A New Look at an Old Cluster: The Membership, Rotation, and Magnetic Activity of Low-Mass Stars in the 1.3-Gyr-Old Open Cluster NGC 752
Marcel Agüeros, Emily Bowsher, John Bochanski, Phill Cargile, Kevin Covey, Stephanie Douglas, Adam Kraus, Alisha Kundert, Nicholas Law, Aida Ahmadi, Héctor Arce
AAccepted to
ApJ
April 5, 2018
Preprint typeset using L A TEX style emulateapj v. 04/17/13
A NEW LOOK AT AN OLD CLUSTER: THE MEMBERSHIP, ROTATION, AND MAGNETIC ACTIVITY OFLOW-MASS STARS IN THE 1.3-GYR-OLD OPEN CLUSTER NGC 752
M. A. Agüeros , E. C. Bowsher , J. J. Bochanski , P. A. Cargile , K. R. Covey , S. T. Douglas , A. Kraus ,A. Kundert , N. M. Law , A. Ahmadi , H. G. Arce Accepted to ApJ April 5, 2018
ABSTRACTThe nearby open cluster NGC 752 presents a rare opportunity to study stellar properties at ages > >
50% increase over previous catalogs. Using a Bayesian framework to fitMESA Isochrones & Stellar Tracks evolutionary models to literature photometry and the Tycho-
Gaia
Astrometric Solution data available for 59 cluster members, we infer the age of, and distance to, NGC752: 1 . ± .
06 Gyr and 438 +8 − pc. We also report the results of our optical monitoring of the clusterusing the Palomar Transient Factory. We obtain rotation periods for 12 K and M cluster members,the first periods measured for such low-mass stars with a well-constrained age > α forcluster stars, finding that members earlier than ≈ M2 are magnetically inactive, as expected at thisage. Forthcoming
Gaia data should solidify and extend the membership of NGC 752 to lower masses,thereby increasing its importance for studies of low-mass stars.
Subject headings: open clusters and associations: individual (NGC 752), stars: rotation, stars: activity INTRODUCTION
A star’s age is one of its most fundamental parame-ters. It is also, for low-mass, main-sequence field stars,notoriously difficult to measure accurately. Over the pastdecade, a number of authors have proposed age-rotationand age-magnetic activity relations as tools for deter-mining ages for < ∼ (cid:12) stars (e.g., Mamajek & Hillen-brand 2008; Barnes 2010; Reiners & Mohanty 2012; Mattet al. 2015). While measurements for small samples ofsolar-type stars with precise, > << Department of Astronomy, Columbia University, 550West 120th Street, New York, NY 10027, USA; [email protected] Department of Chemistry, Biochemistry, and Physics, RiderUniversity, 2083 Lawrenceville Road, Lawrenceville, NJ 08648,USA Harvard-Smithsonian Center for Astrophysics, 60 GardenStreet, Cambridge, MA 02138, USA Department of Physics and Astronomy, Western WashingtonUniversity, Bellingham, WA 98225, USA Department of Astronomy, University of Texas at Austin,2515 Speedway, Stop C1400, Austin, TX 78712, USA Department of Astronomy, University of Wisconsin-Madison, Madison, WI 53706, USA Department of Physics and Astronomy, University of NorthCarolina, Chapel Hill, NC 27599, USA University of Calgary, 2500 University Dr. NW, Calgary, Al-berta T2N 1N4, Canada Department of Astronomy, Yale University, New Haven, CT06520, USA sepe, the Hyades, and the Pleiades; Agüeros et al. 2011;Douglas et al. 2014; Covey et al. 2016).The failure of
Kepler ’s second reaction wheel and themission’s rebirth as K2 (Howell et al. 2014) was an oppor-tunity to measure new rotation periods ( P rot ) for mem-bers of open clusters along the ecliptic. The result hasbeen a notable increase in our understanding of the ro-tational behavior of < ∼ (cid:12) stars in the linchpin clusterslisted above (Douglas et al. 2016; Rebull et al. 2016a,b;Stauffer et al. 2016; Douglas et al. 2017).Unfortunately, with the exception of the ≈ K2 ’s Cam-paign 7, none of the clusters surveyed by Kepler or K2 issufficiently old and close to enable the P rot measurementsneeded to extend our understanding of the rotational evo-lution of low-mass stars to ages > h m , +37 ◦ ), discovered by CarolineHerschel in 1783, could become a benchmark cluster forstudying stellar rotation and activity at 1-2 Gyr. Whilenearby for a cluster of its age (( m − M ) o ≈
8; Danielet al. 1994), NGC 752 has received relatively little at-tention, in part because of how difficult it has been toestablish a high-confidence membership catalog for thecluster. Surveys such as that of Daniel et al. (1994) werelimited to identifying cluster giants and main-sequencemembers earlier than mid-K, mostly due to a lack ofproper motion (PM) data for fainter stars.Later-type NGC 752 members can now be identifiedusing all-sky photometric and astrometric surveys. Aswas demonstrated by Kraus & Hillenbrand (2007, here- a r X i v : . [ a s t r o - ph . S R ] A p r Agüeros et al.
TABLE 1Comparison of the Main NGC 752 MembershipCatalogs
Non- SpTCatalog Members Members Range ‡ Daniel et al. (1994) 157 † · · · Mermilliod et al. (1998) 17 † · · · This work 258 · · ·
F0-M4
Note . — Fourteen of the Mermilliod et al. (1998) probablemembers are also probable members in Daniel et al. (1994).Mermilliod et al. (1998) identify an additional two possiblemembers that do not appear in the Daniel et al. (1994) cat-alog, and reclassify one Daniel et al. (1994) non-member as aprobable member. † Includes both probable and possible members. ‡ Daniel et al. (1994) and Mermilliod et al. (1998) do not pro-vide spectral types for their stars. The Daniel et al. (1994)stars are F-K dwarfs and K-type red giants; the Mermilliodet al. (1998) stars are all red giants. See Figure 3. after KH07), combining data from e.g., the Sloan DigitalSky Survey (SDSS; York et al. 2000), the Two MicronAll Sky Survey (2MASS; Skrutskie et al. 2006), and thethird U.S. Naval Observatory CCD Astrograph Catalog(UCAC3; Zacharias et al. 2010), can yield precise PMswith standard deviations σ ≈ − , spectropho-tometric distances accurate to within about 10%, andspectral types (SpTs) accurate to within about 1 sub-class. Even for sparse and slow-moving clusters such asComa Berenices, these surveys reveal the low-mass stel-lar populations that eluded previous searches.In Section 2, we summarize previous work on NGC752’s membership before providing an improved and ex-panded membership catalog for the cluster. We use thisnew catalog to derive a more accurate age for and dis-tance to the cluster in Section 3. In Section 4, we describeour Palomar Transient Factory (PTF; Law et al. 2009;Rau et al. 2009) observations of NGC 752 and use theresulting data to measure P rot for 12 K and M clustermembers. In Section 5, we describe our spectroscopiccampaign to characterize chromospheric activity in thiscluster. We place these results in context in Section 6 toconstrain the evolution of low-mass stars up to 1.3 Gyr.We conclude in Section 7. CONSOLIDATING AND EXPANDING NGC 752’SMEMBERSHIP
Consolidating membership data from the literature
We began by compiling membership information forNGC 752 from Daniel et al. (1994) and Mermilliod et al.(1998). Daniel et al. (1994) provided the most compre-hensive membership catalog for the cluster, derived fromprevious PM and radial velocity (RV) studies and newRV measurements. This catalog is divided into threemembership levels: probable member, possible member,and probable non-member. A star’s membership is deter-mined from its PM; Daniel et al. (1994) give the resultsof Platais (1991) priority in the case of conflicts in theliterature. The membership status was adjusted if therewas strong evidence for non-membership based on theRV measurements made by Pilachowski et al. (1988) for19 stars or by Daniel et al. (1994) for 79 stars. The finalcatalog of 255 stars contains 109 probable members, 48possible members, and 98 probable non-members. -10 -5 0 5 10 15 20 µ α (mas yr -1 )-20-15-10-505 µ δ ( m a s y r - ) Fig. 1.—
Proper-motion distribution for > ∼ ,
000 stars < ◦ fromthe center of NGC 752 and with DMs between 6.5 and 8.5 mag. Thestars are the 105 probable and possible cluster members identifiedby Daniel et al. (1994) for which we measure proper motions. Thecircle is the 2 σ limit for a M0 cluster member. Mermilliod et al. (1998) conducted an 18-year RV sur-vey of NGC 752’s red giants. The resulting catalog of 30stars includes 15 probable members, two possible mem-bers, and 13 non-members.There is significant overlap between the catalogs: 22 ofthe 30 Mermilliod et al. (1998) stars are in Daniel et al.(1994). The only significant difference concerns Platais172, classified as a non-member by Daniel et al. (1994)and as a probable member by Mermilliod et al. (1998).We therefore adopt the Daniel et al. (1994) catalog as thebedrock of our membership catalog, adding Platais 172and two possible members identified by Mermilliod et al.(1998) that were not studied by Daniel et al. (1994).We also had access to RV measurements for 123 can-didate cluster members. These include RVs published byDaniel et al. (1994) for 92 stars (including 19 RVs fromPilachowski et al. 1988), as well as measurements for 76stars shared with us by C. Pilachowski (45 of which alsohave RVs published in Daniel et al. 1994). For each ofthese 76 stars, ≈
15 spectra were obtained as part of along-term monitoring campaign with the Hydra spectro-graph on the WIYN 3.5-m telescope, Kitt Peak, AZ. The RVs were derived from spectra of the Mg b triplet(5167, 5173, 5184 Å) obtained using the bench-mountedspectrograph with the blue fiber cables. To provide thehighest possible precision, the same fibers were placedon the same stars for every observation. A subset of, onaverage, eight non-variable stars with known RVs wereused to establish the zero point for each frame. Withthis approach, it was possible to obtain relative precisionfrom run to run and night to night of 200 m s − for anindividual star (C. Pilachowski, pers. comm.).We examined a number of other studies of NGC 752 in The WIYN Observatory is a joint facility of the University ofWisconsin-Madison, Indiana University, Yale University, and theNational Optical Astronomy Observatory.
GC 752: Membership, Rotation, and Activity 3order to identify other candidate cluster members. How-ever, these usually relied on the Daniel et al. (1994) mem-bership catalog (e.g., Sestito et al. 2004; Giardino et al.2008; Bartaši¯ut˙e et al. 2011) and did not include new PMor RV data, so we did not take them into account whenmaking membership determinations.
Identifying new candidate members
Past surveys of NGC 752 found many FGK members,but only small numbers of late-K and M dwarfs. Theselow-mass members span much of the dynamic range ofour PTF observations, and correctly identifying them iscritical for interpreting the results of our rotational mon-itoring program. We therefore used the techniques firstdescribed in KH07 to add new candidate, low-mass mem-bers to the catalog described above.Our candidate selection pipeline used astrometric andphotometric data from 2MASS and UCAC3. SinceNGC 752 and most of the surrounding area do nothave SDSS coverage, we adapted our spectral-energy-distribution (SED) fitting procedure from KH07 to useUSNO-B1.0 photometry (Monet et al. 2003); see the Ap-pendix for details and Table 1 for the SED template mag-nitudes in the USNO-B1.0 filters. We combined the as-trometric measurements to calculate PMs and the pho-tometric measurements to calculate spectrophotometricdistances and photometric spectral types (SpTs) for ob-jects within 4 ◦ of the cluster center.For our astrometric analysis, we fitted the absolute po-sitions reported in each catalog with a linear solution inRA and DEC, σ -clipping at 3 σ to remove potentiallyerroneous measurements. For our photometric analysis,we fitted all available photometry against a grid of SEDmodels, where the photometric SpT of the best-fit modelwas adopted as the object’s SpT, and the average differ-ence between the absolute magnitudes of the templateand the apparent magnitudes of the object was used toinfer the distance modulus (DM) and hence distance.After measuring the PMs, SpTs, and DMs, we com-puted a membership probability P mem for each object us-ing the methods described by Sanders (1971) and Francic(1989). We first cut our sample to include only objectswith DMs between 6.5 and 8.5 mag, corresponding to ≈ ≈ We considered allother objects to be likely field stars and removed themfrom our catalog. Figure 1 is the PM diagram for the > ∼ e − r in spatial position and Gaussian inPM, centered on the cluster’s mean position and meanPM of (8, −
11) mas yr − ) and a field-star distribution(distributed constantly in position and as a bivariateGaussian in PM, where the mean and σ of the PM distri-bution was fit independently for each bin). A bivariate This analysis was undertaken before the release of UCAC4and UCAC5. This is 0.2-0.3 mag brighter than more recent estimates forthe cluster’s DM; cf. Bartaši¯ut˙e et al. (2007) and discussion inSection 3.
Gaussian was chosen for the field PM distribution be-cause the cluster PM is low, and hence the traditionalparametrization (e.g., Deacon & Hambly 2004) as an ex-ponential (parallel to the cluster PM vector) and one-dimensional Gaussian (perpendicular to the cluster PMvector) breaks down. The PM diagram for field stars islargely dominated by dwarfs for most bins, but has a sig-nificant contribution from background giants for K stars,so a bivariate Gaussian allows the shape to vary betweenthese extremes as needed.
Fig. 2.—
Constructing our membership catalog. The red asteriskon the far right indicates that not all stars with
J > . P mem <
50% were rejected as members. Four with 10% ≤ P mem <
50% and RVs consistent with the cluster’s were included in our finalcatalog.
Producing an updated membership catalog
To assemble a definitive membership catalog, we be-gan by combining the list of members and non-membersassembled from the literature and the list of new candi-date members constructed above. We then matched thestars in this merged catalog to 2MASS; only one likelymember, with P mem = 89 . J < . J < . J > .
5, we selected the 212with P mem ≥
50% as candidate members. For the 105 stars that are listed as probable and possiblemembers in the literature and for which we calculated a P mem (including five that are brighter than J = 9 . P mem ≥ Agüeros et al.
TABLE 2Stars Identified as Members in the Literature and asNon-Members in This Work
J P mem
RV(mag) (%) (km s − )01561395+3747048 477 9.78 ± . ± . ± . ± . ± . ± . ± · · · ± . ± . ± . − ± . ± · · · ± . ± . ± · · · ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . ± . − ± . ± . ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . ± . ± . †
786 11.70 ± . ± . ± . ± . · · · ± . · · · ± . ± . ± . ± . ± . ± . ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · ± . · · · Note . — “Platais ID” is the catalog number of the star in Platais (1991). P mem is presented for all of the stars for which it was calculated, even thoughfor those with J < . σ werecomputed for each star on each night. The first (top) RV uncertainty corre-sponds to the σ in the RV computed from the set of nightly average RVs (andis used for single stars, for which we define this σ < − ); the second(bottom) uncertainty is the average of the nightly σ (and is used for binaries).The first 11 stars have RVs > σ from the cluster value of 5.5 ± − (Daniel et al. 1994). The bottom 21 stars have P mem < P mem <
10% and are thereforeexcluded. † Candidate binary
We further refined the membership status of the 123candidate members with RV measurements obtained byDaniel et al. (1994) and C. Pilachowski (priv. comm).For the 92 Daniel et al. (1994) measurements we usedthe given RV uncertainties in making our comparison tothe cluster value. For the RVs that appear only inPilachowski et al. (1988) we used the typical σ quotedby these authors of 0.5 km s − .Most of the stars observed as part of the WIYN long-term monitoring campaign were observed multiple timeson multiple nights, and an average RV and σ were com-puted for each star on each night. In addition, an un- Two Daniel et al. (1994) stars lack σ values; for these we usethe average σ derived from the other Daniel et al. (1994) RVs. certainty σ corresponding to the σ in the RV computedfrom the set of nightly average RVs and another σ corre-sponding to the average of the nightly σ were calculated.For one of these stars to be labeled a member, we re-quired that its mean RV be within 2 σ of the cluster RVof 5 . ± . − (Daniel et al. 1994). This require-ment resulted in the rejection of a number of stars thathad been listed as members in the literature. For exam-ple, five stars with
J < . J > . P mem ≥ Mermilliod et al. (1998) found 4 . ± .
11 km s − . For sim-plicity, we use the Daniel et al. (1994) value for our RV tests. GC 752: Membership, Rotation, and Activity 5The 17 stars with σ > − were labeled can-didate binaries; 12 of these were identified as candidatebinaries by Daniel et al. (1994). For the five new systems,we use σ to test for the agreement with the cluster RVand classify four as members. The fifth, Platais 786, hadbeen considered a member, but has a P mem = 0 . ≤ P mem <
50% andRV measurements and identified four whose RVs are con-sistent with cluster membership. As a result, we addedthese stars to our final list of cluster members. Starswith RVs consistent with membership but P mem < χ >
3) and there-fore P mem < J versus ( J − K ) and have PMs and RVsconsistent with the expected values, so they are plausi-ble members. The fifth star, Platais 684, is one of ournewly identified candidate binaries: it has a good SEDfit, photometry marginally consistent with membership(and therefore a low P mem ), but RV variability that sug-gests it is a single-lined spectroscopic binary. Its natureneeds be investigated further.In Table 1 we summarize the properties of our newcatalog and compare it to those of Daniel et al. (1994)and of Mermilliod et al. (1998). Our work has added 125new stars to the cluster, reclassified five stars listed asnon-members in the literature as cluster members, andextended NGC 752’s membership to the mid-M stars.Conversely, we have removed 32 stars, or one-fifth of themerged Daniel et al. (1994) and Mermilliod et al. (1998)catalog, from the list of cluster members (see Table 2).The 258 cluster members are presented in Tables 3 and4. A J versus ( J − K ) color-magnitude diagram (CMD)for NGC 752 is shown in Figure 3.Kharchenko et al. (2013) investigated the membershipof NGC 752 as part of a large-scale survey of Milky Waystar clusters. We compared the P mem we derived for can-didate members to those obtained by Kharchenko et al.(2013) for 568 stars in their NGC 752 catalog; this in-cluded many stars for which we calculated a P mem < P mem to many candidates;these are stars near the cluster core that fall on the CMDsequence and proper-motion locus of the cluster.However, Kharchenko et al. (2013) compute probabil-ities that capture spatial position with a step function,assigning P mem = 0% for all stars outside the tidal ra-dius and otherwise weighing all stars uniformly. Fig-ure 4 therefore also contains a substantial populationin the lower right corner, where we measure a P mem ofnear 0% despite the high P mem estimated by Kharchenkoet al. (2013). These stars are field interlopers that fallnear the cluster sequence and proper-motion locus: sincethese interlopers should be uniformly distributed on thesky, most will be located at large radii from the clustercore (but still within its tidal radius) and will be down-weighted by our algorithm, which fits the radial-density J-K ) (mag)161412108 J ( m a g ) new members M0 M7F0 F5 G2 K0 K5
Fig. 3.—
CMD for NGC 752. Members identified in the literatureare in black; our new high-confidence members are in red. P mem (Kharchenko et al. 2013)0.00.20.40.60.81.0 P m e m ( t h i s w o r k ) Fig. 4.—
A comparison of membership probabilities calculatedby Kharchenko et al. (2013) and in this work. While both catalogsassign high P mem to stars near the cluster core that fall on thecluster’s CMD sequence and proper-motion locus, the Kharchenkoet al. (2013) P mem calculation is more sensitive to field interloperswithin the cluster’s tidal radius, resulting in large numbers of non-members with artificially high P mem being listed in their catalog. profile, more effectively than the step function used byKharchenko et al. (2013).Finally, we note that our statistical approach to mem-bership is bound to result in some contamination, withour catalog including stars with high P mem that would be Agüeros et al. TABLE 3 P mem Selected NGC 752 Members
J K
Mass SpT DM M bol
Binary? a P mem (mag) (mag) (M (cid:12) ) (mag) (mag) (%)01501676+3812369 · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · Note . — The full version of this table is available online. a Based on RV measurements published by Daniel et al. (1994) (“D”) or Mermilliod et al. (1998), or collected by C. Pilachowski (“P”).
TABLE 4Other NGC 752 Members
J K
Mass SpT RV D RV P Binary? a (mag) (mag) (M (cid:12) ) (km s − ) (km s − )01510351+3746343 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · ± · · · · · · · · · ± ± · · · · · · ± · · · · · · ± ± · · · F5.6 · · · ± . ± . · · · ± ± · · · · · · ± − ± . ± . DP01553936+3752525 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · Note . — The full version of this table is available online. a Based on RV measurements published by Daniel et al. (1994) (“D”) or Mermilliod et al. (1998), or collected by C. Pilachowski (“P”). excluded when additional information is included or be-comes available. We expect that the forthcoming releaseof the second
Gaia data release (DR2) will be invaluablefor improving the cluster census.
Calculating masses for cluster members
The availability of 2MASS photometry for nearly allof NGC 752’s member stars—and for members of otherclusters to which we wished to compare NGC 752—droveus to use these 2MASS magnitudes to estimate stellarmasses, as in Agüeros et al. (2011). We calculated eachstar’s absolute K magnitude (M K ), using the source-specific distance modulus associated with the star’s SEDfit. Of the empirical absolute magnitude-mass relationsidentified by Delfosse et al. (2000), the M K -mass relationis the best calibrated, and we used this relation for starswith M K > . α /H] = 0 population (up-dated from the original version published in Dotter et al.2008). Systematic uncertainties in the Delfosse et al.(2000) relation are of order ≈ UPDATING NGC 752’S AGE AND DISTANCE
Previous efforts Available from http://stellar.dartmouth.edu/~models/ . A critical step in establishing NGC 752 as a benchmarkopen cluster is accurately determining its age and dis-tance. Main-sequence and red-giant-branch CMD mod-eling of NGC 752 have produced estimated ages rangingfrom 1 to 2 Gyr and DMs ranging from 7.7 to 8.5 mag(e.g., Meynet et al. 1993; Daniel et al. 1994; Dinescuet al. 1995; Twarog et al. 2015). However, these agesand distances were usually derived using a by-eye com-parison of model isochrones to various color-magnitudedata sets, which does not provide statistically meaningfuluncertainties on the output parameters. Furthermore,these isochrone fits generally used sub-solar metallicityisochrones, which are likely not appropriate for this clus-ter.The two most recent and robust determinations ofNGC 752’s age and distance are those performed byBartaši¯ut˙e et al. (2007, 2011) and Twarog et al. (2015).Bartaši¯ut˙e et al. (2007) used a least-squares minimiza-tion to derive an isochrone age of 1 . ± .
04 Gyr and( m − M ) V = 8 . ± .
14 mag for the upper main sequenceof NGC 752. These authors’ grid-search technique didprovide a goodness-of-fit metric and solved for the best-fit age and DM. However, it did not fully account forcorrelated errors in colors and magnitudes, and the ac-curacy of the Bartaši¯ut˙e et al. (2007) results is limited tothe spacing between isochrones in their model grid. InBartaši¯ut˙e et al. (2011), including newly identified pho- van Leeuwen (2009) used HIPPARCOS parallaxes and thephotometric box method to derive a DM of 8.53 ± GC 752: Membership, Rotation, and Activity 7tometric late-type candidate members led the authors tofind an isochrone age of 1 .
41 Gyr and a DM of 8 . ± . ≈ E ( b − y ) =0.025 ± E ( B − V ) = 0.034 ± − .
07 to − . ± Fig. 5.—
Age and distance estimates derived for 59 NGC 752members with no evidence for a binary companion. Error bars in-dicate the characteristic 1 σ uncertainties associated with each ageand distance estimate, with full probability distribution functions(PDFs) shown on the top and right sides of the main panel foreach star. Individual PDFs are transparent, such that regions ofparameter space favored by fits to multiple stars appear darker.The cluster’s age (1.34 ± +8 − pc) are de-rived by multiplying the individual stellar age/distance PDFs toidentify the maximum likelihood values for the combined clusterpopulation, and are highlighted in the main panel with dashed redlines. Our Bayesian approach and results
We applied a Bayesian framework to cluster memberswith astrometric measurements in the Tycho-
Gaia
As-trometric Solution (TGAS) catalog (Gaia Collaborationet al. 2016;
Gaia
Collaboration et al. 2016). Our analysisused photometry from several publicly available surveys,typically Tycho-2 (Høg et al. 2000) BV , Gaia G , 2MASS JHK , and Wide-field Infrared Survey Explorer (Wrightet al. 2010) W W
2, and W
3. SED-based metallici-ties are inherently uncertain, and we therefore applied aGaussian prior for metallicity for the cluster based on theGuo et al. (2017) spectroscopic analysis. These authorsmeasured metallicities for 36 candidate single membersof the cluster using R ≈ ,
000 spectra obtained withthe Hectochelle multi-object spectrograph, finding that [Fe/H] = − . ± . − . ± . A V = 0 . ± .
1, based on the valuederived by Twarog et al. (2015).We then use MINESweeper, a newly developedBayesian approach for determining stellar parametersusing the newest MESA Isochrones & Stellar Tracks(MIST) evolutionary models (Choi et al. 2016; Dotter2016) to infer probability distribution functions for theage and distance of each cluster member. A detaileddescription of MINESweeper will be given in Cargile etal. (in prep); examples of its use include Rodriguez et al.(2017), Temple et al. (2017), and Dotter et al. (2017).MINESweeper provides full posterior distributions of allpredicted stellar parameters from the MIST models, in-cluding ages, masses, and radii.Since we are modeling each cluster member as a sin-gle star, unresolved binaries result in unreliable stellarparameters due to the influence of the binary on thestellar SED. There are 82 likely members in our cata-log with
Gaia
TGAS astrometric parallax measurements:of these, 23 have been identified as RV variables (seeTables 3 and 4), and we therefore derive estimates ofthe stellar parameters only for the 59 apparently singlestars. To determine cluster-wide values for the stellar param-eters inferred from the MINESweeper fits, we computed akernel density estimation of the individual posterior dis-tributions for the stellar parameters estimated for eachstar. The final combined posterior distributions pro-vide the most probable age, distance, [Fe/H], and A V for NGC 752 given our priors and assuming that all ofthese stars are true cluster members. The maximumlikelihood values for the distance and ages of individ-ual stars are shown in Figure 5, along with the super-positions of the individual age and distance probabilitydistribution functions. The combined probability densityfunctions imply the following maximum likelihood meancluster parameters: an age = 1 . ± .
06 Gyr, a distance= 438 +8 − pc (DM = 8.21 +0 . − . ), a [Fe/H] = 0 . ± . A V = 0 . +0 . − . . These cluster parameters arein agreement with those of Bartaši¯ut˙e et al. (2011) andTwarog et al. (2015) and have more robust uncertaintyestimates.As a consistency check, we investigated the direct as-trometric distances provided for 53 stars with accurateTGAS ( σ π < .
35 mas) parallaxes. For these clusterstars, we find a weighted mean parallax π = 2 . ± . ± d =431 ± These values are consistent with those wehave determined using the MINESweeper analysis, andthe RMS of the
Gaia measurements is 0.324 mas, so thequoted uncertainties are consistent with the scatter.However, there are likely to be spatially correlated sys-tematic uncertainties in the
Gaia data at the level ofthis scatter ( ≈ Gaia
Collaboration et al. In practice, including the RV-variable stars does not changeour results. In this case we do not apply a binary cut, as binarity shouldnot impact the parallax-derived distance.
Agüeros et al.
TABLE 5PTF Observations of NGC 752
PTF Field Field Center Number ofNumber (J2000) Observations110005 01:53:35 +37:19:00 377110006 01:53:53 +38:19:00 413 σ d = +65 − pc and σ DM = +0 . − . mag, respectively. The order-of-magnitude im-provement in the precision of parallaxes and proper mo-tions of Gaia
DR2 relative to its first data release willlikely remove these potential systematic uncertainties,enabling proper motion selection to improve the clus-ter census and providing a precise and accurate distancemeasurement to this benchmark >
31 30 29 28 27 26
RA (deg) D e c ( d e g ) field 110005field 110006 known membersnew memberscontrol sample Fig. 6.—
Spatial distribution of the 128 previously catalogedand the 130 newly identified NGC 752 members. The PTF fieldsare overlaid (110005: dashed lines; 110006: dot-dashed lines); theposition of the dead chip in both fields is indicated by the dottedlines. The orange (empty) stars are selected for our control sampleas part of the validation of our measured P rot ; see Section 4.2. MEASURING STELLAR ROTATION AT 1.3 GYR
PTF observations and photometric data reduction
We monitored NGC 752 from 2010 Aug 22 to 2011 Jan19 using time allocated to two PTF Key Projects: thePTF Open Cluster Survey (POCS; Agüeros et al. 2011;Douglas et al. 2014; Covey et al. 2016; Kraus et al. 2017)and the PTF/M-dwarfs survey (Law et al. 2011, 2012).The PTF infrastructure is described in Law et al. (2009);of primary interest to us was one component, the robotic48-inch Oschin (P48) telescope at Palomar Observatory,CA, which we used to conduct our imaging campaign.The P48 was equipped with the modified CFH12K mo-saic camera, which had 11 working CCDs, 92 megapix-els, 1 sampling, and a 7.26 deg field-of-view (Rahmeret al. 2008). Under typical conditions (1 . full-width half-maximum images that reacheda 5 σ limiting R P T F ≈
21 mag in 60 s (Law et al. 2010).We imaged two overlapping 3.5 ◦ × ◦ fields cover-ing the center of NGC 752. The fields were selected sothat the bulk of the cluster members identified by Daniel et al. (1994) and Mermilliod et al. (1998) fell on one chipin each (see Figure 6). For most of the campaign, thesefields were observed one to four times a night, weatherpermitting. There were gaps in our coverage each monthwhen PTF conducted its g -band and/or H α surveys. Be-cause we shared some of our observing time with thePTF/M-dwarfs survey, a transiting-planet search, therewere multiple nights in our campaign when the clusterwas observed with a higher frequency, resulting in ≈ ra-dius to produce multi-epoch light curves. This generatedphotometry for all objects at each epoch with approxi-mate zero points determined on a chip-by-chip basis us-ing USNO-B1.0 photometry of bright stars. The zeropoints were then refined by a downhill-simplex algorithmthat minimized the median photometric variability overall bright non-variable stars in the images.We then applied a version of the SYSREM algorithm toremove systematic trends from the data, e.g., those dueto atmospheric extinction, detector efficiency, or pointspread function changes (Tamuz et al. 2005). Figure 7shows the impact this had on the photometry from field110005, and in particular the resulting improved perfor-mance at the bright end. Applying SYSREM also al-lowed us to identify a few nights for which the overallphotometric behavior of the chips differed significantlyfrom the median over our entire observing campaign. Period measurement
To detect periodic signals in our light curves, we fol-lowed closely the methods developed for our Pleiadesanalysis (Covey et al. 2016). The 90 PTF light curvesfor NGC 752 members were first cleaned of unreliabledata points—those with errors > > σ re-moved from the mean magnitude—before computing aLomb-Scargle periodogram (Scargle 1982; Press & Ry-bicki 1989) for 8000 candidate P rot spaced logarithmi-cally between 0.1 and 50 d. Each light curve was thenphased on the period initially found to have the max-imum power, and 4 σ outliers from a smoothed, phase-folded light curve were clipped before generating an up-dated periodogram. This clipping and computing pro-cess was performed three times before a final period wasassigned to the star.The error on our P rot measurements was estimated us-ing the width of a Gaussian fit to the corresponding peakin the power spectrum (Lamm et al. 2004). This widthindicates a fundamental uncertainty in the period mea-surement that originates from the frequency resolution ofthe power spectrum and the time sampling of the data. Period validation
The process described above returns a P rot for everylight curve. We modified slightly the Covey et al. (2016)GC 752: Membership, Rotation, and Activity 9 Fig. 7.— σ versus median R PTF magnitude of the ≈ σ when placing the objects in bins of width 1 mag is shown as thered points. The raw data are plotted in the left panel; at the brightend, the scatter exceeds the formal photometric errors by factorsof a few, indicating that the precision is limited by systematic ef-fects rather than by random photometric errors. As can be seen inthe right panel, in addition to removing systematic trends in thedata, applying the SYSREM algorithm significantly improves thephotometric performance for R PTF <
16 mag. approach to select the significant and reliable P rot mea-surements. We identified a sample of 254 field stars thathave PTF light curves, high-quality 2MASS photome-try, and ( J − K ) colors and J magnitudes similar tothose of NGC 752 members (see Figure 8). These stars’PTF photometry will exhibit the same instrumental sig-natures as those of the members, but because these starsshould be older and have lower levels of magnetic activ-ity, they should be less variable (as expected based on theage-activity relation; e.g., Hawley et al. 1999; Soderblomet al. 2001; Douglas et al. 2014).We then tested our ability to recover P rot from datathat reflect the cadence and noise properties of our tar-gets’ data by injecting artificial periodic signals intothese quieter light curves and running the same period-detection algorithm as that applied to cluster members.We first removed every star in our control sample forwhich the periodogram included a peak with a power >
20, thereby selecting a sample of 156 minimally variablestars. For each of these remaining stars, we then gener-ated 1500 periodic light curves in which a sine curve withan amplitude scaled relative to the light curve’s σ anda period randomly selected from a Gaussian distributioncentered at 25 ±
10 d was added to the PTF photometrywhile preserving the original light curve’s timestamps. By applying our period-detection algorithm to the re-sulting 234,000 artificial light curves, we measured thedependence of our recovery rate and accuracy of our P rot measurement on the properties of the input light curveand of the output periodogram. We defined as a suc-cessful recovery any simulation in which the input andrecovered P rot agree to within 3%. Our overall successrate was 73%. This simulation allowed us to set a thresh- amplitude/ σ light curve = 0.3, 0.6, 0.9, 1.2, or 1.5. We required that the injected P rot be between 0.1 and 50 d.This choice of a Gaussian P rot distribution for the simulations isthe main difference with our approach in Covey et al. (2016). J-K ) (mag)161412108 J ( m a g ) control sample M0 M7F0 F5 G2 K0 K5
Fig. 8.—
CMD for cluster members and control stars identifiedto test the robustness of our P rot measurements. The control starsample has a color and magnitude distribution that mirrors thatof NGC 752 stars with PTF light curves, but should be dominatedby old field stars with little inherent variability. old power of 40 for the most significant peak in our peri-odograms as the one to use for identifying robust periodmeasurements.To determine if the star exhibits a single, unambigu-ous period, we also cleaned the periodograms for clustermembers of any aliases and beat periods between thecandidate period and a one-day sampling frequency be-fore searching for secondary peaks with power ≥
60% ofthe primary peak’s. If no secondary peaks were found,the primary period was flagged as a secure detection (i.e.,CLEAN = 1); sources with such secondary peaks wereflagged as having an ambiguous period (CLEAN = 0). Inpractice this step eliminated only three stars with peakperiodogram powers > MEASURING CHROMOSPHERIC ACTIVITY AT 1.3 GYR
We used the WIYN 3.5-m telescope on Kitt Peak, AZ,to obtain spectra for 96 stars; we used the MDM Hilt-ner 2.4-m telescope, also on Kitt Peak, to obtain spec-tra for 180 stars (see Table 7). The resulting sampleis ≈
70% complete for candidate cluster members with P mem ≥
50% but that lacked spectra prior to this work.Our observational set-up and data reduction processesare described below.
WIYN: Set-up and data reduction
We observed NGC 752 with the Hydra multi-objectspectrograph during the nights of 2011 Feb 7 and 8. Weused the bench-mounted spectrograph with the red fiber0 Agüeros et al. P mem = 94%)
600 650 700 750Epoch (d)16.4216.4016.3816.36 R PT F ( m a g ) P rot (d)01020304050 P o w e r R PT F ( m a g ) P rot = 19.49 +/- 4.09 d)0.04 ∆ R PT F ( m a g ) Fig. 9.—
Top:
PTF light curve for a newly identified high-confidence NGC 752 member, 2MASS J01570057+3746131. The x-axisis the number of Julian days since 2009 Jan 1. The error bars on the star at the top right show the median photometric uncertainty.
Middle:
The periodogram calculated via our iterative process (black line), with the peak power, corresponding to a period of 19.49 d,highlighted with an orange diamond. The blue Gaussian, with which we estimate the uncertainty on P rot , is a fit to this peak.Beat periods between this P rot and a one-day alias are flagged with vertical (blue) dot-dashed lines; the power threshold used toflag sources with ambiguous period detections (i.e., other periods with peaks with ≥
60% of the primary peak’s power) is shown asa horizontal dashed line.
Bottom:
The phase-folded light curve. A median-filtered version of this light curve, shown as an orangeline, is subtracted to create a pre-whitened light curve, shown in the sub-panel at the bottom. The periodogram computed fromthis pre-whitened light curve is shown as a gray line in the middle panel. The primary peak and beat periods are not present in theperiodogram of the pre-whitened light curve, indicating that the periodic signature removed during the pre-whitening accounts for allof the significant structure in the star’s light curve.
The journal version of this article includes this figure for the other rotators we identify. cables and an échelle grating with 600 lines mm − setat a blaze angle of 13.9 ◦ . This resulted in coverage from 6050 to 8950 Å with ≈ ≈ TABLE 6NGC 752 Rotators P mem M K Mass R PTF P rot (%) (mag) (M (cid:12) ) (mag) (d)01525891+3803515 K4.7 52.8 4.49 .
92 303 17 . ± . .
28 383 18 . ± . .
58 389 13 . ± . .
83 688 14 . ± . .
39 695 19 . ± . .
87 692 14 . ± . .
94 690 5 . ± . .
74 690 16 . ± . .
66 642 17 . ± . .
33 583 34 . ± . .
94 694 13 . ± . .
02 693 32 . ± . Note . — The formal uncertainty for the SpTs is 0.1 spectral classes. However, the systematic uncertainty inthe underlying definition is ≈ K is calculated for each star using the best-fit distancemodulus determined from the SED fit, and in turn is used to obtain masses. Masses for sources brighter thanM K = 5.5 mag are assigned using the theoretical model of Dotter et al. (2008), while masses for fainter sourcesare assigned using the empirical mass-luminosity relation measured by Delfosse et al. (2000). Although theDelfosse et al. (2000) relation extends to stars with M K = 4.5 mag, the predicted mass values diverge byup to about 5% from those of Dotter et al. (2008) for stars brighter than M K = 5.5 mag. We provide themedian RPTF magnitude of each light curve after filtering on flags and correcting for the (generally verysmall) photometric offset between fields for stars that were in both. The uncertainty on these magnitudes is oforder 1%.
TABLE 7Spectroscopic Statistics
WIYN Hiltner(Hydra) (ModSpec)Targets 96 180... with P mem >
50% 60 106... with spectra in literature 12 40... observed more than once 5 7
Note . — P mem is the membership probability in ourcluster catalog; see §2. (BF) centered at 01 h m s , +37 ◦ , and a faintfield (FF) centered at 01 h m s , +37 ◦ (J2000 co-ordinates). The two fields required exposure times of1800 and 5400 s, respectively, which were split into threesub-exposures for cosmic-ray removal. We placed targetfibers on 59 candidate cluster members in the BF and 42candidate members in the FF; five stars were included inboth fields, for a total of 96 individual targets.The data were reduced using standard routines inthe IRAF Hydra package. Each image was trimmedand instrument biases were removed. The spectra forthe individual fibers were extracted, flat-fielded, anddispersion-corrected. Sky spectra from ≈
30 fibers placedacross the field-of-view were combined and subtractedfrom our target spectra. We throughput-corrected andflux-calibrated each spectrum using the flux standardG191-B2B, which was obtained using the same instru-ment set-up as our targets. We then combined thethree sub-exposures for each object to form a single, highsignal-to-noise spectrum for each candidate cluster mem-ber; four sample Hydra spectra are included in Figure 10.
MDM: Set-up and data reduction
We used the MDM Observatory Modular Spectrograph(ModSpec) on the 2.4 m to obtain spectra of 180 candi-date cluster members over the course of five observingruns between 2010 Dec 1 and 2012 Feb 20. ModSpec Available from http://iraf.noao.edu/tutorials/dohydra/dohydra.html Å ) N o r m a li z e d ( o ff s e t ) f l u x Fig. 10.—
Sample Hydra and ModSpec data. Hydra providedcoverage from 6050 to 8950 Å, while ModSpec covered 4500 to7500 Å. Photometric SpT and P mem are indicated in parentheses.Imperfect flux calibrations are responsible for the structure seenin the continuum of the top three (ModSpec) spectra. The strongA-band telluric features are indicated by the ⊕ . provided coverage from 4500 to 7500 Å with ≈ ≈ <
10% across the full spectral rangeof these observations. Given this near constant disper-sion and the lack of bright sky lines to provide a higherorder solution, a simple linear offset was applied to eachobservation’s wavelength solution to correct for the off-set measured from the 5577 Å line. This line was tooweak to provide accurate offset measurements for expo-sures shorter than 30 s, so the instrumental wavelengthsolution was preserved for spectra with short exposuretimes. The uncertainty in the measurement of the [OI]line center was typically ≈ ≈ − . Six sample ModSpec spectra areshown in Figure 10. Identifying chromospherically active members
To identify chromospherically active cluster members,we measured the H α EqW for each spectrum. The mea-surement window used varied from spectrum to spectrumand was adjusted interactively. Ideally, we would alwaystake the continuum flux to be the average between 6550-6560 Å and 6570-6580 Å. For spectra for which the H α line extended into these windows, the continuum fluxwas measured from 10 Å windows on each side of theline. The resulting H α EqW measurements are shown inFigure 11 as a function of ( r − K ).To estimate the human error in these interactive mea-surements, the same person measured each EqW twice,and we took the difference between the two measure-ments. We then used a Monte Carlo technique to de-termine the statistical significance of our H α measure-ments. Lacking noise spectra for these stars, we addednoise drawn from a Gaussian distribution with a widthequal to the σ of the flux in the continuum region to eachspectrum and remeasured the EqW 2500 times. The twoerror measurements were added in quadrature to producethe EqW uncertainties (for details about this procedure,see Douglas et al. 2014).Using these EqW uncertainties, we identified magnet-ically active stars as those with H α EqW+3 σ <
0. Wefound only three stars that satisfy this criterion; we dis-cuss this result in the next section, in the broader contextof the age-rotation-activity relationship in NGC 752 andother open clusters. PLACING ROTATION AND ACTIVITY AT 1.3 GYR INCONTEXT
Comparing the P rot for NGC 752 to those foryounger clusters and to the Matt et al. (2015)model for rotational evolution As in previous papers, we placed our observations ofrotators in NGC 752 in the context of the rotational evo-lution of low-mass stars (cf. Agüeros et al. 2011; Dou-glas et al. 2016, 2017). Our empirical comparison waswith the ≈ P rot measurements exist for stars downto 0.2 M (cid:12) , and with the ≈ Kepler target NGC6811, which is the only cluster close in age to NGC 752with substantial rotational data. We also compared our (r-K) −2−10123456 H α E q W F2 G2 K2 K7 M0 M2
Fig. 11.— H α EqWs as a function of ( r − K ) color for NGC 752cluster members. Uncertainties in EqW are measured through aMonte Carlo process described in the text. The three emitters inthe mid/late K regime are significant outliers from the remainderof the cluster population and have moderate membership proba-bilities: 50% < P mem < data to the predictions from the Matt et al. (2015) modelfor stellar angular-momentum evolution. Initial condi-tions for this model are set by approximating the mass-period distribution observed in very young clusters. An-gular momentum is then removed by winds using a pre-scription based on the solar wind described by Kawaler(1988) and Matt et al. (2012), and the overall angular-momentum loss scales with mass and radius.Below, we extend our Douglas et al. (2017) test of theMatt et al. (2015) model. We then describe the processof constructing the mass-period sample for NGC 6811,examine evidence for rotational evolution between Prae-sepe, NGC 6811, and NGC 752’s ages, and compare theNGC 752 data to the Matt et al. (2015) model for a1.3-Gyr-old population. Comparing the Praesepe data to the Matt et al.(2015) model
The top row of Figure 12 replicates the comparisonmade in Douglas et al. (2017) between the Matt et al.(2015) model and mass-period data for the ≈ > ∼ (cid:12) stars in Praesepe (and in the Hyades, an-other ≈ (cid:12) stars. The model predicts morerapidly rotating < ∼ (cid:12) stars than are observed; < (cid:12) stars in the model and the data reflect this mismatch:the model predicts that the median rotator should havea P rot = 4 . P e r i o d ( d ) Praesepe data653-Myr model
M4M2M0K5K2G2F5F2
M4M2M0K5K2G2F5F2 O • )0.11.010.0 P e r i o d ( d ) NGC 6811 data963-Myr model O • )510152025 Fig. 12.—
Comparison between the Matt et al. (2015) models (empty red stars) and the mass-period distributions for Praesepe (blackstars, top row; data from Douglas et al. 2017) and NGC 6811 (blue stars, bottom row; data from Meibom et al. 2011). The age of the Mattet al. (2015) model population is indicated in the right panel in each row. On the left, the periods are plotted logarithmically, and on theright linearly. when 26 known and candidate binaries are removed, hasa P rot = 14 . Furthermore, more than half of the cluster 0.6-0.3 M (cid:12) stars have converged to the slow-rotator sequence, whichextends from ≈ (cid:12) , and more than half of theremaining rapid rotators are binaries. At 650 Myr, how-ever, the Matt et al. (2015) model predicts that the slow-rotator sequence should end around 0.6 M (cid:12) . If Praesepeis ≈ Hereafter, we remove known and candidate binaries from ourcatalog when calculating median periods for Praesepe.
Examining rotational evolution between Praesepe andNGC 6811
NGC 6811 (1.00 ± Kepler field, and theonly cluster close in age to NGC 752 for which P rot havebeen obtained (Meibom et al. 2011). We matched the ro-tators listed in Meibom et al. (2011) to 2MASS and usedthe cluster properties determined by Janes et al. (2013)( E ( B − V ) = 0 . m − M ) V = 10 .
22) and the 1 Gyr,[Fe/H] = 0.07 and [ α /H] = 0 (updated) Dotter et al.(2008) model to calculate masses for these stars in themanner described in Section 2.4. In the bottom rowof Figure 12, we show the resulting mass-period distri-bution for this cluster, along with the Matt et al. (2015) Janes et al. (2013) find [Fe/H] = − .
18 for NGC 6811 based onisochrones fits, but an analysis of R ≈ ,
000 spectra of individualmembers by Molenda-Żakowicz et al. (2014) finds a mean [Fe/H]= 0.04 ± O • )0.11.010.0 P e r i o d ( d ) NGC 6811 dataPraesepe data837-Myr model
M4M2M0K5K2G2F5F2 O • )51015 K5K2G2F5
Fig. 13.—
As in Figure 12, but now plotting Praesepe and NGC 6811 together with a Matt et al. (2015) 837-Myr-old population. ThePraesepe and NGC 6811 slow-rotating sequences are well-matched, particularly when the P rot are plotted logarithmically, and both appearto flatten for masses < ∼ (cid:12) , which is not predicted by the model. In the right (linear P rot ) panel, there is evidence of spin down for the0.9-1.1 M (cid:12) stars between the ages of Praesepe and NGC 6811, but the 0.8-0.9 M (cid:12) stars do not appear to have spun down. predictions for the distribution of a 963-Myr-old popula-tion. The Matt et al. (2015) model clearly overestimatesthe spin-down for < ∼ (cid:12) stars. Model stars with massesbetween 0.8 and 1.0 M (cid:12) have a median P rot = 14 . P rot = 10 . ± (cid:12) stars is sur-prisingly small over the ≈
350 Myr that should separateNGC 6811 from Praesepe: the 38 Praesepe stars in thismass range have a median P rot = 9.9 d. In Figure 13,we combine the data for Praesepe and NGC 6811 andshow that the clusters’ two slow-rotating sequences arevery well-matched, especially considering that the bulk ofthe Praesepe data for stars > (cid:12) come from ground-based observations (Delorme et al. 2011; Kovács et al.2014). The quality of those data is not as high as thosefrom Kepler , presumably contributing to the scatter inthe periods for Praesepe stars between 0.8 and 1.2 M (cid:12) relative to what is seen for NGC 6811. The combinedcluster data are well described by the Matt et al. (2015)model population for 837 Myr, although the model con-tinues to over-predict the spin down of stars between ≈ (cid:12) and under-predict the spin down of < ∼ (cid:12) stars.One can draw several possible conclusions from thiscomparison. If we assume that angular-momentum evo-lution is roughly constant with time, then at least one ofthe cluster ages is incorrect. NGC 6811’s age could beyounger than 1 Gyr, which we tested by comparing thecluster data to progressively younger Matt et al. (2015)model populations. While these comparisons do showthat the cluster’s mass-period sequence is better fit (byeye) when using < P rot at progressively highermasses (e.g., at 963 Myr, the single-valued mass-periodsequence begins to fan out at ≈ (cid:12) ; at 837 Myr, at ≈ (cid:12) ; and at 653 Myr, at ≈ (cid:12) ). This spreadis not seen in the NGC 6811 data, suggesting that starswith masses > ∼ (cid:12) (the lowest mass for which we have Kepler data) have all had time to spin down to a slow-rotating sequence and setting a lower limit of ≈
800 Myrfor the cluster’s age.On the other hand, Praesepe could be older than pre-viously thought, as argued by Brandt & Huang (2015),who, by incorporating rotation into their evolutionarymodels, found that the cluster is closer to ≈
800 Myr inage. Increasing Praesepe’s age in this manner requiresexplaining the presence of fast rotators with masses be-tween 0.5 and 1.1 M (cid:12) , since these stars lie outside of therange of P rot predicted by the Matt et al. (2015) model.The cluster of Praesepe stars at ≈ (cid:12) and P rot ≈ > ∼ (cid:12) stars are binaries; in Praesepe, whichhas not been surveyed as extensively for binarity, halfof the rapidly rotating > ∼ (cid:12) stars are confirmed orcandidate binary systems, and the remaining > ∼ (cid:12) fast rotators are not confirmed single stars, because theyhave not been searched for companions. Finally, the mass-period data for the two clusters maybe suggesting that spin down progresses differently forsolar-mass and lower-mass stars. The right panel of Fig-ure 13, where the periods are plotted linearly, shows thatthere is evidence for spin down for the 0.9-1.0 M (cid:12) stars:for Praesepe, the 20 stars in this mass bin have a median P rot = 9.4 d, while for NGC 6811, the 11 stars have a me-dian P rot = 10.8 ± P rot is erased when considering 0.8-0.9 M (cid:12) stars, how-ever: the median for the 18 Praesepe stars is 10.8 d andfor the 15 NGC 6811 members it is 10.8 ± √ age spin down for these stars, The Meibom et al. (2011) periods are only for nominally singlemembers of NGC 6811; these authors have extensive RV data forthe cluster.
GC 752: Membership, Rotation, and Activity 15 P e r i o d ( d ) NGC 752 dataPraesepe dataNGC 6811 data
M2M0K5K2G2F5
M2M0K5K2G2F5 O • )0.11.010.0 P e r i o d ( d ) NGC 752 data1.344-Gyr model O • )10203040 Fig. 14.—
Top —
Comparison between the mass-period distribution for the joint Praesepe (empty black stars) and NGC 6811 (emptycyan stars) sample presented in Figure 13 and for NGC 752 (solid blue stars).
Bottom —
Comparison between the Matt et al. (2015)model for 1.344 Gyr (empty red stars) and the NGC 752 data. As in previous figures, on the left, the periods are plotted logarithmically,and on the right linearly. strengthens the impression that spin down is stalling forthese lower-mass stars: for 0.9-1.0 M (cid:12) stars, the modelpredicts a median P rot = 12 . (cid:12) stars,13.7 d, at 837 Myr. The potential stalling of spin downobserved for 0.8-0.9 M (cid:12) stars needs to be tested withdata from older clusters, with Ruprecht 147 a particu-larly promising cluster for this (J. Curtis, pers. comm.). Comparing NGC 752 to the younger clusters and tothe Matt et al. (2015) model
In Figure 14, we show a comparison of the combinedmass-period data for Praesepe and NGC 6811 and for the12 members of NGC 752 for which we have new P rot mea-surements (top row). The sparseness of the data for NGC752 make it difficult to draw strong conclusions from thiscomparison, although on average, the NGC 752 stars doappear to be rotating more slowly than their youngercounterparts. The difference is not significant, with thelowest-mass stars in NGC 752 in particular being indis-tinguishable in the mass-period plane from their cousins in Praesepe. For the eight 0.6-0.8 M (cid:12) stars in NGC 752,the median P rot = 16.6 ± P rot = 13.8 d.If we remove the two ≈ (cid:12) longest-period rotatorsin NGC 752, which have associated large period uncer-tainties, the median P rot for cluster stars in this mass bindrops to 14.9 ± (cid:12) stars in NGC 752,the median P rot = 18.9 ± P rot = 16.6 d for 83 stars. If we exclude the fast rotatingPraesepe stars in this mass bin, which are likely bina-ries, so as to focus the comparison on the slow-rotatingsequence only, the median Praesepe P rot is 18.1 d for 59stars.The comparison to the Matt et al. (2015) model shownin Figure 14 illustrates the difficulty of calibrating gy-rochronology models at these ages. Rather than a se-quence of slow-rotating, ≈ solar-mass stars as in Fig-ure 12, we have a handful of lower-mass stars with whichto anchor the comparison to the models. Still, it does6 Agüeros et al.appear that the model is significantly over-predicting thespin down for the 0.6-0.8 M (cid:12) stars, with the predictedmedian star in that mass range having a P rot = 21.0 dat 1.344 Gyr, ≈ (cid:12) star predicted to have a P rot = 17.2 d. One pos-sible interpretation is that we are seeing the evolution-ary stalling observed in the comparison of Praesepe andNGC 6811 for 0.8-0.9 M (cid:12) stars shifted to lower masses,with the 0.6-0.8 M (cid:12) stars being the ones now rotatingsignificantly faster than expected at this age. Comparing magnetic activity in NGC 752 and inthe Hyades and Praesepe
Studies of observational tracers of coronal or chromo-spheric activity have uncovered a mass-dependent tran-sition between active and inactive stars in open clus-ters (e.g., Kafka & Honeycutt 2006; Douglas et al. 2014;Núñez & Agüeros 2016; Núñez et al. 2017). The dividingline between these two populations shifts to lower massesin older clusters, indicating that lower-mass stars possesslonger activity lifetimes. For FGK stars, these lifetimesare ≤
650 Myr, as calibrated by observations of open clus-ters younger than NGC 752 (Hawley et al. 1999; Douglaset al. 2014; Núñez et al. 2017).Extending such measurements to older open clustersis a primary motivation of this work. Our knowledge ofthe chromospheric activity lifetimes of lower-mass starscurrently relies on indirect calibrations, such as modelingthe vertical gradient in H α emission line strengths as aconsequence of dynamical heating in the Galactic disk(e.g., West et al. 2008).Our spectroscopic campaign confirms that the bound-ary between active and inactive stars has shifted well intothe M dwarf regime in NGC 752. As noted above, thereare three stars in our spectroscopic sample with formaldetections of H α emission, but we do not consider thesestars indicative of the location of the active/inactiveboundary in this cluster. As Figure 11 demonstrates,even the modest activity signatures measured from thesestars (EqW > < P mem < bona fide cluster members. Calculating themean H α EqW for NGC 752 members in bins of ( r − K ),as shown in Figure 15, indicates that there is no transi-tion to activity in NGC 752, at least within the domainof our spectroscopic survey, which includes stars withspectral types as late as ≈ M2.Indeed, the comparison of EqW loci in Figure 15demonstrates that the activity properties of NGC 752’searly type (SpT < M2) members are fully consistent withthose of the largely inactive field star population. Com-paring the NGC 752 stars with those in Praesepe and theHyades, which exhibit a clear transition to active popu-lations at ( r − K ) ≈
4, also indicates that the locationof the active/inactive boundary shifts to lower masses asstars age from < (r-K) −10−8−6−4−20246 A v e r a g e H α E q W F2 G2 K2 K7 M0 M2 M4 M5Prae epeHyade NGC 752
Fig. 15.—
Average H α EqW as a function of logarithmicallybinned ( r − K ) for stars in NGC 752 compared to the youngerPraesepe and Hyades clusters. The number of stars in each binis along the top; the vertical bars show the σ for the bin, andthe horizontal bars show the extent of the bin. Measurementsof a comparison sample of ≈
752 should lie at a spectral type of ≈ M3. CONCLUSIONS
We present an updated list of likely cluster membersfor NGC 752, one of the only nearby open clusters signifi-cantly older than the Hyades. Our catalog is constructedby supplementing the catalogs of Daniel et al. (1994);Mermilliod et al. (1998) with candidates identified usingupdated photometric and proper motion criteria, and re-fined via radial velocity measurements. We produce alist of 258 probable cluster members, a >
50% increaseover previous catalogs, and in particular provide the firsthigh confidence list of late K- and M-dwarf members ofthe cluster.Using a Bayesian framework to fit MIST isochronesto literature photometry and the
Gaia
TGAS astrome-try available for 59 NGC 752 members, we derive max-imum likelihood mean parameters for the cluster. Wefind an age = 1 . ± .
06 Gyr, a distance = 438 +8 − pc(DM = 8.21 +0 . − . ), a [Fe/H] = 0 . ± .
01, and an A V =0 . +0 . − . . These cluster parameters are in agreementwith those of Bartaši¯ut˙e et al. (2011) and Twarog et al.(2015) and have more robust uncertainty estimates.We report on the results of our optical monitoring ofthe cluster. We targeted NGC 752 with PTF for fivemonths in 2010-2011, producing light curves with 400-700 R P T F measurements for 90 cluster members. We usethese these light curves to identify 12 high-confidence Kand M cluster members with reliable P rot measurements.These are the first periods measured for such low-massGC 752: Membership, Rotation, and Activity 17stars with a well-constrained age > ≈
650 Myr) and NGC 6811 ( ≈ ≈ solar-mass stars losing angular momentum as predicted bya Skumanich-type spin-down law, whereas 0.8-0.9 M (cid:12) stars do not appear to have spun down significantly overthe ≈
350 Myr that separate the two clusters. An alter-native interpretation is that at least one of the ages forthese two clusters is incorrect, as has already been ar-gued for Praesepe by Brandt & Huang (2015), who findits age to be closer to 800 Myr.The sparseness of the NGC 752 P rot data make it dif-ficult to draw strong conclusions from a comparison tothe data for the younger clusters or to the Matt et al.(2015) model. Although it does seem that, on average,the NGC 752 stars are rotating more slowly than theiryounger counterparts, the difference is not significant,and in particular the lowest-mass stars in NGC 752 forwhich we measure P rot are indistinguishable from theircousins in Praesepe. Comparisons with the Matt et al.(2015) model data suggest that the model over-predictsthe angular momentum lost by K and early M stars overtheir first 1.3 Gyr; this excess in the predicted spin downfor these stars was also observed when comparing themodel predictions to the data for the younger clusters.On the other hand, the Matt et al. (2015) model sys-tematically under-predicts the spin down of 0.4-0.6 M (cid:12) stars at Praesepe’s age, but the model P rot are consistentwith the P rot measured for these stars at NGC 752’s age.There are only four < (cid:12) NGC 752 stars for whichwe have these measurements, however.Finally, we discuss spectroscopic observations of over270 candidate cluster members with the MDM 2.4-m andWIYN 3.5-m telescopes. Based on our measurementsof H α , we find that NGC 752’s stars are magneticallyinactive at spectral types of ≈ M2 and earlier, and indeedthat these stars’ activity properties are fully consistentwith those of the largely inactive field-star population.Comparing the NGC 752 stars with those in Praesepeand the Hyades also indicates that the location of theactive/inactive boundary shifts to lower masses as starsage from < ≈ M3.The fraction of NGC 752 members for which we mea-sured P rot , 13%, is smaller than that we obtained in ourPTF Pleiades campaign (19%; Covey et al. 2016), but ishigher than that in our Praesepe campaign (7%; Agüeroset al. 2011). This highlights the challenge in definingappropriate metrics for identifying robust P rot measure-ments. These efforts are essential, however: while yieldsfrom satellite observations are much higher (i.e., essen-tially 100% for the Pleiades with K2 ; Rebull et al. 2016a),an analysis by Douglas et al. (2017) of the properties ofrotators in Praesepe suggested that K2 was not uncov-ering rotators with smaller amplitudes than those iden- tified from the ground. Even in the era of K2 and (soon)the Transiting Exoplanet Survey Satellite , ground-basedsurveys of rotation in clusters still have an importantrole to play. And forthcoming
Gaia data should solidifyand extend the membership of NGC 752 to lower masses,thereby increasing its importance for studies of low-massstars.We thank Eran Ofek for his help scheduling PTF ob-servations. David Fierroz participated in the WIYN andMDM observing runs, and we thank him for his help incollecting the spectra presented here. We thank JulesHalpern and John Thorstensen for their help with theMDM observations, and the WIYN observing specialistsfor their assistance with the Hydra observations. CatyPilachowski very generously shared the results of her RVmonitoring of NGC 752, and Stanislava Bartaši¯ut˙e andJustas Zdanavičius sent us their photometric catalog ofcandidate NGC 752 members, for which we thank them.We are grateful to Sean Matt for running many itera-tions of his model for us, enabling the comparisons pre-sented here, and for his comments on our results. Wethan Jason Curtis and Bruce Twarog for providing de-tailed comments on a draft, and the anonymous refereefor comments that improved the final paper.M.A.A. acknowledges support provided by the NSFthrough grant AST-1255419. P.A.C. acknowledges sup-port provided by the NSF through grant AST-1109612.The MDM Observatory is operated by Dartmouth Col-lege, Columbia University, Ohio State University, OhioUniversity, and the University of Michigan.This paper is based on observations obtained with theSamuel Oschin Telescope as part of the Palomar Tran-sient Factory project, a scientific collaboration betweenthe California Institute of Technology, Columbia Univer-sity, Las Cumbres Observatory, the Lawrence BerkeleyNational Laboratory, the National Energy Research Sci-entific Computing Center, the University of Oxford andthe Weizmann Institute of Science.This research has made use of NASA’s AstrophysicsData System Bibliographic Services, the SIMBADdatabase, operated at CDS, Strasbourg, France, theNASA/IPAC Extragalactic Database, operated by theJet Propulsion Laboratory, California Institute of Tech-nology, under contract with the National Aeronauticsand Space Administration, and the VizieR database ofastronomical catalogs (Ochsenbein et al. 2000).IRAF is distributed by the National Optical Astron-omy Observatories, which are operated by the Associ-ation of Universities for Research in Astronomy, Inc.,under cooperative agreement with the National ScienceFoundation. PyRAF is a product of the Space TelescopeScience Institute, which is operated by AURA for NASA.The Two Micron All Sky Survey was a joint projectof the University of Massachusetts and the Infrared Pro-cessing and Analysis Center (California Institute of Tech-nology). The University of Massachusetts was responsi-ble for the overall management of the project, the ob-serving facilities and the data acquisition. The InfraredProcessing and Analysis Center was responsible for dataprocessing, data distribution and data archiving.8 Agüeros et al.
TABLE 1SED Templates for USNO-B1.0Photometry
SpT M B M R M I M bol (mag) (mag) (mag) (mag)B8 − − − − SED TEMPLATES
We based our SED fitting procedures on those described by KH07, but since NGC 752 is not in the SDSS footprint,we extended the SED templates to use USNO-B1.0 photometry. We calculated the absolute magnitudes in the USNO-B1 filters (photographic
BRI ) by bootstrapping from our highly probable members of Praesepe and Coma Berenices,which span spectral types of A0-M7. For each star, we already had a measurement of m bol and SpT based on SEDfits to SDSS and 2MASS photometry. We then downloaded the USNO-B1.0 magnitudes for those stars and computedthe ( B − m bol ), ( R − m bol ), and ( I − m bol ) colors. Finally, we calculated the average value for these colors for SpT binsof cluster members (i.e., G4.0-G6.4 to correspond to G5 stars, or M0.6-M1.5 to correspond to M1 stars) and combinedthem with the M bol absolute values from KH07 to compute the absolute magnitudes M B , M R , and M I .For B8 stars, we linearly extrapolated the color-SpT relations of early A stars with respect to similar SDSS filters—( g − B ), ( r − R ), and ( i − I )—to compute absolute magnitudes from KH07. For the latest-type stars (M8-L0),we conducted a similar extrapolation on the colors of mid-M stars, then verified them by conducting SED fits on asample of bright ultracool dwarfs (from Leggett et al. 2002) that had photometry in both USNO-B1.0 and SDSS.There were too few ultracool dwarfs with photometry in USNO-B1.0 to justify fitting color relations to those data,but the measurements sufficed to confirm that the extrapolation from mid-M stars was valid. We give M B , M R , M I ,and M bol as a function of SpT in Table 1.Based on the scatter in colors between very similar filters (i.e., ( i − I ) and ( r − R )) in color-SpT relations for oursample of open cluster members, we estimate that the typical photometric uncertainty for USNO-B1.0 magnitudesis σ ≈ .
25 mag. Differences in the emulsions used for the original photographic plates also will introduce somecolor terms; for example, POSS-I conducted B “filtered” observations with a Kodak 103a-O emulsion and no filter,while POSS-II used Kodak IIIaJ emulsions with a GG385 filter. The corresponding southern surveys that contribute toUSNO-B1.0 (which are not relevant to our survey, but could be interpreted using the same SEDs) also used Kodak IIIaJemulsions, but with a slightly redder GG385 filter. The color terms appear to be small compared to the uncertaintyfor individual stars, so we computed a single calibration for all versions of B , R , and I . However, the color terms couldintroduce small systematic uncertainties in SED fits for stellar populations.GC 752: Membership, Rotation, and Activity 19 TABLE 1 P mem Selected NGC 752 Members
J K
Mass SpT DM M bol
Binary? a P mem (mag) (mag) (M (cid:12) ) (mag) (mag) (%)01501676+3812369 · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · TABLE 1 — Continued
J K
Mass SpT DM M bol
Binary? a P mem (mag) (mag) (M (cid:12) ) (mag) (mag) (%)01564860+3729114 622 9.71 ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · ± ± ± ± ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · ± ± ± ± ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± ± ± ± ± ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · GC 752: Membership, Rotation, and Activity 21
TABLE 1 — Continued
J K
Mass SpT DM M bol
Binary? a P mem (mag) (mag) (M (cid:12) ) (mag) (mag) (%)01581427+3700453 · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · ± ± ± ± · · · ± ± · · · ± ± · · · ± ± ± ± · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± · · · ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · · · · ± ± ± ± · · · b · · · · · · · · · · · · M3.7 6.84 ± ± · · · a Based on RV measurements published by Daniel et al. (1994) (“D”) or Mermilliod et al. (1998), or collected byC. Pilachowski (“P”). b P mem = 89.5%) clustermember; the source does not factor in to our subsequent analysis, so we do not attempt to derive a separatenon- K -band based mass estimate.2 Agüeros et al. TABLE 2Other NGC 752 Members
J K
Mass SpT RV D RV P Binary? a (mag) (mag) (M (cid:12) ) (km s − ) (km s − )01510351+3746343 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · ± · · · · · · · · · ± ± · · · · · · ± · · · · · · ± ± · · · F5.6 · · · ± . ± . · · · ± ± · · · · · · ± − ± . ± . DP01553936+3752525 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · ± ± · · · · · · ± ± . ± . · · · ± ± · · · · · · ± ± . ± . · · · · · · ± ± ± · · · D01561550+3738416 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · ± · · · DM01562163+3736084 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± ± · · · D01564975+3801216 626 8.20 ± ± · · · · · · ± ± . ± . · · · · · · ± ± · · · · · · ± · · · DM01565395+3819272 · · · ± ± · · · · · · · · · · · · · · · ± ± · · · · · · · · · ± . ± . · · · · · · ± ± · · · F5.1 · · · · · · · · · ± ± · · · · · · · · · ± . ± . P01570311+3808026 · · · ± ± · · · · · · · · · · · · · · · ± ± · · · F4.0 − ± ± . ± . DP01571709+3726089 · · · ± ± · · · · · · · · · · · · ± ± · · · · · · ± · · · · · · ± ± · · · ± . ± . P01573621+3745101 849 9.05 ± ± · · · · · · ± ± . ± . DP01573696+3702110 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · ± · · · DM01573767+3749008 857 9.07 ± ± · · · · · · ± ± . ± . DP01573895+3746123 · · · ± ± · · · · · · ± · · · M01574472+3759184 · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · ± · · · · · · ± ± · · · · · · · · · ± . ± . · · · · · · ± ± · · · · · · ± · · · D01580924+3728355 993 12.27 ± ± ± ± . ± . · · · ± ± ± ± . ± . · · · · · · ± ± · · · · · · ± · · · · · · ± ± ± ± . ± . · · · ± ± · · · · · · ± ± . ± . DP01584005+3738051 1129 10.91 ± ± · · · · · · ± ± . ± . D01584793+3826078 · · · ± ± · · · F4.8 · · · · · · · · · · · · ± ± · · · · · · ± · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · · · · · · · · · · ± ± · · · · · · · · · a Based on RV measurements published by Daniel et al. (1994) (“D”) or Mermilliod et al. (1998), or collected byC. Pilachowski (“P”).GC 752: Membership, Rotation, and Activity 23
REFERENCESAgüeros, M. A. et al. 2011, ApJ, 740, 110Angus, R., Aigrain, S., Foreman-Mackey, D., & McQuillan, A.2015, MNRAS, 450, 1787Anthony-Twarog, B. J., Deliyannis, C. P., Twarog, B. A., Croxall,K. V., & Cummings, J. D. 2009, AJ, 138, 1171Barnes, S. A. 2010, ApJ, 722, 222Bartaši¯ut˙e, S., Deveikis, V., Straižys, V., & Bogdanovičius, A.2007, Baltic Astronomy, 16, 199Bartaši¯ut˙e, S., Janusz, R., Boyle, R. P., & Philip, A. G. D. 2011,Baltic Astronomy, 20, 27Bertin, E., & Arnouts, S. 1996, A&AS, 117, 393Brandt, T. D., & Huang, C. X. 2015, ApJ, 807, 24Choi, J., Dotter, A., Conroy, C., Cantiello, M., Paxton, B., &Johnson, B. D. 2016, ApJ, 823, 102Covey, K. R. et al. 2016, ApJ, 822, 81—. 2007, AJ, 134, 2398Curtis, J. L., Wolfgang, A., Wright, J. T., Brewer, J. M., &Johnson, J. A. 2013, AJ, 145, 134Daniel, S. A., Latham, D. W., Mathieu, R. D., & Twarog, B. A.1994, PASP, 106, 281Deacon, N. R., & Hambly, N. C. 2004, A&A, 416, 125Delfosse, X., Forveille, T., Ségransan, D., Beuzit, J., Udry, S.,Perrier, C., & Mayor, M. 2000, A&A, 364, 217Delorme, P., Collier Cameron, A., Hebb, L., Rostron, J., Lister,T. A., Norton, A. J., Pollacco, D., & West, R. G. 2011,MNRAS, 413, 2218Dinescu, D. I., Demarque, P., Guenther, D. B., & Pinsonneault,M. H. 1995, AJ, 109, 2090Dotter, A. 2016, ApJS, 222, 8Dotter, A., Chaboyer, B., Jevremović, D., Kostov, V., Baron, E.,& Ferguson, J. W. 2008, ApJS, 178, 89Dotter, A., Conroy, C., Cargile, P., & Asplund, M. 2017, ApJ,840, 99Douglas, S. T. et al. 2014, ApJ, 795, 161Douglas, S. T., Agüeros, M. A., Covey, K. R., Cargile, P. A.,Barclay, T., Cody, A., Howell, S. B., & Kopytova, T. 2016,ApJ, 822, 47Douglas, S. T., Agüeros, M. A., Covey, K. R., & Kraus, A. 2017,ApJ, 842, 83Francic, S. P. 1989, PhD thesis, California Univ., Santa Cruz.Gaia Collaboration et al. 2016, A&A, 595, A1Gallet, F., & Bouvier, J. 2015, A&A, 577, A98Giardino, G., Pillitteri, I., Favata, F., & Micela, G. 2008, A&A,490, 113Guo, X., Johnson, J. A., Mann, A. W., Kraus, A. L., Curtis,J. L., & Latham, D. W. 2017, ApJ, 838, 25Hawley, S. L., Reid, I. N., Gizis, J. E., & Byrne, P. B. 1999, inAstronomical Society of the Pacific Conference Series, Vol. 158,Solar and Stellar Activity: Similarities and Differences, ed.C. J. Butler & J. G. Doyle, 63Høg, E. et al. 2000, A&A, 355, L27Howell, S. B. et al. 2014, PASP, 126, 398