Spectroscopy of M81 Globular Clusters
aa r X i v : . [ a s t r o - ph . C O ] A p r Spectroscopy of M81 Globular Clusters Julie B. Nantais and John P. Huchra
Harvard-Smithsonian Center for Astrophysics60 Garden Street, Cambridge, MA 02138, USA
ABSTRACT
We obtained spectra of 74 globular clusters in M81. These globular clustershad been identified as candidates in an HST ACS I-band survey. 68 of these 74clusters lie within 7 ′ of the M81 nucleus. 62 of these clusters are newly spec-troscopically confirmed, more than doubling the number of confirmed M81 GCsfrom 46 to 108. We determined metallicities for our 74 observed clusters usingan empirical calibration based on Milky Way globular clusters. We combinedour results with 34 M81 globular cluster velocities and 33 metallicities from theliterature and analyzed the kinematics and metallicity of the M81 globular clus-ter system. The mean of the total sample of 107 metallicities is − . ± r (deprojected) = 108 ±
22 km s − overall. This result is likelybiased toward high rotational velocity due to overrepresentation of metal-rich,inner clusters. The rotation patterns among globular cluster subpopulations areroughly similar to those of the Milky Way: clusters at small projected radii andmetal-rich clusters rotate strongly, while clusters at large projected radii andmetal-poor clusters show weaker evidence of rotation. This study uses observations from the MMT Observatory, a joint facility of the Smithsonian Institutionand the University of Arizona.
1. Introduction
The study of globular clusters (GCs) is an excellent way to learn about the develop-mental history of nearby galaxies, for several reasons. First, they provide a “fossil record”of a galaxy’s most intense episodes of star formation. Secondly, since GCs are close ap-proximations of simple stellar populations (SSPs), it is relatively easy to estimate their agesand metallicities spectroscopically. Also, having luminosities of ∼ − M ⊙ , they aremuch brighter than most individual stars and can thus be more easily studied photomet-rically and spectroscopically. Brodie and Strader (2006) provide a comprehensive reviewof our understanding of GC systems (GCSs) and how they formed. One common find-ing is that most elliptical and lenticular galaxies (Larsen et al. 2001; Kundu and Whitmore1998; Lee, Kim, and Geisler 1998) and early- and intermediate-type spiral galaxies (Zinn1985; Huchra, Stauffer, and Van Speybroeck 1982) have bimodal metallicity distributions intheir GCSs, comprised of an old metal-poor (MP) subpopulation and a slightly youngermetal-rich (MR) subpopulation. The mean metallicities of these MR and MP subpopula-tions (Strader, Brodie, and Forbes 2004) and the mean metallicity of the GCSs as a whole(Brodie and Huchra 1991) are found to vary systematically with the mass of the galaxy.While GC metallicity distributions among galaxies of diverse masses and Hubble typesshow a systematic pattern, the kinematic properties of GCSs tend to be qualitatively differenteven among galaxies of similar mass and Hubble type. For instance, while rotation is foundamong the MR and MP GC subpopulations of the Virgo giant ellipticals M87 (Cˆot´e et al.2001) and M60 (Hwang et al. 2008), the GCS of Fornax giant elliptical NGC 1399 does notappear to be rotating at all (Richtler et al. 2004). Such kinematic differences are also seenamong disk galaxies. The GCS of M31 shows strong rotation (Lee et al. 2008; Perrett et al.2002), while in the Milky Way GCS there is only notable rotation among bulge GCs andvery MP halo clusters (Harris 2001). In contrast to both of these galaxies, M104, a nearbymassive S0/a galaxy, shows no signs of net rotation of its GCS as a whole. This result holdseven in sub-populations of the M104 GCS divided according to metallicity or galactocentricdistance (Bridges et al. 2007). Some late-type spiral galaxies in the Sculptor Filament showpossible evidence of strongly rotating GCSs (Olsen et al. 2004), but Nantais et al. (2010a)did not find clear evidence of rotation in the GCS of the Sculptor galaxy NGC 300, andMora et al. (2008) did not find notable rotation in a spectroscopic study of NGC 45. Thesevariable rotation patterns in the kinematics of GCSs suggest that similar galaxies may havehad very different early formation histories that do not necessarily follow a neat patternaccording to mass and/or Hubble type.M81, an Sab spiral galaxy with a distance modulus of 27.7 (Freedman at al. 2001), isan ideal target for furthering our understanding of spiral galaxy GCSs. Three spectroscopic 3 –studies of M81’s GCS have been performed thus far. Brodie and Huchra (1991) (hereafterBH91) observed eight GCs in M81 as part of a multi-galaxy study, with a variance-weightedmean metallicity of -1.46 (unweighted mean -1.08). Perelmuter et al. (1995) (hereafter PBH)estimated the metallicities and radial velocities of 25 GCs from a sample of 82 candidateschosen from the Perelmuter and Racine (1995) (PR95) GC candidate catalog. PBH founda mean metallicity of -1.48 ± ± § § § § §
2. Data
We obtained spectra of a total of 207 extended objects in M81 on the nights of 2006May 3-5 and 2007 November 13 and 17 with Hectospec on the 6.5 m MMT on Mt. Hopkinsin Arizona. Hectospec has 300 fibers of ∼ ′′ diameter. The number of objects we couldobserve was limited by how many fibers could be placed within a small region; we could notobserve more than ∼
75 objects on and near the disk of M81 (within the range of the Nantaiset al. 2009b HST I-band images) in a single pointing. We used the 270 line mm − grating,with spectral coverage from 3700-9150 ˚A and spectral resolution of about 5.1 ˚A, similar tomany other studies on extragalactic GCs. The fibers covered a diameter of 1.5 ′′ on the sky.The objects observed on 2006 May 3 and 4 have total exposure times of 80 minutes(four 20-minute exposures), and the spectra from 2006 May 5 have total exposure times of60 minutes (three 20-minute exposures). Objects for these three exposures were selected fromthe Nantais et al. (2010b) catalog on the basis of magnitude, with bright objects preferredover faint objects. The objects observed in November 2007 had 6 20-minute exposures on eachnight, but four of those exposures on November 17 were affected by clouds, so we combined8 exposures, 6 from November 13 and 2 from November 17, to create the final spectra. For 4 –the November 2007 observations, we gave extra priority to objects with an especially highlikelihood of being GCs, based on both their Nantais et al. (2010b) g-r and r-I colors andtheir visual appearance. We also prioritized observation of objects with low-quality or noprior spectra in either the May 2006 observation run or the existing literature.Ideally, we would accomplish background subtraction by measuring the sky level usingapertures on either side of the object. This would account for both the temporal and spatialvariation in the background. However, we cannot do this with a fiber spectrograph. Instead,we observed several sky offset frames each night, each taken about 5 ′′ to the north, south, east,or west of the object exposures. For the 2006 May 4 observations, only one sky offset exposurewas used to sky-subtract these images, since the rest had sky fluxes either notably brighteror notably fainter than in the object exposures. The spectra extracted using these unreliablesky exposures suffered from either over-subtracted or under-subtracted sky continuum thatdominated the object’s own spectral features. On 2007 November 17, three of the skyexposures were lost to cloudy weather, so the total November 2007 sky offset spectrum wasan average of 9 expsoures. Table 1 lists the observation details for each night: the numberof objects observed, the number of object pointings, the number of sky pointings, and thetotal integration time.Each CCD image was flat-fielded, bias-corrected and wavelength-calibrated by the SPECROAD automated pipeline. This pipeline uses the “hectospec” package of IRAF tasks specially writ-ten at the Harvard-Smithsonian Center for Astrophysics (CfA) to facilitate the reduction ofHectospec data, plus tasks from other IRAF packages such as “mscred” and “rvsao.” All filesused for bias subtraction, flat fielding, and wavelength calibration (lamp files) were taken onor as near as possible to the date of observation. The automated pipeline followed to its fullextent uses the nearest sky fiber from a given pointing in order to subtract the background,which does not account for the very high spatial variation in the background. Therefore, inorder to perform sky subtraction, we used object exposure spectra and sky offset exposurespectra produced by an earlier stage in the pipeline process in which sky subtraction andvelocity determination had not yet been performed. We averaged all individual exposuresof both the object and the sky offset taken during acceptable weather conditions, and sub-tracted the average sky offset spectra from the average object spectra. Flux calibration wasthen performed using a standard star spectrum observed on or as near as possible to thenight of observation. Any remaining cosmic rays and poorly subtracted telluric lines wereremoved individually from the final object spectra using the “x” command in the IRAF taskspecplot in the noao.onedspec package.
3. Object Identification
Objects were identified by a combination of radial velocity and visual inspection of theirspectroscopic features. Any object with a radial velocity greater than ∼
400 km s − and clearevidence of redshifted absorption or emission features was determined to be a backgroundgalaxy. H II regions were recognizable by their emission lines near zero redshift. GCs had ared spectrum and discernible old stellar population absorption features near zero redshift,and radial velocities less than 400 km s − . Objects with no or very faint emission lines,a blue continuum, and absorption lines typical of a young (A-F) stellar population werelabeled as open clusters. The plausibility of a given object identification was confirmed viathe HST I-band images. No objects identified in our catalog as having GC-like morphologywere spectroscopically determined to be young or open clusters. Since our spectroscopiccandidates were chosen based on size information in space-based imaging data, they arealmost certain not to be stars. Figure 1 shows the locations of different objects with respectto the M81 disk. Figure 2 shows sample spectra of GCs, including a combined spectrumof all GCs. We found a total of 74 GCs, 47 H II regions, 23 background galaxies, 7 openclusters, and 56 objects with too low signal-to-noise to determine their nature. Of the 74GCs, 12 have been previously observed: 8 by SBKHP, 3 by PBH, and one by BH91. Wehave thus identified 62 newly-confirmed GCs. Figure 3 shows the distribution of the radialvelocities of all recognized objects - GCs, H II regions, and open clusters combined - withinM81. No recognized object that is not a background galaxy has a radial velocity greaterthan 400 km s − or less than −
400 km s − .One object, 1859 (Figure 4), showed both strong GC continuum and absorption featuresand notable H II emission. Its HST image resembled a GC, with what may have been a smallH II region some 2 ′′ away. We therefore treat it as a GC, and remove the H II emission inorder to calculate its radial velocity. The GC absorption features yield a radial velocity of114 ±
17 km s − , while the H II emission lines yield a radial velocity of 35 ±
65 km s − .Table 2 shows the positions and radial velocities of our GCs and the 34 GCs fromBH91, PBH, SBKHP that we did not re-observe. The first few columns of Table 2 give theID number in our catalog, in the PBH and SBKHP spectroscopic catalogs, and in the PR95and Chandar, Ford, and Tsvetanov (2001) (CFT01) photometric catalogs. The next threecolumns give the RA in hours, the declination in degrees, and the projected galactocentricdistance in arcminutes. The final two columns in Table 2 give the radial velocity and itsuncertainty in km s − . Table 3 shows the positions, velocities, and object types of all HIIregions, galaxies, and open clusters. It is organized similarly to Table 2, but with an extracolumn after the ID number in our catalog listing the object type: galaxy (gx), HII region(hii), or open cluster (oc). 6 –Our spectroscopic identifications of objects can be compared to the visual identifica-tions in the I-band images in Nantais et al. (2010b), as a check on the reliability of thevisual classification scheme in the Nantais et al. (2010b) catalog. Of the 107 GC candidateswe observed, 80 had high enough signal-to-noise to confirm their nature; and of those 80objects, 73 (91%) were confirmed as GCs, 5 (6%) were found to be H II regions, and 2 (3%)were identified as galaxies. Assuming these spectroscopic results are representative of GCcandidates, the vast majority of our GC candidates in Nantais et al. (2010b) are true GCs.We also observed 26 H II region/OB association candidates, 23 galaxy candidates, and 50“Other” or unclassified objects. All 26 H II region and OB association candidates were foundto be either H II regions or open clusters; 24 (92%) were confirmed as H II regions, and theremaining two objects were dominated by blue continuum and were classified as open clus-ters. Of the 23 galaxy candidates, 17 had spectra with sufficient signal-to-noise to determinethe object type, and all 17 were confirmed to be galaxies. Of the 50 “Other” or unclassifiedobjects, only 27 had sufficient signal-to-noise to be identified. One of these spectra (3.7% ofthe identified “Others”) was confirmed as a GC, 17 (63%) were confirmed as H II regions, 5(18.5%) were confirmed as open clusters, and 4 (14.8%) were confirmed as galaxies. Theseresults, if typical of “Other” objects, suggests, as we found in Nantais et al. (2010b), thatthe “Other” category is dominated by young objects such as H II regions, OB associations,and open clusters, with most of the remaining objects probably being galaxies.
4. Metallicity Analysis
Spectral indices and metallicity estimates were determined using the same methods andwavelength definitions as in Nantais et al. (2010a), which was based on the Brodie and Huchra(1990) method but using indices calibrated on the Schiavon et al. (2005) Milky Way GCspectra smoothed to 5 ˚A resolution. GC metallicites from Harris (1996) were used for thecalibration of index strength to metallicity. Since we did not observe any Lick standard stars,we do not correct our fluxes to the Lick system. We calibrated the full list of 25 indices usedin Nantais et al. (2010a), and chose which ones to use on the basis of both the quality of thecalibration and the scatter of the index in the M81 data. We chose MgH, Mg2, Mgb, Fe5270,Fe5335, Fe5406, G4300, δ , and CNR to determine a weighted average of the metallicity. Ta-bles 4 and 5 give the spectral indices we measured. In Tables 4 and 5, the first column isthe ID in our catalog, and the remaining columns give the values of the individual spectralindices named in the column header in magnitudes. Below each row of data beginning withan ID number is a row labeled “ σ .” Each “ σ ” row lists the 1 σ uncertainty in each index, inthat index’s column, for the object whose ID number is listed in the previous row. Figures5 and 6 show the spectral indices used for metallicity estimates as a function of the Mg2 7 –index. Table 6 gives the metallicites we calculated. The table is arranged as three sets ofthree columns each. The first column in each set of 3 (columns 1, 4, and 7) gives the objectID, the second column in each set (columns 2, 5, and 8) gives the metallicity, and the thirdcolumn in each set (columns 3, 6, and 9) give the 1 σ uncertainty in the metallicity. Table 7shows the linear fits to the Milky Way index-metallicity relations, with the index ID in thefirst column, the slope (a) and intercept (b) of the [ F e/H ] = a ( index ) + b in the next twocolumns, the correlation coefficient R in the fourth column, and the R I , σ m , and σ s columnsas defined in Brodie and Huchra (1990) in the next three columns. In Table 7 our σ s (un-certainty of repeat exposures) estimates are based on the largest dispersion of index valuesof the four individual exposures (in the same observing run) of the two brightest GCs: 2029and 743. The dispersions in the index were calculated for each object, and the largest of thetwo dispersions was adopted as σ s . Ideally, we would use several observations of the samehigh signal-to-noise object in two or more separate observing runs on two or more differentnights, but we did not have such observations. The index differences in repeat observationsof objects on separate nights were dominated by the low signal-to-noise of these objects inone or both exposures; therefore, we concluded that the safest way to measure σ s was to userepeat observations of very high signal-to-noise objects on the same night.Our own sample of 74 GCs has a mean metallicity of − . ± .
08. Beyond our ownsample, there are also 34 GCs from SBKHP, PBH, and BH91 that we did not observe, 33of which have spectroscopic metallicity estimates. In Nantais et al. (2010b), we raise thepossibility that six of the PBH clusters and two of the SBKHP clusters may have beenmisidentified stars, but here we shall assume that all objects identified as clusters in theliterature are indeed GCs. For more complete analysis, however, we also perform metallicity,kinematics, and mass analysis excluding these 8 objects. Combining our metallicities withthose of the 33 clusters from the pre-existing literature, the mean metallicity of the M81GCS is < [Fe/H] > = − . ± .
07. If we exclude the 8 clusters flagged as possible stars fromthe total sample, we find a slightly lower mean metallicity, < [Fe/H] > = − . ± .
07. Themean metallicity of the M81 sample, with or without the starlike objects, is comparable tothe divide between the MR and the MP GCs in the metallicity histogram shown in Figure7. Figure 7 displays the metallicities of GCs in the M81, the Milky Way (Harris 1996), andM31 (Barmby et al. 2000) in 0.25 dex bins. The divide in the M81 sample seems lower inmetallicity than the Milky Way and M31 samples, and M81 appears to have somewhat ofan excess of metallicities in the − . − . ∼ harris/mwgc.dat σ − -weighted linear least-squares fit to metallicity as a function of galactocentricradius. The mean metallicity appears to remain nearly constant within ∼ − . ± .
020 dex kpc − . Eliminating the eight starlike clusters fromthe sample, the metallicity gradient increases to − . ± .
020 dex kpc − . 9 –Figure 10 also suggests that there may be a metallicity gradient among the MP clustersalone. A least-squares linear fit to the MP GCs alone gives a slope of − . ± .
012 dexkpc − . No clear evidence of a metallicity gradient is seen in the MR subsample, but largeuncertainties and sample biases make this difficult to determine. In theory, the apparentmetallicity gradient in the MP subsample could be an artifact of differences between themetallicity scales of our work and PBH, SBKHP, and BH91. All confirmed GCs beyond ∼ − . ± .
014 dex kpc − . Furthermore, if we compare the mean of the metallicitiesof the 12 GCs we observed that were also observed by SBKHP, PBH, or BH91, we find thatour mean metallicity is only 0.11 dex higher than theirs and within the scatter in the meanof the 12 metallicities ( < [ F e/H ] > us = − . ± .
14 vs. < [ F e/H ] > literature = − . ± . R >
10 kpc GC sample from the Milky Way and M31reduces their mean metallicities ( − .
30 for the Milky Way and − .
24 for M31) by only 0.05-0.06 dex. If we do have half of the halo population of M81 in our sample, then perhaps theGCS of M81 is more MR than either of these two galaxies. However, the mean metallicityof GCs in the inner 5 kpc of M81 is very similar to the mean metallicities of M31 and MilkyWay GCs within this distance: < [ F e/H ] > R< kpc = − .
03 for M81 ( − .
05 excluding the 8starlike objects), − .
00 for M31, and − .
07 for the Milky Way.
5. GC Kinematics5.1. Rotation, Mean Velocity, and Velocity Dispersion
Figure 12 shows the GC radial velocities as a function of position angle measured indegrees east of north, and Figure 13 shows the mean velocities in 30 ◦ position angle bins.Overplotted are the peak and outer edge H I rotation curves from Rots (1975), and a σ − -weighted least-squares fit to the rotation curve of the M81 GCs. The formula for the rotationcurve is V rot,proj. = V c + V pr sin ( φ − φ ), where V c is the mean velocity, V pr is the projectedradial velocity ( V rot sin ( i )), and φ is the position angle of the rotation axis. To deprojectour rotational velocities, we adopted an inclination angle of 59 ◦ (Rots 1975) and a positionangle of 157 ◦ (de Vaucouleurs et al. 1991). Figure 14 shows the GC velocity vs. positionangle at different distances from M81, following the distance binning of SBKHP: R < ≤ R ≤ > ≥ − .
06, and the MP subpopulationis all GCs with [Fe/H] < − .
06. The second column lists the number of clusters in thesubpopulation. The third column lists V c . The fourth column lists the deprojected rotationalvelocity V r , and the fifth column lists the projected radial velocity V pr = V r sin ( i ) determinedby our least squares fit. The sixth column lists the position angle φ . The final two columnslist the velocity dispersion of each subpopulation with and without correction for the rotationof the subpopulation. 11 –The mean radial velocity of all 108 GCs is -23 ± − ; excluding the 8 “starlike”objects reduces the mean radial velocity to -24 ± − . In order to compare our mean GCvelocity to a velocity for the M81 nucleus derived from a relatively internally consistent setof measurements, we searched the Center for Astrophysics spectroscopy archives for spectraof the M81 nucleus from the Z-Machine and FAST spectrographs. We found 42 spectra,15 from the Z-Machine dating between 1978 and 1980 and 27 from FAST dating between1994 and 1996. All spectra had radial velocities measured via either cross-correlation (likethe radial velocities of our GC spectra) or a combination of cross-correlation and emissionlines. Most also had heliocentric velocity corrections listed. For those that did not haveheliocentric corrections - the older FAST spectra - we calculated corrections using the IRAFtask “bcvcorr” in the rvsao package. We applied the heliocentric corrections to each velocityand found an error-weighted mean of -21 ± − — highly consistent with our meanGC radial velocity.The GCS of M81 shows strong evidence of rotation as a whole, at a rate of 108 ±
22 kms − , roughly half the Rots (1975) peak H I rotation rate of ∼
240 km s − . This rotation rate isonly slightly less than that of the MR subpopulation of the Milky Way (118 km s − , Harris2001), and larger than that of the Milky Way as a whole, yet is modest compared to therapid rotation rate of 190 km s − that Lee et al. (2008) find for the M31 GCS. The exclusionof the 8 starlike objects raises the total M81 GC rotation rate by a statistically insignifcantamount, to 113 ±
23 km s − . The rotation-uncorrected velocity dispersion of the M81 GCSis 145 km s − , and is unaffected if the 8 starlike objects are excluded. Applying the rotationcorrection gives a velocity dispersion of 130 km s − , similar to that of M31 (Lee et al. 2008).Our high overall rotation rate may be attributable to the bias toward inner, bulge and diskGCs in our sample discussed in §
4. We discuss the rotation rates of subpopulations definedaccording to projected distance and metallicity below.In Figure 14 and Table 8, rotation is visibly apparent in the low and intermediateprojected distance bins, but less obvious for GCs at large projected distances. The innersample yields a (deprojected) rotation velocity of 133 ±
40 km s − , while the middle sampleyields a rotation velocity of 87 ±
24 km s − . This contrasts somewhat with the results ofSBKHP, who find the greatest rotation to occur among the intermediate-projected-distanceGCs. The outer sample nominally has a higher rotational velocity, 101 ±
33 km s − , but itis more uncertain. The velocity dispersion of the outer clusters is reduced by only 1 km s − when this rotation is applied, as opposed to 4 km s − for the intermediate-projected-distancepopulation and 17 km s − for the inner population. At intermediate projected distances,the rotation-corrected velocity dispersion is 133 km s − , and at large projected distances itis 132 km s − , similar to the 130 km s − calculated for the entire GCS. At small projecteddistances, the rotation-corrected velocity dispersion is slightly higher, at 138 km s − . 12 –The difference between our results and SBKHP’s results may be due to SBKHP havinga much smaller sample at R < II region, all less than 1.6 ′ from the center of M81) in which the M81background light from the bulge was strong enough to extract and analyze separately as ifit were a star cluster. The metallicities derived for these background bulge light sampleswere all around -0.4 dex with little scatter (0.04 dex), whereas the mean metallicity of ourGCs is -0.96 dex with a scatter of 0.65 dex. Even the very innermost GC sample, within 2kpc of the center of M81 and thus most likely to be polluted by M81 background light, hasa mean metallicity of -0.94 dex and a scatter of 0.59 dex. We therefore conclude that ourinner GC spectra are not simply M81 background light pollution, and thus our velocitiesand metallicities are most likely valid.Figure 15 shows the velocity as a function of position angle and the rotation curve fitsfor the MR and MP subpopulations. The MP subpopulation has one significant outlyingcluster, 1352 (marked with an “x” in Figure 12). It is a very bright (I = 16.83) clusterlocated about 0.66 ′ from the nucleus, with a radial velocity of 353 ±
12 km s − . If thisobject is included in the least-squares weighted fit to the MP rotation curve, the MP GCrotational velocity is increased to a large value (161 km s − ), and correcting for the rotationactually increases the velocity dispersion from 142 km s − to 154 km s − . We find the bestrotational velocity fit, giving the greatest reduction in MP GC velocity dispersion, by using σ − weights and excluding Object 1352. MR GCs rotate at about 122 km s − , while MPGCs rotate at about 67 km s − if they rotate at all. The MP velocity dispersion is reducedby only 1 km s − when the rotation correction is applied, and the uncertainty in the rotationvelocity is quite large (38 km s − , nearly half the magnitude of the rotation velocity itself).This is similar to what is found in the Milky Way (Zinn 1985; Harris 2001), where inner,MR bulge/disk GCs rotate significantly but the MP halo population as a whole does not.We can also revisit the effects of Object 1352 on the total and inner M81 GC rotationrates. Recalculating the total rotation curve without Object 1352 gives V r,tot = 93 ± − (down from 108 km s − ), V = -29 ±
11 km s − , and φ = 280 ◦ . Recalculatingthe inner rotation curve minus Object 1352 gives V r,tot = 102 ±
29 km s − (down from 133km s − ), V = -34 ±
19 km s − , and φ = 271 ◦ . We then compared the rotation-correctedvelocity dispersions including and excluding Object 1352, calculated by subtracting eachGC’s predicted radial velocity using the above rotation curve results from its actual radialvelocity and finding the dispersion of these adjusted velocities. The total rotation-corrected 13 –velocity dispersion (including 1352) increases slightly to 131 km s − , and the inner rotation-corrected velocity dispersion (including 1352) is slightly improved at 138 km s − . Thus,Object 1352 increases the rotation velocities in these subsamples by about 1 σ , but does nothave a drastic effect on the velocity dispersion as with the MP GCs.Overall, the analysis of rotation rate as a function of distance and metallicity suggeststhat, while there may be some rotation among outer GCs and possibly MP GCs, the mostobvious rotation is among inner and MR GC populations, much as in the Milky Way. SinceGCs within the inner 4 kpc make up half our sample, and MR GCs within the inner 4 kpcmake up a full quarter of our sample, the overall rotation rate we observe is most likely biasedtoward high values due to the overrepresentation of M81’s rapidly rotating bulge/disk GCpopulation. However, unlike in the Milky Way, there is some evidence for relatively weakrotation among GCs at distances greater than 4 kpc and possibly (though this is less clear)among MP GCs. More MP halo GCs in M81 would need to be spectroscopically confirmed,from a survey covering larger galactocentric distances, to better understand the kinematicsof M81’s GCS. The velocity dispersion of all 108 GCs, corrected for the mean uncertainty in velocity,is 145 km s − . Using the projected mass estimator (Heisler, Tremaine, and Bahcall 1985),assuming isotropic orbits and subtracting a systemic velocity of -34 km s − from all GC ve-locities, the 108 clusters yield a mass of (1.64 ± × M ⊙ for M81. The 100 non-starlikeclusters yield a total projected mass of (1.56 ± × M ⊙ . Errors are estimated usingthe bootstrap method, randomly resampling from the original set of v r values 10,000 times.Our total M81 mass is considerably smaller than previous determinations ( ∼ × M ⊙ inSBKHP and PBH), but this seems to be because most of our clusters come from the innerregions, leading to smaller v r values on average. Our mass is similar to the Rots (1975)total mass estimate of 1.7 × M ⊙ . Mass estimates as a function of maximum distancefrom the center are shown in Table 9. The first column lists the number of clusters in eachsubsample defined by a maximum galactocentric distance. The second column lists the max-imum galactocentric distance of each subsample. The third column lists the projected massestimate derived from all clusters within that projected distance. 14 –
6. Summary
Overall, we find the GCS of M81 to be similar to those of M31 and the Milky Way.A KMM test provides modest statistical evidence that the M81 GC system has a bimodalmetallicity distribution, with peaks in the metallicity distribution similar to those of M31and the Milky Way. The mean metallicity of 107 GCs is –1.06 ± REFERENCES
Ashman, K. M., Bird, C. M., and Zepf, S. E. 1994, AJ, 108, 2348Barmby, P., Huchra, J. P., Brodie, J. P., Forbes, D. A., Schr¨oder, L. L., and Grillmair, C. J.2000, AJ, 119, 727.Beasley, M. A., Baugh, C. M., Forbes, D. A., Sharples, R. M., and Frenk, C. S. 2002,MNRAS, 333, 383.Bridges, T. J., Rhode, K. L., Zepf, S. E., and Freeman, K. C. 2007, ApJ, 658, 980.Brodie, J. P. and Huchra, J. P. 1990, ApJ, 362, 503.Brodie, J. P. and Huchra, J. P. 1991, ApJ, 379, 157. 15 –Brodie, J. P. and Strader, J. 2006, ARA&A, 44, 193.Chandar, R., Ford, H. C., and Tsvetanov, Z. 2001, AJ, 122, 1330.Cˆot´e, P., Marzke, R. O., and West, M. J. 1998, ApJ, 501, 554.Cˆot´e, P., McLaughlin, D. E., Hanes, D. A., Bridges, T. J., Geisler, D., Merritt, D., Hesser,J. E., Harris, G. L. H., and Lee, M. G. 2001, ApJ, 559, 828.de Vaucouleurs, G., de Vaucouleurs, A., Corwin, H., Buta, R., Paturel, G., and Foqu´e, P.1991, Third Reference Catalog of Bright Galaxies, Vol. 2 (New York: Springer-Verlag),548.Forbes, D. A., Brodie, J. P., and Grillmair, C. J. 1997, AJ, 113, 1652.Forte, J. C., Geisler, D., Ostrov, P. G., Piatti, A. E., and Gieren, W. 2001, AJ, 121, 1992.Freedman, W. L., Madore, B. F., Fibson, B. K., Ferrarese, L., Kelson, D. D., Sakai, S.,Mould, J. R., Kennicutt, R. C., Ford, H., Graham, J. A.. Huchra, J., Hughes, S. M.G., Illingworth, G. D., Macri, L. M., and Stetson, P. 2001, ApJ, 553, 47.Galleti, S., Bellazzini, M., Federici, L., Buzzoni, A., and Fusi Pecci, F. 2007, A&A, 471, 127.Geisler, D., Lee, M. G., and Kim, E. 1996, AJ, 111, 1529.Huchra, J. P., Stauffer, J., and van Speybroeck, L. 1982, ApJL, 259, 57.Harris, W. E. 1996, AJ, 112, 1487.Harris, W. E. 2001, Saas-Fee Advanced Course 28: Star Clusters 28:223.Harris, W. E. 2009, ApJ, 699, 254.Harris, W. E. 2009, ApJ, 703, 939.Heisler, J., Tremaine, S., and Bahcall, J. N. 1985, ApJ, 298, 8.Hwang, H. S., Lee, M. G., Park, H. S., Kim, S. C., Park, J. H., Sohn, Y. J., Lee, S. G., Rey,S. C., Lee, Y. W., and Kim, H. I. 2008, ApJ, 674, 869.Kundu, A. and Whitmore, B. C. 1998, AJ, 116, 2841.Larsen, S. S., Brodie, J. P., Huchra, J. P., Forbes, D. A., and Grillmair, C. J. 2001, AJ, 121,2974. 16 –Lee, M. G., Kim, E., and Geisler, D. 1998, AJ, 115, 947.Lee, M. G., Hwang, H. S., Kim, S. C., Park, H. S., Geisler, D., Sarajedini, A., and Harris,W. E. 2008, ApJ, 674, 886.Mora, M. D., Larsen, S. S., and Kissler-Patig, M. 2008, A&A, 489, 1065.Nantais, J. B., Huchra, J. P., Barmby, P., and Olsen, K. A. G. 2010, AJ, 139, 1178.Nantais, J. B., Huchra, J. P., McLeod, B., and Strader, J. 2009, AJ, in press, eprintarXiv:astro-ph/0908.4091.Olsen, K. A. G., Miller, B. W., Suntzeff, N. B., Schommer, R. A., and Bright, J. 2004, AJ,127, 2674.Perelmuter, J. M. and Racine, R. 1995, AJ, 109, 1055.Perelmuter, J. M., Brodie, J. P., and Huchra, J. P. 1995, AJ, 110, 620.Perrett, K. M., Bridges, T. J., Hanes, D. A., Irwin, M. J., Brodie, J. P., Carter, D., Huchra,J. P., and Watson, F. G. 2002, AJ, 123, 2490.Rhode, K. L., Zepf, S. E., and Santos, M. R. 2005, ApJ, 630, L21.Richtler, T., Dirsch, B., Gebhardt, K., Geisler, D., Hilker, M., Alonso, M. V., Forte, J. C.,Grebel, E. K., Infante, L., Larsen, S., Minniti, D., and Rejkuba, M. 2004, AJ, 197,2094.Rots, A. H. 1975, A&A, 45, 43.Schiavon, R. P., Rose, J. A., Courteau, S., and MacArthur, L.A. 2005, ApJS, 160, 163.Schr¨oder, L. L., Brodie, J. P., Kissler-Patig, M., Huchra, J. P., and Phillips, A. C. 2002, AJ,123, 2473Strader, J., Brodie, J. P., and Forbes, D. A. 2004, AJ, 127, 295.Strader, J., Brodie, J. P., Cenarro, A. J., and Beasley, M. A. 2005, AJ, 130, 1315.Zinn, R. 1985, ApJ, 283, 424.
This preprint was prepared with the AAS L A TEX macros v5.2.
17 –Fig. 1.— Locations of spectroscopically observed objects. The dotted line represents the µ B = 25 mag arcsec − isophote. 18 –Fig. 2.— Spectra of selected M81 GCs. 19 –Fig. 3.— Velocity histogram of all M81 objects (GCs, H II regions, and open clusters). 20 –Fig. 4.— Spectrum and HST image of Object 1859, an apparent M81 GC mixed with H II emission. 21 –Fig. 5.— Blue indices and MgH vs. Mg2 for M81 GCs. Error bars shown are meanuncertainties in all M81 GC indices. 22 –Fig. 6.— MgB and iron lines vs. Mg2 for M81 GCs. Error bars shown are mean uncertaintiesin all M81 GC indices. 23 –Fig. 7.— Histogram of the metallicities of 103 M81 GCs, along with the Milky Way andM31 for comparison. 24 –Fig. 8.— GC metallicity-galaxy luminosity relations with our current M81 mean metallicity. 25 –Fig. 9.— Locations of MP and MR M81 GCs. The dotted line represents the µ B = 25 magarcsec − isophote. 26 –Fig. 10.— Metallicity of GCs as a function of distance from the center of M81. “Litera-ture” objects are from PBH, BH91, and SBKHP. The metallicity gradient is a least-squaresweighted linear fit to metallicity as a function of galactocentric distance. 27 –Fig. 11.— Number of confirmed GCs in 1 kpc bins as a function of projected radius in M81(108 confirmed GCs), the Milky Way (150 confirmed GCs), and Andromeda (294 confirmedGCs). 28 –Fig. 12.— Velocity vs. position angle for all M81 GCs, shown with the peak and outer edgeH I rotation curves and a fit to the GC rotation. 29 –Fig. 13.— Mean velocity vs. position angle in position angle bins of 30 ◦ . 30 –Fig. 14.— Velocity vs. position angle for GC subsamples at various distances from the centerof M81, with fits to the GC rotation. 31 –Fig. 15.— Velocity vs. position angle for MP (top) and MR (bottom) GC subsamples, withfits to the GC rotation. Outlying object 1352, an inner MP GC marked with an “X” insteadof a triangle, was left out of the MP rotation fit. 32 –Table 1. Observation InformationDate N objects N objexp N skyexp T int a b c d d ID Prev. spec PR95 ID CFT01 ID RA Dec Dist. RV σ RV (hours) (deg) (arcmin) (km s − ) (km s − )34 – 51040 – 09:54:04.910 69:09:18.390 9.47 19 22108 – 50934 – 09:54:25.100 69:08:03.760 7.31 103 28173 PBH 50552 50552 87 09:54:38.810 69:04:10.220 4.84 20 15187 (PBH 50462) 50462 97 09:54:39.700 69:03:26.460 4.78 -11 43264 – 51118 – 09:54:46.470 69:10:53.930 8.08 92 41345 – 50652 – 09:54:50.310 69:05:08.410 4.00 -67 74359 – 50912 – 09:54:51.050 69:07:50.080 5.40 160 27464 (PBH 50388*) 50388 – 09:54:56.140 69:02:31.020 3.58 -98 30505 – – – 09:54:58.470 69:08:08.500 5.20 127 41526 (PBH 51053) 51053 30 09:54:59.790 69:09:27.370 6.25 247 76594 PBH 50861 50861 – 09:55:02.780 69:07:29.380 4.46 -30 58605 – 50376 – 09:55:03.310 69:02:23.730 3.06 -2 69628 – 50659 41 09:55:04.380 69:05:15.820 2.89 -19 27676 – 50873 – 09:55:07.260 69:07:34.220 4.30 193 48705 – 50555 – 09:55:08.420 69:04:11.130 2.22 -73 30720 – 50715 39 09:55:08.980 69:05:51.540 2.89 173 31722 – 50590 – 09:55:09.040 69:04:28.390 2.22 -300 20743 – 50548 – 09:55:09.790 69:04:07.640 2.09 23 8839 – 50358 104 09:55:14.280 69:02:06.340 2.46 -106 27861 – 50259 – 09:55:15.160 69:00:25.800 3.82 -122 23863 – 50678 42 09:55:15.240 69:05:24.180 2.18 128 29876 – 50708 37 09:55:15.610 69:05:47.810 2.44 196 65966 – 50713 38 09:55:19.210 69:05:50.110 2.28 157 28993 – 50697 – 09:55:20.160 69:05:37.620 2.06 -2 391029 – 50777 – 09:55:21.920 69:06:37.630 2.87 131 51089 SBKHP 14 50834 52 09:55:25.190 69:07:14.650 3.39 80 241104 – – 106 09:55:25.700 69:01:39.920 2.34 13 481154 – 50263 – 09:55:29.130 69:00:31.010 3.40 -178 501162 – 50655 – 09:55:29.730 69:05:11.680 1.31 85 111172 – 50351 108 09:55:30.100 69:01:59.540 1.94 -3 551257 – 50786 – 09:55:34.410 69:06:42.240 2.78 35 651265 – 50104 – 09:55:34.920 68:58:14.820 5.64 -130 131300 – 50773 – 09:55:37.140 69:06:35.530 2.68 -28 401301 SBKHP 03 50359 – 09:55:37.280 69:02:07.570 1.82 -111 131308 – 50175 – 09:55:37.770 68:59:17.690 4.61 -194 221309 – 50463 – 09:55:37.800 69:03:27.960 0.61 53 251327 – 50802 – 09:55:38.510 69:06:55.020 3.02 185 281341 – 50693 – 09:55:39.350 69:05:32.620 1.71 162 471350 – 50387 – 09:55:40.010 69:02:29.560 1.55 -368 511352 – – – 09:55:40.020 69:04:10.140 0.66 353 121363 – 50681 – 09:55:40.490 69:05:24.700 1.62 179 441393 BH91 HS26 60012 – 09:55:41.930 68:55:00.740 8.90 -150 171413 – 50512 – 09:55:43.400 69:03:51.640 0.91 -102 161428 – 50583 – 09:55:44.230 69:04:24.130 1.09 101 301456 – 50315 – 09:55:46.040 69:01:25.730 2.73 -255 22
34 –Table 2—Continued
ID Prev. spec PR95 ID CFT01 ID RA Dec Dist. RV σ RV (hours) (deg) (arcmin) (km s − ) (km s − )1490 – 50760 – 09:55:47.680 69:06:25.230 2.80 -41 181495 – 50858 – 09:55:48.000 69:07:27.830 3.77 25 751496 SBKHP 07 50514 – 09:55:48.000 69:03:52.020 1.31 -218 331506 – 50744 – 09:55:48.500 69:06:11.990 2.65 163 411512 – 50674 64 09:55:48.760 69:05:22.260 2.00 -51 281524 SBKHP 02 50304 – 09:55:49.180 69:01:15.350 3.00 50 141537 – 50111 – 09:55:50.260 68:58:22.800 5.72 -215 451563 SBKHP 06 50460 – 09:55:51.320 69:03:23.650 1.69 207 261571 SBKHP 15 50889 7 09:55:51.860 69:07:39.490 4.07 -61 121627 SBKHP 05 50418 – 09:55:54.510 69:02:52.620 2.16 -29 151635 SBKHP 01 50285 – 09:55:54.980 69:00:56.160 3.54 -203 141643 – 50486 – 09:55:55.260 69:03:37.520 1.98 -84 171652 – 50386 – 09:55:56.190 69:02:28.500 2.50 -289 581816 – 50847 5 09:56:03.060 69:07:19.550 4.30 11 191859 – 51046 – 09:56:05.010 69:09:21.290 6.10 114 171946 – 50381 – 09:56:08.690 69:02:24.530 3.50 -199 351951 – 50258 – 09:56:08.820 69:00:23.580 4.73 -77 572081 – 50651 76 09:56:14.140 69:05:05.320 3.82 -235 672087 – 50320 – 09:56:14.320 69:01:29.950 4.38 -168 582163 – 50065 – 09:56:17.510 68:57:12.060 7.77 -159 322170 – 50178 – 09:56:17.770 68:59:18.680 6.07 -3 222171 – – – 09:56:17.780 69:03:04.620 4.06 -134 252196 – 50228 – 09:56:18.950 68:59:55.330 5.70 -93 292219 PBH 50415 50415 – 09:56:20.490 69:02:49.050 4.35 -346 312230 (PBH 50355) 50355 – 09:56:21.090 69:02:01.510 4.66 63 182327 – 50301 – 09:56:27.470 69:01:09.900 5.56 -42 122355 (PBH 50254) 50254 – 09:56:31.730 69:03:54.870 5.20 -282 322381 – 50338 – 09:56:36.890 69:01:46.350 6.06 -102 152490 – 50058 – 09:44:01.840 68:56:52.910 10.57 -113 1530244 lit 30244 – 09:44:54.850 68:49:00.400 19.55 100 7040083 lit 40083 – 09:56:38.470 69:22:50.300 19.67 -87 5240165 lit 40165 – 09:55:03.840 69:15:37.800 11.93 6 3040181 lit 40181 – 09:55:44.090 69:14:11.700 10.27 46 3050037 lit 50037 – 09:55:06.360 68:56:26.000 7.82 -18 6650225 lit 50225 – 09:56:40.590 68:59:52.600 7.23 -7 1350233 lit 50233 – 09:55:55.830 69:00:03.400 4.34 -14 4850286 lit 50286 – 09:54:58.870 69:00:58.200 4.24 -9 6850357 lit 50357 – 09:54:11.220 69:02:06.600 7.51 122 4550394 lit 50394 – 09:54:16.500 69:02:34.400 6.95 -70 3050401 lit 50401 – 09:56:31.750 69:02:38.400 5.36 -283 8750696 lit 50696 – 09:55:02.450 69:05:38.100 3.22 -44 2850785 lit 50785 – 09:54:35.880 69:06:43.100 5.79 197 4850826 lit 50826 – 09:55:52.190 69:07:10.400 3.65 -22 8050886 lit 50886 – 09:55:30.990 69:07:38.600 3.71 134 3950960 lit 50960 – 09:55:52.060 69:08:18.700 4.68 77 43
35 –Table 2—Continued
ID Prev. spec PR95 ID CFT01 ID RA Dec Dist. RV σ RV (hours) (deg) (arcmin) (km s − ) (km s − )51027 lit 51027 – 09:54:20.020 69:09:11.100 8.33 300 9760045 lit 60045 – 09:55:56.860 68:52:13.400 11.83 -28 1970319 lit 70319 – 09:53:42.910 69:13:23.900 13.55 -242 6870349 lit 70349 – 09:53:03.240 69:13:47.400 16.47 -50 3680172 lit 80172 – 09:53:51.690 68:57:04.400 11.34 -51 4090103 lit 90103 – 09:53:39.840 68:48:00.070 18.83 -111 71BH91 HS01 lit 40312 – 09:54:52.010 69:19:45.300 16.16 258 27BH91 HS02 lit 50864 – 09:56:25.740 69:07:27.800 5.84 180 94BH91 HS06 lit 60072 – 09:55:08.080 68:49:56.500 14.09 122 77BH91 R012 lit 60126 – 09:55:26.970 68:44:53.600 18.93 -37 39S04 lit 50378 – 09:55:57.872 69:02:23.070 2.68 -318 31S08 lit 50667 – 09:55:22.295 69:05:19.160 1.70 -144 9S09 lit 50690 – 09:55:21.587 69:05:31.980 1.91 -261 27S10 lit 50738 – 09:55:30.277 69:06:06.120 2.19 22 39S11 lit 50759 – 09:55:35.840 69:06:25.530 2.51 11 24S12 lit 50782 – 09:55:33.042 69:06:39.880 2.73 -200 17S13 lit 50787 1 09:56:05.569 69:06:42.970 4.00 -87 49S16 lit 50889 – 09:55:40.194 69:07:30.820 3.63 -100 25Note. — iterature spectroscopy IDs in parentheses represent objects for which low signal-to-noise spectra were obtained byPBH, but the objects were not confirmed as clusters. PR/PBH object 50388 was originally labeled as a star, but we find it tobe a GC. Object 1859, a GC overlapping with an H II region, was cleaned of emission lines before its velocity was determined,so as to find the velocity of the GC rather than the H II region.
36 –Table 3. Non-GC Objects
ID Obj. Type PR95 ID CFT01 ID RA Dec Dist. RV σ RV (hrs.) (deg.) ( ′ ) (km s − ) (km s − )36 gx – – 09 54 06.64 69 06 20.90 8.09 112001 28143 hii – – 09 54 34.65 69 05 54.56 5.59 85 10159 oc 50618 – 09 54 36.76 69 04 46.96 5.11 76 59164 hii 50774 – 09 54 37.67 69 06 36.90 5.64 103 14185 hii 50649 – 09 54 39.63 69 05 07.22 4.93 71 11215 hii 50485 – 09 54 42.47 69 03 38.25 4.54 32 10269 hii 50574 – 09 54 46.63 69 04 22.32 4.18 33 9283 hii – – 09 54 47.58 69 03 23.27 4.11 13 10287 hii – – 09 54 47.72 69 03 22.52 4.10 5 14301 hii 50729 – 09 54 48.53 69 05 59.46 4.49 114 12329 hii 50840 – 09 54 49.30 69 07 18.78 5.18 139 12357 hii – – 09 54 50.98 69 02 50.89 3.92 -20 10371 hii – – 09 54 51.60 69 06 35.39 4.57 106 12395 hii 50997 – 09 54 52.97 69 08 50.38 6.08 173 13406 hii 50883 – 09 54 53.42 69 07 38.63 5.14 154 13469 hii 50996 – 09 54 56.85 69 08 49.26 5.87 165 11540 hii 50183 – 09 55 00.69 68 59 22.77 5.39 -103 15561 hii 40291 09 55 01.58 69 12 56.19 9.44 131 13611 hii 51117 – 09 55 03.45 69 10 54.07 7.47 168 11629 hii 50659 41 09 55 04.43 69 09 56.14 6.54 139 24668 gx 51043 32 09 55 06.65 69 09 20.11 5.91 81798 51802 gx 50331 – 09 55 12.75 69 01 41.81 2.88 69171 16811 gx – 103 09 55 13.25 69 01 34.41 2.95 61522 23859 oc 50979 09 55 15.02 69 08 31.42 4.88 187 46889 hii 51060 – 09 55 16.17 69 09 35.11 5.87 168 21898 hii 51003 – 09 55 16.52 69 08 55.17 5.22 165 11944 hii 50975 – 09 55 18.40 69 08 28.75 4.75 139 11956 hii – – 09 55 18.92 69 10 05.76 6.31 156 14991 hii 50371 105 09 55 20.04 69 02 18.56 1.99 -74 171011 hii 50480 – 09 55 21.17 69 03 36.32 1.12 -20 171027 hii – 54 09 55 21.89 69 07 36.52 3.83 139 141036 gx 50589 – 09 55 22.27 69 04 26.39 1.10 99145 181121 hii 50577 – 09 55 26.87 69 04 22.15 0.72 213 181227 hii – – 09 55 32.76 69 11 55.63 8.01 101 121252 hii 50980 – 09 55 34.21 69 08 31.50 4.61 117 111426 hii 50845 – 09 55 44.17 69 07 19.27 3.54 64 131470 hii 50711 – 09 55 46.85 69 05 48.86 2.26 40 241471 gx – – 09 55 46.86 68 55 43.02 8.29 124246 681473 hii 50179 – 09 55 46.88 68 59 20.18 4.74 -223 121479 gx – – 09 55 47.20 68 53 56.79 10.05 82813 711497 gx 50775 – 09 55 48.09 69 06 35.77 2.99 100229 791515 hii 50113 – 09 55 48.89 68 58 27.49 5.64 -235 121593 hii 50152 – 09 55 52.98 68 59 04.18 5.16 -234 141673 hii 50633 – 09 55 57.70 69 04 55.54 2.41 -36 131685 hii 51008 – 09 55 58.05 69 08 56.62 5.49 23 11
37 –Table 3—Continued
ID Obj. Type PR95 ID CFT01 ID RA Dec Dist. RV σ RV (hrs.) (deg.) ( ′ ) (km s − ) (km s − )1704 hii – – 09 55 58.69 69 09 09.36 5.71 20 101734 gx – – 09 56 00.12 68 57 13.25 7.12 56775 211736 gx – – 09 56 00.13 69 05 33.40 2.91 100396 621743 hii 50531 – 09 56 00.60 69 04 00.31 2.45 -70 101747 gx – – 09 56 00.75 68 56 18.14 8.01 124037 691757 hii – – 09 56 01.03 68 59 02.00 5.49 -233 131772 hii 50098 – 09 56 01.58 68 58 03.84 6.39 -257 161830 hii – – 09 56 03.58 69 08 24.64 5.25 2 251839 hii 51067 – 09 56 04.01 69 09 39.84 6.37 29 631843 gx – – 09 56 04.17 68 56 13.83 8.18 123902 421889 hii 50200 – 09 56 06.19 68 59 34.23 5.26 -228 141922 oc 50430 – 09 56 08.11 69 03 05.31 3.23 -81 461934 hii – 75 09 56 08.28 69 05 17.98 3.42 -23 101970 gx 50962 – 09 56 09.56 69 08 19.47 5.47 32332 171976 hii 50319 – 09 56 09.80 69 01 29.73 4.08 -186 132002 oc 50584 – 09 56 10.66 69 04 23.58 3.38 63 692114 oc 50377 – 09 56 15.33 69 02 22.19 4.08 -149 672146 hii – – 09 56 16.70 69 05 39.23 4.25 -72 112240 oc 50822 – 09 56 21.32 69 07 07.83 5.36 24 852288 hii 50176 – 09 56 24.38 68 59 15.82 6.54 -228 122290 oc 50866 – 09 56 24.49 69 07 28.86 5.80 8 202412 gx 50170 – 09 56 42.89 68 59 11.67 7.83 56450 362462 gx – – 09 56 53.48 69 03 11.55 7.21 128234 762468 gx – – 09 56 55.16 68 58 57.20 8.87 60258 152471 gx – – 09 56 55.37 68 57 59.10 9.47 36073 282479 gx – – 09 56 57.53 68 58 10.81 9.50 127370 562484 gx – – 09 56 59.07 69 01 28.74 8.07 127833 982499 gx 50107 – 09 57 04.67 68 58 13.24 9.99 83038 652500 gx 20246 – 09 57 04.72 69 01 31.67 8.53 63144 102510 gx 50238 – 09 57 09.62 69 00 02.11 9.47 26763 182518 gx 50120 – 09 57 13.18 68 58 28.51 10.49 104981 209– hii – – 09 54 40.90 69 05 01.85 4.80 44 14
38 –Table 4. Brodie and Huchra Indices
ID CNR/CN1 G4300 H β MgH/Mg1 Mg2 Mgb Fe5270 NaD/Na1 CNB H&K MgG δ (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag)34 0.510 0.534 0.014 0.077 0.237 0.174 0.102 0.129 nan 1.287 -0.424 2.954 σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
39 –Table 4—Continued
ID CNR/CN1 G4300 H β MgH/Mg1 Mg2 Mgb Fe5270 NaD/Na1 CNB H&K MgG δ (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
40 –Table 4—Continued
ID CNR/CN1 G4300 H β MgH/Mg1 Mg2 Mgb Fe5270 NaD/Na1 CNB H&K MgG δ (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag)1490 -0.067 0.131 0.078 0.012 0.027 0.051 0.025 0.055 0.007 0.259 -0.228 0.374 σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
41 –Table 4—Continued
ID CNR/CN1 G4300 H β MgH/Mg1 Mg2 Mgb Fe5270 NaD/Na1 CNB H&K MgG δ (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) σ σ σ σ σ σ σ
42 –Table 5. Pseudo-Lick Indices
ID CN2 Ca42 Fe43 Ca44 Fe45 C
46 Fe50 Fe53 Fe54 Fe5709 Fe5782 TiO1 TiO2(mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag)34 0.485 0.577 0.124 0.203 -0.013 0.036 0.082 0.074 0.136 0.081 0.075 0.060 0.061 σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
43 –Table 5—Continued
ID CN2 Ca42 Fe43 Ca44 Fe45 C
46 Fe50 Fe53 Fe54 Fe5709 Fe5782 TiO1 TiO2(mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
44 –Table 5—Continued
ID CN2 Ca42 Fe43 Ca44 Fe45 C
46 Fe50 Fe53 Fe54 Fe5709 Fe5782 TiO1 TiO2(mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag)1490 -0.020 0.089 0.041 0.042 0.048 0.011 0.027 0.012 0.043 -0.013 -0.010 -0.018 0.025 σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ σ
45 –Table 5—Continued
ID CN2 Ca42 Fe43 Ca44 Fe45 C
46 Fe50 Fe53 Fe54 Fe5709 Fe5782 TiO1 TiO2(mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) σ σ σ σ σ σ σ
46 –Table 6. Metallicities of GCs
ID [Fe/H] σ [ F e/H ] ID [Fe/H] σ [ F e/H ] ID [Fe/H] σ [ F e/H ] (dex) (dex) (dex) (dex) (dex) (dex)34 1.1 2.39 1327 -0.95 0.76 2381 -0.3 0.44108 -0.72 0.66 1341 -1.49 0.52 2490 -0.92 0.37173 0.08 0.41 1350 -0.88 0.5 PBH30244 -1.76 1.78187 -1.51 0.36 1352 -1.77 0.3 PBH40083 -1.29 0.8264 -1.63 0.36 1363 -1.38 0.53 PBH40165 -1.57 0.43345 -2.1 0.61 1393 -0.12 0.27 PBH40181 0.64 1.43359 -0.43 0.83 1413 -0.35 0.74 PBH50037 -2.34 0.83464 -0.64 0.58 1428 -1.08 0.42 PBH50225 -0.04 0.59505 -0.44 0.56 1456 -0.65 0.62 PBH50233 -1.75 1.02526 -1.04 2.02 1490 -1.42 0.32 PBH50286 -0.04 1.85594 -0.29 0.76 1495 -1.31 1.63 PBH50357 -3.62 2.97605 -1.69 0.62 1496 -0.4 0.93 PBH50394 -1.5 1.29628 -1.17 0.52 1506 -0.22 1.82 PBH50401 -0.04 1676 1.15 2.96 1512 -1.69 0.24 PBH50696 -1.86 0.5705 -0.81 0.21 1524 -1.08 0.44 PBH50785 -0.72 1.17720 -0.89 0.64 1537 -1.85 0.36 PBH50826 -1.46 1.11722 -1.44 0.35 1563 -1.5 0.44 PBH50886 -1.79 0.87743 -0.81 0.44 1571 -0.57 0.53 PBH50960 -1.79 0.64839 -1.48 0.32 1627 -0.9 0.36 PBH51027 -2.47 1.01861 -0.57 0.46 1635 -0.91 0.54 PBH60045 -1.03 0.97863 -0.79 0.55 1643 0.12 0.56 PBH70319 -2.31 1.69876 -0.54 0.8 1652 -1.81 0.44 PBH70349 -2.41 1.15966 -0.37 0.35 1816 -0.37 0.53 PBH80172 -0.77 0.68993 -0.99 0.44 1859 -0.61 0.33 PBH90103 -2.23 0.991029 -0.86 0.41 1946 -1.1 0.44 BH91 HS01 -2.1 0.971089 -1.37 0.41 1951 -1.94 1.11 BH91 HS02 -0.92 0.41104 -1.26 1.44 2081 -1.26 0.97 BH91 HS06 -1.77 0.831154 -0.05 0.89 2087 -1.63 0.7 BH91 R012 0.19 1.441162 -0.82 0.45 2163 -1.22 0.77 SBKHP04 -0.41 0.091172 -1.06 0.9 2170 -1.08 0.42 SBKHP08 -0.7 0.061257 -1.97 0.56 2171 -1.33 0.39 SBKHP09 -1.21 0.131265 -0.8 0.43 2196 -1.79 0.53 SBKHP10 -1.32 0.361300 -1.86 0.49 2219 -1.64 0.43 SBKHP11 -1.11 0.411301 -0.15 0.5 2230 -1.46 0.35 SBKHP12 – –1308 -0.4 0.53 2327 -1.53 0.29 SBKHP13 -1.06 0.061309 -1.41 0.34 2355 -0.26 0.51 SBKHP16 -0.67 0.04
47 –Table 7. Linear Fits to [
F e/H ] = a ( index ) + b Index ID a b R R I σ m σ s (dex/mag) (dex) δ c V r V r , proj. PA σ V , uncorr. σ V , rot. corr.(km s − ) (km s − ) (km s − ) (deg) (km s − ) (km s − )All 108 -22 ±
12 108 ±
22 93 ±
19 261 ±
11 145 128 < ±
21 133 ±
40 114 ±
34 294 ±
14 155 1414-8 kpc (incl.) 37 -17 ±
17 87 ±
24 75 ±
21 299 ±
19 137 133 > ±
22 101 ±
33 87 ±
28 278 ±
34 133 132MR 53 -33 ±
13 122 ±
18 104 ±
15 290 ±
13 148 125MP 53* -36 ±
19 67 ±
38 57 ±
33 323 ±
21 142 141Note. — MR GCs are those with [Fe/H] ≥ − .
06, and MP GCs are those with [Fe/H] < − .
06. MP GC rotation was calculated by excluding a bright outlier, Object 1352, a MPinner GC. Including this object gave a result inconsistent with the trend outlined by theother MP clusters. 48 –Table 9. Projected Mass of M81N R max M proj (kpc) (10 M ⊙ )54 3.82 0 . ± . . ± . . ± . . ± ..