1: %% fIRST DRAFT Feb. 28, 2005
2:
3: %%\documentclass{aastex}
4:
5: %%\documentstyle[12pt,aasms4]{article}
6: \documentclass[12pt,preprint]{aastex}
7: %% manuscript produces a one-column, double-spaced document:
8: %%\documentclass[manuscript]{aastex}
9: %%\documentclass[10pt, preprint]{aastex}
10: %% preprint2 produces a double-column, single-spaced document:
11:
12: %%\documentclass[10pt, preprint2]{aastex}
13: %%\documentclass{emulateapj}
14: %%\usepackage{lscape}
15: %%\usepackage{epsfig}
16: %%\usepackage{apjfonts}
17: \usepackage{graphics}
18: %%\usepackage{floatfig}
19: %%\usepackage{wrapfig}
20:
21: %% Sometimes a paper's abstract is too long to fit on the
22: %% title page in preprint2 mode. When that is the case,
23: %% use the longabstract style option.
24: %% \documentclass[preprint2,longabstract]{aastex}
25:
26: \shorttitle {The SEGUE Stellar Parameter Pipeline. II. Validation
27: from Galactic Globular and Open Cluster Observations}
28:
29: \shortauthors{Lee et al.}
30:
31: \begin{document}
32:
33: \title{The SEGUE Stellar Parameter Pipeline. II. Validation with Galactic Globular and Open Clusters}
34:
35: \author{Young Sun Lee, Timothy C. Beers, Thirupathi Sivarani}
36: \affil{Department of Physics \& Astronomy, CSCE: Center for the
37: Study of Cosmic Evolution, and JINA: Joint Institute for Nuclear
38: Astrophysics, Michigan State University, East Lansing, MI 48824,
39: USA} \email{lee@pa.msu.edu, beers@pa.msu.edu, thirupathi@pa.msu.edu}
40:
41: \author{Jennifer A. Johnson, Deokkeun An}
42: \affil{Department of Astronomy,\\ Ohio State University,
43: Columbus, OH 43210}
44: \email{jaj@astronomy.ohio-state.edu,
45: deokkeun@astronomy.ohio-state.edu}
46:
47: \author{Ronald Wilhelm}
48: \affil{Department of Physics,\\ Texas Tech University, Lubbock, TX
49: 79409} \email{ron.wilhelm@ttu.edu}
50:
51: \author{Carlos Allende Prieto, Lars Koesterke}
52: \affil{Department of Astronomy, \\ University of Texas, Austin, TX
53: 78712} \email{callende@astro.as.utexas.edu}
54:
55: \author{Paola Re Fiorentin}
56: \affil{Max Planck Institut f$\ddot{u}$r Astronomie, \\
57: K$\ddot{o}$nigstuhl 17, 69117 Heidelberg, Germany}
58: \email{fiorent@mpia-hd.mpg.de}
59:
60: \author{Coryn A.L. Bailer-Jones}
61: \affil{Max Planck Institut f$\ddot{u}$r Astronomie, \\
62: K$\ddot{o}$nigstuhl 17, 69117 Heidelberg, Germany}
63: \email{calj@mpia-hd.mpg.de}
64:
65: \author{John E. Norris}
66: \affil{Research School of Astronomy and Astrophysics, \\
67: Australian National University, Weston, ACT 2611, Australia}
68: \email{jen@mso.anu.edu.au}
69:
70: \author{Brian Yanny}
71: \affil{Fermi National Accelerator Laboratory,\\
72: Batavia, IL 60510}
73: \email{yanny@fnal.gov}
74:
75: \author{Constance Rockosi}
76: \affil{Department of Astronomy,\\
77: University of California, Santa Cruz, CA 95064}
78: \email{crockosi@ucolick.org}
79:
80: \author{Heidi J. Newberg}
81: \affil{Department of Physics $\&$ Astronomy,\\
82: Rensselaer Polytechnical Institute, Troy, NY 12180}
83: \email{newbeh@rpi.edu}
84:
85: \author{Kyle M. Cudworth}
86: \affil{Yerkes Observatory, \\
87: The University of Chicago, Williams
88: Bay, WI 53191} \email{kmc@yerkes.uchicago.edu}
89:
90: \author{Kaike Pan}
91: \affil{Apache Point Observatory, \\
92: Apache Point Observatory, P. O. Box 59, Sunspot, NM 88349}
93: \email{kpan@apo.nmsu.edu}
94:
95: \author{ }
96: \affil{}
97:
98: \clearpage
99:
100: \begin{abstract}
101:
102: We validate the performance and accuracy of the current SEGUE (Sloan Extension
103: for Galactic Understanding and Exploration) Stellar Parameter
104: Pipeline (SSPP), which determines stellar atmospheric parameters (effective
105: temperature, surface gravity, and metallicity) by comparing derived overall
106: metallicities and radial velocities from selected likely members of three
107: globular clusters (M~13, M~15, and M~2) and two open clusters (NGC~2420 and
108: M~67) to the literature values. Spectroscopic and photometric data obtained
109: during the course of the original Sloan Digital Sky Survey (SDSS-I) and its
110: first extension (SDSS-II/SEGUE) are used to determine stellar radial velocities
111: and atmospheric parameter estimates for stars in these clusters. Based on the
112: scatter in the metallicities derived for the members of each cluster, we
113: quantify the typical uncertainty of the SSPP values, $\sigma (\rm [Fe/H])$ =
114: 0.13 dex for stars in the range of 4500 K $\le T_{\rm eff} \le 7500$ K and $2.0
115: \le \log g \le 5.0$, at least over the metallicity interval spanned
116: by the clusters studied ($-2.3 \le {\rm [Fe/H]} < 0 $). The
117: surface gravities and effective temperatures derived by the SSPP are
118: also compared with those estimated from the comparison of the
119: color-magnitude diagrams with stellar evolution models; we find
120: satisfactory agreement. At present, the SSPP underestimates
121: [Fe/H] for near-solar-metallicity stars, represented by members of
122: M~67 in this study, by $\sim$ 0.3 dex.
123:
124: \end{abstract}
125:
126: \keywords{methods: data analysis --- stars: abundances, fundamental parameters
127: --- surveys --- techniques: spectroscopic }
128:
129: \section{Introduction}
130:
131: The Sloan Extension for Galactic Understanding and Exploration (SEGUE) is one of
132: three key projects (LEGACY, SUPERNOVA SURVEY, and SEGUE) in the current
133: extension of the Sloan Digital Sky Survey, known collectively as SDSS-II. The
134: SEGUE program is in the process of obtaining $ugriz$ imaging of some 3500 square
135: degrees of sky outside of the SDSS-I footprint (Fukugita et al. 1996; Gunn et
136: al. 1998, 2006; York et al. 2000; Stoughton et al. 2002; Abazajian et al. 2003,
137: 2004, 2005; Pier et al. 2003), with special attention being given to scans of
138: lower Galactic latitudes ($|b|$ $<$ 35$^{\circ}$) in order to better probe the
139: disk/halo interface of the Milky Way. SEGUE is also obtaining $R$ $\simeq$ 2000
140: spectroscopy over the wavelength range 3800 $-$ 9200\,{\AA} for some 250,000
141: stars in 200 selected areas over the sky available from Apache Point, New
142: Mexico.
143:
144: The SEGUE Stellar Parameter Pipeline (hereafter, SSPP) processes the wavelength-
145: and flux-calibrated spectra generated by the standard SDSS spectroscopic
146: reduction pipeline (Stoughton et al. 2002), obtains equivalent widths and/or
147: line indices for 77 atomic or molecular absorption lines, and estimates $T_{\rm
148: eff}$, log $g$, and [Fe/H] through the application of a number of approaches.
149: The current techniques employed by the SSPP include a minimum distance method
150: (Allende Prieto et al. 2006), neural network analysis (Bailer-Jones 2000;
151: Willemsen et al. 2005; Re Fiorentin et al. 2007), auto-correlation analysis
152: (Beers et al. 1999), and a variety of line index calculations based on previous
153: calibrations with respect to known standard stars (Beers et al. 1999; Cenarro et
154: al. 2001a,b; Morrison et al. 2003). The SSPP employs five different methods for
155: estimation of $T_{\rm eff}$, eight for estimation of log $g$, and nine for
156: estimation of [Fe/H]. Details of the methods used are discussed in detail by Lee
157: et al. (2007a, hereafter Paper I). The use of multiple methods allows for
158: empirical determinations of the internal errors for each parameter, based on the
159: range of reported values -- typical internal errors for stars in the temperature
160: range $4500$~K $\le$ $T_{\rm eff}$ $\le$ 7500~K are $\sim$ 73~K, $\sim$ 0.19
161: dex, and $\sim$ 0.10 dex, in $T_{\rm eff}$, log $g$, and [Fe/H], respectively.
162: Allende Prieto et al. (2007, hereafter Paper III) point out that the internal
163: uncertainties provided by the SSPP underestimate the typical random errors at
164: high signal-to-noise ($S/N$) ratios because most methods in the SSPP make use of
165: similar parameter indicators (e.g., hydrogen lines for effective temperature)
166: and similar atmospheric models. Paper III empirically determines empirically external
167: uncertainties of $\sim$ 130~K, $\sim$ 0.21 dex, and $\sim$ 0.11 dex, for $T_{\rm
168: eff}$, log $g$, and [Fe/H], respectively, by comparison with high-resolution
169: spectroscopy ($7000 < R < 45,000$) of brighter SDSS-I/SEGUE that have been
170: obtained with 8m$-$10m class telescopes. Somewhat larger errors apply to stars
171: with temperatures near the extremes of the range above. The present study of
172: Galactic globular and open cluster stars tests the SSPP's ability to derive accurate
173: results for stars with a wide range of temperatures and gravities appropriate
174: for metal-poor and near-solar-metallicity stellar populations in the Galaxy, and
175: demonstrates that the derived metallicity scale is identical for dwarfs and
176: giants.
177:
178: Although the SSPP will continue to evolve in the near future, it has been frozen
179: for now at the version used for obtaining results for stars with suitable data
180: from SDSS Data Release 6 (DR-6; Adelman-McCarthy et al. 2007b). Previous
181: versions of the SSPP have already been used for the analysis of SDSS-I
182: observations. For example, Allende Prieto et al. (2006) report on the
183: application of one of the methods included in the SSPP to some 20,000 F- and
184: G-type stars from SDSS-I DR-3 (Abazajian et al. 2005). Beers et al. (2006) have
185: compiled a list of over 6000 stars with [Fe/H] $< -2.0$ (including several
186: hundred with [Fe/H] $< -3.0$), based on application of the present SSPP to some
187: 200,000 stars from SDSS-I DR-5 (Adelman-McCarthy et al. 2007a). Carollo et al.
188: (2007) reports on an analysis of the kinematics of relatively bright stars from
189: SDSS-I that have been used as calibration objects during the main survey.
190:
191: In this paper, the second in the SSPP series, we show that estimates of the
192: atmospheric parameters and radial velocities obtained by the SSPP for
193: stars with a reasonable likelihood of membership in previously studied Galactic
194: globular and open clusters are sufficiently accurate to justify the use of the
195: present SSPP parameters for carrying out detailed studies of the halo and thick-disk
196: populations of the Milky Way. In deriving the overall iron abundance for each
197: cluster, we assume it comprises a chemically homogenous population.
198: %This assumption is valid since as Cohen \& Mel$\acute{\rm e}$ndez (2005) showed there
199: %has not been detected dispersion larger than the observational uncertainties in
200: %the heavy elements between Ca and the iron peak. Most of large chemical
201: %inhomogeneity occurs to the light elements such as C, N, O, Na, etc. and we are
202: %not comparing the abundance of these elements.
203:
204: In \S 2, the photometric and spectroscopic data obtained for M~13, M~15, M~2,
205: NGC~2402, and M~67 are described. Section 3 presents the methods used to
206: separate likely cluster members from field stars in the directions toward these
207: clusters. Best estimates of the overall [Fe/H] and radial velocity of each
208: cluster are derived in \S 4. In \S 5 we compare the SSPP determinations of
209: $T_{\rm eff}$ and log $g$ for selected member stars in each cluster with their
210: expected positions on color-magnitude diagrams. A summary and brief conclusions
211: are provided in \S 6.
212:
213:
214: \section{Photometric and Spectroscopic Data}
215:
216: Galactic globular and open clusters are nearly ideal testbeds for validation of
217: the stellar atmospheric parameters estimated by the SSPP. In most clusters, it
218: is expected that their member stars were born simultaneously out of well-mixed,
219: uniform-abundance gas at the same location in the Galaxy. Therefore, with the
220: exception of effects due to post main-sequence evolution, primordial variations
221: in carbon and nitrogen, or contamination from binary companions that have
222: transferred material, the member stars should exhibit very similar elemental
223: abundance patterns. Three of the clusters in our study, M~13, M~15, and M~2,
224: have well-known CN variations that extend to the main-sequence turnoffs (Smith
225: \& Briley 2006 for M13; Cohen, Briley, \& Stetson 2005 for M15; Smith \& Mateo
226: 1990 for M2). However, these abundance variations can be ignored when deriving
227: metallicities from regions of the spectra that do not include CH, CN, or NH
228: features, as is the case with most of our techniques (those that may be affected
229: by the presence of such features are automatically de-selected in the
230: determination of the adopted [Fe/H]).
231:
232: True cluster members should exhibit small radial velocity differences with
233: respect to their parent clusters. Furthermore, it is possible to examine
234: theoretical predictions of temperatures and surface gravities for member stars
235: that lie along the cluster main sequence (MS), red giant branch (RGB), or
236: horizontal branch (HB) in color-magnitude diagrams (CMDs). As part of tests of
237: the SEGUE star-selection algorithm (Adelman-McCarthy et al. 2007b) and the SSPP,
238: and during normal SEGUE operation, we have obtained $ugriz$ photometry and
239: medium-resolution (2.3 {\AA}; $R$ = 2000) spectroscopy for large numbers of
240: stars along lines of sight toward the globular clusters M~13, M~15, and M~2 and
241: the open clusters NGC~2420 and M~67. Below we discuss these photometric and
242: spectroscopic data in more detail.
243:
244: \subsection{Photometric Data}
245:
246: The SDSS obtains scans of the sky using the ARC 2.5m telescope on Apache Point,
247: New Mexico. These data are collected in five broad bands ($u, g, r, i, z$) with
248: central wavelengths 3551, 4686, 6166, 7480, and 8932\,{\AA} (Fukugita et al.
249: 1996), respectively, using an imaging array of 30 ($6 \times 5$) 2048 $\times$
250: 2048 Tektronix CCDs (Gunn et al. 1998). The pixel size is 24 $\mu$m,
251: corresponding to $0.396{''}$ on the sky. A series of software procedures,
252: collectively known as the SDSS PHOTO pipeline (Lupton et al. 2001), processes
253: and reduces the scanned images shortly after data are obtained. As part of these
254: procedures, the instrumental fluxes and astrometric positions (Pier et al.
255: 2003), as well as a determination of whether an object is likely to be stellar
256: (i.e., a \emph{point source}), or not (an \emph{extended} source) are obtained.
257: Afterwards, the photometric data are further calibrated by matching to brighter
258: known standards observed with a smaller calibration telescope on Apache Point
259: (Hogg et al. 2001; Smith et al. 2002; Tucker et al. 2006). The processed
260: photometric data have been shown to exhibit 2\% relative and absolute errors
261: (0.02 magnitudes) in $g$, $r$, and $i$, and $3 \%-5 \%$ errors in $u$ and $z$
262: for all stellar objects brighter than $g = 20$ (Stoughton et al. 2002; Abazajian
263: et al. 2004, 2005; Ivezic et al. 2004). The first-pass photometric data for each
264: of the clusters used in the present study were secured by querying the DR-3
265: (Abazanjian et al. 2005), DR-5 (Adelman-McCarthy et al. 2007a), and DR-6
266: (Adelman-McCarthy et al. 2007b) releases from the SDSS Catalog Archive Server
267: (CAS).
268:
269: Figure 1 illustrates one of the primary challenges in working with data for
270: clusters obtained with SDSS -- the automated PHOTO pipeline (Lupton et al. 2001)
271: was not designed to adequately deal with crowded fields such as the central
272: regions of globular clusters. As a result, essentially all of the stars in this
273: region (which are by definition the most likely ones to be cluster members) do
274: not have reported apparent magnitudes in the SDSS CAS. To circumvent this
275: limitation as much as possible, we have instead performed crowded field
276: photometry for the center of the clusters, using the DAOPHOT/ALLFRAME suite of
277: programs (Stetson 1987; Stetson 1994) in IRAF\footnote{IRAF is distributed by
278: the National Optical Astronomy Observatories, which is operated by the
279: Association of Universities for Research in Astronomy, Inc. under cooperative
280: agreement with the National Science Foundation.}. A full description of the
281: methods used and the photometric measures obtained is provided by Johnson et al.
282: (2007). Briefly, DAOPHOT was run on each image, and the five images of each
283: field (one for each filter) were then simultaneously run through ALLFRAME.
284: DAOGROW (Stetson 1990) was used to derive aperture corrections to the
285: point-spread-function photometry for the SDSS aperture radius of 7.4 arcsecs.
286: Finally, the zeropoint term from the {\it tsField} files was applied to calibrate the
287: data. This procedure also permits a check on the techniques used by the SDSS
288: PHOTO pipeline in regions outside the cluster where the areal density of sources
289: on the sky is sufficiently low that it may be used.
290:
291: After completing the above procedures, we finally combine the results from the
292: PHOTO pipeline with those from the crowded-field photometry to obtain an almost
293: complete catalog of $ugriz$ photometry for stars in the region of each of our
294: program clusters. All photometric data are corrected for extinction and
295: reddening by application of the Schlegel, Finkbeiner, $\&$ Davis (1998) maps.
296: The average reddening ($E(B-V)$) for stars in the direction of these clusters is
297: 0.017, 0.110, 0.045, 0.041, 0.032 for M~13, M~15, M~2, NGC~2420, and M~67,
298: respectively. Comparing with the literature values listed in Table 1, most of
299: the average reddenings of the clusters agree within about 0.02 mags.
300:
301: \subsection{Spectroscopic Data}
302:
303: The spectroscopy discussed in the present paper was obtained during the course
304: of SEGUE tests and normal SEGUE observations. In normal SEGUE operation mode, a pair of
305: plug-plates (referred to as the ``bright'' and ``faint'' plates) are obtained
306: over the $3^{\circ}$ field of the ARC 2.5m. A total of 640 optical fibers are
307: employed to obtain $R=2000$ spectra for on the order of 600 program
308: stars for each plate (the remaining fibers are used for spectrophotometric and
309: reddening calibration objects and observations of the night sky). The exposure
310: time depends on observation conditions. For a bright plate, exposures are set to
311: achieve a total $(S/N)^{2} >$ 15/1 from the two blue-side CCDs on the SDSS
312: spectrographs; the exposure for a faint plate is set such that a total $(S/N)
313: ^{2} >$ 50/1 for all four (red and blue CCDs) on the SDSS spectrographs is
314: achieved. In order to identify and remove cosmic ray hits, each plate must have
315: at least three exposures; the integration time for any single exposure is not
316: longer than 30 minutes. For the purposes of targeting objects on these plates,
317: the boundary between the bright and faint plates is set at $r \sim$ 18.0. The
318: data thus obtained are processed through the SDSS spectroscopic pipeline
319: software (SPECTRO2D and SPECTRO1D), which produces wavelength and
320: flux-calibrated spectra, and also obtains estimates of radial velocities and
321: line indices (Stoughton et al. 2002). Tests of the quality of stellar radial
322: velocities from the SSPP (which uses initial estimates from the SDSS processing
323: pipelines) indicate precisions better than 5 km~s$^{-1}$ are achieved for
324: brighter stars, with zero-point offsets of no more than a few km~s$^{-1}$,
325: respectively (Paper III). These errors degrade for fainter stars, as expected.
326:
327: An initial set of candidate member stars of the globular and open clusters
328: studied in the present paper were selected on the basis of photometric and
329: astrometric data (proper motions) from the literature. The central cores of the
330: clusters were not targeted because the PHOTO pipeline does not resolve the very
331: crowded fields into single star detections, and also due to limitations on the
332: separations of the fibers during the spectroscopic follow-up stage. The primary
333: method for selecting member candidates was performed by plotting a photometric
334: CMD for a given cluster, and choosing stars from regions of this diagram that
335: correspond to location on the MS turnoff or RGB of the cluster. An additional
336: list of bright stars for M~15 and M~2 with previously available proper motions
337: consistent with membership in the clusters was provided by Cudworth (1976 and
338: private communication) and Cudworth $\&$ Rauscher (1987). Other stars in the
339: fields of these clusters were used to fill spectroscopic fibers using the
340: default SEGUE target selection algorithm (Adelman-McCarthy et al. 2007b). While
341: many of these additional targets turned out to be stars from the general field
342: populations, a significant fraction turned out serendipitously to be members of
343: the clusters.
344:
345: For M~13, three specially designed plates were obtained. Two of the three plates
346: followed the standard SEGUE target selection procedure (Adelman-McCarthy et al.
347: 2007b) of sampling stars with a variety of spectral types based on the SDSS
348: imaging and PHOTO processing. An additional set of likely M~13 members,
349: including several stars that were saturated in the SDSS image ($r < 14.5$) and
350: with coordinates from Cudworth $\&$ Monet (1979) and Cudworth (private
351: communication), were added to the target list with high priority (bumping
352: ordinary SEGUE targets), in order to obtain spectra of several likely giant-branch and
353: horizontal-branch members.
354:
355: In the case of NGC~2420, the stars chosen for spectroscopy were primarily
356: targeted from the SDSS photometry obtained by the PHOTO pipeline, using the
357: normal SEGUE target selection algorithm. Additional stars with apparent
358: magnitudes in the range 14.5 $< g <$ 20.5 that fell within 0.5 degrees
359: from the center of NGC~2420 were also targeted for spectroscopy. However, due to
360: crowding, if two objects were within 55${''}$ of one another, then only one
361: received a fiber. Thus, not every star in the central region of NGC~2420 was
362: targeted. There were about 480 objects selected in this way, including a number
363: of non-cluster members that are located in the NGC~2420 field.
364:
365: For M~67, the initial targets came from the SDSS imaging data processed by the
366: PHOTO pipeline. However, for this cluster, many candidate members with
367: positions, magnitudes, and colors from the WEBDA (http:
368: //www.univie.ac.at/webda/) catalogs were added to the target lists. The bright
369: targets (with $r < 14$) saturate the SDSS imaging camera, so these were added
370: from the literature (Sanders 1989; Fan et al. 1996). Such bright stars normally
371: saturate a regular SDSS spectroscopic exposure, so there were exposed for
372: shorter than normal. About 200 very bright stars between about 12 $<$ $g$ $<$ 14
373: were targeted.
374:
375: In total, we obtained SDSS spectroscopy for 1920, 1280, 640, 1280, and 640
376: targets, including sky spectra and calibration object spectra, in the fields of
377: M~13, M~15, M~2, NGC~2420, and M~67 respectively. The reduced spectra were then
378: processed through the SSPP in order to estimate $T_{\rm eff}$, log $g$, and
379: [Fe/H], among other quantities. Table 1 summarizes the global properties of the
380: clusters under consideration in this paper, taken from the compilation of Harris
381: (1996) for the globular clusters and from WEDBA or Gratton (2000) for the open
382: clusters.
383:
384: \subsection{Radial Velocities}
385:
386: There are two estimated radial velocities provided from the SDSS spectroscopic
387: pipeline. One is an absorption redshift obtained by cross-correlating the
388: spectra with templates that were obtained from SDSS commissioning spectra
389: (Stoughton et al. 2002). Another comes from matching the spectra with ELODIE
390: template spectra (Prugniel $\&$ Soubiran 2001). In most cases the velocity based
391: on the ELODIE template matches appears to be the best available estimate, as
392: spectra of ``quality assurance'' stars with multiple measurements show the most
393: repeatable values for this estimator. However, this is not always the case. We
394: proceed to select the best available velocity in the following manner. If the
395: velocity determined by comparison with the ELODIE templates has a reported error
396: of 20 km s$^{-1}$ or less then this velocity is adopted. If the error from the
397: ELODIE template comparison is larger than 20 km s$^{-1}$, and the relative
398: offset between the two radial velocities is less than 40 km s$^{-1}$, we take an
399: average of the two. If none of the criteria above are satisfied, which happens
400: only rarely, and mainly for quite low $S/N$ spectra, or for hot/cool stars
401: without adequate templates, we obtain the calculated radial velocity from a
402: custom routine that examines the wavelengths of a number of prominent absorption
403: features. If none of these methods yield a reasonable estimate of radial
404: velocity, or it appears spurious (i.e., falls outside of the range $\pm 1000$ km
405: s$^{-1}$), we simply ignore the star in subsequent analyses. A more detailed
406: description of the procedures used for determination of the best available
407: radial velocity, and of zero-point offsets of the radial velocities, can be found
408: in Paper I.
409:
410: \section{Membership Selection from the Spectroscopic Samples}
411:
412: Owing to an insufficient number of stars with available spectroscopy for each
413: cluster, it is not possible to obtain a well-defined Color Magnitude Diagram
414: (CMD) based solely on spectroscopically confirmed member stars. Thus, we make
415: use of photometric data in the field of each cluster, and describe below how we
416: obtain a relatively clean CMD for individual clusters, and select likely member
417: stars from the spectroscopic data.
418:
419: \subsection{Likely Member Star Selection for Globular Clusters}
420:
421: One of the primary issues that one needs to address when creating a CMD, or
422: selecting likely member stars, for a star cluster is removal of contamination
423: from field stars. In order to approximately isolate the likely cluster members
424: from the field stars we have made use of the CMD mask algorithm described by
425: Grillmair et al. (1995). We illustrate the basic idea by application of this
426: algorithm to the M~13 field shown in Figure 1. We first select all stars inside
427: the estimated tidal radius (25.2$^{'}$; Harris 1996), shown as the innermost
428: green circle in Figure 1. This is regarded as the cluster region. The red dots
429: represent stars with available photometry from the SDSS PHOTO pipeline (Lupton
430: et al. 2001); the black dots are stars with photometry obtained from DAOPHOT.
431: The blue open circles indicate stars with available spectroscopy. We then choose
432: an annulus outside the cluster region, indicated on the Figure as the region
433: between the two black circles, as the field or background region.
434:
435: We next obtain CMDs of each region, spanning $-1.0 \le (g-r)_0 \le 1.5$ and $12
436: \le g_0 \le 22$, and then subdivide these diagrams such that the size of each
437: sub-grid is 0.2 mag wide in $g_0$ and 0.05 mag wide in $(g-r)_0$ color. The
438: total number of sub-grids for the CMDs in each region is thus 2500
439: (50$\times$50). Figure 2 shows the resulting CMDs of the cluster (left panel)
440: and field (right panel) regions, overplotted with squares representing the
441: selected sub-grids, obtained as described below.
442:
443: We first calculate the signal-to-noise ($s/n$) in each preliminary sub-grid by
444: application of Eqn. (1) over the entire CMD region shown in Figure 2. Here we
445: assume that the field stars outside the tidal radius are uniformly distributed
446: throughout the annulus area.
447:
448: \begin{equation} s/n(i,j) = \frac{n_{c}(i,j) - gn_{f}(i,j)}{\sqrt{n_{c}(i,j) + g^{2}n_{f}(i,j)}}.
449: \label{eq1}
450: \end{equation}
451:
452: \noindent In the above, $n_{c}$ and $n_{f}$ refer to the number of stars in
453: each sub-grid with color index $i$ and magnitude index $j$, counted within the
454: cluster region and field region, respectively. The parameter $g$ represents the
455: ratio of the cluster area to the field area.
456:
457: The following procedures are followed in order to find the optimal range of
458: colors and magnitudes that correspond to the likely members of each cluster.
459: First, we sort the elements of $s/n(i,j)$ in descending order, so that we obtain
460: a one-dimensional array of $s/n(i,j)$ with index $l$; the array element with the
461: highest $s/n(i,j)$ corresponds to $l=1$. The next step is to obtain star counts
462: in gradually larger regions of the CMDs. The accumulated area is represented as
463: $a_{k} = ka_{l}$, where $a_{l} = 0.01$ mag$^{2}$, which is the same for all
464: sub-grids, and is the area of a single sub-grid in the CMD array, and the $k$ is
465: the number of sub-grids to combine. Finally, the cumulative signal-to-noise
466: ratio, $S/N(a_{k}) $, as a function of $a_{k}$, is calculated from:
467:
468: \medskip
469: \begin{equation} S/N(a_{k}) = \frac{N_{c}(a_{k}) - gN_{f}
470: (a_{k})}{\sqrt{N_{c} (a_{k}) + g^{2} N_{f} (a_{k})}}
471: \label{eq3}
472: \end{equation}
473: \medskip
474:
475: \noindent where,
476:
477: \begin{equation}
478: N_{c}(a_{k}) = \sum_{l=1}^{k} n_{c}(l), ~~~N_{f}(a_{k}) = \sum_{l=1}^{k} n_{f}(l) \label{eq4}
479: \end{equation}
480: \medskip
481:
482: \noindent The parameter $n_{c}(l)$ denotes the number of stars within the
483: cluster region having ordered color-magnitude index $l$; $n_{f}(l)$ represents
484: the same quantity for the field region. Based on the maximum value of
485: $S/N(a_{k})$, a threshold value of $s/n$ is picked in order to select high-
486: contrast surface-density areas (i.e, high $s/n$) between the cluster and field
487: regions. These are considered to be the sub-grids that contain likely cluster
488: members. After removing single-star events in areas of the CMDs where the
489: field-star density is low, all stars in sub-grids with $s/n(i,j)$ greater than
490: the threshold value of $s/n$ are selected. These stars are considered as the
491: photometrically likely member stars for a given cluster.
492:
493: The red squares shown in the left panel of Figure 2 are the sub-grids with $s/n$
494: greater than the threshold value; the corresponding sub-grids in the field region
495: are shown as green squares in the right panel of this Figure. Figure 3 depicts
496: the CMD of the selected likely members of M~13 from the photometric data, shown
497: as black dots. The same procedures are performed to differentiate the likely
498: member stars of M~15 and M~2 from the photometric sample.
499:
500: We now proceed to select the stars that are likely members of the globular
501: clusters from the available spectroscopic sample. This step begins by selection
502: of the stars within the cluster tidal radii that pass the photometric criterion
503: for membership, based on their location on the cluster CMDs according to the
504: algorithm described above.
505:
506: Figure 3 displays the cleaned CMD of M~13, overplotted with the likely members
507: from the spectroscopic sample (shown as red circles). The same procedures are
508: carried out to identify likely member stars of M~15 and M~2 from their
509: spectroscopic data. Based on these cuts, at this stage of the analysis there are
510: 296 (338) likely members for M~13, 124 (160) for M~15, and 21 (22) for M~2
511: identified. In the above, the first listed numbers indicate the stars with
512: available estimates of [Fe/H] from the SSPP, while the quantities in parentheses
513: represent the number of stars with available radial velocities (RVs). Additional
514: cuts, based on the derived metallicity estimates and RVs, are described in \S 4.
515:
516: \subsection{Likely Member Star Selection for Open Clusters}
517:
518: Since the fields of nearby open clusters are not as crowded as those of globular
519: clusters, the signal-to-noise ratio between the cluster region and the
520: background region is not sufficiently high to select likely cluster members by
521: means of the CMD mask algorithm. As an alternative, we first obtain a fiducial
522: line for an open cluster (including its main sequence and sub-giant branch, if
523: it exists) by use of an robust polynomial fitting procedure. As an example,
524: Figure 4 shows the CMD of the NGC~2420 field inside a radius of 0.3 degrees from
525: the center of the cluster. According to the open cluster catalog of Dias et al.
526: (2002), the apparent diameter of this cluster is only 5$^{'}$ on the sky, but we
527: prefer to adopt a 20$^{'}$ radius, in order to include as many member stars as
528: possible. The red line is the fiducial line derived from the robust polynomial
529: fit. The blue lines are the upper and lower limits (fiducial $\pm$ 0.06 dex in
530: $(g-r)_{0}$), determined by eye. Stars from the spectroscopic sample that fall
531: within the 20$^{'}$ radius and inside the blue limit lines in Figure 4 are
532: identified.
533:
534: A similar procedure is applied to M~67, except that a 30$^{'}$ (apparent
535: diameter of 25$^{'}$; Dias et al. 2002) radius and fiducial $\pm$ 0.10 mag in
536: $(g-r)_{0}$ is used. Based on this selection method, there are 195 (234) and 61
537: (64) for NGC~2420 and M~67, respectively. The first listed numbers indicate the
538: stars with available estimates of [Fe/H] from the SSPP, while the quantities in
539: parentheses represent the number of stars with available radial velocities (RVs)
540: . Additional cuts, based on the derived metallicity estimates and RVs, are
541: described below.
542:
543:
544: \section{Determination of Overall Metallicities and Radial Velocities of the
545: Clusters}
546:
547: In order to investigate the accuracy of our derived metallicities and RVs, we
548: now consider the global distribution of these parameters obtained from the
549: current version of the SSPP for the likely cluster members. In this section, we
550: describe a method to best isolate ``true member stars'' from the spectroscopic
551: samples described above. We then use these subsamples to determine our best
552: estimates of the overall metallicities and RVs of the clusters considered in
553: this study.
554:
555: \subsection{Selection of True Members}
556:
557: We establish the criteria for carrying out metallicity and RV cuts as follows.
558: %Note that as mentioned in \S 1, because there is no manifest evidence of large
559: %dispersion in the abundance of the iron peak elements (e.g., Cohen \&
560: %Mel$\acute{\rm e}$ndez 2005) for the clusters currently under consideration. we
561: %suppose the clusters have chemically mono population and make use of the
562: %metallicity as a membership criterion.
563:
564: The left panel of Figure 5 illustrates the [Fe/H] distribution for three
565: different subsamples of stars. The first, shown as the black dot-dashed line,
566: represents the distribution of derived metallicities for the 1547 stars with
567: available estimates of [Fe/H] along the line of sight to M~13. Note that this
568: distribution includes numerous stars that cannot be considered members of the
569: cluster, as they cover a much wider range of [Fe/H] than might be expected if
570: they were drawn exclusively from the cluster member population. The dot-dashed
571: line in the right panel of Figure 5 shows the RV distribution of these same
572: stars.
573:
574: The red dashed line in the left panel of Figure 5 is the
575: distribution of [Fe/H] for the 296 likely members selected from the
576: spectroscopic sample as described above. We obtain a Gaussian fit to
577: the {\it highest peak} of the distribution of these stars (solid
578: blue line in Figure 5), and obtain an estimate of the mean
579: ($<$[Fe/H]$>$) and standard deviation ($\sigma$) for this
580: distribution. Similar fits are obtained for the distribution of RVs
581: for the likely members shown in the right panel of Figure 5. On the
582: basis of these fits, we now trim likely outliers by application of a
583: 2-$\sigma$ clipping procedure, for example:
584:
585: \medskip
586: \begin{equation}
587: \rm <[Fe/H]> - 2\sigma_{\rm [Fe/H]} \leq \rm [Fe/H]_{\star} \leq
588: \rm<[Fe/H]> + 2\sigma_{\rm [Fe/H]} \label{eq6}
589: \end{equation}
590:
591: \medskip
592: \begin{equation}
593: \rm <RV> - 2\sigma_{\rm RV} \leq \rm RV_{\star} \leq \rm<RV> +
594: 2\sigma_{\rm RV} \label{eq7}
595: \end{equation}
596:
597: \noindent In the above, [Fe/H]$_{\star}$ and RV$_{\star}$ correspond to the
598: values of these parameters for each star under consideration. The stars
599: surviving both of these clips are considered true cluster members for the
600: purpose of this study. Note that at no point have we considered the external
601: ``known'' values of [Fe/H] and RV for the clusters as a whole.
602:
603: Based on the application of these membership cuts, we now have a total of 169
604: stars identified as true members of M~13, 63 stars as true members of M~15, 9
605: stars as true members of M~2, 195 stars as true members of NGC~2420, and 51
606: stars as true members of M~67. The distribution of [Fe/H] and RV for the
607: surviving members of M~13 are shown as green histograms in the left and right
608: panels of Figure 5, respectively. Similar plots for M~15, M~2, NGC~2420, and
609: M~67 are shown in Figures 6, 7, 8, and 9, respectively. The distribution of
610: [Fe/H] for the selected true member stars of each cluster, as a function of
611: $T_{\rm eff}$, is shown in Figure 10.
612:
613: Table 2 summarizes the results of the above exercise. Column (1) lists the
614: cluster name. Columns (2) and (3) are the lower and upper limits for the
615: 2-$\sigma$ cuts on [Fe/H], respectively. Columns (4) and (5) are the
616: corresponding adopted limits on RV used for these cuts.
617:
618: Tables 4$-$8 list the observed and derived quantities for all of the
619: individual stars considered as true member stars in the analysis of
620: each cluster. The columns are as defined in the table notes for
621: Table 4.
622:
623: \subsection{Determination of Overall Estimates of Mean
624: Cluster Metallicity and Radial Velocity}
625:
626: We now obtain final estimates of the cluster metallicities and RVs based on
627: Gaussian fits to the surviving true member stars for each cluster, as shown by
628: the blue curves in Figures 5$-$9. Table 2 summarizes these determinations. The
629: mean metallicity and 1-$\sigma$ spread of the metallicities of the true member
630: stars are listed in columns (6) and (7), respectively. Similar quantities for
631: the RVs are listed in columns (8) and (9). Column (10) lists the total number of
632: true member stars associated with each cluster, based on our analysis. External
633: estimates of the metallicities and RVs for these clusters, adopted from the
634: Harris (1996) compilation for M~13, M~15, and M~2 and from WEBDA (and references
635: therein) for NGC~2420 and M~67, are listed in columns (11) and (12). Column (13)
636: lists metallicity estimates for these clusters obtained from high-resolution
637: spectroscopy of a limited number of brighter stars by Kraft \& Ivans (2003) for
638: M~15 and M~13, Ivans (private communication) for M~2, and Gratton (2000) for
639: NGC~2420 and M~67.
640:
641: For M~13, our estimate of the mean abundance, $<$[Fe/H]$>$ = $-1.56$, is very
642: close to the Harris (1996) estimate ([Fe/H]$_{\rm H}$ = $-$1.54). However, the
643: recent study of Kraft \& Ivans (2003) reported a revised cluster abundance for
644: M~13, derived from high-resolution spectroscopy of 28 giants. Their value
645: indicates a metallicity for M~13 that is a bit lower than that given by Harris,
646: [Fe/H]$_{\rm HR}$ = $-$1.63, and is lower by 0.07 dex than our estimate. Cohen
647: \& Mel$\acute{\rm e}$ndez (2005) reported [Fe/H] = $-1.50$ from a
648: high-resolution ($R$ = 35,000) analysis of a sample of 25 stars, consisting of
649: stars from the giant branch to near the main-sequence turnoff. Our derived
650: spread in the metallicities of the M~13 true member stars (0.16 dex) is also
651: satisfyingly low, especially considering the wide range of temperatures for true
652: members that are considered here. Our estimate of the mean radial velocity,
653: $<$RV$>$ = $-$245.1 km s$^{-1}$, with a standard deviation of 8.7 km s$^{-1}$,
654: is in good agreement with that given by Harris ($-$245.6 km s$^{-1}$). It is
655: important to note that, as mentioned in Paper I, we have already
656: added $+$7.3 km s$^{-1}$ to all DR-6 (Adelman-McCarthy et al. 2007b) stellar
657: radial velocities. Before the adjustment of this offset, an average offset of
658: $-$8.6 km s$^{-1}$ for M 13 and $-$6.8 km s$^{-1}$ for M 15 is obtained. Thus,
659: together with an offset of $-$6.6 km s$^{-1}$ that was obtained from a
660: preliminary result of a high-resolution spectroscopic analysis of SDSS-I/SEGUE
661: stars (Paper III) before DR-6 (Adelman-McCarthy et al. 2007b), we derived an
662: average offset of $-$7.3 km s$^{-1}$. However, a recent high-resolution
663: spectroscopic analysis of SDSS-I/SEGUE stars indicates an offset of about $-$6.9
664: km s$^{-1}$, resulting in an average of $-$7.4 km s$^{-1}$, Hence, in future data
665: releases (e.g, DR-7), this very minor difference might be reflected. For the
666: analysis of our clusters, all radial velocities have been corrected by $+$7.3 km
667: s$^{-1}$, in order to be consistent with DR-6.
668:
669: Our estimate of the mean abundance of M~15, $<$[Fe/H]$>$ = $-2.12$, is close to
670: the value listed by Harris ([Fe/H]$_{\rm H}$ = $-$2.26). While Kraft $\&$ Ivans
671: (2003) obtained [Fe/H]$_{\rm HR}$ = $-$2.42, based on high-resolution
672: spectroscopy for nine giants in this cluster, Otsuki et al. (2006) reported
673: [Fe/H] = $-$2.29 from an analysis of high-resolution spectra for six giants
674: belonging to this cluster. Our derived spread in the metallicities of true
675: member stars in M~15 is quite low (0.14 dex). Our estimate of the mean radial
676: velocity, $<$RV$>$ = $-$107.4 km s$^{-1}$, with a standard deviation of 10.5 km
677: s$^{-1}$, agrees very well with that of the Harris (1996) value ($-$107.0 km
678: s$^{-1}$).
679:
680: There are only a very small number of true member stars (9) for M~2. Their
681: average metallicity, $<$[Fe/H]$>$ = $-$1.58, is similar to the Harris (1996)
682: value ([Fe/H]$_{\rm H}$ = $-$1.62), and is in very good agreement with the value
683: obtained by Ivans (private communication) ([Fe/H]$_{\rm HR}$ = $-$1.56). The
684: estimated spread in our derived metallicities, 0.08 dex, is quite small. Our
685: estimate of the mean radial velocity, $<$RV$>$ = $+$0.4 km s$^{-1}$, with a
686: standard deviation of 7.7 km s$^{-1}$, is higher (by about 6 km s$^{-1}$) than
687: that provided by Harris ($-$5.3 km s$^{-1}$). Clearly, for the purposes of
688: validation of the SSPP, it would be highly desirable to obtain a larger number
689: of member stars in M~2; a new plate (640 spectra) will be obtained in the near
690: future. Note that M~2 presents a special challenge, since its mean metallicity
691: is quite close to that expected for members of the field halo population, while
692: its radial velocity is buried in the peak of foreground disk stars. Our
693: stringent criterion for true cluster members should remain effective, however,
694: since few non-cluster members will fulfill both the RV and metallicity criteria.
695:
696: There are 130 true member stars selected for the open cluster NGC~2420. The mean
697: iron abundance of the selected true member stars is $<$[Fe/H]$>$ = $-$0.46,
698: which is in excellent agreement with the value ([Fe/H] = $-$0.44) determined by
699: Gratton (2000) from high-resolution spectroscopy of one member star. Friel $\&$
700: Janes (1993) reported [Fe/H] = $-$0.42 for nine member stars, based on medium-
701: and low-resolution spectroscopic data. Friel et al. (2002) determined [Fe/H] =
702: $-$0.38 $\pm$ 0.07, based on medium-resolution spectra of 20 member stars. Most
703: of these literature values are within the spread of our derived value. The
704: derived spread in the metallicities of true member stars (0.12 dex) is very low.
705: The radial velocity for NGC~2420 listed in Table 2, ($+74.0$ km s$^{-1}$), is an
706: average of the values $+84.0$ km s$^{-1}$, $+71.1$ km s$^{-1}$, and $+67.0$ km
707: s$^{-1}$ from Friel (1989), Scott et al. (1995), and Rastorguev et al. (1999),
708: respectively. This value agrees very well with our derived estimate of $+74.8$
709: km s$^{-1}$, with a standard deviation of $6.2$ km s$^{-1}$.
710:
711: For M 67, 51 stars are identified as true member stars. A mean metallicity of
712: $<$[Fe/H]$> = -$0.35 is derived, with a small spread of 0.15 dex. This derived
713: $<$[Fe/H]$>$ differs by 0.37 dex from that of Gratton (2000), [Fe/H] = $+$0.02,
714: who derived this value from a high-resolution study of one member star. Randich
715: et al. (2007) analyzed 10 member stars of this cluster, based on high-resolution
716: ($R \sim 45,000$) spectroscopy, and derived [Fe/H] = $+0.03 \pm 0.01$. [Fe/H] =
717: $+0.02 \pm 0.03$ was determined by Yong et al. (2005) from a high-resolution
718: spectroscopic analysis of three member stars. However, based on
719: medium-resolution spectra of 25 members, Friel et al. (2002) reported [Fe/H] =
720: $-0.15 \pm 0.05$. Other catalogs of open clusters (e.g., Twarog et al. 1997;
721: Chen et al. 2003) also report a solar metallicity for this cluster. The
722: literature values based on high-resolution spectroscopic analyses clearly
723: suggest that the present SSPP tends to under-estimate [Fe/H] by about 0.3 dex
724: for near-solar-metallicity stars. This is perhaps related to the difficulties
725: arising from the strong atomic and molecular lines (and possibly unreliable
726: synthetic spectra) for metal-rich stars. As mentioned by Gratton (2000), it
727: might be desirable to re-calibrate the metallicity scale used for the analysis
728: of medium-resolution spectra for stars with the solar iron abundance to better
729: match the results obtained from high-resolution analyses. The radial velocity of
730: $+32.9$ km s$^{-1}$ for M 67 from the literature listed in Table 2 (the average
731: of the Scott et al. 1995; Friel $\&$ Janes 1993; Rastorguev et al. 1999 values)
732: agrees with our derived mean radial velocity of $+34.9$ km s$^{-1}$ within the
733: standard deviation of our measurements, 5.6 km s$^{-1}$.
734:
735: Thus, taking into account only the scatter in the metallicities and radial
736: velocities calculated from the members of each cluster, we are able to derive
737: estimates of typical external uncertainties for the SSPP values, $\sigma (\rm [Fe/H])$ =
738: 0.13 dex, and $\sigma (\rm RV)$ = 7.7 km~s$^{-1}$ (after 5$\sigma$ clipping is
739: applied).
740:
741:
742: \section{A Comparison of Derived $T_{\rm eff}$ and log $g$ for
743: True Cluster Members with Color-Magnitude Diagrams}
744:
745: In the previous section, we have considered the accuracy with which the SSPP
746: obtains estimates of metallicity and radial velocity. We now consider the
747: accuracy with which the SSPP obtains estimates of effective temperatures,
748: $T_{\rm eff}$, and surface gravities, log $g$. One excellent ``global'' test of
749: these estimates is to examine the locations of the true member stars on the
750: observed CMD (based on the totality of likely photometric member data) for each
751: cluster. One can also compare with corresponding theoretical CMDs.
752:
753: Figures 11$-$15 show plots of the SSPP-estimated temperatures and gravities for
754: true member stars superposed on the photometrically cleaned CMDs for each of our
755: clusters. Note that in order to obtain the theoretical temperature scales (shown
756: along the top of the left-hand panels in each Figure), we make use of a linear
757: relation between $(g-r)_0$ color and $T_{\rm eff}$ by performing a least square
758: fit in this plane to the theoretical models of Girardi et al. (2004). We choose
759: the isochrones from this study that have the closest [Fe/H] to the derived
760: metallicity of each cluster, and adopt an age of 13.5 Gyr for the globular
761: clusters, 2.2 Gyr for NGC 2420, and 4.3 Gyr for M 67 (adopting the ages from
762: WEBDA for the open clusters). A similar procedure is applied for transforming the
763: $g_0$ magnitude to a theoretical log $g$ scale (shown along the far right axis
764: in the right-hand panels of each Figure). Distance moduli from the Harris (1996)
765: compilation for the globular clusters and WEBDA for the open clusters (also
766: listed in Table 1 of this study) are used in order to compute apparent
767: magnitudes.
768:
769: In these Figures, we plot the SSPP-estimated parameters for true member stars in
770: different colors, corresponding to different ranges of temperature and surface
771: gravity (as shown in the legend for each plot). Each color represents a range of
772: 500~K in $T_{\rm eff}$ and 0.5 dex in log $g$. The effective temperature
773: estimated by the SSPP appears in excellent agreement for most of the true member
774: stars, with only a few exceptions. Such stars could either be outright errors in
775: SSPP predictions, or could just be foreground/background stars that survived the
776: various membership cuts we have applied. Inspection of the Figures also reveals
777: the presence of a few stars close to the main-sequence TO in M~13 and M~15 that
778: appear to have slightly lower SSPP-estimated log $g$ than expected from the
779: theoretical scale. The surface gravities of stars along the RGB appear to be
780: very well estimated. Such behavior is perhaps to be expected, since the stars
781: close to the TO region are at the low end of the $S/N$ range that we accept for
782: analysis, and thus are subject to greater errors in the determination of their
783: atmospheric parameters. The RGB stars are among the brightest, and hence are
784: likely to have the best-determined estimates.
785:
786: Inspection of Figure 14 for NGC 2420 indicates that gravity estimates for
787: most of the main-sequence stars are well-estimated from the SSPP, with the
788: exception of the faintest stars. These stars have only low $S/N$ spectra
789: available, resulting in higher uncertainties in determinations of their surface
790: gravities. It should also be recalled that surface gravity is a difficult
791: parameter to estimate, especially from spectra of the resolving power obtained
792: by the SDSS. Overall, we are pleased to see as good a behavior in the estimates
793: of this parameter as is demonstrated in Figures 11$-$15.
794:
795: In addition, using the derived relations between $(g-r)_0$ and $T_{\rm eff}$,
796: and $g_0$ and log~$g$ from the isochrones, we predicted $T_{\rm eff}$ and
797: log~$g$ from the observed $(g-r)_0$ and $g_0$, respectively. Table 3 lists the
798: averages and standard deviations of the residuals of the effective temperatures
799: and surface gravities between the SSPP estimates and the calculated values. Even
800: though we have employed a simple relationship between $(g-r)_0$ with $T_{\rm
801: eff}$, we see good agreement between the SSPP estimates and the theoretical
802: values in $T_{\rm eff}$. Although, as expected, we notice a rather large offset
803: and scatter in the gravity, indicating a more complex function is needed, the
804: scatters are still within each bin size (0.5 dex) in Figures 11$-$15. M~2
805: exhibits a larger scatter in both $T_{\rm eff}$ and log~$g$ than the other
806: clusters, owing to the small number of member stars selected.
807:
808:
809: \section{Summary and Conclusions}
810:
811: Based on photometric and spectroscopic data reported in SDSS-I and
812: SDSS-II/SEGUE, we have examined estimates of stellar atmospheric parameters and
813: heliocentric radial velocities obtained by the SEGUE Stellar Parameter
814: Pipeline (SSPP) for likely members of three Galactic globular clusters, M~13,
815: M~15, and M~2, and two open clusters, NGC~2420 and M~67, and compared them with
816: those obtained by external estimates for each cluster as a whole.
817:
818: From the derived scatters in the metallicities and radial velocities obtained
819: for the likely members of each cluster, we quantify the typical external
820: uncertainties of the SSPP-determined values, $\sigma (\rm [Fe/H])$ = 0.13 dex,
821: and $\sigma (\rm RV)$ = 7.7 km~s$^{-1}$, respectively. These uncertainties apply
822: for stars in the range of 4500 K $\le T_{\rm eff} \le 7500$ K and $2.0 \le \log
823: g \le 5.0$, at least over the metallicity interval spanned by the clusters
824: studied ($-2.3 \le {\rm [Fe/H]} < -0.4$). Therefore, the metallicities and
825: radial velocities obtained by the SSPP appear sufficiently accurate to be used
826: for studies of the kinematics and chemistry of the metal-poor and moderately
827: metal-rich stellar populations in the Galaxy. We have also confirmed that
828: $T_{\rm eff}$ and log $g$ are sufficiently well-determined by the SSPP to
829: distinguish between different luminosity classes through a comparison with
830: theoretical predictions.
831:
832: A comparison of the analysis of the available high-resolution spectroscopy of
833: SDSS-I/SEGUE stars (Paper III) with the SSPP predictions indicates that the
834: uncertainty in radial velocities adopted by the SSPP is no more than 5
835: km~s$^{-1}$ (after adjusting for an empirical offset of +7.3 km~s$^{-1}$). The
836: empirically determined precisions in estimated atmospheric parameters are $\sim$
837: 130~K for effective temperature, $\sim$ 0.21 dex for surface gravity, and $\sim$
838: 0.11 dex for [Fe/H]. These errors apply to the brightest stars obtained by
839: SDSS-I/SEGUE observations, on the order of $14.0 \le g \le 15.5$, and are
840: expected to degrade somewhat for fainter stars. We also found that the SSPP
841: tends to underestimate [Fe/H] for near-solar-metallicity stars (represented by
842: members of M~67 in this study), by $\sim$ 0.3 dex.
843:
844: In future papers we will compare the predictions of the SSPP with
845: intermediate-metallicity clusters ([Fe/H] $\sim$ $-$0.7) and with additional
846: near-solar-metallicity populations, as sampled by metal-rich globular clusters
847: and nearby open clusters. Additionall metal-poor clusters will also be examined.
848: Further refinements in the SSPP, which hopefully will be better able to
849: recover accurate abundances for near-solar-metallicity stars, are anticipated.
850:
851: \acknowledgements
852:
853: Funding for the SDSS and SDSS-II has been provided by the Alfred P.
854: Sloan Foundation, the Participating Institutions, the National
855: Science Foundation, the U.S. Department of Energy, the National
856: Aeronautics and Space Administration, the Japanese Monbukagakusho,
857: the Max Planck Society, and the Higher Education Funding Council for
858: England. The SDSS Web Site is http://www.sdss.org/.
859:
860: The SDSS is managed by the Astrophysical Research Consortium for the
861: Participating Institutions. The Participating Institutions are the
862: American Museum of Natural History, Astrophysical Institute Potsdam,
863: University of Basel, University of Cambridge, Case Western Reserve
864: University, University of Chicago, Drexel University, Fermilab, the
865: Institute for Advanced Study, the Japan Participation Group, Johns
866: Hopkins University, the Joint Institute for Nuclear Astrophysics,
867: the Kavli Institute for Particle Astrophysics and Cosmology, the
868: Korean Scientist Group, the Chinese Academy of Sciences (LAMOST),
869: Los Alamos National Laboratory, the Max-Planck-Institute for
870: Astronomy (MPIA), the Max-Planck-Institute for Astrophysics (MPA),
871: New Mexico State University, Ohio State University, University of
872: Pittsburgh, University of Portsmouth, Princeton University, the
873: United States Naval Observatory, and the University of Washington.
874:
875: Y.S.L., T.C.B., and T.S. acknowledge partial funding of this work
876: from grant PHY 02-16783: Physics Frontiers Center / Joint Institute
877: for Nuclear Astrophysics (JINA), awarded by the U.S. National
878: Science Foundation. NASA grants (NAG5-13057, NAG5-13147) to C.A.P.
879: are thankfully acknowledged. J.E.N acknowledges support from
880: Australian Research Council Grant DP0663562. C.B.J and P.R.F
881: acknowledge support from the Deutsche Forschungsgemeinschaft (DFG)
882: grant BA2163.
883:
884:
885: \clearpage
886:
887:
888: \begin{thebibliography}{}
889:
890: \bibitem[]{} Abazajian, K., et al. 2003, \aj, 126, 2081
891:
892: \bibitem[]{} Abazajian, K., et al. 2004, \aj, 128, 502
893:
894: \bibitem[]{} Abazajian, K., et al. 2005, \aj, 129, 1755
895:
896: \bibitem[]{} Adelman-McCarthy, J. K., et al. 2007a, \apjs, in press
897:
898: \bibitem[]{} Adelman-McCarthy, J. K., et al. 2007b, \apjs, accepted
899:
900: \bibitem[]{} Allende Prieto, C., et al. 2007, \aj, submitted (Paper III)
901:
902: \bibitem[]{} Allende Prieto, C., Beers, T. C., Wilhelm, R., et al.
903: 2006, \apj, 636, 804
904:
905: \bibitem[]{} Bailer-Jones, C. A. L. 2000, \aap, 357, 197
906:
907: \bibitem[]{} Beers, T. C., et al. 2006, BAAS, 38, 168.08
908:
909: \bibitem[]{} Beers, T. C., Rossi, S., Norris, J. E., Ryan, S. G.,
910: $\&$ Shefler, T. 1999, \apj, 506, 892
911:
912: \bibitem[]{} Carollo, D., et al. 2007, Nature, submitted (astro-ph/0706.3005)
913:
914: \bibitem[]{} Cenarro A.J., Cardiel N., Gorgas J., Peletier R.F., Vazdekis A., $\&$ Prada F. 2001a, \mnras, 326, 959
915:
916: \bibitem[]{} Cenarro A. J., Gorgas J., Cardiel N., Pedraz S., Peletier R.F., $\&$ Vazdekis, A. 2001b, \mnras, 326, 981
917:
918: \bibitem[]{} Chen, L., Hou, J.L., $\&$ Wang, J.J. 2003, \aj, 125, 1397,
919:
920: \bibitem[]{} Cohen, J. G., Briley, M. M., $\&$ Stetson, P. B. 2005, \aj, 130, 1177
921:
922: %\bibitem[]{} Cohen, J. G. $\&$ Mel$\acute{\rm e}$ndez, J. 2005, \aj, 129, 303
923:
924: \bibitem[]{} Cudworth, K. M. 1976, \aj, 81, 519
925:
926: \bibitem[]{} Cudworth, K. M. $\&$ Monet, D. G. 1979, \aj, 84, 774
927:
928: \bibitem[]{} Cudworth, K. M. $\&$ Rauscher, B. 1987, \aj, 93, 856
929:
930: \bibitem[]{} Dias, W. S., Alessi, B. S., Moitinho, A., $\&$ Lepine, J. R. D. 2002, \aap, 389, 871
931:
932: \bibitem[]{} Fan, X., et al. 1996, \aj, 112, 628
933:
934: \bibitem[]{} Friel, E. D.1989, \pasp, 101, 244
935:
936: \bibitem[]{} Friel, E. D. $\&$ Janes, K. A. 1993, \aap, 267, 75
937:
938: \bibitem[]{} Friel, E. D., Janes, K. A., Tavarez, M., Scott, J., et al. 2002, \aj, 124, 2693
939:
940: \bibitem[]{} Fukugita, M., Ichikawa, T., Gunn, J.E., Doi, M., Shimasaku, K.,
941: $\&$ Schneider, D.P. 1996, \aj, 111, 1748
942:
943: \bibitem[]{} Girardi, L., Grebel, E. K., Odenkirchen, M., $\&$ Chiosi, C. 2004, \aap, 422, 205
944:
945: \bibitem[]{} Gratton, R. 2000, in Stellar Clusters and Associations: Convection,
946: Rotation, and Dynamos, ASP Conference Series
947: (eds. R. Pallavicini, G. Micela, \& S. Sciortino), 198, p. 225
948:
949: \bibitem[]{} Grillmair, C. J., Freeman, K. C., Irwin, M., $\&$ Quinn, P. J. 1995, \aj, 109, 2553
950:
951: \bibitem[]{} Gunn, J. E., et al. 1998, \aj, 116, 3040
952:
953: \bibitem[]{} Gunn, J. E., et al. 2006, \aj, 131, 2332
954:
955: \bibitem[]{} Harris, W. E. 1996, \aj, 112, 1487
956:
957: \bibitem[]{} Hogg, D.W., Finkbeiner, D.P., Schlegel, D.J., $\&$ Gunn, J.E. 2001,
958: \aj, 122, 2129
959:
960: \bibitem[]{} Ivezic, Z., et al. 2004, Astron. Nach., 325, 583
961:
962: \bibitem[]{} Johnson, J. A. et al. 2007, in preparation
963:
964: \bibitem[]{} Kraft, R. P. $\&$ Ivans, I. I. 2003, \pasp, 115, 143
965:
966: \bibitem[]{} Lee, Y. S., et al. 2007a, \aj, submitted (Paper I)
967:
968: \bibitem[]{} Lupton, R., et al. 2001, in ASP Conf. Ser. 238, Astronomical Data Analysis Software and Systems
969: X, ed. F. R. Harnden, Jr., F. A. Primini, and H. E. Payne (San Francisco: Astr. Soc. Pac.), p. 269
970:
971: \bibitem[]{} Morrison, H. L., Norris, J., Mateo, M., et al. 2003, \aj, 125, 2502
972:
973: \bibitem[]{} Moultaka, J., Ilovaisky, S. A., Prugniel, P., $\&$ Soubiran, C. 2004, \pasp, 116, 693
974:
975: \bibitem[]{} Otsuki, K., Honda, S., Aoki, W., Kajino, T., $\&$ Mathews, G. 2006, \apj, 641L, 117
976:
977: \bibitem[]{} Pier, J.R., Munn, J.A., Hindsley, R.B., Hennessy, G.S.,
978: Kent, S.M., Lupton, R.H., $\&$ Ivezic, Z. 2003, \aj, 125, 1559
979:
980: \bibitem[]{} Prugniel, Ph. $\&$ Soubiran, C., 2001, \aap, 369,1048
981:
982: \bibitem[]{} Randich, S., Sestito, P., Primas, F., Pallavicini, R., $\&$ Pasquini, L., 2006, \aap, 450, 557
983:
984: \bibitem[]{} Rastorguev, A.S., Glushkova, E.V., Dambis, A.K., $\&$
985: Zabolotskikh M.V. 1999, Astron. Letters, 25, 689
986:
987: \bibitem[]{} Re Fiorentin, P., Bailer-Jones, C. A. L., Lee, Y. S. et al. 2007, \aap, 467, 1373
988:
989: %\bibitem[]{} S\'anchez-Bl\'azquez, P., Peletier, R. F, Jim\'enez-Vicente, J., et al. 2006, \mnras, 371, 703
990:
991: \bibitem[]{} Sanders, W. L. 1989, Rev., Mex. Astron. Astro. 17, 31
992:
993: \bibitem[]{} Schlegel, D. J., Finkbeiner, D. P., $\&$ Davis, M., 1998, \apj, 500, 525
994:
995: \bibitem[]{} Scott, J. E., Friel, E. D., $\&$ Janes, K. A. 1995, \aj, 109, 1706
996:
997: \bibitem[]{} Smith, G. H $\&$ Briley, M. M. 2006, \pasp, 118, 740
998:
999: \bibitem[]{} Smith, G. H. $\&$ Mateo, M. 1990, \apj, 353, 533
1000:
1001: \bibitem[]{} Smith, J.A., et al 2002, \aj, 123, 2121
1002:
1003: \bibitem[]{} Stetson, P. B. 1987, \pasp, 99, 191
1004:
1005: \bibitem[]{} Stetson, P. B. 1990, \pasp, 102, 932
1006:
1007: \bibitem[]{} Stetson, P. B. 1994, \pasp, 106, 250
1008:
1009: \bibitem[]{} Stoughton, C., et al. 2002, \aj, 123, 485
1010:
1011: \bibitem[]{} Twarog, B.A., Ashman, K.M., $\&$ Anthony-Twarog, B.J. 1997, \aj, 114, 2556
1012:
1013: \bibitem[]{} Tucker, D., et al. 2006, AN, 327, 821
1014:
1015: \bibitem[]{} Willemsen, P.G., Hilker, M., Kayser, A., $\&$
1016: Bailer-Jones, C. A. L. 2005, \aap, 436, 379
1017:
1018: \bibitem[]{} Yong, D., Carney, B. W., $\&$ Teixera de Almeida, M. L. 2005, \aj, 130, 597
1019:
1020: \bibitem[]{} York, D. G., et al. 2000, \aj, 120, 1579
1021:
1022: \end{thebibliography}
1023:
1024:
1025: \clearpage
1026:
1027:
1028: %% Figure 1
1029: \begin{figure}
1030: \centering
1031: \plotone{f1.eps}
1032: %\plotone{f1.eps}
1033: \caption{Stars with
1034: available photometry in the field of M~13. The red dots represent
1035: photometry from the SDSS PHOTO pipeline, while the black dots are
1036: from the crowded-field photometry analysis. Blue open circles
1037: indicate stars with available SDSS spectroscopy. The green circle is
1038: the tidal radius. The region inside this radius is regarded as the
1039: cluster region; the annulus between the two black circles is
1040: considered the field region.}
1041: \end{figure}
1042: \clearpage
1043:
1044:
1045: %% Figure 2
1046: \begin{figure}
1047: \centering
1048: \includegraphics[angle=90,scale=0.60]{f2.eps}
1049: %\includegraphics[angle=90,scale=0.60]{f5.eps}
1050: \caption{Color-Magnitude Diagrams of the M~13 stars inside the tidal
1051: radius (left panel), and inside the field region (right panel),
1052: shown as black dots. The selected sub-grids from the $S/N$ cut are
1053: shown as red squares in the left panel and green squares in the
1054: right panel. These selected sub-grids are used in the analysis.}
1055:
1056: \end{figure}
1057: \clearpage
1058:
1059: %% Figure 3
1060: \begin{figure}
1061: \centering \plotone{f3.eps}
1062: %\plotone{f6.eps}
1063: \caption{The M~13 Color-Magnitude Diagram based on the likely member
1064: stars (black dots) selected from the photometric sample. The likely
1065: members identified from the spectroscopic sample are indicated with
1066: red circles.}
1067: \end{figure}
1068: \clearpage
1069:
1070: %% Figure 4
1071: \begin{figure}
1072: \centering \plotone{f4.eps}
1073: %\plotone{n2420cmdsel}
1074: \caption{Color-Magnitude Diagram of the NGC~2420 field. The red
1075: line is the fiducial line obtained by application of a robust
1076: fourth-order polynomial fit. The stars inside the blue lines
1077: (fiducial $\pm$ 0.06 mag in $(g-r)_0$) are regarded as likely member
1078: stars from the photometric sample.}
1079: \end{figure}
1080: \clearpage
1081:
1082: %% Figure 5
1083: \begin{figure}
1084: \centering
1085: \includegraphics[angle=90,scale=0.60]{f5.eps}
1086: %\includegraphics[angle=90,scale=0.60]{m13mtrvhist.eps}
1087: \caption{[Fe/H] and radial velocity distributions for stars in the
1088: direction of M~13. Gaussian fits (blue solid curves) to the
1089: distribution of the selected true members, shown in the green
1090: histograms, are over-plotted. }
1091: \end{figure}
1092: \clearpage
1093:
1094: %% Figure 6
1095: \begin{figure}
1096: \centering
1097: \includegraphics[angle=90,scale=0.60]{f6.eps}
1098: %\includegraphics[angle=90,scale=0.60]{m15mtrvhist.eps}
1099: \caption{Same as Fig. 5 but for M~15.}
1100: %\caption{[Fe/H] and radial velocity
1101: %distributions for stars in the direction of M~15. Gaussian fits
1102: %(blue solid curves) to the distribution of the selected true
1103: %members, shown in the green histograms, are over-plotted.}
1104: \end{figure}
1105: \clearpage
1106:
1107: %% Figure 7
1108: \begin{figure}
1109: \centering
1110: \includegraphics[angle=90,scale=0.60]{f7.eps}
1111: %\includegraphics[angle=90,scale=0.60]{m2mtrvhist.eps}
1112: \caption{Same as Fig. 5 but for M~2.}
1113: %\caption{[Fe/H] and radial velocity
1114: %distributions for stars in the direction of M~2. Gaussian fits (blue
1115: %solid curves) to the distribution of the selected true members,
1116: %shown in the green histograms, are over-plotted.}
1117: \end{figure}
1118: \clearpage
1119:
1120: %% Figure 8
1121: \begin{figure}
1122: \centering
1123: \includegraphics[angle=90,scale=0.60]{f8.eps}
1124: %\includegraphics[angle=90,scale=0.60]{n2420mtrvhist.eps}
1125: \caption{Same as Fig. 5 but for NGC~2420.}
1126: %\caption{[Fe/H] and radial velocity
1127: %distributions for stars in the direction of NGC~2420. Gaussian fits
1128: %(blue solid curves) to the distribution of the selected true
1129: %members, shown in the green histograms, are over-plotted.}
1130: \end{figure}
1131: \clearpage
1132:
1133: %% Figure 9
1134: \begin{figure}
1135: \centering
1136: \includegraphics[angle=90,scale=0.60]{f9.eps}
1137: %\includegraphics[angle=90,scale=0.60]{m67mtrvhist.eps}
1138: \caption{Same as Fig. 5 but for M~67.}
1139: %\caption{[Fe/H] and radial velocity
1140: %distributions for stars in the direction of M~67. Gaussian fits
1141: %(blue solid curves) to the distribution of the selected true
1142: %members, shown in the green histograms, are over-plotted.}
1143: \end{figure}
1144:
1145: %% Figure 10
1146: \begin{figure}
1147: \centering
1148: \includegraphics[scale=0.8]{f10.eps}
1149: %\includegraphics[scale=0.8]{gcmt.eps}
1150: \caption{Distribution of [Fe/H] as a function of effective
1151: temperature for selected true member stars of M~13, M~15, M~2, NGC
1152: 2420, and M 67. The mean [Fe/H] determined for each cluster based on
1153: these estimates is shown with the blue dashed line; the red solid
1154: line represents the adopted literature value in each panel.}
1155: \end{figure}
1156: \clearpage
1157:
1158: %% Figure 11
1159: \begin{figure}
1160: \centering
1161: \includegraphics[angle=90,scale=0.60]{f11.eps}
1162: %\includegraphics[angle=90,scale=0.60]{m13cmdtmgr.eps}
1163: \caption{Temperature and gravity distributions of the selected true
1164: member stars for M~13. Each color represents a temperature range of
1165: width 500~K, and a surface gravity range of 0.5 dex, respectively.
1166: The temperature scales on the top of the left hand panel come from a
1167: linear relation between $(g-r)_0$ color and $T_{\rm eff}$ by
1168: performing a least squares fit to the theoretical models of Girardi
1169: et al. (2004). A similar procedure is applied for transforming the
1170: $g_0$ magnitude to a theoretical log $g$ scale on the ordinate in
1171: the right-hand panel.}
1172: \end{figure}
1173: \clearpage
1174:
1175: %% Figure 12
1176: \begin{figure}
1177: \includegraphics[angle=90,scale=0.60]{f12.eps}
1178: %\includegraphics[angle=90,scale=0.60]{m15cmdtmgr.eps}
1179: \caption{Same as Fig. 11 but for M~15.}
1180: %\caption{Temperature and gravity
1181: %distributions of the selected true member stars for M~15. Each color
1182: %represents a temperature range of width 500~K, and a surface gravity
1183: %range of 0.5 dex, respectively. The temperature scales on the top of
1184: %the left hand panel come from a linear relation between $(g-r)_0$
1185: %color and $T_{\rm eff}$ by performing a least squares fit to the
1186: %theoretical models of Girardi et al. (2004). A similar procedure is
1187: %applied for transforming the $g_0$ magnitude to a theoretical log
1188: %$g$ scale on the ordinate in the right-hand panel.}
1189: \end{figure}
1190: \clearpage
1191:
1192: %% Figure 13
1193: \begin{figure}
1194: \centering
1195: \includegraphics[angle=90,scale=0.60]{f13.eps}
1196: %\includegraphics[angle=90,scale=0.60]{m2cmdtmgr.eps}
1197: \caption{Same as Fig. 11 but for M~2.}
1198: %\caption{Temperature and gravity
1199: %distributions of the selected true member stars for M~2. Each color
1200: %represents a temperature range of width 500~K, and a surface gravity
1201: %range of 0.5 dex, respectively. The temperature scales on the top of
1202: %the left hand panel come from a linear relation between $(g-r)_0$
1203: %color and $T_{\rm eff}$ by performing a least squares fit to the
1204: %theoretical models of Girardi et al. (2004). A similar procedure is
1205: %applied for transforming the $g_0$ magnitude to a theoretical log
1206: %$g$ scale on the ordinate in the right-hand panel.}
1207: \end{figure}
1208: \clearpage
1209:
1210: %% Figure 14
1211: \begin{figure}
1212: \centering
1213: \includegraphics[angle=90,scale=0.60]{f14.eps}
1214: %\includegraphics[angle=90,scale=0.60]{n2420cmdtmgr.eps}
1215: \caption{Same as Fig. 11 but for NGC~2420.}
1216: %\caption{Temperature and gravity distributions of the selected true
1217: %member stars for NGC~2420. Each color represents a temperature range
1218: %of width 500~K, and a surface gravity range of 0.5 dex,
1219: %respectively. The temperature scales on the top of the left hand
1220: %panel come from a linear relation between $(g-r)_0$ color and
1221: %$T_{\rm eff}$ by performing a least squares polynomial fit to the
1222: %theoretical models of Girardi et al. (2004). A similar procedure is
1223: %applied for transforming the $g_0$ magnitude to a theoretical log
1224: %$g$ scale on the ordinate in the right-hand panel.}
1225: \end{figure}
1226: \clearpage
1227:
1228: % Figure 15
1229: \begin{figure}
1230: \centering
1231: \includegraphics[angle=90,scale=0.60]{f15.eps}
1232: %\includegraphics[angle=90,scale=0.60]{m67cmdtmgr.eps}
1233: \caption{Same as Fig. 11 but for M~67.}
1234: %\caption{Temperature and gravity
1235: %distributions of the selected true member stars for M~67. Each color
1236: %represents a temperature range of width 500~K, and a surface gravity
1237: %range of 0.5 dex, respectively. The temperature scales on the top of
1238: %the left hand panel come from a linear relation between $(g-r)_0$
1239: %color and $T_{\rm eff}$ by performing a least squares fit to the
1240: %theoretical models of Girardi et al. (2004). A similar procedure is
1241: %applied for transforming the $g_0$ magnitude to a theoretical log
1242: %$g$ scale on the ordinate in the right-hand panel.}
1243: \end{figure}
1244: \clearpage
1245:
1246: \input{tab1.tex}
1247: \clearpage
1248:
1249: \input{tab2.tex}
1250: \clearpage
1251:
1252: \input{tab3.tex}
1253: \clearpage
1254:
1255: \input{tab4.tex}
1256: %\input{m13tab.tex}
1257: \clearpage
1258:
1259: \input{tab5.tex}
1260: %\input{m15tab.tex}
1261: \clearpage
1262:
1263: \input{tab6.tex}
1264: %\input{m2tab.tex}
1265: \clearpage
1266:
1267: \input{tab7.tex}
1268: %\input{2420tab.tex}
1269: \clearpage
1270:
1271: \input{tab8.tex}
1272: %\input{m67tab.tex}
1273: \clearpage
1274: \end{document}
1275: