Y) and
1398: old (O) Mt.\,Wilson stars, as classified by Vaughan (1980).
1399: Note that both techniques segregate the Y and O stars. The demarcation is
1400: sharper in chromospheric age, as it ought to be, since this is the criterion
1401: chosen to classify the stars as young or old.
1402: The figure shows that, apart from a slight tendency towards shorter gyro ages
1403: (discussed further below), there is general agreement between the
1404: chromospheric- and gyro ages for this sample.
1405: Note that except for a few stars discussed below, there are no stars with
1406: widely discrepant ages, unlike the corresponding comparison with isochrone ages,
1407: where discrepancies are routine (c.f. Fig.\,2 in Barnes 2001).
1408:
1409: %{\bf Gyro ages better-defined than chromospheric ages for very young/old stars: }\\
1410: Note especially that the gyro ages are well-behaved for every single star here,
1411: ranging from just under 100\,Myr to just over 10\,Gyr. In contrast, there
1412: are three stars whose chromospheric ages are almost certainly incorrect
1413: (see table\,4).
1414: The two stars HD82443 and HD129333 have chromospheric ages of 0.67\,Myr
1415: and 0.002\,Myr respectively. These stars are undoubtedly young but the
1416: interpretation of these numbers eludes us. The corresponding gyro ages
1417: for these stars are $164 \pm 18$\,Myr and $73 \pm 9$\,Myr,
1418: which suggest that they are essentially on the Zero Age Main Sequence (ZAMS).
1419: Also, for one star, HD95735, the chromospheric age is 20\,Gyr, greater than
1420: that of the universe (dotted lines). This cannot be correct. The gyro age for
1421: this star is $3.2 \pm 0.5$\,Gyr, definitely younger than the Sun.
1422: At least in this restricted sense and for specific stars, the gyro ages are
1423: better defined than chromospheric ages.
1424:
1425: %{\bf Further details }\\
1426: Fig.\,8 elaborates on the difference between the chromospheric- and gyro ages.
1427: The solid line again denotes equality, while the dashed line,
1428: at $t_{gyro}=0.74 \times t_{chrom}$, bisects the data points. This shows that
1429: the gyro ages are roughly 25\% lower than the chromospheric ages overall.
1430: The nature of the disagreement can be probed by segregating the stars by color.
1431: Thus, stars bluer than $B-V=0.6$ and redder than $B-V=0.8$ are plotted using
1432: blue crosses and red asterisks respectively, while those with intermediate
1433: colors are plotted using green squares.
1434: This exercise shows that there is good agreement for stars redward of
1435: $B-V=0.6$ and that the above discrepancy pertains only to the blue stars.
1436:
1437: \subsection{Discussion of the disagreement between the techniques}
1438:
1439: %{\bf Introductory para: }\\
1440: This disagreement can be probed further than merely stating that the errors in
1441: the chromospheric ages are greater than those of the gyro ages. It must
1442: originate in either the gyro ages for blue stars being systematically shorter
1443: or the chromospheric ages for these being systematically longer, or both.
1444: The discussion below, and the results of testing binaries, performed in
1445: Section\,8, suggest that the chromospheric ages are the more problematical.
1446:
1447: %{\bf Problems with gyro ages for blue stars: }\\
1448: With respect to the gyro ages, one defect is that
1449: the open cluster sample used to define the mass dependence of rotation
1450: does not contain stars with $B-V$ colors blueward of 0.5 because it is not yet
1451: possible to distinguish between very blue C- and I-type stars. This means that
1452: $f(B-V)$ is an extrapolation for stars with $ 0.4 < B-V < 0.5$. The fitting
1453: function $f(B-V)$, blueward of $B-V=0.6$, appears to be somewhat elevated with
1454: respect to the data points displayed in Fig.\,5. This would tend to lower the
1455: gyro ages. If $f(B-V)$ were lowered in this region by $\sim$20\%, the gyro ages
1456: would be raised by a factor of $\sim$1.5, which is doable considering the
1457: main sequence lifetime of the F\,stars, but a reduction of $\sim$30\% would
1458: double the gyro ages, and might run afoul of standard stellar evolution,
1459: because the main sequence lifetime of a late-F star is $\sim$5\,Gyr.
1460:
1461: %{\bf Problems with chromospheric ages for blue stars: }\\
1462: The chromospheric ages are not blamefree in this regard either, and it is
1463: almost certain that they have been overestimated for F\,stars\footnote{The embedded mass dependence in the chromospheric ages can be traced to Noyes et al. (1984), where the mass dependence of chromospheric emission was based on the Rossby Number, and theoretical estimates of the variation of convective turnover timescale with stellar mass. The residual mass dependence could be removed eventually with the availability of larger samples of stars, especially those in open clusters.}.
1464: Four F\,stars have chromospheric ages in excess of 7\,Gyr and one in excess of
1465: 6\,Gyr. These values exceed the main sequence lifetime of a late-F star, which
1466: is 5\,Gyr.
1467: In comparison, all of these five F\,stars have shorter gyro ages, with the
1468: oldest of them assigned a gyro age of 2.3\,Gyr.
1469: These stars are also listed in Table\,4. These stars are located at higher
1470: chromospheric ages than 5\,Gyr, marked in Fig.\,8 with thin dashed blue lines.
1471: Thus, while the gyro ages have possibly been slightly underestimated for
1472: F\,stars, it is almost certain that the corresponding chromospheric ages have
1473: been overestimated.
1474:
1475: \subsection{Additional issues with chromospheric ages}
1476:
1477: %{\bf General issues with chromospheric emission: }\\
1478: The derivation of a chromospheric age for a star is complicated by the
1479: natural variability of chromospheric emission with stellar rotational phase
1480: and stellar cycle (e.g. Wilson, 1963).
1481: In fact, rotation-related variations in chromospheric emission are the
1482: preferred way of deriving rotation periods for old stars.
1483: Binarity or other effects could result in additional variability.
1484: These variations make it necessary for repeated measurement on a suitable
1485: timescale of the chromospheric emission from a star to ensure that the
1486: measured average is a good representation of the chromospheric emission at
1487: that age for the star.
1488: Therefore, it is unlikely that one can make a single measurement of
1489: chromospheric emission and derive a good age for a star.
1490:
1491: %{\bf Giampapa paper: }\\
1492: Some of the issues with chromospheric ages are illustrated by the recent work
1493: of Giampapa et al. (2006) on the chromospheric properties of the sun-like
1494: stars in the open cluster M\,67, averaged over several seasons of observing.
1495: This cluster is known to be $\sim$4\,Gyr old (e.g. VandenBerg \& Stetson, 2004).
1496: There is no evidence that the stars in this cluster are not coeval.
1497:
1498: Correspondingly, Giampapa et al. (2006) derive mean and median ages in the
1499: range 3.8\,Gyr to 4.3\,Gyr.
1500: What is surprising is that the chromospheric ages for individual stars
1501: range from under 1\,Gyr to 7.5\,Gyr (see Fig.\,13 in their paper).
1502: Admittedly, the vast majority of the stars have chromospheric ages between
1503: 2\,Gyr and 6\,Gyr, but this range is not small either.
1504: This result seems to cast doubts on the precision in ages for single stars
1505: obtainable even in principle with chromospheric emission because measuring
1506: the age for M\,67 using a random cluster member could result in such large
1507: age variability.
1508:
1509: %{\it Having tested the rotational clock, we now use it on an illustrative
1510: %group of field stars.}
1511:
1512: \clearpage
1513:
1514: \begin{deluxetable}{llcrrrrr}
1515: \tabletypesize{\scriptsize}
1516: \tablecaption{Gyrochronology ages and errors for the Mt.\,Wilson stars. \label{tbl-3}}
1517: \tablewidth{0pt}
1518: \tablehead{
1519: \colhead{HD} & \colhead{$B-V$} &\colhead{$P_{rot}$(d)\tablenotemark{A}} & \colhead{$-log<R'_{HK}>$} & \colhead{$t_{chrom}$/Myr } & \colhead{$t_{iso}$/Myr\tablenotemark{B}} & \colhead{$t_{gyro}$/Myr} & \colhead{$\delta t_{gyro}$/Myr}
1520: %\colhead{$\Theta$\tablenotemark{b}}
1521: }
1522: \startdata
1523: Sun &0.642 &26.09\tablenotemark{C} &4.901 &3895 &...... &4566 &770\\
1524: 1835 &0.66 &7.78 &4.443 &601 &$<$1760 &408 &54\\
1525: 2454 &0.43 &3 &4.792 &2609 &...... &790 &350\\
1526: 3229 &0.44 &2 &4.583 &1251 &...... &260 &91\\
1527: 3651 &0.85 &44 &4.991 &5411 &$>$11800 &6100 &990\\
1528: 4628 &0.88 &38.5 &4.852 &3250 &$>$6840 &4370 &680\\
1529: 6920 &0.60 &13.1 &4.793 &2618 &...... &1510 &240\\
1530: 10476 &0.84 &35.2 &4.912 &4056 &$>$8840 &4070 &630\\
1531: 10700 &0.72 &34 &4.958 &4802 &$>$12120 &5500 &910\\
1532: 10780 &0.81 &23 &4.681 &1764 &10120 &1945 &280\\
1533: 16160 &0.98 &48.0 &4.958 &4802 &540 &5370 &850\\
1534: 16673 &0.52 &7 &4.664 &1664 &...... &820 &150\\
1535: 17925 &0.87 &6.76 &4.311 &74 &$<$1200 &157 &16\\
1536: 18256 &0.43 &3 &4.722 &2033 &...... &790 &350\\
1537: 20630 &0.68 &9.24 &4.420 &489 &$<$2760 &522 &69\\
1538: 22049 &0.88 &11.68 &4.455 &659 &$<$600 &439 &52\\
1539: 25998 &0.46 &2.6 &4.401 &398 &...... &270 &70\\
1540: 26913 &0.70 &7.15 &4.391 &352 &...... &294 &36\\
1541: 26965 &0.82 &43 &4.872 &3499 &$>$9280 &6320 &1030\\
1542: 30495 &0.63 &7.6\tablenotemark{D} &4.511 &923 &6080 &450 &62\\
1543: 35296 &0.53 &3.56 &4.378 &294 &...... &202 &33\\
1544: 37394 &0.84 &11 &4.454 &654 &$<$1360 &432 &52\\
1545: 39587 &0.59 &5.36 &4.426 &518 &4320 &286 &41\\
1546: 45067 &0.56 &8 &5.094 &7733 &5120 &760 &120\\
1547: 72905 &0.62 &4.69 &4.375 &281 &...... &187 &24\\
1548: 75332 &0.49 &4 &4.464 &703 &1880 &387 &79\\
1549: 76151 &0.67 &15 &4.659 &1635 &1320 &1380 &200\\
1550: 78366 &0.60 &9.67 &4.608 &1370 &$<$680 &840 &130\\
1551: 81809 &0.64 &40.2 &4.921 &4193 &...... &10600\tablenotemark{E} &1900\\
1552: 82443 &0.77 &6 &4.211 &0.7 &...... &164 &18\\
1553: 89744 &0.54 &9 &5.120 &8421 &1880 &1110 &190\\
1554: 95735 &1.51 &53 &5.451 &20028 &...... &3070 &460\\
1555: 97334 &0.61 &8 &4.422 &499 &$<$2920 &551 &80\\
1556: 100180 &0.57 &14 &4.922 &4209 &3800 &2070 &350\\
1557: 101501 &0.72 &16.68 &4.546 &1082 &$>$11320 &1400 &200\\
1558: 106516 &0.46 &6.91 &4.651 &1591 &...... &1770 &480\\
1559: 107213 &0.50 &9 &5.103 &7966 &2040 &1630 &330\\
1560: 114378 &0.45 &3.02 &4.530 &1010 &...... &445 &130\\
1561: 114710 &0.57 &12.35 &4.745 &2205 &$<$1120 &1630 &270\\
1562: 115043 &0.60 &6 &4.428 &528 &...... &335 &47\\
1563: 115383 &0.58 &3.33 &4.443 &601 &$<$760 &122 &17\\
1564: 115404 &0.93 &18.47 &4.480 &779 &...... &950 &120\\
1565: 115617 &0.71 &29 &5.001 &5609 &8960 &4200 &680\\
1566: 120136 &0.48 &4 &4.731 &2098 &1640 &443 &97\\
1567: 129333 &0.61 &2.80 &4.152 &0.002 &$<$1440 &73 &9\\
1568: \hline
1569: 131156A &0.76 &6.31\tablenotemark{F} &4.363 &232 &$<$760 &187 &21\\
1570: 131156B &1.17 &11.94\tablenotemark{G} &4.424 &508 &$>$12600 &265 &28\\
1571: \hline
1572: 141004 &0.60 &25.8 &5.004 &5669 &6320 &5570 &990\\
1573: 143761 &0.60 &17 &5.039 &6413 &9720 &2490 &410\\
1574: 149661 &0.82 &21.07 &4.583 &1251 &$<$4160 &1600 &220\\
1575: 152391 &0.76 &11.43 &4.448 &625 &720 &587 &75\\
1576: 154417 &0.57 &7.78 &4.533 &1023 &4200 &670 &105\\
1577: \hline
1578: 155885 &0.86 &21.11\tablenotemark{H} &4.559 &1141 &...... &1440 &200\\
1579: 155886 &0.86 &20.69\tablenotemark{I} &4.570 &1191 &...... &1390 &190\\
1580: 156026 &1.16 &18.0\tablenotemark{J} &4.622 &1439 &$<$480 &593 &71\\
1581: \hline
1582: 160346 &0.96 &36.4 &4.795 &2637 &...... &3280 &490\\
1583: 165341A\tablenotemark{K} &0.86 &20 &4.548 &1091 &...... &1300 &180\\
1584: 166620 &0.87 &42.4 &4.955 &4750 &$>$11200 &5400 &860\\
1585: 178428 &0.70 &22 &5.048 &6616 &...... &2560 &390\\
1586: 185144 &0.80 &27 &4.832 &3019 &...... &2730 &410\\
1587: 187691 &0.55 &10 &5.026 &6128 &3200 &1250 &210\\
1588: 190007 &1.17 &28.95 &4.692 &1832 &$<$1760 &1460 &200\\
1589: 190406 &0.61 &13.94 &4.797 &2657 &3160 &1610 &250\\
1590: 194012 &0.51 &7 &4.720 &2019 &...... &900 &170\\
1591: \hline
1592: 201091 &1.18 &35.37 &4.764 &2359 &$<$440 &2120 &300\\
1593: 201092 &1.37 &37.84\tablenotemark{L} &4.891 &3753 &$<$680 &1870 &260\\
1594: \hline
1595: 206860 &0.59 &4.86 &4.416 &470 &$<$880 &237 &33\\
1596: 207978 &0.42 &3 &4.890 &3740 &...... &1270 &810\\
1597: 212754 &0.52 &12 &5.073 &7207 &...... &2300 &440\\
1598: 219834B\tablenotemark{M} &0.91 &43 &4.944 &4563 &$>$13200 &5040 &800\\
1599: 224930 &0.67 &33 &4.875 &3538 &...... &6330 &1080\\
1600: \enddata
1601:
1602: %% Text for table notes should follow after the \enddata but before
1603: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
1604: %% in the table for each \tablenotetext.
1605:
1606: \tablenotetext{A}{Only {\it measured} periods for unevolved stars are listed.
1607: They are taken, in order of priority, from Donahue, Saar \& Baliunas (1996),
1608: Baliunas et al. (1983), and Baliunas, Sokoloff \& Soon (1996). The first of
1609: these lists the average rotation period of several seasonal periods (and the
1610: differential rotation), hence the priority assigned to this paper, the second
1611: a single best period determined from an intensive chromospheric monitoring
1612: program in 1980-81 (with the error of that single determination), and the third
1613: a mean rotation period (to lower precision than the previous two publications)
1614: based on the entire extant intensive sampling database.}
1615: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
1616: \tablenotetext{B}{The isochrone ages listed in this and subsequent tables are taken from Takeda et al. (2007).}
1617: \tablenotetext{C}{The mean solar period of 26.09d, taken from Donahue et al.
1618: (1996), represents the average of 8 determinations, and is presumably
1619: representative of the mean latitude of sunspot persistence, while the $\sim$25d
1620: period usually listed is the mean equatorial rotation period.}
1621: \tablenotetext{D}{Baliunas et al. (1996) list a significantly different period
1622: of 11d.}
1623: \tablenotetext{E}{The gyro age should be treated with caution because this star
1624: is a spectroscopic binary (Pourbaix, 2000).}
1625: \tablenotetext{F}{Period is from Donahue et al. (1996). Baliunas et al. (1996) simply list a period of 6d.}
1626: \tablenotetext{G}{Period is from Donahue et al. (1996). Baliunas et al. (1996) list a period of 11d.}
1627: \tablenotetext{H}{This period is from Donahue et al. (1996). Baliunas et al. (1983) list a very similar period of 22.9$\pm$0.5d.}
1628: \tablenotetext{I}{This period is from Donahue et al. (1996). Baliunas et al. (1983) list a very similar period of 20.3$\pm$0.4d.}
1629: \tablenotetext{J}{This period is from Baliunas et al. (1983). Baliunas et al. (1996) list a period of 21d for all 3 components HD155885, HD155886, \& HD156026.}
1630: \tablenotetext{K}{The second component, HD165341B, of this binary is also in the Mt.\,Wilson sample (e.g. Baliunas et al. 1996), but the period of 34d is one calculated from chromospheric emission.}
1631: \tablenotetext{L}{The periods listed for HD201091 and HD201092 are from Donahue et al. (1996). Baliunas et al. (1996) list similar periods of 35 and 38d respectively, while Baliunas et al. (1983) list the somewhat discrepant periods of 37.9$\pm$1.0 and 48d respectively.}
1632: \tablenotetext{M}{The other component in this system, HD219834A, is also in the Mt.\,Wilson dataset, but it seems to be evolved, and so is excluded here.}
1633: %\tablecomments{Occasionally, authors wish to append a short
1634: %paragraph of explanatory notes that pertain to the entire table, but
1635: %which are different than the caption. Such notes should be placed in
1636: %a {\tt tablecomments} command like this.}
1637:
1638: \end{deluxetable}
1639:
1640:
1641: \clearpage
1642:
1643: \begin{deluxetable}{llrrl}
1644: \tabletypesize{\scriptsize}
1645: \tablecaption{Stars with suspect chromospheric ages. \label{tbl-4}}
1646: \tablewidth{0pt}
1647: \tablehead{
1648: \colhead{Star} & \colhead{$B-V$} & \colhead{$Age_{chromo}$} & \colhead{$Age_{gyro}$} & \colhead{Comment}
1649: %\colhead{$\Theta$\tablenotemark{b}}
1650: }
1651: \startdata
1652: HD45067 & 0.56 & 7.73\,Gyr & 0.76$\pm$0.1\,Gyr & Chromo age $>$ lifetime\\
1653: HD82443 & 0.77 & 0.7\,Myr & 164$\pm$18\,Myr & Chromo age too small?\\
1654: HD89744 & 0.54 & 8.42\,Gyr & 1.11$\pm$0.19\,Gyr & Chromo age $>$ lifetime\\
1655: HD95735 & 1.51 & 20\,Gyr & 3.1$\pm$0.46\,Gyr & Chromo age $>$ Age of universe\\
1656: HD107213 & 0.50 & 7.97\,Gyr & 1.63$\pm$0.33\,Gyr & Chromo age $>$ lifetime\\
1657: HD129333 & 0.61 & 0.002\,Myr & 73$\pm$9\,Myr & Chromo age too small?\\
1658: HD187691 & 0.55 & 6.13\,Gyr & 1.25$\pm$0.21\,Gyr & Chromo age $>$ lifetime\\
1659: HD212754 & 0.52 & 7.21\,Gyr & 2.3$\pm$0.44\,Gyr & Chromo age $>$ lifetime\\
1660: \enddata
1661:
1662: %% Text for table notes should follow after the \enddata but before
1663: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
1664: %% in the table for each \tablenotetext.
1665:
1666: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
1667: %with the deluxetable environment}
1668: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
1669:
1670: %\tablecomments{Occasionally, authors wish to append a short
1671: %paragraph of explanatory notes that pertain to the entire table, but
1672: %which are different than the caption. Such notes should be placed in
1673: %a {\tt tablecomments} command like this.}
1674:
1675: \end{deluxetable}
1676:
1677:
1678:
1679:
1680: %\newpage
1681: \section{Ages for young field stars from the Vienna-KPNO (Strassmeier et al. 2000) survey}
1682:
1683: %{\bf Introduction: }\\
1684: Another group of stars amenable to the calculation of ages via gyrochronology
1685: is the field star sample of Strassmeier et al. (2000). Unlike the older
1686: Mt.\,Wilson star sample, this group contains some stars for which gyro ages
1687: are not yet appropriate, and here we demonstrate how to identify and excise
1688: these stars and calculate ages for the rest.
1689:
1690: %{\bf Initial cut of the sample: }\\
1691: The full Strassmeier et al. (2000) sample consists of 1058 Hipparcos stars with
1692: various measured parameters, including chromospheric emission.
1693: Of these 1058 stars, 140 have measured rotation periods, and of these,
1694: we are interested here only in stars on the main sequence. Using the Luminosity
1695: classes supplied by Strassmeier et al., we have simply selected the dwarf stars,
1696: and excised the others.
1697: This leaves us with 101 dwarf stars with measured rotation periods.
1698:
1699: %{\bf Removal of C/g stars: }\\
1700: These 101 stars with measured periods are plotted in a color-period diagram in
1701: Fig.\,9. We superimpose an I sequence curve corresponding to 100\,Myr, and
1702: assume that the 16 stars below this curve are C-\,or g\,stars, while those
1703: above are I\,sequence stars similar to those in the Mt.\,Wilson sample.
1704: In the scenario from Barnes (2003a), the stars below are either on the
1705: C\,sequence appropriate to their age, or in the transition, g, between the
1706: C-\,and I\,sequences.
1707: This cut is undoubtedly conservative, since there are open clusters younger
1708: than 100\,Myr known to possess I sequences, but we prefer to lose a few stars
1709: rather than risk over-extending the technique.
1710: This leaves us with 85 potential I\,sequence stars amenable to gyrochronology.
1711:
1712: %{\bf Preliminary Interpretation: }\\
1713: These 85 I\,sequence stars are again plotted in Fig.\,10, where now we have
1714: superimposed isochrones corresponding to ages of
1715: 100-, 200-, \& 450\,Myr, and 1-, 2-, \& 4.5\,Gyr.
1716: We see that the Strassmeier et al. (2000) main sequence sample with measured
1717: periods consists mainly of stars younger than 1\,Gyr, and all but 4 younger than
1718: 2\,Gyr. In fact the median age for the sample is 365\,Myr, in keeping with the
1719: selection of this sample for activity.
1720:
1721: %{\bf Ages for each star: }\\
1722: Gyro ages are calculated as above for each star, and listed in Table\,5, along
1723: with their basic measured properties.
1724: Almost all of these stars are redder than the Sun.
1725: For such stars, as shown in the previous section, there is very good agreement
1726: between gyro- and chromospheric ages, and consequently, some confidence can be
1727: attached to the calculated ages.
1728: We have also calculated the errors on these ages, using equation\,(16) from
1729: section\,4, and listed these in the final column of the table.
1730:
1731: %{\bf Corroborating evidence: }\\
1732: At present, no good test of these ages is possible.
1733: Although the chromospheric emission has been measured, the measurements have
1734: been made with a small telescope, and are not long-term averages, so that
1735: the values quoted cannot be treated with the confidence associated, for
1736: instance, with the Mt.\,Wilson measurements, and there is considerable scatter,
1737: as the few repeat measurements demonstrate.
1738: Furthermore, no photospheric correction has been performed, so they
1739: are on a different scale, and the relationships of Soderblom et al. (1991) and
1740: Donahue (1998) do not apply.
1741: However, it is possible to plot the $R_{HK}$ values provided against the gyro
1742: ages calculated above to make sure that gross errors are absent, and we perform
1743: this exercise in Fig.\,11.
1744: The figure demonstrates that, as expected, the chromospheric activity declines
1745: steadily with stellar age, and thus, that the gyro ages are reasonable.
1746:
1747: %{\it We now use the rotational clock on another illustrative
1748: %group of field stars.}
1749:
1750:
1751: \clearpage
1752:
1753: \begin{deluxetable}{lrcrrrr}
1754: \tabletypesize{\scriptsize}
1755: \tablecaption{Gyrochronology ages and errors for the Vienna-KPNO survey (Strassmeier et al. 2000) stars. \label{tbl-5}}
1756: \tablewidth{0pt}
1757: \tablehead{
1758: \colhead{HD} & \colhead{$B-V$} &$P_{rot}$(d) & \colhead{$R_{HK}$} & \colhead{$t_{iso}$/Myr} & \colhead{$t_{gyro}$/Myr} & \colhead{$\delta t_{gyro}$/Myr}
1759: %\colhead{$\Theta$\tablenotemark{b}}
1760: }
1761: \startdata
1762: HD691 &0.76 &6.105 &7.2E-5 &$<$1040 &175 &20\\
1763: HD5996 &0.76 &12.165 &6.0E-5 &...... &662 &86\\
1764: HD6963 &0.73 &20.27 &4.9E-5 &2240 &1960 &290\\
1765: HD7661 &0.75 &7.85 &7.1E-5 &...... &294 &35\\
1766: HD8997a &0.97 &10.49 &3.5E-5 &...... &292 &32\\
1767: HD8997b &0.97 &10.49 &1.8E-5 &...... &292 &32\\
1768: HD9902b &0.65 &7.41 &8.9E-5 &...... &389 &52\\
1769: HD10008 &0.80 &7.15 &6.1E-5 &...... &211 &23\\
1770: HD12786 &0.83 &15.78 &5.6E-5 &...... &890 &120\\
1771: HD13382 &0.68 &8.98 &6.0E-5 &...... &494 &65\\
1772: HD13507 &0.67 &7.60 &6.6E-5 &$<$1320 &373 &48\\
1773: HD13531 &0.70 &7.52 &7.0E-5 &$<$3520 &324 &40\\
1774: HD13579A &0.92 &6.79 &2.8E-5 &$>$8840 &141 &14\\
1775: HD16287 &0.94 &11.784 &5.3E-5 &$<$2360 &390 &45\\
1776: HD17382 &0.82 &$>$50 &6.1E-5 &...... &$>$8450\tablenotemark{A} &$>$1400\\
1777: HD18632 &0.93 &10.055 &5.3E-5 &$>$7800 &293 &33\\
1778: HD18955a &0.86 &8.05 &6.1E-5 &...... &225 &25\\
1779: HD19668 &0.81 &5.41 &4.1E-5 &...... &120 &12\\
1780: HD19902 &0.73 &$>$50 &5.4E-5 &...... &$>$11200 &$>$2000\\
1781: HD20678 &0.73 &5.95 &6.6E-5 &...... &185 &21\\
1782: HD27149a &0.68 &8.968 &5.7E-5 &...... &492 &65\\
1783: HD27149b &0.68 &8.968 &5.1E-5 &...... &492 &65\\
1784: HD28495 &0.76 &7.604 &9.3E-5 &...... &268 &31\\
1785: HD31000 &0.75 &7.878 &8.2E-5 &...... &296 &35\\
1786: HD53157 &0.81 &10.88 &6.2E-5 &...... &460 &56\\
1787: HD59747 &0.86 &8.03 &6.5E-5 &$<$920 &224 &24\\
1788: HD73322 &0.91 &16.41 &5.1E-5 &...... &788 &100\\
1789: HD75935 &0.77 &8.19 &6.3E-5 &...... &299 &35\\
1790: HD77825 &0.96 &8.64 &5.2E-5 &...... &205 &22\\
1791: HD79969 &0.99 &43.4 &3.6E-5 &...... &4340 &670\\
1792: HD82443 &0.78 &5.409 &9.4E-5 &...... &130 &14\\
1793: HD83983 &0.88 &10.92 &3.9E-5 &...... &386 &45\\
1794: HD87424 &0.89 &10.74 &5.3E-5 &$<$1720 &365 &42\\
1795: HD88638 &0.77 &4.935 &8.4E-5 &...... &113 &12\\
1796: HD92945 &0.87 &13.47 &7.5E-5 &$<$1280 &592 &73\\
1797: HD93811 &0.94 &8.47 &5.0E-5 &...... &206 &22\\
1798: HD94765 &0.92 &11.43 &4.6E-5 &$<$1480 &384 &44\\
1799: HD95188 &0.76 &7.019 &7.1E-5 &$<$960 &230 &26\\
1800: HD95724 &0.94 &11.53 &5.2E-5 &...... &374 &43\\
1801: HD95743 &0.97 &10.33 &4.0E-5 &...... &284 &31\\
1802: HD101206 &0.98 &10.84 &4.1E-5 &...... &305 &34\\
1803: HD103720 &0.95 &17.16 &4.2E-5 &...... &787 &99\\
1804: HD105963A &0.88 &7.44 &6.6E-5 &...... &184 &19\\
1805: HD105963B &0.88 &7.44 &6.2E-5 &...... &184 &19\\
1806: HD109011a &0.94 &8.31 &5.8E-5 &...... &199 &21\\
1807: HD109647 &0.95 &8.73 &5.3E-5 &...... &214 &23\\
1808: HD110463 &0.96 &11.75 &4.5E-5 &...... &371 &42\\
1809: HD111813 &0.89 &7.74 &6.5E-5 &...... &194 &21\\
1810: HD113449 &0.85 &6.47 &6.9E-5 &...... &152 &16\\
1811: HD125874 &0.88 &7.52 &6.1E-5 &...... &188 &20\\
1812: HD128311 &0.97 &11.54 &4.5E-5 &$<$960 &351 &40\\
1813: HD130307 &0.89 &21.79 &3.9E-5 &1520 &1425 &190\\
1814: HD139194 &0.87 &9.37 &3.8E-5 &...... &294 &33\\
1815: HD139837 &0.73 &6.98 &8.0E-5 &...... &251 &30\\
1816: HD141272 &0.80 &14.045 &6.7E-5 &...... &773 &100\\
1817: HD141919 &0.88 &13.62 &4.4E-5 &...... &590 &72\\
1818: HD142680 &0.97 &33.52 &2.5E-5 &...... &2740 &400\\
1819: HD144872 &0.96 &26.02 &2.7E-5 &...... &1720 &240\\
1820: HD150511 &0.88 &10.58 &5.0E-5 &...... &363 &42\\
1821: HD153525 &1.00 &15.39 &1.7E-5 &...... &577 &69\\
1822: HD153557 &0.98 &7.22 &3.7E-5 &...... &140 &14\\
1823: HD161284 &0.93 &18.31 &4.2E-5 &...... &930 &120\\
1824: HD168603 &0.77 &4.825 &6.4E-5 &...... &108 &11\\
1825: HD173950 &0.83 &10.973 &5.2E-5 &...... &442 &53\\
1826: HD180161 &0.80 &5.49 &2.7E-5 &...... &127 &13\\
1827: HD180263 &0.91 &14.16 &3.7E-5 &...... &593 &72\\
1828: HD189733 &0.93 &12.039 &4.6E-5 &...... &415 &48\\
1829: HD192263 &0.94 &23.98 &4.1E-5 &2560 &1530 &210\\
1830: HD198425 &0.94 &22.64 &4.3E-5 &...... &1370 &185\\
1831: HD200560 &0.97 &10.526 &5.2E-5 &...... &294 &32\\
1832: HD202605 &0.74 &13.78 &5.3E-5 &...... &900 &120\\
1833: HD203030 &0.75 &6.664 &7.1E-5 &...... &215 &25\\
1834: HD209779 &0.67 &10.29 &6.2E-5 &10000 &670 &92\\
1835: HD210667 &0.81 &9.083 &5.3E-5 &$<$4200 &325 &38\\
1836: HD214615AB &0.76 &6.20 &7.5E-5 &...... &181 &20\\
1837: HD214683 &0.94 &18.05 &4.2E-5 &...... &886 &110\\
1838: HD220182 &0.80 &7.489 &6.4E-5 &...... &230 &26\\
1839: HD221851 &0.85 &12.525 &3.8E-5 &...... &541 &66\\
1840: HD258857 &0.91 &19.98 &3.9E-5 &...... &1150 &150\\
1841: HIP36357 &0.92 &11.63 &4.7E-5 &...... &397 &46\\
1842: HIP43422 &0.75 &11.14 &3.3E-4 &...... &578 &74\\
1843: HIP69410 &0.96 &9.52 &5.0E-5 &...... &248 &27\\
1844: HIP70836 &0.94 &21.84 &3.2E-5 &...... &1280 &170\\
1845: HIP77210ab &0.83 &13.83 &6.3E-5 &...... &690 &87\\
1846: HIP82042 &0.96 &13.65 &3.2E-5 &...... &496 &59\\
1847: \enddata
1848:
1849: %% Text for table notes should follow after the \enddata but before
1850: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
1851: %% in the table for each \tablenotetext.
1852:
1853: \tablenotetext{A}{The gyro age should be treated with caution because this star
1854: is a spectroscopic binary (Latham et al. 2002).}
1855: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
1856: %with the deluxetable environment}
1857: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
1858:
1859: %\tablecomments{Occasionally, authors wish to append a short
1860: %paragraph of explanatory notes that pertain to the entire table, but
1861: %which are different than the caption. Such notes should be placed in
1862: %a {\tt tablecomments} command like this.}
1863:
1864: \end{deluxetable}
1865:
1866:
1867:
1868:
1869:
1870:
1871:
1872:
1873:
1874: %\newpage
1875: \section{Ages for young- and intermediate-age field stars from Pizzolato et al. (2003) with X-ray measurements}
1876:
1877: %{\bf Introduction }\\
1878: There exists another comparably large group of main sequence field stars for
1879: which rotation periods and other relevant information are available. This group
1880: has been assembled by Pizzolato et al. (2003) in connection with a study of
1881: X-ray activity. There are 110 stars in this group. We remove 2 of these,
1882: HD82885 and HD136202, suspected to be evolved, leaving 108 stars.
1883: 51 of these 108 stars are in also in the Mt.\,Wilson sample, but the
1884: remainder do not overlap with the Strassmeier et al. (2000) stars either,
1885: and hence warrant attention. Furthermore, these stars also have measured X-ray
1886: fluxes, listed conveniently in Pizzolato et al. (2003),
1887: which allow a crude comparison with the gyro ages we derive below.
1888:
1889: %{\bf Division and cleaning of the sample: }\\
1890: The Pizzolato et al. (2003) stars must follow the same rotational patterns as
1891: the open cluster, Mt.\,Wilson, and Strassmeier et al. (2000) stars.
1892: We can use the same condition that we used with the Strassmeier et al. (2003)
1893: stars to excise the C/g stars from the sample.
1894: We plot the color-period diagram for this sample in Fig.\,12.
1895: Plotting a 100\,Myr isochrone as before, we excise all stars below it,
1896: since these are either C- or g\,type stars, or only ambiguously I\,type.
1897: Note that of the excised stars, the ones with periods below 1 day are
1898: almost certainly C\,sequence stars.
1899: We also excise GL551 ($B-V=1.90$, $P=42d$) because it is fully convective
1900: and therefore unable to sustain an interface dynamo.
1901: (Note also that apart from this one object, there are no slow rotators redward
1902: of $B-V=1.55$. This is consistent with the prediction by Barnes (2003a) for
1903: the terminus of the I sequence at the point of full convection.)
1904: This leaves us with 79 stars that are potentially on the I sequence in this
1905: sample.
1906:
1907: %{\bf Some details: }\\
1908: These 79 stars are suitable for gyrochronology.
1909: We calculate the gyro ages as before, and list them, their errors, and other
1910: relevant information for these stars in Table\,6.
1911: Fig.\,13 displays the color-period diagram for these 79 stars, with isochrones
1912: at 100\,Myr, 200\,Myr, 450\,Myr, 1\,Gyr, 2\,Gyr, 4.5\,Gyr, \& 10\,Gyr.
1913: Fig.\,13 shows that this sample spans a substantial range of ages,
1914: from 100\,Myr to 6\,Gyr (all but 4 of them),
1915: although most of them are younger than the Sun, and the median age for the
1916: sample is 1.2\,Gyr.
1917:
1918: %{\bf Isochrone/X-rays: }\\
1919: These results are reasonably consistent with Sandage et al. (2003) who suggest a
1920: (classical isochrone) age for the oldest stars in the local Galactic disk of
1921: 7.4 to 7.9\,Gyr ($\pm 0.7$\,Gyr) depending on whether or not the stellar
1922: models allow for diffusion.
1923: All the Pizzolato et al. (2003) stars except for HD81809,
1924: which is known to be a spectroscopic binary (Pourbaix, 2000),
1925: have calculated gyro ages shortward of this age.
1926:
1927: The available X-ray data for these same stars also suggest that the gyro
1928: ages are reasonable. In Fig.\,14, we plot the X-ray emission from these stars
1929: against their gyro ages. We see that the X-ray emission declines steadily,
1930: as expected, and in fact, there are no widely discrepant data points.
1931:
1932: %{\bf Henry et al work: }\\
1933: Finally, we note that in addition to the 51 stars in this group that are common
1934: to the Mt.\,Wilson sample, 19 are present in the chromospheric emission survey
1935: of Southern stars by Henry et al. (1996), where their $R'_{HK}$ values are
1936: published. These are on the same system as the Mt.\,Wilson data. Thus, it is
1937: possible to compute their chromospheric ages, and compare them with the ages
1938: from gyrochronology. This comparison is shown in Fig.\,15.
1939: The stars can be seen to scatter around the line of equality, and in fact,
1940: the agreement between the gyro and chromospheric ages for all but two of them
1941: is within a factor of two (see Fig.\,15).
1942:
1943: %{\it We now do the strictest test on the gyro clock.}
1944:
1945: \clearpage
1946:
1947: \begin{deluxetable}{llcrrrr}
1948: \tabletypesize{\scriptsize}
1949: \tablecaption{Gyrochronology ages and errors for the Pizzolato et al. (2003) stars. \label{tbl-6}}
1950: \tablewidth{0pt}
1951: \tablehead{
1952: \colhead{Star} & \colhead{$B-V$} &$P_{rot}$(d) &$log(L_x/L_{bol})$ & \colhead{$t_{iso}$/Myr} & \colhead{$t_{gyro}$/Myr} & \colhead{$\delta t_{gyro}$/Myr}
1953: %\colhead{$\Theta$\tablenotemark{b}}
1954: }
1955: \startdata
1956: Sun &0.66 &25.38 &-6.23 &........ & 3980 & 650\\
1957: GL338B &1.42 &10.17 &-4.90 &........ & 140 & 14\\
1958: GL380 &1.36 &11.67 &-5.04 &........ & 196 & 20\\
1959: GL673 &1.36 &11.94 &-4.96 &........ & 205 & 21\\
1960: GL685 &1.45 &18.60 &-4.92 &........ & 435 & 50\\
1961: HD1835 &0.66 & 7.70 &-4.67 &$<$1760 & 400 & 53\\
1962: HD3651 &0.85 &48.00 &-5.70 &$>$11800 & 7200 & 1200\\
1963: HD4628 &0.88 &38.00 &-6.01 &$>$6840 & 4260 & 660\\
1964: HD10360 &0.88 &30.00 &-5.97 &$<$600 & 2700 & 400\\
1965: HD10361 &0.86 &39.00 &-5.94 &$<$520 & 4710 & 740\\
1966: HD10476 &0.84 &35.20 &-6.59 &$>$8840 & 4070 & 630\\
1967: HD10700 &0.72 &34.50 &-6.21 &$>$12120 & 5660 & 930\\
1968: HD11507 &1.43 &15.80 &-4.76 &........ & 324 & 36\\
1969: HD13445 &0.82 &30.00 &-5.56 &$>$8480 & 3160 & 480\\
1970: HD14802 &0.60 & 9.00 &-4.50 &........ & 732 & 110\\
1971: HD16160 &0.97 &45.00 &-5.71 & 540 & 4840 & 760\\
1972: HD16673 &0.52 & 7.40 &-5.03 &........ & 910 & 170\\
1973: HD17051 &0.56 & 7.90 &-5.02 & 2720 & 740 & 120\\
1974: HD17925 &0.87 & 6.60 &-4.51 &$<$1200 & 150 & 15\\
1975: HD20630 &0.68 & 9.40 &-4.62 &$<$2760 & 540 & 72\\
1976: HD22049 &0.88 &11.30 &-4.92 &$<$600 & 412 & 48\\
1977: HD25998 &0.52 & 3.00 &-4.40 &........ & 159 & 27\\
1978: HD26913 &0.70 & 7.20 &-4.18 &........ & 298 & 37\\
1979: HD26965 &0.82 &37.10 &-5.59 &$>$9280 & 4750 & 750\\
1980: HD30495 &0.64 & 7.60 &-4.86 & 6080 & 428 & 58\\
1981: HD32147 &1.06 &47.40 &-5.87 &$<$5450 & 4510 & 700\\
1982: HD35296 &0.53 & 5.00 &-4.52 &........ & 388 & 66\\
1983: HD36435 &0.78 &11.20 &-4.90 &........ & 531 & 66\\
1984: HD38392 &0.94 &17.30 &-4.77 &........ & 816 & 100\\
1985: HD39587 &0.59 & 5.20 &-4.51 & 4320 & 270 & 38\\
1986: HD42807 &0.66 & 7.80 &-4.83 &........ & 410 & 54\\
1987: HD43834 &0.72 &32.00 &-6.05 & 8760 & 4900 & 800\\
1988: HD52698 &0.90 &26.00 &-4.74 &........ & 1960 & 280\\
1989: HD53143 &0.81 &16.40 &-4.67 &........ & 1010 & 130\\
1990: HD72905 &0.62 & 4.10 &-4.47 &........ & 144 & 18\\
1991: HD75332 &0.52 & 4.00 &-4.35 & 1880 & 277 & 48\\
1992: HD76151 &0.67 &15.00 &-5.24 & 1320 & 1380 & 200\\
1993: HD78366 &0.60 & 9.70 &-4.75 &$<$680 & 850 & 130\\
1994: HD81809 &0.64 &40.20 &-6.25 &........ & 10600\tablenotemark{A} & 1900\\
1995: HD82106 &1.00 &13.30 &-4.64 &$<$600 & 435 & 50\\
1996: HD95735 &1.51 &48.00 &-5.12 &........ & 2530 & 370\\
1997: HD97334 &0.60 & 7.60 &-4.51 &$<$2920 & 529 & 77\\
1998: HD98712 &1.36 &11.60 &-4.08 &........ & 194 & 20\\
1999: HD101501 &0.74 &16.00 &-5.17 &$>$11320 & 1200 & 170\\
2000: HD114613 &0.70 &33.00 &-5.85 & 5200 & 5600 & 930\\
2001: HD114710 &0.58 &12.40 &-5.50 &$<$1120 & 1530 & 250\\
2002: HD115383 &0.58 & 3.30 &-4.82 &$<$760 & 120 & 16\\
2003: HD115404 &0.92 &19.00 &-5.25 &........ & 1020 & 130\\
2004: HD128620 &0.71 &29.00 &-6.45 & 7840 & 4200 & 670\\
2005: HD128621 &0.88 &42.00 &-5.97 &$>$11360 & 5170 & 820\\
2006: HD131156A &0.72 & 6.20 &-4.70 &$<$760 & 207 & 24\\
2007: HD131977 &1.11 &44.60 &-5.38 &$<$600 & 3690 & 560\\
2008: HD141004 &0.60 &18.00 &-6.18 & 6320 & 2780 & 460\\
2009: HD147513 &0.62 & 8.50 &-4.61 &$<$680 & 587 & 84\\
2010: HD147584 &0.55 &13.00 &-4.58 &........ & 2080 & 370\\
2011: HD149661 &0.81 &23.00 &-4.96 &$<$4160 & 1950 & 280\\
2012: HD152391 &0.75 &11.10 &-4.66 & 720 & 574 & 73\\
2013: HD154417 &0.58 & 7.60 &-4.91 & 4200 & 597 & 91\\
2014: HD155885 &0.86 &21.11 &-4.71 &........ & 1440 & 200\\
2015: HD155886 &0.85 &20.69 &-4.65 &........ & 1420 & 190\\
2016: HD156026 &1.16 &18.00 &-5.23 &$<$480 & 593 & 71\\
2017: HD160346 &0.96 &36.00 &-5.36 &........ & 3210 & 480\\
2018: HD165185 &0.62 & 5.90 &-4.43 &........ & 291 & 39\\
2019: HD165341 &0.86 &19.70 &-5.18 &........ & 1260 & 170\\
2020: HD166620 &0.87 &42.00 &-6.19 &$>$11200 & 5300 & 845\\
2021: HD176051 &0.59 &16.00 &-5.70 &........ & 2350 & 390\\
2022: HD185144 &0.79 &29.00 &-5.58 &........ & 3220 & 490\\
2023: HD187691 &0.55 &10.00 &-5.97 & 3200 & 1250 & 210\\
2024: HD190007 &1.12 &29.30 &-5.01 &$<$1760 & 1620 & 220\\
2025: HD190406 &0.61 &14.50 &-5.58 & 3160 & 1730 & 270\\
2026: HD191408 &0.87 &45.00 &-6.42 &$>$7640 & 6050 & 980\\
2027: HD194012 &0.51 & 7.00 &-5.49 &........ & 900 & 170\\
2028: HD201091 &1.17 &37.90 &-5.51 &$<$440 & 2450 & 350\\
2029: HD201092 &1.37 &48.00 &-5.32 &$<$680 & 2960 & 440\\
2030: HD206860 &0.58 & 4.70 &-4.62 &$<$880 & 237 & 34\\
2031: HD209100 &1.06 &22.00 &-5.69 &........ & 1030 & 130\\
2032: HD216803 &1.10 &10.30 &-4.54 &$<$520 & 223 & 23\\
2033: HD219834B &0.91 &42.00 &-5.49 &$>$13200 & 4820 & 760\\
2034: HD224930 &0.67 &33.00 &-5.90 &........ & 6330 & 1100\\
2035:
2036: \enddata
2037:
2038: %% Text for table notes should follow after the \enddata but before
2039: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
2040: %% in the table for each \tablenotetext.
2041:
2042: \tablenotetext{A}{This gyro age should be treated with caution because this star is a spectroscopic binary (see text, and Pourbaix, 2000).}
2043: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
2044: %with the deluxetable environment}
2045: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
2046:
2047: %\tablecomments{Occasionally, authors wish to append a short
2048: %paragraph of explanatory notes that pertain to the entire table, but
2049: %which are different than the caption. Such notes should be placed in
2050: %a {\tt tablecomments} command like this.}
2051:
2052: \end{deluxetable}
2053:
2054:
2055:
2056:
2057:
2058:
2059:
2060:
2061: %\newpage
2062: \section{Ages via gyrochronology for components of wide binaries}
2063:
2064: As we have seen, testing these (or other) stellar ages is complicated because
2065: no star apart from the Sun has an accurately determined age. However, it is
2066: possible to test the ages in a relative manner by asking whether the individual
2067: components of binary stars yield the same age. This test has been applied, with
2068: mixed results, to chromospheric ages by Soderblom et al. (1991) and
2069: Donahue (1998).
2070: We show here that gyrochronology yields substantially similar ages for both
2071: components of the three main sequence wide binary systems where {\it measured}
2072: rotation periods are available for the individual stars.
2073: [This latter requirement excludes otherwise interesting systems like
2074: 16\,Cyg\,A/B (HD186408/HD186427; e.g. Cochran et al. 1997), where the rotation
2075: periods are derived quantities (Hale 1994), and 70\,Oph\,A/B (HD165341A/B), in
2076: the Mt.\,Wilson sample, where only the A component has a measured period
2077: (Noyes et al. 1984; Baliunas et al. 1996).]
2078:
2079: %\subsection{$\kappa^1$ Ceti (HD 20630)}
2080: %$\kappa^1$Ceti is an active nearby G5 dwarf that was selected as a target for
2081: %the MOST (Microvariability and Oscillations of Stars) satellite. Exquisite
2082: %photometry of the star, obtained over an almost uninterrupted baseline of a
2083: %month showed evidence for two large starspots with rotation periods of 8.9 and
2084: %9.3 days (Rucinski et al. 2004). We use the mean of these,
2085: %9.1d, as the rotation period, and the $B-V$ color of 0.68 listed in
2086: %Rucinski et al. (2004) to plot it in the color-period diagram in Fig.\,15.
2087: %Gyrochronology yields an age of 520\,Myr, in rough agreement with, and
2088: %possibly superceding, the estimate of 750\,Myr of Gudel et al. (1997),
2089: %and the 650\,Myr estimate of Dorren \& Guinan (1994).
2090:
2091: \subsection{$\xi$\,Boo\,A/B (HD131156A/B)}
2092: $\xi$\,Boo\,A/B is a wide main sequence binary (G8V+K4V) in the Mt.\,Wilson
2093: sample. The orbit calculated by Hershey (1977) gives a period of 152\,yr and
2094: eccentricity of 0.51, suggesting no rotational interaction between the
2095: components.
2096: The Mt.\,Wilson datasets (Noyes et al. 1984; Baliunas et al. 1996; Donahue et
2097: al. 1996)
2098: provide separate color and period measurements for both components,
2099: making the system particularly valuable as a test of the mass dependence of
2100: rotation, under the assumption that binarity does not affect their rotation.
2101: Since the components of binaries are usually considered to be coeval,
2102: gyrochronology ought to give the same age for the individual components.
2103: For this system, gyrochronology yields ages of 187\,Myr and 265\,Myr for the
2104: bluer and redder components respectively (Table\,7), which gives a formal mean
2105: age for the system of 226\,$\pm$18Myr.
2106: The individual values, though not in agreement within the formal errors,
2107: are closer together than those provided by other methods.
2108: For example, the chromospheric ages for the components are 232\,Myr and
2109: 508\,Myr respectively, which also suggest a young age for the system.
2110: As regards isochrone ages, Fernandes et al. (1998) have derived
2111: an isochrone age for the system of $2 \pm 2$Gyr.
2112: More recently, Takeda et al. (2007) have derived isochrone ages
2113: for the A and B components of $<$0.76\,Gyr and $>$12.60\,Gyr respectively,
2114: attesting to the difficulty of applying the isochrone method to field stars.
2115:
2116: \subsection{61\,Cyg\,A/B (HD201091/HD201092)}
2117: There is a second, lower mass, main sequence wide binary (K5V+K7V) in the
2118: Mt.\,Wilson sample for which measured colors and periods are available.
2119: This is the 61\,Cyg\,A/B visual binary system, whose parameters (from Donahue
2120: et al. 1996, see also Baliunas et al. 1996 and Hale, 1994)
2121: are listed in Table\,7.
2122: The orbit from Allen et al. (2000) suggests a semi-major axis of 85.6 AU and
2123: eccentricity of 0.32, while that from Gorshanov et al. (2005) suggests a period
2124: of 659yr and eccentricity of 0.48.
2125: Neither of these suggest an interaction between the components.
2126: Gyrochronology yields ages of 2.12\,Gyr and 1.87\,Gyr for the A and B
2127: components respectively, suggesting a mean age for the system of
2128: 2.0$\pm$0.2\,Gyr (see Fig.\,16), where the large differential rotation of the
2129: components contributes significantly to the error.
2130: The corresponding chromospheric ages for the same stars are 2.36\,Gyr and
2131: 3.75\,Gyr respectively, again in reasonable agreement, but not as close as the
2132: gyro ages.
2133: The isochrone ages for these stars, upper limits of $<$0.44\,Gyr and
2134: $<$0.68\,Gyr respectively (Takeda et al. 2007), seem somewhat short.
2135:
2136: \subsection{$\alpha$CenA/B (HD128620/HD128621)}
2137: We now consider the famous older system $\alpha$CenA/B, its G2V and K1V
2138: components bracketing the Sun in mass, a system much studied by many
2139: researchers over the years (e.g. Guenther and Demarque, 2000;
2140: Miglio \& Montalban 2005), and of special interest to asteroseismologists.
2141: Heintz (1982) has calculated an orbit with period of $\sim$80yr and eccentricity
2142: of 0.516, suggesting that the components have not suffered rotational
2143: interactions.
2144: The published ages for the system range from 4\,Gyr to 8\,Gyr, depending on the
2145: details of the models used (see, e.g. Guenther \& Demarque 2000).
2146: Guenther and Demarque (2000) themselves derive an age range of 7.6-6.8\,Gyr,
2147: somewhat older than the Sun, depending on whether or not $\alpha$CenA has a
2148: convective core. Eggenberger et al. (2004) suggest an age of 6.5$\pm$0.3\,Gyr.
2149: Using the rotation periods provided by Ed Guinan (2006, personal communication),
2150: %http://www.renoir.vill.edu/\~astronom/obs95/node19.html,
2151: 28$\pm$3d and 36.9$\pm$1.8d for the A and B components respectively
2152: \footnote{Pizzolato et al. (2003) lists periods of 29d and 42d respectively, sourced from Saar and Osten (1997), which in turn sources the first to Hallam, Aliner \& Endal (1991), and states that the latter is estimated from CaII measurements.},
2153: and $B-V$ colors\footnote{SAB thanks David Frew for his trouble researching these colors.}
2154: of 0.67$\pm$0.02 and 0.87$\pm$0.02, we
2155: derive ages for the components of 4.6\,Gyr and 4.1\,Gyr, with a mean of
2156: 4.4$\pm$0.5\,Gyr, toward the lower end of the published ages\footnote{The Pizzolato et al. (2003) periods would yield a slightly older gyro age of 4.6\,Gyr for the system.},
2157: but in good agreement with one another\footnote{There is a third component in the $\alpha$Cen system, $\alpha$Cen\,C (Proxima Centauri), and it too has a measured period, 31$\pm$2d, but its spectral type is M5V, so it is not on the interface sequence (and hence not considered here), and it ought to follow the age dependence appropriate for the C sequence stars, but this dependence is not yet known well.}.
2158: These stars, and the corresponding isochrone are also plotted in Fig.\,16.
2159: The chromospheric ages for the $\alpha$Cen\,A and B components using $R'_{HK}$
2160: values from Henry et al. (1996) are 5.62 and 4.24\,Gyr respectively, again
2161: comparable, if not as close.
2162: In comparison, the isochrone ages from Takeda et al. (2007) for the A and B
2163: components, derived separately, are 7.84\,Gyr and $>$11.36\,Gyr respectively.
2164:
2165:
2166: \subsection{36\,Oph\,A/B/C (HD155886/HD155885/HD156026)}
2167: Finally, we consider the triple system 36\,Oph\,A/B/C, included in the
2168: Mt.\,Wilson sample. A and B are two chromospherically active K1 dwarfs, while
2169: the distant tertiary, C, is a K5 dwarf. The AB orbit has a period of
2170: $\sim$500yr, but a very high eccentricity of $\sim$0.9, implying a closest
2171: approach of A and B of order 6AU (Brosche 1960; Irwin et al. 1996).
2172: The latter fact suggests proceeding with caution, because A and B could
2173: potentially have interacted rotationally.
2174:
2175: We have used the observed periods of 20.69d, 21.11d, and 18.0d, listed in
2176: Donahue et al. (1996) and Baliunas et al. (1983) for the A, B, and C components
2177: respectively to plot these in the color-period diagram displayed in Fig.\,17\footnote{Baliunas et al (1996) list a joint period of 21d for all three components. Pizzolato et al. (2003) reference Saar \& Osten (1997) for the 20.69d and 21.11d periods for A and B, and Hempelmann et al. (1995), who in turn references Noyes et al. (1984) for the 18.0d (observed) period for C. Saar \& Osten (1997) themselves reference Donahue et al. (1996) for the A and B periods, and say that the 18.5d period is estimated from CaII measurements.}.
2178: The gyro ages for A and B are both nominally 1.43\,Gyr, but that for C is only
2179: 590$\pm$70\,Myr.
2180: We favor the lower age here because the C component is distant, while
2181: the A and B components seem to have interacted and presumably spun down to
2182: their $\sim$21d periods from the $\sim$13.4d periods that would otherwise be
2183: expected for the 590\,Myr age for the system.
2184:
2185: Interestingly, the chromospheric ages for the A, B, and C components range from
2186: 1.1\,Gyr to 1.4\,Gyr, similar to the gyro age for the A/B pair.
2187: The isochrone age for the C component only, provided by
2188: Takeda et al. (2007) is $<$480\,Myr, again suggesting a youthful system.
2189: The fact that the A and B components have essentially the same mass provides a
2190: simplification that could be quite useful to further studies of this system.
2191:
2192: For the present state of gyrochronology, we consider the particular cases
2193: presented above to represent success.
2194:
2195: %{\it We now show that the gyro clock is superior to the isochrone clock.}
2196:
2197: \clearpage
2198:
2199: \begin{deluxetable}{rlrrrr}
2200: \tabletypesize{\scriptsize}
2201: \tablecaption{Ages for wide binary systems. \label{tbl-7}}
2202: \tablewidth{0pt}
2203: \tablehead{
2204: \colhead{Star} & \colhead{$B-V$} & \colhead{$\bar{P}_{rot}$(d)\tablenotemark{a}} &\colhead{$Age_{chromo}$} & \colhead{$Age_{iso}$} & \colhead{$Age_{gyro}$\tablenotemark{b}}
2205: %\colhead{$\Theta$\tablenotemark{b}}
2206: }
2207: \startdata
2208: HD131156A & 0.76 & 6.31(0.05) & 232\,Myr &$<$760\,Myr & 187$\pm$21\,Myr\\
2209: HD131156B & 1.17 & 11.94(0.22) & 508\,Myr &$>$12600\,Myr & 265$\pm$28\,Myr\\
2210: Mean & & & & & {\bf 226$\pm$18\,Myr}\\
2211: \hline
2212: HD201091 &1.18 &35.37(1.3) & 2.36\,Gyr &$<$0.44\,Gyr & 2.12$\pm$0.3\,Gyr\\
2213: HD201092 &1.37 &37.84(1.1) & 3.75\,Gyr &$<$0.68\,Gyr & 1.87$\pm$0.3\,Gyr\\
2214: Mean & & & & & {\bf 2.0$\pm$0.2\,Gyr}\\
2215: \hline
2216: HD128620 & 0.67 & 28(3) & 5.62\,Gyr &7.84\,Gyr & 4.6$\pm$0.8\,Gyr\\
2217: HD128621 & 0.87 & 36.9(1.8) & 4.24\,Gyr &$>$11.36\,Gyr & 4.1$\pm$0.7\,Gyr\\
2218: Mean & & & & & {\bf 4.4$\pm$0.5\,Gyr}\\
2219: \hline
2220: \hline
2221: HD155886 & 0.85 & 20.69(0.4) & 1.1\,Gyr & ...... & 1.42$\pm$0.19\,Gyr\\
2222: HD155885 & 0.86 & 21.11(0.4) & 1.2\,Gyr & ...... & 1.44$\pm$0.20\,Gyr\\
2223: HD156026 & 1.16 & 18.0(1.0) & 1.4\,Gyr &$<$0.48\,Gyr &{\bf 0.59$\pm$0.07\,Gyr}\\
2224: \enddata
2225:
2226: %% Text for table notes should follow after the \enddata but before
2227: %% the \end{deluxetable}. Make sure there is at least one \tablenotemark
2228: %% in the table for each \tablenotetext.
2229:
2230: %\tablenotetext{a}{Sample footnote for table~\ref{tbl-1} that was generated
2231: %with the deluxetable environment}
2232: %\tablenotetext{b}{Another sample footnote for table~\ref{tbl-1}}
2233: \tablenotetext{a}{Differential rotation is the main contributor to the period errors in the parentheses.}
2234: \tablenotetext{b}{Boldface figures denote the final gyro age for each system}
2235:
2236: %\tablecomments{Occasionally, authors wish to append a short
2237: %paragraph of explanatory notes that pertain to the entire table, but
2238: %which are different than the caption. Such notes should be placed in
2239: %a {\tt tablecomments} command like this.}
2240:
2241: \end{deluxetable}
2242:
2243:
2244: %\newpage
2245: \section{Comparison with isochrone ages}
2246:
2247: %{\bf Introduction: }\\
2248: A uniform comparison of gyro- and isochrone ages was not possible until
2249: Takeda et al. (2007) submitted a manuscript to the Astrophysical Journal
2250: Supplement subsequent to this submission.
2251: This paper contains a very careful derivation of isochrone ages for the
2252: $\sim$1000 stars in the Spectroscopic Properties of Cool Stars Catalog (SPOCS).
2253: This catalog (Valenti \& Fischer 2005) itself consists of high-resolution
2254: echelle spectra and their detailed analysis of over 1000 nearby F-, G- and
2255: K-type stars obtained through the Keck, Lick, and Anglo-Australian Telescope
2256: planet search programs, including $\sim$100 stars with known planetary
2257: companions.
2258:
2259: %{\bf What Takeda et al did: }\\
2260: Takeda et al. (2007) conduct a Bayesian analysis of the stellar parameters
2261: using reasonable priors to generate a probability distribution function (PDF)
2262: for the age of each star. This method permits the identification of a
2263: `well-defined' age for a star if the PDF peaks appropriately, or just as
2264: importantly, the derivation of an isochrone upper- or lower limit for the age.
2265: Indeed, for most of the stars common to our sample and theirs, they derive
2266: only such a limit, as a glance at the column for isochrone ages in Tables 3, 5
2267: and 6 shows. However, for 26 of these (common) stars, Takeda et al. (2007) list
2268: well-defined ages, and these can be compared to the corresponding gyro ages.
2269:
2270: %{\bf The comparison: }\\
2271: This comparison is shown in Fig.\,18. Of these 26 stars, only 3 lie above the
2272: line of equality, and 13 have isochrone ages within a factor of two of
2273: the gyro ages, all higher than the corresponding gyro ages. In fact, the median
2274: isochrone age is 2.7 times the median gyro age. Evidently, the Bayesian
2275: technique used still does not eliminate the known
2276: bias in the isochrone ages towards older values.
2277:
2278: %{\bf Their binary ages suck: }\\
2279: In fact, the same test applied to the binary systems in the previous
2280: section with respect to gyro ages yields uncertain results with respect to
2281: these isochrone ages. Indeed, of the 9 stars under consideration, only one
2282: ($\alpha$\,Cen\,A) has a well-defined isochrone age, and the rest upper- or
2283: lower limits. These stars are also plotted in Fig.\,18, with dashed lines
2284: joining the binary components, and arrows indicating upper- or lower limits.
2285:
2286: %{\bf Summation: }\\
2287: In summary, it would seem that the isochrone ages are still problematical,
2288: despite the careful analysis of Takeda et al. (2007). Of course, as we have
2289: noted in the introduction, it is perhaps not fair and evidently not possible,
2290: to use slowly varying parameters to derive precise ages for stars on the main
2291: sequence. The two methods are, however, complementary in that it might be
2292: preferable to use gyro ages on the main sequence, and isochrone ages off it.
2293:
2294: %\newpage
2295: \section{Conclusions}
2296:
2297: %{\bf Introduction, and basis of the technique: }\\
2298: %We have offered additional evidence (over Barnes 2003a) for the existence of
2299: %distinct sequences in the rotation period distributions for late-type stars.
2300: The rotation period distributions of solar and late-type stars suggest that
2301: coeval stars are preferentially located on one of two sequences.
2302: The mass- and age dependencies of one of these sequences, the interface
2303: sequence, are shown to be universal, shared by both cluster and field stars,
2304: and we have specified them using simple functions, generalizing the dependence
2305: originally suggested by Skumanich (1972). The mass dependence is derived
2306: observationally using a series of open clusters, and the age dependence,
2307: roughly $\sqrt t$, is specified via a solar calibration.
2308:
2309: %{\bf Development of the technique: }\\
2310: %These functions enable the identification of interface sequence stars among
2311: %field stars, and the excising of non-interface sequence stars from the sample.
2312: %Furthermore, their systematic behavior with mass and age permits their use
2313: %for gyrochronology, the derivation of ages for individual late-type main
2314: %sequence field or cluster stars from their rotation periods and colors/masses.
2315: %Errors are calculated for such ages, based on the data currently available,
2316: %and shown to be roughly 15\% for individual stars.
2317:
2318: %{\bf Development of the technique: }\\
2319: The dependencies are inverted to provide the age of a star as a function
2320: of its rotation period and color, a procedure we call gyrochronology.
2321: Errors are calculated for such ages, based on the data currently available,
2322: and shown to be roughly 15\% (plus possible systematic errors) for
2323: individual stars.
2324: Because the dependencies are universal, they must also apply to field stars,
2325: but the derivation of such ages requires excising pre-I\,sequence stars,
2326: facilitated by their location below the I\,sequence in color-period diagrams.
2327: The short lifetime of this pre-I\,sequence phase assures us that all such stars
2328: are less than a couple of hundred million years in age.
2329:
2330: %{\bf Demonstration of how to use the technique: }\\
2331: Using this formalism, we have calculated ages via gyrochronology for individual
2332: stars in three illustrative groups of field stars, and listed them along with
2333: the errors.
2334: For the first group, the Mt.\,Wilson stars, these ages are shown to be in
2335: general agreement with the chromospheric ages, except that stars bluer than the
2336: Sun have systematically higher chromospheric ages, the median chromospheric age
2337: being higher by about 33\%.
2338: The majority of the second group, from Strassmeier et al. (2000), are shown to
2339: be younger than 1\,Gyr, in keeping with the selection of the sample
2340: for activity, which correlates negatively, as expected, with gyro age.
2341: The third group, from Pizzolato et al. (2003), are shown to be somewhat older,
2342: partially due to an overlap with the Mt.\,Wilson sample, and their X-ray fluxes
2343: are shown to decay systematically with gyro age.
2344: We have shown that gyrochronology yields similar ages for both components of
2345: three wide binary systems, $\xi$\,Boo\,A/B, 61\,Cyg\,A/B, and
2346: $\alpha$\,Cen\,A/B.
2347: The 36\,Oph\,A/B/C triple system shows signs of rotational interaction between
2348: the A and B components.
2349: Finally, the recent Takeda et al. (2007) isochrone ages appear to be inferior
2350: to the gyro ages for the same main sequence stars.
2351:
2352: %{\bf One sentence summary: }\\
2353: Thus, we have re-investigated the use of a rotating star as a clock,
2354: clarified and improved its usage, calibrated it using the Sun,
2355: and demonstrated that it keeps time well.
2356:
2357: %% If you wish to include an acknowledgments section in your paper,
2358: %% separate it off from the body of the text using the \acknowledgments
2359: %% command.
2360:
2361: %% Included in this acknowledgments section are examples of the
2362: %% AASTeX hypertext markup commands. Use \url without the optional [HREF]
2363: %% argument when you want to print the url directly in the text. Otherwise,
2364: %% use either \url or \anchor, with the HREF as the first argument and the
2365: %% text to be printed in the second.
2366:
2367: \acknowledgments
2368:
2369: The word ``gyrochronology'' was inspired by the work of A.\,E.\,Douglass
2370: on dendrochronology at Lowell Observatory.
2371: SAB would like to acknowledge Sabatino Sofia as a constant source of
2372: intellectual and moral support and many discussions, and Charles Bailyn for
2373: initially suggesting the removal of the age dependence. Marc Buie, Will
2374: Grundy, Wes Lockwood, Bob Millis, Byron Smith, Brian Skiff and my other
2375: colleagues at Lowell have supported me in numerous ways. Stephen Levine read
2376: the manuscript closely, and found an algebraic error.
2377: David James, Heather Morrison, Steve Saar, Sukyoung Yi and an anomymous referee
2378: are gratefully acknowledged for input on a prior version of the paper.
2379: The paper owes much to the baristas at Late For The Train, Flagstaff.
2380: Finally, this material is based upon work partially supported by the National
2381: Science Foundation under Grant No. AST-0520925.
2382:
2383:
2384:
2385: \appendix
2386: \section{Appendix: Derivation of the error on the index $n$}
2387:
2388: By definition,
2389: \begin{equation}
2390: {
2391: P = f(B-V).g(t) = a\,x^b\,t^n
2392: }
2393: \end{equation}
2394: Taking natural logarithms and rearranging, we get
2395: \begin{equation}
2396: {
2397: n = \frac{ln\,P_{\odot} - ln\,a - b\,ln\,x_{\odot}}{ln\,t_{\odot}} = \frac{U}{V}
2398: }
2399: \end{equation}
2400: Differentiating yields
2401: \begin{equation}
2402: {
2403: \frac{dn}{n} = \frac{dU}{U} - \frac{dV}{V}
2404: }
2405: \end{equation}
2406: or
2407: \begin{equation}
2408: {
2409: \frac{dn}{n} = \frac{1}{U} [\frac{dP_{\odot}}{P_{\odot}} - \frac{da}{a}
2410: - b \frac{dx_{\odot}}{x_{\odot}} - ln\,x_{\odot}\,db]
2411: - \frac{dt_{\odot}}{t_{\odot}\,ln\,t_{\odot}}
2412: }
2413: \end{equation}
2414: Adding the errors in quadrature yields
2415: \begin{equation}
2416: {
2417: (\frac{\delta n}{n})^2 = (\frac{\delta t_{\odot}}{t_{\odot}\,ln\,t_{\odot}})^2
2418: + \frac{1}{U^2} [ (\frac{\delta P_{\odot}}{P_{\odot}})^2
2419: + (\frac{\delta a}{a})^2
2420: + (b\frac{\delta x_{\odot}}{x_{\odot}})^2
2421: + (ln\,x_{\odot}\,\delta b)^2
2422: ]
2423: }
2424: \end{equation}
2425: For the error in the age of the Sun (4566\,Myr; $ln\,t_{\odot}=8.426$),
2426: we adopt the value of 50\,Myr\footnote{Allegre et al. (1995) list the
2427: impressively small error of $+2/-1$\,Myr
2428: (in agreement with the present day precision of radioactive dating techniques)
2429: for the age of the formation of the Allende refractory inclusions,
2430: generally accepted as the age of the Earth/meteorites/Solar System. However,
2431: we astronomers do not know what event in the Sun's history corresponds to this
2432: point. Is this the zero age main sequence, or the birthline 43\,Myr earlier
2433: (Barnes \& Sofia, 1996), or some other event entirely? In view of these
2434: uncertainties, we adopt an error of 50\,Myr in the age of the Sun.},
2435: for that in the rotation period, 1\,d (consistent with the measured range
2436: in the solar rotation period - see section 4 and Donahue et al. 1996),
2437: and for that in the solar $B-V$ color ($x = B-V_{\odot} = 0.242$),
2438: we adopt the value 0.01.
2439: From section\,2, $a = 0.7725 \pm 0.011$ and $b = 0.601 \pm 0.024$.
2440: Input of these values yields
2441: \begin{equation}
2442: {
2443: (\frac{\delta n}{n})^2 = (\frac{50}{4566 \times 8.43})^2
2444: + \frac{1}{4.37^2} [ (\frac{1}{26.09})^2
2445: + (\frac{0.011}{0.7725})^2
2446: + (0.601\frac{0.01}{0.242})^2
2447: + (-1.419 \times 0.024)^2
2448: ]
2449: }
2450: \end{equation}
2451: or
2452: \begin{equation}
2453: {
2454: (\frac{\delta n}{n})^2 = 1.69 \times 10^{-6}
2455: + 10^{-6} [ 77.4 + 10.6 + 32.3 + 60.7 ]
2456: = 182.6 \times 10^{-6}
2457: }
2458: \end{equation}
2459: or\footnote{Note that the largest terms come from the differential rotation of
2460: the sun and the index $b$, while the age error of the sun contributes little to
2461: the error in $n$.},
2462: \begin{equation}
2463: {
2464: \frac{\delta n}{n} = 1.37 \times 10^{-2}
2465: }
2466: \end{equation}
2467: so that
2468: \begin{equation}
2469: {
2470: n = 0.5189 \pm 0.0070
2471: }
2472: \end{equation}
2473: which shows that the index $n$ is determined well.
2474:
2475: %% The reference list follows the main body and any appendices.
2476: %% Use LaTeX's thebibliography environment to mark up your reference list.
2477: %% Note \begin{thebibliography} is followed by an empty set of
2478: %% curly braces. If you forget this, LaTeX will generate the error
2479: %% "Perhaps a missing \item?".
2480: %%
2481: %% thebibliography produces citations in the text using \bibitem-\cite
2482: %% cross-referencing. Each reference is preceded by a
2483: %% \bibitem command that defines in curly braces the KEY that corresponds
2484: %% to the KEY in the \cite commands (see the first section above).
2485: %% Make sure that you provide a unique KEY for every \bibitem or else the
2486: %% paper will not LaTeX. The square brackets should contain
2487: %% the citation text that LaTeX will insert in
2488: %% place of the \cite commands.
2489:
2490: %% We have used macros to produce journal name abbreviations.
2491: %% AASTeX provides a number of these for the more frequently-cited journals.
2492: %% See the Author Guide for a list of them.
2493:
2494: %% Note that the style of the \bibitem labels (in []) is slightly
2495: %% different from previous examples. The natbib system solves a host
2496: %% of citation expression problems, but it is necessary to clearly
2497: %% delimit the year from the author name used in the citation.
2498: %% See the natbib documentation for more details and options.
2499:
2500: %\clearpage
2501:
2502: \begin{thebibliography}{}
2503: %\bibitem[Adams et al. (1998)] {aww98} Adams, N.R., Walter, F.M., \& Wolk, S.J.,
2504: % 1998, AJ, 116, 237
2505: \bibitem[Allain et al. (1996)] {abp96} Allain, S., Bouvier, J., Prosser, C.,
2506: Marschall, L.A. and Laaksonen, B.D., 1996, A\&A, 305, 498
2507: \bibitem[Allegre et al. (1995)]{aea95} Allegre, C.J., Manhes, G., \& Gopel, C.,
2508: 1995, Geochemica et Cosmochimica Acta, 59(8), 1445
2509: \bibitem[Allen et al. (2000)]{aea00} Allen, C., Poveda, A, \& Herrera, M.A.,
2510: 2000, A\&A, 356, 529
2511: %\bibitem[Attridge \& Herbst (1992)] {ah92} Attridge, J.M., \& Herbst, W., 1992,
2512: % ApJL, 398, 61
2513: %\bibitem[Babcock (1961)] {b61} Babcock, H.W., 1961, ApJ, 133, 572
2514: \bibitem[Baliunas et al. (1983)]{bea83} Baliunas, S., Vaughan, A.H., Hartmann,
2515: L., Middelkoop, F., Mihalas, D., Noyes, R.W., Preston, G.W., Frazer, J.,
2516: \& Lanning, H., 1983, ApJ, 275, 752
2517: \bibitem[Baliunas et al. (1995)]{bea95} Baliunas, S., \& 27 co-authors, 1995,
2518: ApJ, 438, 269
2519: \bibitem[Baliunas et al. (1996)]{bss96} Baliunas, S., Sokoloff, D. \& Soon, W.,
2520: 1996, ApJL, 457, 99
2521: \bibitem[Barnes (1998)]{b98} Barnes, S.A., 1998, PhD Thesis, Yale University
2522: \bibitem[Barnes (2001)]{b01} Barnes, S.A., 2001, ApJ, 561, 1095
2523: \bibitem[Barnes (2003a)]{b03a} Barnes, S.A., 2003a, ApJ, 586, 464
2524: \bibitem[Barnes (2003b)]{b03b} Barnes, S.A., 2003b, ApJL, 586, 145
2525: \bibitem[Barnes \& Sofia (1996)]{bs96} Barnes, S.A. \& Sofia, S., 1996, ApJ,
2526: 462, 746
2527: \bibitem[Barnes \& Sofia (1998)]{bs98} Barnes, S.A. \& Sofia, S., 1998, BAAS,
2528: 30, 917
2529: %\bibitem[Barnes et al. (2001)]{bsp01} Barnes, S.A., Sofia, S. \& Pinsonneault,
2530: % M.H., 2001, ApJ, 548, 1071.
2531: \bibitem[Barnes et al. (1999)]{bsps99} Barnes, S.A., Sofia, S., Prosser, C.F.
2532: \& Stauffer, J.R., 1999, ApJ, 516, 263
2533: \bibitem[Basri et al. (1996)]{bea96} Basri, G., Marcy, G.W., \& Graham, J.,
2534: 1996, ApJ, 458, 600
2535: %\bibitem[Basri et al. (1996)]{bea96} Basri, G., Marcy, G.W., Oppenheimer, B.,
2536: % Kulkarni, S., \& Nakajima, T., 1996, in 9th Cambridge Workshop on
2537: % Cool Stars, Stellar Systems, and the Sun, Pallavicini \& Dupree (eds),
2538: % ASP Conf. Series, 109, 587
2539: %\bibitem[Bouvier et al. (1997)]{bfa97} Bouvier, J., Forestini, M., \& Allain,
2540: % S., 1997, A\&A, 326, 1023
2541: %\bibitem[Brandenburg et al. (1998)]{bst98} Brandenburg, A., Saar, S.H., \&
2542: % Turpin, C.R., 1998, ApJL, 498, 51
2543: \bibitem[Brosche (1960)]{b60} Brosche, P., 1960, AN, 285, 261
2544: %\bibitem[Cameron \& Li (1994)] {cl94}Cameron, A.C. \& Li, J., 1994,
2545: % A\&A, 269, 1099
2546: %\bibitem[Cameron \& Campbell (1993)] {cc93} Cameron, A.C. \& Campbell, C.G.,
2547: % 1993, A\&A, 274, 309
2548: %\bibitem[Cameron et al. (1995)] {ccq95} Cameron, A.C., Campbell, C.G. \&
2549: % Quaintrell, H., 1995, A\&A, 298, 133
2550: %\bibitem[Chaboyer et al. (1995)]{cdp95} Chaboyer, B., Demarque, P. \&
2551: % Pinsonneault, M., 1995, ApJ, 441, 865
2552: %\bibitem[Charbonneau \& MacGregor (1997a)]{cm97a} Charbonneau, P. \& MacGregor,
2553: % K.B., 1997a, ApJ, 486, 484
2554: %\bibitem[Charbonneau \& MacGregor (1997b)]{cm97b} Charbonneau, P. \& MacGregor,
2555: % K.B., 1997b, ApJ, 486, 502
2556: %\bibitem[Choi \& Herbst (1996)]{ch96} Choi, P.I. \& Herbst, W., 1996, AJ, 111,
2557: % 283
2558: \bibitem[Cochran et al. (1997)]{cea97} Cochran, W.D., Hatzes, A.P., Butler, R.P.
2559: , \& Marcy, G.W., 1997, ApJ, 483, 457
2560: \bibitem[Cowley (19870]{c87} Cowley, A.P., 1976, PASP, 88, 95
2561: %\bibitem[Delfosse et al. (1998)]{dfpm98} Delfosse, X., Forveille, T., Perrier,
2562: % C., \& Mayor, M., 1998, A\&A, 331, 581
2563: \bibitem[Demarque \& Larson (1964)]{dl64} Demarque, P.D. \& Larson, R.B., 1964,
2564: AJ, 140, 544
2565: \bibitem[Demarque et al. (2004)]{dea04} Demarque, P., Woo, J.-H., Kim, Y.-C.,
2566: Yi, S.K., 2004, ApJS, 155, 667
2567: %\bibitem[Dolan \& Mathieu (2002)]{dm02} Dolan, C.J. \& Mathieu, R.D., 2002,
2568: % AJ, 123, 387
2569: \bibitem[Donahue (1998)]{d98} Donahue, R.A., 1998, in ASP Conf. Ser. 154,
2570: Tenth Cambridge Workshop on Cool Stars, Stellar Systems and the Sun,
2571: ed. R.A. Donahue and J.A. Bookbinder, 1235
2572: \bibitem[Donahue et al. (1996)]{dea96} Donahue, R.A., Saar, S.H. \& Baliunas,
2573: S.L., 1996, ApJ, 466, 384
2574: \bibitem[Dorren \& Guinan (1994)]{dg94} Dorren, J.D., \& Guinan, E.F., 1994,
2575: ApJ, 428, 805
2576: \bibitem[Duncan (1984)]{d84} Duncan, D.K., 1984, AJ, 89, 515
2577: \bibitem[Duncan et al. (1984)]{dea85} Duncan, D.K., Baliunas, S.L., Noyes, R.W.,
2578: Vaughan, A.H., Frazer, J., \& Lanning, H.H., 1984, PASP, 96, 707
2579: %\bibitem[Durney et al. (1993)]{ddr93} Durney, B.R., De\,Young, D.S., \&
2580: % Roxburgh, I.W., Solar Physics, 145, 207
2581: %\bibitem[Duvall et al. (1984)]{duv84} Duvall, T.L., Jr., Dziembowski, W.A.,
2582: % Goode, P.R., Gough, D.O., Harvey, J.W. \& Leibacher, J. W., 1984,
2583: % Nature, 310, 22
2584: \bibitem[ESA, 1997]{esa97} ESA, 1997, The Hipparcos and Tycho Catalogues,
2585: ESA SP-1200
2586: %\bibitem[Edwards et al. (1993)]{esh93} Edwards, S. Strom. S. E., Herbst, W.,
2587: % Attridge, J., Michael Merill, K. , Probst, R. and Gatley, I., 1993,
2588: % AJ, 106, 372
2589: %\bibitem[Eff-Darwich et al. (2002)]{ekj02} Eff-Darwich, A., Korzennik, S.G.,
2590: % Jimenez-Reyes, S.J., 2002, ApJ, 573, 857
2591: \bibitem[Eggenberger et al. (2004)]{eea04} Eggenberger, P., Charbonnel, C.,
2592: Talon, S., Meynet, G., Maeder, A., Carrier, F., \& Bourban, G., 2004,
2593: A\&A, 422, 247
2594: %\bibitem[Endal and Sofia (1981)]{es81} Endal, A.S. \& Sofia, S., 1981, ApJ,
2595: % 243, 625
2596: %\bibitem[Feigelson et al. (2003)]{fei03} Feigelson, E.D., Gaffney III, J.A.,
2597: % Garmire, G., Hillenbrand, L.A. \& Townsley, L., 2003, ApJ, in press
2598: % (astro-ph/0211049)
2599: \bibitem[Fernandes et al. (1998)]{fea98} Fernandes, J., Lebreton, Y., Baglin, A.
2600: \& Morel, P., 1998, A\&A, 338, 455
2601: %\bibitem[Gaidos, Henry \& Henry (2000)] {ghh00} Gaidos, E.J., Henry, G.W., \&
2602: % Henry, S.M., 2000, AJ, 120, 1006
2603: \bibitem[Giampapa et al. (2006)]{gea06} Giampapa, M.S., Hall, J.C., Radick,
2604: R.R., \& Baliunas, S.L., 2006, ApJ, 651, 444
2605: \bibitem[Girardi et al. (2002)]{gea02} Girardi, L., Bertelli, G., Bressan, A.,
2606: Chiosi, C., Groenewegen, M.A.T., Marigo, P., Salasnich, B., Weiss, A.,
2607: 2002, A\&A, 391, 195
2608: %\bibitem[Goode et al. (1991)]{goo91} Goode, P.R., Dziembowski, W.A.,
2609: % Korzennik, S.G., Rhodes, E.J. Jr., 1991, ApJ, 367, 649
2610: \bibitem[Gorshanov et al. (2005)]{gea05} Gorshanov, D.L., Shakht, N.A.,
2611: Kisselev, A.A., \& Poliakow, E.V., 2005, in {\it Dynamics of
2612: Populations of Planetary Systems}, eds. Z. Knezevic \& A. Milani, IAU
2613: Coll 197, p.91
2614: %\bibitem[Gough (1982)]{gou82} Gough, D.O., 1982, Nature, 298, 334
2615: \bibitem[Gudel (2004)]{g04} Gudel, M., 2004, A\&ARv, 12, 71
2616: %\bibitem[Gudel et al. (1997)]{gea97} Gudel, M., Guinan, E.F., \& Skinner, S.L.,
2617: % 1997, ApJ, 483, 947
2618: \bibitem[Guenther \& Demarque (2000)]{gd00} Guenther, D.B., \& Demarque, P.,
2619: 2000, ApJ, 531, 503
2620: \bibitem[Hale (1994)]{h94} Hale, A., 1994, AJ, 107, 306
2621: \bibitem[Hallam et al. (1991)]{hea91} Hallam, K.L., Aliner, B., \& Endal, A.S.,
2622: 1991, ApJ, 372, 610
2623: %\bibitem[Hauser \& Marcy (1999)]{hm99} Hauser, H.M. \& Marcy, G.W., 1999, PASP,
2624: 111, 321
2625: \bibitem[Heintz (1982)]{h82} Heintz, W.D., 1982, Observatory, 102, 42
2626: \bibitem[Hempelmann et al. (1995)]{hea95} Hempelmann, A., Schmitt, J.H.M.M.,
2627: Schultz, M., Rudiger, G., \& Stepien, K., 1995, A\&A, 294, 515
2628: \bibitem[Henry et al. (1995)]{hea95} Henry, G.W., Fekel, F.C. \& Hall, D.S.,
2629: 1995, AJ, 110, 2926
2630: \bibitem[Henry et al. (1996)]{hea96} Henry, T.J., Soderblom, D.R., Donahue, R.A.
2631: , \& Baliunas, S.L., 1996, AJ, 111, 439
2632: %\bibitem[Herbst et al. (2001)]{hbm01} Herbst, W. Bailer-Jones, C.A.L., \&
2633: % Mundt, R., 2001, ApJL, 554, 197
2634: \bibitem[Hershey (1977)]{h77} Hershey, J.L., 1977, AJ, 82, 179
2635: \bibitem[Holmberg et al. (2006)]{Hol05} Holmberg, J., Flynn, C. \& Portinari,
2636: L., 2006, MNRAS, 367, 449
2637: \bibitem[Ihaka \& Gentleman (1996)]{IG96} Ihaka \& Gentleman, 1996, Journal of
2638: Computational and Graphical Statistics, 5, 299
2639: \bibitem[Irwin et al. (1996)]{iea96} Irwin, A.W., Yang, S.L.S., \& Walker,
2640: G.A.H., 1996, PASP, 108, 580
2641: \bibitem[James et al. (2007)]{jea07} James, D.J.J., Barnes, S.A., \& Meibom, S.,
2642: 2007, ApJ, submitted.
2643: %\bibitem[James et al. (2000)]{jam00} James, D.J., Jardine, M.M., Jeffries, R.D.
2644: % , Randich, S., Cameron, A.C. \& Ferreira, M., 2000, MNRAS, 318, 1217
2645: \bibitem[Jorgensen \& Lindegren (2005)]{jl05} Jorgensen, B.R. \& Lindegren, L.,
2646: 2005, A\&A, 436, 127
2647: %\bibitem[Kawaler (1988)]{k88} Kawaler, S.D., 1988, ApJ, 333, 236
2648: \bibitem[Kawaler (1989)]{k89} Kawaler, S.D., 1989, ApJ, 343, 65
2649: %\bibitem[Keppens et al. (1995)]{kmc1995} Keppens, R., MacGregor, K.B., \&
2650: % Charbonneau, P., 1995, A\&A, 294, 469
2651: %\bibitem[Kim \& Demarque (1996)]{kd96} Kim, Y.-C. \& Demarque, P.D., 1996,
2652: % ApJ, 457, 340
2653: \bibitem[Kim et al. (2002)]{kea02} Kim, Y.-C., Demarque, P., Yi, S.K.,
2654: Alexander, D.R., 2002, ApJS, 143, 499
2655: \bibitem[Kraft (1967)]{k67} Kraft, R., 1967, ApJ, 150, 551
2656: %\bibitem[Krishnamurthi et al. (1997)]{kea97} Krishnamurthi, A. , Pinsonneault,
2657: % M.H., Barnes, S., \& Sofia, S., 1998, ApJ, 480, 303
2658: \bibitem[Krishnamurthi et al. (1998)]{kea98} Krishnamurthi, A. \& 15 co-authors,
2659: 1998, ApJ, 493, 914
2660: \bibitem[Kunte et al. (1988)]{kea88} Kunte, P.K., Rao, A.R., Vahia, M.N., 1988,
2661: Ap\&SS, 143, 207
2662: \bibitem[Lachaume et al. (1999)] {lea99} Lachaume, R., Dominik, C., Lanz, T.,
2663: \& Habing, H. J., 1999, A\&A, 348, 897
2664: %\bibitem[Lamm et al. (2002)] {lmbh02} Lamm, M., Mundt, R., Bailer-Jones,
2665: % C.A.L, Herbst, W., 2002, AAS, 200, 70.02
2666: \bibitem[Latham et al. (2002)]{lea02} Latham, D.W., Stefanik, R.P., Torres, G.,
2667: Davis, R.J., Mazeh, T., Carney, B.W., Laird, J.B. \& Morse, J.A., 2002,
2668: AJ, 124, 1144
2669: %\bibitem[Leighton (1969)] {l69} Leighton, R.B., 1969, ApJ, 156,1
2670: %\bibitem[MacGregor \& Brenner (1991)]{mb91} MacGregor, K.B. \& Brenner, M.,
2671: % 1991, ApJ, 376, 204
2672: %\bibitem[Martin \& Zapatero-Osorio (1997)]{mzo97} Martin, E.L. \&
2673: % Zapatero-Osorio, M.R., 1997, MNRAS, 286, 17
2674: \bibitem[Matthews et al. (2000)]{mea00} Matthews, J.M., Kuschnig, R., Walker,
2675: G.A., Pazder, J., Johnson, R., Skaret, K., Shkolnik, E., Lanting, T.,
2676: Morgan, J.P., \& Sidhu, S., 2000, in The Impact of Large-Scale Surveys
2677: on Pulsating Star Research, ASP Conf. Series, 203, 74
2678: %\bibitem[Mestel (1968)]{m68} Mestel, L., 1968, MNRAS, 138, 359
2679: %\bibitem[Mestel \& Spruit (1987)]{ms87} Mestel, L., \& Spruit, H.C., 1987,
2680: % MNRAS, 226, 57
2681: \bibitem[Miglio \& Montalban (2005)]{mb05} Miglio, A. \& Montalban, J., 2005,
2682: A\&A, 441, 615
2683: %\bibitem[Mohanty \& Basri (2002)]{mb02} Mohanty, S. \& Basri, G., 2002,
2684: % astro-ph/0201455
2685: %\bibitem[Mohanty et al. (2002)]{mbs02} Mohanty, S., Basri, G., Shu, F., Allard,
2686: % F. \& Chabrier, G., 2002, ApJ, 571, 469
2687: \bibitem[Murthy \& Patterson (1962)]{mp62} Murthy, V.R. \& Patterson, C.C.,
2688: 1962, J. Geophys. Res., 67, 1161
2689: \bibitem[Noyes et al. (1984)]{nhb67} Noyes, R.W., Hartmann, L.W., Baliunas,
2690: S.L., Duncan, D.K. \& Vaughan, A.H., 1984, ApJ, 279, 763
2691: %\bibitem[Olah, Jurcsik \& Strassmeier (2003)]{ojs03} Olah, K., Jurcsik, J., \&
2692: % Strassmeier, K.G., 2003, A\&A, 410, 685
2693: %\bibitem[Parker (1993)]{p93} Parker, E.N., 1993, ApJ, 408, 707
2694: \bibitem[Patten \& Simon (1996)]{ps96} Patten, B.M. \& Simon, T., 1996, ApJS,
2695: 106, 489
2696: \bibitem[Patterson (1953)]{p53} Patterson, C.C., 1953, in Report by the
2697: Subcommittee on Nuclear Processes in Geological Settings, National
2698: Research Council, National Academy of Sciences, Washington, D.C., p.36
2699: \bibitem[Patterson (1955)]{p55} Patterson, C.C., 1955, Geochimica et
2700: Cosmochimica Acta, 7, 151
2701: \bibitem[Patterson (1955)]{p56} Patterson, C.C., 1956, Geochimica et
2702: Cosmochimica Acta, 10, 230
2703: \bibitem[Patterson et al. (1955)]{pea55} Patterson, C.C., Tilton, G.R., \&
2704: Ingraham, M.G., 1955, Science, 121, 69
2705: \bibitem[Pallavicini et al. (1981)]{pea81} Pallavicini, R., Golub, L., Rosner,
2706: R., Vaiana, G.S., Ayres, T., \& Linsky, J.L., 1981, ApJ, 248, 279
2707: \bibitem[Perryman et al. (1997)]{pea97} Perryman, M.A.C, Lindegren, L.,
2708: Kovalevsky, J., Hog, E., \& 16 co-authors, A\&A, 323, L49
2709: \bibitem[Pinsonneault et al. (1989)]{pksd89} Pinsonneault, M. H., Kawaler, S.
2710: D., Sofia, S. and Demarque, P., 1989, ApJ, 338, 424
2711: %\bibitem[Pinsonneault et al. (1990)]{pkd90} Pinsonneault, M. H., Kawaler, S.
2712: % D. and Demarque, P., 1990, ApJS, 74, 501
2713: \bibitem[Pizzolato et al. (2003)]{pea03} Pizzolato, N., Maggio, A., Micela, G.,
2714: Sciortino, S., \& Ventura, P., 2003, A\&A, 397, 147
2715: \bibitem[Pojmanski (2001)]{p01} Pojmanski, G., in Small Telescope Astronomy on
2716: Global Scales, eds. B. Paczynski, W.-P. Chen, \& C. Lemme, ASP Conf.
2717: Series, 246, 53
2718: \bibitem[Pont \& Eyer (2004)]{pe04} Pont, F. \& Eyer, L., 2004, MNRAS, 351, 487
2719: \bibitem[Pourbaix (2000)]{p00} Pourbaix, D., 2000, A\&AS, 145, 215
2720: \bibitem[Prosser \& Grankin (1997)]{pg97} Prosser, C.F. \& Grankin, K., 1997,
2721: CfA preprint 4539
2722: %\bibitem[Prosser et al. (1996)]{pea96} Prosser, C.F., Randich, S., Stauffer,
2723: % J.R., Schmitt, J.H.M.M. \& Simon, T., 1996, AJ, 112, 1570
2724: %\bibitem[Prosser et al. (1993)]{pea93} Prosser, C.F., et al., 1993, PASP, 105,
2725: % 1407
2726: %\bibitem[Queloz et al. (1998)]{que98} Queloz, D., Allain, S., Mermilliod,
2727: % J.-C., Bouvier, J. \& Mayor, M., 1998, A\&A, 335, 183
2728: \bibitem[Radick et al. (1987)]{rtl87} Radick, R.R., Thompson, D.T., Lockwood,
2729: G.W., Duncan, D.K, and Baggett, W.E., 1987, ApJ, 321, 459
2730: \bibitem[Radick et al. (1990)]{rsl90} Radick, R.R., Skiff, B.A. \& Lockwood,
2731: G.W., 1990, ApJ, 353, 524
2732: %\bibitem[Radler et al. (1990)]{rea90} Radler, K.-H., Wiedemann, E.,
2733: % Brandenburg, A., Meinel, R. \& Tuominen, I., 1990, A\&A, 239, 413
2734: \bibitem[Randall et al. (2005)]{rea05} Randall, S.K., Matthews, J.M.,
2735: Fontaine, G., Rowe, J., Kuschnig, R., Green, E. M., \& 8 co-authors,
2736: 2005, ApJ, 633, 460
2737: %\bibitem[Randich (1997)] {r97} Randich, S., 1997, in Cool Stars in Clusters
2738: % and Associations: Magnetic Activity and Age Indicators, eds. Micela,
2739: % Pallavicini and Sciortino, Mem. Soc. Ast. Italiana, 1997, 68-4, 971
2740: %\bibitem[Randich (2000)] {r00} Randich, S., 2000, in ASP Conf. Ser., 198,
2741: % Stellar Clusters and Associations: Convection, Rotation, and Dynamos,
2742: % ed. R. Pallavicini, G. Micela \& S. Sciortino (San Francisco: ASP), 401
2743: %\bibitem[Randich et al. (1995)] {rea95} Randich, S., Schmitt, J.H.M.M.,
2744: % Prosser, C.F., \& Stauffer, J.R., 1995, A\&A, 300, 134
2745: %\bibitem[Randich et al. (1996)] {rea96} Randich, S., Schmitt, J.H.M.M.,
2746: % Prosser, C.F., \& Stauffer, J.R., 1996, A\&A, 305, 785
2747: \bibitem[Rebolo et al. (1992)]{Rea92} Rebolo, R., Martin, E.L., \& Maguzzu, A.,
2748: 1992, ApJL, 389, 83
2749: %\bibitem[Rebull (2001)] {r01} Rebull, L.M., 2001, AJ, 121, 1676
2750: %\bibitem[Rebull et al. (2002)] {rea02} Rebull, L.M., Wolff, S.C., Strom,
2751: % S.E., \& Makidon, R.B., 2002, AJ, 124, 546
2752: \bibitem[Rucinski et al. (2004)]{rea04} Rucinski, S.M., \& 10 co-authors, 2004,
2753: PASP, 116, 1093.
2754: %\bibitem[Saar \& Brandenburg (1999)]{sb99} Saar, S.H. \& Brandenburg, A.,
2755: % 1999, ApJ, 524, 295
2756: \bibitem[Saar \& Osten (1997)]{so97} Saar, S.H., \& Osten, R.A., 1997, MNRAS,
2757: 284, 803
2758: \bibitem[Saio et al. (2007)]{sea07} Saio, H., and 12 co-authors, 2007, ApJ, 654,
2759: 544
2760: \bibitem[Sandage (1962)]{s62} Sandage, A., 1962, ApJ, 135, 349
2761: \bibitem[Sandage et al. (2003)]{slv03} Sandage, A., Lubin, L.M., \& VandenBerg,
2762: D.A., 2003, PASP, 115, 1187
2763: %\bibitem[Schatzman (1962)]{s62} Schatzman, E., 1962, Annales d'Astrophysique,
2764: % 25, 18
2765: %\bibitem[Sills et al. (2000)] {spt00} Sills, A., Pinsonneault, M.H. \&
2766: % Terndrup D.M., 2000, 534, 335
2767: \bibitem[Skumanich (1972)]{sku72} Skumanich, A., 1972, ApJ, 171, 565
2768: %\bibitem[Smith (1979)]{smi79} Smith, M.A., 1979, PASP, 91, 737
2769: %\bibitem[Soderblom (1983)]{sod83} Soderblom, D.R., 1983, ApJS, 53, 1
2770: \bibitem[Soderblom (1985)]{sod85} Soderblom, D.R., 1985, AJ, 90, 2103
2771: \bibitem[Soderblom \& Clements (1987)]{sc87} Soderblom, D.R. \& Clements, S.D.,
2772: 1987, AJ, 93, 920
2773: \bibitem[Soderblom et al. (1991)]{sdj91} Soderblom, D.R., Duncan, D.K., \&
2774: Johnson, D.R.H., 1991, ApJ, 375, 722
2775: %\bibitem[Soderblom et al. (1993a)]{sshj93a} Soderblom, D.R., Stauffer, J.R.,
2776: % Hudon, J.D., \& Jones, B.F., 1993, ApJS, 85, 315
2777: %\bibitem[Soderblom et al. (1993b)]{ssmj93b} Soderblom, D.R., Stauffer, J.R.,
2778: % MacGregor, K.B., \& Jones, B.F., 1993, ApJ, 409, 624
2779: %\bibitem[Spiegel \& Weiss (1980)]{sw80} Spiegel, E.A. \& Weiss, N.O., 1980,
2780: % Nature, 287, 616
2781: %\bibitem[Stassun et al. (1999)]{smmv99} Stassun, K.G., Mathieu, R.D., Mazeh,
2782: % T., \& Vrba, F.J., 1999, AJ, 117, 2941
2783: \bibitem[Stauffer (2000)]{s00} Stauffer, J., 2000, in ASP Conf. Ser., 198,
2784: Stellar Clusters and Associations: Convection, Rotation, and Dynamos,
2785: ed. R. Pallavicini, G. Micela \& S. Sciortino (San Francisco: ASP), 255
2786: %\bibitem[Stauffer \& Hartmann (1987)]{sh87} Stauffer, J.R. \& Hartmann, L.W.,
2787: % 1987, ApJ, 318, 337
2788: %\bibitem[Stauffer et al. (1984)]{shsb84} Stauffer, J.R., Hartmann, L.W.,
2789: % Soderblom, D.R., \& Burnham, N., 1984, ApJ, 280, 202
2790: %\bibitem[Steenbeck \& Krause (1969)]{sk69} Steenbeck, M. \& Krause, F., 1969,
2791: % Astron. Nachr., 291, 49
2792: \bibitem[Strassmeier (2006)]{s06} Strassmeier, K.G., 2006, Ap\&SS, 304, 397
2793: \bibitem[Strassmeier et al. (2000)]{sgb00} Strassmeier, K.G., Granzer, T.,
2794: Boyd, L.J., Epand, D.H., 2000, Proc. SPIE, eds R.I. Kibrick \& A.
2795: Wallander, 4011, 157
2796: \bibitem[Strassmeier et al. (2000)]{sea00} Strassmeier, K.G., Washuettl, A.,
2797: Granzer, Th., Scheck, M., \& Weber, M., 2000, A\&AS, 142, 275
2798: \bibitem[Takeda et al. (2007)]{tea06} Takeda, G., Ford, E.B., Sills, A., Rasio,
2799: F.A., Fischer, D.A. \& Valenti, J.A., 2007, ApJS, 168, 297
2800: %\bibitem[Terndrup et al. (1999)]{tea99} Terndrup, D.M., Krishnamurthi, A.,
2801: % Pinsonneault, M.H. \& Stauffer, J.R., 1999, 118, 1814
2802: %\bibitem[Terndrup et al. (2000)]{tea00} Terndrup, D.M., Stauffer, J.R.,
2803: % Pinsonneault, M.H., Sills, A., Yuan, Y., Jones, B.F., Fischer, D., \&
2804: % Krishnamurthi, A., 2000, AJ, 119, 1303
2805: \bibitem[Valenti \& Fischer (2005)]{vf05} Valenti, J.A., \& Fischer, D.A., 2005,
2806: ApJS, 159, 141
2807: %\bibitem[Van Leeuwen \& Alphenaar (1982)]{va82} Van Leeuwen, F., \& Alphenaar,
2808: % P., 1982, ESO Messenger, 28, 15
2809: \bibitem[Van Leeuwen et al. (1987)]{vam87} Van Leeuwen, F., Alphenaar, P., \&
2810: Meys, J.J.M., 1987, A\&AS, 67, 483
2811: \bibitem[VandenBerg \& Stetson (2004)]{vs04} VandenBerg, D.A. \& Stetson, P.B.,
2812: 2004, PASP, 116, 997
2813: \bibitem[VandenBerg et al. (2006)]{vea06} VandenBerg, D.A., Bergbusch, P.A., \&
2814: Dowler, P.D., 2006, ApJS, 162, 375
2815: \bibitem[Vauclair (1972)]{v72} Vauclair, S. 1972, in Ages des Etoiles, Proc.
2816: IAU Coll. 17, ed. G. Cayrel de Strobel \& A. M. Delplace, 38
2817: \bibitem[Vaughan (1980)]{v80} Vaughan, A.H., 1980, PASP, 92, 392
2818: %\bibitem[Vaughan \& Preston (1980)]{vp80} Vaughan, A.H. \& Preston, G.W.,
2819: % 1980, PASP, 92, 385
2820: \bibitem[Wilson (1963)]{w63} Wilson, O.C., 1963, ApJ, 138, 832
2821: \bibitem[Wozniak et al. (2004)]{woz04} Wozniak, P.R., Vestrand, W.T., \& 16
2822: co-authors, 2004, AJ, 127, 2436
2823: \bibitem[Wright et al. (2004)]{wea04} Wright, J.T., Marcy, G.W., Butler, R.P.,
2824: Vogt, S.S., 2004, ApJS, 152, 261
2825: \bibitem[Yi et al. (2001)]{yea01} Yi, S.K., Demarque, P., Kim, Y.-C., Lee,
2826: Y.-W., Ree, C.H., Lejeune, T., \& Barnes, S.A., 2001, ApJS, 136, 417
2827:
2828:
2829: \end{thebibliography}
2830:
2831:
2832:
2833: %% Use the figure environment and \plotone or \plottwo to include
2834: %% figures and captions in your electronic submission.
2835:
2836:
2837: \clearpage
2838:
2839: \begin{figure} %Fig1 $P/\sqrt{t}$ vs $(B-V)_0$ for only the open clusters
2840: %\plotone{fig1_b03c.eps}
2841: \includegraphics[scale=1.25]{f1.eps}
2842: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2843: \caption{Plot of $P/\sqrt{t}$ vs $(B-V)_0$ for open cluster stars only
2844: ($P$=rotation period; $t$=cluster age).
2845: Symbol sizes and colors correspond to cluster age (blue/small = young,
2846: red/large = old). The densest concentration of stars in the vicinity of the
2847: solid line represents the interface sequence. Note how the interface sequences
2848: of all the open clusters coincide. Also note the clearly visible convective
2849: sequence along the lower edge of the upper panel. The solid line represents
2850: $f(B-V)$. Dotted lines are at $2f$ and $4f$. Some stars in the vicinity of the
2851: dashed lines could be spurious periods or non-members.
2852: The same data are plotted in both panels, on a linear scale in the upper panel,
2853: and on a logarithmic scale in the lower panel.
2854: \label{fig1}}
2855: \end{figure}
2856:
2857:
2858: \begin{figure} %Fig2 $P/\sqrt{t}$ for the (grouped) young Mt.Wilson stars
2859: %\plotone{fig2_b03c.eps}
2860: \includegraphics[scale=1.25]{f2.eps}
2861: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2862: \caption{Plot of $P/\sqrt{t}$ vs $(B-V)_0$ for the young Mt.\,Wilson stars
2863: (small black asterisks), assumed to be 780\,Myr old, the
2864: median chromospheric age for this sample, overplotted on the open cluster data.
2865: Note how the young Mt.\,Wilson stars overlie the interface sequences for the
2866: open clusters, and that no young Mt.\,Wilson stars are on the C sequence.
2867: The non-coeval nature of the young
2868: Mt.\,Wilson sample probably accounts for much of the dispersion observed.
2869: \label{fig2}}
2870: \end{figure}
2871:
2872: \begin{figure} %Fig3 $P/\sqrt{t}$ for the (grouped) old Mt.Wilson stars
2873: %\plotone{fig3_b03c.eps}
2874: \includegraphics[scale=1.25]{f3.eps}
2875: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2876: \caption{Plot of $P/\sqrt{t}$ vs $(B-V)_0$ for the old Mt.\,Wilson stars
2877: (large black asterisks), assumed to be 4.24\,Gyr old, the
2878: median chromospheric age for this sample, overplotted on the open cluster data.
2879: Note how the old Mt.\,Wilson stars overlie the interface sequences for the open
2880: clusters, and that no Mt.\,Wilson stars are located near the C sequence.
2881: The non-coeval nature of the old
2882: Mt.\,Wilson sample probably accounts for much of the dispersion observed.
2883: \label{fig3}}
2884: \end{figure}
2885:
2886: \begin{figure} %Fig4 $P/\sqrt{t}$ for the (individual) Mt.Wilson stars
2887: %\plotone{fig3_b03c.eps}
2888: \includegraphics[scale=1.25]{f4.eps}
2889: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2890: \caption{Plot of $P/\sqrt{t}$ vs $(B-V)_0$ for the
2891: individually age-corrected (chromospheric ages) Mt.\,Wilson stars.
2892: Note how the Mt.\,Wilson stars (small green circles = young;
2893: large red circles =old) lie on top of the interface sequences
2894: for the open clusters. The solid line represents $f(B-V)$, as before, and the
2895: dotted lines are a factor of $0.8-$ and $1.25 \times f$ ($\pm$25\%).
2896: \label{fig4}}
2897: \end{figure}
2898:
2899: \begin{figure} %Fig5 Mass dependence from R
2900: %\includegraphics[angle=-90,width=6.5in]{fig_str_CgI.eps}
2901: \plotone{f5.eps}
2902: %\plotfiddle{fig_P_R_CgI.eps}{4in}{90}{80}{80}{-290}{-150}
2903: \caption{The fit to the mass dependence (solid line), using R:
2904: $f(B-V)=(0.7725 \pm 0.011) \times (B-V_0-0.4)^{0.601 \pm 0.024}$.
2905: The abscissa gives $(B-V_o-0.4)$ and the ordinate $P/\sqrt{t}$ for individual
2906: I sequence stars in the main sequence open clusters listed in the text.
2907: The dashed line shows a smooth trend curve plotted using the function
2908: {\it lowess} in the R statistics package.
2909: Note the similarity of the two curves, which demonstrates
2910: that the fitting function is appropriate for these data.
2911: \label{fig5}}
2912: \end{figure}
2913:
2914: \begin{figure} %Fig6 $P/g(t)$ for the (individual) Mt.Wilson stars
2915: %\plotone{fig3_b03c.eps}
2916: \includegraphics[scale=1.25]{f6.eps}
2917: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2918: \caption{Plot of $P/\sqrt(t)$ vs $(B-V)_0$ for individually age-corrected
2919: (chromospheric ages) Mt.\,Wilson stars (small green circles = young;
2920: large red circles = old), with the old (dashed) and new (solid) functions, $f$,
2921: overplotted. Note that the new function accommodates bluer stars.
2922: The discrepancy arises from the assumed chromospheric ages for the stars,
2923: which are almost certainly overestimated for the F\,stars (see text).
2924: The same data are plotted in both panels, on a linear scale in the
2925: upper panel, and on a logarithmic scale in the lower panel.
2926: \label{fig6}}
2927: \end{figure}
2928:
2929: \begin{figure} %Fig7 Chromospheric age vs gyrochronology for Mt.Wilson stars
2930: %\plotone{fig3_b03c.eps}
2931: %\includegraphics[scale=0.75]{wilson_agecomp.ps}
2932: \includegraphics[scale=1]{f7.eps}
2933: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2934: \caption{Comparison of gyro- and chromospheric ages for the Mt.\,Wilson stars.
2935: The young (Y) and old (O) Mt.\,Wilson stars are marked with small green and
2936: large red circles respectively. The line indicates equality. Note that the
2937: gyro ages are well-behaved for the youngest stars, where the chromospheric
2938: ages are suspect. The dotted lines represent the age of the universe, and the
2939: cross indicates typical gyro/chromospheric age errors quoted for this sample.
2940: \label{fig7}}
2941: \end{figure}
2942:
2943: \begin{figure} %Fig8 Chromospheric age vs gyrochronology for Mt.Wilson stars
2944: %\plotone{fig3_b03c.eps}
2945: %\includegraphics[scale=0.75]{wilson_agecomp.ps}
2946: \includegraphics[scale=1]{f8.eps}
2947: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
2948: \caption{Comparison of gyro and chromospheric ages for the Mt.\,Wilson stars.
2949: Blue crosses indicate stars bluer than $B-V=0.6$, red asterisks stars redder
2950: than $B-V=0.8$, and green squares those with colors between.
2951: The upper (solid) line indicates equality, while the lower (dashed) line at
2952: $Age_{gyro}=0.74 \times Age_{chromo}$ bisects the data.
2953: Note that both techniques are in general agreement about the youth or antiquity
2954: of any particular star, but that the gyro ages are roughly 25\% lower on
2955: average. The figure also shows that the bluer stars contribute most to this
2956: discrepancy. The thick and thin dotted lines represent the age of the universe
2957: and the lifetime of F\,stars (5\,Gyr) respectively.
2958: The cross indicates typical gyro/chromospheric age errors quoted for this
2959: sample.
2960: \label{fig8}}
2961: \end{figure}
2962:
2963: \begin{figure} %Fig9 Removal of Strassmeier Cg stars
2964: %\includegraphics[angle=-90,width=6.5in]{fig_str_CgI.eps}
2965: \includegraphics[scale=1]{f9.eps}
2966: %\plotone{fig_str1.eps}
2967: %\plotfiddle{fig_P_R_CgI.eps}{4in}{90}{80}{80}{-290}{-150}
2968: \caption{ Division of the Strassmeier et al. (2000) sample into I sequence
2969: (suitable for gyrochronology) and C/g (unsuitable) categories.
2970: The solid line separates the two categories of stars, and represents an
2971: isochrone for 100\,Myr.
2972: \label{fig9}}
2973: \end{figure}
2974:
2975: \begin{figure} %Fig10 Gyro ages for Strassmeier CgI stars
2976: %\includegraphics[angle=-90,width=6.5in]{fig_str_CgI.eps}
2977: \includegraphics[scale=1]{f10.eps}
2978: %\plotone{fig_str_2.eps}
2979: %\plotfiddle{fig_P_R_CgI.eps}{4in}{90}{80}{80}{-290}{-150}
2980: \caption{ Ages for the Strassmeier et al. (2000) I sequence stars may be read
2981: off this figure. Isochrones correspond to ages of 100\,Myr, 200\,Myr, 450\,Myr,
2982: 1\,Gyr, 2\,Gyr, and 4.5\,Gyr. Note that all but 4 of the stars are less than
2983: 2\,Gyr in age.
2984: \label{fig10}}
2985: \end{figure}
2986:
2987: \begin{figure} %Fig11 R_HK for Strassmeier CgI stars
2988: %\includegraphics[angle=-90,width=6.5in]{fig_str_CgI.eps}
2989: \includegraphics[scale=1]{f11.eps}
2990: %\plotone{fig_str_2.eps}
2991: %\plotfiddle{fig_P_R_CgI.eps}{4in}{90}{80}{80}{-290}{-150}
2992: \caption{ $R_{HK}$ vs. Gyro age for the Strassmeier et al. (2000) I sequence
2993: stars. Note the declining trend of $R_{HK}$ with age. The trend is obvious
2994: despite the fact that the $R_{HK}$ values are not long-term averages, and have
2995: not been corrected for photospheric contributions or variation with color.
2996: \label{fig11}}
2997: \end{figure}
2998:
2999: \begin{figure} %Fig12 Removal of Pea03 Cg stars
3000: %\plotone{fig_pea1.eps}
3001: \includegraphics[scale=1]{f12.eps}
3002: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3003: \caption{ Division of the Pizzolato et al. (2003) stars into
3004: C/g- \& I\,categories.
3005: The solid line separates the two categories of stars, and represents a
3006: rotational isochrone for 100\,Myr. The dotted line indicates the approximate
3007: color ($B-V=1.6$; M\,3) for the onset of full convection.
3008: \label{fig12}}
3009: \end{figure}
3010:
3011: \begin{figure} %Fig13 Gyro ages for the Pea03 stars
3012: %\plotone{fig_pea2.eps}
3013: \includegraphics[scale=1]{f13.eps}
3014: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3015: \caption{ Ages for the Pizzolato et al. (2003) stars may be read off this
3016: figure. Rotational isochrones correspond to ages of
3017: 100\,Myr, 200\,Myr, 450\,Myr, 1\,Gyr, 2\,Gyr, 4.5\,Gyr, \& 10\,Gyr, as marked.
3018: \label{fig13}}
3019: \end{figure}
3020:
3021: \begin{figure} %Fig14 X-rays vs. Gyro ages for the Pea03 stars
3022: %\plotone{fig_pea3.eps}
3023: \includegraphics[scale=1]{f14.eps}
3024: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3025: \caption{ X-ray emission vs. gyro age for the Pizzolato et al. (2003) stars.
3026: Note the steady decline in X-ray emission with gyro age, as expected.
3027: The line drawn has a slope of $-5/4$, as expected from MHD turbulence.
3028: \label{fig14}}
3029: \end{figure}
3030:
3031: \begin{figure} %Fig15 Chromospheric age vs gyrochronology for Pea stars
3032: %\plotone{fig3_b03c.eps}
3033: %\includegraphics[scale=0.75]{wilson_agecomp.ps}
3034: \includegraphics[scale=1.00]{f15.eps}
3035: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3036: \caption{Comparison of gyro and chromospheric ages for the 19 Pizzolato et
3037: al. (2003) stars in the Southern Chromospheric Survey of Henry et al. (1996).
3038: Note that almost all the stars scatter about the line of equality (solid).
3039: The dashed lines indicate factors of two above and below the gyro ages.
3040: Typical error bars are indicated.
3041: The dotted lines indicate the age of the universe.
3042: \label{fig15}}
3043: \end{figure}
3044:
3045:
3046: \begin{figure} %Fig16 Ages for three wide binary systems
3047: %\plotone{fig3_b03c.eps}
3048: \includegraphics[scale=1.00]{f16.eps}
3049: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3050: \caption{Color-period diagram for three wide binary systems,
3051: $\xi$\,Boo\,A/B, 61\,Cyg\,A/B, \& $\alpha$\,Cen\,A/B.
3052: Rotational isochrones are drawn for ages of 226\,Myr, 2.0\,Gyr \& 4.4\,Gyr
3053: respectively, and the errors are indicated with dashed lines.
3054: Note that for all three wide binary systems, both components give substantially
3055: the same age.
3056: The dotted line corresponds to the age of the universe.
3057: \label{fig16}}
3058: \end{figure}
3059:
3060:
3061:
3062: \begin{figure} %Fig17 Age for the 36Oph triple system
3063: %\plotone{fig3_b03c.eps}
3064: \includegraphics[scale=1.00]{f17.eps}
3065: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3066: \caption{Color-period diagram for the 36\,Oph\,ABC triple system.
3067: Isochrones are drawn for ages of 590\,Myr (solid) and 1.43\,Gyr (thick dashed).
3068: The distant companion, C, gives the 590\,Myr age for the system.
3069: The error is indicated with thin dashed lines.
3070: The A and B components appear to have interacted and spun down to $\sim$20d
3071: against a nominally expected period of $\sim$13d.
3072: The dotted line corresponds to the age of the universe.
3073: \label{fig17}}
3074: \end{figure}
3075:
3076:
3077: \begin{figure} %Fig18 Isochrone age vs gyrochronology for SPOCS stars
3078: %\plotone{fig3_b03c.eps}
3079: %\includegraphics[scale=0.75]{wilson_agecomp.ps}
3080: \includegraphics[scale=1.00]{f18.eps}
3081: %\plotfiddle{fig1_b02_lin.eps}{9in}{0}{80}{80}{-290}{-150}
3082: \caption{Comparison of gyro and isochrone ages for the 26 Takeda et al. (2007)
3083: stars with `well-defined' ages in common with the gyrochronology sample
3084: presented in this paper. The solid line denotes equality and the dotted lines
3085: the age of the universe. There is no strong correlation between the two ages,
3086: except that the median isochrone age is a factor of 2.7 times higher than
3087: the median gyro age. Takeda et al. (2007) stars with upper- or lower limits
3088: (arrows) are not plotted, except for the wide binaries (the components
3089: are connected by dashed lines) discussed in the text. It would appear that the
3090: gyro ages supercede the isochrone ages for main sequence stars.
3091: \label{fig18}}
3092: \end{figure}
3093:
3094:
3095: %% If you are not including electronic art with your submission, you may
3096: %% mark up your captions using the \figcaption command. See the
3097: %% User Guide for details.
3098: %%
3099: %% No more than seven \figcaption commands are allowed per page,
3100: %% so if you have more than seven captions, insert a \clearpage
3101: %% after every seventh one.
3102:
3103:
3104: %% Tables should be submitted one per page, so put a \clearpage before
3105: %% each one.
3106:
3107: %% Two options are available to the author for producing tables: the
3108: %% deluxetable environment provided by the AASTeX package or the LaTeX
3109: %% table environment. Use of deluxetable is preferred.
3110: %%
3111:
3112: %% Three table samples follow, two marked up in the deluxetable environment,
3113: %% one marked up as a LaTeX table.
3114:
3115:
3116: %% In this first example, note that the \tabletypesize{}
3117: %% command has been used to reduce the font size of the table.
3118: %% Note also that the \label command needs to be placed
3119: %% inside the \tablecaption.
3120:
3121:
3122: %\clearpage
3123:
3124: %% The following command ends your manuscript. LaTeX will ignore any text
3125: %% that appears after it.
3126:
3127: \end{document}
3128:
3129: %%
3130: %% End of file `sample.tex'.
3131: